CHAPTER 11 Immunologysun-lab.med.nyu.edu/files/sun-lab/attachments/CPMB.Ch.11.Immunology.pdf · in this book documents the importance of immunological methods to the puriﬁcation

CHAPTER 11Immunology

INTRODUCTION

C ertain technological advances in the field of molecular biology were made possiblein part by earlier progress in the field of immunology. A review of the earlier chapters

in this book documents the importance of immunological methods to the purification ofproteins as well as to the identification of specific cDNA clones. Specific antibodies havegreatly facilitated the purification of proteins by immunoaffinity chromatography (UNIT

10.11) and immunoprecipitation (UNIT 10.16). One limitation of immunoaffinity chromatog-raphy has been that the harsh dissociation conditions required to elute bound antigensfrom high-affinity antibodies sometimes denature the eluted antigens. UNIT 11.18 presents amethod to circumvent this problem by utilizing polyol-responsive antibodies that releasetheir bound antigens under gentle dissociation conditions, employing a combination ofvarious low molecular weight polyhydroxylated compounds (e.g., ethylene glycol) andnonchaotropic salts (e.g., ammonium sulfate). These polyol-responsive antibodies canbe readily identified and isolated from typical fusions, prepared by standard hybridomaprocedures. When pure protein has been unavailable for deducing the complementaryoligonucleotide sequence, specific antibodies have been utilized to screen recombinantDNA libraries for the desired cDNA clones (UNIT 6.7) and selected mRNA for the trans-lation of desired protein (UNIT 6.8). Specific antibodies have also been utilized to identifyantigen by western blotting (UNIT 10.8).

Just as immunology has facilitated the advances made in the field of molecular biology,the latter in turn has contributed to a better understanding of the basis for antibodydiversity. The clonal selection theory proposed by Sir Macfarlane Burnet in 1959 isnow an accepted concept: each B cell differentiates into a plasma cell committed to theproduction of antibodies specific for one antigen—i.e., the antibodies are monoclonalin nature. “Clonal selection” refers to the fact that when an antigen binds to one ofthese antibodies on the membrane of the B cell, the cell is stimulated to proliferate (atwhich point some variation may be introduced in the “monoclonal” cell line). Generally,many clones respond to a single antigen, as most proteins carry multiple antigenic sites(called epitopes). The overall immune response is polyclonal, with specific recognitionof multiple, discrete epitopes.

An understanding of the genetic mechanisms responsible for antibody (or immunoglob-ulin) diversity requires some knowledge of antibody structure. Man has five major im-munoglobulin classes: IgG, IgA, IgD, IgE, and IgM, which share the same type of com-bining site for antigen. The immunoglobulin molecule is similar for the first four classes;it consists of four polypeptides—two heavy chains and two light chains—arranged inthe shape of the letter “Y,” with a molecular weight of ∼150,000. The IgM class, witha molecular weight of ∼800,000, consists of five Y-shaped molecules arranged in acyclic pentamer, with the antigen-binding sites facing outward. Although the differentimmunoglobulin classes can share the same κ or λ light chains, they are each distin-guished by their unique heavy chains, designated γ (IgG), α (IgA), δ (IgD), ε (IgE),and µ (IgM). The heavy and light chains are each composed of constant and variableregions. The antigen-binding site, a cleft of about 15 A × 20 A × 10 A deep formed by

Contributed by John A. SmithCurrent Protocols in Molecular Biology (2005) 11.0.1-11.0.3Copyright C© 2005 by John Wiley & Sons, Inc.

Immunology

11.0.1

Supplement 72

Introduction

11.0.2

Supplement 72 Current Protocols in Molecular Biology

interactions of hypervariable regions of the heavy- and light-chain variable regions, isunique for each antibody.

For many years it was assumed that the mammalian germ line must include a separategene for every polypeptide that ultimately appears in an antibody; this model presup-poses a vast number of immunoglobulin genes. In the past decade, however, recombinantDNA technology has shown that diversity in antigen-binding sites arises through geneticrecombination in somatic cells—i.e., while B lymphocytes are maturing and differen-tiating in the bone marrow. Located on different chromosomes are approximately 50genes coding for the “constant” C regions, the “variable” V regions, the “joining” Jsegments (which combine with the C and V regions to make up the antibody’s lightchain) and the “diversity” D segments (which combine with C, J, and V regions to com-prise the antibody’s heavy chain). Mouse germ cells have a few hundred V segments,approximately 20 D segments, and 4 J segments, which can be assembled in >10,000combinations. Subsequent assemblage of heavy and light chains could yield >10 millionspecific antigen-binding sites. (For an excellent review of the molecular biology of theimmune system, see Tonegawa, 1985.)

This chapter presents the methodologies for the preparation of both monoclonal and poly-clonal antibodies. Section I describes the enzyme-linked immunosorbent assay (ELISA),a highly sensitive, versatile, and quantitative technique that requires little equipment andfor which critical reagents are readily available. The preparation of enzyme-antibodyconjugates, which forms the basis of this assay, is described in UNIT 11.1. The versatilityof ELISAs is demonstrated by the six distinct ELISA protocols presented in UNIT 11.2.These provide general methods for the detection of specific antibodies, soluble antigens,or cell-surface antigens. Protocols for determining the isotype (i.e., serological class) ofantibodies are described in UNIT 11.3.

The pioneering studies of Kohler and Milstein (1975) enable investigators to obtain mil-ligram quantities of specific monoclonal antibodies after immunizing mice with relativelyimpure antigen. The spleen is removed from a previously immunized mouse that has asufficient antibody titer. After separation into individual cells, B cells from the spleenare fused with myeloma cells of B cell origin to produce immortal antibody-secretinghybridoma cells of predetermined specificity. Each hybridoma cell is capable of pro-ducing an unlimited supply of a single, antigen-specific monoclonal antibody. SectionII describes the preparation of these antigen-specific monoclonal antibodies in separateprotocols that cover immunization of mice (UNIT 11.4), cell preparation and cell fusion forgenerating hybridoma cell lines (UNITS 11.5-11.7), cloning by limiting dilution to ensure theproduction of truly monoclonal antibodies derived from a single antibody-secreting cell(UNIT 11.8), freezing and recovery of hybridoma cell lines (UNIT 11.9), production of cellculture supernatants of monoclonal antibodies in ascites fluid (UNIT 11.10), and purifica-tion of these monoclonal antibodies by affinity chromatography (UNIT 11.11). Detection ofantibody in serum, hybridoma supernatants (micrograms per milliliter), and ascites fluid(milligrams per milliliter) by ELISA is described in UNIT 11.2.

Although monoclonal antibodies can be made available in unlimited quantities andwithout the need to purify the antigen to homogeneity, the reliance upon only monoclonalantibodies for detection and identification of antigen and cDNA clones can produceequivocal results. Because monoclonal antibodies may be specific for short peptidesequences, there is a possibility of obtaining false positives, since unrelated proteins canshare small regions of homology. One way in which this uncertainty can be minimizedis to utilize several different monoclonal antibodies specific for different sites on theantigen. Another disadvantage of using a monoclonal antibody is that it may have arelatively low affinity for a given antigenic site.

Immunology

11.0.3

Current Protocols in Molecular Biology Supplement 72

These problems caused by the use of monoclonal antibodies may be circumvented bygenerating polyclonal antibodies, which consist essentially of numerous monoclonalantibodies with different epitope specificities (Section III). When a purified antigen isavailable in sufficient amount for immunization, it is possible to obtain specific poly-clonal antibodies with high affinity after repeated immunizations (UNIT 11.12; Klinman andPress, 1975). Choice of animal is determined by the amount of antiserum required forsubsequent experiments. Although animals such as goats, sheep, or horses can providelarger volumes of antiserum, few institutions have adequate facilities for their care andmaintenance. Mice, rats, and guinea pigs, on the other hand, may not yield sufficientvolumes of antiserum. For these reasons, rabbits have become the animal of choice forthe generation of polyclonal antibodies. UNIT 11.12 describes the proper preparation ofantigen as well as various routes of immunization in rabbits to optimize the antibodyresponse. Although a schedule for immunization and boosting is provided, this procedureis only a recommendation of what has worked for the author; optimal conditions shouldbe determined empirically. UNIT 11.13 describes systems for in vitro antibody production,and subsequent measurement of secreted antibodies. UNIT 11.14 discusses the purificationfrom serum, ascites fluid, or hybridoma supernatant of the immunoglobulin G fraction,which becomes the predominant antibody class after the booster injection.

If purified antigen is in limited supply, polyclonal (as well as monoclonal) antibodies canstill be raised by immunization with synthetic peptides whose sequences are based onthat of the protein, which it is designed to mimic (Section IV). In this case, the selectionof an immunogenic peptide is vital for obtaining a good antibody response. UNIT 11.15

discusses the necessary parameters to consider in the selection of a particular peptidesequence that will elicit an antibody that recognizes the native form of the protein. Toenhance the immunogenicity of the peptide, it can be chemically cross-linked to a carriermolecule (UNIT 11.16). Such cross-linking of the peptide has been demonstrated to behelpful in generating an antibody response to peptides that might not otherwise elicitantibody production.

The quantitation of specific antibody (as well as its isotypes) in polyclonal antisera, ascitesfluid, or hybridoma supernatant by solid-phase radioimmunoassay (RIA) is describedin UNIT 11.17. Although this method is more laborious than the nonradioactive ELISAdescribed in UNIT 11.2, the disadvantages are offset by greater sensitivity and betterreproducibility from assay to assay. A solution-phase RIA is also presented in UNIT 10.24.This method is also a sensitive assay for quantitation of protein in unknown samples.Because it requires only one antibody against the antigen of interest, it is especiallyuseful for short peptides, which may have only a single antigenic site.

LITERATURE CITEDBurnet, F.M. 1959. The Clonal Selection Theory of Acquired Immunity. Vanderbilt University Press,

Nashville.

Klinman, N.R. and Press, J. 1975. The B Cell specificity repertoire: Its relationship to definable subpopula-tions. Transplant. Rev. 24:41-83.

Kohler, G. and Milstein, C. 1975. Continuous cultures of fused cells secreting antibody of predefinedspecificity. Nature (Lond.) 256:495-497.

Tonegawa, S. 1985. The molecules of the immune system. Sci. Am. 253:122-131.

John A. Smith

SECTION I IMMUNOASSAYSAntigen can be detected and quantitated using an enzyme-linked immunosorbent assay(ELISA). The direct and sandwich ELISAs discussed in UNIT 11.2 are particularly usefulfor determining the presence or amount of antigen in samples ranging from crude bacteriallysates to highly purified protein antigen preparations. These assays require the prepara-tion of enzyme-antibody conjugates. How these conjugates are prepared and whatenzymes should be linked to the antigen-specific antibody in the conjugate are presentedin UNIT 11.1.

UNIT 11.1 Conjugation of Enzymes to AntibodiesConjugation of enzymes to antibodies involves the formation of a stable, covalentlinkage between an enzyme [e.g., horseradish peroxidase (HRPO), urease, or alkalinephosphatase] and an antigen-specific monoclonal or polyclonal antibody in whichneither the antigen-combining site of the antibody nor the active site of the enzyme isfunctionally altered. The chemistry of cross-linking HRPO or urease to immunoaffinity-purified monoclonal or polyclonal antibodies (IgG) is presented in Figures 11.1.1 and

NO2 NO2

NO2 NO2NH NH

Na IO4 HRPOHRPOCH2OH

HO OHOH OH

OO

CH2OH

NO2NO2

NO2 NO2Na BH4

NHNH

lgG

NH2

HRPOHRPO CH2OH CH2OH

NONHHO

lgGlgG

Schiff baseformation

stableconjugate

activatedperoxidase

O

OO

O

Figure 11.1.1 Conjugation of horseradish peroxidase (HRPO) to antibody (IgG) using the perio-date oxidation method. The method involves three chemical steps: (1) sodium periodate (NaIO4)oxidation of the carbohydrate side chains of HRPO, (2) Schiff base formation between activatedperoxidase and amino groups of the antibody, and (3) sodium borohydride (NaBH4) reduction ofthe Schiff base to form a stable conjugate.

Contributed by Scott E. Winston, Steven A. Fuller, Michael J. Evelegh, and John G.R. HurrellCurrent Protocols in Molecular Biology (2000) 11.1.1-11.1.7Copyright © 2000 by John Wiley & Sons, Inc.Supplement 50

11.1.1

Coagulation ofEnzymes toAntibodies

11.1.2, respectively. The chemistry of cross-linking alkaline phosphatase to antibodies ispresented in Figure 11.16.2.

BASICPROTOCOL

CONJUGATION OF HORSERADISH PEROXIDASE TO ANTIBODIESHorseradish peroxidase–antibody conjugates (Tijssen and Kurstak, 1984) can be used inELISA (enzyme-linked immunosorbent assay; UNIT 11.2) and western blotting (UNIT 10.8).

Materials

1 mg/ml antibody solution (affinity-purified polyclonal or monoclonal antibodies;UNIT 11.11)

0.1 M phosphate buffer, pH 6.8Horseradish peroxidase (HRPO; Sigma Type VI #P8375)0.1 M carbonate buffer, pH 9.2Sodium periodate (NaIO4) solution, freshly preparedSodium borohydride (NaBH4) solution, freshly preparedSaturated ammonium sulfate [(NH4)2SO4] solutionTris/EDTA/NaCl (TEN) buffer, pH 7.2Bovine serum albumin (BSA)Glycerol

Dialysis membrane (see reagents and solutions and APPENDIX 3)Pasteur pipet fitted with glass woolSephadex G-25, medium (size of gel matrix)

1. Dialyze 1 mg/ml antibody solution against 2 liters of 0.1 M phosphate buffer, pH 6.8,overnight at 4°C, stirring gently.

lgG

lgG

lgG

urease

urease

O

O

C

N

S

N

C

H

H

N OH

N

H

N

ON

NMBS

NH2

SH

C

O

O

OO

O

O

O

O

O

Figure 11.1.2 Conjugation of urease to antibody (IgG) with m-maleimidobenzoyl N-hydroxysuc-cinimide ester (MBS). The first step involves benzoylation of the amino groups of the antibody (IgG)oxidate oxidation method. The second step involves the thiolation of the maleimide moiety by thesulfhydryl groups of the urease enzyme.


11.1.2

Immunology

Antibody may be polyclonal or monoclonal, purified as described in UNIT 11.11.

A280/1.44 = mg IgG/ml. Concentration of antibody should be at least 1 mg/ml.

Soak dialysis membrane 1 hr in 50% (v/v) aqueous ethanol, 1 hr in 10 mM NaHCO3, and1 hr in 1 mM EDTA. Then rinse twice in distilled water and store at 4°C in phosphate buffercontaining 0.01% (w/v) sodium azide.

2. Dissolve 10 mg HRPO in 1 ml 0.1 M carbonate buffer, pH 9.2.

3. Mix 0.25 ml freshly prepared NaIO4 solution with 0.25 ml of 10 mg/ml HRPO/carbonate mixture from step 2, cap tightly, and incubate at room temperature for 2 hrin the dark (NaIO4 is light sensitive).

4. Into a Pasteur pipet fitted with glass wool and blocked at the tip with Parafilm, add1 ml of the dialyzed 1 mg/ml antibody solution from step 1 to 0.5 ml of the 10 mg/mlHRPO solution from step 3. Add 0.25 g Sephadex G-25 to the antibody/HRPOmixture.

Addition of Sephadex increases the concentration of antibody and HRPO by absorbingwater. This enhances the conjugation of enzyme to antibody.

5. Incubate 3 hr at room temperature in the dark.

6. Wash column with 0.75 ml carbonate buffer to elute conjugate.

7. Add 38 µl freshly prepared NaBH4 solution to the eluate and incubate 30 min at roomtemperature in the dark.

8. Add 112 µl freshly prepared NaBH4 and incubate 60 min in the dark.

9. Add 0.9 ml of saturated (NH4)2SO4 solution and stir gently for 30 min at 4°C.Centrifuge 15 min at 10,000 × g, 4°C.

10. Decant, discard supernatant, and resuspend pellet in 0.75 ml TEN buffer.

11. Dialyze resuspended pellet overnight at 4°C against 2 liters TEN buffer. Change TENsolution in the morning and continue dialyzing for 4 hr.

12. Remove conjugate from dialysis membrane and add sufficient BSA to bring theconjugate solution to a final concentration of 20 mg BSA/ml.

13. Add an equal volume of glycerol and store at −20°C.

ALTERNATEPROTOCOL

CONJUGATION OF UREASE TO ANTIBODIES

Urease conjugates can be used in ELISA (UNIT 11.2) but not western blotting (UNIT 10.8).

Additional Materials

20 mg/ml urease (Sigma Type VII #U0376; source is important) in0.1 M phosphate buffer

m-Maleimidobenzoyl N-hydroxysuccinimide ester in dimethylformamide(MBS/DMF solution)

0.143 M 2-mercaptoethanol (prepare from 14.3 M stock)Phosphate-buffered saline (PBS; APPENDIX 2)

12 × 75–mm glass tubes1.5 × 5–cm PD-10 column (Pharmacia)Nitrogen tank


11.1.3


1. Repeat step 1 of the basic protocol.

2. Dialyze 0.25 ml of 20 mg/ml urease in 0.1 M phosphate buffer against 2 liters 0.1 Mphosphate buffer overnight at 4°C while stirring gently.

3. In the morning replace the phosphate buffer and continue to dialyze for 2 hr.

4. Remove antibody and urease solutions from dialysis membranes and place in separateglass tubes. Read A280 of antibody solution and dilute with phosphate buffer to 0.5mg/ml.

5. Add 0.075 ml MBS/DMF solution to 1.5 ml of 0.5 mg/ml dialyzed antibody solutionin a glass tube (MBS/antibody molar ratio, 120:1). Place a magnetic stir bar in thetube and stir gently at room temperature for 30 min.

6. Load on a PD-10 column (see support protocol, UNIT 10.9) preequilibrated with 100ml phosphate buffer. Run column with phosphate buffer and collect 0.6-ml fractions.Read A280 of fractions and collect first peak that elutes.

The first peak contains activated antibody and the second peak contains free MBS.

7. Pool first peak (generally about 2.5 to 3.0 ml) and add 0.15 ml of 20 mg/ml ureasein 0.1 M phosphatase buffer (urease/antibody weight ratio, 4:1).

8. Stir at room temperature under N2 for 1.5 hr or until solution appears cloudy.

9. Add 0.143 M 2-mercaptoethanol to a final concentration of 2 mM (0.014 mlmercaptoethanol solution/ml urease-antibody conjugate) and stir at room tempera-ture for 30 min.

10. Dialyze overnight at 4°C against 2 liters PBS. In the morning, replace PBS anddialyze 4 hr.

11. Add an equal volume of glycerol, divide into small aliquots, and store at −20°C (stablefor 1 year).

ALTERNATEPROTOCOL

CONJUGATION OF ALKALINE PHOSPHATASE TO ANTIBODIES

Alkaline phosphatase conjugates can be used in ELISA (UNIT 11.2) and western blotting(UNIT 10.8).


5 mg/ml antibody solution (affinity-purified polyclonal or monoclonal antibodies, UNIT 11.11)

10 mg/ml alkaline phosphatase (enzyme immunoassay grade; BoehringerMannheim; source is important)

25% glutaraldehyde in H2OTris/ovalbumin solutionSodium azide

1. Dialyze 5 mg/ml antibody solution as in step 1 of the basic protocol, except dialyzeagainst PBS.

2. Remove antibody solution from dialysis membrane and place in a tube. Read A280

and dilute with PBS to 3 mg/ml.

3. Add 100 µl dialyzed antibody solution to 90 µl of 10 mg/ml alkaline phosphatase ina 1.5-ml microcentrifuge tube.


11.1.4

Immunology

4. Add 5 ml 25% glutaraldehyde and mix gently. Let stand at room temperature.

5. Remove 25-µl samples at time 0, 5, 10, 15, 30, 60, and 120 min and place in separate1.5-ml microcentrifuge tubes. Add 125 µl PBS to each sample, then add 1.1 mlTris/ovalbumin solution. Store each sample on ice until the time course is completed.

6. Dialyze the samples against PBS as described in step 1 of this protocol. Test eachsample for alkaline phosphatase activity using a direct ELISA assay (UNIT 11.2) todetermine which conjugation time yields the most active enzyme conjugate.

7. Repeat steps 1 to 4, but in step 4 allow the reaction to proceed for the optimalconjugation time, as determined in step 6.

8. Add sodium azide to 0.1% and store the conjugate protected from light at 4°C for upto 1 year. Alternatively, add an equal volume of glycerol and store the conjugates at−20°C for 1 year.

REAGENTS AND SOLUTIONS

0.1 M carbonate buffer, pH 9.21.36 g sodium carbonate7.35 g sodium bicarbonate950 ml H2OAdjust pH to 9.2 with 1 M HCl or 1 M NaOH, if necessaryAdd H2O to 1 liter

m-Maleimidobenzoyl N-hydroxysuccinimide ester in dimethylformamide(MBS/DMF solution)

Prepare 0.25% (w/v) solution by adding 2.3 mg MBS (Pierce #22310) to 0.92 mlDMF. Use within 1 hr of preparation.

MBS will deteriorate upon prolonged storage if repeatedly thawed, opened, and refrozen.Dispense into 10-mg aliquots and store desiccated at −20°C.

0.1 M phosphate buffer, pH 6.8Stock A, 0.2 M: 31.2 g NaH2PO4 in 1 liter H20Stock B, 0.2 M: 28.39 g Na2HPO4 in 1 liter H20Mix 51 ml Stock A, 49 ml Stock B, and 100 ml H20

Saturated ammonium sulfate [(NH4)2SO4] solution

Prepare 0.01 M Tris solution by adding 1.21 g Tris base to 990 ml water, adjust topH 7.0, and bring to a final volume of 1 liter. Weigh 767 g (NH4)2SO4 and dissolvein 1 liter 0.01 M Tris by stirring and gently warming. Adjust the pH to 7.0 and storeat 4°C. (NH4)2SO4 crystals should be seen at the bottom of the solution at 4°C.

Sodium borohydride (NaBH4) solution5 mg NaBH4/ml 0.1 mM NaOH (use immediately)

Sodium periodate (NaIO4) solution1.7l mg NaIO4/ml H20 (use immediately)

Tris/EDTA/NaCl (TEN) buffer, pH 7.2To 930 ml H2O, add:6.06 g Tris base0.37 g Na2EDTA8.77 g NaClAdjust pH to 7.2 with HClAdd H2O to 1 liter


11.1.5


Tris/ovalbumin buffer0.05 M Tris⋅Cl, pH 8.05% ovalbumin5 mM MgCl2

0.5% NaN3

0.5% mertiolate

COMMENTARY

Background InformationDirect conjugation of enzymes to antibodies

has greatly simplified the development andperformance of many different types of immu-noassays. The conjugation of HRPO (Nakaneand Kawaoi, 1974) to antibody is dependent onthe generation of aldehyde groups by periodateoxidation of the carbohydrate moieties onHRPO. Combination of these active aldehydeswith amino groups on the antibody forms Schiffbases that, upon reduction by sodium borohy-dride, become stable. Horseradish peroxidase(HRPO) conjugates are useful in all types ofimmunological assays, but are generally lessstable than urease conjugates. In addition, en-dogenous peroxidases may cause false positivereactions. For urease conjugation (Healey et al.,1983), cross-linking the enzyme and antibodywith MBS is achieved through benzoylation offree amino groups on antibody. This is followedby thiolation of the maleimide moiety of MBSby the cysteine sulfhydryl groups of urease. Theadvantages of urease conjugates are their sta-bility in solution at normal working dilutions,the rapid turnover rate of the enzyme, the easilydiscernible color change when substrate isadded, and the fact that urease is not found inmost mammalian or bacterial systems. The dis-advantage is that since no precipitable substrateis available, urease conjugates cannot be usedfor immunohistology or western blotting.

Alkaline phosphatase conjugates are usefulfor all types of immunological assays depend-ing on the alkaline phosphatase substrate used(i.e., p-nitrophenyl phosphate in diethanol-amine is the preferred substrate for ELISA withcolorimetric detection, 4-methylumbelliferylphosphate is useful for ELISA with fluorimet-ric detection, and nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate is the pre-ferred substrate for western blotting). Alkalinephosphatase conjugates are as stable as ureaseconjugates and more stable than HRPO conju-gates. Endogenous phosphatases can causefalse positive reactions. However, levamisolewill inhibit alkaline phosphatase in many mam-malian tissues but not the alkaline phosphatase(i.e., bovine intestinal) used in the conjugates,

and for this reason levamisole may be added tothe substrate solution.

The one-step glutaraldehyde method (Volleret al., 1976) is the most simple available proce-dure for preparing alkaline phosphatase–anti-body conjugates. Various alternative proce-dures for preparing alkaline phosphatase con-jugates have been compared (Jeanson et al.,1988).

The sensitivity that can be achieved witheither HRPO, urease, or alkaline phosphataseconjugates is comparable and between 1 ng/mland 10 ng/ml of antigen can be detected.

Critical ParametersThe most critical parameters of both conju-

gation methods are the quality of enzyme andthe cross-linking reagents. Several lots of thesereagents should be tested as described in theprotocol before conjugating to larger quantitiesof antibodies. It is imperative that the m-male-imidobenzoyl N-hydroxysuccinimide ester(MBS), sodium periodate (NaIO4), and sodiumborohydride (NaBH4) be stored in a desicca-tor and that solutions containing these chemi-cals be prepared immediately prior to use.The method described is applicable to mostantibodies and should produce conjugatesthat are useful for developing an ELISA fordetecting sensitively and specifically a givenantigen. However, not all antibodies conjugatein an identical manner. It may be necessary tovary the ratio of MBS/antibody or urease/anti-body for the urease conjugation and theNaIO4/HRPO and HRPO/antibody ratios for agiven HRPO conjugation.

The quality and grade of alkaline phos-phatase is crucial to the generation of effectiveconjugates. Immunoassay grade material is rec-ommended over lower grades, and the enzymeshould not be conjugated beyond its expirationdate. In the case of polyclonal antisera, thespecificity and titer of the antiserum will bereflected in the conjugate and any purificationprocedures that increase these values, such asimmunoaffinity chromatography (UNIT 10.11)will enhance conjugate performance.

The selection of an optimal conjugation time


11.1.6

Immunology

for preparing alkaline phosphatase–antibodyconjugates varies for different antibodies, inparticular when monoclonal antibodies areused. In contrast, polyclonal antibodies may bereliably conjugated in 120 min.

TroubleshootingThere are several factors that may contribute

to the production of poor enzyme-antibodyconjugates. It is important to determine firstwhether a poor conjugate is the result of inac-tivation of either the antibody or the enzyme(or both) or the result of insufficient or exces-sive cross-linking. The affinity of the antibodyfor substrate can be measured by determiningthe presence of bound antibody with anotherimmunoassay employing anti-antibody conju-gated to a different enzyme. Enzyme activitycan be measured by cleavage of substrate atdifferent enzyme concentrations. Precipitationof material in the conjugate solution or opaquesolutions are indicative of excessive cross-link-ing. Sodium dodecyl sulfate–polyacrylamidegel electrophoresis (UNIT 10.2) is useful for moni-toring the extent of cross-linking by determin-ing the Mr of the cross-linked species.

Insufficient cross-linking usually resultsfrom the use of inactive or poor-quality cross-linking agents. Try using fresh reagents or dif-ferent lots of reagent. Excessive cross-linkingand inactivation of antibody or enzyme can beeliminated by either reducing the concentrationof antibody and enzyme or by reducing the timeof reaction.

It may not be possible to generate effectivealkaline phosphatase conjugates with all anti-bodies using the one-step glutaraldehydemethod. An alternative is to try a differentconjugation technique (see Jeanson et al.,1988). Another alternative is to use an anti-spe-cies antibody– alkaline phosphatase conjugateto detect the antibody in question. These re-agents may be purchased or prepared using theabove technique.

Anticipated ResultsMost monoclonal antibodies will couple to

alkaline phosphatase very quickly, and a 5-minconjugation time will often be optimal. In gen-eral, polyclonal antibodies will take longer toconjugate, usually between 1 and 2 hr. The yieldand titer of the resultant conjugate will dependon the original antibody’s properties and spe-

cific application.It is difficult to estimate the yield or working

dilution of the conjugates, as it is dependent onnumerous factors such as antibody affinity, typeof ELISA, and quality of antigen. In general,the working dilutions range from 1:100 to1:10,000.

Time ConsiderationsThe total time for conjugation is 1 to 3 days,

with working times of several hours per conju-gation. The majority of time is spent dialyzingor stirring. Once started, a protocol should becompleted as described because the reactantproducts and solutions are unstable.

Literature CitedHealey, K., Chandler, H.M., Cox, J.C., and Hurrell,

J.G.R. 1983. A rapid semi-quantitative capillaryenzyme immunoassay for digoxin. Clin. Chim.Acta 134:51-58.

Jeanson, A., Cloes, J.-M., Bouchet, M., and Rentier,B. 1988. Comparison of conjugation proceduresfor the preparation of monoclonal antibody-en-zyme conjugates. J. Immunol. Methods 111:261-270.

Nakane, P.K. and Kawaoi, A. 1974. Peroxidase-la-beled antibody. A new method of conjugation. J.Histochem. Cytochem. 22:1084-1091.

Tijssen, P. and Kurstak, E. 1984. Highly efficient andsimple methods for the preparation of peroxidaseand active peroxidase-antibody conjugates forenzyme immunoassay. Anal. Biochem. 136:451-457.

Voller, A., Bidwell, D.E., and Barlett, A. Enzymeimmunoassays in diagnostic medicine. Bull.W.H.O. 53:55-65.

Key ReferenceVan Vunakis, H. and Langone, J.J., eds. 1980. Im-

munochemical techniques. Methods Enzymol.70:1-525.

An excellent collection of articles on immunoassaytechniques, including several on enzyme-antibodyconjugation techniques.

Contributed by Scott E. Winston, Steven A. Fuller, and Michael J. EveleghADI DiagnosticsRexdale, Ontario

John G.R. HurrellBoehringer MannheimIndianapolis, Indiana


11.1.7


UNIT 11.2Enzyme-Linked Immunosorbent Assays(ELISA)This unit describes six different ELISA systems for the detection of specific antibodies,soluble antigens, or cell-surface antigens. In all six systems, soluble reactants are removedfrom solution after specifically binding to solid-phase reactants. Table 11.2.1 summarizesthe different ELISA protocols, which are illustrated in Figures 11.2.1-11.2.6.

In the first four protocols, solid-phase reactants are prepared by adsorbing an antigen orantibody onto plastic microtiter plates; in the next two protocols, the solid-phasereactants are cell-associated molecules. In all protocols, the solid-phase reagents areincubated with secondary or tertiary reactants covalently coupled to an enzyme.

Table 11.2.1 Summary of ELISA Protocols

ELISA protocol Uses Required reagents Comments

Indirect Antibody screening;epitope mapping

Antigen, pure or semipure: test solutioncontaining antibody;enzyme conjugate thatbinds Ig of immunizedspecies

Does not require theuse of preexistingspecific antibodies;requires relativelylarge amounts ofantigen

Direct competitive Antigen screening;detect soluble antigen

Antigen, pure or semipure; test solutioncontaining antigen;enzyme-antibody conjugate specific forantigen

Rapid assay withonly two steps;excellent formeasuring antigeniccross-reactivity

Antibody-sandwich Antigen screening;detect soluble antigen

Capture antibody (purified or semi-purifiedspecific antibody); test solution containing antigen; enzyme-antibody conjugate specific for antigen

Most sensitiveantigen assay;requires relativelylarge amounts ofpure or semi-purespecific antibody(capture antibody)

Double antibody–sandwich

Antibody-screening;epitope mapping

Capture antibody:(specific for Ig ofimmunized species);test solution containingantigen; enzyme-antibody conjugate specific for antigen

Does not requirepurified antigen;relatively long assaywith five steps

Direct cellular Screen cells forexpression of antigen; measurecellular antigenexpression

Cells that express antigen of interest;enzyme-antibody conjugate specific for cellular antigen

Sensitive assay forbulk screening;insensitive toheterogeneity ofexpression in mixedpopulation of cells

Indirect cellular Screen for antibodies againstcellular antigens

Cells used for immunizing; test solution containingantibodies; enzymeconjugate that binds Ig of immunized species

May not detectantibodies specificfor cellular antigensexpressed at a lowdensity

Supplement 15

Contributed by Peter Hornbeck, Scott E. Winston, and Steven A. FullerCurrent Protocols in Molecular Biology (1991) 11.2.1-11.2.22Copyright © 2000 by John Wiley & Sons, Inc.

11.2.1

Immunology

Unbound conjugates are washed out and a chromogenic or fluorogenic substrate is added.As the substrate is hydrolyzed by the bound enzyme conjugate, a colored or fluorescentproduct is generated. Finally, the product is detected visually or with a microtiter platereader. The amount of product generated is proportional to the amount of analysate in thetest mixture. The first support protocol can be used to optimize the different ELISAs. Thesecond support protocol provides a method for preparing lysates for use as test antigenfrom bacterial cultures containing expressed protein.

BASICPROTOCOL

INDIRECT ELISA TO DETECT SPECIFIC ANTIBODIES

This assay is useful for screening antisera or hybridoma supernatants for specificantibodies when milligram quantities of purified or semipurified antigen are available (1mg of purified antigen will permit screening of 80 to 800 microtiter plates; Fig. 11.2.1).Antibodies are detected by coating the wells of microtiter plates with antigen, incubatingthe coated plates with test solutions containing specific antibodies, and washing awayunbound antibodies. A solution containing a developing reagent, (e.g., alkaline phos-phatase conjugated to protein A, protein G, or antibodies against the test solutionantibodies) is then added to the plate. After incubation, unbound conjugate is washedaway and substrate solution is added. After a second incubation, the amount of substratehydrolyzed is assessed with a spectrophotometer or spectrofluorometer. The measuredamount is proportional to the amount of specific antibody in the test solution. Visualinspection can also be used to detect hydrolysis.

Materials

Developing reagent: protein A–alkaline phosphatase conjugate (Sigma #P9650),protein G–alkaline phosphatase conjugate (Calbiochem #539304), oranti-Ig-alkaline phosphatase conjugate (UNIT 11.1)

Antigen solutionPBS (APPENDIX 2) containing 0.05% NaN3 (PBSN)Water, deionized or distilledBlocking bufferTest antibody samples4-methylumbelliferyl phosphate (MUP) or p-nitrophenyl phosphate (NPP)

substrate solution0.5 M NaOH (optional)

Multichannel pipet and disposable pipet tipsImmulon 2 (Dynatech #011-010-3450), Immulon 4 (Dynatech #011-010-3850), or

equivalent microtiter platesPlastic squirt bottlesMicrotiter plate reader (optional)—spectrophotometer with 405-nm filter or

spectrofluorometer (Dynatech #011-970-1900) with 365-nm excitation filterand 450-nm emission filter

Determine developing reagent and antigen concentrations1. Determine the optimal concentration of the developing reagent (conjugate) by

criss-cross serial dilution analysis (see first support protocol).

Good conjugates of many specificities are available commercially. Choice of developingreagent (i.e., conjugate) is determined by the goals of the assay. If it is necessary to detectall antibodies that bind to antigen, conjugates prepared with antibodies specific for Ig κand λ light chains should be used. Alternatively, protein A– or protein G–enzyme conjugatesmay be preferable when screening monoclonal antibodies. Specific monoclonal antibodiesthat bind protein A or protein G are easy to purify and characterize.


11.2.2

Enzyme-LinkedImmunosorbent

Assays

2. Determine the final concentration of antigen coating reagent by criss-cross serialdilution analysis (see first support protocol). Prepare an antigen solution in PBSN atthis final concentration. The final concentration of antigen is usually 0.2 to 10.0µg/ml. Prepare ∼6 ml antigen solution for each plate.

Pure antigen solution concentrations are usually ≤2 �g/ml. Although pure antigen prepa-rations are not essential, >3% of the protein in the antigen solution should be the antigen.The total concentration of protein in the antigen solution should be increased for semipu-rified antigen preparations. Do not raise the total protein concentration in the antigensolution to >10 �g/ml, since this concentration usually saturates >85% of the availablesites on Immulon microtiter plates. For some antigens, coating may occur more efficientlyat different pHs.

Coat plate with antigen3. Using a multichannel pipet and tips, dispense 50 µl antigen solution into each well

of an Immulon microtiter plate. Tap or shake the plate to ensure that the antigensolution is evenly distributed over the bottom of each well.

Ag Ag

Ag Ag

Ab Ab

Ag Ag

Ab Ab

E

Ab

E

Ab

Ag Ag

Ab Ab

E

Ab

E

Ab

Ab = detected Ab

coat well with anitgen

incubate with antibody

incubate with antibody-enzyme conjugate

add substrate and observecolor change or fluorescence

wash

wash

block

Figure 11.2.1 Indirect ELISA to detect specific antibodies. Ag = antigen; Ab = antibody; E =enzyme.


11.2.3

Immunology

4. Wrap coated plates in plastic wrap to seal and incubate overnight at room temperatureor 2 hr at 37°C.

Individual adhesive plate sealers are sold commercially but plastic wrap is easier to useand works as well. Sealed plates can be stored at 4°C with antigen solution for months.

5. Rinse coated plate over a sink by filling wells with deionized or distilled waterdispensed either from a plastic squirt bottle or from the tap. Flick the water into thesink and rinse with water two more times, flicking the water into the sink after eachrinse.

Block residual binding capacity of plate6. Fill each well with blocking buffer dispensed as a stream from a squirt bottle and

incubate 30 min at room temperature.

Residual binding capacity of the plate is blocked in this step. Tween 20 (0.05%) by itself ismore effective at blocking than any protein tested, but because the combination of proteinand Tween 20 may be more effective than Tween 20 alone in some cases, bovine serumalbumin (BSA; 0.25%) is included in the blocking buffer.

7. Rinse plate three times in water as in step 5. After the last rinse, remove residual liquidby wrapping each plate in a large paper tissue and gently flicking it face down ontoseveral paper towels laying on the benchtop.

Rinsing with water is cheaper and easier than rinsing with buffered solutions and is aseffective.

Add antibody to plate8. Add 50 µl antibody samples diluted in blocking buffer to each of the coated wells,

wrap plate in plastic wrap, and incubate ≥2 hr at room temperature.

While enough antibody may be bound after 1 to 2 hr to generate a strong signal, equilibriumbinding is generally achieved after 5 to 10 hr. Thus, the specific signal may be significantlyincreased by longer incubations.

For this and all steps involving the delivery of aliquots of many different solutions tomicrotiter plates with multichannel pipets, such as the primary screening of hybridomasupernatants, the same pipet tips can be reused for hundreds of separate transfers. Washtips between transfers by expelling any liquid remaining in the tips onto an absorbentsurface of paper tissues, rinsing tips five times in blocking buffer, and carefully expellingany residual liquid from tips onto the tissues. Avoid bubbles in the tips; any tip withintractable bubbles should be replaced.

Wash the plate9. Rinse plate three times in water as in step 5.

10. Fill each well with blocking buffer, vortex, and incubate 10 min at room temperature.

Plates are vortexed to remove any reagent remaining in the corners of the wells.

11. Rinse three times in water as in step 5. After the final rinse, remove residual liquidas in step 7.

Add developing reagent to plate12. Add 50 µl developing reagent in blocking buffer (at optimal concentration determined

in step 1) to each well, wrap in plastic wrap, and incubate ≥2 hr at room temperature.

The strength of the signal may be increased by longer incubations (see annotation to step 8).


11.2.4


Assays

13. Wash plates as in steps 9 to 11.

After final rinsing, plates may be wrapped in plastic wrap and stored for months at 4°Cprior to adding substrate.

Add substrate and measure hydrolysis14. Add 75 µl MUP or NPP substrate solution to each well and incubate 1 hr at room

temperature.

15. Monitor hydrolysis qualitatively by visual inspection or quantitatively with a mi-crotiter plate reader (see below). Hydrolysis can be stopped by adding 25 µl of 0.5M NaOH.

a. Visually, hydrolysis of NPP can be detected by the appearance of a yellow color.If using a microtiter plate reader to measure NPP hydrolysis, use a 405-nm filter.

b. Visually, hydrolysis of MUP can be monitored in a darkened room by illuminationwith a long-wavelength UV lamp. If using a microtiter plate spectrofluorometerto measure MUP hydrolysis, use a 365-nm excitation filter and a 450-nm emissionfilter.The fluorogenic system using the MUP substrate is 10 to 100 times faster than thechromogenic system using NPP. Furthermore, the rate of spontaneous hydrolysis of MUPis much lower than that of NPP.

To detect bound antibodies that are present at low concentration, measure hydrolysis at alater time. To calculate when to measure hydrolysis the second time, remember that theamount of hydrolysis is almost directly proportional to the time of hydrolysis. For example,if the hydrolysis in the wells of interest reads 200 at 1 hr and a reading of 2000 is desired,incubate the plate ∼10 hr before taking the second reading.

ALTERNATEPROTOCOL

DIRECT COMPETITIVE ELISA TO DETECT SOLUBLE ANTIGENS

This assay is used to detect or quantitate soluble antigens and is most useful when botha specific antibody and milligram quantities of purified or semipurified antigen areavailable (Fig. 11.2.2). To detect soluble antigens, plates are coated with antigen and thebinding of specific antibody-enzyme conjugates to antigen-coated plates is inhibited bytest solutions containing soluble antigen. After incubation with mixtures of the conjugateand inhibitor in antigen-coated wells, unbound conjugate is washed away and substrateis added. The amount of antigen in the test solutions is proportional to the inhibition ofsubstrate hydrolysis and can be quantitated by interpolation onto an inhibition curvegenerated with serial dilutions of a standard antigen solution.

The direct assay may also be adapted as an indirect assay by substituting specific antibodyfor specific antibody-enzyme conjugate. The amount of specific antibody bound is thendetected using a species-specific or isotype-specific conjugate as a tertiary reactant.


Specific antibody–alkaline phosphatase conjugate (UNIT 11.1)Standard antigen solutionTest antigen solutionsRound- or cone-bottom microtiter plates

1. Determine the optimal concentration of coating reagent and antibody–alkaline phos-phatase conjugate by criss-cross serial dilution analysis in which the concentrationsof both the antigen (coating reagent) and the conjugate (developing reagent) arevaried (see first support protocol). Prepare a 2× conjugate solution by diluting the


11.2.5

Immunology

specific antibody–alkaline phosphatase conjugate in blocking buffer to twice theoptimal concentration.

The final concentration is usually 25 to 500 ng antibody/ml. Prepare 3 ml antibody–alkalinephosphatase conjugate for each plate.

2. Coat and block wells of an Immulon microtiter plate with 50 µl antigen solution asin steps 2 to 7 of the basic protocol.

3. Prepare six 1:3 serial dilutions of standard antigen solution in blocking buffer (seefirst support protocol for preparation of serial dilutions)—these antigen concentra-tions will be used in preparing a standard inhibition curve (see step 10).

Antigen concentrations should span the dynamic range of inhibition. The dynamic rangeof inhibition is defined as that range of inhibitor concentrations wherein changes ofinhibitor concentration produce detectable changes in the amount of inhibition. Thedynamic range of inhibition is empirically determined in an initial assay in which antigenconcentration is typically varied from the micromolar (10−6 M) to the picomolar (10−12 M)range. For most protein antigens, initial concentration should be ∼10 �g/ml, followed bynine 1:4 serial dilutions in blocking buffer. These antigen dilutions are assayed for theirability to inhibit the binding of conjugate to antigen-coated plates under standard assayconditions. From this initial assay, six 1:3 antigen dilutions spanning the dynamic rangeof inhibition are selected for further use as standard antigen-inhibitor dilutions. Prepare≥75 �l of each dilution for each plate to be assayed.

Inhibitor curves are most sensitive in the region of the curve where small changes ininhibitor concentrations produce maximal changes in the amount of inhibition. This

Ag

E

Ab

Ag

E

Ab

Ag Ag AgAg

Ag Ag

E

Ab

Ag

E

Ab

Ag

E

Ab

with inhibitor antigen without inhibitor antigen

block

wash

coat well with antigen

incubate with antibody-enzyme conjugate with orwithout inhibitor antigen

add substrate and measureinhibition of color change orfluorescence

= detected AG

Ab

Ag Ag

E

Ab

E

Ag

Ag

Figure 11.2.2 Direct competitive ELISA to detect soluble antigens. Ag = antigen; Ab = antibody;E = enzyme.


11.2.6


Assays

region of the curve normally spans 15% to 85% inhibition. In most systems, this range ofinhibition is produced by concentrations of inhibitor between 1 and 250 ng/ml.

4. Mix and incubate conjugate and inhibitor by adding 75 µl of 2× conjugate solution(from step 1) to each well of a round- or cone-bottom microtiter plate, followed by75 µl inhibitor—either test antigen solution or standard antigen solution (from step3). Mix the conjugate and inhibitor solutions by pipetting up and down in the pipettip three times (see annotation to step 8 in the basic protocol) and incubate ≥30 minat room temperature.

For accurate quantitation of the amount of antigen in the test solutions, test antigensolutions should inhibit conjugate binding between 15% to 85%. It may be necessary toassay two or three different dilutions of the test solutions to produce inhibitions within thisrange.

5. Prepare uninhibited control samples by mixing equal volumes of 2× conjugatesolution and blocking buffer.

6. Transfer 50 µl of the mixture of conjugate plus inhibitor (from step 4) or conjugateplus blocking buffer (from step 5) to an antigen-coated plate (from step 2) andincubate 2 hr at room temperature.

If samples are to be assayed in duplicate, the duplicates should be in adjacent columns onthe same plate. Reserve column 11 for uninhibited control samples (step 5) and column 12for substrate alone without any conjugate. If the concentration of antigen in the test samplesis to be accurately quantitated, dilutions of homologous antigen solutions (step 3) shouldbe included on each plate.

7. Wash plate as in steps 9 to 11 of the basic protocol.

8. Add 75 µl of MUP or NPP substrate solution to each well and incubate 1 hr at roomtemperature.

9. Read plates on the microtiter plate reader after ≥1 hr, at which time enough substratehas been hydrolyzed in the uninhibited reactions to permit accurate measurement ofthe inhibition.

10. Prepare a standard antigen-inhibition curve constructed from the inhibitions pro-duced by the dilutions of the standard antigen solutions from step 3. Plot antigenconcentration on the x axis, which is a log scale, and fluorescence or absorbance onthe y axis, which is a linear scale.

11. Interpolate the concentration of antigen in the test solutions from the standardantigen-inhibition curve.

The dynamic range of the inhibition curve may deviate from linearity if the specificantibodies are heterogeneous and possess significantly different affinities or if the standardantigen preparation contains heterogeneous forms of the antigen. Antigen concentrationin test samples can be accurately interpolated from the inhibition curve as long as the testantigen is antigenically identical to the standard antigen and the concentration of testantigen falls within the dynamic range of inhibition.


11.2.7

Immunology

ALTERNATEPROTOCOL

ANTIBODY-SANDWICH ELISA TO DETECT SOLUBLE ANTIGENS

Antibody-sandwich ELISAs may be the most useful of the immunosorbent assays fordetecting antigen because they are frequently between 2 and 5 times more sensitive thanthose in which antigen is directly bound to the solid phase (Fig. 11.2.3). To detect antigen,the wells of microtiter plates are coated with specific (capture) antibody followed byincubation with test solutions containing antigen. Unbound antigen is washed out and adifferent antigen-specific antibody conjugated to enzyme (i.e., developing reagent) isadded, followed by another incubation. Unbound conjugate is washed out and substrateis added. After another incubation, the degree of substrate hydrolysis is measured. Theamount of substrate hydrolyzed is proportional to the amount of antigen in the testsolution.


Specific antibody or immunoglobulin fraction from antiserum or ascites fluid, orhybridoma supernatant (UNIT 11.10), or bacterial lysate (second support protocol)

E E

E

Ag

Ab Ab

Ab Ab

Ab

Ag

Ab

Ab

Ag

Ab

Ag

E

Ab

Ag

Ab

Ab

Ag

Ab

wash

wash

block

coat well with antibody

incubate with antigen

incubate withantibody-enzyme conjugate


Ag = detected Ag

Figure 11.2.3 Antibody-sandwich ELISA to detect antigen. Ag = antigen; Ab = antibody; E =enzyme.


11.2.8


Assays

1. Prepare the capture antibody by diluting specific antibody or immunoglobulinfraction in PBSN to a final concentration of 0.2 to 10 µg/ml.

The capture antibodies can be monoclonal or polyclonal.

If the immunoglobulin fraction from an antiserum or ascites fluid is used, the concentrationof total protein may need to be increased to compensate for the lower content of specificantibody. Little advantage is gained by increasing the total protein concentration in thecapture antibody solution beyond 10 �g/ml.

2. Determine the concentration of capture antibody and conjugate necessary to detectthe desired concentration of antigen by criss-cross serial dilution analysis (see firstsupport protocol). Prepare a capture antibody solution in PBSN at this concentration.

3. Coat wells of an Immulon plate with capture-antibody solution as in steps 3 to 5 ofthe basic protocol.

4. Block wells as in steps 6 and 7 of the basic protocol.

5. Prepare a standard antigen-dilution series by successive 1:3 dilutions of the homolo-gous antigen stock in blocking buffer (see first support protocol).

In order to measure the amount of antigen in a test sample, the standard antigen-dilutionseries needs to span most of the dynamic range of binding. This range typically spans from0.1 to 1000 ng antigen/ml. The dynamic range of binding is defined as that range of antigenconcentrations wherein small, incremental changes in antigen concentration producedetectable differences in the amount of antigen bound (see annotation to step 3, in thepreceding alternate protocol). In most assay systems, the amount of antigen in a testsolution is most accurately interpolated from the standard curve if it produces between15% to 85% of maximal binding.

NOTE: While standard curves are necessary to accurately measure the amount of antigenin test samples, they are unnecessary for qualitative “yes/no” answers.

6. Prepare dilutions of test antigen solutions in blocking buffer.

It may be necessary to assay one or two serial dilutions of the initial antigen test solutionto ensure that at least one of the dilutions can be accurately measured. For most assaysystems, test solutions containing 1 to 100 ng/ml of antigen can be accurately measured.

7. Add 50-µl aliquots of the antigen test solutions and the standard antigen dilutions(from step 5) to the antibody-coated wells and incubate ≥2 hr at room temperature.

For accurate quantitation, samples should be run in duplicate or triplicate, and thestandard antigen-dilution series should be included on each plate (see step 5). Pipettingshould be performed rapidly to minimize differences in time of incubation between samples.


9. Add 50 µl specific antibody–alkaline phosphatase conjugate and incubate 2 hr atroom temperature.

The conjugate concentration is typically 25 to 400 ng specific antibody/ml.

When the capture antibody is specific for a single determinant, the conjugate must beprepared from antibodies which recognize different determinants that remain availableafter the antigen is bound to the plate by the capture antibody.


11. Add 75 µl of MUP or NPP substrate solution to each well and incubate 1 hr at roomtemperature.


11.2.9

Immunology

12. Read the plate on a microtiter plate reader.

To quantitate low-level reactions, the plate can be read again after several hours ofhydrolysis.

13. Prepare a standard curve constructed from the data produced by serial dilutions ofthe standard antigen (step 5). Plot antigen concentration on the x axis which is a logscale, and fluorescence or absorbance on the y axis which is a linear scale.

14. Interpolate the concentration of antigen in the test solutions from the standard curve.

ALTERNATEPROTOCOL

DOUBLE ANTIBODY–SANDWICH ELISA TO DETECTSPECIFIC ANTIBODIES

This assay is especially useful when screening for specific antibodies in cases when asmall amount of specific antibody is available and purified antigen is unavailable (Fig.11.2.4). Additionally, this method can be used for epitope mapping of different mono-clonal antibodies that are directed against the same antigen. Plates are coated with captureantibodies specific for immunoglobulin from the immunized species. The test antibodysolution is incubated on the plates coated with the capture antibodies. Plates are thenwashed, incubated with antigen, washed again, and incubated with specific antibodyconjugated to an enzyme. After incubation, unbound conjugate is washed out and substrateis added. Wells that are positive for hydrolysis may contain antibodies specific for theantigen.


Capture antibodies specific for immunoglobulin from the immunized speciesSpecific antibody–alkaline phosphatase conjugate

1. Coat wells of an Immulon microtiter plate with 50 µl of 2 to 10 µg/ml captureantibodies as in steps 2 to 5 of the basic protocol.

NOTE: Capture antibodies must not bind the antigen or conjugate antibodies. Whenanalyzing hybridoma supernatants or ascites fluid, coat plates with 2 �g/ml captureantibody. When analyzing antisera, coat plates with 10 �g/ml capture antibody.

2. Block wells as in steps 6 and 7 of the basic protocol.

3. Prepare dilutions of test antibody solutions in blocking buffer. Add 50 µl to coatedwells and incubate ≥2 hr at room temperature.

Hybridoma supernatants, antisera, or ascites fluid can be used as the test samples. Dilutehybridoma supernatants 1:5 and antisera or ascites fluid 1:200.


5. Prepare an antigen solution in blocking buffer containing 20 to 200 ng/ml antigen.

Although purified antigen preparations are not essential, the limit of detectability for mostprotein antigens in this type of system is 2 to 20 ng/ml. A concentration of 20 to 200 ngantigen/ml is recommended.

6. Add 50-µl aliquots of the antigen solution to antibody-coated wells and incubate ≥2hr at room temperature.


8. Add 50 µl specific antibody–alkaline phosphatase conjugate to the wells and incubate2 hr at room temperature.


11.2.10


Assays

The conjugate antibodies must not react with the capture antibody or the test antibody. Theconjugate concentration is typically between 25 to 500 ng specific antibody/ml, and shouldbe high enough to result in ∼0.50 absorbance units/hr at 405 nm when using NPP as asubstrate or a signal of 1000 to 1500 fluorescence units/hr when using MUP as a substrate.If no specific antibodies from the appropriate species are available to serve as a positivecontrol, then a positive control system should be constructed out of available reagents.Such reagents can be found in Linscott’s Directory of Immunological and BiologicalReagents.

E

Ab

Ag

Ab

E

Ab

Ag

Ab

E

Ab

Ab

E

Ab

Ag

Ab

Ag

Ab

Ag

Ab

Ab Ab

Ab Ab

coat well with capture antibody

block

incubate with antibody

incubate with antigen

incubate withantibody-enzyme conjugate


Ab Ab

Ab Ab

Ab Ab

Ab Ab

Ab = detected Ab

wash

wash

wash

Ag

Figure 11.2.4 Double antibody–sandwich ELISA to detect specific antibodies. Ag = antigen; Ab= antibody; E = enzyme.


11.2.11

Immunology


10. Add 75 µl of MUP or NPP substrate solution to each well and incubate 1 hr at roomtemperature. After 1 hr, examine hydrolysis visually or spectrophotometrically (seestep 15 of the basic protocol).

In order to detect low-level reactions, the plate can be read again after several hours ordays of hydrolysis.

11. Check for false positives by rescreening samples that test positive for antigen-specificantibody. For each positive sample, coat four wells with capture antibody and armthe capture antibody with test antibody (steps 1 to 4). Incubate two of the wells withantigen (steps 5 to 7) and two of the wells with blocking buffer. Add conjugate andsubstrate to all four wells (steps 8 to 10) and measure hydrolysis after 1 hr.

This procedure will eliminate false positives resulting from test antibodies that react withthe enzyme-antibody complex.

ALTERNATEPROTOCOL

DIRECT CELLULAR ELISA TO DETECT CELL-SURFACE ANTIGENS

The expression of cell-surface antigens or receptors is measured using existing antibodiesor other ligands specific for cell-surface molecules (Fig. 11.2.5). Cells are incubated withenzyme conjugated to antibodies that are specific for a cell-surface molecule. Unboundconjugate is washed away and substrate is added. The level of antigen expression isproportional to the amount of substrate hydrolysis. This procedure can be as sensitive asflow cytometry analysis in quantitating the level of antigen expression on a population ofcells (Coligan et al., 1991). Unlike the flow cytometry analysis, however, this method isnot sensitive for mixed populations. This assay can be converted to an indirect assay bysubstituting biotinylated antibody for the enzyme-antibody conjugate, followed by asecond incubation with avidin–alkaline phosphatase.

E

Ab

C

E

Ab

C

E

Ab

C

E

Ab

Ag

C

wash andcentrifuge

incubate cells withantibody-enzyme conjugate

add substrate, resuspend cells,and observe color changeor fluorescence

Ag

Ag Ag

Ag = detected Ag

Figure 11.2.5 Direct cellular ELISA to detect cell-surface antigens. Ab = antibody; E = enzyme;C = cell.


11.2.12


Assays


Cell samplesSpecific antibody–alkaline phosphatase conjugate (see second support protocol)Wash buffer, ice-cold

Cone- or round-bottom microtiter platesSorvall H-1000B rotor (or equivalent)

1. Determine the optimal number of cells per well and the antibody-conjugate concen-tration by criss-cross serial dilution analysis (see first support protocol) using variablenumbers of positive- and negative-control cell samples and varying concentrationsof antibody-biotin conjugate.

Titrate cells initially at 1-5 × 105/well and conjugate at 0.5 to 10 �g/ml. For preparationand handling of cells, consult steps 2 to 5.

Because eukaryotic cells express variable amounts of alkaline phosphatase, test cells mustbe assayed in a preliminary experiment for alkaline phosphatase by incubation withsubstrate alone. If the test cells express unacceptable levels of alkaline phosphatase,another enzyme conjugate such as β-galactosidase should be used. Both chromogenic andfluorogenic substrates are available for β-galactosidase.

2. Centrifuge cell samples in a table-top centrifuge 5 min in Sorvall H-1000B rotor at1500 rpm (450 × g), 4°C, in a 15- to 50-ml centrifuge tube. Count cells (APPENDIX 3)and resuspend in ice-cold wash buffer at 1-5 × 106 cells/ml.

If the surface antigen retains its antigenicity after fixation, cells may be fixed at thebeginning of the experiment—but do not fix cells unless it can be demonstrated that theantigenicity is retained after fixation. Fix cells by suspending in glutaraldehyde (0.5% final;from a 25% stock, EM grade Sigma #G5882), and incubating 30 min at room temperature.Pellet cells, resuspend in PBSLE (see second support protocol), and incubate for 30 minat 37°C. Wash twice in PBSLE and resuspend in wash buffer. Cells can be kept for monthsat 4°C after fixation.

3. Dispense 100 µl of cell suspension (1-5 × 105 cells) into wells of cone- or round-bot-tom microtiter plates, and centrifuge 1 min at 450 × g, 4°C. Remove supernatant byvacuum aspiration, and disrupt pellet by briefly shaking microtiter plate on a vortexmixer or microtiter plate shaker.

4. Resuspend pellet in 100 µl of conjugate in ice-cold wash buffer at the optimalconcentration (see step 1). Incubate 1.5 hr at 4°C, resuspending cells by gentlyshaking at 15-min intervals.

Be careful not to splash cell suspensions out of wells.

5. Centrifuge cells 1 min at 450 × g, 4°C, remove supernatant by vacuum aspiration,briefly vortex pellet, and resuspend in 200 µl ice-cold wash buffer. Repeat three times.

6. Add 100 µl MUP or NPP substrate solution. Incubate 1 hr at room temperature,resuspending cells by gently shaking at 15-min intervals during hydrolysis.

7. Determine extent of hydrolysis by visual inspection or using a microtiter plate reader.


11.2.13

Immunology

ALTERNATEPROTOCOL

INDIRECT CELLULAR ELISA TO DETECT ANTIBODIESSPECIFIC FOR SURFACE ANTIGENS

This assay is designed to screen for antibodies specific for cell-surface antigens (Fig.11.2.6). Antibodies against surface antigens are detected by incubating whole cells witha test solution containing the primary antibody. The unbound antibody is washed awayand the cells are then incubated with an enzyme conjugated to antibodies specific for theprimary antibody. Unbound enzyme conjugate is washed away and substrate solutionadded. The level of bound primary antibody is proportional to the amount of substratehydrolysis.


Positive-control antibodies (i.e., those that react with the experimental cells and are from the immunized species)

Negative-control antibodies (i.e., those that do not react with the experimental cells)

Test antibody solutionAntibody or F(ab′)2 (against immunoglobulin from the immunized

species) conjugated to alkaline phosphataseCone- or round-bottom microtiter plates

C

Ag Ag

Ab Ab

C

Ab = detected Ab

incubate cells with antibody

incubate with antibody-enzymeconjugate

add substrate, resuspend cells,and observe color changeor fluorescence

wash andcentrifuge

C

Ag Ag

Ab Ab

C

wash andcentrifuge

E

Ab

E

Ab

C

Ag Ag

Ab Ab

C

E

Ab

E

Ab

Figure 11.2.6 Indirect cellular ELISA to detect antibodies specific for surface antigens. Ab =antibody; E = enzyme; C = cell.


11.2.14


Assays

1. Centrifuge and resuspend cell samples as in step 2 of the previous alternate protocolat 1-5 × 106 cells/ml.

Because this technique detects antibodies against uncharacterized epitopes, fixation priorto analysis is not recommended. Fixation may destroy the antigenicity of the epitope. Allsteps must be performed at 4°C in physiological buffers containing NaN3.

Because eukaryotic cells express variable amounts of alkaline phosphatase, test cells mustbe assayed for alkaline phosphatase activity. If the endogenous alkaline phosphatase levelis too high, another enzyme should be substituted for alkaline phosphatase in the antibody-enzyme conjugate (see annotation to step 1 of the previous alternate protocol).

2. In preliminary assays, determine the optimal number of cells per well and conjugateconcentration by criss-cross serial dilution analysis using positive- and negative-con-trol antibodies instead of test antibodies (see first support protocol). In adapting thecriss-cross serial dilution analysis, whole cells replace the solid-phase coatingreagent; see techniques for handling cells are outlined in steps 3 to 8. Set up titrationsby varying the number of cells between 1 × 105 and 5 × 105/well, the concentrationof positive- and negative-control antibodies between 0.1 and 10 µg/ml, and theconcentration of antibody-enzyme conjugate between 0.1 and 10 µg/ml.

3. Dispense 100 µl of cell suspension (1-5 × 105 cells) into wells of round- orcone-bottom microtiter plates. Centrifuge 1 min at 1500 rpm, 4°C, remove super-natant by vacuum aspiration, and disrupt pellet by briefly shaking microtiter plate onthe vortex mixer.

4. Resuspend cells in 100 µl solutions containing 1 to 10 µg/ml test antibody or controlantibodies in ice-cold wash buffer. Incubate 1.5 hr at 4°C, resuspending cells bygently shaking at 15-min intervals.

Be careful not to splash cell suspensions out of wells.

5. Centrifuge cells 1 min at 1500 rpm, 4°C, remove supernatant by vacuum aspiration,briefly vortex pellet, and resuspend in 200 µl ice-cold wash buffer. Repeat twice.

6. Resuspend pellet in 100 µl enzyme-antibody conjugate or F(ab′)2-enzyme conjugatediluted in ice-cold wash buffer. The optimal concentration of antibody, determinedin step 2, is usually 100 to 500 ng/ml. Incubate 1.5 hr at 4°C, resuspending cells bygently shaking at 15-min intervals.

When working with cells that may express Fc receptors, it is best to use enzyme conjugatedto F(ab′)2 fragments. F(ab′)2 fragments have had the Fc portion of the antibody enzymati-cally removed and no longer bind to Fc receptors.

7. Wash cells as in step 5. Repeat three times.

8. Add 100 µl MUP or NPP substrate solution. Allow hydrolysis to proceed until thesignal has reached the desired levels; resuspend cells by gently shaking at 15 minintervals during hydrolysis. If desired, stop hydrolysis by adding 25 µl of 0.5 MNaOH.

9. Determine extent of hydrolysis by visual inspection or spectrophotometrically usinga microtiter plate reader.


11.2.15

Immunology

SUPPORTPROTOCOL

CRISS-CROSS SERIAL DILUTION ANALYSIS TO DETERMINEOPTIMAL REAGENT CONCENTRATIONS

Serial dilution titration analyses are performed to determine optimal concentrations ofreagents to be used in ELISAs. In this protocol, all three reactants in a three-stepELISA—a primary solid-phase coating reagent, a secondary reagent that binds theprimary reagent, and an enzyme-conjugated tertiary developing reagent that binds to thesecondary reagent—are serially diluted and analyzed by a criss-cross matrix analysis (Fig11.2.7). Once the optimal concentrations of reagents to be used under particular assayconditions are determined, these variables are kept constant from experiment to experi-ment. The coating (primary), secondary, and developing (tertiary) reagents will varydepending upon which of the previous protocols needs to be optimized.


Coating reagentSecondary reagentDeveloping reagent17 × 100–mm and 12 × 74–mm test tubes

Columns

1 2 3 4 5 6 7 8 9 10 11 12

0 0 0 0 0 0 0 0 0 0 0 0

31.25 2700 2100 1200 410 120 0 60 10 10 10 0

62.5 3600 4000 2270 790 240 0 120 30 10 10 10

125 over over 3650 1370 360 0 195 40 10 10 0

250 over over over 2060 560 0 300 80 20 0 0

500 over over over 3200 1000 0 500 120 40 20 10

(ng/ml) 200 50 12.5 3.12 0.78 0 200 50 12.5 3.12 0.78 0

G

F

E

D

C

B

A

H

Secondary reactant

homologous(antigen)

heterologous(antigen)

Row

s

Ter

tiary

rea

ctan

t(a

ntib

ody-

alka

line

phos

phat

ase)

Figure 11.2.7 Results of a criss-cross serial dilutionanalysis (for optimization of secondary and tertiary reac-tant concentrations) of an antibody-sandwich ELISA todetect antigen. The numbers in columns 1 to 11 and rowsB to G represent relative fluorescence units observed foreach well on a 96-well microtiter plate.

Plates were coated overnight with the capture antibodyat 2 µg/ml. The secondary reactants, 4-fold serial dilu-tions of the homologous antigen and a non-cross-reac-tive heterologous antigen, were incubated on the plate 2hr. The tertiary reactant, 2-fold serial dilutions of specificantibody–alkaline phosphatase conjugates, were incu-bated on the plate 2 hr. After 1 hr of incubation with thesubstrate MUP, the fluorescence was read in a microtiterplate spectrofluorometer.

Reagent concentrations depend upon individual assayvariables that are set by the investigator. If the time ofhydrolysis is set at 1 hr, the relative fluorescence at∼1000 relative fluorescence units, and the sensitivity at780 pg/ml of homologous antigen, then 500 ng/ml ofenzyme-antibody conjugate must be used in the ELISA.If, however, the assay has to detect only 3.12 ng/ml ofhomologous antigen, then the concentration of conju-gate can be reduced to 125 ng/ml. It should be noted bycomparing the homologous with the heterologous reac-tions (wells B5 versus B11 and D4 versus D10) that boththe specificity and the signal-to-noise ratio for this assayare excellent.


11.2.16

Prepare coating-reagent dilutions1. Place four 17 × 100–mm test tubes in a rack and add 6 ml PBSN to the last three

tubes. In tube 1, prepare a 12-ml solution of coating reagent at 10 µg/ml in PBSN.Transfer 6 ml of tube 1 solution to tube 2. Mix by pipetting up and down five times.Repeat this transfer and mix for tubes 3 and 4; the tubes now contain the coatingreagent at 10, 5, 2.5, and 1.25 µg/ml.

2. Using a multichannel pipet, dispense 50 µl of the coating reagent solutions into wellsof four Immulon microtiter plates (i.e., each plate is filled with one of the fourdilutions). Incubate overnight at room temperature or 2 hr at 37°C.

3. Rinse and block plates with blocking buffer as in steps 5 to 7 of the basic protocol.

Prepare secondary-reagent dilutions4. Place five 12 × 75–mm test tubes in a rack and add 3 ml blocking buffer to the last

four tubes. In tube 1, prepare a 4-ml solution of secondary reagent at 200 ng/ml inPBSN. Transfer 1 ml of tube 1 solution to tube 2. Pipet up and down five times. Repeatthis transfer and mix for tubes 3 to 5; the tubes now contain the secondary reactantat 200, 50, 12.5, 3.125, and 0.78 ng/ml. If possible, prepare and test serial dilutionsof a nonreactive heterologous form of the secondary reactant in parallel (Fig. 11.2.7).

If the assay is especially insensitive, it may be necessary to increase the secondary reactantconcentrations so the tube-1 solution is 1000 ng/ml.

5. Dispense 50 µl of the secondary reagent solutions into the first five columns of allfour coated plates. The most dilute solution is dispensed into column 5, whilesolutions of increasing concentration are added successively into columns 4, 3, 2, and1. Thus, the fifth column contains 0.78 ng/ml and the first column 200 ng/ml. Incubate2 hr at room temperature.

6. Wash plates as in steps 9 to 11 of the basic protocol.

Prepare developing-reagent dilutions7. Place five 17 × 100–mm test tubes in a rack and add 3 ml blocking buffer to the last

four tubes. In tube 1, prepare a 6-ml solution of developing reagent at 500 ng/ml inblocking buffer. Transfer 3 ml of tube 1 solution into tube 2 and mix. Repeat thistransfer and mixing for tubes 3 and 4—the tubes now contain the developing reagentat 500, 250, 125, 62.5, and 31.25 ng/ml.

8. Dispense 50 µl of the developing reagent solutions into the wells of rows 2 to 6 ofeach plate, dispensing the most dilute solution into row 6 and solutions of increasingconcentration successively into rows 5, 4, 3, and 2. Incubate 2 hr at room temperature.

9. Wash plates as in steps 9 to 11 of the basic protocol.

Measure hydrolysis10. Add 75 µl MUP or NPP substrate solution to each well, incubate 1 hr at room

temperature, and measure the degree of hydrolysis visually or with a microtiter platereader. An appropriate assay configuration results in 0.50 absorbance units/hr at 405nm when using NPP as a substrate or 1000 to 1500 fluorescence units/hr when usingMUP as a substrate.

These results can be used to adjust optimal concentrations in the basic and alternateprotocols.


11.2.17

Immunology

SUPPORTPROTOCOL

PREPARATION OF BACTERIAL CELL LYSATE ANTIGENS

A culture of E. coli containing proteins expressed from cloned genes is lysed for use astest antigen in any of the first three protocols of this unit. For more extensive discussionon protein expression for antigen production, see UNITS 16.4-16.7 (expression by fusionprotein vectors).

Materials

Escherichia coli culture in broth or agar (UNITS 1.2 & 1.3)Cell resuspension bufferLysozyme solutionTris/EDTA/NaCl (TEN) buffer (UNIT 11.1)10% sodium dodecyl sulfate (SDS)8 M urea (optional)

Nylon-tipped applicator (Falcon #2069, Becton Dickinson)

1. For liquid culture, centrifuge 5 ml of cells at 2500 rpm in a tabletop centrifuge for 10min. Decant supernatant and resuspend pellet in 5 ml cell resuspension buffer byvortexing gently. For agar culture, remove about 10 colonies from the plate using anylon-tipped applicator and resuspend in 2 ml cell resuspension buffer. Press swabagainst side of tube to remove as much liquid as possible.

Yield of expressed protein may vary with growth phase. Samples should be taken foranalysis at various periods of growth (e.g., mid-log and stationary phases). If samples aretaken from agar plates, the culture should be grown overnight at 37°C.

2. Place 1 ml of resuspended cells in a microcentrifuge tube on ice.

3. Add 0.2 ml lysozyme solution to the tube and leave 5 min on ice.

4. Microcentrifuge 5 min. Decant supernatant and save. Resuspend pellet in 1.2 ml TENbuffer.

Since many expressed proteins are insoluble, it is worthwhile to assay both the pellet andsupernatant for activity.

5. Add 0.065 ml of 10% SDS solution to each sample. Incubate 10 min at 37°C. Samplesare ready for ELISA at this point. Store frozen if not used within several hours.

Alternatively, add urea to a final concentration of 8 M (4.8 g to a final volume of 10 ml) todenature and solubilize proteins.


Borate-buffered saline (BBS)0.017 M Na2B4O7⋅10H2O0.12 M NaClAdjust to pH 8.5 with NaOH

Blocking bufferBBS (see above) containing:0.05% Tween 201 mM EDTA0.25% bovine serum albumin (BSA)0.05% NaN3

Store at 4°C

Gelatin may be substituted for BSA; 5% instant milk has been successfully used but mayinterfere nonspecifically with antibody binding.


11.2.18


Assays

Cell resuspension buffer (10 mM HEPES)2.38 g HEPESAdd H2O to 1 liter

Lysozyme solution5 mg chicken egg white lysozyme (Sigma Grade VI #L2879)1 ml TEN buffer (UNIT 11.1)Make fresh immediately before use

MUP substrate solution0.2 mM 4-methylumbelliferyl phosphate (MUP; Sigma #M8883)0.05 M Na2CO3

0.05 mM MgCl2

Store at room temperature

NPP substrate solution3 mM p-nitrophenyl phosphate (NPP; Sigma #104-0)0.05 M Na2CO3

0.05 mM MgCl2

Store at 4°C

Test antibody solutionHybridoma supernatants (UNIT 11.10) can usually be diluted 1:5 and ascites fluid andantisera (UNIT 11.12) diluted 1:500 in blocking buffer and still generate a strongpositive signal. Dilutions of nonimmune ascites or sera should be assayed as anegative control. Prepare antibody dilutions in cone- or round-bottom microtiterplates before adding them to antigen-coated plates.

Sources of appropriate antibodies and conjugates can be found in Linscott’s Directory ofImmunological and Biological Reagents.

Test antigen solution0.2 to 10 µg/ml antigen, purified or partially purified in PBSN; store at 4°C

Wash bufferHanks balanced salt solution (HBSS; APPENDIX 2)1% fetal calf serum (FCS; heat-inactivated 60 min, 56°C)0.05% NaN3

Store at 4°C

COMMENTARY

Background InformationSince their first description in 1971 (Engvall

and Perlman), ELISAs have become the systemof choice when assaying soluble antigens andantibodies. Factors that have contributed totheir success include their sensitivity, the longshelf-life of the reagents (alkaline phosphataseconjugates typically lose only 5% to 10% oftheir activity per year), the lack of radiationhazards, the ease of preparation of the reagents,the speed and reproducibility of the assays, andthe variety of ELISA formats that can be gen-erated with a few well-chosen reagents. Addi-tionally, no sophisticated equipment is neces-sary for many ELISA applications, including

screening hybridoma supernatants for specificantibodies and screening biological fluids forantigen content.

The ELISAs described here combine thespecial properties of antigen-antibody interac-tions with simple phase separations to producepowerful assays for detecting biological mole-cules. The multivalency of antibodies can resultin the formation of long-lived antigen-antibodycomplexes, thus allowing long periods of timeduring which such complexes can be measured.By designing an assay so that a capture reagentinitiates the binding of antigen-antibody com-plexes and enzyme conjugates onto a solidphase, the unbound reagents can be easily and


11.2.19

Immunology

rapidly separated from the solid phase. Thesolid phase is washed and the amount of boundconjugate is visualized by incubating the solidphase with a substrate that forms a detectableproduct when hydrolyzed by the bound en-zyme. ELISAs are similar in principle to ra-dioimmunoassays, except that the radioactivelabel is replaced by an enzyme conjugate.

A number of different enzymes have beensuccessfully used in ELISAs, including alka-line phosphatase, horseradish peroxidase, β-galactosidase, glucoamylase, and urease. Alka-line phosphatase—perhaps the most widelyused conjugated enzyme—is recommendedbecause of its rapid catalytic rate, excellentintrinsic stability, availability, ease of conjuga-tion, and resistance to inactivation by commonlaboratory reagents. Additionally, the sub-strates of alkaline phosphatase are nontoxic andrelatively stable. Solutions of p-nitrophenylphosphate (NPP) are stable for months at 4°C,while solutions of 4-methylumbelliferyl phos-phate (MUP) can be kept for months at roomtemperature without any significant spontane-ous hydrolysis. The biggest disadvantage ofalkaline phosphatase is that if NPP is used as asubstrate, the yellow color of the nitrophenylproduct is relatively difficult to detect visually.Using the substrate MUP instead of NPP cangreatly enhance the sensitivity of the assay. Thefluorogenic system using MUP is 10 to 100times faster than the chromogenic system usingNPP, and appears to be as sensitive as an en-hanced chromogenic assay in which alkalinephosphatase generates NAD+ from NADP(Macy et al., 1988). The disadvantage of usingfluorogenic substrates is that they require amicroplate fluorometer costing twice as muchas a high-quality microtiter plate spectro-photometer.

Cellular ELISAs have been shown to be assensitive as flow cytometry analysis in detectingsome cell-surface antigens (Bartlett and Noelle,1987) and are potentially of great value in rap-idly screening hybridoma supernatants for anti-bodies against surface molecules (Feit et al.,1983). Using ELISAs for screening large num-bers of hybridoma supernatants has been hin-dered by the large number of cells required andhigh background signal. The increased sensitiv-ity of the fluorogenic system should reduce thenumber of cells needed by a factor of 5, makingthe system more useful as a screening assay.

In addition to the methods described here,hundreds of other ELISA applications havebeen described, including the determination of

antibody affinities (Beatty et al., 1987; Schotset al., 1988), the detection of antibodies specificfor hormone receptors (Quinn et al., 1988;Wang and Leung, 1985), expression cloning oflymphokine receptors (Harada et al., 1990), andhomogeneous assays in which a solid phase isnot needed because the antigen-antibody inter-action itself modifies the enzymatic activity(Rubenstein et al., 1972). A number of booksare devoted to ELISAs and can be consulted forfurther discussion (Maggio, 1981; Kurstak,1986).

Critical ParametersSensitive ELISAs require antibodies of high

affinity and high specificity. Since the sensitiv-ity of an ELISA depends upon the affinity ofthe antibodies involved, antibodies with thehighest affinities should be used when settingup ELISAs. Antibodies should be screened forunwanted cross-reactions. For instance, cap-ture antibodies must not bind conjugate anti-bodies and vice versa. There are many commer-cial sources of reliable reagents. Linscott’s Di-rectory of Immunological and BiologicalReagents is an excellent source book for locat-ing reagents used in ELISAs. If reagents fromone source are inadequate, try another.

When screening for expressed proteins in E.coli, it is important to utilize conjugates withantibodies that recognize nonnative and nativemolecules. Many foreign proteins expressed inE. coli will not assume their native conforma-tion, and expression of such proteins will notbe detected if antibody specific for the nativeform is used. It is also important to test enzyme-antibody conjugates for cross-reactivity or non-specific binding to host cell molecules. Thispotential problem can be eliminated by incor-porating these as control antigens in the screen-ing procedures used to select the original anti-bodies in the basic protocol.

When coating plates with antigen, the anti-gen preparation need not be pure, but shouldgenerally comprise >3% of the protein in thecoating solution. In some situations, dilution ofthe antigen solution with BSA has greatly im-proved the sensitivity of the ELISA (Jitsukawaet al., 1989).

All steps after coating the microtiter platesshould be carried out in solutions containing0.05% Tween 20 and a carrier protein (0.25%BSA or gelatin).

When using ELISAs for quantitative deter-minations of antigen or antibody concentra-tions, four guidelines should be followed.


11.2.20


Assays

First, it is essential that all experimental condi-tions up to the final wash after incubation withconjugate—including incubation times, washtimes, reagent concentrations, and tempera-ture—be kept constant between experiments.This is especially important in assays usingpolyvalent antibodies and complex mixtures ofantigens. The optimal concentrations of all re-agents for each system should be determinedin an initial criss-cross serial dilution experi-ment (see first support protocol). Second, be-cause the efficiency of binding and other mi-cro-environmental conditions can vary fromplate to plate, a standard curve should be in-cluded on each plate. Third, all samples mustbe analyzed at least in duplicate. Fourth, theconcentration of the reagent being quantitatedmust lie within the dynamic range of the stand-ard curve.

Anticipated ResultsAntibody-sandwich assays are generally the

most sensitive ELISA configuration and candetect concentrations of protein antigens be-tween 100 pg/ml and 1 ng/ml. ELISAs in whichantigen is directly bound to plates are usuallyan order of magnitude less sensitive than sand-wich techniques.

Either the direct or sandwich ELISA may beused to detect and quantitate a bacterially ex-pressed antigen or a purified or partially puri-fied antigen in the range of 1 ng/ml to 1 µg/ml.A major disadvantage of the direct ELISA isthat when an impure antigen preparation like abacterial lysate is coated directly onto the sur-face of the microtiter well, the antigen mustcompete with all the other macro-molecules inthe lysate for binding to the plastic and verylittle of the desired antigen may be bound. Thesandwich ELISA bypasses this problem byrelying on selective adsorption of an antigen toan antigen-specific antibody-coated surface.

Time ConsiderationsThese assays are designed to take ∼6 hr, but

the incubation times may be abbreviated orexpanded as needed. Since equilibrium bindingbetween the soluble and solid phases frequentlytakes 5 to 10 hr, stronger specific signals canusually be obtained by longer incubations.

Fluorogenic ELISAs are generally 10 to 100times faster than assays using chromogenicsubstrates.

Literature CitedBartlett, W.C. and Noelle, R.J. 1987. A cell-surface

ELISA to detect interleukin 4–induced class IIMHC expression on murine B cells. J. Immunol.Methods 105:79-85.

Beatty, J.D., Beatty, B.G., and Vlahos, W.G. 1987.Measurement of monoclonal affinity by non-competitive immunoassay. J. Immunol. Methods100:173-179.

Coligan, J.E., Kruisbeek, A.M., Margulies, D.H.,Shevach, E.M., and Strober, W., eds. 1991. Cur-rent Protocols in Immunology, Chapter 5: Im-munofluorescence and cell sorting. Greene Pub-lishing and Wiley-Interscience, New York.

Engvall, E. and Perlman, P. 1971. Enzyme-linkedimmunosorbent assay (ELISA): Quantitative as-say of immunoglobulin G. Immunochemistry8:871-879.

Feit, C., Bartal, A.H., Tauber, G., Dymbort, G., andHirshaut, Y. 1983. An enzyme-linked immu-nosorbent assay (ELISA) for the detection ofmonoclonal antibodies recognizing antigens ex-pressed on viable cells. J. Immunol. Methods58:301-308.

Harada, N., Castle, B.E., Gorman, D.M., Itoh, N.,Schreurs, J., Barrett, R.L., Howard, M., and Mi-yajima, A. 1990. Expression cloning of a cDNAencoding the murine interleukin 4 receptor basedon ligand binding. Proc. Natl. Acad. Sci. U.S.A.87:857-861.

Jitsukawa, T., Nakajima, S., Sugawara, I., andWatanabe, H. 1989. Increased coating efficiencyof antigens and preservation of original antigenicstructure after coating in ELISA. J. Immunol.Methods 116:251-257.

Kurstak, E. 1986. Enzyme Immunodiagnosis. Aca-demic Press, San Diego.

Linscott’s Directory of Immunological and Biologi-cal Reagents, Santa Rosa, Calif.

Macy, E., Kemeny, M., and Saxon, A. 1988. En-hanced ELISA: How to measure less than 10picograms of a specific protein (immunoglobu-lin) in less than 8 hours. FASEB J. 2:3003-3009.

Maggio, E.T. 1981. Enzyme Immunoassay. CRCPress, Boca Raton, Fla.

Quinn, A., Harrison, R., Jehanli, A.M.T., Lunt, G.G.,and Walsh, S.S. 1988. An ELISA for the detec-tion of anti-acetylcholine receptor antibodies us-ing biotinylated α-bungarotoxin. J. Immunol.Methods 107:197-203.

Rubenstein, K.E., Schneider, R.S., and Ulmann,E.L. 1972. Homogeneous enzyme immunoas-say: A new immunochemical technique. Bio-chem. Biophys. Res. Commun. 47:846.

Schots, A., Van der Leede, B.J., De Jongh, E., andEgberts, E. 1988. A method for the determinationof antibody affinity using a direct ELISA. J.Immunol. Methods 109:225-233.


11.2.21

Immunology

Wang, K.C. and Leung, B.S. 1985. FluorometricELISA methods for rapid screening of anti-es-trogen receptor antibody production in hybri-doma cultures. J. Immunol. Methods 84:279.

Key ReferenceLinscott’s Directory. See above.

Highly recommended publication listing sources ofimmunological reagents, kits, and cells/organisms,including addresses and phone numbers of commer-cial suppliers (updated quarterly).

Contributed by Peter HornbeckUniversity of MarylandBaltimore, Maryland

Scott E. Winston (bacterial cell lysate antigens)Univax BiologicsRockville, Maryland

Steven A. Fuller (bacterial cell lysate antigens)Allelix Inc.Mississauga, Ontario


11.2.22


Assays

UNIT 11.3Isotype Determination of AntibodiesFrequently it is necessary to know the amount and serological class of antibodies madeby an immunized animal, produced by hybridomas, or present in the serum of patientswith inflammatory or neoplastic conditions. The immunologist’s approach to such aproblem is to consider the antibody or immunoglobulin molecules themselves as antigensand to use anti-immunoglobulin antibodies as the specific and sensitive agents ofdetection. This unit describes two methods for measurement and classification of super-natant or serum immunoglobulins—an ELISA (first basic protocol) and a methodemploying electrophoresis and immunofixation (second basic protocol).

BASICPROTOCOL

SANDWICH ELISA FOR ISOTYPE DETECTION

The speed and sensitivity of sandwich ELISAs make them the assays of choice for isotypedetermination. The antibody-sandwich ELISA to measure soluble antigen (UNIT 11.2; Fig.11.2.3) is adapted in this unit for isotype detection. Microtiter wells are coated withisotype-specific capture antibodies followed by incubation with test solutions containingthe antibodies to be isotyped. After test antibodies have been bound to the plate by reactingwith capture antibodies, unbound test antibodies are washed out. Developing reagent isadded to the wells, followed by another incubation. Unbound conjugate is washed outand substrate is added. Substrate hydrolysis indicates that the test solution contained theappropriate isotype. A lack of hydrolysis indicates that the test solution did not containthe appropriate isotype. UNIT 11.2 should be consulted for additional information.

Materials

Capture anti-isotype antibodies: heavy-chain class-specific antibodies (anti-µ, -α,-γ, -δ, -ε), heavy-chain subclass-specific antibodies (anti-γ1, -γ2a, -γ2b, -γ3,-γ4), or light-chain isotype-specific antibodies (anti-κ, -λ)

PBS (APPENDIX 2) containing 0.05% NaN3 (PBSN)Test antibodies: hybridoma supernatants, ascites fluid, or antiseraBlocking buffer (UNIT 11.2)Standard isotype antibodies (i.e., purified antibodies of known isotypes)Developing reagent: anti-Ig antibody (specific for all heavy-chain

classes)–alkaline phosphatase conjugate (see UNIT 11.1; Southern Biotechnologyor Linscott’s Directory)

MUP or NPP substrate solution (UNIT 11.2)

Immulon 2 or 4 microtiter plates (or equivalent; UNIT 11.2)

Additional reagents and equipment for ELISA (UNIT 11.2)

1. Prepare the capture anti-isotype antibodies by diluting in PBSN to 2 µg/ml final.

Capture antibodies can be monoclonal or polyclonal.

When screening hybridoma supernatants for IgG (anti-γ), it is advisable to initially use acapture-antibody preparation that recognizes all IgG isotypes. Subsequent analysis usingsubclass-specific capture antibodies can determine which IgG subclass is expressed. It iscrucial that the capture antibodies do not bind to the antibodies in the developing reagent.

2. Coat each well of an Immulon microtiter plate with 50 µl capture antibody solutionas in steps 3 to 5 of the basic protocol in UNIT 11.2.

3. Block wells as in steps 6 and 7 of the basic protocol in UNIT 11.2.

4. Prepare dilutions of the test antibodies in blocking buffer—typically, hybridomasupernatants are diluted 1:5 while sera are diluted 1:500.

Supplement 18

Contributed by Peter Hornbeck, Thomas A. Fleisher, and Nicholas M. PapadopoulosCurrent Protocols in Molecular Biology (1992) 11.3.1-11.3.6Copyright © 2000 by John Wiley & Sons, Inc.

11.3.1

Immunology

5. Prepare the standard isotype antibodies at ∼500 ng/ml to serve as positive controls.

6. Transfer 50-µl aliquots of test or standard isotype antibodies into the antibody-coatedwells and incubate >2 hr at room temperature.

Specificity controls consisting of wells coated with capture antibodies and a series ofstandard antibodies of different isotypes should be included on each plate. Negativecontrols should be included on each plate and consist of wells coated with captureantibodies but receive no test or control antibodies.

7. Wash plate as in steps 9 to 11 of the basic protocol in UNIT 11.2.

8. Prepare the developing reagent so that the final conjugate solution contains ∼200 nganti-Ig/ml.

Crucial that antibodies in developing reagent do not bind to capture antibodies.

9. Add 50 µl developing reagent to each well and incubate 2 hr at room temperature.

All test wells, positive-control wells, and negative-control wells should receive the devel-oping reagent.

10. Wash as in steps 9 to 11 of the basic protocol in UNIT 11.2.

11. Add 75 µl MUP or NPP substrate solution to each well and incubate at roomtemperature. Periodically check the plate for substrate hydrolysis.

Hydrolysis of NPP results in liberation of a yellow product that can be detected by visualinspection in ambient light. Hydrolysis of MUP results in the liberation of a fluorescentproduct that can be detected by visual inspection under a long-wavelength UV lamp in adarkened room. For more quantitative estimates of isotype concentrations, plates can beread with microtiter plate reader. Positive-control wells give strong signals by 1 hr. Weakerreactions can be detected by incubation for many hours or overnight.

BASICPROTOCOL

DETECTING AND ISOTYPING ANTIBODIES BYELECTROPHORESIS AND IMMUNOFIXATION

Qualitative identification and quantitative determination of serum (and other biologicalfluid) proteins provide useful information concerning pathologic conditions of the lym-phoid system. High-resolution zone electrophoresis is a simple method to separate serumproteins based on their classification defined by five electrophoretic zones: albumin,α1-globulin, α2-globulin, β-globulin, and γ-globulin. Following electrophoresis, theproteins can be detected by staining with amido black, or immunofixation can beperformed first to achieve more precise identification. In immunofixation proceduredescribed below, immunoglobulins within separated protein bands are identified, andclonality is established, using antisera to the α, γ, µ, ε, and δ heavy chains and κ and λlight chains of immunoglobulins. Other serum proteins—including glycoproteins, trans-ferrin, and C3—can be identified in electrophoresed sample by same technique.

The serum (or other fluid) is loaded on an agarose gel–covered microscope slide andelectrophoresed. Proteins are detected after zone electrophoresis and can be identified byimmunofixation.

Materials

Serum (or other biological fluid)Normal saline95% methanol/5% acetic acid1% amido black (1 g in 100 ml of 2.5% acetic acid)2.5% (v/v) acetic acid


11.3.2

IsotypeDetermination

of Antibodies

2 × 2–cm cellulose acetate stripMonospecific anti-Ig, heavy-chain-specific (α, γ, µ, ε, or δ) or

light-chain-specific (κ or λ)0.85% (w/v) NaCl

Agarose gel–covered microscope slidesPlexiglas electrophoresis cell with two agarose bridgesElectrophoresis power supply (e.g., Pharmacia EPS 500/400)

Electrophorese the sample1. Cut a narrow slit with a razor blade in the middle of an agarose gel–covered

microscope slide. For standard fixation and staining (steps 3 and 4), place 1.5 µlundiluted serum in the slit. For immunofixation (steps 5 to 8), dilute serum in normalsaline to generate 1 to 2 mg/ml of the specific protein being evaluated (e.g., IgG, IgA,or IgM) and place in the slit.

2. Place the slide in an electrophoresis cell with two agarose bridges having the samecomposition as the gel on the slide. Electrophorese 12 min at 140 V. Proceed to steps3 and 4 for standard fixation and staining. Proceed to steps 5 to 8 for immunofixation.

Standard fixation and staining3. Fix slides by submerging in 95% methanol/5% acetic acid 15 min. Air dry with filter

paper placed on the gel.

4. Stain agarose gel by submerging in 1% amido black 10 min. Destain in 2.5% aceticacid until the background clears, then rinse in water. Dry 5 min at 60°C in a dryingoven for visual inspection (Fig. 11.3.1).

Immunofixation5. Transfer the unstained and unfixed gel (from step 2) to a petri dish containing damp

filter paper. Overlay the γ-globulin zone of the gel with a cellulose acetate stripimpregnated with a monospecific anti-Ig.

The γ-globulin zone is that nearest the cathode (Fig. 11.3.1).

The cellulose acetate strip is cut from any commercial membrane and dipped into theappropriate antibody solution (see critical parameters). The concentration of the antibodysolution provided by supplier (usually 0.5 to 1 mg/ml) is suitable.

6. Allow the antibody to diffuse into the gel 15 min at room temperature.

7. Remove the the slides and immerse in 0.85% NaCl for 2 hr to wash out unfixedproteins. Air dry with filter paper placed on the gel.

8. Stain as described in step 4 and visually inspect for immunoprecipitated protein bands(Fig. 11.3.2).

globulins

albu

min

+ –

α β γ

Figure 11.3.1 Normal serum proteinelectrophoretic pattern obtained by agarose gelzone electrophoresis demonstrating the fiveprotein zones.


11.3.3

Immunology


Agarose gel–covered microscope slidesAdd 0.5 g agarose (Seakem agarose HE, FMC Bioproducts) to 100 ml of 0.05 Mbarbital buffer (Sigma #B0500); store at room temperature and heat to 100°C. Stiruntil agarose has dissolved and allow to cool to 70°C. Pour 2.5 ml of the agarosesolution on each microscope slide and allow to gel 2 to 3 min at room temperature.Store the prepared slides at 4°C in a humidified chamber. Set up a chamber byhalf-filling a petri dish with the above agarose solution. Allow to cool and place ina refrigerator. This humidified chamber can be used to store agarose slides up to2 weeks.

COMMENTARY

Background Information

Sandwich ELISAAntibodies are heteromeric molecules con-

sisting of heavy and light chains, each of whichcontains a variable and a constant region.Heavy-chain constant regions include µ, α, γ1,γ2a, γ2b, γ3, γ4, δ, or ε, depending upon thespecies; light-chain constant regions include κand λ. Immunoglobulin constant regions, com-monly referred to as isotypes, determine manyof the biological and immunochemical proper-ties of the antibody molecule including com-plement fixation, binding to Fc receptors, andbinding to proteins A and G. Because the iso-

type may influence the method of purification(Andrew and Titus, 1991), it is routine to deter-mine the isotypes of monoclonal antibodies orother specific antibody preparations as part oftheir initial characterization. The identificationof antibody isotypes can easily be performedwith an ELISA employing commercially avail-able anti-isotype reagents. Alternatively, iso-types can be determined using electrophore-sis/immunofixation (second basic protocol) ora double-immunodiffusion assay (Hornbeck,1991).

See UNIT 11.2 for a full discussion of theELISA technique.

globulins

– al

bum

in

+

βα γ

– patient

anti-µ

anti-κ

A

B

C

monoclonalband

+ –

+ –

Figure 11.3.2 Agarose gel zone electrophresis of patient serum demonstrating a monoclonalband in the γ-globuling zone (A; arrow). Immunofixation electrophoresis with anti-µ (B) and anti-κ(C) demonstrate that the band is a µ-κ monoclonal immunoglobulin. There is no reactivity with theantisera to the other heavy (α, γ, δ, ε) and light (λ) chains (data not shown).


11.3.4


of Antibodies

Electrophoresis and immunofixationZone electrophoresis is based on the princi-

ple that charged particles migrate at differentrates in an electric field based on the net chargeof the particle. The application of this methodto the evaluation of serum proteins was firstdescribed in a seminal paper by Tiselius (1937).In this paper it was shown that serum proteinswere separable into defined zones (albumin,α-globulins, β-globulins, and γ-globulins).While protein separation by zone electropho-resis is excellent, protein quantitation using thismethod is poor. To overcome these shortcom-ings, various support media have been em-ployed, including cellulose acetate, agar gel,and, more recently, agarose gel. Agarose gelprovides improved stability and clarity, as wellas greater sensitivity than other materials (Pa-padopoulos et al., 1982).

Immunofixation electrophoresis combinesthe high resolution of agarose gel zone electro-phoresis and the unique specificity of an anti-gen-antibody reaction (Johnson, 1982). Afterelectrophoretic separation, the antigen of inter-est is reacted with an overlayed, monospecificantibody to form an immunoprecipitate, whichcan be easily detected. The location of theimmunoprecipitate depends on the electro-phoretic migration of the specific protein anti-gen. The unreacted proteins and antibody re-agents are washed out of the agarose gel andthe precipitin band is stained for visualization.Using this method, polyclonality, oligoclonal-ity, or monoclonality can be ascertained (Fig.11.3.2). Commercial sources for complete im-munofixation electrophoresis setups are avail-able. Immunoelectrophoresis, a classic methodin which diffusion of antibody into the gel iscombined with electrophoresis, is an alterna-tive method for evaluation of protein clonality.However, this approach is less sensitive andmore difficult to interpret as compared withimmunofixation (Johnson, 1986).

Critical Parametersand Troubleshooting

Sandwich ELISASources of isotype-specific antibodies can

be found in Linscott’s Directory of Immu-nological and Biological Reagents (see keyreferences, UNIT 11.2). Isotype-specific antibod-ies should have no detectable cross-reactivityagainst other isotypes and should not cross-re-act with other antibodies that might be used inthe assay. Check isotype-specific reagentsagainst standard isotype proteins to confirm

their specificity. Sources of purified isotypesfrom the species of interest, to be used asexperimental standards, can also be found inLinscott’s Directory. Alternatively, standardisotype proteins can be prepared from myelomaand hybridoma lines of known isotypes usingstandard techniques (UNIT 11.8). Many myelomasand hybridomas of defined isotypes and speci-ficities are available from ATCC and othersources; see full listing in Knapp et al. (1991).

The concentration of the developing reagentshould be adjusted so that the positive controlgives a strong signal by 1 hr. Since the hydroly-sis of MUP is at least 10-fold easier to detectthan the hydrolysis of NPP, assays using MUPcan be significantly faster than those using NPP.Test solutions should be scored as positive onlywhen they give 3-fold higher signals than thenegative controls.

While this assay is designed to qualitativelydetermine the presence or absence of a givenisotype in the test solution, it can be easilymodified to quantitate the concentration of iso-type by including serial dilutions of standardisotype proteins (for details of quantitation us-ing a standard curve, see UNIT 11.2 protocol forantibody-sandwich ELISA to measure solubleantigen).

Electrophoresis and immunofixationAntigen and antibody interaction at or near

the point of equivalence results in the formationof immune complexes that produce an insol-uble precipitate. In the case of a monoclonal(homogeneous) protein, this precipitate isfound in a very narrow band, while a polyclonalprotein will generate a broad band. Immuno-precipitation is optimal at antigen/antibodyequivalence. It is often useful to quantitate totalimmunoglobulin levels in the sample to allowfor the dilution of the serum (see step 1 of basicprotocol). Antigen excess will result in clearspots (lack of detectable immunoprecipitate) inthe location where the band(s) are anticipated.If this occurs, dilute the sample and repeatimmunofixation electrophoresis.

Failure to see a clearly discernible electro-phoretic protein pattern after step 4 suggests atechnical problem at some point during steps 1through 4. Immunofixation should not be per-formed and the initial electrophoresis proce-dure should be repeated.

Absence of a detectable immunoprecipitateafter step 8 (in the presence of a gamma globu-lin band after step 4) suggests antigen excess.Quantitate immunoglobulins, dilute serumaccordingly, and repeat entire experiment.


11.3.5

Immunology

Anticipated ResultsSandwich ELISAs are typically sensitive to

0.5 to 2.0 ng/ml of antibody isotype.During electrophoresis and immunofixa-

tion, the specific bands observed in im-munofixation electrophoresis are dependent onboth the presence of particular proteins and theappropriate antisera being used in the detectionprocess. In the case of normal immunoglobu-lins, all three major isotypes should be ob-served. Monoclonal or oligoclonal immuno-globulins produce single or multiple darker(clonal) bands (Fig. 11.3.2). The detection limitfor a specific protein using this immunofixationtechnique is 5 to 10 µg/ml.

Time ConsiderationsThe sandwich ELISA requires 6 to 8 hr.

However, the times allotted for the variousincubation steps can usually be reduced by half,so results can be obtained in 3 to 4 hr.

For electrophoresis and immunofixation,running and developing the gel takes 2 to 3 hr.Additional time (10 to 15 min) is needed toprepare and titrate the serum sample if immu-noglobulin levels are increased.

Literature CitedAndrew, S.M. and Titus, J.A. 1991. Purification of

immunoglobulin G. In Current Protocols in Im-munology (J.E. Coligan, A.M. Kruisbeek, D.H.Margulies, E.M. Shevach, and W. Strober, eds.)pp. 2.7.1-2.7.12. Greene Publishing and Wiley-Interscience, New York.

Hornbeck, P. 1991. Double-immunodiffusion assayfor detecting specific antibodies. In Current Pro-tocols in Immunology (J.E. Coligan, A.M. Kruis-beeck, D.H. Margulies, E.M. Shevach, and W.Strober, eds.) pp. 2.3.1-2.3.4. Greene Publishingand Wiley-Interscience, New York.

Johnson, M.A. 1982. Immunofixation electrophore-sis. Clin. Chem. 28:1797-1800.

Johnson, M.A. 1986. Immunoprecipitation in gels.In Manual of Clinical Laboratory Immunology.(N.R. Rose, H. Friedman, and J.L. Fahey, eds.)pp. 14-24. Am. Soc. Microbiol., Washington,D.C.

Knapp, W., Stockinger, H., Majdic, O., and Shevach,E.M. 1991. The CD system of leukocyte surfacemolecules. In Current Protocols in Immunology(J.E. Coligan, A.M. Kruisbeeck, D.H. Margulies,E.M. Shevach, and W. Strober, eds.) pp. A.4.1-A.4.28. Greene Publishing and Wiley-Inter-science, New York.

Papadopoulos, N.M., Elin, R.J., and Wilson, D.M.1982. Incidence of γ banding in a healthy popu-lation by high-resolution immunofixation elec-trophoresis. Clin. Chem. 28:707-708.

Tiselius, A. 1937. A new apparatus for electro-phoretic analysis of colloidal mixtures. Trans.Faraday Soc. 33:524-526.

Key ReferencesJohnson, 1986. See above.

A concise discussion of the principles of immuno-precipitation with specific reference to immunofixa-tion electrophoresis.

Maggio, E.T. 1981. Enzyme Immunoassay. CRCPress, Boca Raton, Fla.

A valuable reference describing parameters ofELISA technology.

Contributed by Peter Hornbeck (sandwich ELISA)University of MarylandBaltimore, Maryland

Thomas A. Fleisher and Nicholas M. Papadopoulos (electrophoresis and immunofixation)Warren Grant Magnuson Clinical CenterBethesda, Maryland


11.3.6


of Antibodies

SECTION IIPREPARATION OF MONOCLONALANTIBODIESThe preparation of monoclonal antibodies should be undertaken carefully, since theproduction of monoclonal antibodies is expensive and time-consuming. Figure 11.4.1summarizes the experimental procedures that must be carried out to prepare monoclonalantibodies. The various procedures are presented in individual protocols and described insufficient detail to allow an individual with no prior experience to carry out a cell fusion,to produce monoclonal antibodies in ascites fluid, and to purify antigen-specific mono-clonal antibodies.

isolation of spleen cells preparation of myeloma cells

feeder cells

cell fusion

ELISAscreening

of hybridomasupernatants

expansion and selection of cultures to be cloned

feeder cells freeze cells

cloning by limiting dilution

isolation and expansion of clones

freeze cells

recovery of frozen cells

production of ascites fluids

purification of monoclonal antibodies

ELISAscreening

of sera

immunization of mice

Figure 11.4.1 Flow chart for preparation of monoclonal antibodies.

Supplement 18

Contributed by Steven A. Fuller, Miyoko Takahashi, and John G.R. HurrellCurrent Protocols in Molecular Biology (1992) 11.4.1-11.4.6Copyright © 2000 by John Wiley & Sons, Inc.

11.4.1

Immunology

UNIT 11.4 Immunization of MiceAntigen is prepared for injection either by emulsifying an antigen solution with Freundsadjuvant or by homogenizing a polyacrylamide gel slice containing the protein antigen.Mice are immunized at 2- to 3-week intervals. Test bleeds are collected 7 days after eachbooster immunization to monitor serum antibody levels. Mice are chosen for hybridomafusions when a sufficient antibody titer is reached.

BASICPROTOCOL

PRODUCTION OF IMMUNE SPLEEN CELLS: IMMUNIZATION WITHSOLUBLE ANTIGEN

Materials

Phosphate-buffered saline (PBS; APPENDIX 2)AntigenComplete Freunds adjuvantAny strain mice, 6 to 8 weeks oldIncomplete Freunds adjuvant

22-G needles3-ml syringes with locking hubs (Luer-Lok, Becton Dickinson)Double-ended locking hub connector (Luer-Lok, Becton Dickinson)Sterile sharp scissorsSterile razor blades or scalpel bladesWooden applicator sticks200-µl pipettor

Additional reagents and equipment for ELISA (UNITS 11.2 & 11.3) and western blotting (optional; UNIT 10.8)

1. Prepare an emulsion (200 to 400 µl/mouse) of equal volumes PBS containing 25 to100 µg antigen and complete Freunds adjuvant. Using a 22-G needle, inject miceintraperitoneally. For each antigen, 3 to 5 mice are immunized.

Complete Freunds adjuvant contains mycobacteria—incomplete Freunds does not.

An emulsion is most readily prepared by linking two locking syringes, one loaded withantigen and the other loaded with adjuvant, using a double-ended locking connector (seeFig. 11.4.2). Press syringe barrels back and forth, transferring contents from one syringeto the other, for 5 to 10 min until a stable emulsion is produced. For best antibodyproduction, inject antigen in as small an emulsion volume as practicable.

A stable emulsion is an oil-in-water emulsion which will not disperse when dropped intowater. This is a useful check for the emulsification endpoint. Further, at the endpoint theemulsion will thicken noticeably.

Mice may be restrained for immunization in the following manner: Place mouse on grilledcage top. Lift mouse by the tail (generally, when mice are lifted by the tail they will grabthe bars of the cage top with their front feet, thus stabilizing themselves for restraint).Immobilize the mouse’s head by pinching together the skin at the base of the skull betweenthumb and forefinger. Turn hand over so that mouse is lying with its back against the palm.Wrap fourth finger around tail and stretch mouse over arched palm for intraperitonealinjection.

CAUTION: Handle Freunds adjuvant carefully, since self-injection can cause a positiveTB test and lead to a granulomatous reaction.

2. Boost mice 3 weeks later by intraperitoneally injecting an emulsion (200 to 400 µl)of equal volumes PBS containing 10 to 50 µg antigen and incomplete Freundsadjuvant. The emulsion is prepared and injected as in step 1.


11.4.2

Immunizationof Mice

3. Bleed mice 7 days after second immunization by cutting off 0.5 cm of the tail withsterile sharp scissors or a razor blade. Collect 100 to 200 µl blood into a 1.5-mlmicrocentrifuge tube. After clot formation, rim the clot with a wooden applicatorstick to dislodge the clot from the surface of the tube, but do not break up the clot.After clot retraction, transfer the serum into another microcentrifuge tube with a200-µl pipettor. If test bleeds are collected more than three times, it will be necessaryto cut the tail vein to obtain further samples rather than cutting off additional lengthsof the tail itself. This is done by nicking one of the lateral tail veins with a razor blade.

The collection of blood may be facilitated by using a heat lamp to warm the mouse for 30sec to 1 min prior to cutting of the tail. Additionally, if blood flow from the cut tail is slow,the tail may be “milked” from base to the cut tip with thumb and forefinger.

4. Determine the antibody titer in the serum by ELISA (UNITS 11.2 & 11.3). If desired,further characterize the antibody specificity by western blotting (UNIT 10.8).

Antibody titer is operationally defined as that dilution of serum that results in 0.2absorbance units above background in the ELISA procedure.

5. If the antibody titer is considered too low (≥1⁄1000) for cell fusion, mice can be boostedevery 2 weeks until an adequate response is achieved. Bleed the mice and test theserum with an ELISA.

6. When the antibody titer is sufficient (>1⁄1000), boost mice by injecting 10 to 50 µgantigen in PBS intraperitoneally (200 to 400 µl), or intravenously (50 to 100 µl) viathe tail veins, 3 days before fusion but >2 weeks after previous immunization.

In general, the higher the serum antibody titer, the more antigen-specific antibody-produc-ing hybridomas are obtained per fusion.

If an antibody against a nonimmunodominant epitope is desired, the cell fusion may bedone at an earlier or later time, since the percentage of antibody-producing cells in thespleen directed at these less immunogenic regions of the antigen may vary with time in anunpredictable fashion.

7. Perform cell fusion (UNIT 11.7) 3 days after the immunization (step 6).

3-ml glass syringes

double-endedLuer-Lok connectoror 3-way stopcock

antigen and adjuvant

Figure 11.4.2 Double-syringe device for preparation of antigen-adjuvant emulsions.


11.4.3

Immunology

ALTERNATEPROTOCOL

IMMUNIZATION WITH COMPLEX ANTIGENS (MEMBRANES, WHOLECELLS, AND MICROORGANISMS)

1. Prime the mice and boost intraperitoneally with adjuvant (i.e., complete Freunds forpriming and incomplete Freunds for booster immunizations) as described for solubleantigen (see basic protocol, steps 1 and 2) or suspend antigen in PBS and inject. Use1 to 2 × 107 cells for mammalian species or 108 to 109 bacterial or yeast cells.

2. Bleed the mice and determine the antibody titer of the serum as described for solubleantigen (see basic protocol, steps 3 to 6).

3. Perform cell fusion (UNIT 11.7) 3 days after final immunization.

ALTERNATEPROTOCOL

IMMUNIZATION WITH ANTIGEN ISOLATED BY ELECTROPHORESIS

In some instances the antigen under investigation can be purified most conveniently bygel electrophoresis (UNIT 10.2). Mice can be immunized with protein antigens still containedin a polyacrylamide gel slice, as described in this protocol.


0.1 M KCl, coldTissue grinder

Additional reagents and equipment for denaturing (SDS) discontinuous gel electrophoresis (UNIT 10.2)

1. Apply a protein mixture containing 10 to 50 µg of the desired protein antigen to anappropriate denaturing (SDS) discontinuous gel electrophoresis system (e.g., theLaemmli gel system) and complete the electrophoresis as described in UNIT 10.2.

2. Soak gel 5 to 15 min in cold 0.1 M KCl. Protein bands will appear as white precipitatesagainst a clear gel background.

3. Cut out the appropriate bands from the gel with a razor blade or scalpel blade.

4. Prepare gel suspension by homogenizing the gel slice in a minimum volume of PBSusing a tissue grinder. Minimum volume is defined by adding successive 100-µlvolumes of PBS until the homogenized gel is liquid.

Alternatively, the gel may be air dried for 1 to 2 hr, smashed with a glass rod, and suspendedin a minimum volume of PBS.

5. Immunize each mouse with 200 to 400 µl gel suspension containing 10 to 50 µgantigen via an intraperitoneal injection.

Amount of antigen is estimated from prior observation of the proportion of desired proteinantigen to other antigens in the sample as determined by the relative intensity of stainedbands on the polyacrylamide gel (see UNIT 10.6 for staining procedures).

6. Boost mice after 3 weeks with 200 to 400 µl gel suspension containing 10 to 25 µgantigen.

7. Bleed the mice and determine the antibody titer of the serum as described for solubleantigen (see basic protocol).

Mice immunized repeatedly with polyacrylamide tend to form adhesions that can makeaseptic removal of the spleen difficult.

8. Perform cell fusion (UNIT 11.7) 3 days after final immunization.


11.4.4

Immunizationof Mice

COMMENTARY

Background InformationThe stimulation of an effective humoral im-

mune response in mice is critical to the produc-tion of monoclonal antibodies directed at aparticular antigen. The variety and quality ofthe monoclonal antibodies prepared is gener-ally directly proportional to the serum antibodytiter in the particular mouse used for cell fusion.Any means of antigen preparation, antigen de-livery, or immunization schedule that increasesantibody titer in the serum of the immunizedmouse will potentiate the isolation of hybrido-mas secreting monoclonal antibodies of inter-est. We have described two methods of antigenpreparation: (1) antigen emulsified in Freundsadjuvants (probably the most common tech-nique used) and (2) antigen isolated in apolyacrylamide gel slice and homogenized.Other preparation methods (e.g., adsorption ofantigen to supports such as aluminum hydrox-ide or aluminum phosphate, polystyrene beads,or nitrocellulose paper, and alternate sites ofinjection such as footpads) are discussed in thekey references.

Critical ParametersIt is desirable to use antigen of the highest

available purity for immunizations, particu-larly for primary immunizations. Contami-nants may be more immunogenic than the an-tigen of interest and as such may result in a lowspecificity antibody. Mice given primary im-munizations of highly pure antigen may beboosted with less pure material (containing aslittle as one-third specific antigen in a complexprotein mixture).

TroubleshootingPoor success in raising an adequate antibody

titer to an antigen of interest can be attributedto several factors. Improperly prepared emul-sion when using Freunds adjuvant (i.e., theaqueous and oil phases separate upon standing)is ineffective in stimulation of an immune re-sponse. Contaminants in an antigen preparationmay be more immunogenic, necessitating amore homogeneous preparation of the desiredantigen. Other parameters that can be varied inan effort to produce a higher antibody titer andincreased specificity include presentation ofantigen (Freunds adjuvant emulsion versus

polyacrylamide gel slice), site of immunization(intraperitoneal versus footpad or tail vein),antigen dose, and frequency of immunization.Alternate immunization protocols are pre-sented in the key references below.

Anticipated ResultsIsolation of high-quality monoclonal anti-

bodies correlates with high-serum antibody tit-ers. A serum ELISA titer of 1⁄1000 is the mini-mum level before attempting a cell fusion.Titers for most antigens (particularly from ani-mals injected with highly purified antigen) willrange from 1⁄1000 to 1⁄100,000 after 3 to 4 immu-nizations. Occasional serum samples will titerat greater than 106. The proportion of mono-clonal antibodies of IgG class rather than IgMclass generally increases proportionally to theduration of the immunization schedule, al-though this can vary dramatically among dif-ferent antigens. [In general, IgG class antibod-ies are more suitable for immunoassays, west-ern blott ing (UNIT 10.8), immunoaffinitychromatography (UNIT 10.11), and immunopre-cipitation (UNIT 10.16)].

Time ConsiderationsA primary immunization followed by two

booster immunizations and test bleeds will oc-cupy 6 weeks. For many antigens, however, anadequate antibody response in the mice isachieved only after several months and multipleimmunizations.

Key ReferencesHurrell, J.G.R., ed. 1982. Monoclonal Hybridoma

Antibodies: Techniques and Applications. CRCPress, Boca Raton, Fla.

Langone, J.J. and Van Vunakis, H., eds. 1986. Im-munological techniques, Part I: Hybridoma tech-nology and monoclonal antibodies. Methods En-zymol. 121:1-947.

Contributed by Steven A. Fuller and Miyoko TakahashiADI DiagnosticsRexdale, Ontario

John G.R. HurrellBoehringer Mannheim DiagnosticsIndianapolis, Indiana


11.4.5

Immunology

UNIT 11.5 Preparation of Myeloma Cells

BASICPROTOCOL

Myeloma cells are cultured with 8-azaguanine to ensure their sensitivity to the HATselection medium (see UNIT 11.6) used after cell fusion (UNIT 11.7). One week prior to cellfusion, myeloma cells are grown in medium without 8-azaguanine. Cell culture conditionsare adjusted such that the Sp2/0 cells are in the log phase of growth and exhibit highviability at the time of collection for fusion (UNIT 11.7).

Materials

Sp2/0 murine myeloma cell line (ATCC #CRL 1581)Complete culture medium20 µg/ml 8-azaguanine

Tissue culture flasks, 25 cm2 or 75 cm2

8% CO2-in-air gas mixtureHumidified 37°C, 8% CO2 incubatorInverted microscope

1. Recover frozen cells from liquid N2 storage, as described in UNIT 11.9.

2. Grow Sp2/0 cells overnight in complete medium in tissue culture flasks at 37°C in aCO2 incubator in 8% CO2-in-air atmosphere with 98% relative humidity.

3. Determine that the cells are growing by examining the cell cultures in the flasks withan inverted microscope and return culture flask to CO2 incubator for continuation ofcell growth.

4. To ensure that the Sp2/0 cells remain aminopterin sensitive for the selection processfollowing fusion, supplement the complete culture medium with 8-azaguanine at 20µg/ml during maintenance. One week prior to fusion, culture cells in medium without8-azaguanine.

A seeding cell density of 2.5 to 5 × 104 cells/ml works well with Sp2/0 cells.

Sp2/0 cells will grow to a maximum density of 6 to 9 × 105 cells/ml, with a doubling timeof 10 to 15 hr. When this density is reached, there is a rapid decline in cell viability. TheSp2/0 cultures are split every 2 to 3 days either by discarding an appropriate volume fromthe old flask and replacing with fresh medium or by transferring an appropriate volume ofcells to a new flask and adding fresh medium. A 1-in-10 or 1-in-20 split is recommended.

5. A total of 1 × 107 Sp2/0 cells (i.e., 1:10 ratio to immune spleen cells) is used forfusion. Cell viability at the time of collection should be greater than 95%. To ensurethat cells are collected in log phase of growth, adjust the cell density to 2 × 105

cells/ml the day before the fusion by adding fresh medium. Determine cell viabilityusing the trypan blue exclusion method (see support protocol, below) on cellssuspended in serum-free medium or PBS.

SUPPORTPROTOCOL

CELL VIABILITY TEST BY TRYPAN BLUE EXCLUSION

This procedure is used to determine the number of viable cells present in the cell culture.A non-viable cell will have a blue cytoplasm; a viable cell will have a clear cytoplasm.


Phosphate-buffered saline (PBS; APPENDIX 2) or serum-free complete culture medium

0.4% trypan blue solution

Contributed by Steven A. Fuller, Miyoko Takahashi, and John G.R. HurrellCurrent Protocols in Molecular Biology (1988) 11.5.1-11.5.3Copyright © 2000 by John Wiley & Sons, Inc.Supplement 18

11.5.1

Preparationof Myeloma Cells

Binocular microscopeHemacytometer

1. Centrifuge 1 ml cell suspension at 100 × g for 5 min.

2. Resuspend the cell pellet in 1 ml PBS or serum-free complete culture medium.

Serum proteins stain with trypan blue and can produce misleading results. Determinationsmust be made in serum-free solution.

3. Mix 1 part of trypan blue solution and 1 part cell suspension (1⁄2 dilution).

4. Using a binocular microscope, count the unstained (viable) and stained (dead) cellsseparately in a hemacytometer. Each of the four corner squares (composed them-selves of 16 smaller squares) have 1 mm sides and are 0.1 mm deep (0.1 mm3). Countall cells within each of the four corner squares, including those that lie on the bottomand left-hand perimeters, but not those that lie on the top and right-hand perimeters.Count any clumps of cells as one cell. Calculate the mean number of cells per 0.1-mm3

volume. Multiply by 104 to obtain the number of cells/ml (i.e., cells/cm3). Applydilution factor for trypan blue (2×) to obtain the number of cells per milliliter ofculture.

5. Calculate the percentage of viable cells as follows:

Viable cells (%) = Number of viable cells

Total number of cells (dead and viable) × 100


Complete culture medium

Dulbecco modified Eagle medium (DMEM), high-glucose formula (4.5 g glucose/liter; GIBCO/BRL #430-2100) supplemented to the indicated concentrations withthe following additives: 2.8 g/liter sodium bicarbonate (33.3 mM) 4.8 g/liter HEPES (20 mM) 10% fetal calf serum (v/v) 10 ml/liter L-glutamine (2 mM) 10 ml/liter sodium pyruvate (1 mM) 10 ml/liter penicillin (50 IU/ml) and streptomycin (50 µg/ml)

The last four additives are available as 100× solutions from GIBCO/BRL and othermajor suppliers of cell culture media. Penicillin and streptomycin are combined inone solution.

Samples of fetal calf serum lots should be tested for ability to support efficient cellgrowth and cloning before a large purchase because there is much variabilitybetween lots of a given supplier. The fetal calf serum must be mycoplasma free. Iflow volume usage of fetal calf serum precludes testing of serum lots, purchase ofmycoplasma-free, virus-free, low endotoxin sera from suppliers such asGIBCO/BRL, Flow Laboratories, or Sigma will generally provide satisfactoryresults. Horse or bovine serum is not an adequate substitute!


11.5.2

Immunology

COMMENTARY

Background InformationThe Sp2/0 cell line was chosen as the fusion

partner for immune spleen cells because of itsgood rate of growth, the efficiency with whichhybridomas are obtained after fusion, and, mostimportantly, because it does not synthesize orsecrete any immunoglobulin heavy or lightchains itself. The Sp2/0 myeloma cell line wasdeveloped by Schulman et al. (1978). Othercommonly used cell lines are P3X63-Ag8.653(Kearney et al., 1979), which does not secreteimmunoglobulins, and NS-1 (Kohler and Mil-stein, 1976), which produces only κ lightchains.

Critical ParametersOptimal growth of myeloma cells is density

dependent. Cultures should be split at regularintervals to maintain >95% viability. Do notculture Sp2/0 cells longer than 1 month to avoidgenetic drift and development of antibiotic-re-sistant contaminants. Maintain several aliquotsof Sp2/0 cells in liquid nitrogen storage.

Anticipated ResultsProper care yields a healthy log phase

myeloma cell culture able to sustain good pro-duction of hybridomas upon fusion.

Time ConsiderationsDepending on culture conditions, 105 cells

can be expanded to the 107 cells required forfusion in 4 to 6 days.

Literature CitedKearney, J.F., Radbruch, A., Liesegang, B., and Ra-

jewsky, K. 1979. A new mouse myeloma cell linethat has lost immunoglobulin expression but per-mits the construction of antibody-secreting hy-brid cell lines. J. Immunol. 123:1548-1550.

Kohler, G. and Milstein, C. 1976. Fusion betweenimmunoglobulin-secreting and nonsecretingmyeloma cell lines. Eur. J. Immunol. 6:511-519.

Schulman, M., Wilde, C.D., and Kohler, G. 1978. Abetter cell line for making hybridomas secretingspecific antibodies. Nature 276:269-270.

Contributed by Steven A. Fuller, Miyoko Takahashi, and John G.R. HurrellAllelix Inc.Mississauga, Ontario


11.5.3

Preparationof Myeloma Cells

UNIT 11.6Preparation of Mouse Feeder Cells forFusion and Cloning

BASICPROTOCOL

Chilled sucrose solution is injected intraperitoneally into mice. When withdrawn, thesolution contains feeder cells (macrophages and other cells) that are placed in the wellsof microtiter plates 1 day prior to seeding of hybridomas from cell fusion (UNIT 11.7) orcloning (UNIT 11.8) procedures.

Materials

0.34 M sucrose solution, sterile and chilledMice (any strain)70% ethanolHAT medium, chilledSterile phosphate-buffered saline (PBS; APPENDIX 2)

10-ml syringe, sterile18-G needle, sterile50-ml conical centrifuge tube, sterileDissecting boardForceps, sterileScissors, sterile96-well microtiter plates8% CO2-in-air gas mixtureHumidified CO2 incubator

Additional reagents and equipment for estimating cell viability by trypan blueexclusion (support protocol, UNIT 11.5)

1. Just prior to sacrificing a mouse, fill a 10-ml syringe with 8 ml chilled sucrose solutionand attach 18-G needle.

To avoid macrophages adhering to plastic surfaces, it is important to use chilled solutionsto optimize cell harvest.

2. Chill the 50-ml conical centrifuge tube in ice.

3. Kill mouse by cervical dislocation.

This is accomplished by firmly holding a thick pencil or similar rod-shaped object to theneck of the mouse just behind the skull and quickly and firmly pulling the tail.

4. Immerse the mouse in a 100-ml beaker containing 70% ethanol.

5. Lay out mouse on dissecting board.

6. Snip skin at diaphragm level and pull skin back, exposing the lower part of the ribcage and abdomen.

With forceps pull skin from underlying tissue at the diaphragm level and snip with ascissors. With forceps or sterile gloved hands, pull skin back at both sides of the incisionto expose the lower part of the rib cage and abdomen.

Care must be taken not to tear or cut the peritoneal membrane.

7. Insert the needle into the peritoneal cavity at the base of the sternum and rest the tipof the needle over the liver. Inject sucrose solution. Gently squeeze the abdomen twoor three times.

8. Harvest the peritoneal feeder cells by withdrawing as much solution as possible intothe syringe.

Care must be taken not to puncture the digestive organs, which may lead to fecalcontamination of the feeder cells.

Supplement 1


11.6.1

Immunology

Enough peritoneal feeder cells can usually be isolated from one mouse to seed ∼100 to 300wells. However, some mice do not yield effective feeder cells. Depending on the totalnumber of wells that must be seeded with mouse feeder cells, an appropriate number ofmice must be killed. Peritoneal exudate feeder cells can be prepared up to 3 days prior touse.

9. Transfer the feeder cell–containing sucrose solution into the 50-ml centrifuge tube.

10. In a sterile fume hood, add 20 ml chilled HAT medium.

11. Centrifuge at 100 × g for 5 min at room temperature.

12. Resuspend the pellet in 1 ml chilled HAT medium and perform cell viability test bytrypan blue exclusion as described in the support protocol, UNIT 11.5.

13. Suspend the cell pellet in chilled HAT medium at 1 × 105 cells/ml.

14. Add 100 µl cell suspension to each of the 60 inner wells of the 96-well plates. Theperipheral 36 wells are filled with sterile PBS.

Plates having 24 wells may be used. If this is the case, add 1 ml cell suspension/well.

15. Incubate plates overnight at 37°C in a CO2 incubator in 8% CO2-in-air with 98%relative humidity.


The following solutions are sterilized by filtration through a 0.22-µm membrane. Asuitable sterilization system is a disposable filter unit (e.g., Nalgene #120-0020). Glass-distilled water should be used for all preparations.

0.34 M sucrose solution58.2 g sucroseH2O to 500 mlFilter sterilize and store at 4°C in 100-ml aliquots

HAT (hypoxanthine/aminopterin/thymidine) mediumComplete culture medium (see reagents and solutions, UNIT 11.5) supplemented tothe indicated concentrations with the following additives: 20% (v/v) fetal calf serum 0.1 mM nonessential amino acids 100 µM hypoxanthine 0.4 µM aminopterin 16 µM thymidineThese additives may be purchased in concentrated and sterile solutions from themajor suppliers of cell culture media and reagents. Concentrated solutions ofhypoxanthine and thymidine (HT) and aminopterin may also be prepared in thelaboratory (see following recipes).

100× HT solutionWeigh 340.3 mg hypoxanthine and 96.9 mg thymidine; add water to 250 ml. Heatto 70°C to dissolve. Filter sterilize and store in 20-ml aliquots at −20°C. Thaw at70°C for 10 to 15 min.

1000× aminopterin solutionWeigh 17.6 mg aminopterin. Add 60 ml water and dissolve by adding 0.1 M NaOHdropwise. Titrate with HCl to pH ∼8.5. Adjust volume to 100 ml and filter sterilize.Make 100× working solution by diluting stock in complete culture medium. Storein 5-ml aliquots at −20°C.

Aminopterin precipitates at low pH and is light sensitive.


11.6.2

Preparation ofMouse Feeder

Cells for Fusionand Cloning

COMMENTARY

Background InformationTo maximize the yield of hybrids from the

fusion and cloning procedures, feeder cells arerequired to be cocultured with the hybrids,while hybrid cell density is low. Mouse perito-neal cells, most of which are macrophages,have been found to be convenient and effectivefeeder cells which are a source of solublegrowth factors for hybridoma cells.

Critical ParametersFeeder cells such as peritoneal cells provide

best support of hybridoma growth when used1 to 3 days after harvest. Use of chilledsolutions is necessary for optimum cell harvest,to prevent macrophages adhering to plastic sur-faces.

Anticipated ResultsFrom 1 to 3 × 106 peritoneal feeder cells are

harvested from one mouse. The number offeeder cells will be enough to seed 100 to 300wells.

Time ConsiderationsPeritoneal feeder cells from one mouse can

be processed in 1 hr or less.



11.6.3

Immunology

UNIT 11.7 Fusion of Myeloma Cells with ImmuneSpleen Cells

BASICPROTOCOL

Freshly harvested spleen cells and myeloma cells are copelleted by centrifugation andfused by addition of polyethylene glycol solution to the pellet. Cells are centrifuged againand the PEG solution diluted by slow addition of medium. Fused cells are centrifuged,resuspended in selection medium, and aliquoted into 96-well microtiter plates. Hybrido-mas are grown to 10 to 50% confluence and then assayed for production of antigen-spe-cific antibody.

Materials

Any strain immunized mouse (UNIT 11.3)Sp2/0 murine myeloma cells in active log phase (Am. Type Culture

Collection #CRL 1581; UNIT 11.5)Diethyl ether70% ethanolDulbecco modified Eagle medium (DMEM) with supplementsSterile polyethylene glycol (PEG) solutionHAT medium (UNIT 11.6)HT mediumPhosphate-buffered saline (PBS; APPENDIX 2)

15- and 50-ml centrifuge tubesGlass desiccator or metal can with lidDissecting board10.5-cm scissors (Irex #IR-105), sterile10.5-cm forceps (Irex #IR-1393), sterile60- and 100-mm petri dishesStainless-steel strainer (Cellector; GIBCO #1985-8500), sterile3-cc glass syringes with 26-G needle5-ml serological pipets37° C water bathStopwatch8% CO2-in-air gas mixtureHumidified CO2 incubatorPolyvinyl or polystyrene 96-well microtiter platesInverted microscope

Additional reagents and equipment for estimating cell viability by trypan blue exclusion (UNIT 11.5) and for detection of antibodies (UNIT 11.4)

Preparation of myeloma and spleen cells1. Just prior to sacrificing the mouse, transfer 1 × 107 Sp2/0 murine myeloma cells

(prepared as described in UNIT 11.5) to a 50-ml centrifuge tube. Check the percentageof viable cells using the trypan blue exclusion method (support protocol, UNIT 11.5).

2. Sacrifice the mouse by anesthetizing with diethyl ether in a closed container (e.g., aglass desiccator or a metal can with a lid).

At this point, a blood sample may be collected from the mouse by severing the blood vesselsof one forelimb. Collect the blood with a Pasteur pipet and place the blood in a microcen-trifuge tube.

3. Immerse the mouse in a beaker containing 70% ethanol and lay out on a dissectingboard.


11.7.1

Fusion of MyelomaCells with Immune

Spleen Cells

4. Using sterile forceps, lift skin over the thorax area and snip with sterile scissors. Peelskin over both sides to expose left side of the rib cage.

5. Using another set of sterile forceps and scissors, remove the spleen from the left upperabdomen of the mouse. The spleen is a small dark red organ.

6. Place the spleen in a 60-mm petri dish containing 3 ml supplemented DMEM.

7. Take spleen in petri dish to sterile hood and carefully dissect away surface fat andother adhering tissue by using a sterile forceps and scissors.

8. Transfer spleen to sterilized, stainless-steel strainer in 100-mm petri dish with 10 mlDMEM.

9. Fill the 3-cc syringe with 2 ml supplemented DMEM. Using a 26-G needle, fill thespleen with DMEM by injecting at several sites.

10. With sterile scissors, cut the supplemented DMEM–filled spleen in 3 or 4 places.

11. Using circular movements, press the spleen against the screen of the stainless-steelstrainer with the glass syringe plunger of the 3-cc syringe until only fibrous tissueremains on top of the strainer screen. The tissue that is forced through the strainer iscollected in a sterile petri dish underneath.

12. Rinse the screen with 2 ml supplemented DMEM.

13. Transfer the suspension of spleen cells to a 15-ml centrifuge tube. Using a 5-mlserological pipet, disperse the clumps by drawing up and expelling several times.

14. Let suspension stand for 3 min at room temperature.

15. Transfer the top 95% of the cell suspension to a 15-ml centrifuge tube.

16. Perform a viable cell count using the trypan blue exclusion procedure (supportprotocol, UNIT 11.5); record the percentage of viable cells.

Ignore red blood cells that are substantially smaller than the nucleated cells.

17. Transfer 1 × 108 viable spleen cells into a 15-ml centrifuge tube.

18. Wash myeloma cells (from step 1) twice with supplemented DMEM followed bycentrifugation at 200 × g for 5 min and resuspend the cells in 5 ml supplementedDMEM.

19. Add spleen cells to myeloma cells in the 50-ml tube and fill the tube with DMEM.

20. Centrifuge the suspension at 200 × g for 5 min at room temperature.

21. Resuspend the cell pellet in 50 ml supplemented DMEM and centrifuge as above.

22. Warm the cell pellet by placing the tube in a 37°C water bath in a beaker for 2 min.

23. Loosen the pellet by flicking the tip of the tube gently.

Cell fusion24. Fuse spleen and Sp2/0 cells with sterile PEG solution:

Over the first 1 min—Add 1 ml PEG solution at 37°C. Mix gently.Over the next 2 min—Spin at 100 × g (2 min, total time).Over the next 3 min—Add 4.5 ml supplemented DMEM.Over the next 2 min—Add 5 ml supplemented DMEM.

Fill the tube with supplemented DMEM.


11.7.2

Immunology

Timing is critical and should be monitored with a stopwatch.

25. Centrifuge at 100 × g for 5 min at room temperature.

26. Aspirate supernatant.

27. Resuspend the cell pellet in 35 ml HAT medium.

Do not force the dispersion of small cell clumps.

28. Incubate cell suspension at 37°C in a CO2 incubator in 8% CO2-in-air with 98%relative humidity for a minimum of 30 min.

Plating and culture of fused cells29. Add 100 µl cell suspension to each of the 60 inner wells of six 96-well plates.

Peripheral 36 wells are filled with sterile PBS.

Twenty-four (24) hr prior to use, the 60 inner wells are conditioned with 1 × 104 mouseperitoneal macrophages per well in 100 �l HAT medium.

30. Incubate plates at 37°C in CO2 incubator in 8% CO2-in-air with 98% relativehumidity. (This is day 1 of the culture).

31. On day 5 of the culture, add 100 µl HAT medium to each well.

32. On day 7 of the culture, remove 100 µl from each well and add 100 µl fresh HATmedium.

33. Repeat step 32 every other day until hybrid cell growth covers 10% to 50% of thesurface area of the wells. This is monitored by examining the bottom of the wellswith an inverted microscope. At this time, the wells should be screened for antibody(UNIT 11.4).

34. Grow the hybrids in HAT medium for 2 weeks after fusion.

35. After 2 weeks, change the medium to HT medium.

36. The hybrids are grown in the HT medium until the completion of two cloningprocedures (as described in the protocol for cloning of hybridoma cells by limitingdilution, UNIT 11.8).


Dulbecco modified Eagle medium (DMEM), high-glucose formula (GIBCO/BRL#430- 2100), supplemented to the indicated concentrations with the followingadditives:

2.8 g/liter sodium bicarbonate (33.3 mM)4.8 g/liter HEPES (20 mM)

HT medium

Prepare as described in UNIT 11.6 for HAT medium but without aminopterin solution.

PEG (polyethylene glycol) solution

Weigh 10 g PEG 4000 (Merck) into a 100-ml glass bottle. Autoclave PEG for 10min at 121°C. Cool the molten PEG to 50°C. Mix with 10 ml prewarmed supple-mented DMEM (50°C) that contains 5% (v/v) dimethylsulfoxide (DMSO). Aliquotthe mixture into small glass bottles (3 ml/bottle). Store at 4°C in the dark.

PEG 4000, Merck’s gas chromatography grade, appears to be the best PEG for cell fusionsusing Sp2/0 myeloma cells regardless of lot number. Remove PEG from the autoclave as soonas the pressure is down in order to avoid prolonged heating of PEG. Incubate PEG solutionat 37°C for 24 hr prior to use to test its sterility.


11.7.3

Fusion of MyelomaCells with Immune

Spleen Cells

Alternatively, if the temperature control of the autoclave is uncertain, melt the PEG at 65°Con the day of fusion. Mix the PEG with supplemented DMEM to make a 45% PEG solutionand sterilize by filtration through a 0.22-�m membrane filter.

COMMENTARY

Background InformationMurine spleen cells, some of which are in-

volved in production of the desired antibodies,are fused with a murine myeloma cell line toform a stable antibody-producing hybridomacell line. The myeloma cells (Sp2/0) are hypox-anthine–guanine phosphoribosyltransferasedeficient (HGPRT−) and therefore are unable touse the purine salvage pathway when de novopurine synthesis is blocked by aminopterin,which is included in the HAT selection me-dium. See discussion of selectable markers inUNIT 9.5. Hypoxanthine and thymidine in theHAT selection medium allow HGPRT+ cells,the spleen cell–myeloma hybrids, to surviveand grow. Unfused spleen cells eventually die.

Literature ReviewKohler and Milstein (1975) first demon-

strated that somatic cell fusion could be usedto generate a hybridoma cell line producing amonoclonal antibody of predetermined speci-ficity. Cell fusion was initially accomplishedby addition of Sendai virus. The fusion proce-dure used in this protocol is a modification ofthe method of Gefter et al. (1977), which usespolyethylene glycol (PEG) as the fusogen. Asimilar procedure is presented by Oi and Her-zenberg (1980). Each protocol encompassescareful timing of the PEG addition to the cellpellet and its subsequent dilution after fusion.

Critical ParametersThe choice and use of PEG in the fusion

protocol is the most critical factor. PEG canvary dramatically in efficiency between manu-facturers and among lots of a particular manu-facturer. The suggested source (Merck) hasprovided the most consistent lots of PEG. Caremust be taken to autoclave the PEG at 121°Cfor only 10 min in order to minimize the pro-duction of toxic aldehydes. In the fusion pro-cedure itself, it is important to adhere carefullyto the time schedule established. Extended in-cubation of the cells with PEG results in de-creased cell viability. Dilution of the PEG bymedium must be done carefully to avoid lysisof the cells.

TroubleshootingWhen poor fusion results (i.e., poor hybri-

doma growth) are obtained despite close adher-ence to the protocol, the likely causes are thePEG, the myeloma cells used as fusion partner,or the culture conditions. Most critical is thePEG (see critical parameters). The quality ofthe fusion partner (i.e., the myeloma culture;see UNIT 11.5 for details) is also very important.The CO2 incubator must provide a stable tem-perature, pH, and humidity for optimal hybri-doma growth.

Anticipated ResultsHybridoma growth should be observed with

the aid of an inverted microscope in nearly allwells after a few days in culture. For wells withviable cells, generally 10 to 50% will containantigen-specific antibody. Results can varywidely, of course, and it has been observed that0 to 99% of the supernatants in the wells willcontain specific antibody.

Time ConsiderationsThe fusion procedure requires 3 to 4 hr to

complete. For best results, it should be accom-plished without interruption. Wells with hybri-doma growth can be assayed for specific anti-body 7 to 12 days after fusion.

Literature CitedGefter, M.L., Margulies, D.H., and Scharff, M.D.

1977. A simple method for polyethylene glycol-promoted hybridization of mouse myelomacells. Somat. Cell Genet. 3:231-236.

Kohler, G. and Milstein, C. 1975. Continuous cul-tures of fused cells secreting antibody of prede-fined specificity. Nature 256:495-497.

Oi, V.T. and Herzenberg, L.A. 1980. Immunoglobu-lin-producing hybrid cell lines. In SelectedMethods in Cellular Immunology (B.B. Mishelland S.M. Shiigi, eds.) pp. 351-372. W.H. Free-man, San Francisco.



11.7.4

Immunology

UNIT 11.8 Cloning of Hybridoma Cell Lines byLimiting Dilution

BASICPROTOCOL

Hybridomas to be cloned are diluted to 0.8 cells/well. This dilution provides 36% of wellswith 1 cell/well by Poisson statistics. When cultures are 10 to 50% confluent, antibody isassayed by ELISA. Two or more cloning procedures are carried out until >90% of thewells containing single clones are positive for antibody production.

Materials

HAT medium (UNIT 11.6)HT medium (UNIT 11.7)Complete culture medium (UNIT 11.5)

Polyvinyl or polystyrene 96-well microtiter plates6- and 24-well culture plates8% CO2-in-air gas mixtureHumidified CO2 incubatorCryotubes (Nunc #3-63401)2-, 5-, and 10-ml serological pipetsMultichannel pipet and tipsInverted microscope

Additional reagents and equipment for preparing mouse feeder cells for fusion and cloning (UNIT 11.6), for ELISA screening (UNIT 11.4), and for estimating cellviability by trypan blue exclusion (UNIT 11.5)

1. The day before cloning, isolate mouse feeder cells and prepare 96-well plates withfeeder cells in appropriate medium, as described in UNIT 11.6.

Choice of medium depends on the current stage of the cloning process. The first cloninguses HAT medium, and the second cloning uses HT medium. Any other clonings usecomplete culture medium.

2. Transfer all of the cells from each well containing antigen-specific antibody in itshybridoma supernatant (as determined by an ELISA, UNIT 11.4) into a separate well ofa 24-well plate that has been preincubated with 0.5 ml of an appropriate culturemedium (see note in step 1) and culture overnight at 37°C, 8% CO2-in-air, and 98%humidity in a CO2 incubator.

If there are greater than 40 to 50 positive wells, it is difficult to manage conveniently thecloning procedure. The ELISA assays (UNIT 11.4) or western blotting (UNIT 10.8) can beperformed using the 24-well culture supernatants in order to select the most promisingsamples to be cloned. The remaining positive samples can be transferred to cryotubes andfrozen, as described in UNIT 11.9.

Cloning efficiency is always improved when hybrid cells are grown to log phase in 24-wellplates. The efficiency is still better if the 24 wells are preconditioned with feeder cells priorto transferring the hybrids.

3. Perform cell viability count using trypan blue exclusion method (support protocol,UNIT 11.5) on the overnight cultures in 24-well plates.

4. Using a 6-well plate, make dilutions of cells from overnight cultures in HT orcomplete medium. In the first well, make a 1:100 dilution in a total of 3 ml; in thesecond well, dilute an aliquot of the first dilution to 80 cells/ml in 5 ml; in the thirdwell, prepare 8 cells/ml in 10 ml (i.e., a 1:10 dilution from the second well).

Choice of medium is discussed in step 1. Using 6-well plates is far easier for preparing celldilutions than using tubes. However, using 6-well plates is expensive.

5. With a multichannel pipet, fill the upper 50 wells of the inner 60 wells of the96-well plate from step 1 with 100 µl of 8 cells/ml dilution (i.e., 0.8 cells/well)


11.8.1

Cloning ofHybridoma

Cell Lines byLimiting Dilution

and the 10 wells of the bottom row with 100 µl of 80 cells/ml dilution (i.e., 8cells/well).

6. Incubate at 37°C in a CO2 incubator in 8% CO2-in-air with 98% relative humidity(day 1).

As a precaution, the hybridoma cells remaining in the 24-well plate can be transferred tocryotubes and frozen, as described in UNIT 11.9.

7. On day 6, feed the culture with the addition of 100 µl/well of fresh medium, using amultichannel pipet. Thereafter, if necessary, refeed the culture every other day byremoving 100 µl media from each well and adding 100 µl fresh media.

8. When cell growth in the bottom of the wells is 10 to 50% confluent (as monitoredusing an inverted microscope), assay for specific antibody in the hybridoma super-natants using an ELISA (UNIT 11.4).

9. Transfer 2 to 3 selected positive subclones from each plate into a 24-well plate (asfor step 2) and incubate overnight.

Expand the subclones and freeze one aliquot for each subclone in a cryotube. This is doneas a precaution in case one fails to recover positive clones.

10. Repeat cloning procedure from the beginning until a stable and single hybridoma cellline is established.

Hybridomas that yield >90% antibody-positive cultures upon recloning are considered tobe stable. Those that yield <90% positive cultures are subjected to further cloning. Whenclones become stable, reduce the 20% fetal calf serum level in the HT medium to 10%,gradually reduce HT level, and finally remove HT from the medium entirely. Some clonesare more sensitive to this HT weaning process than others.

11. Once established as stable cell lines, hybridomas are maintained in complete culturemedium in a similar manner to the myeloma cell line (UNIT 11.5). Cells are then propagatedfor liquid N2 storage (UNIT 11.9) and for antibody production in ascites fluid (UNIT 11.10).

Recloning of established hybridoma lines may become necessary when they are culturedfor longer than 30 days. Somatic mutation or chromosome loss may occur during anextended culturing, which could lead to a loss of antibody production.

COMMENTARY

Background InformationCloning by limiting dilution is a method based

on the Poisson distribution. Dilution of cells to anappropriate number per well can maximize theproportion of wells that contain one single clone.The cloning protocol presented here is a modifica-tion of the method of Galfre and Milstein (1981).

Critical ParametersAccurate cell counting is necessary to obtain

the proper dilution of 0.8 cells/well at step 5.Clone as soon as antibody-positive hybridomasare identified in order to avoid overgrowth bynonsecreting hybrids.

Anticipated ResultsDepending on accuracy of cell counts and growth

characteristics of each hybridoma, 20 to 50% ofwells seeded at 0.8 cells/well can be expected to

exhibit growth. Normally, all wells at 8 cells/well will exhibit growth. Upon ELISA screeningof the first cloning, some hybridomas will havelost production of specific antibody. The propor-tion of hybridomas that lose antibody productiongenerally decreases at each successive cloning.

Time ConsiderationsTwo to three clonings require 4 to 6 weeks.

Literature CitedGalfre, G. and Milstein, C. 1981. Preparation of

monoclonal antibodies: Strategies and proce-dures. Meth. Enzymol. 73(B):3-46.



11.8.2

Immunology

UNIT 11.9 Freezing and Recovery of Hybridoma Cell LinesBASIC

PROTOCOLFREEZING OF CELL LINESHybridoma cells are suspended in dimethyl sulfoxide/fetal calf serum and frozen rapidlyin a dry ice–ethanol and glycerol bath followed by transfer to liquid nitrogen storage.Cells are recovered by thawing rapidly at 37°C, with immediate replacement of freezingmedium by culture medium.

Materials

Freezing medium95% ethanol10% glycerol

Cryotubes (Nunc #3-63401)15-ml plastic centrifuge tubesLiquid N2 freezer

Additional reagents and equipment for estimating cell viability by trypan blue exclusion (UNIT 11.5)

Ideally, hybridoma cells to be frozen should be in the log phase of growth (as describedfor the Sp2/0 myeloma cells, UNIT 11.5, step 5).

1. Perform a cell viability count using the trypan blue exclusion procedure (supportprotocol, UNIT 11.5).

2. Using a pencil, label cryotubes with identification and date.

3. Centrifuge cell suspension at 100 × g for 5 min at room temperature.

4. Aspirate the supernatant.

5. Resuspend cell pellet in freezing medium to give a cell density of 1 × 107 viablecells/ml.

6. Aliquot 0.5 ml/cryotube (i.e., 5 × 106 cells/tube).

7. Freeze in dry ice/ethanol and glycerol bath for 60 min (see Fig. 11.9.1).

For optimal cell viability, it is important that the time interval between steps 5 and 7 is notless than 15 min or greater than 30 min.

Alternatively, place the tubes in a styrofoam box, then place the box in a −70°C freezerovernight. The styrofoam box slows the rate of freezing of the medium in the tubes.

8. Transfer cryotubes to liquid nitrogen freezer.

BASICPROTOCOL

RECOVERY OF FROZEN CELL LINES

Materials

Complete culture medium (as described for Sp2/0 cells, UNIT 11.5)37°C water bathAlcohol swabs15-ml centrifuge tubes25-cm2 tissue culture flask (Costar or Falcon)

1. Thaw cryotubes completely in 37°C water bath.

Thawing should be completed within 1 min. Immerse tubes only to level of contents.

2. Wipe top of cryotube with alcohol swab, transfer cells to 15-ml centrifuge tube andadd 5 ml of complete culture medium warmed in a 37°C water bath.


11.9.1

Freezing andRecovery ofHybridoma

Cell Lines

3. Centrifuge at 100 × g for 5 min.

4. Aspirate supernatant.

5. Resuspend cell pellet in 5 ml warm complete culture medium (37°C).

6. Transfer suspended cells to a 25-cm2 tissue culture flask.

7. Incubate overnight at 37°C in a CO2 incubator in 8% CO2-in-air with 98% humidity.Keep flask upright.

8. Add 5 ml warm complete culture medium (37°C). Lay flask flat.

9. Propagate cells in complete culture media as described in UNIT 11.10, step 3.


Freezing medium10% (v/v) dimethyl sulfoxide, analytical grade90% (v/v) fetal calf serum

Medium is prepared on day of use and chilled to 4°C before use.

Figure 11.9.1 Apparatus for freezing of hybridoma cell lines in dry ice-ethanol and glycerol bath.


11.9.2

Immunology

COMMENTARY

Background InformationLiquid nitrogen storage is the method of

choice for long-term safekeeping of hybridomacell lines. Frozen aliquots of originally isolatedhybridomas provide insurance against loss ofantibody production and vigor during culture.There are many variations of cell freezing meth-ods in use. However, for freezing hybridomasand lymphoid cells in general, this protocol issimple and has been successful.

Critical ParametersCell viability in the freezing and recovery

protocols is drastically affected by the lengthof time spent in liquid freezing medium priorto and after frozen storage. Adhere carefully totime limits of the protocols.

Anticipated ResultsViability upon recovery of frozen aliquots

ranges from 50% to 95%. Any recovery lessthan 50% viable is considered substandard.

Time ConsiderationsFreezing of cells can be accomplished in 1.5

to 2 hr. Recovery of hybridoma cell lines shouldrequire less than 10 min.

Contributed by Steven A. Fuller, Miyoko Takahashi, and John G.R. HurrellAllelix Inc.Mississauga, Ontario, Canada


11.9.3

Freezing andRecovery ofHybridoma

Cell Lines

UNIT 11.10Production of Monoclonal AntibodySupernatant and Ascites Fluid

Wayne M. Yokoyama1

1Howard Hughes Medical Institute and Washington University School of Medicine,St. Louis, Missouri

ABSTRACT

This unit details methods for production of monoclonal antibodies. Two methods aregiven for production of hybridoma supernatants, including one for large-scale production.A protocol for large-scale production of hybridomas or cell lines is presented for usein isolation of cellular proteins. Finally, a method is given for producing and obtainingmouse ascites fluid containing monoclonal antibody. Recommendations developed by theCommittee on Methods of Producing Monoclonal Antibodies (Institute of LaboratoryAnimal Research, National Research Council) on the use of animals for producingmonoclonal antibodies are also discussed. These recommendations are designed to fosterthe judicious use of animals and to strongly encourage the use of in vitro methodswhenever possible. Curr. Protoc. Mol. Biol. 83:11.10.1-11.10.10. C© 2008 by John Wiley& Sons, Inc.

Keywords: antibody production � monoclonal � ascites

INTRODUCTION

A major advantage of using monoclonal antibodies over polyclonal antisera is the poten-tial availability of large quantities of the specific monoclonal antibody. In general, prepa-rations containing the monoclonal antibody include a hybridoma supernatant, ascites fluidfrom a mouse inoculated with the hybridoma, and purified monoclonal antibody. Hy-bridoma supernatants are easy to produce, especially for large numbers of different mon-oclonal antibodies, but are relatively low in monoclonal antibody concentration. Ascitesfluid contains a high concentration of the monoclonal antibody, but the fluid is not a puremonoclonal antibody preparation. To obtain a purified preparation of the monoclonal an-tibody, affinity chromatography (UNITS 10.9, 10.10, & 11.11) of culture supernatants or ascitesfluid can be performed; however, this obviously requires more effort (see Commentary).

Procedures detailing the production of monoclonal antibody supernatants (see BasicProtocol 1), including the production of amounts on a larger scale (liters; see AlternateProtocol 1) are presented here. The protocol for large-scale production of hybrido-mas or cells (e.g., for isolation of cellular proteins) involves a similar procedure(see Alternate Protocol 2). A method for producing and obtaining ascites fluid containingthe monoclonal antibody is also presented (see Basic Protocol 2).

The unit ends with a set of recommendations developed by the Committee on Methods ofProducing Monoclonal Antibodies, Institute of Laboratory Animal Research, NationalResearch Council. The complete report is published by the National Academy Press(NRC, 1999). The recommendations are designed to foster the judicious use of animalsfor the production of monoclonal antibodies and to strongly encourage the use of in vitromethods whenever possible.

NOTE: All protocols using live animals must first be reviewed and approved by an Insti-tutional Animal Care and Use Committee (IACUC) and must follow officially approvedprocedures for the care and use of laboratory animals.

Current Protocols in Molecular Biology 11.10.1-11.10.10, July 2008Published online July 2008 in Wiley Interscience (www.interscience.wiley.com).DOI: 10.1002/0471142727.mb1110s83Copyright C© 2008 John Wiley & Sons, Inc.

Immunology

11.10.1

Supplement 83

MonoclonalAntibody

Supernatant andAscites Fluid

11.10.2


BASICPROTOCOL 1

PRODUCTION OF A MONOCLONAL ANTIBODY SUPERNATANT

There are a variety of methods for producing monoclonal antibody (MAb) supernatants.In the easy version presented here, the hybridoma is grown and split 1:10. The cellsare then overgrown until cell death occurs. The supernatant is harvested and the titerdetermined. If the titer is high, the hybridoma can be used for large-scale production inanticipation of purification or ascites production.

NOTE: All incubations are performed in a humidified 37◦C, 5% CO2 incubator.

Materials

Hybridoma of interest (see UNIT 11.7)Complete DMEM-10 medium (APPENDIX 3F)

175-cm2 tissue culture flasks50-ml conical centrifuge tubes, sterileBeckman TH-4 rotor (or equivalent)

1. Place hybridoma in a 175-cm2 tissue culture flask in complete DMEM-10 medium.Incubate until vigorously growing and ready to split.

Most cell lines need to be split into new medium or new flasks when cell density reaches1–2 × 106 cells/ml. Tissue culture flasks can be inspected with an inverted microscopeand cell viability and density determined. In addition, the culture can be monitored forcontamination. Cells should never be allowed to become so crowded that cell death occursbecause this crisis phase increases the likelihood of phenotypic change. With experience,most investigators will be able to determine whether or not to split the cells by theirmacroscopic appearance and the color of the medium—i.e., cells may need to be splitwhen the medium in the flask becomes turbid and more yellow than medium in the DMEMbottle (macroscopic turbidity in the absence of a color change is often a sign of bacterialcontamination). The density can then be confirmed by microscopic examination.

2. Split the cells 1:10 in a new 175-cm2 flask. Fill the flask with complete DMEM-10to 100 ml total and incubate until the cells are overgrown, medium becomes acidic(yellow), and cells die (∼5 days).

Alternatively, add hybridoma cells at high density (1–2 × 106 cells/ml) in fresh mediumto a tissue culture flask. Incubate 2 to 3 days, at which time the cells will die and thesupernatant can be collected (step 3).

3. Transfer the flask contents to sterile 50-ml conical centrifuge tubes. Centrifuge10 min at 1500 × g (2700 rpm in a TH-4 rotor), room temperature. Collect thesupernatant and discard the pellet.

4. Assay the titer of the MAb supernatant by an appropriate method (see Commentary).

5. Store the supernatant under sterile conditions; it is generally stable at 4◦C for weeksto months, at −20◦C for months to years, and indefinitely at −70◦C. Minimizethawing and refreezing by storing in several aliquots.

ALTERNATEPROTOCOL 1

LARGE-SCALE PRODUCTION OF MONOCLONAL ANTIBODYSUPERNATANT

The first step in monoclonal antibody purification by affinity chromatography (UNIT 11.11)is the production of large amounts of the culture supernatant. The procedure describedbelow utilizes readily available equipment and supplies and a special tissue cultureroller apparatus. Cells are first grown in smaller flasks, then gradually expanded intolarge-volume roller flasks. The supernatant is harvested and stored until needed.

Immunology

11.10.3


Additional Materials (also see Basic Protocol 1)

Complete DMEM-10 medium (APPENDIX 3F) with 5 to 10 mM HEPES, pH 7.2 to 7.470% ethanol

850-cm2 roller flaskRoller apparatus in 37◦C room or incubator250-ml conical centrifuge tubes, sterileBeckman JS-5.2 rotor (or equivalent)

1. Repeat step 1 of Basic Protocol 1 and split 1:10 in complete DMEM-10/HEPES to100 ml total.

Each 175-cm2 flask will ultimately seed 2.35 to 2.5 liters of culture medium. Scale upthe experiment as desired. If the supernatant will be used for MAb purification by affinitychromatography, the yield will be 1 to 10 mg MAb/liter. It is usually not necessary toadapt and grow cells in serum-free media (which frequently decreases yield); however, ifthe supernatant will be used for MAb purification by protein A–affinity chromatography,test the culture medium with FBS alone for contaminants (i.e., other proteins) that maycopurify with the MAb. Bovine newborn serum frequently contains significant amounts ofIg that will bind to protein A and should not be used as a medium supplement.

2. When cells are ready to split, transfer the contents of the 175-cm2 flask (100 ml)to an 850-cm2 roller flask. Add an additional 150 ml complete DMEM-10/HEPES(250 ml total volume). Cap tightly, place on the roller apparatus in 37◦C room orincubator, and grow 1 to 2 days.

Timing of the split will depend on the cell line (see step 1 of Basic Protocol 1).

3. Wipe the cap and neck of the roller flask with a sterile gauze sponge soaked in 70%ethanol.

4. Open the roller flask and add 250 ml complete DMEM-10/HEPES (500 ml totalvolume). Cap tightly and incubate on the roller apparatus 1 to 2 days at 37◦C.

5. Repeat wiping as in step 3.

6. Open the roller flask and add ∼2 liters complete DMEM-10/HEPES until the flask isalmost full (∼2.5 liters total volume depending on capacity of flask). Cap tightly andincubate on the roller apparatus at 37◦C until the medium turns yellow (∼5 days).

Avoid foaming by first pouring in growth medium without FBS followed by FBS to 10%final concentration.

Avoid prolonged rolling, as cell fragmentation will occur and the debris will be difficultto pellet with the large centrifuge tubes.

7. Pour the culture into sterile 250-ml conical centrifuge tubes. Harvest the supernatantby centrifuging 20 min at 250 × g (1000 rpm in a JS-5.2 rotor), room temperature.Collect the supernatant and discard the pellet. Freeze the supernatant in aliquots.

If the supernatant will not be used in a bioassay, add 10% sodium azide to 0.02% final con-centration. If the supernatant will undergo affinity chromatography or salt fractionation,sterile filtration through a 0.45-μm filter is recommended to eliminate debris.

ALTERNATEPROTOCOL 2

LARGE-SCALE PRODUCTION OF HYBRIDOMAS OR CELL LINES

The following procedure is used to produce large amounts of cells which can be usedto isolate cellular components such as membrane proteins. Individual small flasks aregrown, then each is used to inoculate a larger roller flask. The cells are gradually ex-panded by addition of fresh medium and are harvested when the cells are near saturationdensities.

MonoclonalAntibody


11.10.4



Complete DMEM-10 medium (APPENDIX 3F) with 5 to 10 mM HEPES, pH 7.2 to 7.470% ethanolPhosphate-buffered saline (PBS; APPENDIX 2), 4◦C

850-cm2 roller flaskRoller apparatus in 37◦C room or incubator250-ml conical centrifuge tubes, sterileBeckman JS-5.2 rotor (or equivalent)

1. Follow steps 1 to 6 for large-scale production of MAb supernatants (AlternateProtocol 1) but harvest when the density is appropriate or if the cell growth plateaus.

Estimate the amount of cells needed. This procedure will yield ∼106 cells/ml. More cellscan be obtained with faster-growing cell lines or lines that can tolerate higher densities,but we usually grow 80 liters (33 flasks) and obtain 1010 to 1011 cells.

Seed (introduce cells into) the number of 175-cm2 flasks necessary to produce the amountof cells required (one 175-cm2 flask for every 2.4 liters of medium). Each flask should betreated as an independent culture. Thus, if there is contamination at any time after the175-cm2 flasks are initially seeded, the contamination should spread vertically (i.e., stayin the flask and roller bottle seeded by the contaminated flask) and not horizontally (i.e.,not involve any other flasks or roller bottles).

Estimate the time of harvesting by macroscopic inspection for medium color and turbidityand by taking daily cell counts and checking viability in several flasks. When the cellconcentration reaches a plateau, harvesting is indicated. Do not allow cells to overgrowor cell viability will drop precipitously.

2. Pour the cells into sterile 250-ml conical centrifuge tubes, centrifuge 15 min at 250 ×g (1000 rpm in a JS-5.2 rotor), 4◦C, and discard the supernatant.

Each tube can be used for two spins. Harvesting 80 liters of cells requires at least threecentrifuges to spin four to six tubes (1 to 1.5 liters) each, two or three people, and nearlyone day. Conical centrifuge tubes are recommended because pellets in the flat-bottombottles are harder to work with.

3. Place the cell pellets on ice. Pool ten cell pellets into one tube by resuspending thecells in 250 ml of 4◦C PBS. Centrifuge at 250 × g, 4◦C, and discard the supernatant.

4. Repeat and further consolidate the tubes into one tube.

After three washes, the cells are ready for further processing (i.e., cell lysis, radiolabeling,or other procedures).

Alternatively, lyse smaller cell pellets as they become ready after three washes.

BASICPROTOCOL 2

PRODUCTION OF ASCITES FLUID CONTAINING MONOCLONALANTIBODY

High-titer monoclonal antibody preparations can be obtained from the ascites fluid ofmice inoculated intraperitoneally with monoclonal antibody–producing hybridoma cells.The fluid is collected several times after injection of the cells. It is heat-inactivated, titered,and stored.

NOTE: All protocols using live animals must first be reviewed and approved by anInstitutional Animal Care and Use Committee (IACUC) or must conform to governmentalregulations regarding the care and use of laboratory animals.

NOTE: For a discussion of recently adopted guidelines on the use of animals for produc-tion of monoclonal antibodies, see Background Information.

NOTE: Basic methods for handling mice, including handling and restraint, intraperitonealinjections, and euthanasia can be found in Donovan and Brown (2006a,b,c).

Immunology

11.10.5


Materials

Nude mice, 6 to 8 weeks old and specific-pathogen free, or syngeneic host ifmouse-mouse hybridomas are injected

Pristane (2,6,10,14-tetramethylpentadecane; Aldrich)Hybridoma of interest (see UNIT 11.7)Complete DMEM-10 medium (APPENDIX 3F) with 10 mM HEPES and 1 mM

sodium pyruvatePhosphate buffered saline (PBS) or HBSS (APPENDIX 2), sterile and without FBS

20- or 22-G needle and 18-G needle175-cm2 tissue culture flask50- and 15-ml polypropylene conical centrifuge tubes, sterileBeckman TH-4 rotor (or equivalent)56◦C water bath0.45-μm filter

Additional reagents and equipment for performing ELISA (UNIT 11.2) and countingcells (UNIT 11.5)

1. Using a 20- or 22-G needle, inject mice intraperitoneally with 0.5 to 1 ml Pristaneper mouse 1 week prior to inoculation with hybridoma cells (Donovan and Brown,2006a).

Alternatively, the injections can be done at the same time, but the week interval isrecommended to avoid leakage through two needle sites. Mice should be maintained inspecific-pathogen free (spf) facility.

2. Grow hybridoma cells in a 175-cm2 tissue culture flask in complete DMEM-10/HEPES/pyruvate under conditions that promote log-phase growth.

Before injecting the mice (step 7), test the supernatants for MAb activity by ELISA(UNIT 11.2) or appropriate assay, preferably before the cells are expanded.

To minimize the risk of introducing a pathogen into the rodent colony, screen cells forpathogens by antibody-production assay (Donovan and Brown, 2007).

3. Transfer the culture to 50-ml conical centrifuge tubes. Centrifuge 5 min at 500 × g(1500 rpm in a TH-4 rotor), room temperature, and discard supernatant.

4. Wash the cells by resuspending in 50 ml sterile PBS or HBSS without FBS, thencentrifuging 5 min at 500 × g, room temperature, and discarding the supernatant.Repeat twice and resuspend the cells in 5 ml PBS or HBSS.

Avoid washing in FBS-containing medium because the mouse will produce antibodies tothe FBS.

5. Count the cells and determine viability by trypan blue exclusion (UNIT 11.5).

Cells should be nearly 100% viable.

6. Adjust the cell concentration to 2.5 × 106 cells/ml with PBS or HBSS without FBS.

7. Draw up the cells in a 10-ml sterile syringe. Using a 22-G needle, inject nude miceintraperitoneally with 2 ml cells. Wait for ascites to form (1 to 2 weeks).

Typically, three mice are injected at one time. In most cases, at least one and frequentlyall of the mice will develop ascites.

8. Harvest ascites by grasping and immobilizing the mouse in one hand in such a wayas to stretch the abdominal skin taut. With the other hand, insert an 18-G needle 1to 2 cm into the abdominal cavity. Enter either the left or right lower quadrants toavoid the vital organs in the upper quadrants and the major vessels in the midline.Allow the ascites to drip into a sterile 15-ml polypropylene conical centrifuge.

MonoclonalAntibody


11.10.6


If the mouse has a large amount of ascites and the fluid stops dripping from the 18-Gneedle, it may be necessary to reposition the needle tip by withdrawing it slowly andreinserting it in a different plane. If no ascites fluid accumulates, the mouse may bereinjected (see Commentary).

Occasionally, the ascites is under such high pressure that a large amount squirts out assoon as the needle is inserted. For this reason, be sure that the hub of the needle is pointedinto a tube before inserting the needle into the peritoneal cavity.

Rather than tapping the mouse as soon as the ascites is apparent, allow the fluid to buildup (3 to 7 days) to obtain the highest yield. Frequently, 5 to 10 ml (sometimes >40 ml) ofascites can be collected from each mouse.

9. Centrifuge the ascites 10 min at 1500 × g (2700 rpm in a TH-4 rotor), roomtemperature. Harvest the supernatants and discard the pellet. Store ascites fluid at4◦C until all collection is completed (<1 week).

If the fluid clots, “rim” the clot by passing a wooden applicator stick around its edge(between clot and tube) before centrifugation. The clot may adhere to the applicator stickand thus may be discarded or it will remain in the tube and become part of the cell pelletafter centrifugation.

10. Allow the mouse to reaccumulate ascites (2 to 3 days) before reharvesting as instep 8. Process the ascites as in step 9. Repeat this process until no further ascitesaccumulate, the fluid cannot be collected, or the mouse becomes ill. The mouseshould be euthanized at this point.

11. Pool ascites fluid collected on different days and heat-inactivate 45 min in a 56◦Cwater bath. If a clot reforms, remove it by rimming and centrifuge as in step 9.

12. Assay the titer of MAb-containing ascites by the appropriate method (seeCommentary).

Saturating concentration (maximal activity) of the MAb should be apparent at 0.5% orhigher dilutions. Lower titers usually are the result of unstable hybridomas that stopproducing MAb or too many (>2) in vivo serial passes of the hybridomas.

13. Dilute >1:10 and filter sterilize through a 0.45-μm filter. Aliquot and freeze at−70◦C, avoiding repeated freezing and thawing. Shelf life should be several years.

Add sodium azide to 0.02% final concentration if the ascites will not be used for bioassay.

COMMENTARY

Background Information

Production of MAb supernatantThere are three basic preparations that con-

tain monoclonal antibodies: supernatant froma MAb-producing hybridoma, ascites froma mouse inoculated with the hybridoma, orpurified MAb. In the Basic Protocol 1, a MAb-containing supernatant is produced. The hy-bridoma continues to secrete MAb into theculture fluid until cell death occurs. Becausethe MAb is not metabolized, it accumulatesin the culture supernatant. Hybridoma su-pernatants are advantageous because smallamounts (<100 ml) can be easily obtainedin a few days (<1 week) and from multiplehybridomas with relatively little effort. Suchsmall quantities of the MAb can be used in

preliminary experiments to determine if theMAb has the desired property before expend-ing the effort required to produce ascites orpurified MAb.

The first step in MAb purification, usuallyby affinity chromatography, is the productionof large amounts of the MAb. For the non-commercial laboratory, several liters of culturesupernatant are adequate for most MAb-purification protocols. The use of bioreactorsor large spinner flasks would require expen-sive, somewhat fragile equipment. AlternateProtocol 1 involves the use of sealed rollerbottles that can be rotated on a roller appara-tus in any 37◦C room because the pH of themedium is maintained by the HEPES insteadof a bicarbonate-based buffer system and aCO2 atmosphere.

Immunology

11.10.7


Production of ascites containing MAbA convenient source of large concentra-

tions of the desired MAb is the ascites fluidof mice inoculated with the appropriate hy-bridoma cells. These fluids often have a titer>100-fold more than culture supernatants andthus can be diluted significantly. Frequently,high-titer ascites preparations will have satu-rating MAb concentrations of 10−3 to 10−4

by flow cytometry analysis of cell-surface-antigen binding (Holmes et al., 2001a). Thishigh titer minimizes the nonspecific functionaleffects (e.g., in proliferation assays) of equiv-alent concentrations of MAb in spent culturesupernatants. However, ascites used at concen-trations >0.5% may also have significant non-specific effects. In addition to the appropriateisotype control MAb ascites, ascites harvestedfrom mice inoculated with the nonsecretorypartner cell line (i.e., SP2/0) also provides auseful nonspecific control.

The general procedure involves the elicita-tion of nonspecific inflammation in the peri-toneal cavity of an appropriate host mouse,usually with Pristane, and injection of the hy-bridoma cells (see Basic Protocol 2). The tu-mor cells then grow as an ascites tumor andshould continue to secrete the MAb. Eventu-ally, the mouse develops a monoclonal gam-mopathy similar to the human disease, mul-tiple myeloma, in that a monoclonal proteinreaches high titer in the serum. Since thecombination of the ascites tumor and Pristaneresults in an inflammatory exudate in the peri-toneal cavity, the ascites should contain con-centrations of the MAb similar to that in serum.

NRC recommendations for producingmonoclonal antibodies in animals

In 1998, the National Institutes of Healthasked the National Research Council to forma committee to develop recommendations forusing animals to produce monoclonal antibod-ies. The following is a summary of the com-mittee’s recommendations (NRC, 1999).

1. There is a need for the scientific commu-nity to avoid or minimize pain and sufferingby animals. Therefore, over the next severalyears, as tissue culture systems are further de-veloped, tissue culture methods for the pro-duction of monoclonal antibodies should beadopted as the routine method unless there isa clear reason why they cannot be used or whytheir use would represent an unreasonable bar-rier to obtaining the product at a cost consis-tent with the realities of funding of biomedicalresearch programs in government, academia,and industry. This could be accomplished by

establishing tissue culture production facilitiesin institutions.

There are several reasons why the mousemethod of producing monoclonal antibodiescannot be abandoned: some cell lines do notadapt well to tissue culture conditions; in ap-plications where several different mouse mon-oclonal antibodies at high concentrations arerequired for injection into mice, the in vitromethod can be inefficient; rat cell lines usu-ally do not efficiently generate monoclonalantibodies in rats and adapt poorly to tissueculture conditions but do produce monoclonalantibodies in immunocompromised mice;downstream purification or concentration fromin vitro systems can lead to protein de-naturation and decreased antibody activity;tissue-culture methods can yield monoclonalantibodies that do not reflect the normal mod-ification of proteins with sugars, and thisabnormality might influence binding capac-ity and other critical biologic functions ofmonoclonal antibodies; contamination of valu-able cell lines with fungi or bacteria requiresprompt passage through a mouse to save thecell line; and inability of some cell lines that doadapt to tissue-culture conditions to maintainadequate production of monoclonal antibod-ies poses a serious problem. For these reasons,the committee concludes that there is a sci-entific necessity to permit the continuation ofthe mouse ascites method of producing mon-oclonal antibodies. However, over time, as invitro methods improve, the need for the mouseascites method will decrease.

2. The mouse ascites method of producingmonoclonal antibodies should not be banned,because there is and will continue to be scien-tific necessity for this method.

There does not appear to be convincing ev-idence that significant pain or distress is as-sociated with the injection into the mouse ofPristane (a chemical that promotes the growthof the tumor cells), but during the accumu-lation of ascites there is likely to be pain ordistress, particularly when some cell lines thatare tissue-invasive are used and in situationsof significant ascites development. Therefore,after injection of hybridoma cells, mice shouldbe evaluated at least daily, including weekendsand holidays, after development of visible as-cites and should be tapped before fluid accu-mulation becomes distressful. A limit shouldbe placed on the number of taps, and multipletaps should be allowed only if the animal doesnot exhibit signs of distress.

3. When the mouse ascites method for pro-ducing monoclonal antibodies is used, every

MonoclonalAntibody


11.10.8


reasonable effort should be made to minimizepain or distress, including frequent observa-tion, limiting the numbers of taps, and prompteuthanasia if signs of distress appear.

Two of thirteen monoclonal antibodiesapproved by the Food and Drug Administra-tion for therapeutic use cannot be producedby in vitro means, or converting to an in vitrosystem for their production would require (be-cause of federal regulations) proof of bioequi-valence, which would be unacceptably ex-pensive. Furthermore, many commerciallyavailable monoclonal antibodies are routinelyproduced by mouse methods, particularlywhen the amount to be produced is <10 g, an-other situation where it would be prohibitivelyexpensive to convert to tissue-culture condi-tions, However, with further refinement oftechnologies, media, and practices, productionof monoclonal antibodies in tissue culture forresearch and therapeutic needs will probablybecome comparable to the costs of the mouseascites method and could replace the ascitesmethod.

4. Monoclonal antibodies now being com-mercially produced by the mouse ascitesmethod should continue to be so produced,but industry should continue to move towardthe use of tissue-culture methods.

In a few circumstances, the use of the mouseascites method for the production of mono-clonal antibodies might be required. It is sug-gested that the following be used as examplesof criteria to be used by an Institutional An-imal Care and Use Committee (IACUC) inestablishing guidelines for the production ofmonoclonal antibodies in mice by the ascitesmethod.

a. When a supernatant of a dense hybridomaculture grown for 7 to 10 days (stationary batchmethod) yields a monoclonal antibody con-centration of <5 μg/ml. If hollow-fiber reac-tors or semipermeable-membrane systems areused, 500 μg/ml and 300 mg/μl, respectively,are considered low monoclonal antibody con-centrations.

b. When >5 mg of monoclonal antibodyproduced simultaneously by each of five ormore different hybridoma cell lines is needed.It is technically difficult to produce thisamount of monoclonal antibody because it re-quires more monitoring and processing capa-bility than the average laboratory can achieve.

c. When analysis of monoclonal antibodyproduced in tissue culture reveals that a desiredantibody function is diminished or lost.

d. When a hybridoma cell line grows and isproductive only in mice.

e. When >50 mg of functional monoclonalantibody is needed, and previous poor perfor-mance of the cell line indicates that hollow-fiber reactors, small-volume membrane-basedfermentors, or other techniques cannot meetthis need during optimal growth and produc-tion.

It is emphasized that the above criteria arenot all-inclusive and that it is the responsibilityof the IACUCs to determine whether animaluse is required for scientific or regulatory rea-sons. Criteria have not been developed to de-fine a cell line that is low-producing or whentissue-culture methods are no longer a usefulmeans of producing monoclonal antibody.

In summary, the committee recommendsthat tissue culture methods be used routinelyfor monoclonal antibody production, espe-cially for most large-scale production of mon-oclonal antibodies. When hybridomas fail togrow or fail to achieve a product consistentwith scientific goals, the investigator is obligedto show that a good-faith effort was made toadapt the hybridoma to in vitro growth condi-tions before using the mouse ascites method.

Critical ParametersMonoclonal antibody production by hy-

bridomas is an unstable phenotype. Hy-bridoma cells always should be grown un-der log-phase growth conditions. Prolongedin vitro culture and in vivo passage should beavoided. Thus, the most critical parameter iswhether the hybridoma of interest is secret-ing a high titer of the MAb. This should bechecked before a major effort is made to growlarge amounts of supernatant or to produce as-cites fluid. The MAb titer can be determinedby serial dilution of the culture supernatant inthe assay appropriate for that MAb, such asELISA (UNIT 11.2) or flow cytometry (Holmeset al., 2001a,b). Titers of ≥1:10 should be sat-urating if spent culture supernatants are exam-ined. If necessary, the hybridoma can be re-cloned by limiting dilution (UNIT 11.8) to findhigh-producing clones. If cells are known toproduce MAb at high titers, aliquots frozenimmediately after cloning (Yokoyama, 1997)will retain this phenotype.

The most critical parameter in the large-scale production of cell lines and hybrido-mas is the adaptation of the cells to rollerflasks. Most hybridomas and other nonadher-ent cells that grow in suspension can be eas-ily adapted. If the cells (particularly adher-ent cells) cannot be adapted, other methodsshould be tried. For example, large-scale pro-duction of cells for use in isolating cellular

Immunology

11.10.9


components can be performed using multiple175-cm2 flasks instead of roller flasks. Unfor-tunately, the relative surface area is small, andtherefore the number of flasks required can be-come prohibitive. Adherent cell lines are lesseasily adaptable to growth in roller flasks. Thesurface area for growing adherent cells can beincreased by the use of dextran beads (e.g.,Cytodex beads from Pharmacia). These beadscan increase the surface area of a culture flaskseveral fold.

If there is any suspicion that the cells maybe mycoplasma contaminated, diagnosis andtreatment are indicated (Fitch et al., 1997).Mycoplasma-contaminated lines will producea much poorer yield of final cell numbers be-cause they do not grow to as high a cell densityas normal cells.

Supernatants frequently contain 1 to10 μg/ml of MAb, but the concentration iscell-line dependent. The supernatants couldbe concentrated by salt precipitation, but thisis not generally recommended because largevolumes of culture supernatants have to beconcentrated to derive the amount of MAb insmall amounts of ascites. Moreover, the FBSin the supernatant will also be concentrated.While the hybridoma could be adapted to cul-ture in serum-free medium, this requires addi-tional testing and yields may decrease. Affinitypurification of the culture supernatants wouldtake a similar amount of effort and producepurified MAb at high concentrations. Thus,instead of concentrating supernatants (if highconcentrations of MAb are desired), it is rec-ommended to produce purified MAb by firstgrowing hybridomas at a larger scale (liters)or to produce ascites.

For ascites production, it is important toconsider the appropriate host for the hy-bridoma since an injection of allogeneic orxenogeneic cells may result in rejection.For most mouse-mouse hybridomas, an F1hybrid—between the BALB/c strain (origin ofthe commonly used SP2/0 fusion partner) andthe strain from which the normal cells wereobtained—could be used. For xenogeneic hy-bridomas, nude mice or low-dose irradiatednormal mice are potential hosts. Outbred nudemice are somewhat more expensive than nor-mal mice but do not require irradiation. It isnot necessary to use the prohibitively expen-sive inbred nude mouse strains.

Because the level of normal immunoglob-ulin in mouse serum is in the same range asascites fluid (mg/ml), ascites fluid can be onlypartially purified by salt fractionation or anti-

mouse-Ig- or protein A–affinity chromatogra-phy (Andrew and Titus, 1997). However, it is aconvenient source of raw material from whichto affinity purify rat MAb with a mouse anti-rat κ MAb (e.g., MAR 18.5) column (Andrewand Titus, 1997).

TroubleshootingIt is possible that ascites fluid will not form.

The reasons for this are unclear but are prob-ably related to a property of the individualhybridoma. If the mice die without any as-cites forming, particularly within 2 weeks ofinoculation, try fewer cells. If the mice do notform detectable ascites after 2 weeks and theyappear healthy, inject those mice—as well asnaive, Pristane-primed mice—with more cells.If solid tumors form, tease cells into suspen-sion and inject the tumor cells into anotherPristane-primed mouse. Even if a little ascitesforms, the fluid can be transferred to anothermouse (∼0.5 ml/mouse), and large amounts ofascites should accumulate. Once the ascites isformed, the mouse-adapted cells can be frozenand used to reinoculate mice in the future.

Anticipated ResultsMost culture supernatants will have saturat-

ing MAb titers of ≥1:10 when tested at 100 μlfor 106 cells. If the spent culture supernatantis used for MAb purification by affinity chro-matography (UNITS 10.9 & 10.10; Andrew andTitus, 1997), 1 to 10 mg of purified MAb/litercan be anticipated. If a much lower titer oryield is achieved, recloning of the hybridomaline may be indicated. Hamster-mouse hy-bridomas are particularly notorious forinstability.

Most hybridomas can be grown as ascitestumors. The saturating concentration of theMAb in such fluids should be detected at dilu-tions of 1:500. If MAb titers are significantlylower, the hybridoma may be a poor producer.If ascites do not form, see Troubleshootingabove.

Most tumor cells that grow in suspensionshould be amenable to growth in roller flasks,and densities of >106 cells/ml should be at-tained. Careful work should result in no con-tamination.

Time ConsiderationsFor high-titer and large-scale production of

MAb supernatants, a flask split 1:10 will beovergrown, with cell viability definitely de-creasing by day 5 to 6, at which time the su-pernatants can be harvested. If several liters

MonoclonalAntibody


11.10.10


of supernatant are required, ∼10 days are re-quired to expand a 25-cm2 flask (10 ml) to2.4 liters.

For production of ascites fluid containingMAb, 4 to 6 weeks are necessary for growthof the cells for inoculation, ascites accumu-lation, tapping the fluid, centrifugation, anddetermination of the MAb titer.

Once the 175-cm2 flasks are seeded forlarge-scale production of hybridomas and celllines, <2 weeks are required to reach saturat-ing cell densities in the roller bottles.

Literature CitedAndrew, S.M. and Titus, J.A. 1997. Purification

of immunoglobulin G. Curr. Protoc. Immunol.21:2.7.1-2.7.12.

Donovan, J. and Brown, P. 2006a. Parenteralinjections. Curr. Protoc. Immunol. 73:1.6.1-1.6.10.

Donovan, J. and Brown, P. 2006b. Handling and re-straint. Curr. Protoc. Immunol. 73:1.3.1-1.3.6.

Donovan, J. and Brown, P. 2006c. Euthanasia. Curr.Protoc. Immunol. 73:1.8.1-1.8.4.

Donovan, J. and Brown, P. 2007. Animal health as-surance. Curr. Protoc. Immunol. 76:1.1.1-1.1.3.

Fitch, F.W., Gajewski, T.F., and Yokoyama, W.M.1997. Diagnosis and treatment of mycoplasma-contaminated cell cultures. Curr. Protoc.Immunol. 21:A.3E.1-A.3E.4.

Holmes, K.L., Lantz, L.M., Fowlkes, B.J., Schmid,I., and Giorgi, J.V. 2001a. Preparation of cellsand reagents for flow cytometry. Curr. Protoc.Immunol. 44:5.3.1-5.3.24.

Holmes, K.L., Otten, G., and Yokoyama, W.M.2001b. Flow cytometry analysis using theBecton Dickinson FACS Caliber. Curr. Protoc.Immunol. 49:5.4.1-5.4.22.

NRC. 1999. Monoclonal Antibody Production:A Report of the Committee on Methods ofProducing Monoclonal Antibodies, Instituteof Laboratory Animal Research, NationalResearch Council. National Academy Press,Washington, D.C.

Yokoyama, W.M. 1997. Cryopreservation of cells.Curr. Protoc. Immunol. 21:A.3G.1-A.3G.3.

UNIT 11.11Purification of Monoclonal Antibodies

BASICPROTOCOL

PURIFICATION USING PROTEIN A–SEPHAROSE

Ascites fluid is diluted in high pH buffer and loaded onto protein A–Sepharose. Thecolumn is washed and bound antibody is eluted with a succession of buffers of decreasingpH. The procedure for other affinity columns is similar, except that only one eluting bufferis used. Antibodies are concentrated by ultrafiltration.

Materials

Protein A–Sepharose CL-4B (Pharmacia)Tris buffer, pH 8.6Ascites fluid (UNIT 11.10)Citrate buffer, pH 5.5Acetate buffer, pH 4.3Glycine⋅Cl buffer, pH 2.3Neutralizing buffer, pH 7.7

2.5-cm (inner diameter) glass chromatography columnUV flowthrough detector or UV spectrophotometer and cuvetteUltrafiltration cells and XM50 membranes (Amicon)

Additional reagents and equipment for ELISA screening (UNIT 11.4)

1. Fully swell protein A–Sepharose in Tris buffer (50-fold excess, vol/vol) in a beaker.Pour the protein A–Sepharose into a 2.5-cm glass chromatography column andequilibrate protein A–Sepharose with Tris buffer at room temperature (see UNIT 10.11A;Figure 10.11A.1).

The size of the protein A–Sepharose column and the volume of ascites used are subject tothe requirements of the researcher. A column of 10-ml bed volume (∼3 g dry weight proteinA–Sepharose) is sufficient to purify antibody from 10 to 20 ml of ascites.

2. Dilute ascites fluid in 3 vol Tris buffer and apply to column at a flow rate of 1 to 5ml/min.

Ascites can be applied directly to the top of the column by a Pasteur pipet or a flow adapter.

3. Wash column with Tris buffer until entire unbound proteins are eluted. This can bemonitored by measuring the column effluent at A280 using a UV flowthrough detectoror a UV spectrophotometer and cuvette.

4. Collect unbound protein and successively eluted protein peaks by fraction collectoror by pooling an entire buffer elution volume.

5. Elute bound proteins successively with 2 to 3 column volumes of citrate, acetate, andglycine⋅Cl buffers directly into test tubes in a fraction collector containing neutraliz-ing buffer in an amount equal to one-quarter of the collected volume.

Alternatively, bound material may be eluted in one step using glycine⋅Cl buffer only.

Collecting individual buffer fractions provides a better separation of monoclonal antibodyfrom nonspecific antibody.

6. Eluted fractions or pools may be assayed for antigen-specific monoclonal antibodyby ELISA (UNIT 11.4).

Supplement 37


11.11.1

Immunology

7. Concentrate appropriate eluates containing monoclonal antibodies to 1 to 5 mg/mlusing an ultrafiltration cell with an XM50 ultrafiltration membrane. Store at −20°C.

Antibody concentration may be estimated by measuring absorbance at 280 nm. A mouseIgG solution at 1 mg/ml = 1.44 absorbance units.

ALTERNATEPROTOCOL

ALTERNATIVE BUFFER SYSTEM FOR PROTEIN A–SEPHAROSE

Pharmacia recently suggested another buffer system that results in higher binding capacityfor certain monoclonal antibodies. The protocol was as described above; however, thebuffers used were different. These alternative buffers are as listed below.

1. Glycine⋅OH buffer. Replaces Tris buffer in the basic protocol.

2. Elution buffers (0.04 M sodium citrate/0.02 M NaCl at pH 6.0, 5.0, 4.0, and 3.2).These buffers replace those in the basic protocol as follows:

Elution buffers, pH 6.0 and pH 5.0. Replace citrate buffer.Elution buffer, pH 4.0. Replaces acetate buffer.Elution buffer, pH 3.2. Replaces glycine⋅Cl buffer.

ALTERNATEPROTOCOL

PURIFICATION BY ANTIGEN-SEPHAROSE AND ANTI-MOUSEIMMUNOGLOBULIN-SEPHAROSE

This protocol describes the preparation of other affinity columns for purification ofmonoclonal antibodies if protein A–Sepharose will not bind the antibody (e.g., IgM orsome IgG1 antibodies). It is useful for coupling protein antigens or anti-mouse immuno-globulin antibodies to commercially available CNBr-activated Sepharose 4B (see supportprotocol, UNIT 10.16 for alternate procedure). Columns to which antigen or anti-mouseimmunoglobulin antibody are attached may be used as immunoadsorbents to isolatespecific antibody (antigen-Sepharose) or all mouse immunoglobulins (anti-mouse immu-noglobulin–Sepharose) from ascites fluid.


CNBr-activated Sepharose 4B (Pharmacia)1 mM HClCoupling bufferProtein antigen (previously purified; see Chapter 10) or anti-mouse

immunoglobulin antibody (commercially available)1 M ethanolamine, pH 8.0Phosphate-buffered saline (PBS; APPENDIX 2)Washing buffer

60- or 150-ml sintered glass funnel, medium porosity50-ml conical plastic centrifuge tube

1. Allow required amount of CNBr-activated Sepharose 4B to swell in 1 mM HCl.

One gram of dry gel swells to about 3 ml. It is convenient to use 5 to 10 mg of ligand permilliliter of gel. The remainder of the protocol assumes that 10 ml of gel is being prepared.

2. Transfer to sintered glass funnel and wash successively with (a) 1 mM HCl, 200 to500 ml, and (b) coupling buffer, 200 to 500 ml (wash over a period of 0.5 hr).

3. Transfer swollen gel to a 50-ml conical plastic centrifuge tube. Add 50 to 100 mg ofligand at a concentration of 2 to 5 mg/ml.

4. Mix gel and ligand on rocker platform or end-over-end mixer overnight at 4°C.


11.11.2

Purification ofMonoclonalAntibodies

5. Centrifuge at 250 × g for 10 min.

6. Fill the tube with coupling buffer. Centrifuge as in step 5.

7. Incubate with 20 to 40 ml of 1 M ethanolamine, pH 8.0, on rocker platform for 4 to5 hr at 4°C.

8. Centrifuge at 250 × g for 10 min. Wash gel successively by filling the tube with PBSand washing buffer followed by centrifugation.

9. Transfer gel to a 2.5-cm glass column and equilibrate with 100 ml PBS.

10. Ascites fluid is diluted with 3 vol PBS and applied to the column at a flow rate of 1to 5 ml/min at room temperature.

11. Wash unbound protein from the column with 30 ml PBS.

12. Elute bound protein using glycine⋅Cl buffer from the basic protocol (p. 11.11.1,step 5).

13. Concentrate eluate containing monoclonal antibodies to 1 to 5 mg/ml using ultrafil-tration cell and XM50 ultrafiltration membrane. Store at −20°C.

Antibody concentration may be estimated by measuring absorbance at 280 nm. A mouseIgG solution at 1 mg/ml = 1.44 absorbance units.


Acetate buffer, pH 4.3To 900 ml H2O, add:6.80 g sodium acetate (trihydrate) (0.05 M)8.77 g NaCl (0.15 M)Titrate with acetic acid to pH 4.3

Add H2O to 1 liter

Citrate buffer, pH 5.5

}(0.05 M citrate)2.45 g citric acid (anhydrous)10.96 g trisodium citrate dihydrate8.77 g NaCl (0.15 M)Add H2O to 1 liter

Coupling buffer, pH 8.3 (0.125 M phosphate)Solution A: 17.75 g Na2HPO4 (anhydrous)

Add H2O to 1 literSolution B: 1.95 g NaH2PO4⋅2H2O

Add H2O to 100 mlTitrate solution A with solution B to pH 8.3

Elution buffers (0.04 M sodium citrate/0.02 M NaCl at pH 6.0, 5.0, 4.0, and 3.2)To 900 ml H2O, add:11.76 g trisodium citrate dihydrate1.17 g NaClTitrate with HCl to pH 6.0, 5.0, 4.0, or 3.2Add H2O to 1 liter

1 M ethanolamine, pH 8.061.1 ml ethanolamineTitrate with HCl to pH 8.0Add H2O to 1 liter


11.11.3

Immunology

Glycine⋅Cl buffer, pH 2.3To 900 ml H2O, add:3.75 g glycine (0.05 M)8.77 g NaCl (0.15 M)Titrate with HCl to pH 2.3Add H2O to 1 liter

Glycine⋅OH bufferTo 700 ml H2O, add:108.9 g glycine (1.45 M)175.3 g NaCl (3 M)Titrate with NaOH to pH 8.9Add H2O to 1 liter

Neutralizing buffer, pH 7.7 (0.5 M phosphate)Solution A: 70.98 g Na2HPO4 (anhydrous)

Add H2O to 1 literSolution B: 7.80 g NaH2PO4⋅2H2O

Add H2O to 100 mlTitrate solution A with solution B to pH 7.7

Tris buffer, pH 8.6To 900 ml H2O, add:6.06 g Tris⋅Cl (0.05 M)8.77 g NaCl (0.15 M)0.2 g NaN3 (0.02%)Titrate with HCl to pH 8.6Add H2O to 1 liter

Washing bufferTo 900 ml H2O, add:13.6 g sodium acetate (trihydrate) (0.1 M)29.2 g NaCl (0.5 M)Titrate with acetic acid to pH 4.0Add H2O to l liter

COMMENTARY

Background InformationThe uses of monoclonal antibodies as en-

zyme conjugates and as immunoaffinity re-agents require their purification from the crudeascites fluid. Protein A–Sepharose chromatog-raphy and affinity chromatography are superiorto ammonium sulfate precipitation, gel filtra-tion, or ion-exchange chromatography for thepreparation of contaminant-free antibody.

Literature ReviewStaphylococcal protein A is known to bind

immunoglobulins of several mammalian spe-cies. Ey et al. (1978) were able to use proteinA–Sepharose to fractionate murine IgG anti-bodies using a stepwise pH elution. This pro-tocol uses the buffer systems described by Oiand Herzenberg (1980), which allow efficient

purification of most IgGs and occasional IgMmonoclonal antibodies at room temperature.The alternate protocol for preparation of spe-cific affinity columns is a modification of theprocedure suggested by Pharmacia.

Critical ParametersLow-pH buffer elutions should be collected

into vessels containing neutralizing buffer toavoid undue loss of antibody activity becauseof denaturation.

Anticipated ResultsFor most antibodies bound by the affinity

columns, yield will be quantitative. ProteinA–Sepharose will bind some IgM, most IgG1,and nearly all IgG2a, IgG2b, and IgG3 mono-clonal antibodies. Yield of other affinity tech-


11.11.4

Purification ofMonoclonalAntibodies

niques depends on the antibody-ligand interac-tion.

Time ConsiderationsThe time involved in affinity chromatogra-

phy depends largely on sample size and columndimensions. Research scale procedures (lessthan 20 ml of ascites on a 1 × 10 cm column)can be accomplished in 2 to 4 hr.

Literature CitedEy, P.L., Prowse, S.J., and Jenkin, C.R. 1978. Isola-

tion of pure IgG1, IgG2a and IgG2b immuno-globulins from mouse serum using protein A–Sepharose. Immunochemistry 15:429-436.

Oi, V.T. and Herzenberg, L.A. 1980. Immunoglobu-lin-producing hybrid cell lines. In SelectedMethods in Cellular Immunology (B.B. Mishelland S.M. Shiigi, eds.) pp. 351-372. W.H. Free-man, San Francisco.

Key ReferencesHurrell, J.G.R. ed. 1982. Monoclonal Hybridoma

Antibodies: Techniques and Applications. CRCPress, Boca Raton, FL.

Langone, J.J. and Van Vunakis, H. eds., 1986. Im-munological techniques, Part I: Hybridoma tech-nology and monoclonal antibodies. Meth. Enzy-mol. 121: 1-947.

These volumes present a large body of material onbasic hybridoma methodology and detail the use ofmonoclonal antibodies in the study of hormones,structural proteins, viruses, parasites, and mammal-ian cell types.

Galfre, G. and Milstein, C. 1981. Preparation ofmonoclonal antibodies: Strategies and proce-dures. Meth. Enzymol. 73(B):3-46.

Details the strategies and procedures for the prepa-ration of monoclonal antibodies.



11.11.5

Immunology

SECTION IIIPREPARATION OF POLYCLONALANTISERA

UNIT 11.12Production of Polyclonal Antisera

Helen M. Cooper1 and Yvonne Paterson2

1Queensland Brain Institute and University of Queensland, Brisbane, Australia2University of Pennsylvania, Philadelphia, Pennsylvania

ABSTRACT

Much of modern biology and biochemistry relies on the availability of highly specificantibodies for use in such ubiquitous techniques as immunohistochemistry, ELISAs, im-munoprecipitation, and immunoblotting. Thus, the generation of large quantities of spe-cific antibodies directed against proteins or peptides of interest is essential to the successof both basic and applied research programs. In addition, with the advent of antibody-based proteomic strategies for profiling protein expression and post-translational modifi-cation, a requirement for timely production of specific antibodies has been exemplified.Polyclonal antibodies derived from animals immunized with purified proteins or peptidesare particularly valuable for use in the laboratory. This unit provides protocols for theproduction of polyclonal antisera specific for protein antigens using rabbits, rats, mice,and hamsters. Curr. Protoc. Mol. Biol. 81:11.12.1-11.12.10. C© 2008 by John Wiley &Sons, Inc.

Keywords: polyclonal antiserum � adjuvant � antigen � immunization

INTRODUCTION

Antibodies are serum immunoglobulins with binding specificity for particular antigens.Although antibodies can be identified in the serum of individuals or patients that havebeen exposed to particular pathogens, the usual methods for eliciting antibodies involveimmunization with purified or partially purified antigen preparations. Antigens used aremost commonly proteins or peptides, but carbohydrates, nucleic acids, small organicmolecules (haptens) conjugated to appropriate protein carriers, cells, and cell and tissueextracts can also be employed.

The first consideration is usually whether polyclonal or monoclonal antibodiesare needed. Polyclonal antibodies are particularly valuable for immunoprecipitation(UNIT 10.16) and immunoblotting (UNIT 10.8), whereas monoclonal antibodies can haveexquisite specificity and can be derived for almost any purpose. Choice of the species ofanimal to be used for immunization is based in part on whether antibodies of great speci-ficity are required—in which case genetically defined strains can be very helpful—orantibodies of wide cross-reactivity are needed.

The amount of antibody needed must also be evaluated. Clones of hybridomas (somaticcell hybrids of B cells from an immunized animal’s spleen and myeloma tumors permis-sive for the production of monoclonal immunoglobulins) provide an essentially limitlesssupply of a constant reagent. Nevertheless, the initial investment in producing a mono-clonal antibody is quite large, whereas relatively large amounts of a polyclonal antiserumcan be obtained from a single rabbit or from several genetically identical rats or mice.

Current Protocols in Molecular Biology 11.12.1-11.12.10, January 2008Published online January 2008 in Wiley Interscience (www.interscience.wiley.com).DOI: 10.1002/0471142727.mb1112s81Copyright C© 2008 John Wiley & Sons, Inc.

Immunology

11.12.1

Supplement 81

Production ofPolyclonal

Antisera

11.12.2


In this unit, the Basic and Alternate Protocols describe the production of polyclonalantisera specific for protein antigens in rabbits, rats, mice, and hamsters. The SupportProtocol presents a method for preparing serum from blood. Polyclonal antipeptideantisera can be produced by substituting carrier-conjugated peptides (UNIT 11.16) for thepurified protein antigens.

STRATEGIC PLANNING

Production of good antisera depends in large part upon the quality, purity, and amountof available antigen as well as on the specificity and sensitivity of the assay. For proteinantigens, if possible, the material should be biochemically homogeneous and, dependingon the intended use, should be in either a native or denatured conformation. Be awarethat minor contaminants are often (unfortunately) more antigenic than the immunogenof interest, and antisera resulting from immunization may have more activity againstthe contaminants than against the protein of interest. Antisera to be used for screeningbacterial expression cDNA libraries or for immunoblots are best made against denaturedprotein, whereas those to be used for screening cDNAs expressed in eukaryotic transfec-tion systems or for immunoprecipitation of native-cell-synthesized structures might bestbe made against native protein.

Although the advances offered by the development of monoclonal antibody techniqueshave revolutionized the specificity, uniformity, and quantity of antibodies, thereremain many circumstances in which polyclonal antibodies are more desirable thanmonoclonal antibodies. Production of polyclonal antisera takes less time and effortthan production of monoclonal antibodies, requires relatively simple and readilyavailable equipment, and produces reagents that can be used for immunoprecipitation,immunoblotting, and enzyme-linked immunosorbent assays (ELISAs). In addition, withthe advent of antibody-based proteomic strategies for profiling protein expression andpost-translational modification, a requirement for high-throughput antibody productionhas emerged. Multiplexing of peptide and protein immunizations within a single rabbitallows rapid generation of specific antibodies while minimizing the number of animalsused (Larsson et al., 2006). This approach has obvious benefits in terms of animalwelfare and cost efficiency.

Choice of animal for the production of antibodies depends upon the amount of antiserumdesired, the evolutionary distance between the species from which the protein of interesthas been derived and the species of the animal to be immunized, and prior experiencewith the immunogens. Rabbits are the usual animal of choice because they are geneticallydivergent from the human and mouse sources of the proteins most often studied. Rabbitsprovide as much as 25 ml of serum from each bleed without significant harmful effects.For smaller-scale experiments, or for those that rely on precisely defined antibody speci-ficities, inbred mouse strains may be the system of choice. Because mice are smaller,the volume of antigen suspension used for immunization is significantly less, but theamount of serum that can be obtained from a single bleed does not exceed 0.5 ml. Ratsand hamsters may be used when larger amounts of serum are needed, or when the greaterevolutionary distance is advantageous. With repeated bleeding, as much as 5 ml of serumcan be obtained from these species. Additional discussion of the choice of species forthe production of monoclonal antibodies is given in UNIT 11.10.

The choice of adjuvant for in vivo animal use has become problematic in recent years.Freund’s adjuvant has been reliably and widely used for over fifty years (Freund et al.,1937). However, there is a degree of distress and discomfort to the animal associated withits use, which requires that the responsible investigator explore alternatives (McWilliamand Niemi, 1988). In most studies that have compared available alternatives to Freund’sadjuvant, the Basic Protocol given here, which uses a CFA/IFA immunization regimen,

Immunology

11.12.3


gives far superior antibody titers than the commonly used, commercially available ad-juvants (Johnston et al., 1991; Deeb et al., 1992; Smith et al., 1992). However, theTiterMax adjuvant has now emerged as an effective alternative to CFA/IFA (Bennettet al., 1992; also see http://www.titermax.com). TiterMax has two significant advantagesover CFA/IFA in that it is less toxic and can deliver high, sustained titers after a singleimmunization. The use of TiterMax is described in the Alternate Protocol, and the re-searcher should seriously consider employing this adjuvant over CFA/IFA. It should bekept in mind that not all antigens are equal; therefore, one adjuvant may perform betterthan another for a given antigen.

NOTE: All protocols using live animals must first be reviewed and approved by anInstitutional Animal Care and Use Committee (IACUC) and must follow officially ap-proved procedures for the care and use of laboratory animals. The ethical issues as-sociated with the production of polyclonal antisera in animals is addressed on manyuniversity Web sites. For example, the Duke University and Medical Center Web siteprovides a comprehensive guide to the selection of animals and alternative adjuvantchoices (http://vetmed.duhs.duke.edu/guidelines for adjuvant use.htm). It is also recom-mended that, before embarking on the production of antiserum, the researcher read thereport and recommendations of the European Center for the Validation of AlternativeMethods (ECVAM) on the production of polyclonal antibodies in laboratory animals(http://altweb.jhsph.edu/publications/ECVAM/ecvam35.htm). It is highly recommendedthat appropriately trained animal technicians within the institute’s animal facility carryout the immunization and bleeding of the animals.

BASICPROTOCOL

IMMUNIZATION TO PRODUCE POLYCLONAL ANTIBODIES USINGFREUND’S ADJUVANT

In the presence of adjuvant, the protein antigen is injected intramuscularly (i.m.), intra-dermally (i.d.), or subcutaneously (s.c.) into an animal of the chosen species. Boosterimmunizations are started 4 to 8 weeks after the priming immunization and continued at2- to 3-week intervals. Prior to the priming immunization and following the primary andeach booster immunization, the animal is bled and serum prepared from whole blood(see Support Protocol).

Instructions on the different strategies for immunization (i.m., i.d., or s.c.) can be foundin Donovan and Brown (2006a) and for bleeding (from marginal vein or artery of ear forrabbit; various other sites for mouse, rat, or hamster) in Donovan and Brown (2006b). Fac-tors important in preparing specific high-titer antisera, procedures for modifying proteinantigens to enhance their immunogenicity, and choice of host animal are discussed in theCommentary. Protocols for immunization prior to production of monoclonal antibodies(UNIT 11.10) should be reviewed for these purposes.

Materials

Rabbit, rat, mouse, or hamster of appropriate strainComplete Freund’s adjuvant (CFA; Sigma)1 to 2 mg/ml purified protein antigen in PBS (APPENDIX 2)Incomplete Freund’s adjuvant (IFA; Sigma)

50-ml disposable polypropylene centrifuge tubes3-ml glass syringes with 19-, 21-, and 22-G needlesDouble-ended locking hub connector (Luer-Lok, Becton Dickinson) or plastic

3-way stopcock

CAUTION: CFA is an extremely potent inflammatory agent, particularly if introducedintradermally or into the eyes, and may cause profound sloughing of skin or loss of sight.Self-injection can cause a positive TB test and lead to a granulomatous reaction. Usegloves and protective eyewear when handling CFA.


Antisera

11.12.4


1. Bleed the animal prior to immunization and collect blood sample in a 50-ml cen-trifuge tube. Prepare serum from blood, then assay and store (see Support Protocol).

This preimmune bleed is critical as a control to ensure that the antibody activity detectedin later bleeds is due to the immunization.

2. Shake CFA to disperse insoluble heat-killed Mycobacterium tuberculosis bacilli.Add 2 ml CFA to 2 ml of 0.25 to 0.5 mg/ml purified protein antigen in PBS at 4◦C.

These volumes produce immunogen sufficient to immunize 4 rabbits or up to 80 mice. Donot use Tris-based buffers for generating the emulsion.

An effective and simple method for preparing purified protein antigen is by preparativeSDS-PAGE (UNIT 10.2A). As much as 2 mg of a homogeneous protein can be loaded acrossmultiple precast gels (1.5-mm-thick gel, 0.5-cm-wide comb). Following electrophoresis,an edge can be sliced off with a razor blade, fixed and stained, and used to identifythe region containing the protein band (UNIT 10.6). The gel slice containing the proteinmay then be directly added to several milliliters of PBS (APPENDIX 2) and emulsified asdescribed below with an equal volume of CFA. The acrylamide serves as an additionalcomponent for the protein depot provided by the adjuvant.

3. Draw up the CFA/antigen mixture into a 3-ml glass syringe with a 19-G needle.Remove needle, expel as much air as possible, and attach syringe to the double-endedlocking hub connector or the plastic 3-way stopcock (see Fig. 11.12.1). Attach anempty 3-ml glass syringe at the other end and force the mixture back and forthfrom one syringe to the other repeatedly. When the mixture is homogeneous andwhite, disconnect the connector or stopcock, attach a 21-G needle, and test whetherthe emulsion is stable by extruding a small drop onto the surface of 50 ml of coldwater in a 100-ml beaker. A good oil-in-water emulsion should hold together as adroplet on the surface of the water. If the drop disperses, mix the antigen using thehub-connected syringes until it forms an emulsion.

Heat will be generated by this procedure. Chill on a bed of ice from time to time to keepthe mixture as close to 4◦C as possible.

Images of stable emulsions can be found in Koh et al. (2006).

4. Transfer all of the adjuvant-antigen emulsion to one syringe and remove the connectoror stopcock. Attach a 22-G needle to the syringe and remove air bubbles.

Figure 11.12.1 Double-syringe device for preparation of antigen-adjuvant emulsions.

Immunology

11.12.5


5. Restrain the animal and inject the adjuvant/antigen emulsion into multiple i.m., i.d.,or s.c. sites.

Discard the unused immunogen. For extremely valuable antigens, the emulsion may bestored at 4◦C for several weeks and re-emulsified before use. However, denaturation ofprotein antigens may take place under these conditions. For immunization of small rodents(e.g., mice), it is often better to carry out injections intraperitoneally (i.p.).

6. Bleed the animal 10 to 14 days following the priming immunization and collectblood sample. Prepare serum from blood (see Support Protocol).

7. Prepare antigen for booster immunizations, following steps 2 to 4. When CFA is theprimary adjuvant, use IFA as the adjuvant for all subsequent immunizations.

8. Administer the first booster immunization 4 to 8 weeks after the priming immuniza-tion, bleed the animal 7 to 14 days later, and collect blood sample. Prepare serumfrom blood (see Support Protocol).

Some investigators will administer the first booster immunization as early as 2 weeksafter the primary immunization.

9. Administer further booster immunizations at 2- to 3-week intervals. Bleed animal10 to 14 days after each boost and collect blood sample. Prepare serum from blood(see Support Protocol).

Repeated i.d. immunization should be avoided as it can cause skin ulceration. Followingprimary i.d. or s.c. immunization, it is preferable to boost with i.m. injections for therabbit. Some investigators prefer primary i.m. injections with boosters at other sites.

ALTERNATEPROTOCOL

IMMUNIZATION TO PRODUCE POLYCLONAL ANTISERUM USINGTITERMAX

For highly immunogenic antigens, the use of Freund’s adjuvant can certainly be avoided.For other immunogens it may be necessary to test different adjuvant systems.

Additional Materials (also see Basic Protocol)

TiterMax #R-1 (CytRx, Sigma, or http://www.titermax.com; store <24 months at4◦C)

1-ml all-plastic syringes

1. Emulsify aqueous antigen with TiterMax adjuvant. See Basic Protocol, steps 1through 5, but use 0.5 ml antigen and a 0.5-ml vial of TiterMax in step 2 and anall-plastic syringe in step 3.

TiterMax contains microparticulate silica coated with the nonionic block copolymerCRL-8941 and squalene.

Although glass syringes are recommended for Freund’s adjuvant emulsions, all-plasticsyringes should be used with TiterMax.

Each reconstituted 0.5-ml vial will immunize 20 mice or 10 rabbits. Unused anti-gen/adjuvant emulsion can be stored at 4◦C, −20◦C, or −70◦C for as long as the antigenis stable. It may be necessary to re-emulsify before using.

2. Transfer the antigen emulsion to a 1-ml syringe, attach a 22-G needle to the syringe,and remove air bubbles.

3. Restrain the animal and inject the adjuvant/antigen emulsion.

Rabbits should receive either 40 µl TiterMax/antigen emulsion i.m. in each thigh, or four100-µl s.c. injections over the shoulders and hind limbs, or ten 40-µl i.d. injections alongthe back. High antibody titers have been obtained with 30 to 50 µg antigen per rabbit.


Antisera

11.12.6


4. Bleed the animal and prepare antigen for booster immunization (see Basic Protocol,steps 6 and 7).

Boosting with TiterMax may not be necessary for all antigens. If a second immunizationis necessary, use soluble antigen in place of antigen/adjuvant emulsion at 4 weeks. Iftiters are still low after 10 to 14 days, a booster dose of antigen/TiterMax adjuvant canbe given immediately. Increasing the amount of antigen may also help.

5. Administer booster immunizations at 4, 8, and 12 weeks. Bleed the animal 10 to14 days after each booster immunization. Prepare serum from blood (see SupportProtocol) and cease immunization when high antigen-specific titers have beenachieved.

SUPPORTPROTOCOL

PREPARATION OF SERUM FROM BLOOD

Each blood sample is allowed to stand 4 hr at room temperature and overnight at 4◦Cuntil a clot forms. After removal of the clot and debris, the serum is assayed and storedat −20◦C.


Blood samples (see Basic Protocol)Sigma 4K15 centrifuge and 11150 rotor (or equivalent)

Additional reagents and equipment for immunoblotting (UNIT 10.8),immunoprecipitation (UNIT 10.16), and ELISA (UNIT 11.2)

1. Allow blood to stand in the 50-ml centrifuge tube 4 hr at room temperature to allowclot to form, then place overnight at 4◦C to allow clot to retract.

2. Gently loosen the clot from the sides of the tube with a wooden applicator stick (donot break up the clot), then remove the clot from tube with the applicator.

If a clot has not formed, initiate clotting by placing a wooden applicator stick into thetube containing the collected blood, then begin again at step 1.

3. Transfer serum to a 50-ml centrifuge tube. Pellet any remaining blood cells anddebris by centrifuging 10 min at 2700 × g, 4◦C, and save supernatant.

4. Assay antibody titer by the appropriate method: immunoprecipitation, immunoblot-ting, ELISA, or double-immunodiffusion assay in agar.

5. Store serum in aliquots in screw-top tubes at −20◦C.

Some sera lose activity on repeated freezing/thawing; others aren’t stable at 4◦C.

COMMENTARY

Background InformationThe kinetics of development of a spe-

cific antibody response upon immunization ofa rabbit with antigen are illustrated in Fig-ure 11.12.2. After the primary immunization,naive B cells are stimulated to differentiateinto antibody-secreting plasma cells. For mostsoluble protein antigens, specific antibody be-gins to appear in the serum 5 to 7 days afterthe animal is injected. The antibody concentra-tion (titer) continues to rise and peaks aroundday 12, after which it decreases. Similar kinet-ics are observed with mice, rats, hamsters, andrabbits.

In addition to differentiating into antibody-forming cells, the antigen-stimulated B cellsproliferate to form a large population of mem-ory B cells, which quickly become activatedafter the booster injection is administered.Thus, the lag period before the appearanceof the specific antibody is much shorter af-ter a booster injection than that observed forthe initial immunization. In addition, a sig-nificantly higher titer of specific antibody isachieved and sustained for a longer period oftime. The peak of antibody production occurs7 to 14 days after boosting. As a consequenceof the existence of the memory B cells, less

Immunology

11.12.7


Figure 11.12.2 Kinetics of development of the specific antibody response. Arrows indicate whenpriming and boosting immunizations were administered. Actual amounts of specific antibody pro-duced will vary considerably depending on immunogenicity of the protein.

antigen is required to stimulate a strong sec-ondary response. Memory B cells are long-lived; therefore, a specific antibody responsecan be elicited as much as 6 months to ayear after the last booster. Finally, the averageaffinity and degree of specificity of the anti-body population for the antigen increase withrepeated immunizations (Klinman and Press,1975). A more in-depth discussion of the im-munological concepts underlying the genera-tion of the humoral immune response, includ-ing the potency of a variety of adjuvants, isgiven in Aguilar and Rodrıguez (2007) andLiang et al. (2006).

Adjuvants greatly enhance the specific an-tibody titer, as they allow the antigen to bereleased slowly, thus ensuring the continualpresence of antigen to stimulate the immunesystem. Freund’s adjuvant has been used ex-tensively in the preparation of antigen becauseit induces a high, long-lasting antibody titerthat is often still measurable 25 weeks ormore after boosting. The presence of killedmycobacteria in complete Freund’s adjuvant(CFA) activates the T cell population, provid-ing necessary lymphokines for B cell stimu-lation and maturation. CFA may cause gran-uloma and subsequent necrotic abscesses, soit should be used only for primary immuniza-tion. Some countries are restricting the use ofCFA in laboratory animals for this reason. In-complete Freund’s adjuvant (IFA) is adequatefor booster injections. Wherever possible, toreduce animal discomfort, less noxious ad-juvants should be used, as described in theAlternate Protocol.

Over recent years many different adjuvantsystems have been formulated for use in hu-mans in the clinical context (reviewed in Lianget al., 2006; Aguilar and Rodrıguez, 2007).Although, in general, these adjuvants do notboost the immune response to the same ex-tent as CFA/IFA or TiterMax, they are lesstoxic and therefore may be worth consider-ing, especially when the antigen is expectedto be highly immunogenic. Published compar-isons of commercial adjuvants with the basicCFA/IFA protocol vary widely in their conclu-sions. In most cases, however, CFA/IFA pro-duces higher titers of higher-affinity antibodiesin a shorter time period (Johnston et al., 1991;Deeb et al., 1992; Smith et al., 1992; Lianget al., 2006; Aguilar and Rodrıguez, 2007).For this reason, CFA/IFA has always been fa-vored for the generation of polyclonal antiserain animals for laboratory studies. Presently,the TiterMax adjuvant appears to be one ofthe only viable alternatives to CFA/IFA, andshould therefore be the adjuvant of choice dueto its lower toxicity and ability to generate highantibody titers after a single immunization.

Critical ParametersNew Zealand red or white rabbits are gener-

ally the best source of specific antisera because30 to 50 ml of whole blood can be obtainedat each bleed. The life span of a rabbit is 5to 6 years, so a continual source of specificantiserum can be provided over a period oftime by one rabbit after booster injections. Inthis regard, the recommended times betweenbooster injections are not critical; the animal


Antisera

11.12.8


may be rested for several months between sub-sequent boosters, after the primary and sec-ondary booster injections. Blood collection,however, must take place 7 to 14 days aftereach booster to ensure a high titer.

Preimmune serum from the same animalis the preferred negative control. If addi-tional control serum is required, either immuneserum from animals immunized with totallyunrelated antigens or pooled serum from naiveanimals will be adequate. Occasionally, spu-rious antibody activities from nonimmunizedanimals may mimic the activity of the immuneserum.

Antibody specificity may vary widely be-tween individual animals with respect to thedominant antigenic epitopes recognized on agiven protein antigen. Therefore, antiserumfrom a single animal should be used through-out a study. If more than one animal mustbe used for particular antisera, the antiserashould be pooled. Large-scale production ofantisera can be carried out in goats, sheep,and horses with appropriate veterinary guid-ance. If serum is taken from inbred animalstrains, the variability in antibody specificity,as observed in outbred rabbits, is less of aproblem.

The most important factor in producing ahighly specific polyclonal antiserum is the pu-rity of the antigen preparation used for immu-nization. The immune system is very sensi-tive to the presence of foreign proteins. Anycontaminating proteins in the antigen prepara-tion can potentially induce a strong immuneresponse when injected in the presence of ad-juvant. When antisera are employed in sensi-tive techniques such as immunoblotting or thescreening of cDNA libraries, significant an-tibody titers to protein contaminants can bea major problem. Thus, the antigen prepara-tion should contain no significant contaminat-ing proteins. Ideally, there should be no visiblecontaminating bands when 10 to 20 µg are an-alyzed on an SDS-polyacrylamide gel stainedwith Coomassie brilliant blue (UNITS 10.2 &

10.6).If the antiserum is to be used in functional

assays, extra care must be taken to ensurethat the immunizing antigen is in its nativeform, because antibodies directed against de-natured forms of the protein antigen will in-teract weakly, if at all, with the antigen in itsnative conformation. On the other hand, anti-bodies used in immunoblots, immunoprecip-itation of primary in vitro translation prod-ucts, and immunoscreening of cDNA expres-sion libraries may be most effective if gener-

ated against a denatured protein with reducedand carboxymethylated disulfide bonds. Sim-ilar considerations are also relevant to the useof antisera in proteomic applications.

TroubleshootingInability to attain high-titer antiserum af-

ter several booster injections may be due to avariety of factors as described below.

Use of inappropriate adjuvant. Some ex-perimentation may be necessary to optimizethe antigen/adjuvant ratio for different anti-gens. If the Alternate Protocol still fails to pro-duce a good antibody titer after three immu-nizations, switch to the Basic Protocol. Furtheradjuvant systems are described in Liang et al.(2006) and Aguilar and Rodrıguez (2007).

Inadequate antigen emulsification. If theemulsion fails the drop-on-water test de-scribed in the Basic Protocol (step 3; also seeKoh et al., 2006), repeat the emulsificationprocess. Be sure to use phosphate-bufferedsaline. Avoid plastic syringes and Tris-basedbuffers with CFA and IFA. Alternative emul-sification techniques include homogenization,sonication, or vortexing (Koh et al., 2006;http://www.titermax.com).

The antigen is a poor immunogen. In gen-eral, the immunogenicity of a protein is re-lated to the degree to which it differs from“self” proteins (Benjamin et al., 1984). Largebacterial or viral proteins such as hemag-glutinin or bacterial-coat proteins are highlyimmunogenic, whereas proteins from mam-malian sources, such as polypeptide hormonesor cell- surface receptors, may be poorly im-munogenic due to tolerance. Protein antigenscan be made more immunogenic in two ways.First, they can be chemically linked to a carrierprotein that is known to be a good immuno-gen. Common carrier proteins include keyholelimpet hemocyanin (KLH), fowl immunoglob-ulin, and bovine serum albumin (BSA). Cou-pling peptides to carrier proteins is describedin UNIT 11.16; the same protocols can be used tocouple the protein antigen of interest to the de-sired carrier. Second, the immunogenicity ofan antigen may be enhanced by its polymeriza-tion into large aggregates via a cross-linkingagent such as glutaraldehyde. The protocol inUNIT 11.16 for the coupling of peptide antigensto a carrier protein with glutaraldehyde canalso be used to polymerize any protein antigen.With both the coupling and polymerizationprocedures, any insoluble antigen complexesformed should be removed prior to immuniza-tion by centrifuging 10 min at 15,000 × g,4◦C.

Immunology

11.12.9


Host animal’s immune system may be com-promised by bacterial or viral infection. Re-fer to Donovan and Brown (2007a,b) for dis-cussion of the consequences of poor ani-mal husbandry. Utilize animals from reliable,pathogen-free sources and maintain them inappropriate infection-free facilities.

Only a few animals have been immunized.Because of the vagaries of immune-responsegenes in outbred animals such as rabbits, someantigens may not induce a good antibody re-sponse in a significant proportion of randomlyselected animals. Thus, it is best to immunizeseveral different animals and to screen the serafor the best responder. Obviously, this is lessof a problem in homozygous inbred strains,but with a new antigen it is wise to test severalstrains for their antibody response.

An insufficient amount of antigen wasused. Although recommended concentrationsof antigen for rabbits are 0.25 to 0.5 mg/mlinjected into multiple sites, for a total of 1 to2 ml in the same animal, good results can beobtained with 1/10 to 1/20 of the concentra-tion in the same volume. It is always temptingto use less of a precious antigen, but often toolow a dose leads to too low a response.

Anticipated ResultsFor large or nonevolutionarily related pro-

teins, a titer of 5 to 10 mg/ml of serum canbe expected after repeated boosts (hyperim-munization). When immunizing with small orhighly conserved protein species, a titer of 1 to2 mg/ml of specific antibodies is more likely.Antibody titers and affinity for the antigen willbe low after primary immunization and thefirst booster immunization, but both titer andaffinity will increase with subsequent immu-nizations.

Time ConsiderationsPreparation of the immunogen and immu-

nization will take ∼3 hr on each occasion. Col-lection of antisera will take 1 to 2 hr, dependingon the number and species of animals.

When using the CFA/IFA adjuvant system,collection of antiserum after the primary im-munization will be at 10 to 14 days. This willbe a low-titer, low-affinity serum. The firstbooster normally is given 4 to 8 weeks afterthe primary immunization but can be given asearly as 2 weeks after the primary if time is crit-ical. Ideally, there should be at least 19 days be-tween the primary and the secondary bleeds. Asecond booster is given at 6 weeks with a bleedon day 52 to 59. This will usually be the first

high-titer bleed. If a titer of <1 mg/ml of spe-cific antibody is obtained, subsequent boostingimmunization will be necessary. When usingTiterMax as the adjuvant, only a single immu-nization is required. In this case, a low titer isobserved (equivalent to that seen after primingwith CFA). Subsequently, the titer continues torise without further boosting, allowing serumto be collected at biweekly intervals. If titersbegin to decrease, the rabbit should be boostedas described in the Alternate Protocol.

Literature CitedAguilar, J.C. and Rodrıguez, E.G. 2007. Vaccine

adjuvants revisited. Vaccine 25:3752-3762.

Benjamin, D.C., Berzofsky, J.A., East, I.J., Gurd,F.R.N., Hannum, C., Leach, S.J., Margoliash,E., Michael, J.G., Miller, A., Prager, E.M.,Reichlin, M., Sercarz, E.E., Smith-Gill, S.J.,Todd, P.E., and Wilson, A.C. 1984. The anti-genic structure of proteins: A reappraisal. Annu.Rev. Immunol. 2:67-101.

Bennett, B., Check, I.J., Olsen, M.R., and Hunter,R.L. 1992. A comparison of commercially avail-able adjuvants for use in research. J. Immunol.Methods 153:31-40.

Deeb, B.J., DiGiacomo, R.F., Kunz, L.L., andStewart, J.L. 1992. Comparison of Freund’s andRibi adjuvants for inducing antibodies to thesynthetic antigen (TG)-AL in rabbits. J. Im-munol. Methods 152:105-113.

Donovan, J. and Brown, P. 2006a. Parenteral injec-tions. Curr. Protoc. Immunol. 73:1.6.1-1.6.10.

Donovan, J. and Brown, P. 2006b. Blood collection.Curr. Protoc. Immunol. 73:1.7.1-1.7.9.

Donovan, J. and Brown, P. 2007a. Animal health as-surance. Curr. Protoc. Immunol. 76:1.1.1-1.1.3.

Donovan, J. and Brown, P. 2007b. Managing im-munocompromised animals. Curr. Protoc. Im-munol. 77:1.2.1-1.2.5.

Freund, J., Casals, J., and Hismer, E.P. 1937. Sen-sitization and antibody formation after injectionof tubercle bacilli and paraffin oil. Proc. Soc.Exp. Biol. Med. 37:509.

Johnston, B.A., Eisen, H., and Fry, D. 1991. Anevaluation of several adjuvant emulsion regi-mens for the production of polyclonal antiserain rabbits. Lab. Anim. Sci. 41:15-21.

Klinman, N.R. and Press, J. 1975. The B cell speci-ficity repertoire: Its relationship to definablesubpopulations. Transplant. Rev. 24:41-83.

Koh, Y.T., Higgins, S.A., Weber, J.S., and Kast,W.M. 2006. Immunological consequences of us-ing three different clinical/laboratory techniquesof emulsifying peptide-based vaccines in incom-plete Freund’s adjuvant. J. Transl. Med. 4:42-54.

Larsson, K., Wester, K., Nilsson, P., Uhlen, M.,Hober, S., and Wernerus, H. 2006. MultiplexedPrEST immunization for high-throughputaffinity proteomics. J. Immunol. Meth. 315:110-120.


Antisera

11.12.10


Liang, M.T., Davies, N.M., Blanchfield, J.T., andToth, I. 2006. Particulate systems as adjuvantsand carriers for peptide and protein antigens.Curr. Drug Deliv. 3:379-388.

McWilliam, A. and Niemi, S.M. 1988. Freund’sadjuvant. Canadian Council on Animal CareResource. 12:1.

Smith, D.E., O’Brien, M.E., Palmer, V.J., andSadowski, J.A. 1992. The selection of an ad-juvant emulsion for polyclonal antibody pro-duction using a low-molecular-weight antigenin rabbits. Lab. Anim. Sci. 42:599-601.

Key ReferenceColigan, J.E., Bierer, B.E., Margulies, D.H.,

Shevach, E.M., and Strober, W. (eds.) 2007.Current Protocols in Immunology. John Wiley& Sons, Hoboken, N.J.

UNIT 11.13In Vitro Antibody Production

This unit describes the antigenic stimulation of in vitro antibody production by B cellsand the subsequent measurement of secreted antibodies. Such in vitro systems have beenextremely useful for studying the regulation of interacting cell types involved in antibodyproduction—namely B cells, T cells, and macrophages—as well as their antigenicstimulants. In addition, these studies have revealed important information about cellularrequirements for growth and differentiation factors during antibody production.

A generalized system for inducing in vitro antibody production, and which can accom-modate various types of antigens under study, is first presented (see Basic Protocol 1).Secreted antibodies can then be measured with an enzyme-linked immunosorbent assay(ELISA; UNIT 11.2) or other soluble-antibody detection systems. Alternatively, the numberof antibody-producing cells can be quantified by plaque-forming cell (PFC) assays: theCunningham-Szenberg technique (see Basic Protocol 2) and the Jerne-Nordin technique(see Basic Protocol 3). Both methods employ specially prepared slide chambers (seeSupport Protocols 1 and 2) in which the antibody-producing B cells are mixed withcomplement and indicator sheep red blood cells (SRBC), or with trinitrophenol-modifiedSRBC (TNP-SRBC; see Support Protocol 3), with subsequent lysis and counting ofplaques. Because IgM antibodies fix complement efficiently, whereas IgG and IgAantibodies do not, unmodified PFC assays measure only IgM antibodies. The assay canbe modified, however, to measure all classes of antibodies or to enumerate total immu-noglobulin-secreting B cells (see Alternate Protocol). Yet another method of measuringthe number of antibody-producing B cells (in a class-specific fashion) is to use theELISPOT technique (Lycke and Coico, 1996). The resting B cells used in these proceduresare prepared as described in Support Protocol 5 for Percoll gradient centrifugation.

NOTE: All solutions and equipment coming into contact with cells must be sterile, andproper sterile technique must be used accordingly.

STRATEGIC PLANNING

Choice of Antigen

In setting up an in vitro system for antibody production and measurement, the antigen tobe used is a critical factor that will determine the types of cells and lymphokines necessaryfor an adequate response.

A first consideration is whether the antigen to be used is one that contains immunogenicepitopes that the animal from which cells are to be obtained has “seen” in vivo. Generallyspeaking, if the antigen is new and a primary response is being elicited in vitro, theresponse will be small, perhaps below the limits of measurement, because the cells utilizeddo not contain the number of specific precursors necessary for a measurable in vitroresponse. Responses to most protein antigens fall into this category, so that in vitroinduction with such antigens require the use of cells from in vivo primed animals. Anapparent exception to this rule is the particulate antigen SRBC, which provides goodresponses even with unprimed cells. This, however, can be attributed to the fact that SRBCcontain immunogenic epitopes that cross-react with those present in environmentalantigens, so that the cell donor has, in fact, been exposed to this antigen in vivo. It shouldbe noted in this regard that SRBC induce only IgM antibody responses in in vitro cellsobtained from unprimed animals, probably reflecting the fact that cells from unprimedanimals do not contain enough SRBC-specific precursors of isotype-switched B cells.

Supplement 50

Contributed by James J. Mond and Mark BrunswickCurrent Protocols in Molecular Biology (2000) 11.13.1-11.13.17Copyright © 2000 by John Wiley & Sons, Inc.

11.13.1

Immunology

Thus, to obtain IgG responses with SRBC antigen, one must use cells from SRBC-primedanimals.

A second consideration is whether the antigen is a T cell–dependent or –independentantigen. T cell–dependent antigens (most protein antigens) require the presence ofantigen-primed T cells which provide the necessary factors (lymphokines) for B cellproliferation and differentiation. In addition, for these antigens there is increasing evi-dence that T cells are required for promoting cell-cell interactions necessary for antibodyproduction. Perhaps the one exception to this rule is that certain T cell– dependent antigenssuch as SRBC can stimulate primed B cells in the absence of T cells, provided that T celllymphokines are present.

T cell–independent antigens, as defined originally, can stimulate B cells in vitro in theabsence of T cells. Type I antigens, such as trinitrophenylated lipopolysaccharide (LPS),have intrinsic polyclonal activating properties, whereas type II antigens, such as trini-trophenylated Ficoll and dextran, do not possess this characteristic. While both stimulatein vivo responses in nude mice and in vitro responses in T cell–depleted spleen cells, typeI antigens can stimulate responses in immature neonatal B cells and in cells from xidimmune-defective mice, but type II antigens stimulate responses only in immunologicallymature B cells. The T cell–independent designation for these antigens is now outdated,as it has become clear that T cell–derived lymphokines are necessary for the responsewhen very pure resting B cells are used (Mond et al., 1983; Thompson et al., 1984).Consequently, T cell–derived lymphokines are now usually added to systems that use Tcell–independent antigens as a stimulant, It is therefore most correct to refer to this classof antigen as “T cell–regulated” or as “T cell–independent type I or type II.”

Decisions concerning the type of antigen used (and the cell/lymphokine requirements)must be made in order that the response to be studied is precisely defined. In this regard,B cells responding to complex, multiepitopic antigens of any kind are a clonally diversemixture of cells; the antibody response will therefore be heterogeneous and will compriseantibodies with specificity for the various immunodominant determinants on the antigen.However, if a modified antigen is utilized—i.e., an antigen conjugated to a chemicallydefined low-molecular-weight antigenic epitope—the response to this epitope can thenbe measured. In this case the known epitope is called the hapten and the molecule to whichit is attached is called the carrier; since B cell responses are being quantified, the B cellsin these systems recognize the hapten and T cells recognize other epitopes on the carrier(Mitchison, 1971). TNP is frequently employed as the hapten; this chemical group canbe placed on various carriers, including soluble proteins (ovalbumin), to measure highlydefined T cell–dependent B cell responses, as well as on other T cell–independent carrierssuch as Brucella abortus (BA) and LPS to measure type I responses, or on Ficoll anddextran to measure type II responses.

Table 11.13.1 summarizes lymphokine requirements of cellular responses to some anti-gens. This table is incomplete in that the type of lymphokine that is added may haveconsiderable effect both on the quality of the responses (such as the isotype profile) andon the magnitude of the response. For example, in most systems IL-4 and IFN-γ arerequired lymphokines for stimulation of IgE and IgG2a responses, respectively.

A final consideration relating to choice of antigen concerns the need for macrophages,either as antigen-presenting cells (APC) or as sources of cytokines in the culture system.Whereas B cells can present those antigens for which they express specific surface Igreceptors, they are not generally sufficient in the usual in vitro culture systems becausesuch antigen-specific cells are present in low numbers. Macrophages, on the other hand,take up and present antigen in an antigen-nonspecific manner and are thus the primary


11.13.2

In Vitro AntibodyProduction

choice of APC in in vitro systems. Thus, for responses to T cell–dependent antigens, whereT cell activation is required for inducing a response, the need for macrophages is absolute.T cell–independent responses, however, are relatively independent of the requirement formacrophages. It is important to note that the presence of too many macrophages may besuppressive for some in vitro responses, because of the release of inhibitory substances(e.g., prostaglandins and certain cytokines).

Preparation of Cells

Prior to setting up cultures to measure in vitro antibody production, considerable thoughthas to be given to the methods for preparing the cell populations to be studied. Unprimed,unseparated cell populations may be obtained from a variety of lymphoid organs. Morefrequently, however, purified T and B cell populations are used (Hathcock, 1991a,b;Mage, 1993), or purified populations of resting B cells alone can be used. Priming of Tcells is described in Kruisbeek and Shevach (1991); priming of B cells is accomplishedby immunizing mice with hapten-carrier conjugate emulsified in complete Freundsadjuvant 3 to 5 weeks prior to cell harvest (using a different carrier than that used for Tcell priming). It is best to size-separate B cells on Percoll gradients (see Support Protocol5) after enrichment as in Hathcock (1991b) when studying T cell–independent responses,since responses of resting and partially activated B cells vary (Thompson et al., 1984).This cannot be done when studying T cell–dependent responses, as antigen-primed(activated) B cells are required.

BASICPROTOCOL 1

INDUCTION OF ANTIGEN-SPECIFIC AND POLYCLONALANTIBODY PRODUCTION

Antibody production is induced by culturing B cells, desired antigens (either sheep redblood cells or trinitrophenylated carriers), appropriate lymphokines, and T cells for 4 to5 days (see Strategic Planning, above). Secreted antibodies can be measured in thesupernatant with ELISA assays (UNIT 11.2). Alternatively, the number of antibody-produc-ing cells can be measured as in Basic Protocol 2 or Basic Protocol 3.

Materials

Antigens (see recipe): SRBC, TNP-BA, TNP-Ficoll, TNP-dextran,TNP-ovalbumin (TNP-OVA), or TNP-LPS

Recombinant lymphokines (for use with purified B cells): rIL-2 (Cetus orSchering) and/or rIL-1 (Hoffman La Roche)

Complete RPMI-10 medium (see recipe) or complete DMEM-10 medium (seerecipe)

Table 11.13.1 Cells and Lymphokines Required for Stimulation of In Vitro AntibodyProductiona

Antigen T Cells B Cells Lymphokines PFC/culture

TNP-Ficoll − + IL-1, IL-2, or IL-5 400 (IgM)TNP-BA − + IL-1, IL-2, or IL-5 2000 (IgM)SRBC (T-independent) − + IL-1, IL-2, or IL-5 1000 (IgM)SRBC (T-dependent) T′ + − 3000 (IgM)

4000 (IgG)TNP-OVA T′ + − 500 (IgM)

300 (IgG)aAbbreviations: BA, Brucella abortus; OVA, ovalbumin; PFC, plaque-forming cells; SRBC, sheep red bloodcells; T′, antigen-primed T cells; TNP, trinitrophenylated.


11.13.3

Immunology

B lymphocytes purified from mice (Support Protocol 5) either nonimmunized orimmunized 3 to 5 weeks previously with specific hapten-carrier conjugate (UNIT

11.12 and Table 11.13.1)T cells primed to antigen or carrier (Kruisbeek and Shevach, 1991; Table 11.13.1)

96-well flat-bottom microtiter plates (Costar)Refrigerated low-speed centrifuge (e.g., IEC 7R with 216 rotor)Sterile gauze

1. Add antigens and recombinant lymphokines diluted to the appropriate concentrationsin supplemented complete medium in a total volume of 0.15 ml to the wells of a96-well microtiter plate. Set up in triplicate to permit calculation of standard error.

The amount of antigen to be used must be determined separately in each system. A widerange of concentrations should be tested in pilot experiments and one or two concentrationsin later (more definitive) studies.

Approximate final concentrations in the wells for representative antigens:

2 × 106 cells/well SRBC10 �g/ml TNP-OVA1:100 dilution of stock TNP-BA10 ng/ml TNP-Ficoll5 to 50 �g/ml TNP-LPS or unconjugated LPS

Lymphokines are added when responses to T cell–independent antigens are studied. IfTNP-Ficoll or SRBC are used as the stimulating antigen, rIL-1 and rIL-2 (or alternatively,rIL-5) are recommended. When TNP-BA is used as antigen, only IL-2 or IL-5 is necessary.rIL-1 is used at 10 U/ml and rIL-2 at 100 U/ml. When T cells are added, lymphokines arenot required.

2. Add 105 to 106 B cells in 50 µl to each well with multichannel pipettor to give a finalvolume of 0.2 ml.

Appropriate cell number must be determined empirically. Use lower cell concentrationsfor responses elicited by polyclonal stimuli, and higher cell concentrations for responsesdriven by specific antigens, but never use more than 106 cells per well.

3. Incubate plates for 4 to 5 days at 37°C, 5% CO2. If an assay of secreted antibody isdesired, go to step 4. If a plaque assay is to be performed, go to step 7.

During the culture period, preparation can be made for analysis of culture supernatantsby an enzyme-linked immunosorbent assay (UNIT 11.2).

4. Centrifuge plates 10 min at 1000 rpm (170 × g), 4°C. In a tissue culture hood, discardsupernatant by inverting and flicking plate, then blot dry with sterile gauze.

5. Add 0.2 ml fresh supplemented complete medium to each well; repeat step 4 twice.

6. Add 0.2 ml fresh supplemented complete medium to each well and incubate plates24 to 48 hr in a humidified 37°C, 5% CO2 incubator. Remove and save 150 µlsupernatant from each well with a multichannel pipettor if secreted antibody is to bemeasured using ELISA or other suitable technique.

7. Centrifuge plates 10 min at 170 × g, 4°C. Invert plates in one quick motion and flickoff supernatant. Without reinverting plates, remove excess fluid by blotting plates onabsorbent paper.

8. Add 0.2 ml supplemented complete medium to each well and repeat step 7.

9. Add 0.2 ml supplemented complete medium to each well. Tap plates gently or placeon vibrating platform to ensure complete resuspension of cells.


11.13.4


Repeated washing of the wells is required to remove free antibody before the subsequentassay of antibody production. Antibody-producing B cells are now ready to be measuredas described for plaque-forming cell assays (see Basic Protocols 2 and 3).

PLAQUE-FORMING CELL ASSAYS

B cells prepared in Basic Protocol 1 can be assayed for antibody production using aplaque-forming cell (PFC) assay. In this approach, the B cells are plated in (or on) a layerof hapten- or antigen-conjugated SRBC under conditions in which the secreted antibodywill cause local SRBC lysis, resulting in pin-point areas of clearing in the lawn of SRBC(plaques) that can be counted. The two widely used PFC assays presented below are alikein principle but different in detail. Antibody-producing B cells are mixed with a sourceof complement and indicator RBC if SRBC are used as stimulating antigen, or TNP-RBCif TNP-conjugated antigens are used as the stimulating antigen. The B cell/SRBC mixtureis then processed in Cunningham-Szenberg chambers or with Jerne-Nordin slides. Thistechnique measures only cells secreting IgM since only antibody of the latter isotype fixescomplement under the conditions of the assay. It can, however, be modified to measurecells secreting antibody of any class by the addition of isotype-specific rabbit anti-Igantibody that binds to secreted immunoglobulin and forms complexes that fix comple-ment. The PFC assays can also be modified to enumerate total immunoglobulin-secretingB cells stimulated by a polyclonal B cell activator, e.g., LPS (see Alternate Protocol).

There are advantages and disadvantages of both methods. The principle advantage of theCunningham-Szenberg assay is that six to eight plates can be processed in one day. Itsdisadvantage is that extra time and effort are required for preparing the slide chambers,which cannot be reused, and for preabsorbing the complement. The Jerne-Nordin assay,on the other hand, accommodates only three 96-well plates in an 8-hr period. However,while this technique requires construction of a tray for holding the slides, the tray can bereused indefinitely; in addition, complement needn’t be preabsorbed. Thus, for IgG orprotein A plaquing, Jerne-Nordin is the assay of choice.

BASICPROTOCOL 2

Cunningham-Szenberg Assay (Cunningham and Szenberg, 1968)

Materials

Supplemented complete RPMI-10 (see recipe) or DMEM-10 medium (see recipe)7.5% (v/v) SRBC or 15% (v/v) TNP-SRBC (see Support Protocol 3) in

supplemented complete mediumSource of complement: 50% (v/v) guinea pig serum (Life Technologies; also see

recipe) in supplemented complete mediumWax (tissue preparation grade; Fisher)

96-well round-bottom microtiter plate (Linbro # 36-311-05)Cunningham-Szenberg chambers (see Fig. 11.13.1 and Support Protocol 1)120-cm2 glass petri dishDissecting microscope

NOTE: All reagents should be room temperature before addition to chambers, becausecold reagents result in bubble formation during incubation.


11.13.5

Immunology

1. Prepare molten wax by heating in a 120-cm2 glass petri dish on a hot plate; maintainat low heat.

2. To each well of an empty 96-well microtiter plate, add the following:

50 µl supplemented complete medium50 µl 7.5% SRBC or 15% TNP-SRBC50 µl suspended B cells (from step 9 of Basic Protocol 1)50 µl diluted guinea pig serum

Mix the contents of each well with pipettor and fill Cunningham-Szenberg chambersby slowly expelling the contents of the pipet with the tip placed at an angle at theoverlap edges (see Fig 11.13.1B).

Add guinea pig serum to only 24 wells at a time to prevent inactivation of the complementduring the period the slides are kept at room temperature. The time needed to fill and seal24 slides is at the upper limit necessary to prevent the chambers from drying out.

3. After filling 24 slide chambers, seal each side with molten wax (see Fig. 11.13.1).Incubate chambers 45 to 60 min at 37°C.

Metal trays from standard 37°C incubators are excellent for carrying slides; each trayholds ∼72 slides.

Seal with wax within 20 min to prevent the chamber from drying out.

4. Remove chambers from incubator and count plaques under a dissecting microscopeat 10× magnification.

Care must be taken to differentiate between air bubbles and real plaques. Air bubbles haveclearly defined borders and are highly reflective. Plaques have hazy borders and arenonreflective (see Fig. 11.13.2).

tape

glass slides

double-sided tape

cells introduced into thetwo chambers overlap edges

wax seal

petri dish onhot plate

wax sealinga chamber

A

B C

slides

Figure 11.13.1 The Cunningham-Szenberg chamber. (A) The chambers as they are made instrips. (B) The whole chamber. (C) The chambers being dipped into a petri dish and sealed with hotwax.


11.13.6


BASICPROTOCOL 3

Jerne-Nordin Assay (Jerne and Nordin, 1963)

Materials

1% agarose (SeaPlaque; FMC Bioproducts)2× basal Eagle medium (Life Technologies)10% (v/v) SRBC or 20% (v/v) TNP-SRBC (see Support Protocol 3) in

supplemented complete mediumPhosphate-buffered saline (PBS; APPENDIX 2), 4°CSource of lyophilized guinea pig serum (Life Technologies)

5-ml pipets and 5-ml glass test tubes, prewarmed42°C water bathPrecoated agarose slides (see Support Protocol 2)Tray for holding slides (Fig. 11.13.3; not commercially available)

1. Mix equal amounts of 1% agarose with 2× basal Eagle medium in a 50-ml centrifugetube. Using a prewarmed pipet, add 0.4 ml of this plaquing medium to glass test tubes(maintained in a rack in a 42°C water bath).

Maintain this mixture at 42°C until the B cells are added and placed onto the slides. Thisis critical, because at lower temperatures agarose slowly and imperceptibly gels and givesrise to false plaques. The cooled agarose can no longer be used.

2. To each test tube, first add 1 drop of 10% SRBC or 20% TNP-SRBC, then add (6tubes at a time) 0.1 ml B cell suspension (from step 9 of Basic Protocol 1) aftercarefully resuspending cells.

3. Vortex tubes and immediately pour and spread suspensions onto precoated agaroseslides. Allow slides to harden and turn upside down on a tray (see Fig. 11.13.3).

Trays are designed such that eight slides (or less) can be placed upside-down.

4. Cover tray with a damp cloth and place 60 to 90 min in an incubator at 37°C.

Even if a humidified incubator is used, do not omit the damp cloth.

5. Prepare complement by adding 5 ml of cold PBS to 5 ml lyophilized guinea pig serum.When serum is dissolved, add it to 95 ml of cold PBS.

Figure 11.13.2 Typical plaques from PFCassay. The plaques show pinpoint clearing in alawn of RBC. This is due to lysis by complementand antibody being secreted by the B cell(center of plaque).


11.13.7

Immunology

6. Add 2 ml of freshly diluted guinea pig serum per slide to the shallow well in the trayholding the slides (i.e., 16 ml/tray; see Fig. 11.13.3). Place slides in 37°C incubatorfor 90 min.

There is no further need for a damp cloth, as the complement maintains the moisture.

7. Remove slides from incubator and count plaques using a dissecting microscope at10× magnification (Fig. 11.13.2).

ALTERNATEPROTOCOL

MEASUREMENT OF ISOTYPE-SPECIFIC ANTIBODYAND POLYCLONAL ANTIBODY RESPONSES

Isotype-Specific Antibody Response

The Jerne-Nordin PFC assay (see Basic Protocol 3) can be modified for measurement ofIgG-and IgA-producing B cells as follows. At the end of step 1, add 50 µl goat anti-IgMantibody to each tube containing plaquing medium to inhibit IgM (direct) plaques. Then,after the incubation in step 4 (before preparing complement in step 5), add developinganti-IgG (or anti-IGA) to the slides to develop IgG (or IgA)-producing B cells. This isdone by filling the shallow well in the tray holding the slides with 2 ml of antibody.Incubate 1 hr at 37°C, then draw off the developing antibody with a Pasteur pipetconnected to a vacuum suction flask. Proceed with the preparation and addition ofcomplement as in steps 5 and 6.

Appropriate dilution of rabbit anti-IgM, -IgG, and -IgA antibodies (Jackson Labs) insupplemented complete medium must be predetermined for each batch of antibody, but isusually in the range of 1:100. Addition of developing antibodies is critical, since IgG andIgA antibodies fix complement poorly in their absence.

21.3 cm

cross-section

cross-section of tray

5.7 cm

A

B

C

1.5 0.60.60.2

0.2 0.1

nail

slide upside down

tray

agarose on slidecomplement or anti-lgG antibody in tray

Figure 11.13.3 Jerne-Nordin slide assembly. (A) Top view of tray sufficient for eight slides. In thecorners, nails can be worked through to facilitate stacking of multiple trays. (B) Amplified cross-sec-tion of the tray with detailed dimensions (in cm). (C) Cross-section of tray with agarose-coated slidein place for incubation.


11.13.8


Polyclonal Antibody Response

The Jerne-Nordin PFC assay (see Basic Protocol 3) can also be modified to measure totalIg-secreting B cells stimulated by a polyclonal B cell activator—e.g., LPS. In this case,Ig of unknown antigenic specificity is measured; thus, an indicator cell must be usedwhich binds Ig of all isotypes. The indicator cells are SRBC conjugated to protein A (seeSupport Protocol 4), a substance that binds both mouse and rabbit IgG with high avidity.Protein A–SRBC thus binds a developing rabbit anti-mouse Ig antibody that has beenadded to the B cell preparation and subsequently binds secreted mouse Ig. The resultingcomplex is then able to activate complement and effect SRBC lysis.

To assay polyclonal antibody response, prepare protein A–SRBC indicator cells accordingto Support Protocol 3. Proceed with step 1 of the Jerne-Nordin protocol (see BasicProtocol 3), then add 1 drop of 15% protein A–SRBC (see Support Protocol 4) suspendedin supplemented complete medium (instead of unmodified SRBC or TNP-SRBC) in step2. Proceed with steps 3 and 4, then add developing antibody (rabbit anti-IgM and anti-IgG)to the shallow well in the tray holding the slides. Incubate slides 1 hr at 37°C, then drawoff the developing antibodies with a Pasteur pipet connected to a vacuum suction flask.Proceed with the preparation and addition of complement as in steps 5 and 6.

It is best (but not essential) to use affinity-purified antibodies (Jackson Immunoresearch)that have been pretitrated to determine optimum dilution in supplemented completemedium.

SUPPORTPROTOCOL 1

PREPARATION OF CUNNINGHAM-SZENBERG CHAMBERS

Materials

70% ethanolMicroscope slides1⁄4-in. double-sided tape (3M)

1. Place 15 to 30 microscope slides side-by-side on a bench in a row.

2. Stick double-sided 1⁄4-in. tape on top and bottom edges. Place a third strip of tapebetween the other two, dividing slide into two sections (see Fig. 11.13.1).

3. Wipe slides with a paper towel soaked with 70% ethanol and allow to air dry.

4. Remove backing of double-sided tape and wipe a second slide with 70% ethanol, airdry, and place on first slide in the row. Overlap the slides on the side edges as in Figure11.13.1 to facilitate filling the chamber. Repeat this process so the entire row of slideson the bench are covered with a second slide.

5. Apply slight pressure with a roller to ensure that the slides are stuck firmly together.

Each slide chamber can hold just under 0.1 ml of fluid on either side of the tape for a totalof 0.2 ml per slide.


11.13.9

Immunology

SUPPORTPROTOCOL 2

PREPARATION OF GLASS SLIDES FOR THEJERNE-NORDIN ASSAY

Materials

1% agarose (SeaPlaque; FMC Bioproducts)Microscope slides2 × 2–in. sterile gauze pad (Johnson & Johnson)

1. Prepare microscope slides as for Cunningham-Szenberg chambers (see SupportProtocol 1), but omit the double-sided tape (i.e., follow steps 1 and 3).

2. Boil 1% agarose and dilute 1:10 in water (0.1% final).

3. Apply a thin film of 0.1% agarose to prepared slides (frosted end up) with a sterilegauze pad.

4. Allow slides to dry.

Precoated slides can be stored for 2 months at room temperature in a dust-free environment.

SUPPORTPROTOCOL 3

PREPARATION OF MODIFIED INDICATOR SRBC FOR PFC ASSAYS

TNP-Modified SRBC (Rittenberg and Pratt, 1969)


Modified barbital buffer (MBB; see recipe)2,4,6-trinitrobenzenesulfonic acid, sodium salt (TNBS; Eastman Kodak)Cacodylate buffer (see recipe)120 mg glycylglycine (gly-gly; Sigma #10022) in 15 ml MBB (see recipe for

MBB)15-ml graduated conical tube (Costar)

1. Remove SRBC from stock bottle and centrifuge 15 min in IEC 216 rotor at 3000 rpm(1525 × g). Aspirate supernatant and transfer 3 ml of packed SRBC to a tubecontaining 50 ml MBB.

2. Wash SRBC with 50 ml MBB by centrifuging 10 min at 1525 × g. Repeat the washingprocedure two more times. Resuspend SRBC in 50 ml MBB.

3. Add 60 mg TNBS to 21 ml cacodylate buffer in a 50-ml foil-covered tube (TNBS islight sensitive). Mix and allow the tube to stand 10 min at room temperature.

4. Slowly add 5 ml SRBC/MBB suspension to TNBS/cacodylate solution, while shak-ing. Cap tube and invert occasionally over the next 10 min.

5. Fill tube with MBB. Centrifuge 10 min in IEC 216 rotor at 2000 rpm (675 × g), 4°C,and discard supernatant.

6. Add glycylglycine solution to tube and resuspend SRBC. Place tube on ice 10 min.

Formation of glycylglycine solution is such that 20 mg of glycylglycine is used for every 10mg of TNBS.

7. Centrifuge 10 min at 675 × g and discard supernatant.

The supernatant should be yellow. If it is not, the conjugation step was not successful.

8. Wash SRBC twice in 50 ml MBB and at least once in supplemented completemedium, using the conditions in step 2. Continue washing until there is no evidenceof SRBC lysis (clear supernatant).


11.13.10


9. Spin cells at 675 × g in a graduated 15-ml conical tube to determine final packed cellvolume. Discard supernatant and resuspend SRBC in supplemented complete me-dium at a final concentration of 15% or 20% (v/v) for Cunningham-Szenberg orJerne-Nordin assays respectively.

TNP-SRBC can be stored ∼1 week but should be washed prior to use and should not beused until lysis is absent. (i.e., until supernatant stays clear after washing). Fresh SRBCmake better targets and are more likely to survive the conjugation procedure.

SUPPORTPROTOCOL 4

Protein A–Modified SRBC (Gronowicz et al., 1976)

Materials

Saline (0.15 M NaCl; Life Technologies)0.5 mg/ml protein A (Pharmacia Biotech) in saline (store frozen in saline at 1

mg/ml)1× chromic chloride solution (see recipe)15-ml conical test tube (Costar)

1. Wash SRBC in 50 ml saline by centrifuging 10 min in IEC-216 rotor at 2000 rpm(675 × g), 4°C. Repeat the washing procedure two more times.

2. Pipet 0.5 ml packed cells into a 15-ml conical test tube.

3. Add 0.5 ml freshly prepared 0.5 mg/ml protein A and 5 ml of 1× chromic chloridesolution to SRBC. Incubate 1 hr at room temperature. Invert tube gently every 10min.

4. Wash twice in 50 ml saline and once in 50 ml supplemented complete medium asdescribed in step 1.

5. Resuspend SRBC at a final concentration of 15% (v/v) in supplemented completemedium.

Protein A–SRBC are best used on the same day, but will keep for 3 days at 4°C.

SUPPORTPROTOCOL 5

ISOLATION OF B CELLS BY PERCOLL GRADIENT CENTRIFUGATION

Resting B cells are isolated from a suspension of spleen cells by elimination of T cellswith an anti-Thy-1 MAb (Hathcock, 1991b), followed by enrichment using flotation onPercoll gradients (also see Kruisbeek and Shevach, 1991). However, instead of using thetwo layers of 55% and 70% Percoll described in Kruisbeek and Shevach (1991) (forenrichment of accessory cells), B cell enrichment requires four layers of 55%, 60%, 65%,and 70% Percoll. All other conditions are identical to those described in Kruisbeek andShevach (1991). Resting B cells are collected from the 65% to 70% interface aftercentrifugation.


Hank’s balanced salt solution (HBSS; APPENDIX 2), ice coldPercoll (store at 4°C; Pharmacia Biotech #17-0891-01)Percoll mix solution (see recipe)

15- and 50-ml polypropylene centrifuge tubesSorvall H-1000B rotor (or equivalent)

Additional reagents and equipment for preparation of depleted B cell suspensions(Hathcock, 1991b; Mage, 1993), making cell suspensions (Kruisbeek, 1993),and counting cells (APPENDIX 3F)


11.13.11

Immunology

1. Prepare T cell–depleted B cell suspension from spleen. Resuspend cells in HBSSusing 1 ml of HBSS for each spleen.

Be sure to include anti-Thy-1, anti-CD4, and anti-CD8 antibodies to deplete all T cells. Ifdesired, deplete RBC present in spleen cell suspension using ACK lysing buffer, pH 7.4 (seerecipe); however, small RBC contamination usually does not interfere with proliferationstudies.

2. Prepare 70% Percoll by mixing 290 ml Percoll with 170 ml Percoll mix solution.Keep on ice at all times.

3. Prepare 50%, 60% and 66% Percoll as follows:

50% 60% 65% 70% Percoll, ice-cold (ml) 21.42 25.71 27.86HBSS, ice-cold (ml) 8.58 4.29 2.14

4. Prepare 50%/60%/65%/70% Percoll gradients in 15-ml conical centrifuge tubes:

a. Add 2.5 ml 70% Percoll to 15-ml tube.

b. Gently pipet 2.5 ml of 65% Percoll over 70% Percoll layer.

c. Gently pipet 2.5 ml of 60% Percoll over 65% Percoll layer.

d. Gently pipet 2.5 ml of 50% Percoll over 60% Percoll layer.Careful layering of Percoll is essential to create the sharp gradient interfaces necessaryfor good cell separation.

Use polypropylene tubes to withstand high-speed centrifugation. Make ≥4 gradients fora suspension of cells from 10 spleens and ≥6 gradients for a suspension of cells from20 spleens.

5. Allow prepared gradients to sit on ice for ≥15 min before adding cells and centrifug-ing.

Good cell separations depend on use of cold gradients.

6. Gently layer 2.5 ml of cell suspension (from step 1) onto the surface of each gradient.Add cell suspension slowly and down side of tube so as not to break gradient surface.

7. Centrifuge gradients 13 min in Sorvall H-1000B rotor at 3000 rpm (∼1900 × g), 4°C.The brake must be off.

8. Aspirate HBSS and 50% Percoll layer; then, with a pipet, separately collect cells ateach interface between remaining layers and place into 50-ml centrifuge tubes.

Cells at the 65%/70% interface are the small, resting B cells, whereas those at the 50%/60%interface are the large, activated B cells. Cells at the intermediate 60%/65% interface aremainly small, resting B cells; these cells can be used as resting cells if absolutely necessary.

9. Wash cells 3 times with HBSS, centrifuging at 1200 to 1500 rpm (300 × g) each time.

10. Count cells after final wash.


11.13.12



ACK lysing buffer8.29 g NH4Cl (0.15 M)1 g KHCO3 (10.0 mM)37.2 mg Na2EDTA (0.1 mM)Add 800 ml H2OAdjust pH to 7.2 to 7.4 with 1 N HClAdd H2O to 1 literFilter sterilize through a 0.2-µm filterStore at room temperature

AntigensThe following antigens can be used for in vitro antibody production (Basic Protocol1). Note that when used in conjunction with nonimmunized, purified B cells, theseantigens must be used with lymphokines.Sheep red blood cells (SRBC): T cell–dependent, particulate antigen which servesas a relatively potent in vitro immunogen; does not require the use of cells from invivo primed animals (for IgM response). Several batches may need to be screenedto obtain cells that stimulate reproducible responses; once found, use that sourceexclusively. This antigen results in nonspecific B cell response to various epitopes.SRBC can be purchased from Becton Dickinson (#11943).Ovalbumin (OVA): Soluble protein that is a relatively weak immunogen and requiresthe use of cells from in vivo primed animals. Elicits nonspecific B cell response;however, trinitrophenylation of the protein (TNP-OVA) results in production ofantibodies to the specific epitope. However, in this case, T cells must be primed tothe carrier molecule (OVA), and B cells must be primed to the hapten molecule(TNP). TNP-OVA can be prepared by the method of Eisen et al. (1953).Trinitrophenylated Ficoll and dextran (TNP-Ficoll and TNP-dextran): “Type 2,”high-molecular-weight multiepitope antigens that have no intrinsic polyclonalactivating properties and stimulate responses only in immunologically mature Bcells (CBA/N mice do not respond, but nude mice do). Use of these antigens requiresthe addition of T cell-derived lymphokines (IL-1 and IL-2; alternatively, IL-5) orthe use of B cells partially activated in vivo (see commentary discussion of Tcell-regulated antigens). TNP-Ficoll can be purchased from Biosearch or preparedby the method of Inman (1975).Trinitrophenylated Brucella abortus and lipopolysaccharide (TNP-BA and LPS):“Type 1” antigens that have intrinsic polyclonal activating properties and canstimulate responses in cell from all mouse strains (immature neonatal B cells as wellas B cells from xid immune-defective mice). Use of these T cell–regulated antigensrequires addition of lymphokines (IL-2 only) or use of B cells partially activated invivo (see commentary). Killed B. abortus is available from the U.S. Department ofAgriculture (National Veterinary Services Labs, P.O. Box 844, Ames, IA 50010)and is trinitrophenylated as described by Mond et al. (1978). LPS is available fromDifco as E. coli 0111:B4.

Barbital buffer, modified (MBB), 5× stock solution2.875 g barbituric acid1.875 g sodium barbital0.1103 g CaCl2⋅2H2O0.238 g MgCl2⋅6H2O42.5 g NaClH2O to 1 literDilute 1:5 for working solution


11.13.13

Immunology

Cacodylate buffer (0.28 M)Dissolve 24 g cacodylic acid in deionized water (∼300 ml). Adjust to pH 6.9 withconcentrated HCl. Dilute to 400 ml. Store at room temperature ≤6 months.

Chromic chloride solution (2.5 × 10−4 M)For 200× stock, dissolve 134 mg chromic chloride in 10 ml saline (not buffered).Mix well. Dilute to 1× with saline just before use.

DMEM-10 medium, supplemented, completePrepare in modified Dulbeccos minimal essential medium (DMEM):10% fetal bovine serum (FBS; APPENDIX 3F)1 mM sodium pyruvate2 mM L-glutamine50 µg/ml gentamycin0.1 mM nonessential amino acids0.3% sodium bicarbonate25 mM HEPES50 µM 2-mercaptoethanol (2-ME)

Modified DMEM contains L-glutamine, 1000 mg glucose/liter, and phenol red but does notcontain sodium bicarbonate. L-glutamine is labile and is therefore added fresh. The mediumcan be stored up to 1 month at 4°C.

Fetal bovine serum (FBS) is a critical element in obtaining good in vitro responses for anyprotocols in this unit. It is not unusual to have to screen 10 to 20 different batches of FBS toobtain a batch that supports high specific responses but low background responses. Reliablesources of FBS are Hyclone and Armour.

Guinea pig serumTo use as a source of complement, preabsorb with TNP-SRBC (see Support Protocol3) 30 min on ice using 1 ml packed cells and 5 ml serum. Complement may berefrozen and stored at −70°C. Commercial sources such as Life Technologies haveworked well in the PFC assays.Complement absorbed with TNP-SRBC can also be used for unmodified SRBC.Complement does not have to be absorbed for the Jerne-Nordin assay (BasicProtocol 3), as the agarose absorbs cytotoxic activity during the assay.

Percoll mix solution45 ml 10× phosphate-buffered saline (PBS; APPENDIX 2)3 ml 0.6 N HCl (1:20 dilution of 12 N HCl; J.T. Baker)132 ml H2OCheck that pH is 7.0 to 7.2Filter sterilize using 0.22-µm filter; store at 4°C

RPMI-10 medium, supplemented, completePrepare in complete RPMI/10% FBS (APPENDIX 3F):25 mM HEPES1 mM sodium pyruvate50 µg/ml gentamycin0.25 µg/ml Fungizone (this supplementation includes 100 penicillin and 100

µg/ml streptomycin; Whittaker #17-745A)

See note regarding testing of FBS in recipe above for supplemented DMEM-10.


11.13.14


COMMENTARY

Background InformationIn vitro antibody production by B cells fol-

lowing stimulation with T cell–dependent an-tigens (i.e., antigens that activate B cells onlyif antigen-specific T cells are present to provideessential help) involves a number of discretecellular events. Initially, antigen is taken up bythe B cell as a result of binding to and cross-linkage of specific Ig receptors on the cellsurface; this is followed by expression of theprocessed antigen on the cell surface in thecontext of MHC class II antigens. Next, thereis B cell presentation of the processed antigento the T cell, which leads to interactions be-tween the T cell receptor and the antigen–MHCclass II complex as well as between adhesionantigens on the T cell and B cell. These cell-surface events lead to stimulation of both inter-acting cells—the T cells producing growth anddifferentiation factors (lymphokines) and the Bcells differentiating into antibody-producingplasma cells. Macrophages also play a role inthis process as antigen-presenting cells thatinteract with T cells early in the process to bringabout T cell activation and lymphokine produc-tion; such interaction may be critical since Bcells require T cell–derived cytokines for opti-mum activation and differentiation. It shouldbe noted that a given clone of B cells recognizesone determinant (usually called the haptenicdeterminant) on the antigen via the Ig receptor,whereas the T cell usually recognizes anotherdeterminant (usually called the carrier determi-nant) expressed on the processed antigen.

B cells may also be activated and induced tosecrete immunoglobulin in the absence of an-tigen-specific T cells by cross-linkage of the Igreceptor by antigens containing multiple copiesof the same determinant (multivalent antigens).These so-called T cell–independent antigens(such as TNP-Ficoll or TNP-dextran) wereoriginally felt to be capable of activating B cellsin the complete absence of T cells; however, itis now known that even these antigens requirethe help of T cell-derived lymphokines to in-duce antibody production by resting B cells. Itis therefore more appropriate to refer to thisgroup of antigens as T cell–regulated antigens.These antigens can, however, induce B cellsthat have been partially activated in vivo torespond even in the absence of cytokine-medi-ated help.

The culture system for measuring in vitroantibody production described in this unit isessentially a miniaturized version of the system

originally developed by Mishell and Dutton(1966, 1967). Much early information about Tcell and B cell function was derived using thissystem in conjunction with mouse spleen cellsas a source of T and B cells and heterologousred blood cells as a source of antigen. Oneadvantage of using SRBC to study in vitro Tand B cell responses is that this antigen inducesresponses in cells obtained from unprimed ani-mals, probably because animals are naturallyprimed in vivo with antigens that cross-reactwith those associated with SRBC. However,under the usual in vitro conditions, the responseof unprimed cells consists mainly of IgM anti-body production; thus, either priming of the Tcell/B cell source or addition of lymphokinesto the culture system is necessary if IgG re-sponses are desired. One disadvantage of usingSRBC (and other complex antigens) is that alarge number of B cell clones with poorlydefined anti-RBC specificities are activatedduring the response. To study more restrictedresponses (of individual cell clones), it is nec-essary to use purified peptide antigens as theimmunogen; however, in this case the responseis poor with regard to levels of antibody pro-duced unless T cells from primed animals areused or lymphokines are added to the cultures.

Critical ParametersStimulation of antibody production in vitro

can lead to inconsistent and, at times, unreliableresults. It is recommended that a laboratorystarting this technology obtain reagents andcells from an experienced laboratory group.The novice should also consult the originalliterature cited in this unit to gain a fuller senseof the many parameters affecting in vitro re-sponse, as well as review the strategic planningsection at the beginning of this unit.

Responder cells. B cells from pathogen-freemice should be used. B cells obtained frommice that are chronically infected with endemicviruses show poor in vitro responses. Healthy,uninfected mice are available from JacksonLaboratory (APPENDIX 4) and the National Can-cer Institute (see Silverman and La Via, 1976).A purer and more homogeneous population ofB lymphocytes permits a more reliable inter-pretation of the results. However, as mentionedin strategic planning, many culture systemsrequire the presence of macrophages. Finally,defined T cell populations should be utilized.In this regard, the use of T cell populationsdefined by various surface antigens, organs or


11.13.15

Immunology

origin—as well as other parameters orclonal/hybridoma T cell populations—shouldbe considered.

Media components. Fetal calf serum of dif-ferent sources and batches vary in their abilityto support in vitro B cell responses (Mishell andDutton, 1967; Shiigi and Mishell, 1975). It isbest to test as many different types of FCS aspossible to find one that gives a low backgroundresponse and high specific response. When dif-ferent sources of FCS are to be tested, be surethat the supplier has available >20 to 30 litersof a particular lot, as it is not worth the time andeffort to test a batch containing only a few liters.2-mercaptoethanol is essential for most in vitromurine B cell responses. The 2-ME stockshould be replaced every 2 months.

RBC stimulator cells. For use as an antigenin in vitro B cell systems and in PFC assays,the SRBC should be as fresh as possible. Fordirect anti-SRBC responses, they should be <2months old, and different batches should betested for efficacy of response (McCarthy andDutton, 1975a,b). If anti-TNP or –protein Aresponses are being examined, the cells shouldbe <1 week old before conjugation, becauseSRBC become increasingly fragile with age.When older SRBC are conjugated, they lysevery easily, making it difficult to enumerateplaques.

Incubator. The quality of the incubator isvery important. It must be capable of maintain-ing temperature, humidity, and percent CO2

extremely accurately.Cunningham-Szenberg assay. The guinea

pig serum must be absorbed properly as de-scribed in reagents and solutions. In addition,the complement should be used within 20 minof thawing (during the assay). All reagentsshould be at room temperature before they areadded to the chambers, because if they are cold,air bubbles form in the chambers during incu-bation, making plaques difficult to count.Chambers should be sealed with wax within 20min of filling to prevent the chamber fromdrying out.

Jerne-Nordin assay. It is essential that theagarose does not cool below 42°C before it ispoured onto the slides. After the slides arepoured, the agarose must be solidified by al-lowing to cool to room temperature before theslides are inverted and put on the racks.

Antigen removal. When assaying super-natants for specific antibody responses, it isimportant to wash the cells to remove the anti-gen prior to collecting supernatants for theELISA. The timing of antigen removal and

collection of supernatants will depend on thesystem being employed. In general, antigen canbe removed 3 to 6 days after the onset of culture,and supernatant can be collected after 4 to 7days. In studying polyclonal immunoglobulinsecretion induced by lipopolysaccharide orlymphokine, there is no need to remove thesestimuli, as they do not interfere with the ELISA.

Manipulations of cells. Most importantly,when working with cells in vitro one shouldalways be aware that any manipulation, how-ever minor, may influence the responsivenessof the cell and may lead to variability in theresults. To minimize these occurrences, atten-tion should be paid to treating cells with care—e.g., not centrifuging at speeds greater than1000 rpm, not vortexing for prolonged periodsof time, and maintaining cells on ice in mediumthat is appropriately buffered and supple-mented with FBS.

Anticipated ResultsFor antibody secretion, the number of PFC

generated will depend on the nature of thestimulus and/or the presence of added lympho-kines. However, typical responses for 106 puri-fied B cells cultured with stimulatory lympho-kines are presented in Table 11.13.1.

Time ConsiderationsPreparing purified resting B cells by Percoll

gradient centrifugation (Support Protocol 5)takes ∼4 hr, while preparing antigen and dis-tributing the cells into the plates normally takes∼2 hr. The exact length of time will depend onthe size and complexity of the experiment andthe skill of the investigator.

Supernatants for ELISA—washing theplates to remove excess antigen takes 0.5 hr;harvesting supernatants for assay takes 0.5 hr(the ELISA takes ∼7 hr).

Preparation of modified SRBC—TNP-SRBC and Protein A–SRBC both take ∼3 hr.

Jerne-Nordin assay—precoating the slidestakes ∼15 min; the maximum size of an experi-ment is three 96-well plates, which would takean experienced worker 8 hr.

Cunningham-Szenberg assay—preparationof slides takes ∼1 hr per 120 slides; absorptionof complement takes ∼2 hr; an experiencedworker can assay six 96-well plates in 8 hr.

Literature CitedCunningham, A.J. and Szenberg, A. 1968. Further

improvements in the plaque technique for detect-ing single antibody forming cells. Immunology14:599-600.


11.13.16


Eisen, H., Belman, S., and Carsten, M. 1953. Thereaction of 2,4-dinitrobenzene sulfonic acid withfree amino groups of proteins. J. Am. Chem. Soc.75:4583-4591.

Gronowicz, E., Coutinho, A., and Melchers, F. l976.A plaque assay for all cells secreting Ig of a giventype or class. Eur. J. Immunol. 6:588-590.

Hathcock, K. 1991a. T cell enrichment by cytotoxicelimination of B cells and accessory T cells. InCurrent Protocols in Immunology (J.E. Coligan,A.M. Kruisbeek, D.H. Margulies, E.M. Shevach,and W. Strober, eds.), pp. 3.3.1-3.3.5. John Wiley& Sons, New York.

Hathcock, K. 1991b. T cell depletion by cytotoxicelimination. In Current Protocols in Immunol-ogy (J.E. Coligan, A.M. Kruisbeek, D.H. Mar-gulies, E.M. Shevach, and W. Strober, eds.), pp.3.4.1-3.4.3. John Wiley & Sons, New York.

Inman, J. K., 1975. Thymus-independent antigens:The preparation of covalent, hapten-Ficoll con-jugates. J. Immunol. 114:704-709.

Jerne, N.K. and Nordin, A.A. 1963. Plaque forma-tion in agar by single antibody producing cells.Science 140:405-408.

Kruisbeek, A.M. and Shevach, E. 1991. Proliferativeassays for T cell function. In Current Protocolsin Immunology (J.E. Coligan, A.M. Kruisbeek,D.H. Margulies, E.M. Shevach, and W. Strober,eds.), pp. 3.12.1-3.12.14. John Wiley & Sons,New York.

Lycke, N.Y. and Coico, R. 1996. Measurement ofimmunoglobulin synthesis using the ELISPOTassay. In Current Protocols in Immunology (J.E.Coligan, A.M. Kruisbeek, D.H. Margulies, E.M.Shevach, and W. Strober, eds.), pp. 7.14.1-7.14.9. John Wiley & Sons, New York.

Mage, M.G. 1993. Fractionation of T cells and Bcells. In Current Protocols in Immunology (J.E.Coligan, A.M. Kruisbeek, D.H. Margulies, E.M.Shevach, and W. Strober, eds.), pp. 3.5.1-3.5.6.John Wiley & Sons, New York.

McCarthy, M.M. and Dutton, R.W. 1975a. The hu-moral response of mouse spleen cells to twotypes of sheep erythrocytes. I. Genetic control ofthe response to H and L SRBC. J. Immunol.115:1316-1321.

McCarthy, M.M. and Dutton, R.W. 1975b. The hu-moral response of mouse spleen cells to twotypes of sheep erythrocytes. II. Evidence forgene expression in the B lymphocyte. J. Immu-nol. 115: 1322-1329.

Mishell, R.I. and Dutton, R.W. 1966. Immunizationof normal mouse spleen cell preparations in vi-tro. Science 153:1004-1006.

Mishell, R.I. and Dutton, R.W. 1967. Immunizationof dissociated spleen cell cultures from mice. J.Exp. Med. 126:423-442.

Mitchison, N.A. 1971. The carrier effect in the sec-ondary response to hapten-protein conjugates.II. Cellular cooperation. Eur. J. Immunol. 1:18-27.

Mond, J.J., Scher, I., Mozier, D.E., Blacsy, M., andPaul, W.E. l978. T independent responses in Bcell defective CBA/N mice to Brucella abortusand to trinitrophenyl (TNP) conjugates ofBrucella abortus. Eur. J. Immunol. 8:459-463.

Mond, J.J., Farrar, W.E., Paul W.E., Fuller-Farrar, J.,Schaeffer, M., and Howard, M., 1983. T celldependence and factor reconstitution of in vitroantibody responses to TNP-B. abortus and res-toration of depleted responses with chroma-tographed fractions of a T cell derived factor. J.Immunol. 131:633-637.

Rittenberg, M.B. and Pratt, C. 1969. Anti-trini-trophenyl (TNP) plaque assay. Primary responseof BALB/c mice to soluble and particulate im-munogen. Proc. Soc. Exp. Biol. Med. l32:575.

Shiigi, S.M. and Mishell, R.I. 1975. Sera and the invitro induction of immune responses. I. Bacterialcontamination and the generation of good fetalbovine sera. J. Immunol. 115:741-744.

Silverman, M.S. and LaVia, M.F. 1976. Letters tothe editor, J. Immunol. 117: 2270-2279.

Thompson, C-B., Scher, I., Schaeffer, M., Lindsten,T., Finkelman F.D. and Mond, J.J. 1984. Sizedependent B lymphocyte subpopulations: Rela-tionship of cell volume to surface phenotype, cellcycle, proliferative response and requirement forantibody production to TNP-Ficoll and TNP-BA. J. Immunol. 133:2333-2342.

Contributed by James J. Mond and Mark BrunswickUniformed Services University of the Health SciencesBethesda, Maryland


11.13.17

Immunology

UNIT 11.14Purification of Immunoglobulin G Fractionfrom Antiserum, Ascites Fluid, or HybridomaSupernatant

This unit describes the isolation of the immunoglobulin G (IgG) fraction (containingantibodies of all specificities) from a complex protein mixture such as antiserum, ascitesfluid, or hybridoma supernatant. The Basic Protocol utilizes saturated ammonium sulfatesolution to precipitate the IgG fraction, while the Alternate Protocol describes fractiona-tion of IgG by chromatography on DEAE–Affi-Gel Blue resin. The purification of IgGby affinity chromatography utilizing staphylococcal protein A or antigen-Sepharose isdescribed in UNIT 11.11.

BASICPROTOCOL

PRECIPITATION OF IgG WITH SATURATED AMMONIUM SULFATE

Antiserum or ascites fluid is adjusted to 33% with respect to the concentration of saturatedammonium sulfate, resulting in the precipitation of the IgG. At this concentration ofsaturated ammonium sulfate, a large percentage—but not all—of the contaminatingprotein species present in the antiserum or ascites remains in solution.

Materials

Saturated ammonium sulfate (SAS) solution (see recipe)33% SAS solution (see recipe)Immunoglobulin-containing antiserum, ascites, or tissue culture supernatant

Additional reagents and equipment for dialysis (APPENDIX 3C)

1. Add 1 vol SAS solution dropwise with a Pasteur pipet to 2 vol antiserum, ascites, ortissue culture supernatant with constant mixing at 4°C.

2. Allow precipitate to form over a period of 2 to 4 hr at 4°C with constant mixing. Pelletprecipitate by centrifugation for 20 min at 12,000 × g, 4°C.

3. Wash pellet by resuspending (vortexing) it in a volume of cold 33% SAS solutionequivalent to the original volume of antiserum, ascites, or tissue culture supernatant.Repeat wash step once.

4. Dissolve pellet in appropriate cold buffer by gentle vortexing.

A convenient volume for solubilizing the IgG fraction is 5% to 10% of the originalantiserum, ascites, or hybridoma supernatant volume.

5. Dialyze IgG solution over 48 hr at 4°C against three changes of the desired buffer (4liters per change) to fully remove the ammonium sulfate.

See APPENDIX 3C for preparing dialysis membrane (molecular weight cutoff 12,000 to14,000; Spectrapor 2 or equivalent) and for large-volume dialysis.

Supplement 50

Contributed by Helen M. Cooper and Yvonne PatersonCurrent Protocols in Molecular Biology (2000) 11.14.1-11.14.5Copyright © 2000 by John Wiley & Sons, Inc.

11.14.1

Immunology

ALTERNATEPROTOCOL

FRACTIONATION OF IgG BY CHROMATOGRAPHY ONDEAE–AFFI-GEL BLUE

This alternate procedure can be used to isolate the IgG fraction from either ascites orantiserum or to further purify the IgG fraction obtained after SAS precipitation. Underlow salt concentrations, many protein species present in ascites or antiserum, includingIgG, are bound to DEAE–Affi-Gel Blue resin with varying affinities. When increasingthe salt concentration in small increments, the IgG fraction is selectively eluted, leavingthe majority of the contaminating protein species still bound to the resin. This protocoldescribes the fractionation of murine IgG.


Loading buffer (see recipe)Elution buffer (see recipe)NaN3, crystalline formDEAE–Affi-Gel Blue (Bio-Rad)

Additional reagents and equipment for gel-filtration chromatography (UNIT 10.9)and SDS-polyacrylamide gel electrophoresis (UNIT 10.2)

1. Prepare a column of DEAE–Affi-Gel Blue according to manufacturer’s instructions(see Fig. 10.9.1) and equilibrate with five bed volumes of loading buffer.

It is recommended that a total bed volume of 7 ml/ml antiserum or ascites be used.

This and subsequent steps should be carried out at 4°C.

2. Dialyze IgG-containing sample against two changes of loading buffer (4 liters perchange) for ∼40 hr at 4°C.

3. Apply sample to the column at a flow rate of 10 ml/hr. Elute unbound protein withthree bed volumes of loading buffer at the same flow rate.

Alternatively, elution of the unbound material may be carried out overnight; however, atleast three bed volumes of loading buffer must be run down the column before the specificelution of the IgG fraction.

4. Elute bound IgG fraction with elution buffer at a flow rate of 10 ml/hr. Collect 10-mlfractions and store at 4°C until they are pooled.

Transferrin (Mr76,000) usually coelutes with the IgG fraction.

5. Identify fractions containing IgG by analyzing 30 to 50 µl of each fraction on a 10%SDS–polyacrylamide gel (UNIT 10.2), under reducing conditions (see UNIT 10.2).

Under reducing conditions, the IgG heavy and light chains will run separately withmolecular weights of 50,000 and 25,000, respectively.

The amount of transferrin contaminating the IgG will be determined at this step. Theamount present will vary depending on the source of the antibody. It can be removed bygel-filtration chromatography (UNIT 10.9).

6. Pool fractions containing IgG and dialyze over 40 hr at 4°C against two changes ofthe desired buffer (4 liters per change).

7. Re-equilibrate DEAE–Affi-Gel Blue column with five bed volumes of loading buffer,add a few crystals of NaN3 to retard bacterial growth, and store at 4°C.


11.14.2

Purification of IgGFraction from

Antiserum, AscitesFluid, or Hybridoma

Supernatant


Elution buffer (20 mM Tris⋅Cl/50 mM NaCl, pH 8.0)2.4 g Tris base2.92 g NaCl900 ml H2OAdjust to pH 8.0 with concentrated HClH2O to 1 liter

Loading buffer (20 mM Tris⋅Cl/30 mM NaCl, pH 8.0)2.4 g Tris base1.75 g NaCl900 ml H2OAdjust to pH 8.0 with concentrated HClH2O to 1 liter

Saturated and 33% saturated ammonium sulfate (SAS) solution450 g ammonium sulfateH2O to 500 mlHeat solution on a heated stirring plate until ammonium sulfate dissolves com-pletely. Filter solution while it is still warm and then allow it to cool. Upon cooling,crystals will form and should not be removed. Check pH of cooled solution; adjustto pH 7.5 with ammonium hydroxide. For 33% SAS solution, mix 33 ml SASsolution with 67 ml phosphate-buffered saline (PBS; APPENDIX 2).

A quantity of 760 g of ammonium sulfate is required per 1000 ml for a 100% saturatedsolution. A total of 450 g/500 ml is used to ensure sufficient crystal formation to maintainsaturation. A slightly different recipe for saturated ammonium sulfate solution in Tris bufferis contained in UNIT 11.1 and may be substituted in this protocol.

COMMENTARY

Background InformationAffinity chromatography, in which the rele-

vant protein antigen or peptide antigen is cou-pled to an affinity resin (see UNIT 11.11), is themethod of choice for the purification of specificantibody. However, under certain circum-stances, such purification procedures are eitherunnecessary or not applicable (e.g., when theantigen is not available in adequate quantities).In this unit two procedures are described for theisolation of the IgG fraction from ascites orantiserum. In both cases, however, the total IgGcomponent is isolated.

The precipitation of protein using ammo-nium sulfate is achieved by dehydration in themicroenvironment of the protein molecule. Insolution, a large number of water molecules arebound to the sulfate ion (SO4

=), significantlyreducing the amount of water available to in-teract with the protein molecules. At a particu-lar concentration of ammonium sulfate an in-sufficient quantity of unbound water will re-main to keep a given protein species in solution,resulting in the precipitation of that protein. Theexact concentration of ammonium sulfate re-

quired to precipitate a protein will be deter-mined by the characteristics of the amino acidside chains exposed to the solvent. Precipitationof the IgG fraction of antiserum or ascites fluidat an ammonium sulfate concentration of 33%efficiently isolates the IgG fraction from majorcontaminating proteins such as albumin andhemoglobin, yielding an IgG preparation thatis relatively pure. However, a significant num-ber of other proteins will coprecipitate. There-fore, this fractionation protocol serves as anexcellent initial step in the purification of IgG.The SAS-precipitated IgG fraction can then befurther purified by immunoaffinity chromatog-raphy or on a DEAE–Affi-Gel Blue column(see Alternate Protocol). In addition, the IgGfraction can be stored for a long period of time(1 to 2 years) as a suspension of precipitate inSAS solution without loss of antibody-bindingactivity. Fresh antiserum or ascites can bebrought to the desired ammonium sulfate con-centration and safely stored until further puri-fication of the IgG fraction is required.

Affinity chromatography using DEAE–Affi-Gel Blue is a combination of dye-interac-


11.14.3

Immunology

tion and ion-exchange (IEX) chromatography.The DEAE–Affi-Gel Blue is a bifunctionalresin bearing both diethylaminoethyl groupsand Cibacron F3GA molecules. The dye-inter-action chromatographic function is based onthe ability of the dye, Cibacron blue F3GA, tobind a large variety of enzymes and other pro-teins with a range of affinities. A comprehen-sive discussion on the nature and use of thisdye, when attached to an affinity chromatogra-phy matrix, is given in the technical literaturefound on the Bio-Rad website (http://www.bio-rad.com). Blue Sepharose 6 Fast Flow is theequivalent matrix produced by AmershamPharmacia Biotech (http://www.apbiotech.com). In addition, the IEX moieties on the resininteract with the remaining acidic proteins inthe antiserum or ascites. Overall, a large per-centage (70% to 80%) of contaminating pro-teins are separated from the IgG fraction underthe elution conditions described. The chroma-tographic conditions given in the Basic Proto-col for the selective elution of the immuno-globulin fraction from the DEAE–Affi-GelBlue column have been optimized for murineIgG. The manufacturers recommend that 0.02M K2HPO4 (adjusted to pH 8.0 with KOH)buffer be used when fractionating the IgG com-ponent from human antiserum. For isolating theIgG fraction from rabbit antiserum, it is recom-mended that 0.02 M Tris⋅Cl, pH 8.0/0.03 MNaCl (i.e., the loading buffer used in the Alter-nate Protocol) be used. It is stated that thehuman and rabbit IgG fractions do not bind tothe column when using these respective buffers.

Pure preparations of immunoglobulins ofthe G and M isotypes can be achieved by avariety of other chromatographic procedures,including gel filtration and IEX. These proto-cols are described in this manual (UNITS 10.9 &

10.10). The application of electrophoresis to theisolation of pure immunoglobulin species isalso described in the former volume.

Critical ParametersThe addition of SAS solution to the antis-

erum or ascites must be carried out slowly. Highlocal concentrations of ammonium sulfate willresult in the coprecipitation of other contami-nating protein species that usually precipitateat SAS concentrations >33%. The precipitationshould be carried out between 4° and 25°C.Outside this temperature range, the degree ofsaturation of ammonium sulfate deviates by>3%. Also, to maintain total saturation of theSAS solution, ensure that a significant quantity

of ammonium sulfate crystals are always pre-sent at the bottom of the saturated solution.

The optimal elution conditions isolatingmurine IgG using DEAE–Affi-Gel Blue havebeen described. As stated in background infor-mation, the optimal conditions for fractionatingIgG from other species will vary. It is alsopossible that individual monoclonal antibodiesmay elute under slightly different salt concen-trations than those given in the Basic Protocol.Therefore, the buffer system should be testedby running a small volume of the antibodycontaining sample down a mini-column ofDEAE–Affi-Gel Blue resin before committinga larger quantity of sample to the column.

TroubleshootingIn general, immunoglobulins are a very ho-

mogeneous family of proteins and will there-fore precipitate when the SAS solution is at aconcentration of 33%. In some instances, how-ever, precipitation may be only partial or maynot occur at all under these conditions. Thisproblem is more commonly observed whenattempting to precipitate an individual mono-clonal antibody from ascites fluid. Under suchcircumstances, it is necessary to increase theammonium sulfate saturation to 40% or 50%.It must be noted that the number of contami-nating protein species will increase with in-creasing ammonium sulfate concentration. Ifthe IgG fraction is not selectively eluted fromthe DEAE–Affi-Gel Blue column under theconditions described in the basic protocol andin background information, it may be necessaryto adjust the NaCl concentration in the elutionbuffer. However, do not increase the NaCl con-centration above 75 mM, as contaminating pro-tein species begin to elute above this level ofsalt concentration.

Anticipated ResultsAn efficient precipitation or ion-exchange

purification of the IgG fraction will be achievedusing these procedures. However, in manycases, transferrin will coprecipitate or copurifyand may be removed by gel-filtration chroma-tography (UNIT 10.9; IgG: MW 150,000; trans-ferrin: MW 76,000). The preferred procedurefor purifying antibodies from serum or ascitesfluid is by immunoaffinity chromatography ona ligand-bearing affinity matrix or with proteinA–Sepharose (UNIT 11.11).

Time ConsiderationsPrecipitation of the IgG fraction using SAS

can be completed within 4 to 6 hr. After allow-


11.14.4

Purification of IgGFraction from

Antiserum, AscitesFluid, or Hybridoma

Supernatant

ing 24 hr for dialysis, affinity chromatographyon DEAE–Affi-Gel Blue can be completedover a 24-hr period. It is usually convenient towash the unbound material through the columnovernight, using the loading buffer.

Contributed by Helen M. CooperWalter and Eliza Hall InstituteMelbourne, Australia

Yvonne PatersonUniversity of PennsylvaniaPhiladelphia, Pennsylvania


11.14.5

Immunology

SECTION IVPREPARATION OF ANTIPEPTIDEANTIBODIES

UNIT 11.15Introduction to Peptide Synthesis

DEVELOPMENT OF SOLID-PHASE PEPTIDE-SYNTHESISMETHODOLOGY

A number of synthetic peptides are signifi-cant commercial or pharmaceutical products,ranging from the dipeptide sugar substituteaspartame to clinically used hormones such asoxytocin, adrenocorticotropic hormone, andcalcitonin. Rapid, efficient, and reliable meth-odology for the chemical synthesis of thesemolecules is of utmost interest. The stepwiseassembly of peptides from amino acid precur-sors has been described for nearly a century.The concept is a straightforward one, wherebypeptide elongation proceeds via a coupling re-action between amino acids, followed by re-moval of a reversible protecting group. The firstpeptide synthesis, as well as the creation of theterm “peptide,” was reported by Fischer andFourneau (1901). Bergmann and Zervas (1932)created the first reversible Nα-protecting groupfor peptide synthesis, the carbobenzoxy (Cbz)group. DuVigneaud successfully applied early“classical” strategies to construct a peptide withoxytocin-like activity (duVigneaud et al.,1953). Classical, or solution-phase methods forpeptide synthesis have an elegant history andhave been well chronicled. Solution synthesiscontinues to be especially valuable for large-scale manufacturing and for specialized labo-ratory applications.

Peptide synthesis became a more practicalpart of present-day scientific research follow-ing the advent of solid-phase techniques. Theconcept of solid-phase peptide synthesis(SPPS) is to retain chemistry that has beenproven in solution but to add a covalent attach-ment step that links the nascent peptide chainto an insoluble polymeric support (resin). Sub-sequently, the anchored peptide is extended bya series of addition cycles (Fig. 11.15.1). It isthe essence of the solid-phase approach thatreactions are driven to completion by the useof excess soluble reagents, which can be re-moved by simple filtration and washing with-out manipulative losses. Once chain elongationhas been completed, the crude peptide is re-leased from the support.

In the early 1960s, Merrifield proposed theuse of a polystyrene-based solid support forpeptide synthesis. Peptides could be assembledstepwise from the C to N terminus using Nα-protected amino acids. SPPS of a tetrapeptidewas achieved by using Cbz as an α-amino-pro-tecting group, coupling with N,N′-dicyclo-hexylcarbodiimide (DCC), and liberating thepeptide from the support by saponification orby use of HBr (Merrifield, 1963). SPPS waslater modified to use the t-butyloxycarbonyl(Boc) group for Nα protection (Merrifield,1967) and hydrogen fluoride (HF) as the re-agent for removal of the peptide from the resin(Sakakibara et al., 1967). SPPS was thus basedon “relative acidolysis,” where the Nα-protect-ing group (Boc) was labile in the presence ofmoderate acid (trifluoroacetic acid; TFA),while side-chain-protecting benzyl (Bzl)–based groups and the peptide/resin linkagewere stable in the presence of moderate acidand labile in the presence of strong acid (HF).The first instrument for automated synthesis ofpeptides, based on Boc SPPS, was built byMerrifield, Stewart, and Jernberg (Merrifield etal., 1966). From the 1960s through the 1980s,Boc-based SPPS was fine-tuned (Merrifield,1986). This strategy has been utilized for syn-thesis of proteins such as interleukin-3 andactive enzymes including ribonuclease A andall-L and all-D forms of HIV-1 aspartyl protease.

In 1972, Carpino introduced the 9-fluo-renylmethoxycarbonyl (Fmoc) group for Nα

protection (Carpino and Han, 1972). The Fmocgroup requires moderate base for removal, andthus offered a chemically mild alternative to theacid-labile Boc group. In the late 1970s, theFmoc group was adopted for solid-phase appli-cations. Fmoc-based strategies utilized t-butyl(tBu)–based side-chain protection and hy-droxymethylphenoxy-based linkers for peptideattachment to the resin. This was thus an “or-thogonal” scheme requiring base for removalof the Nα-protecting group and acid for removalof the side-chain protecting groups and libera-tion of the peptide from the resin. The milderconditions of Fmoc chemistry as compared toBoc chemistry—which include elimination ofrepetitive moderate acidolysis steps and the

Supplement 63

Contributed by Gregg B. FieldsCurrent Protocols in Molecular Biology (2002) 11.15.1-11.15.9Copyright © 2003 by John Wiley & Sons, Inc.

11.15.1

Immunology

final strong acidolysis step—were envisionedas being more compatible with the synthesis ofpeptides that are susceptible to acid-catalyzedside reactions. In particular, the modification ofthe indole ring of Trp was viewed as a particularproblem during Boc-based peptide synthesis(Barany and Merrifield, 1979), which could bealleviated using Fmoc chemistry. One exampleof the potential advantage of Fmoc chemistryfor the synthesis of multiple-Trp-containingpeptides was in the synthesis of gramicidin A.Gramicidin A, a pentadecapeptide containingfour Trp residues, had been synthesized pre-viously in low yields (5% to 24%) using Bocchemistry. The mild conditions of Fmoc chem-

istry dramatically improved the yields of grami-cidin A, in some cases up to 87% (Fields et al.,1989, 1990). A second multiple-Trp-containingpeptide, indolicidin, was successfully assem-bled in high yield by Fmoc chemistry (King etal., 1990). Thus, the mild conditions of Fmocchemistry appeared to be advantageous for cer-tain peptides, as compared with Boc chemistry.

One of the subsequent challenges for prac-titioners of Fmoc chemistry was to refine thetechnique to allow for construction of proteins,in similar fashion to that which had beenachieved with Boc chemistry. Fmoc chemistryhad its own set of unique problems, includingsuboptimum solvation of the peptide/resin,

A

X

NH CH C OH

O

+ linker resin

anchoring

A NH CH C

O

linker resin

N -deprotectionα

A NH CH C OH

O

H N CH C

O

linker resin2+

coupling

(1) N -deprotection(2) cleavage(3) side-chain deprotection

α

A NH CH C

OY

NH CH C

O

linker resin

H N2 CH C

OZ

NH CH C

OY

NH CH C

ONH2OHNH CH C

OR

n

repetitive cycle X

XY

X

X

Figure 11.15.1 Generalized approach to solid-phase peptide synthesis. Symbols: A, Nα-protect-ing group; circle, side-chain protecting groups; R, X, Y, and Z, side-chain functionalities.


11.15.2

Introduction toPeptide Synthesis

slow coupling kinetics, and base-catalyzed sidereactions. Improvements in these areas of Fmocchemistry (Atherton and Sheppard, 1987;Fields and Noble, 1990; Fields et al., 2001)allowed for the synthesis of proteins such asbovine pancreatic trypsin inhibitor analogs,ubiquitin, yeast actin-binding protein 539-588,human β-chorionic gonadotropin 1-74, mini-collagens, HIV-1 Tat protein, HIV-1 nucleocap-sid protein Ncp7, and active HIV-1 protease.

The milder conditions of Fmoc chemistry,along with improvements in the basic chemis-try, have led to a shift in the chemistry employedby peptide laboratories. This trend is best ex-emplified by a series of studies (Angeletti et al.,1997) carried out by the Peptide SynthesisResearch Committee (PSRC) of the Associa-tion of Biomolecular Resource Facilities(ABRF). The PSRC was formed to evaluate thequality of the synthetic methods utilized in itsmember laboratories for peptide synthesis. ThePSRC designed a series of studies from 1991to 1996 to examine synthetic methods and ana-lytical techniques. A strong shift in the chem-istry utilized in core facilities was observedduring this time period—i.e., the more seniorBoc methodology was replaced by Fmoc chem-istry. For example, in 1991 50% of the partici-pating laboratories used Fmoc chemistry, while50% used Boc-based methods. By 1994, 98%of participating laboratories were using Fmocchemistry. This percentage remained constantin 1995 and 1996. In addition, the overall qual-ity of the peptides synthesized improvedgreatly from 1991 to 1994. Possible reasons forthe improved results were any combination ofthe following (Angeletti et al., 1997):

1. The greater percentage of peptides syn-thesized by Fmoc chemistry, where cleavageconditions are less harsh.

2. The use of different side-chain protectinggroup strategies that help reduce side reactionsduring cleavage.

3. The use of cleavage protocols designedto minimize side reactions.

4. More rigor and care in laboratory tech-niques.

The present level of refinement of solid-phase methodology has led to numerous, com-mercially available instruments for peptidesynthesis (Table 11.15.1).

The next step in the development of solid-phase techniques includes applications for pep-tides containing non-native amino acids, post-translationally modified amino acids, andpseudoamino acids, as well as for organic mole-cules in general. Several areas of solid-phase

synthesis need to be refined to allow for thesuccessful construction of this next generationof biomolecules. The solid support must beversatile so that a great variety of solvents canbe used, particularly for organic-molecule ap-plications. Coupling reagents must be suffi-ciently rapid so that sterically hindered aminoacids can be incorporated. Construction of pep-tides that contain amino acids bearing post-translational modifications should take advan-tage of the solid-phase approach. Finally, ap-propriate analytical techniques are needed toassure the proper composition of products.

THE SOLID SUPPORTEffective solvation of the peptide/resin is

perhaps the most crucial condition for efficientchain assembly during solid-phase synthesis.Swollen resin beads may be reacted and washedbatch-wise with agitation, then filtered eitherwith suction or under positive nitrogen pres-sure. Alternatively, they may be packed in col-umns and utilized in a continuous-flow modeby pumping reagents and solvents through theresin. 1H, 2H, 13C, and 19F nuclear magneticresonance (NMR) experiments have shownthat, under proper solvation conditions, thelinear polystyrene chains of copoly(styrene-1%-divinylbenzene) resin (PS) are nearly asaccessible to reagents as if free in solution. 13Cand 19F NMR studies of Pepsyn (copolymer-ized dimethylacrylamide, N,N′-bisacry-loylethylenediamine, and acryloylsarcosinemethyl ester) have shown similar mobilities atresin-reactive sites as PS. Additional supportscreated by grafting polyethylene glycol(polyoxyethylene) onto PS—either by control-led anionic polymerization of ethylene oxideon tetraethylene glycol–PS (POE-PS) or bycoupling Nω-Boc– or Fmoc–polyethylene gly-col acid or –polyethylene glycol diacid toamino-functionalized PS (PEG-PS)—combinethe advantages of liquid-phase synthesis (i.e.,a homogeneous reaction environment) andsolid-phase synthesis (an insoluble support).13C NMR measurements of POE-PS showedthe polyoxyethylene chains to be more mobilethan the PS matrix, with the highest T1 spin-lat-tice relaxation times observed with POE ofmolecular weight 2000 to 3000. Other supportsthat have been developed that show improvedsolvation properties and/or are applicable toorganic synthesis include polyethylene glycolpolyacrylamide (PEGA), cross-linked acrylateethoxylate resin (CLEAR), and augmented sur-face polyethylene prepared by chemical trans-formation (ASPECT). As the solid-phase


11.15.3

Immunology

method has expanded to include organic-mole-cule and library syntheses, the diversity of sup-ports will enhance the efficiency of these newapplications.

Successful syntheses of problematic se-quences can be achieved by manipulation of thesolid support. In general, the longer the synthe-sis, the more polar the peptide/resin will be-come (Sarin et al., 1980). One can alter thesolvent environment and enhance coupling ef-ficiencies by adding polar solvents and/orchaotropic agents (Fields and Fields, 1994).Also, using a lower substitution level of resinto avoid interchain crowding can improve thesynthesis (Tam and Lu, 1995). During difficultsyntheses, deprotection of the Fmoc group canproceed slowly. By spectrophotometricallymonitoring deprotection as the synthesis pro-ceeds, one can detect problems and extendbase-deprotection times and/or alter solvationconditions as necessary.

COUPLING REAGENTSThe classical examples of in situ coupling

reagents are N,N′-dicyclohexylcarbodiimide(DCC) and the related N,N′-diisopropylcar-bodiimide (Rich and Singh, 1979). The gener-ality of carbodiimide-mediated couplings isextended significantly by the use of either 1-hydroxybenzotriazole (HOBt) or 1-hydroxy-7-azabenzotriazole (HOAt) as an additive, eitherof which accelerates carbodiimide-mediatedcouplings, suppresses racemization, andinhibits dehydration of the carboxamide sidechains of Asn and Gln to the correspondingnitriles. Protocols involving benzotriazol-1-yl-oxy-tr is(dimethylamino)phosphoniumhexafluorophosphate (BOP), benzotriazol-1-y l -oxy-t r i s(pyrrol idino)phosphoniumhexafluorophosphate (PyBOP), 7-azaben-zotriazol-1-yl-oxytris(pyrrolidino)phosphonium hexafluorophosphate (PyAOP), O-ben-zotriazol-1-yl-N,N,N′,N′-tetramethyluroniumhexafluorophosphate (HBTU), O-(7-azaben-

Table 11.15.1 Instruments for Solid-Phase Synthesis

Suppliera Instrument model Fmoc Boc Batch Flow Monitoring Scale (mmol) No. ofpeptides

Peptide synthesis systemsAdvanced ChemTech 90 Yes Yes Yes No No 0.1–12 1–2

Apex 396 Yes Yes Yes No No 0.005–1 96348 Yes Yes Yes No No 0.005–0.15 48357 Yes Yes Yes No No 0.005–0.25 42

Bachem Bioscience SP4000-LAB Yes Yes Yes No No 0.25–5 1SP4000-PRO Yes No Yes No No 5–50 1

Gilson/Abimed AMS422 Yes No Yes No Yes 0.005–1 1Perkin-Elmer ABI433A Yes Yes Yes No Yes 0.05–1 1

Pioneer-MPS Yes No No Yes Yes 0.005–0.1 16Rainin PS3 Yes Yes Yes No No 0.1–0.25 3

Sonata/Pilot Yes Yes Yes No No 0.1–50 1Symphony/Multiplex Yes No Yes No No 0.005–0.35 12

CS Bio CS100 Yes Yes Yes No No 0.05–1 1CS336 Yes Yes Yes No No 0.05–0.25 3036 Yes Yes Yes No No 0.1–2.5 1136 Yes Yes Yes No No 0.1–2.5 1CS536 Yes Yes Yes No No 0.2–25 1CS936S Yes Yes Yes No No ≤600 1CS936 Yes Yes Yes No No ≤12,500 1

Intavis AG AutoSpot Yes No Yes No No 3–4 nmol/mm2 384Multiple organicsynthesis unitsAdvanced ChemTech 384 Yes Yes Yes No No — 4 × 96

Vantage Yes Yes Yes No No — 96aFor contact information, see APPENDIX 4.


11.15.4


zotriazol-1-yl)-N,N,N′,N′-tetramethyluroniumhexafluorophosphate (HATU), and O-ben-zotriazol-1-yl-N,N,N′,N′-tetramethyluroniumtetrafluoroborate (TBTU) result in couplingkinetics even more rapid than that obtained withcarbodiimides. Amino acid halides have alsobeen applied to solid-phase peptide synthesis(SPPS). Nα-protected amino acid chlorideshave a long history of use in solution synthesis.Fmoc–amino acid chlorides and fluorides reactrapidly under SPPS conditions in the presenceof HOBt/N,N-diisopropylethylamine (DIEA)and DIEA, respectively, with very low levels ofracemization. For convenience, tetra-methylfluoroformamidinium hexafluorophos-phate (TFFH) can be used for automated prepa-ration of Fmoc–amino acid fluorides. Aminoacid fluorides have been found to be especiallyuseful for the preparation of peptides contain-ing sterically hindered amino acids, such aspeptaibols. All of the coupling reagents andadditives discussed here are commerciallyavailable (see Table 11.15.2).

SYNTHESIS OF MODIFIEDRESIDUES AND STRUCTURES

Peptides of biological interest often includestructural elements beyond the 20 geneticallyencoded amino acids. Particular emphasis hasbeen placed on peptides containing phosphory-lated or glycosylated residues or disulfidebridges. Incorporation of side-chain-phospho-rylated Ser and Thr by solid-phase peptide syn-thesis (SPPS) is especially challenging, as thephosphate group is decomposed by strong acidand lost with base in a β-elimination process.Boc-Ser(PO3phenyl2) and Boc-Thr(PO3phenyl2)have been found to be useful derivatives, wherehydrogen fluoride (HF) or hydrogenolysiscleaves the peptide/resin and hydrogenolysis re-moves the phenyl groups. Fmoc-Ser(PO3Bzl,H)and Fmoc-Thr(PO3Bzl,H) can be used in con-junction with Fmoc chemistry with some care.Alternatively, peptide/resins that were built up byFmoc chemistry to include unprotected Ser or Thrside chains may be subject to “global” or post-as-sembly phosphorylation. Side-chain-phosphory-lated Tyr is less susceptible to strong-acid decom-

Table 11.15.2 Coupling Reagents and Additives Used in Solid-Phase Peptide Synthesis and Suppliers

Reagent Abbreviation Supplier(s)a

N,N′-dicyclohexylcarbodiimide DCC A, ACT, AO, CI, CN, F, PI, PL, Q, S

N,N′-diisopropylcarbodiimide DIPCDI A, ACT, AO, CI, F, PE, Q, S

O-benzotriazol-1-yl-N,N,N′,N′-tetramethyluronium hexafluorophosphate

HBTU A, ACT, AS, CI, CN, F, NS, PI, PL, Q, S

O-benzotriazol-1-yl-N,N,N′,N′-tetramethyluronium tetrafluoroborate

TBTU A, ACT, B, CI, CN, F, NS, PE, PI, PL, Q, S

O-(7-azabenzotriazol-1-yl)-N,N,N′,N′- tetramethyluronium hexafluorophosphate

HATU PE

Benzotriazol-1-yl-oxy-tris(dimethylamino) phosphonium hexafluorophosphate

BOP A, ACT, AO, B, CI, CN, F, NS, PL, PI, Q, S

Benzotriazol-1-yl-oxy-tris(pyrrolidino)- phosphonium hexafluorophosphate

PyBOP A, ACT, AO, CI, CN, F, S

7-azabenzotriazole-1-yl-oxy-tris(pyrrolidino)- phophonium hexafluorophosphate

PyAOP PE

Tetramethylfluoroformamidinium hexafluorophosphate

TFFH ACT, PE

1-hydroxybenzotriazole HOBt A, ACT, AO, AS, CI, CN, NS, PE, PI, Q, S

1-hydroxy-7-azabenzotriazole HOAt PE

N,N-diisopropylethylamine DIEA A, ACT, AO, CI, F, PE, Q, S

N-methylmorpholine NMM A, AO, CI, F, S

aAbbreviations: A, Aldrich; ACT, Advanced ChemTech; AO, Acros Organics; AS, AnaSpec; B, Bachem; CI, Chem-Impex; CN,Calbiochem-Novabiochem; F, Fluka; NS, Neosystem/SNPE; PE, Perkin-Elmer; PI, Peptides International; PL, PeninsulaLaboratories; Q, Quantum Biotechnologies; S, Sigma. For contact information, see APPENDIX 4.


11.15.5

Immunology

position and is not at all base-labile. Thus, SPPShas been used to incorporate directly Fmoc-Tyr(PO3methyl2), Fmoc-Tyr(PO3tBu2), Fmoc-Tyr(PO3H2), and Boc-Tyr(PO3H2). Phosphoryla-tion may also be accomplished on-line, directlyafter incorporation of the Tyr, Ser, or Thr residuebut prior to assembly of the whole peptide.

Methodology for site-specific incorporationof carbohydrates during chemical synthesis ofpeptides has developed rapidly. The mild con-ditions of Fmoc chemistry are more suited forglycopeptide syntheses than Boc chemistry, asrepetitive acid treatments can be detrimental tosugar linkages. Fmoc-Ser, -Thr, -5-hydroxylys-ine (-Hyl), -4-hydroxyproline (-Hyp), and -Asnhave all been incorporated successfully withglycosylated side chains. The side-chain glyco-syl is usually hydroxyl-protected by either ben-zoyl or acetyl groups, although some SPPSshave been successful with no protection ofglycosyl hydroxyl groups. Deacetylation anddebenzylation are performed with hydra-zine/methanol prior to glycopeptide/resincleavage or in solution with catalytic methox-ide in methanol.

Disulfide-bond formation has beenachieved on the solid-phase by air, K3Fe(CN)6,dithiobis(2-nitrobenzoic acid), or diiodoethaneoxidation of free sulfhydryls, by direct depro-tection/oxidation of Cys(acetamidomethyl)residues using thallium trifluoroacetate or I2,by direct conversion of Cys(9-fluorenyl-methyl) residues using piperidine, and by nu-cleophilic attack by a free sulfhydryl on eitherCys(3-nitro-2-pyridinesulfenyl) or Cys(S-car-boxymethylsulfenyl). The most generally ap-plicable and efficient of these methods is directconversion of Cys(acetamidomethyl) residuesby thallium trifluoroacetate.

Intra-chain lactams are formed between theside-chains of Lys or Orn and Asp or Glu toconformationally restrain synthetic peptides,with the goal of increasing biological potencyand/or specificity. Lactams can also be formedvia side-chain-to-head, side-chain-to-tail, orhead-to-tail cyclization (Kates et al., 1994). Theresidues used to form intra-chain lactams mustbe selectively side-chain deprotected, while allside-chain protecting groups of other residuesremain intact. Selective deprotection is bestachieved by using orthogonal side-chain pro-tection, such as allyloxycarbonyl or 1-(4,4-di-methyl-2,6-dioxocyclohex-1-ylidene)ethylprotection for Lys and allyl or N-[1-(4,4,-di-methyl-2,6-dioxocyclohexylidene)-3-methylbutyl]aminobenzyl protection for Asp/Glu incombination with an Fmoc/tBu strategy. Cycli-

zation is carried out most efficiently with BOPin the presence of DIEA while the peptide isstill attached to the resin.

The three-dimensional orthogonal protec-tion scheme of Fmoc/tBu/allyl protectinggroups is the strategy of choice for head-to-tailcyclizations. An amide linker is used for side-chain attachment of a C-terminal Asp/Glu(which are converted to Asn/Gln) and the α-carboxyl group is protected as an allyl ester. Forside-chain-to-head cyclizations, the N-terminalamino acid (head) can simply be introduced asan Nα-Fmoc derivative while the peptide-resinlinkage and the other side-chain protectinggroups are stable to dilute acid or carry a thirddimension of orthogonality.

PROTEIN SYNTHESISThere are three general chemical approaches

for constructing proteins. First is stepwise syn-thesis, in which the entire protein is synthesizedone amino acid at a time. Second is “fragmentassembly,” in which individual peptide strandsare initially constructed stepwise, purified, andfinally covalently linked to create the desiredprotein. Fragment assembly can be divided intotwo distinct approaches: (1) convergent synthe-sis of fully protected fragments, and (2) che-moselective ligation of unprotected fragments.Third is “directed assembly,” in which individ-ual peptide strands are constructed stepwise,purified, and then noncovalently driven to as-sociate into protein-like structures. Combina-tions of the three general chemical approachesmay also be employed for protein construction.

Convergent synthesis utilizes protected pep-tide fragments for protein construction (Al-bericio et al., 1997). The advantage of conver-gent protein synthesis is that fragments of thedesired protein are first synthesized, purified,and characterized, ensuring that each fragmentis of high integrity; these fragments are thenassembled into the complete protein. Thus, cu-mulative effects of stepwise synthetic errors areminimized. Convergent synthesis requiresready access to pure, partially protected peptidesegments, which are needed as building blocks.The application of solid-phase synthesis to pre-pare the requisite intermediates depends onseveral levels of selectively cleavable protect-ing groups and linkers. Methods for subsequentsolubilization and purification of the protectedsegments are nontrivial. Individual rates forcoupling segments are substantially lower thenfor activated amino acid species by stepwisesynthesis, and there is always a risk of racemi-zation at the C-terminus of each segment. Care-


11.15.6


ful attention to synthetic design and executionmay minimize these problems.

As an alternative to the segment condensa-tion approach, methods have been developedby which unprotected peptide fragments maybe linked. “Native chemical ligation” resultsin an amide bond being generated betweenpeptide fragments (Muir et al., 1997). A pep-tide bearing a C-terminal thioacid is convertedto a 5-thio-2-nitrobenzoic acid ester and thenreacted with a peptide bearing an N-terminalCys residue (Dawson et al., 1994). The initialthioester ligation product undergoes spontane-ous rearrangement, leading to an amide bondand regeneration of the free sulfhydryl on Cys.The method was later refined so that a rela-tively unreactive thioester can be used in theligation reaction (Dawson et al., 1997; Ayerset al., 1999). Safety-catch linkers are used inconjunction with Fmoc chemistry to producethe necessary peptide thioester (Shin et al.,1999).

SIDE-REACTIONSThe free Nα-amino group of an anchored

dipeptide is poised for a base-catalyzed intra-molecular attack of the C-terminal carbonyl.Base deprotection of the Fmoc group can thusrelease a cyclic diketopiperazine while a hy-droxymethyl-handle leaving group remains onthe resin. With residues that can form cis pep-tide bonds, e.g., Gly, Pro, N-methylamino ac-ids, or D-amino acids, in either the first orsecond position of the (C → N) synthesis,diketopiperazine formation can be substantial.The steric hindrance of the 2-chlorotrityl linkermay minimize diketopiperazine formation ofsusceptible sequences during Fmoc chemistry.

The conversion of side-chain protected Aspresidues to aspartimide residues can occur byrepetitive base treatments. The cyclic aspar-timide can then react with piperidine to formthe α- or β-piperidide or α- or β-peptide. Aspar-timide formation can be rapid, and is dependentupon the Asp side-chain protecting group. Se-quence dependence studies of Asp(OtBu)-Xpeptides revealed that piperidine could induceaspartimide formation when X = Arg(2,2,5,7,8-pentamethylchroman-6-sulfonyl; Pmc),Asn(triphenylmethyl; Trt), Asp(OtBu),Cys(Acm), Gly, Ser, Thr, and Thr(tBu) (Laueret al., 1995). Aspartimide formation can alsobe conformation-dependent. This side-reactioncan be minimized by including 0.1 M HOBt inthe piperidine solution (Lauer et al., 1995), orby using an amide backbone protecting group(i.e., 2-hydroxy-4-methoxybenzyl) for the resi-

due in the X position of an Asp-X sequence(Quibell et al., 1994).

Cys residues are racemized by repeatedpiperidine deprotection treatments duringFmoc SPPS. Racemization of esterified (C-ter-minal) Cys can be reduced by using 1% 1,8-diazabicyclo[5.4.0]undec-7-ene in N,N-di-methylformamide (DMF). Additionally, thesteric hindrance of the 2-chlorotrityl linkerminimizes racemization of C-terminal Cysresidues. When applying protocols for Cys in-ternal (not C-terminal) incorporation which in-clude phosphonium and aminium salts as cou-pling agents, as well as preactivation in thepresence of suitable additives and tertiaryamine bases, significant racemization is ob-served. Racemization is generally reduced byavoiding preactivation, using a weaker base(such as collidine), and switching to the solventmixture DMF-dichloromethane (DCM) (1:1).Alternatively, the pentafluorophenyl ester of asuitable Fmoc-Cys derivative can be used.

The combination of side-chain protectinggroups and anchoring linkages commonly usedin Fmoc chemistry are simultaneously depro-tected and cleaved by TFA. Cleavage of thesegroups and linkers results in liberation of reac-tive species that can modify susceptible resi-dues, such as Trp, Tyr, and Met. Modificationscan be minimized during TFA cleavage byutilizing effective scavengers. Three efficientcleavage “cocktails” quenching reactive spe-cies and preserving amino acid integrity, areTFA-phenol-thioanisole-1,2-ethanedithiol-H2O(82.5:5:5:2.5:5) (reagent K) (King et al., 1990),TFA-thioanisole-1,2-ethanedithiol-anisole(90:5:3:2) (reagent R) (Albericio et al., 1990),and TFA-phenol-H2O-triisopropylsilane(88:5:5:2) (reagent B) (Solé and Barany, 1992).The use of Boc side-chain protection of Trp alsosignificantly reduces alkylation by Pmc or2,2,4,6,7-pentamethyldihydro-benzofuran-5-sulfonyl (Pbf) groups.

PURIFICATION AND ANALYSISOF SYNTHETIC PEPTIDES

Each synthetic procedure has limitations,and even in the hands of highly experiencedworkers, certain sequences defy facile prepara-tion. The maturation of high-performance liq-uid chromatography (HPLC) has been a majorboon to modern peptide synthesis, because theresolving power of this technique facilitatesremoval of many of the systematic low-levelby-products that accrue during chain assemblyand upon cleavage. Peptide purification is mostcommonly achieved by reversed-phase HPLC


11.15.7

Immunology

(RP-HPLC; UNIT 10.14). Either alternatively to orin tandem with RP-HPLC, ion-exchangeHPLC (UNIT 10.10) and gel-filtration HPLC (UNIT

10.9) can be used for isolation of desired peptideproducts. The progress of peptide purificationcan be monitored rapidly by matrix-assistedlaser desorption/ionization time-of-flight massspectrometry (MALDI-TOF-MS) or ion-trapelectrospray MS (UNITS 10.21 & 10.22).

The homogeneity of synthetic materialsshould be checked by at least two chroma-tographic or electrophoretic techniques—e.g.,RP-HPLC (UNIT 10.14), ion-exchange HPLC(UNIT 10.10), and capillary zone electrophoresis(UNIT 10.20). Also, determination of a molecularion by MS (UNIT 10.21) using a mild ionizationmethod is important for proof of structure.Synthetic peptides must be checked routinelyfor the proper amino acid composition, and insome cases sequencing data are helpful. ThePSRC studies (see discussion of Developmentof Solid-Phase Peptide Synthesis Methodol-ogy) have allowed for a side-by-side compari-son of a variety of analytical techniques. Effi-cient characterization of synthetic peptides bestbeen obtained by a combination of RP-HPLCand MS, with sequencing by either Edmandegradation sequence analysis or tandem MS(UNIT 10.21) being used to identify the positionsof modifications and deletions. Proper peptidecharacterization by multiple techniques is es-sential.

LITERATURE CITEDAlbericio, F., Kneib-Cordonier, N., Biancalana, S.,

Gera, L., Masada, R.I., Hudson, D., and Barany,G. 1990. Preparation and application of the 5-(4-(9-fluorenylmethyloxycarbonyl)aminomethyl-3,5-dimethoxyphenoxy)valeric acid (PAL) han-dle for the solid-phase synthesis of C-terminalpeptide amides under mild conditions. J. Org.Chem. 55:3730-3743.

Albericio, F., Lloyd-Williams, P., and Giralt, E.1997. Convergent solid-phase peptide synthesis.Methods Enzymol. 289:313-336.

Angeletti, R.H., Bonewald, L.F., and Fields, G.B.1997. Six year study of peptide synthesis. Meth-ods Enzymol. 289:697-717.

Atherton, E. and Sheppard, R.C. 1987. The fluo-renylmethoxycarbonyl amino protecting group.In The Peptides, Vol. 9 (S. Udenfriend and J.Meienhofer, eds.) pp. 1-38. Academic Press,New York.

Ayers, B., Blaschke, U.K., Camarero, J.A., Cotton,G.J., Holford, M., and Muir, T.W. 1999. Intro-duction of unnatural amino acids into proteinsusing expressed protein ligation. Biopolymers(Peptide Sci.) 51:343-354.

Barany, G. and Merrifield, R.B. 1979. Solid-phasepeptide synthesis. In The Peptides, Vol. 2 (E.Gross and J. Meienhofer, eds.) pp. 1-284. Aca-demic Press, New York.

Bergmann, M. and Zervas, L. 1932. Über ein allge-meines Verfahren der Peptidsynthese. Ber.Dtsch. Chem. Ges. 65:1192-1201.

Carpino, L.A. and Han, G.Y. 1972. The 9-fluorenyl-methoxycarbonyl amino-protecting group. J.Org. Chem. 37:3404-3409.

Dawson, P.E., Muir, T.W., Clark-Lewis, I., and Kent,S.B.H. 1994. Synthesis of proteins by nativechemical ligation. Science 266:776-779.

Dawson, P.E., Churchill, M.J., Ghadiri, M.R., andKent, S.B.H. 1997. Modulation of reactivity innative chemical ligation through the use of thioladditives. J. Am. Chem. Soc. 119:4325-4329.

duVigneaud, V., Ressler, C., Swan, J.M., Roberts,C.W., Katsoyannis, P.G., and Gordon, S. 1953.The synthesis of an octapeptide amide with thehormonal activity of oxytocin. J. Am. Chem. Soc.75:4879-4880.

Fields, C.G. and Fields, G.B. 1994. Solvents forsolid-phase peptide synthesis. In Methods inMolecular Biology, Vol. 35: Peptide SynthesisProtocols (M.W. Pennington and B.M. Dunn,eds.) pp. 29-40. Humana Press, Totowa, N.J.

Fields, G.B. and Noble, R.L. 1990. Solid phasepeptide synthesis utilizing 9-fluorenylmethoxy-carbonyl amino acids. Int. J. Peptide Protein Res.35:161-214.

Fields, C.G., Fields, G.B., Noble, R.L., and Cross,T.A. 1989. Solid phase peptide synthesis of 15N-gramicidins A, B, and C and high performanceliquid chromatographic purification. Int. J. Pep-tide Protein Res. 33:298-303.

Fields, G.B., Otteson, K.M., Fields, C.G., and No-ble, R.L. 1990. The versatility of solid phasepeptide synthesis. In Innovation and Perspec-tives in Solid Phase Synthesis: Peptides,Polypeptides and Oligonucleotides, Macro-or-ganic Reagents and Catalysts (R. Epton, ed.) pp.241-260. Solid Phase Conference Coordination,Ltd., Birmingham, U.K.

Fields, G.B., Lauer-Fields, J.L., Liu, R.-q., andBarany, G. 2001. Principles and practice of solid-phase peptide synthesis. In Synthetic Peptides: AUser’s Guide, 2nd ed. (G.A. Grant, ed.) in press.W.H. Freeman, New York.

Fischer, E. and Fourneau, E. 1901. Über einigeDerivate des Glykocoils. Ber. Dtsch. Chem. Ges.34:2868-2877.

Kates, S.A., Solé, N.A., Albericio, F., and Barany,G. 1994. Solid-phase synthesis of cyclic pep-tides. In Peptides: Design, Synthesis and Bio-logical Activity (C. Basava and G.M. Anan-tharamaiah, eds.) pp. 39-57. Birkhaeuser, Bos-ton.

King, D.S., Fields, C.G., and Fields, G.B. 1990. Acleavage method which minimizes side reactionsfollowing Fmoc solid phase peptide synthesis.Int. J. Peptide Protein Res. 36:255-266.


11.15.8


Lauer, J.L., Fields, C.G., and Fields, G.B. 1995.Sequence dependence of aspartimide formationduring 9-fluorenylmethoxycarbonyl solid-phasepeptide synthesis. Lett. Peptide Sci. 1:197-205.

Merrifield, R.B. 1963. Solid phase peptide synthesisI: The synthesis of a tetrapeptide. J. Am. Chem.Soc. 85:2149-2154.

Merrifield, R.B. 1967. New approaches to thechemical synthesis of peptides. Recent Prog.Hormone Res. 23:451-482.

Merrifield, R.B. 1986. Solid phase synthesis. Sci-ence 232:341-347.

Merrifield, R.B., Stewart, J.M., and Jernberg, N.1966. Instrument for automated synthesis ofpeptides. Anal. Chem. 38:1905-1914.

Muir, T.W., Dawson, P.E., and Kent, S.B.H. 1997.Protein synthesis by chemical ligation of unpro-tected peptides in aqueous solution. MethodsEnzymol. 289:266-298.

Quibell, M., Owen, D., Packman, L.C., andJohnson, T. 1994. Suppression of piperidine-me-diated side product formation for Asp(OBut)-containing peptides by the use of N-(2-hydroxy-4-methoxybenzyl) (Hmb) backbone amide pro-tection. J. Chem. Soc. Chem. Commun.2343-2344.

Rich, D.H. and Singh, J. 1979. The carbodiimidemethod. In The Peptides, Vol. 1 (E. Gross and J.Meienhofer, eds.) pp. 241-314. Academic Press,New York.

Sakakibara, S., Shimonishi, Y., Kishida, Y., Okada,M., and Sugihara, H. 1967. Use of anhydrous HFin peptide synthesis I: Behavior of various pro-tective groups in anhydrous HF. Bull. Chem. Soc.Jpn. 40:2164-2167.

Sarin, V.K., Kent, S.B.H., and Merrifield, R.B. 1980.Properties of swollen polymer networks: Solva-

tion and swelling of peptide-containing resins insolid-phase peptide synthesis. J. Am. Chem. Soc.102:5463-5470.

Shin, Y., Winans, K.A., Backes, B.J., Kent, S.B.H.,Ellman, J.A., and Bertozzi, C.R. 1999. Fmoc-based synthesis of peptide-αthioesters: Applica-tion to the total chemical synthesis of a glycopro-tein by native chemical ligation. J. Am. Chem.Soc. 121:11684-11689.

Solé, N.A. and Barany, G. 1992. Optimization ofsolid-phase synthesis of [Ala8]-dynorphin A. J.Org. Chem. 57:5399-5403.

Tam, J.P. and Lu, Y.-A. 1995. Coupling difficultyassociated with interchain clustering and phasetransition in solid phase peptide synthesis. J. Am.Chem. Soc. 117:12058-12063.

KEY REFERENCESAtherton, E. and Sheppard, R.C. 1989. Solid Phase

Peptide Synthesis: A Practical Approach. IRLPress, Oxford.

An extensive collection of Fmoc-based syntheticmethods and techniques.

Barany and Merrifield, 1979. See above.

The definitive, comprehensive overview of the solid-phase method.

Fields, G.B. 1997. Solid-phase peptide synthesis.Methods Enzymol. Vol. 289.

A contemporary collection of SPPS techniques andapplications.

Contributed by Gregg B. FieldsFlorida Atlantic UniversityBoca Raton, Florida


11.15.9

Immunology

UNIT 11.16Synthetic Peptides for Production ofAntibodies that Recognize Intact Proteins

Antibodies that recognize intact proteins can be produced through the use of syntheticpeptides based on short stretches of the protein sequence, without first having to isolatethe protein. The procedure for selecting stretches of protein sequence likely to be antigenicis relatively straightforward. However, no procedure will identify a single sequenceguaranteed to be effective, nor will it usually identify the best single sequence to use.Rather, several sequences will be identified that have a higher-than-average probabilityof producing an effective antigen.

The steps to produce an effective antibody include: (1) designing the peptide sequencebased on the sequence of the protein; (2) synthesizing the peptide; (3) preparing theimmunogen either by coupling the synthetic peptide to a carrier protein or through theuse of a multiple antigenic peptide (MAP); (4) immunizing the host animal; (5) assayingantibody titer in the host animal’s serum; and (6) obtaining the antiserum and/or isolatingthe antibody. This unit covers steps 1 and 3; step 2 requires a laboratory with expertise inpeptide synthesis. Peptide synthesis services are widely available both academically andcommercially.

The best method to select potentially effective sequences is via a computer-assistedstrategy (see Basic Protocol 1). An alternative manual method is also described (seeAlternate Protocol 1) but is not recommended to replace the use of algorithms if there isa choice. A small synthetic peptide is usually insufficiently immunogenic on its own, andtwo methods have been developed to solve this problem. The first (see Basic Protocol 2)involves chemically coupling the synthetic peptide to a carrier protein to boost the immuneresponse. The second method (see Alternate Protocol 2) entails direct synthesis of a MAPcovalent multimer of the simple peptide sequence. Both methods have proven effectiveand it is a matter of personal preference which to use. Coupling to a carrier protein requiresadditional chemical manipulations after synthesis of the peptide, while the MAP iscomplete and ready for immunization at the conclusion of the synthetic protocol.Disadvantages of MAPs are that they are more difficult to produce homogeneously andto analyze postsynthetically. They also may be more prone to insolubility problems.

A carrier protein is a relatively large molecule capable of stimulating an immune responseindependently. A synthetic peptide coupled to a carrier protein acts as a hapten andproduces antibodies specific for the hapten (antibodies against the carrier protein are alsoproduced). The most commonly used carrier proteins are keyhole limpet hemocyanin(KLH) and bovine or rabbit serum albumin (BSA or RSA). KLH is usually preferred,because it tends to elicit a stronger immune response and is evolutionarily more remotefrom mammalian proteins. A common problem with KLH, however, has been its solubil-ity. Pierce Chemical Company sells a preparation of KLH purported to have bettersolubility properties (see below).

Alternatively, peptides can be coupled to carrier proteins through either their amino (seeAlternate Protocol 3) or carboxyl groups (see Alternate Protocol 4). These two alternateprotocols are not recommended as a first choice for coupling, but are included becausethey have been used successfully and may be advantageous for certain special applicationsdiscussed in the Commentary. Also presented are methods for assaying free sulfhydrylcontent and for reducing disulfide bonds in synthetic peptides (see Support Protocols 1and 2).

Supplement 59

Contributed by Gregory A. GrantCurrent Protocols in Molecular Biology (2002) 11.16.1-11.16.19Copyright © 2002 by John Wiley & Sons, Inc.

11.16.1

Immunology

Once the coupling procedure has been performed, it is possible to determine the approxi-mate degree of coupling by amino acid analysis (see Support Protocol 3). However, inmost instances this is unnecessary and the product can be used directly.

BASICPROTOCOL 1

COMPUTER-ASSISTED SELECTION OF APPROPRIATE ANTIGENICPEPTIDE SEQUENCES

An antibody produced in response to a simple linear peptide will most likely recognize alinear epitope in a protein. Furthermore, that epitope must be solvent-exposed to beaccessible to the antibody. The general features of protein structure that correspond tothese criteria are turns or loop structures, which are generally found on the protein surfaceconnecting other elements of secondary structure, and areas of high hydrophilicity,especially those containing charged residues. As a consequence, computer algorithms thatpredict protein hydrophilicity and tendency to form turns are very useful. Several analyticprograms or algorithms that attempt to do this have been developed. Although the choiceof method may rely on availability or personal preference, there tends to be a high levelof agreement among them. As stated earlier, none of the methods will identify the onesingle sequence guaranteed to produce an effective antibody against any given protein.Rather, the methods will offer several good candidates, one or several of which can beused.

Many of these algorithms may already be available on a local computer system. They areincluded in many commercial software packages such as GCG (Genetics ComputerGroup; see APPENDIX 4). The ExPASy Web site of the University of Geneva offers freeaccess to a variety of different programs over the Internet at http://expasy.org/tools.

The following protocol utilizes the hydropathy index developed by Kyte and Doolittle(1982) and the secondary structure prediction method for β turns developed by Chou andFasman (1974) found in the tool “Protscale” at the ExPASy Internet address.

1. Using the selected algorithms, compute the hydropathy index and the tendency forβ-turns of the protein sequence. Use a window size of 7 or 9 and give equal weightto each amino acid. Record the results in either graphical or numerical form, or both.

As an example, the graphical representation of these results for the protein sequence shownin Figure 11.16.1 is presented in Figure 11.16.2.

A window size determines the number of amino acids to be used in computing a value forthe amino acid at the center of the window. For example, a window size of 9 includes 4

10 20 30 40 50 60

MAKVSLEKDK IKFLLVEGVH QKALESLRAA GYTNIEFHKG ALDDEQLKES IRDAHFIGLR

SRTHLTEDVI NAAEKLVAIG CFCIGTNQVD LDAAAKRGIP VFNAPFSNTR SVAELVIGEL 120

LLLLRGVPEA NAKAHRGVWN KLAAGSFEAR GKKLGIIGYG HIGTQLGILA ESLGMYVYFY 180

DIENKLPLGN ATQVQHLSDL LNMSDVVSLH VPENPSTKNM MGAKEISLMK PGSLLINASR 240

GTVVDIPALC DALASKHLAG AAIDVFPTEP ATNSDPFTSP LCEFDNVLLT PHIGGSTQEA 300

QENIGLEVAG KLIKYSDNGS TLSAVNFPEV SLPLHGGRRL MHIHENRPGV LTALNKIFAE 360

QGVNIAAQYL QTSAQMGYVV IDIEADEDVA EKALQAMKAI PGTIRARLLY

Figure 11.16.1 The amino acid sequence of a 410-residue protein analyzed by the methodpresented in Basic Protocol 1. The results are shown in Figure 11.16.2.


11.16.2

SelectingSynthetic

Peptides forProduction of

Antibodies

amino acids on each side of the central amino acid. The value computed for the centralamino acid is the simple average of the values for each amino acid in the window.

2. Compare the results of the two analyses and look for areas of sequence that are highin turn tendency and high in hydrophilicity (low in hydrophobicity).

In Figure 11.16.2, these areas correspond to positive peaks in the Chou-Fasman analysisand negative peaks in the Kyte-Doolittle analysis. The three best areas in terms of amplitudeand correlation are shaded. These correspond to the sequences underlined in Figure 11.16.1.(Note the alignment of these peak optima as compared to the peaks around residue 300.)

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

Sco

re

A

–2

3

Sco

re

B

50 100 150 200 250 300 350 400

Position

–1

0

1

2

Figure 11.16.2 Graphical representation of the results generated by a computer algorithm for thesequence in Figure 11.16.1, analyzed by the method presented in Basic Protocol 1. The shadedareas represent three regions in the sequence meeting criteria for selection as potential immuno-gens. (A) Analysis for β turns (Chou and Fasman, 1974). (B) Analysis for hydrophobicity (Kyte andDoolittle, 1982).


11.16.3

Immunology

3. Examine the sequences for glycosylation site motifs and discard any sequences thatcontain them unless it is known that the protein is not glycosylated.

Amino acids in glycosylated regions may be shielded from presentation to an antibody bymasking carbohydrates.

Amino-linked carbohydrate chains can occur at Asn-X-Ser or Asn-X-Thr sequences.Hydroxyl-linked carbohydrate chains do not appear to have a set motif. A program to assistin the prediction of mucin-type GalNAc O-glycosylation sites in mammalian lipoproteinsis found in the tool “NetOGlc” at the Expasy site (http://expasy.org/tools). However, beforeusing read the documentation carefully and keep in mind that such prediction methodscannot always be successful.

4. Select the best sequences resulting from this analysis to use as antigenic peptides.These are sequences where the largest positive values (peaks with positive deflection)for turn propensity correspond in position to the largest negative values (peaks withnegative deflection) for hydrophobicity. The values obtained in these analyses arerelative and dependent on the individual protein’s composition, so it is not possibleto set an arbitrary minimum value as a cutoff for rejecting a particular peak. Rather,always select the peaks of greatest magnitude in any given sequence. In addition, theimmediate amino-terminal and carboxyl-terminal regions of proteins are often ex-posed to solvent. If these areas appear to be hydrophilic in nature, they are alsoacceptable candidates. Thus each analysis may provide several potential sequences.How many peptides to make (see Anticipated Results) is a matter of individual choice.

ALTERNATEPROTOCOL 1

MANUAL INSPECTION TO SELECT APPROPRIATE PEPTIDE SEQUENCESIf computer algorithms are not available, it is possible to select potential sequences bymanual inspection. Although there is no evidence that a manual method is any lesseffective than the use of computer algorithms, there is a greater probability of overlookingpotentially important areas of sequence. It is therefore recommended that computeranalysis be used whenever it is available. Although it can be done, it would be very timeconsuming and labor intensive to manually calculate values for every overlapping peptideoffset by a single amino acid in the same way that the algorithms do. For this reason, areasrich in polar residues are selected for manual calculation of hydrophilicity and turnpropensity.

1. Visually inspect the protein sequence and select areas that contain at least two to threecharged residues (Lys, Arg, His, Asp, Glu) within a 10- to 15-residue span.

If this criterion cannot be met, select sequences with the greatest number of chargedresidues.

2. From the sequences identified in step 1, select a subset of sequences that are thehighest in Ser, Thr, Asn, Gln, Pro, and Tyr content.

3. Calculate average hydrophilicity and turn propensity for each amino acid in theselected sequences using the values given in Table 11.16.1 and a window of 9 residues(see Basic Protocol 1, step 1).

Be sure to include the residues flanking the selected sequence for calculation of values forthe residues at the ends of the selected sequence. In other words, do not use different sizewindows.

4. Plot the values for each amino acid of a chosen sequence.

Sequences whose optimal values for hydrophilicity and turn propensity correspond (as inFig. 11.16.2) are considered good candidates.

5. Inspect sequences for glycosylation motifs and discard these candidates (see BasicProtocol 1, step 3).


11.16.4

SelectingSynthetic


Antibodies

Amino acids of glycosylated regions may be masked in native proteins, so an antibodyraised against them would be ineffective.

6. Select the best sequences (see Basic Protocol 1, step 4 for criteria), choosing a highturn-propensity-to-hydrophobicity ratio.

BASICPROTOCOL 2

DESIGNING A SYNTHETIC PEPTIDE FOR COUPLING TOA CARRIER PROTEIN

Although there is no direct evidence to show that the state of the termini of the peptideaffects its ability to produce antibodies that will react with the protein, most proceduressuggest that the termini of the peptide should mimic their native state. Thus, sequenceswhose terminal residues normally are in peptide linkage in the protein can have theiramino-terminal and carboxyl-terminal groups modified by acetylation and amidation,respectively, during synthesis.

Modification of the amino or carboxyl termini will decrease the polarity of the peptide insolution and could have a significant effect on the peptide’s solubility. If the peptide lackssufficient protonatable side chains, modification of the termini can be omitted. A generalrule to predict solubility is that the total number of charges at a given pH should be atleast 20% of the number of residues in the peptide.

1. Choose a sequence of 10 to 15 amino acid residues for the synthetic peptide.

Longer peptides are more difficult and expensive to make, and they are usually unnecessary.

Table 11.16.1 Hydrophobic and β-Turn Indices of AminoAcids

Amino acid Symbols Hydrophobicityvaluea

β-turnpropensityb

Arginine Arg (R) −4.5 0.95

Lysine Lys (K) −3.9 1.01

Aspartic acid Asp (D) −3.5 1.46

Glutamic acid Glu (E) −3.5 0.74

Asparagine Asn (N) −3.5 1.56

Glutamine Gln (Q) −3.5 0.98

Histidine His (H) −3.2 0.95

Proline Pro (P) −1.6 1.52

Tyrosine Tyr (Y) −1.3 1.14

Tryptophan Trp (W) −0.9 0.96

Serine Ser (S) −0.8 1.43

Threonine Thr (T) −0.7 0.96

Glycine Gly (G) −0.4 1.56

Alanine Ala (A) 1.8 0.66

Methionine Met (M) 1.9 0.60

Cysteine Cys (C) 2.5 1.19

Phenylalanine Phe (F) 2.8 0.60

Leucine Leu (L) 3.8 0.59

Valine Val (V) 4.2 0.50

Isoleucine Ile (I) 4.5 0.47

aKyte and Doolittle (1982).bChou and Fasman (1974).


11.16.5

Immunology

Try to choose a stretch of sequence that contains some charged residues such as Arg, Lys,His, Glu, and Asp. In addition to the high likelihood of these amino acids being located onthe surface of the protein, they aid handling of the synthetic product by promoting solubility.

2a. If the selected sequence does not contain an internal cysteine: Place a cysteine oneither the amino or carboxyl terminus for use in coupling to a carrier protein with aheterobifunctional cross-linking reagent such as MBS, m-maleimidobenzoyl-N-hy-droxysuccinimide ester (see Basic Protocol 3).

Cross-linking with heterobifunctional reagents is the recommended procedure for mostpeptides (see Basic Protocol 3). As an alternative to using a chemical cross-linking reagent,any peptide, regardless of amino acid content, can also be photochemically linked to acarrier protein if p-benzoyl benzoic acid is added to the peptide during synthesis (seeAlternate Protocol 5). Although photochemical cross-linking is effective, it is not widelyused.

If the sequence includes the immediate amino or carboxyl terminal sequence of the protein,the cysteine should be placed on the end that would normally be engaged in the internalpeptide bond. For sequences internal to the protein, the cysteine may be placed at eitherend according to the preference of the synthetic chemist. However, if amino-terminalcapping (acetylation) is used after the coupling of each amino acid during synthesis, it ispreferable to place the cysteine on the amino-terminal end of the peptide since then onlythe full-length peptide will contain the cysteine residue. In this way, if synthetic difficultiesare encountered, only the full-length peptide will couple to the carrier. If placed on thecarboxyl terminal end, the cysteine residue tends to racemize during synthesis unless achlorotrityl resin is used. However, this should not have an effect on the rest of the peptideor the generation of antibodies.

2b. If the sequence contains an internal cysteine residue: Do not add a terminal cysteinefor MBS cross-linking. Rather, use an alternative coupling procedure (see AlternateProtocols 3, 4, or 5) or synthesize a multiple antigenic peptide (MAP; see AlternateProtocol 2).

Internal cysteine sulfhydryl groups will also cross-link to the carrier protein, and multiplecysteines will result in a peptide attached at multiple points. If the sulfhydryl will eventuallybe important for antibody recognition of the protein, the immunization may not produceeffective antibodies. Furthermore, the additional constraint produced by the existence ofmultiple points of coupling may affect the ultimate ability of the antibody to recognize theprotein.

3a. For sequences whose terminal residues are in peptide linkage within the protein: Ifthe peptide is coupled using a heterobifunctional reagent such as MBS (see BasicProtocol 3), modify the amino and carboxyl ends by acetylation and amidation,respectively, during the synthetic procedure. If coupling is performed with a homo-bifunctional reagent that reacts with amino groups such as glutaraldehyde (seeAlternate Protocol 3) or by the photochemical method (see Alternate Protocol 5),only amidate the carboxyl terminus. If coupling is performed with EDC (see AlternateProtocol 4), only acetylate the amino terminus.

3b. For sequences that are amino or carboxyl terminal to the protein:

i. If the sequence is the immediate amino or carboxyl terminal sequence of the proteinand the peptide will be coupled with a heterobifunctional reagent such as MBS(see Basic Protocol 3), leave the end that is not in peptide linkage (and the endthat does not contain the additional cysteine residue for MBS coupling) as the freeamino or carboxyl group unless it is known that they are normally blocked.

ii. If the sequence is the immediate amino terminal sequence of the protein and thepeptide will be coupled with a homobifunctional reagent that reacts with amino


11.16.6

SelectingSynthetic


Antibodies

groups such as glutaraldehyde (see Alternate Protocol 3) or photochemically (seeAlternate Protocol 5), amidate the C-terminus.

iii. If the sequence is the immediate carboxyl terminal sequence of the protein andthe peptide will be coupled with glutaraldehyde (see Alternate Protocol 3) orphotochemically (see Alternate Protocol 5), leave both termini free.

4. If the peptide will be coupled with 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide(EDC; see Alternate Protocol 4), block the amino terminus by acetylation.

This should be done even if the sequence is at the immediate amino terminus of the protein,since treatment with EDC can result in the production of covalent multimers of the peptidethrough reaction with the N-terminal amino group. Some procedures recommend usingcitraconylation to temporarily protect the amino group. However, this also adds a freecarboxyl group that could result in multiple site attachment of the peptide to the carrier.

HNε

O

CHN

NH2

C

O

CO

C

O

OC

O

HN

C

C

NH

NH2

aan

Lys

Lys

aa2 aa2

aa1

aa2 aa2aan aanaa3 aa3

aan

aa3 aa3

Lys

aa1

β-Ala

aa1 aa1

O

HNε

O

C

HOC

NH

OC

HN

αH

N α

OCH

N

NH

O

NεH

OC

NαH

OC

NH

OC

NH

OC

H2N

O

CHNC

O

HN NH2

O

CHN C

O

HN

C

NH

O

Figure 11.16.3 Representation of a four-branched multiple antigenic peptide (MAP). The MAP core is usuallysynthesized with a β-alanine residue attached to the solid phase support followed by a scaffold of lysine residues.The first lysine residue, attached through its α-carboxyl group to β-alanine, provides two primary amino groups(α and ε) for chain elongation. The next level of the lysine scaffold consists of two lysine residues which in turnprovide four primary amino groups for chain elongation. MAPs with eight branches are formed by providing oneadditional layer of lysine to that depicted here.


11.16.7

Immunology

ALTERNATEPROTOCOL 2

DESIGNING A SYNTHETIC MULTIPLE ANTIGENIC PEPTIDE

A multiple antigenic peptide (MAP; Posnett et al., 1988; Tam, 1988) is an effectivealternative to coupling a simple linear peptide to a carrier protein. MAPs are covalentconstructs consisting of a simple peptide sequence synthesized on a branched core (Fig.11.16.3) with one copy of the peptide sequence on each of four or eight branches. Oneadvantage of a MAP is that it is suitable for use as an immunogen at the conclusion of thesynthetic process. On occasion, MAPs have produced effective protein antibodies whenthe conventional peptide coupled to carrier protein has not. Thus, MAPs represent aneffective alternative approach for antiserum production.

1. Select a sequence between 10 and 15 residues in length.

Longer sequences are unnecessary and increase the probability of synthetic problems. Thepresence of internal cysteine residues are not a concern with MAPs, but if present, takeprecautions to keep them reduced.

2. Synthesize the MAPs utilizing a four-branch core.

Synthesize the MAP core de novo, or purchase resins for solid-phase peptide synthesis withfour- or eight-branched cores (available commercially from Advanced ChemTech, Nova-biochem, Applied Biosystems, AnaSpec, and Bachem Bioscience). Eight-branched coresare suitable if the peptide is no more than 12 to 14 residues and has a high degree ofhydrophilicity. There are more synthetic problems with eight-branched MAPs, presumablydue to the higher density of structure during synthesis: they are more difficult to charac-terize and probably raise a diverse antibody population against some synthetic artifacts(see Mints et al., 1997).

3. Optional. If the selected sequence was not the amino terminus of the protein, acetylatethe new amino terminus.

In the case of a MAP, the carboxyl terminus will remain in covalent linkage to the branchedcore.

4. Use the MAP directly as an immunogen.

Coupling to a carrier protein as described in Basic Protocol 3 is usually not necessary.

BASICPROTOCOL 3

COUPLING SYNTHETIC PEPTIDES TO A CARRIER PROTEIN USINGA HETEROBIFUNCTIONAL REAGENT

If the synthetic peptide was designed with a cysteine residue at one terminus (see BasicProtocol 2, step 2a), the following procedure should be followed for coupling to keyholelimpet hemocyanin (KLH) or other carrier proteins. Care must be taken to assure that thecysteine sulfhydryl group has remained reduced. Under normal synthetic conditions, ifthe peptide was lyophilized and stored dry immediately after synthesis, the sulfhydrylusually remains in the reduced state. The presence of free sulfhydryl groups in the peptidecan be determined with Ellman’s reagent (see Support Protocol 1) just prior to use;alternatively, high-resolution mass spectrometry can be used. If reduction is needed,follow the cysteine reduction procedure (see Support Protocol 2) before starting thecoupling process.

The reagent most commonly used for this purpose is m-maleimidobenzoyl-N-hydroxy-succinimide ester (MBS). However, several related reagents (all available from Pierce)offer some additional features. Sulfo-MBS (Pierce) is a water-soluble alternative to MBS.Succinimidyl 4-(N-maleimidomethyl)-cyclohexane-1-carboxylate (SMCC) and its sul-fonated analog sulfo-SMCC provide the same chemistry with a more pH-stable maleimide(see step 1). The MBS and SMCC reagents can be used interchangeably in this protocol(the sulfo reagents can be dissolved in aqueous solution, while the others must bedissolved in an organic solvent; the concentrations listed are appropriate for all four).


11.16.8

SelectingSynthetic


Antibodies

Materials

Keyhole limpet hemocyanin (KLH; Pierce, Sigma, Calbiochem, or BoehringerMannheim)

0.01 M sodium phosphate buffer, pH 7.5 (APPENDIX 2)10 mg/ml MBS in fresh N,N-dimethylformamide (DMF)0.05 M and 0.1 M sodium phosphate buffer, pH 7.0 (APPENDIX 2)Synthetic peptide with a reduced cysteine residue at either the N- or C-terminus6 M guanidine⋅HCl (see recipe)

Small glass vial with flat bottom∼0.9 × 15–cm gel filtration column with Sephadex G-25 or G-50 (Pharmacia

Biotech) or Bio-Gel P2 or P4 (Bio-Rad) resin; or prepacked PD-10 column(Pharmacia Biotech)

Additional reagents and equipment for gel filtration chromatography (UNIT 10.9)

NOTE: Do not use Tris or other buffers with primary amino groups in this procedure.

CAUTION: MBS is a moisture-sensitive irritant. Read the Material Safety Data Sheetbefore use.

1. Dissolve 5 mg KLH in ∼0.5 ml of 0.01 M sodium phosphate buffer, pH 7.5, in a small,flat-bottomed vial.

A pH range of 7.0 to 7.5 offsets competing reactions. Although the unprotonated form ofthe amine reacts with the N-hydroxysuccinimide ester and would be optimal at pH >8.0,hydrolysis of the ester bond and reaction of the maleimide group with amines is enhancedat higher pH.

2. Add 100 µl of 10 mg/ml MBS/DMF solution and stir gently with a micro stir-bar 30min at room temperature.

A small amount of precipitate may form during this procedure and is acceptable. However,if the precipitate is large, perform the procedure again with fresh components.

As an alternative to performing this coupling procedure from scratch, it is possible topurchase MBS-activated KLH (Pierce; Boehringer Mannheim) or kits containing MBS-ac-tivated KLH and an alternate MBS-activated protein for use in ELISA assays (Pierce).

3. Separate MBS-activated KLH from free MBS on a ∼0.9 × 15–cm gel filtrationcolumn, equilibrating and eluting the column with 0.05 M sodium phosphate buffer,pH 7.0. Collect 0.5-ml fractions and read their absorbance at 280 nm (UNIT 10.9).

The first peak to elute is the KLH-MBS conjugate. These fractions may appear cloudy. Thesecond peak is uncoupled MBS.

4. Pool the KLH-MBS conjugate fractions in a separate tube.

5. Dissolve 5 mg of the synthetic peptide in 0.01 M sodium phosphate buffer, pH 7.0,immediately prior to use. If the peptide is poorly soluble, use 6 M guanidine⋅HCl.

Maleimide groups react specifically with sulfhydryls at slightly acid to neutral pH.

6. Add the peptide solution to the KLH-MBS conjugate. Stir gently with a micro stir-bar3 hr at room temperature.

The coupling may be continued overnight.

7. Dialyze against 4 liters distilled water overnight at 4°C. Use for immunizations within24 hr.


11.16.9

Immunology

N

N

OH

OH

H2C

SS

SH

SH

HS

SSC

H2

CH

OH

CH

CH

2S

H

OH

(DT

T)

SC

H

OH

CH

OH

SC

H2

CH

2S

H

OH

R1

O

NC

NR

2C

O

O

N

CCN

H

R1

R2

KLH

NH

2

KLH

N H

O C

CO

N HR

1N H

R2

N

KLH

NH

2

O O

OCO

N

O O

O O

OHKLH

N HCO

O O

SH

KLH

N HCO

N

O O

S

O2N

SH

HO

OC

CO

OH

SS

(Ellm

an's

Rea

gent

)

NO

2O

2N HO

OC

SS

O2N HO

OC

SH

(A=

412

nm)

KLH

NH

2H

CO

(CH

2)3

CH

O

H2N

KLH

N H(C

H2)

3CO

CO

N H

A B C D E

(ED

C)

CH

2H C

C H

Fig

ure

11.

16.4

R

epre

sent

atio

ns o

f th

e ch

emis

trie

s de

scrib

ed i

n th

is u

nit.

( A)

Cro

ss-li

nkin

g w

ith M

BS

; ( B

) cr

oss-

linki

ng w

ith g

luta

rald

ehyd

e;( C

) cr

oss-

linki

ng w

ith E

DC

; (D

) re

actio

n of

free

sul

fhyd

ryls

with

Ellm

an’s

rea

gent

to p

rodu

ce a

col

orim

etric

pro

duct

; (E

) re

duct

ion

of d

isul

fides

with

DT

T. A

bbre

viat

ions

: KLH

, ke

yhol

e lim

pet

hem

ocya

nin;

MB

S,

m-m

alei

mid

oben

zoyl

-N-h

ydro

xysu

ccin

imid

e es

ter;

ED

C,

1-et

hyl-3

-(3-

dim

ethy

lam

ino-

prop

yl)

carb

odiim

ide;

DT

T, N

,N-d

ithio

thre

itol.


11.16.10

SelectingSynthetic


Antibodies

SUPPORTPROTOCOL 1

ASSAY OF FREE SULFHYDRYLS WITH ELLMAN’S REAGENT

Free sulfhydryl content of a peptide can be quantitatively determined with Ellman’sreagent, 5,5′-dithio-bis(2-nitrobenzoic acid). The molar extinction coefficient at 412 nmof thionitrobenzoate, the colored species generated when the reagent reacts with a freethiol, is 14,150 in 0.1 M sodium phosphate buffer (see Fig. 11.16.4). The sensitivity ofthe reaction is in the low nmol/ml range for sulfhydryl groups, making it well suited forsynthetic peptides. By dry weight most synthetic peptides are only ∼60% to 75% peptide,the remainder consisting of counterions and water of hydration. Amino acid analysis (seeUNIT 11.9) is needed to establish the actual peptide content unambiguously, but such precisemeasurement is not usually necessary for qualitative evaluation of the free sulfhydrylcontent of a peptide sample.

Materials

Cysteine standard stock solution (see recipe)0.1 M sodium phosphate, pH 8.0 (APPENDIX 2)Peptide to be assayedEllman’s reagent solution (see recipe)13 × 100–mm glass test tubes

1. Prepare a cysteine standard curve by adding 25 µl, 50 µl, 100 µl, 150 µl, 200 µl, and250 µl of cysteine standard stock solution to separate 13 × 100–mm tubes. Add ≤250µl of each peptide to be tested to separate tubes. Bring the volume in each tube to250 µl with 0.1 M sodium phosphate, pH 8.0. Add 250 µl of 0.1 M sodium phosphate,pH 8.0, to a blank tube.

The cysteine content of the peptide to be assayed should fall within the range of the standardcurve (37.5 to 375 nmol).

2. Add 50 µl Ellman’s reagent solution and 2.5 ml of 0.1 M sodium phosphate, pH 8.0,to each tube. Mix and incubate 15 min at room temperature.

3. Measure absorbance at 412 nm (A412).

4. Plot the A412 values of the standards after subtracting the value for the blank to producea standard curve. Use this curve to determine the free sulfhydryl content of thepeptides.

SUPPORTPROTOCOL 2

REDUCING CYSTEINE GROUPS IN PEPTIDES

When a peptide is synthesized with a terminal cysteine residue to be used for couplingwith MBS (see Basic Protocol 3), the cysteine must be in the reduced state (present asfree -SH rather than as a disulfide) in order to participate in the reaction with the couplingreagent. If peptides are lyophilized immediately after extraction from the resin cleavagecocktail or reversed-phase HPLC and used immediately after reconstitution, oxidation ofcysteine side chains is usually not a problem. However, if oxidation to disulfides hasoccurred, the peptide can be reduced prior to use with the protocol presented here.

Dithiothreitol (DTT) is preferred to 2-mercaptoethanol (2-ME) as a reducing agentbecause its lower redox potential allows it to be effective at lower concentrations, and thereaction goes to completion because formation of the six-membered ring containing aninternal disulfide is energetically favorable (see Fig. 11.16.4).

To determine if reduction is necessary, quantitate the level of free sulfhydryl groups withEllman’s reagent (see Support Protocol 1).

Additional methods for reducing disulfides include using sodium borohydride (Gailit,1993) and Tris(2-carboxyethyl)phosphine (TCEP; Getz et al., 1999).


11.16.11

Immunology

Materials

Synthetic peptide0.1 M sodium phosphate, pH 8.0 (APPENDIX 2)1 M aqueous dithiothreitol (DTT)1 N HCl

100- or 250-µl polypropylene tubesNitrogen gas source

Additional reagents and equipment for reversed-phase HPLC of peptides (see UNIT

10.14)

1. Dissolve 5 to 10 mg of peptide in 0.1 M sodium phosphate, pH 8.0.

2. Add 100 µl of 1 M DTT.

3. Flush nitrogen over the surface of the liquid, seal the tube, and incubate 1 hr at 37°C.

4. Acidify with 1 N HCl and desalt by reversed-phase HPLC (UNIT 10.14).

5. Pool peptide fractions and lyophilize. Store lyophilized at 4°C until ready to use (upto several days).

The oxidation state of the peptide can usually be followed by analytical monitoring of itselution position on reversed-phase HPLC. Disulfide-linked dimers of peptides generallyelute later than the monomeric peptide.

ALTERNATEPROTOCOL 3

COUPLING SYNTHETIC PEPTIDES TO A CARRIER PROTEIN USINGA HOMOBIFUNCTIONAL REAGENT

The available homobifunctional reagents couple compounds through primary aminogroups. Therefore, peptides with internal lysine residues should not be used in thisprocedure. The reagent most commonly used for this procedure is glutaraldehyde, but itshould not be used with peptides containing internal Cys, Tyr, or His residues. Otherhomobifunctional cross-linking reagents that can be used in the same way as glutaralde-hyde, but do not cross react with Cys, Tyr, or His residues, are also available: disuccin-imidyl suberate (DSS), disuccinimidyl glutarate (DSG), and bis(sulfosuccinimidyl)suberate (BS3; all available from Pierce). However, these reagents are not widelyemployed for coupling peptides to proteins and are not considered the method of choice,and methods for their use have not been formalized. This is probably because couplingof the synthetic peptide to itself and aggregation of the carrier protein can occur withhomobifunctional reagents such as these. Glutaraldehyde, on the other hand, althoughalso subject to this limitation, has generally been used successfully.


50 mM sodium borate buffer, pH 8.0: adjust pH with HClGlutaraldehyde solution (see recipe)1 M glycine in 50 mM sodium borate buffer, pH 8.0

NOTE: Do not use Tris or other buffers with primary amino groups in this procedure.

1. Dissolve 5 mg KLH in ∼1.0 ml of 50 mM sodium borate buffer, pH 8.0.

2. Add 5 mg synthetic peptide.

3. Slowly add 1 ml fresh glutaraldehyde solution with gentle mixing at room tempera-ture. Allow to react for an additional 2 hr with gentle mixing.

Formation of a yellowish color or milkiness is normal and does not affect the sample.


11.16.12

SelectingSynthetic


Antibodies

4. Add 0.25 ml of 1 M glycine to bind unreacted glutaraldehyde.

A darker yellow to brown color may develop.

5. Dialyze the reaction mixture overnight at 4°C against 4 liters of 50 mM sodium boratebuffer, pH 8.0, and then overnight against water. Use immediately.

ALTERNATEPROTOCOL 4

COUPLING SYNTHETIC PEPTIDES TO A CARRIER PROTEIN USINGA CARBODIIMIDE

This procedure couples amino groups to carboxyl groups by way of activation of thecarboxyl group with a water-soluble carbodiimide. Since the procedure is most easilyperformed in one step, peptides containing internal Asp, Glu, Lys, Tyr, or Cys residuesshould not be used. Also, in order to avoid making polymers of the peptide, the aminoterminus should be blocked by acetylation during synthesis (see Basic Protocol 2). Whilethis method has been used successfully, it is not considered to be the method of choiceexcept in special situations.


1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC; Pierce), used fresh orstored desiccated and frozen

0.1 N HCl

NOTE: Buffers containing amino or carboxyl groups should not be used in this procedure.According to some reports, buffers containing phosphate groups should also be avoided.Water is the safest choice as a solvent.

1. Dissolve 5 mg synthetic peptide in 1 ml water.

2. Add 25 mg EDC and carefully adjust pH to 4.0 to 5.0 by adding small amounts of0.1 N HCl. Allow to react for 5 to 10 min at room temperature with gentle mixing.

pH paper suffices to monitor this adjustment.

3. Dissolve 5 mg KLH in 0.5 ml water and add to solution from step 2. React 2 hr atroom temperature with gentle mixing.

4. Dialyze against 4 liters of water overnight at 4°C. Use immediately.

ALTERNATEPROTOCOL 5

COUPLING SYNTHETIC PEPTIDES TO A CARRIER PROTEINPHOTOCHEMICALLY

Most synthetic peptides today are made with 9-fluorenylmethyloxycarbonyl (Fmoc)chemistry (see Fig. 18.1.1), and therefore the following protocol links the photoactivegroup to the free amino terminus of the peptide. An alternative approach if t-butyloxycar-bonyl (Boc) chemistry is used is to link it to an ε-amino group of a terminal lysine (Gorkaet al., 1989).

Additional Materials (see also Basic Protocol 3)

4-benzoyl benzoic acid (Sigma or Aldrich)Quartz spectrophotometry cuvettes

1. Have the synthetic chemist who is making the peptide attach a benzoyl benzoic acidgroup to the amino terminus simply by treating the reagent as the terminal residueduring normal synthesis.

This blocks the amino terminus.


11.16.13

Immunology

2. Dissolve 5 mg KLH in ∼0.5 ml of 0.01 M sodium phosphate buffer, pH 7.5.

3. Add 2 mg of synthetic peptide containing the benzoyl benzoate adduct.

4. Place in a 1-cm quartz cuvette and irradiate with 366-nm light for 3 hr at a distanceof 0.5 cm. Use immediately.

The peptide can be used directly for immunization.

SUPPORTPROTOCOL 3

CALCULATION OF THE MOLAR RATIO OF PEPTIDE TOCARRIER PROTEIN

The molar ratio of peptide to carrier protein coupling efficiency can be calculated todetermine the level of substitution achieved by the coupling procedure. This informationcan be obtained using the results of amino acid compositional analysis. By performingthe calculations presented in this protocol, the molecules of peptide in the conjugate permolecule of carrier protein in the conjugate can be determined.

1. Obtain the amino acid composition of the carrier protein, the peptide, and thepeptide/carrier conjugate. Amino acid compositional analysis (of these hydrolysates)is usually available at sources that provide automated peptide synthesis (see UNIT

11.15). Be sure that the conjugate is free of unconjugated peptide (i.e., it should bewell dialyzed).

2. Determine a scaling factor (SF) that relates the moles of protein in the unconjugatedcarrier protein to the moles of protein in the peptide/carrier conjugate. This is doneby comparing the molar ratio of ≥3 amino acids present in the carrier protein andpeptide/carrier conjugate but not present in the peptide. For example, if the peptideTGLRDSC (Table 11.16.2) is coupled to a carrier protein, choose A, K, and I. Thecalculation is done as:


Table 11.16.2 Sample Calculation of the Extent of Coupling of the Peptide TGLRDSCto Carrier Proteina

Amino acid

Composition of carrier protein

Composition of peptide/carrier conjugate

Amount of carrier protein amino acids in conjugate

Amount of peptide amino acids in conjugate

D 80 185 160 25E 110 222 220 —G 95 215 190 25S 65 150 130 20T 70 163 140 23H 10 19 20 —P 25 51 50 —A 103 206 206 —M 5 11 10 —V 60 118 120 —F 22 45 44 —L 55 133 110 23I 65 135 130 —C 7 22 14 8Y 13 25 26 —K 65 125 130 —R 75 177 150 27

Total pmolamino acid 925 2002 1850 151

11.16.14

SelectingSynthetic


Antibodies

SF = [(pmol A in conjugate/pmol A in carrier) + (pmol K in conjugate/pmol K in carrier) + (pmol I in conjugate/pmol I in carrier)]/3.

For these amino acids, the carrier protein yields are as follows: A = 103 pmol, K =65 pmol, and I = 65 pmol. For these same amino acids, the peptide/carrier conjugateyields: A = 206 pmol, K = 125 pmol, and I = 135 pmol.

From these values, the relative amount of carrier protein in the conjugate versus theunconjugated carrier protein (SF) can be calculated as follows: (206/103 + 125/65 +135/65)/3 = 2.0, indicating that there is twice as much carrier protein in the pep-tide/carrier conjugate hydrolysate as in the carrier-protein hydrolysate.

3. Calculate the moles of peptide present in the conjugate by subtracting the moles ofamino acid present in the carrier from the moles of amino acid present in theconjugate. Choose ≥3 amino acids present in the peptide. The relative amount (SF)of protein present in the carrier protein versus the amount in the conjugate ascalculated in step 2, must also be considered as follows:

Therefore, the amount of peptide in the conjugate hydrolysate for the example shownin Table 9.4.1, calculated using the amino acids G, L, and R, is {[215 − (2 × 95)] +[133 − (2 × 55)] + [177 − (2 × 75)]}/3 = 25 pmol.

4. Calculate the number of moles of protein in the conjugate hydrolysate as follows:

where total pmol carrier protein amino acids = SF × (total amino acid compositionof carrier in pmol) and 110 is the average molecular weight of an amino acid.

In this example, there are 1850 pmol of carrier protein amino acids in the conjugate;therefore, 1850 pmol × (110/100,000) = 2.04 pmol carrier protein in conjugate.

5. Determine the ratio of peptide to carrier protein as follows:

molecules peptide in conjugate/molecules carrier protein in conjugate =pmol peptide in conjugate/pmol carrier protein in conjugate.

Using the values calculated in steps 3 and 4, the result is: 25 pmol peptide inconjugate/2.04 pmol carrier protein in conjugate = 12.2 molecules peptide in conju-gate per molecule carrier protein in conjugate.

− ×− ×

pmol peptide in conjugate = {[pmol G in conjugate (SF pmol G in carrier)]+

[pmol L in conjugate (SF pmol L in carrier)]+

− × [pmol R in conjugate (SF pmol R in carrier)]}/3

= ×total pmol carrier protein amino acidspmol carrier protein in conjugate 110

molecular weight of carrier protein


11.16.15

Immunology


Use Milli-Q-purified water or equivalent for the preparation of all buffers. For common stock solutions,see APPENDIX 2; for suppliers, see APPENDIX 4.

Cysteine standard stock solutionDissolve 26.3 mg cysteine hydrochloride monohydrate in 100 ml of 0.1 M sodiumphosphate, pH 8.0 (APPENDIX 2). Prepare immediately before use.

Ellman’s reagent solutionDissolve 4 mg Ellman’s reagent, 5,5′-dithio-bis-(2-nitrobenzoic acid) (Pierce), in 1ml of 0.1 M sodium phosphate, pH 8.0 (APPENDIX 2). Prepare immediately before use.

Glutaraldehyde solution, 0.15%Add 30 µl of 25% aqueous glutaraldehyde solution to 5 ml of 50 mM sodium boratebuffer, pH 8.0 (pH adjusted with HCl). Prepare fresh and use immediately. If theglutaraldehyde precipitates, check the pH. It should not be above 8.0; a slightly lowerpH can be used (pH 7 to 8).

CAUTION: Glutaraldehyde is a sensitizing agent that should be handled in a hood and onlyaccording to the recommendations in the Material Safety Data Sheet. When mixing solutionsor performing reactions, keep the container covered to prevent vapors from escaping intothe atmosphere.

Guanidine⋅HCl, 6 MDissolve 1 g guanidine⋅HCl in 1 ml of 0.05 M sodium phosphate, pH 7.0 (APPENDIX

2). Store up to several weeks at room temperature.

The resulting 1.8-ml solution should be ∼0.025 M phosphate/6 M guanidine⋅HCl at pH 7.0.

COMMENTARY

Background InformationSynthetic peptides are linear arrays of amino

acids that in most instances possess a randomstructure in solution. While it is not difficult toproduce antipeptide antibodies, it does not nec-essarily follow that the antibodies will recog-nize a protein containing the same stretch ofsequence found in the peptide. In order for thisto occur, the amino acids in the protein must beoriented to the antibody in a way similar to thatof the synthetic peptide. This generally requiresthree basic features of the protein: (1) that thestretch of sequence be exposed to solvent; (2)that the sequence be a continuous stretch ofamino acids; and (3) that it not possess a higher-order structure that renders it unrecognizableby the antibody population.

The large number of model protein struc-tures now available indicate that almost all ofthe ionized groups in water-soluble proteins areon the protein surface. Asp, Glu, Lys, and Argresidues, on the average, comprise 27% of theprotein surface and only ∼4% of the proteininterior. The fraction of residues that are at least95% buried range from 0.36 to 0.60 for nonpo-lar residues and 0.01 to 0.23 for polar residues.Only 1% of Arg and 3% of Lys residues fall

into the 95% buried range (Creighton, 1993).Therefore, it is reasonable to expect solvent-ex-posed areas of proteins to display relativelyhigh levels of polar and charged residues, par-ticularly Arg and Lys.

Proteins display three kinds of secondarystructure: α-helices, β-sheets, and turns orloops. Turns or loops generally connect ele-ments of α-helices and β-sheets, and can eitherfit one of several rather strict motifs with rec-ognizable hydrogen bonding patterns or be ofa more extended, random nature. These turn orloop structures appear to be most useful forantibody production because they tend to befound on the surface of proteins connectinglarger arrays of helices and sheets, and theyconsist of continuous stretches of amino acids.Although many amino acid residues in helicesand sheets are also exposed at the surface, theregular geometry of amino acids containedwithin them makes them less suitable for thispurpose. For instance, in β-sheet structures theside chain of each successive amino acid in theβ-sheet strand points in the opposite directionto the ones immediately preceding and follow-ing it. Thus, even if the amino acid side chainsare not predominantly buried in the interior of


11.16.16

SelectingSynthetic


Antibodies

the protein, only every other side chain is ex-posed on the same surface of the sheet. This canhinder recognition by an antibody producedwith a linear peptide capable of assuming amore random structure. A similar situation ex-ists for α-helices. Although the change in di-rection of the side chains of successive aminoacids is perhaps not as abrupt as in β-sheets,only approximately every third or fourth sidechain is found on the same surface of the helix.Epitopes in proteins have been identified inamphipathic helices, but unless the syntheticpeptide assumes a similar helical structure insolution, recognition by the antibody may beproblematic.

These considerations have led to more use-ful methods for predicting sequences that willproduce antibodies recognizing intact proteins.A variety of different indices that predict hy-drophilicity or hydrophobicity and secondarystructure are available. In addition, predictivemethods based on segmental mobility, sidechain accessibility, and sequence variability(see Van Regenmortel et al., 1988) have alsobeen proposed. All of these methods generallytend to yield similar results, but it must be notedthat these procedures were developed for (andwork best with) water-soluble proteins com-posed of a single globular structure. Additionalcomplications can arise with multisubunit pro-teins, where normally exposed structures maybe shielded by subunit interactions, or mem-brane proteins with large sections shieldedfrom the solvent.

The method presented in this unit utilizesthe correlation between the hydrophilic char-acter of a peptide sequence (Kyte and Doolittle,1982) and its propensity to form β-turn struc-tures (Chou and Fasman, 1974). Free access tothese and many other algorithms is provided atthe ExPASy Web site of the University of Genevaat http://expasy.org.tools.

After selection of the peptide sequence, aneffective immunogen is generally produced bycoupling the peptide to a carrier protein or bysynthesizing a multiple antigenic peptide(MAP), with four or eight identical peptidesassembled simultaneously on the α and εamines of the terminal lysines of a branchedcore (see Fig. 11.16.3).

Critical ParametersAnalyzing protein sequences with algo-

rithms or tables of assigned values for aminoacids is a well-established procedure, but evalu-ating these results and selecting the candidate

sequences requires some consideration. To takefull advantage of the results, choose areas ofsequence that give the maximum values for theproperties being evaluated and that also showthe highest degree of residue-by-residue corre-lation. In other words, choose areas of maxi-mum amplitude where the centers of the peakscorrespond to the same sequence with a diver-gence of no more than two to three residues.Examples of this are given in Figure 11.16.2,which shows results from the method presentedin Basic Protocol 1 for the sequence shown inFigure 11.16.1. The top panel in Figure 11.16.2predicts β-turns as calculated by the method ofChou and Fasman (1974). The bottom panel isa prediction of hydrophobicity using the pa-rameters of Kyte and Doolittle (1982). The dataare analyzed by looking for areas of high turnpropensity (maximum positive deflection in thetop panel) and high hydrophilicity (maximumnegative deflection in the bottom panel). Theshaded areas in Figure 11.16.2 designate threesegments that meet these criteria. Note that themaximum and minimum values of these threestretches of protein sequence correlate verywell. Additional areas of high hydrophilicity(bottom panel) are found near residues 64, 132,137, 149, and 345, although the β-turn valuesof these secondary candidates are not as highas those of the three shaded areas. Two equallyhydrophilic areas at residues 49 and 299 corre-spond to downward deflections in the β-turnprofile and are thus not good candidates basedon this analysis.

Many different chemistries are available forcoupling synthetic peptides to carrier proteinsto produce effective immunogens (Van Regen-mortel et al., 1988). In many cases, however,side reactions or incompatibilities in chemistrybetween the coupling agent and the residuespresent in the peptide can be problematic. Inorder to simplify the process and present thegreatest probability of success in most cases,only a few coupling methods are presented inthis unit. In this regard, the recommended cou-pling procedure is cross-linking of the peptidevia cysteine residues to keyhole limpet hemo-cyanin (KLH) with the heterobifunctional re-agent m-maleimidobenzoyl-N-hydroxysuccin-imide ester (MBS; see Basic Protocol 3). Thiseffective method has enjoyed great success andcan be used for virtually any peptide. The onecaveat is that it is not recommended for peptideswith internal cysteine residues, since they willalso link to the carrier. It is also critical, whencoupling with MBS through an added terminal


11.16.17

Immunology

cysteine residue, that the sulfhydryl group ofthe peptide be present in the free or reducedform (see Support Protocols 1 and 2).

In addition to MBS coupling, other proce-dures commonly used (see Alternate Protocols3 and 4) are included as alternatives for use inspecial situations, but these are not recom-mended as a general alternative to MBS be-cause they are more restrictive and have thepotential for undesirable side reactions. Glu-taraldehyde coupling (see Alternate Protocol 3)should not be used with peptides containinginternal Lys, Cys, Tyr, or His residues and, sinceit is a homobifunctional reagent, cross-linkingof the peptide to itself and the carrier to itselfcan occur. The latter lowers antigenicity andcan result in extensive aggregation and precipi-tation of the carrier. 1-ethyl-3(3-dimethylami-nopropyl) carbodiimide (EDC; see AlternateProtocol 4) is a water-soluble carbodiimide andshould not be used with peptides containinginternal Lys, Glu, Asp, Tyr, or Cys residues.Alternate Protocol 5 describes a simple photo-chemical coupling strategy (Gorka et al., 1989).

Another good alternative for most peptides isthe production of a multiple antigenic peptide(MAP; see Alternate Protocol 2). With thismethod the composition of the peptide is not aconcern beyond its potential solubility properties.In most cases, since hydrophilic sequences areselected, this also is not a major problem. Bothfour- and eight-branched MAPs have been foundto be effective. However, four-branched MAPsare recommended because they are less prone tosynthesis problems and are easier to characterize.

As with any synthetic peptide, the productmust be well characterized before use. If thepeptide is not what it was intended to be, thisdecreases the probability of generating anti-bodies that will recognize the protein. At thevery least, check synthetic peptides for homo-geneity by analytical HPLC and correct massby mass spectrometry (see UNITS 10.21 & 10.22).Characterization of MAP can be more problem-atic due to their multibranched nature (Mints etal., 1997): HPLC and mass spectrometric analy-sis can be compromised by the presence of fourto eight peptide chains per molecule, each ofwhich may have only a small percentage of modi-fication at any particular residue but which in theaggregate contribute to broad spectra. However,this feature of MAPs usually does not tend tocompromise their ability to form antigens of theproper peptide since the correct sequence is usu-ally present in high enough concentration that asignificant amount of specific antibody is pro-

duced among the polyclonal population.Amino acid analysis (UNIT 10.1B), which is lesssensitive to multiple small differences, tends togive a reasonable assessment of the MAP integ-rity.

Anticipated ResultsThe methods outlined in this unit produce

an effective polyclonal antiserum against anintact protein from a single peptide sequence∼50% to 70% of the time. Therefore, it isadvisable to prepare two or three different pep-tides from a given protein to increase the prob-ability of at least one of them being effective.

Time ConsiderationsComputer-assisted analysis of a protein se-

quence and inspection of the data to selectseveral candidate sequences takes from 5 to 30min. Manual analysis of a protein sequence cantake several hours but can certainly be accom-plished in <1 day. Selection of peptide designand manner of synthesis as well as selection ofa coupling method will take <1 hr. Actual prepa-ration of the peptide can be accomplished in 3to 4 days, but this may vary depending on theturnaround time of the synthetic laboratory.Coupling a synthetic peptide to a carrier proteintakes from 1 to 2 days. Although not coveredin this unit, production of the antisera will varywith the animal and protocol used, but gener-ally requires 2 to 3 months. It is thereforeadvisable to inject several animals with differ-ent peptides at one time.

Literature CitedChou, P.Y. and Fasman, G.D. 1974. Prediction of

protein conformation. Biochemistry 13:222-245.

Creighton, T.E. 1993. Proteins: Structure and Molecu-lar Properties, 2nd ed. W.H. Freeman, New York.

Gailit, J. 1993. Restoring free sulfhydryl groups insynthetic peptides. Anal. Biochem. 214:334-335.

Getz, E.B., Xiao, M. Chakrabarty, T., Cooke, R., andSelvin, P.R. 1999. A comparison between thesulfhydryl reductants Tris(2-carboxyethyl)phosphine and dithiothreitol for use in proteinbiochemistry. Anal. Biochem. 272:73-80.

Gorka, J., McCourt, D.W., and Schwartz, B.D. 1989.Automated synthesis of a C-terminal photoprobeusing combined Fmoc and t-Boc synthesisstrategies on a single automated peptide synthe-sizer. Peptide Res. 2:376-380.

Kyte, J. and Doolittle, R.F. 1982. A simple methodfor displaying the hydropathic character of aprotein. J. Mol. Biol. 157:105-132.

Mints, L., Hogue Angeletti, R., and Nieves, E. 1997.Analysis of MAPS peptides. ABRF News 8:22-26.


11.16.18

SelectingSynthetic


Antibodies

Posnett, D.N., McGrath, H., and Tam, J.P. 1988. Anovel method for producing anti-peptide anti-bodies. Production of site-specific antibodies tothe T-cell antigen beta-chain. J. Biol. Chem.263:1719-1725.

Tam, J.P. 1988. Synthetic peptide vaccine design:Synthesis and properties of a high density mul-tiple antigenic peptide system. Proc. Natl. Acad.Sci. U.S.A. 85:5409-5413.

Van Regenmortel, M.H.V., Briand, J.P., Muller, S.,and Plaué, S. 1988. Synthetic polypeptides asantigens. In Laboratory Techniques in Biochem-istry and Molecular Biology, Vol. 19 (R.H. Bur-don and P.H. van Knippenberg, eds.). El-sevier/North-Holland, Amsterdam.

Key ReferencesVan Regenmortel et al., 1988. See above.

Comprehensive treatment of theory and method.

Tam, J.P. 1988. High density multiple antigen pep-tide system for preparation of anti-peptide anti-bodies. Methods Enzymol. 168:7-15.

Original methods article for MAPs.

Internet Resourcehttp://expasy.org/tools.html

Web site for programs to analyze protein sequences.

Contributed by Gregory A. GrantWashington University School of MedicineSt. Louis, Missouri


11.16.19

Immunology

SECTION VDETERMINATION OF SPECIFIC ANTIBODYTITER AND ISOTYPE

UNIT 11.17Determination of the Specific Antibody Titer

The amount of specific antibody present in polyclonal antiserum, ascites fluid, orhybridoma supernatant can be quantitated by either solid-phase radioimmunoassay (RIA)or by direct enzyme-linked immunosorbent assay (ELISA; UNIT 11.2). In the solid-phaseassay described here (see Basic Protocol), serially diluted antiserum is incubated inmicrotiter wells previously coated with the relevant antigen. Bound antibody is detectedby employing 125I-labeled anti-immunoglobulin antibodies. The amount of specificantibody in the antiserum is then determined from a standard curve generated with aspecific antibody of known concentration. The unknown antiserum and the standardantibody are assayed in parallel. The support protocols describe the chloramine T (seeSupport Protocol 1) and IODO-GEN (see Support Protocol 2) procedures for radioiodi-nation of the anti-immunoglobulin reagent. The use of the solid-phase RIA procedure todetermine the light-chain (κ and λ) and heavy-chain (γ, µ, α) isotypes present inpolyclonal antisera and fluids containing monoclonal antibodies is also described asSupport Protocol 3.

BASICPROTOCOL

SOLID-PHASE RADIOIMMUNOASSAY (RIA) FORDETERMINATION OF ANTIBODY TITER

In this assay, a specific antigen is used to coat the wells of the microtiter plate.

Materials

AntigenCoating buffer (see recipe)Control antigen (non-cross-reactive protein)Phosphate-buffered saline (PBS; APPENDIX 2)[125I]anti-immunoglobulin reagent (see recipe; Ab must be specific for the species

in which the test antibody was raised—e.g., [125I]goat anti-mouse for mousehybridoma or [125I]goat anti-rabbit for rabbit sera)

Wash buffer (UNIT 11.2)1% bovine serum albumin in phosphate-buffered saline (BSA/PBS)Standardized antibody solution (5 mg/ml in diluting buffer)Diluting buffer (see recipe)

96-well microtiter plates compatible with γ counter such as Wallac ScintiStripsRepeater pipet (e.g., Eppendorf with disposable Combi-tips)Multichannel pipetAutomated γ counter that counts 96-well plates such as a Wallac MicroBeta TriLux

1. Prepare antigen and control antigen solutions (0.005 to 0.01 mg/ml) in coating buffer.

The optimal antigen concentration will depend on the nature of the antigen employed. Todetermine the optimal coating concentration, incubate a relatively concentrated dilutionof antiserum (1:100 dilution) in microtiter wells previously coated with serial dilutions ofantigen. Bound antibody is then detected as described below in this protocol. Counts perminute (cpm) are plotted against coating antigen concentration. The optimal coatingconcentration is that at which coating antigen is saturating (i.e., at an antigen concentra-tion corresponding to a point on the plateau of the curve).

Supplement 50

Contributed by Helen M. Cooper and Yvonne PatersonCurrent Protocols in Molecular Biology (2000) 11.17.1-11.17.13Copyright © 2000 by John Wiley & Sons, Inc.

11.17.1

Immunology

The antigen need not be pure, although for accurate and reproducible antibody titers, it isdesirable to use a preparation containing a consistent antigen concentration. Bovine serumalbumin (BSA) is a frequently used control antigen; however, the best control antigen is animmunologically non–cross-reactive protein of similar molecular weight to the antigen ofinterest.

If the antigen is stable, the antigen-coated plate can be prepared in advance and stored inPBS for up to 1 week at 4°C.

2. Pipet 50 µl of appropriately diluted antigen into columns 1, 2, and 3, and columns 7,8, and 9 of rows A to H of a 96-well microtiter plate, as shown in Figure 11.17.1. Addcontrol antigen to wells 4, 5, and 6, and 10, 11, and 12 of rows A to H. Cover plateto avoid evaporation and incubate 2 hr at room temperature or overnight at 4°C.

It is desirable to assay each antiserum dilution in triplicate. In general, eight serialdilutions of antiserum are sufficient.

The volumes of antigen, antibody, and [125I]anti-immunoglobulin reagent may be halvedif antigen and/or antibody are precious.

3. Remove coating antigens by shaking into sink. Add 100 µl wash buffer and incubate5 min at room temperature. Remove wash buffer and repeat wash cycle twice.

If the antigen is difficult to obtain, the coating antigen solution may be removed with arepeater pipet, stored at −20°C, and reused. The length of time it may be stored and thenumber of times it may be reused depend on the antigen, but it should not be used >5 times

4. Pipet 100 µl BSA/PBS into each coated well, cover plate, and incubate 1 hr at roomtemperature or overnight at 4°C. Shake out blocking buffer and wash three times withwash buffer as described above.

An alternative to BSA/PBS is 5% preimmune horse serum prepared in PBS.

5. Make serial dilutions of polyclonal antiserum (or ascites fluid or hybridoma super-natant) and standardized antibody solution (5 mg/ml) in diluting buffer. Dilutions canbe made in small glass tubes or plastic 1-ml microcentrifuge tubes. Initially, a 1:5dilution of antiserum is made, followed by seven five-fold dilutions as follows:

a. Label tubes from 1 to 8.

b. Pipet 400 µl diluting buffer into all tubes.

c. Pipet 100 µl undiluted antiserum into tube 1 and mix well by pipetting up anddown without producing foam.

d. Transfer 100 µl diluted antiserum from tube 1 to tube 2 and mix well. Continueto transfer 100 µl diluted antiserum from tubes 2 to 8 as described. Finally, removeand discard 100 µl from tube 8.

6. Follow step 5 for serial dilutions of the standard antibody solution using tubes 9 to 16.

The standardized antibody solution (5 mg/ml) is ideally a purified solution of antibodyprepared by affinity chromatography (UNIT 11.11), with the same antigen specificity as theantiserum of interest and raised in the same host animal species. The concentration of theantibody is determined from the A280 and adjusted to 5 mg/ml. (A solution of 1 mg/mlimmunoglobulin has an A280 of 1.4).

7. Pipet 50 µl diluted antiserum from tubes 1 to 8 into the first six columns of rows Ato H, respectively, as indicated in Figure 11.17.1. In a similar fashion, pipet 50 µl ofeach dilution of standard antibody from tubes 9 to 16 into wells 7 to 12 of rows A toH, respectively. Cover plate and incubate 2 hr at room temperature.

Dilutions may be made while coating antigen is incubating on the plate and then stored at4°C until required.


11.17.2

Determination ofthe Specific

Antibody Titer

8. Shake diluted antiserum and standard antibody from the plate and wash three timeswith wash buffer as described in step 3.

9. Pipet 50 µl of appropriately diluted [125I]anti-immunoglobulin reagent into each well.Cover plate and incubate ≥4 hr at room temperature, or overnight at 4°C.

Steps 9 to 12 should be carried out in a well-ventilated fume hood designated forradioactive work.

A repeater pipet is convenient for delivering the 50-�l aliquots of [125I]anti-immunoglobu-lin reagent.

10. Remove radioactive supernatant from wells using a multichannel pipet and discardwaste into appropriate radioactive waste container. Add 100 µl wash buffer to eachwell and gently pipet up and down three times with the multichannel pipet. Removewash buffer and discard it into radioactive waste container. Repeat this proceduretwice. Wash wells five more times with 100 µl wash buffer per well.

The final five washes may be poured into the sink, as the radioactivity will be negligible;however, first check radiation safety regulations.

11. Allow plates to dry for 4 hr at room temperature or under a heat lamp for 30 min.

12. Count each plate in gamma counter for one minute.

13. Determine amount of specific antibody bound from antiserum of interest and fromstandardized antibody solution as follows:

a. Calculate average cpm (cpmav) for triplicate values obtained for each antibodydilution.

b. For each dilution of antiserum, subtract cpmav obtained when a given dilution isincubated in wells coated with control antigen (background) from cpmav measuredon wells coated with specific antigen (cpmbound).

c. Plot cpmbound versus log10 (antiserum dilution; see Fig. 11.17.2A).

d. Plot cpmbound versus log10 (standard antibody in mg/ml; see Fig. 11.17.2B).

e. Select a value of cpmbound from linear section of binding curve (y in Fig. 11.17.2A)constructed for the antiserum of interest. Translate this value onto the standard

Specificantigen

Controlantigen

Specificantigen

Controlantigen

Antibodydilutionfactor

Tube

3.1 x 103

1.6 x 104

7.8 x 104

3.9 x 105

12345678

525

125625

ABCDEFGH

1 2 3 4 5 6 7 8 9 10 11 12

Antiserum Standard antibody

Figure 11.17.1 The microtiter plate setup for the determination of specific antibody titer. Wells areinitially coated with the specific antigen or the control antigen. Serial dilutions of specific antiserum,ascites fluid or hybridoma supernatant are then incubated in the coated wells. Bound antibody isdetected using a [125I]anti-immunoglobulin reagent.


11.17.3

Immunology

curve and determine standard antibody concentration corresponding to thiscpmbound (z in Fig. 11.17.2B). This concentration is equal to that of specificantibody present in polyclonal antiserum at the dilution giving the selectedcpmbound.

f. Calculate initial specific antibody concentration (i.e., titer) in undiluted serum bymultiplying standard antibody concentration (z) derived from selected cpmbound (y)by corresponding dilution factor of antiserum (x in Fig. 11.17.2A).

SUPPORTPROTOCOL 1

IODINATION OF ANTI-IMMUNOGLOBULIN REAGENTUSING CHLORAMINE T

This protocol is a clean and efficient method for covalently linking 125I to tyrosine residuesin the protein of interest. Unbound 125I is quenched with a saturated tyrosine solution andseparated from the protein-bound fraction by passage down a small desalting column.

CAUTION: Under no circumstances should the iodination reaction be carried out on anopen bench. Use a well-ventilated fume hood (confirmed by radiation safety department);do not use tissue culture hood. During the reaction, significant quantities of free 125I2 areliberated which can concentrate in the thyroid gland. Therefore, as 125I is a strong gammaemitter, it cannot be stressed strongly enough that extreme care must be taken to containthe radioactive reaction products within a well-ventilated fume hood so that they areisolated from the environment.

Ensure that lead bricks line the front of the fume hood to protect against gamma emission.Additionally, a lab coat and two layers of protective gloves should be worn throughoutthe iodination procedure.

It is recommended that the experimenter have his or her thyroid tested by the institutionalradioactive safety department before and within 1 week after carrying out this reaction.If all safety precautions are followed correctly, contamination will not occur.

cpm

boun

d

y

A

x(less dilute) (more dilute)

Antiserum dilution log10

cpm

boun

d y

z

Standard antibody log10 (mg/ml)

B

Figure 11.17.2 Theoretical binding curves for the determination of specific antibody titer. (A) Theexperimental curve. Serial dilutions of specific antiserum, ascites fluid, or hybridoma supernatantare incubated in antigen-coated wells. (B) The standard curve. Serial dilutions of the standardantibody are incubated in the antigen-coated wells. The specific antibody titer is calculated asshown: Specific antibody titer = x × z, where x is the antibody dilution that yields y cpmbound and zis the standard antibody concentration that yields y cpmbound.


11.17.4


Antibody Titer

Materials

Sephadex G-25 (Amersham Pharmacia Biotech), washed and equilibrated in PBS(see APPENDIX 2 for PBS)

DEAE-Sephadex (Amersham Pharmacia Biotech), washed and equilibrated in PBSMixed bed resin (Bio-Rad #AG 501-X8), washed and equilibrated in PBS1% bovine serum albumin in phosphate-buffered saline (BSA/PBS)Protein to be iodinated (1 mg/ml, preferably in iodinating buffer)Iodinating buffer (see recipe)Saturated tyrosine solution (see recipe)Na[125I] (1 mCi in 10 to 5 µl of 0.1 M NaOH, specific activity 17 Ci/mg; NEN

Life Sciences)Chloramine T solution (see recipe)

Plastic tubing (3 mm i.d., 10 cm long) with clampsSilanized glass wool (UNIT 5.6)Small column stand1-ml syringes equipped with 22-G needlesPlastic collection tubes with capsLead pig

1. Prepare desalting column by pushing plastic tubing securely onto tip of a Pasteurpipet and attaching a small clamp to tubing. Place a small wad of siliconized glasswool into neck of pipet. Clamp pipet to a small column stand.

The glass wool is the bed on which the chromatography resins are packed. Therefore, it isimportant that the wool is packed at the neck of the pipet tightly enough to stop the resinbeads from passing through, yet loosely enough to allow the column to flow freely.

2. Layer 1.5 to 2.0 cm Sephadex G-25 onto glass wool. After the Sephadex G-25 hassettled, gently layer 3 to 4 cm DEAE-Sephadex on top.

The DEAE-Sephadex removes aggregated protein.

3. Allow DEAE-Sephadex to settle and then layer 1.5 to 2.0 cm mixed bed resin on top.

The mixed bed resin adsorbs any free 125I− remaining in the solution.

4. Pour 50 ml BSA/PBS down the column.

The 1% BSA in the buffer saturates the resin bed and inhibits the nonspecific adsorptionof iodinated protein during the desalting procedure.

5. Draw up 50 µl saturated tyrosine solution into a 22-G needle connected to a 1-mlsyringe and 250 µl saturated tyrosine solution into a second 22-G needle connectedto a 1-ml syringe.

Do not remove air bubble from inside syringe barrel. The air bubble should separate thebase of the plunger from the solution so that the full contents of the syringe can be expelledwhen required.

6. Take all reagents and necessary equipment to a well-ventilated fume hood that hasbeen designated previously for radioactive use (preferably one used specifically foriodination of proteins using 125I; see cautionary statements preceding materials list).Place the following equipment in the hood ready for use:

a. The desalting column and collection tubes. Remove BSA/PBS from top ofdesalting column. Remove caps from collection tubes and place them close by thecolumn.

b. A number of Pasteur pipets and pipet bulbs for column elution.


11.17.5

Immunology

c. The two syringes containing saturated tyrosine solution (from step 5).

d. One 22-G needle.

e. The lead pig containing a vial of Na[125I].

f. A box of protective gloves beside the hood, within easy reach.

7. Into a third 1-ml syringe (connected to a 22-G needle), take up the following: 50 µlprotein solution (do not remove air bubble between base of plunger and proteinsolution), 40 µl air space, and 30 µl chloramine T solution.

This step can be carried out outside the fume hood, but it must be done immediately priorto commencing the reaction. Because chloramine T is a strong oxidant, the protein mustbe in contact with it for as short a time as possible.

8. Remove vial of Na[125I] from lead pig and remove cap from inner container, leavingthe small glass vial containing the radioactivity inside the inner container.

The inner container will provide shielding from the radioactivity. Under no circumstancesshould the open vial be removed from the fume hood.

9. Insert 22-G needle into rubber lid of vial containing Na[125I].

The 22-G needle acts as a vent when the protein/chloramine T solution is injected into thevial. Ensure that the needle tip lies just below the lid, well above the liquid in the vial.

10. Inject protein/chloramine T solution into vial and allow reaction to proceed oneminute. Stop the reaction by injecting 50 µl saturated tyrosine solution into vial.

11. Remove contents of vial using the syringe that contained 50 µl saturated tyrosinesolution and load contents carefully onto the desalting column.

12. Rinse vial by injecting 250 µl saturated tyrosine solution (from step 5). Remove thiswash solution and load it onto the column.

13. Allow sample to penetrate into column bed and elute with BSA/PBS. Collect twenty0.5-ml fractions.

The 125I-labeled protein is usually eluted within the first 2 to 3 ml.

14. Count 1-µl aliquots from each fraction on a gamma counter to determine where theiodinated protein peak has eluted. Pool fractions containing the radiolabeled proteinand store them at −20°C in a lead pig.

The sample may be stored in aliquots (e.g., 100 �l), depending on the volume required tomake the appropriate dilutions for the RIA.

CAUTION: Dispose of all contaminated solutions and equipment, including the desaltingcolumn, in appropriate radioactive waste receptacles.

SUPPORTPROTOCOL 2

IODINATION OF ANTI-IMMUNOGLOBULIN REAGENTUSING IODO-GEN

IODO-GEN (1,3,4,6-tetrachloro-3α,6α-diphenylglycouril) can be substituted for chlor-amine T as the iodination agent. This reagent is reported to be gentler than chloramine T.Because IODO-GEN has a longer reaction time (10 to 15 min) than chloramine T (1 min),the investigator can vary this parameter for labeling unstable proteins. The chloramine Tsupport protocol (see Support Protocol 1) is modified according to the steps below whenusing IODO-GEN.


11.17.6


Antibody Titer

Materials

IODO-GEN (Pierce #28600T)Methylene chlorideIODO-GEN–coated glass test tubes (Pierce #28601T or see recipe)Drierite

Additional reagents and equipment for iodination of anti-immunoglobulin reagentusing chloramine T (see Support Protocol 1)

1. Proceed with steps 1 to 6 of the chloramine T protocol (see Support Protocol 1).

2. Rinse IODO-GEN–coated glass tube with 200 µl iodinating buffer to remove anyloose microscopic flakes of IODO-GEN. If flakes are visible, reagent has not beenproperly plated and another tube should be tested. Add 100 µl protein solution.

3. Proceed with step 8 of the chloramine T protocol (see Support Protocol 1). Insertthird syringe into vial containing Na[125I] and transfer contents into the IODO-GEN–coated vial containing protein to be labeled. Incubate mixture for 10 to 15 min atroom temperature.

4. Proceed as described in steps 11 to 14 of the chloramine T protocol (see SupportProtocol 1).

SUPPORTPROTOCOL 3

DETERMINATION OF ANTIBODY ISOTYPES

The optimal dilutions of the isotyping reagents (anti-µ, γ, α, λ, and κ) are determinedusing a set of standard monoclonal antibodies of the µ, γ, α, λ, or κ isotype, respectively.This protocol describes the determination of the optimal dilution of an anti-µ isotypingreagent, but the same protocol can be used for optimization of isotyping reagents of allspecificities. Optimization assays for all isotyping reagents must be carried out in parallel.

After establishing the optimal dilutions of isotyping reagents, their use in determining theisotypes of mouse antiserum is described for the µ, γ1, γ2, and γ3 heavy chains and κ andλ light chains. Initially, the specific antibody of interest is bound to the wells of a 96-wellmicrotiter plate coated previously with the antigen. The appropriately diluted isotypingreagents are then incubated within the wells. The extent of specific binding is determinedusing a second [125I]anti-immunoglobulin antibody specific for the animal species fromwhich the isotyping reagents were derived.


Isotype standards (0.1 µg/ml)Isotyping reagents (see recipe)

Determine the optimal dilutions for isotyping reagents1. Pipet 25 µl of µ, γ, α, λ, and κ isotype standards (0.1 µg/ml) and control antigen (0.1

µg/ml) into all wells of columns 1 and 2, 3 and 4, 5 and 6, 7 and 8, 9 and 10, and 11and 12, respectively, of a microtiter plate (see Fig. 11.17.3 for diagram). Cover plateto avoid evaporation and incubate 2 hr at room temperature or overnight at 4°C (seeFig. 11.17.4).

2. Remove isotype standards from plate by flicking contents into the sink. Add 100 µlwash buffer to each well, cover plate, and incubate 5 min at room temperature.Remove wash buffer and repeat the wash step twice.


11.17.7

Immunology

3. Pipet 100 µl blocking buffer onto each coated well, cover plate, and incubate 1 hr atroom temperature or overnight at 4°C. Shake out blocking buffer and wash three timesas in step 2.

4. Label a set of 1.5-ml microcentrifuge tubes 1 to 8.

5. Add 990 µl diluting buffer to tube 1 and 800 µl diluting buffer to tubes 2 to 8.

6. Pipet 10 µl anti-µ isotyping reagent into tube 1 and mix well by pipetting up anddown.

7. Remove 200 µl of 1:100 dilution anti-µ isotyping reagent from tube 1, add it to tube2, and mix well. Continue to transfer 200 µl diluted anti-µ reagent from tube 2 to 8as described. Finally, remove and discard 200 µl from tube 8.

8. Pipet 50 µl of each dilution of the anti-µ reagent from tubes 1 to 8 into wells of rowsA to H, respectively, and incubate plate 2 hr at room temperature.

Each dilution of the isotyping reagent is assayed in duplicate.

9. Remove isotyping reagent from plate and wash plate as described in step 2.

10. Carry out the remainder of the assay (see Basic Protocol, steps 9 to 12).

11. Average the cpm of the duplicate values (cpmav) for each dilution of isotyping reagentbound to each isotype standard.

12. Subtract the cpmav value obtained for a given dilution of anti-µ reagent incubated inwells coated with control antigen from the cpmav value calculated for that dilutionincubated in wells coated with each of the isotype standards, respectively (cpmbound).

13. Choose the dilution of anti-µ isotyping reagent at which minimal, if any, cross-reac-tivity with the nonspecific isotype standards is observed.

The dilutions chosen for all isotyping reagents must yield equivalent cpmav values to ensurecomparable sensitivity of detection for each isotype within a given antibody population.

Isotypingreagentdilution

Tube

12345678

102

5 x 102

2.5 x 103

1.3 x 104

6.3 x 104

3.1 x 105

1.6 x 106

7.8 x 106

ABCDEFGH

1 2 3 4 5 6 7 8 9 10 11 12

µ γ α λ κControlantigen

Figure 11.17.3 Microtiter plate setup for the determination of the optimal dilutions of the isotypingreagents. The wells are coated with the appropriate isotype standards. Serial dilutions of theisotyping reagent are incubated in the coated wells. Bound isotyping reagent is detected using a[125I]anti-immunoglobulin reagent specific for the animal species from which the isotyping reagentis derived.


11.17.8


Antibody Titer

Determine the isotypes14. Pipet 50 µl coating antigen (i.e., specific antigen) into all wells of rows A and B of

the microtiter plate and 50 µl control antigen into all wells of rows C and D. Coverplate to avoid evaporation and incubate 2 hr at room temperature, or overnight at 4°C.

Each isotyping reagent will be assayed in duplicate. The volumes of antigen, antibody, and[125I]anti-immunoglobulin reagent may be reduced if antigens and/or antibody are inlimited supply.

15. Remove coating antigens and block wells (see Basic Protocol, steps 3 and 4).

In this assay only a single specific antibody concentration (10 �g/ml) is required.

16. Add 50 µl antiserum (diluted to give a specific antibody concentration of 10 µg/ml)to each well. Incubate plate 2 hr at room temperature. After incubation, shake outantibody, add 100 µl wash buffer to each well, and incubate plate 5 min at roomtemperature. Remove wash buffer and repeat wash step twice.

17. Pipet 50 µl of each of the appropriately diluted isotyping reagents (in duplicate) intothe designated wells: anti-µ into rows 1 and 2, anti-γ1 into rows 3 and 4, anti-γ2 intorows 5 and 6, anti-γ3 into rows 7 and 8, anti-κ into rows 9 and 10, and anti-λ into rows11 and 12. Cover plate and incubate 2 hr at room temperature.

When using anti-γ isotyping reagents, nonspecific binding to the � heavy chain can besignificantly reduced by adding 5 �l of 0.1 M 2-mercaptoethanol to the 50 �l of antigen-specific antibody in each well prior to the 2-hr incubation period.

18. Wash wells as described in step 16 above. Complete remainder of assay (see BasicProtocol, steps 9 to 12).

19. Determine amount of each isotyping reagent specifically bound in the wells asfollows:

a. Average the cpm for each set of duplicates (cpmav).

b. For each isotype specificity, subtract the cpmav observed in wells coated withcontrol antigen from the cpmav observed in wells coated with specific antigen(cpmbound).

cont

rol

antig

ensp

ecifi

can

tigen A

B

C

D

Isotyping reagents

anti-µ anti-γ1 anti-κ anti-λanti-γ2 anti-γ3

1 2 3 4 5 6 7 8 9 10 11 12

2580 2612 347 302 106 80 91 76 1386 1235 83 67

80 123 2464 2398 44 98 105 125 1068 1039 55 69

106 54 81 93 24 51 118 91 59 78 96 148

103 51 108 102 81 68 62 80 75 37 97 66

Figure 11.17.4 Microtiter plate setup for the determination of the isotype/s of the specific antibodypopulation. The specific antibody (10 µg/ml) is incubated in the antigen (specific or control) coatedwells. The appropriately diluted isotyping reagent is then added to the wells. Bound isotyping reagentis detected using a [125I]anti-immunoglobulin reagent specific for the animal species from which theisotyping reagent is derived. Typical (data as cpm [125I]anti-immunoglobulin antibody bound) isshown for a mouse µκ antibody in rows A and C and a mouse g1k antibody in rows B and D.


11.17.9

Immunology

20. Compare cpmav values obtained for each heavy-and light-chain isotyping reagent todetermine the immunoglobulin isotypes present. See Figure 11.17.4 for typical resultsfor a mouse γ1κ antibody and a mouse µκ antibody.


Chloramine T solutionPrepare an 0.56 mg/ml solution by adding 5.6 mg chloramine T to 10 ml iodinatingbuffer (recipe below). Use immediately.

Coating buffer (0.1 M carbonate buffer, pH 9.6)Stock A, 0.2 M: 21.2 g Na2CO3 (anhydrous) per liter H2OStock B, 0.2 M: 16.8 g NaHCO3 per liter H2OFor pH 9.6, add 80 ml A + 170 ml B + 250 ml H2O

Diluting buffer0.5 g Tween 202.5 g bovine serum albumin (BSA; 0.25%)1.0 g sodium azide (0.1%)Add phosphate-buffered saline (PBS; APPENDIX 2) to 1 liter

2.5 g BSA may be replaced with 10 ml preimmune horse serum (1% final concentration).

NOTE: Do not add sodium azide if the HRPO-antibody conjugate is being used.

[125I]anti-immunoglobulin reagentChoose anti-immunoglobulin reagent specific for the animal species from which theantibody population being assayed is derived and iodinate according to SupportProtocol 1 or 2. Dilute the stock solution of [125I]anti-immunoglobulin reagent indiluting buffer to give 20,000 cpm/50 µl.

5 ml of diluted [125I]anti-immunoglobulin reagent are required per 96-well plate.

Exercise extreme caution at all times when using radioactive materials, no matter how dilutethe sample. Follow all institutional radiation safety regulations with respect to pertinentbiological hazards and radioactive containment procedures.

Iodinating buffer (0.5 M phosphate buffer, pH 7.5)Stock A: 137.03 g Na2HPO4⋅7H2O per liter H2OStock B: 68.99 g NaH2PO4 (anhydrous) per liter H2OAdd 400 ml stock A to 100 ml stock B and check pH

IODO-GEN–coated glass test tubesPipet 100 µl methylene chloride solution containing 10 µg IODO-GEN into a 12 ×75 mm glass tube (suitable for 100 µl protein at 1 mg/ml) and leave overnight toevaporate in a well-ventilated fume hood under a gentle stream of nitrogen.

IODO-GEN is insoluble in aqueous media and will remain immobilized on the reactionvessel wall after the iodinated sample has been removed. It is stable and can be stored coatedon the glass up to 2 months at −20°C over Drierite in a desiccator.

Isotyping reagentsAnti-rabbit isotyping reagents are available from ICN Immunobiologicals. Anti-mouse isotyping reagents are available from a number of immunochemical suppli-ers.

Saturated tyrosine solutionMix 50 g tyrosine and 50 ml iodinating buffer in a test tube. Cap the tube and incubate30 min in a boiling water bath. Allow to cool before use.


11.17.10


Antibody Titer

COMMENTARY

Background InformationThe solid-phase RIA described here pro-

vides a very sensitive method for the determi-nation of the specific antibody titer within agiven antiserum. The protein antigen is immo-bilized on the walls of the microtiter wells,which results in the denaturation of a significantproportion of the antigen. Thus, the solid-phaseRIA allows the detection of antibodies thatrecognize both the native and denatured formsof the antigen. The immobilization of the pro-tein antigen greatly enhances the antigen-anti-body interaction, making the assay relativelyaffinity independent. The specific antibody titerdetermined using this assay includes both con-formationally dependent and independent an-tibody specificities, as well as antibodies whoseaffinity for the antigen is extremely low (Ka =101 to 102 M−1). In addition, the solid-phaseRIA is extremely useful for studying antibodyspecificity when quantities of protein antigenare limited.

In the Basic Protocol, [125I]anti-immuno-globulin reagent is in the RIA as an alternativeto the ELISA (UNIT 11.4), which can also be usedto detect bound antibody. A wide variety ofspecies-specific anti-immunoglobulin reagentscan now be easily obtained from many biotech-nology companies and can be labeled with 125Iusing either of the procedures described in thesupport protocols. In some cases, it may bemore convenient to use 125I-labeled protein Ain place of the anti-immunoglobulin reagent.However, protein A does not bind all species ofimmunoglobulin. A list of immunoglobulinspecies to which protein A binds is shown inTable 11.17.1. A more extensive list is providedby Goding (1978).

Support Protocol 1 describes the covalentcoupling of 125I to protein molecules usingchloramine T. Although this procedure yieldshigh specific activities (12 to 16 µCi/µg ofimmunoglobulin), given a sufficient number ofsurface-exposed tyrosine residues, chloramineT is a strong oxidant and can cause significant

losses in antibody activity after limited expo-sure (i.e., >1 min). In addition, some proteinsmay be much more susceptible to denaturationthrough oxidation than immunoglobulins. Insuch cases, a less harsh method of radio-iodina-tion should be used: the IODO-GEN proceduredescribed in Support Protocol 2 or the lactop-eroxidase procedure (see Hudson and Hay,1980; Mishell and Shiigi, 1980). Both of theseprocedures also link 125I to proteins throughtyrosines residues but yield lower specific ac-tivities. Finally, the addition of a large atom,such as the iodine atom, to the protein surfacecan potentially lead to some degree of confor-mational change, regardless of the radio-io-dination procedure employed. Therefore, it isimportant that no more than one atom of 125I isincorporated per protein molecule. The aboveprotocols yield a substitution level well belowthis level.

Monoclonal antibodies are used in a varietyof techniques such as immunoprecipitation(UNIT 10.16), western blotting (UNIT 10.8), ELISAs(UNIT 11.2), radioimmunoassays, and immuno-histochemistry. In such techniques a secondantibody reagent, tagged with a readily detect-able molecule (e.g., 125I, alkaline phosphatase,or horseradish peroxidase) is often required toenhance the sensitivity and/or selectability ofthe technique. As a consequence, it is necessaryto determine the isotype of the monoclonalantibody to be used. Since polyclonal antiseraare also frequently employed in these tech-niques, it is desirable to know the distributionof isotypes within the specific antibody popu-lation. As a result of its monoclonality, theisotype of a given monoclonal antibody is sin-gular; however, a polyclonal antiserum willusually contain specific antibodies bearing avariety of isotypes. The heavy-chain isotypesfound in rabbits are µ, γ1, γ2, α1, α2, and ε.Similar isotypes are found in mice and humans,although these species have more subtypes inthe µ, γ, α, and ε families. The light-chainisotypes in all mammals are either λ or κ but in

Table 11.17.1 Binding of Immunoglobulin Classes andSubclasses to Protein A

Species Classes whichbind well

Classes whichbind weakly

Rabbit IgG IgMMouse IgG2a IgG2b IgG3 IgG1 IgMHuman IgG1 IgG2 IgG4 IgA2 IgMRat IgG IgM


11.17.11

Immunology

varying ratios. Thus, although these assays aredescribed for polyclonal rabbit antisera, theymay also be used to isotype sera or monoclonalantibodies from other species. The RIA willgive only a qualitative estimation of the spec-trum of specific isotypes present in an antis-erum. A more quantitative estimation of thespecific isotypes present may be obtained byaffinity chromatography, as described in UNIT

11.11, using anti-isotype reagents conjugated toaffinity resins.

Critical ParametersFor the accurate determination of specific

antibody titer using the solid-phase RIA, it iscritical that the standardized antibody solutionbe of similar composition to that of the antis-erum being titered. First, both standard andunknown antibody populations must be derivedfrom the same animal species and be specificfor the same antigen. Second, because the dif-ferent immunoglobulin isotypes will vary intheir affinity for antigen and in their interactionwith the anti-immunoglobulin reagent, the iso-type composition within the standard antibodypopulation and antiserum of interest should beequivalent. Third, it is important that the stand-ard antibody retains maximum binding activity.A significant population of denatured or other-wise inactive standard antibody will result inlow binding activity per milligram of protein(standard antibody) and therefore an underes-timation of specific antibody titer.

All anti-immunoglobulin reagents cross-re-act to a certain extent with different isotypes.Therefore, preliminary assays must be carriedout on each isotyping reagent to determine theoptimal dilution that will allow minimal non-specific cross-reactivity while yielding a sensi-tivity equivalent to that attained for the otherisotyping reagents.

To achieve unambiguous and reproducibleresults, the optimal working dilutions of theisotyping reagents must be determined accu-rately. It is important that a working dilution bechosen at which little, if any, cross- reactivitywith other nonspecific isotypes occurs. At thesame time, the dilution of each isotyping re-agent should be chosen such that the maximumnumbers of cpm specifically bound by eachreagent, at the chosen dilution, are equivalent.This ensures that the RIA will be able to detectthe individual immunoglobulin isotypes, viathe specific recognition of the respective iso-typing reagents, with equivalent sensitivity.

TroubleshootingThe solid-phase RIA is very reliable as long

as all reagents are prepared carefully. However,two potential problems may arise. First, a highlevel of nonspecific binding may occur whenassaying the antiserum of interest. For thehigher dilutions of antiserum (i.e., 1:125 to1:390,000), the nonspecific binding should be<5% of the maximum specific binding. Ifhigher values are obtained, it is likely that otherserum components are adhering to the walls ofthe plastic wells or to the antigen and sub-sequently interacting nonspecifically with theRIA reagents. Protein aggregates can be re-moved from the antiserum by ultracentrifuga-tion for 1 hr at 145,000 × g. Initially dilutingthe antiserum 1:10 or 1:25, instead of 1:5, alsomay help to reduce the background. If theseprocedures do not alleviate the problem, it maybe necessary to remove the nonspecificallybound serum proteins using DEAE–Affi-GelBlue chromatography (see UNIT 11.14, AlternateProtocol).

A second potential problem is the denatura-tion of the anti-immunoglobulin reagent duringthe iodination procedure. As stated in SupportProtocol 1, chloramine T is a strong oxidantwhich should not be in contact with protein forany longer than specified. If no radioactivity isdetected when the wells are counted in thegamma-counter, the 125I-labeled anti-immuno-globulin reagent has most likely lost its bindingactivity. Repeat the iodination procedure ensur-ing that the antiimmunoglobulin reagent is incontact with the chloramine T for as short a timeas possible. Alternatively, the IODO-GEN pro-cedure (see Support Protocol 2) can be used.

In addition, the following should be consid-ered: although the working dilutions of theisotyping reagents are optimized so that cross-reactivity with nonspecific isotype standardsare minimal, significant cross-reactivity amongthe heavy-chain isotypes often occurs whenassaying polyclonal antisera.

Anticipated ResultsThe titer of specific antibody in a given

antiserum is dependent on the immunogenicityof the antigen (see introduction to this chapter).A titer of 5 to 10 mg/ml of specific antibodycan be expected for a plant or bacterial antigen.A mammalian protein antigen may produce atiter significantly less than this, 0.1 to 2 mg/mldepending on the evolutionary relationship be-tween the protein antigen and the analogousprotein in the host.


11.17.12


Antibody Titer

Time ConsiderationsThe solid-phase RIA can be completed in 1

or 2 days, depending on the length of the coat-ing and blocking incubations. Labeling theanti-immunoglobulin reagent with 125I can becompleted within 60 to 90 min. The actualreaction time is 1 to 5 min.

Literature CitedGoding, J.W. 1978. Use of staphylococcal protein A

as an immunological reagent. J. Immunol. Meth.20:241-253.

Hudson, L. and Hay, F.C. 1980. Practical Immunol-ogy, 2nd ed., Blackwell Scientific Publishers,Oxford.

Key ReferenceMishell, B.B. and Shiigi, S.M. (eds.) 1980. Solid-

phase radioimmune assays. In Selected Methodsin Cellular Immunology, pp. 373-397. W.H.Freeman, N.Y.

Contributed by Helen M. CooperWalter and Eliza Hall InstituteMelbourne, Australia

Yvonne PatersonUniversity of PennsylvaniaPhiladelphia, Pennsylvania


11.17.13

Immunology

SECTION VIPREPARATION AND USE OF SPECIALIZEDANTIBODIES

UNIT 11.18Identification of Polyol-ResponsiveMonoclonal Antibodies for Use inImmunoaffinity Chromatography

The extreme specificity of an antigen-antibody reaction can be used as a powerful step inprotein purification processes. For most applications, the antibody is covalently coupledto a supporting matrix (such as cross-linked agarose). A crude or partially purifiedpreparation of material containing the protein antigen is passed over the antibody-matrix,resulting in the selective capture of the target protein. However, one of the limitations ofthis technique, known as immunoaffinity chromatography (UNIT 10.11A), has been thathigh-affinity antigen-antibody complexes are difficult to dissociate, often leading toinactivation of the protein product during elution from the immobilized antibody.

Some antigen-antibody complexes are dissociated in the presence of a combination of alow-molecular-weight polyhydroxylated compound (polyol) and a nonchaotropic salt.These conditions seem to be generally nondenaturing and, in some cases, even protein-stabilizing. This type of antibody is designated “polyol-responsive.” These antibodies canbe easily identified and isolated as monoclonal antibodies (MAbs) from a typical fusion,using standard hybridoma procedures. They have proven to be very valuable reagents forthe immunoaffinity purification of active, labile, multi-subunit protein complexes.

BASICPROTOCOL

The basic protocol describes pertinent steps in hybridoma production (steps 1 to 6),screening of MAbs for polyol responsiveness (steps 7 to 17), optimization of elutingconditions (steps 18 to 23), preparation of the MAb and the MAb-conjugated resin (steps24 to 40), use of the MAb-conjugated resin in immunoaffinity chromatography (steps 41to 47), and, finally, regeneration and storage of the MAb-conjugated resin (steps 48 to50).

Screening of the MAbs for polyol responsiveness is essential to identifying the appropri-ate MAb early in the process and can be performed even at the master-well stage duringstandard hybridoma production, after the fusion but before cloning. Alternatively, anexisting collection of MAbs can be screened by this method before a large amount ofMAb is produced for preparation of the MAb-conjugated resin. The screening procedureis a modified enzyme-linked-immunosorbent assay (ELISA), termed “ELISA-elutionassay.” The ELISA-elution assay can also be used to help optimize the elution conditions.

Materials

Purified or partially purified immunogenMice for immunization (UNIT 11.4)1× phosphate buffered saline (PBS; see recipe for 10×)HAT medium (UNIT 11.6)1% nonfat dry milk (see recipe)PBST (see recipe)1× TE buffer, pH 7.9 (see recipe for 10×)1× TE buffer, pH 7.9 (see recipe for 10×) containing salt-polyol at appropriate

concentrationPolyols (e.g., ethylene glycol, propylene glycol, 2, 3-butanediol, glycerol)

Supplement 54

Contributed by Nancy E. Thompson and Richard R. BurgessCurrent Protocols in Molecular Biology (2001) 11.18.1-11.18.9Copyright © 2001 by John Wiley & Sons, Inc.

11.18.1

Immunology

Salts (e.g., ammonium sulfate, sodium chloride, potassium glutamate)Anti-mouse IgG conjugated to horseradish peroxidase (HRPO)OPD substrate (see recipe; or use other suitable substrate for HRPO)1 M H2SO4

Mouse sub-isotyping kit (e.g., American Qualex; also see UNIT 11.3)Protein A-agarose (Repligen) or DE-52 cellulose (Whatman)Cyanogen bromide–activated Sepharose 4B (Sigma, or see UNIT 10.16)1 mM HClBicarbonate coupling buffer (see recipe)1 M ethanolamine, pH 8.3 (see recipe)Acetate washing buffer (see recipe)2% (w/v) NaN3

Crude material containing protein of interest1× TE buffer, pH 7.9 (see recipe for 10×) containing 2 M potassium thiocyanate

(KSCN)1× TE buffer, pH 7.9 (see recipe for 10×) containing 100 mM NaCl and 0.02%

NaN3

96-well cell culture platesMicrotiter plate reader capable of reading at 490 nm30-ml sintered glass filter and vacuum flaskEnd-over-end rotator10-ml disposable polypropylene column (Bio-Rad)

Additional reagents and equipment for production of monoclonal antibodies (UNITS

11.4-11.11), ELISA (UNIT 11.2), immunoblotting (UNIT 10.8), preparation ofantibody-Sepharose (UNIT 10.16), dialysis (APPENDIX 3C), and protein assays (UNIT 10.1)

Prepare mouse hybridomas1. Immunize mice with desired immunogen (UNIT 11.4) and monitor the response by an

indirect ELISA (UNIT 11.2) and immunoblotting (UNIT 10.8). Identify mice that have atiter well over 1:1000.

Several injections, administered over 4 to 6 weeks, are usually necessary to achieve sucha titer.

2. About 1 month after the last immunization dose, inject the mouse to be used for thefusion intraperitoneally or intravenously with soluble immunogen dissolved in 1×PBS. Use ∼2 to 3 times the amount of immunogen that was used in the lastimmunization dose.

3. Three days later, harvest the spleen, prepare a single-cell suspension, and fuse withthe desired plasmacytoma cell line (UNIT 11.7).

4. Plate cells, suspended in HAT medium, into 96-well cell culture plates (∼10 to 20plates depending upon past history of fusion frequency) as described in UNIT 11.7.

5. About 10 to 14 days after the fusion, screen the hybridomas for specific antibodyproduction, using 100 µl of cell culture medium and a standard ELISA procedure(UNIT 11.2).

If an impure antigen is used, a secondary screen, using a microimmunoblot assay (UNIT

10.8) might be necessary.

6. Add back 100 µl of HAT medium to the positive wells. After 1 to 2 days of incubation,perform the ELISA-elution assay as described in the following steps.

Screen for polyol responsiveness by the ELISA-elution assay7. Coat a microtiter plate with antigen (30 to 100 ng/well, contained in 50 µl of PBS)

for 1 hr at room temperature. Remove the antigen solution and then block with 1%


11.18.2

Polyol-ResponsiveMonoclonalAntibodies

nonfat dry milk at 200 µl/well by incubating for at least 2 hr at room temperature, orovernight at 4°C.

8. Remove the blocking solution. Add 50 µl of the cell culture medium from antigen-specific antibody positive wells to each of two wells of the microtiter plate.

One well will serve as the buffer control, the other well will serve as the polyol-responsivetest well.

9. Incubate at room temperature 1 to 1.5 hr, then wash wells three times with PBST.Remove residual PBST.

10. To one well, add 100 µl of 1× TE buffer, pH 7.9. To the other well add 100 µl of 1×TE buffer, pH 7.9, containing the polyol/salt combination.

For screening, a combination of 0.75 M ammonium sulfate and 40% (v/v) propylene glycolworks well. This combination seems to identify more polyol-responsive MAbs than the 1 MNaCl and 50% (v/v) ethylene glycol originally reported.

11. Incubate 20 min at room temperature, tapping the plate about every 5 min to ensuremixing of the solution in the wells, then wash wells three times with PBST. Removeresidual PBST.

12. Add 50 µl of the proper dilution (usually 1:1000) of HRPO-conjugated anti-mouseIgG antibody in 1% non-fat dried milk. Incubate 1 hr at room temperature.

13. Wash plate eight times with PBST. Remove residual PBST.

14. Add 100 µl of OPD substrate. Allow color to develop for 1 to 5 min.

CAUTION: OPD is a carcinogen. Use appropriate care to avoid contact. Other substratesfor HRPO can be used, but OPD seems to be the most sensitive.

15. Add 50 µl of 1 M H2SO4 to stop the reaction. Stop each control and test reaction inparallel.

Different hybridomas will produce MAbs that have different titers and affinities. Therefore,different reaction times (1 to 5 min) might be needed for the individual antibodies. Othersubstrates may require use of different stop solutions.

16. Read the reactions on a microtiter plate reader at 490 nm.

A MAb is considered to be polyol-responsive if the OD reading of the well treated with thepolyol/salt is 50% or less than the OD reading of the control well. The above wavelengthis for OPD; other substrates may require a different wavelength.

17. Clone the polyol-responsive MAb-producing hybridomas at least twice by limitingdilution (UNIT 11.8). Freeze the hybridomas for future antibody production (UNIT 11.9).

Save the cell culture supernatant. It can be used for optimizing the elution conditions anddetermining the isotype of the MAb.

Optimize elution conditions by ELISA-elution assay18. Coat with antigen and block a microtiter plate (see step 7) for each of the polyol-re-

sponsive MAbs.

19. Prepare the desired polyol/salt combinations in 1× TE buffer, pH 7.9.

These solutions can be prepared at many different concentrations. Generally a range ofeach polyol and salt is assayed. Convenient salt concentrations are 0 M, 0.1 M, 0.25 M,0.5 M, and 0.75 M and convenient polyol concentrations are 0%, 10%, 20%, 30%, 40%,and 50% (v/v). It is sometimes possible to use higher salt concentrations, but some saltsare not soluble at high concentration in the higher percentages of polyols. These solutions


11.18.3

Immunology

can be made ahead and stored in the refrigerator, but should be warmed up to roomtemperature before use.

20. Add 50 µl of the cell culture supernatant to each well. and incubate 1 to 1.5 hr at roomtemperature. Wash the wells three times with PBST.

21. Set up a grid on the microtiter plate so that all each polyol/salt combination is testedin duplicate. Run several different combinations of polyol/salt on one plate.

22. Add 100 µl of each polyol/salt solution to each of two wells. Incubate 20 min at roomtemperature, tapping the plate about every 5 min to ensure mixing of the solution inthe wells. Wash the plate three times with PBST. Remove any residual PBST.Continue with the ELISA-elution assay (steps 12 to 16).

23. Average the OD readings for the duplicate samples and plot the value as a functionof polyol/salt concentrations.

A typical set of data is shown in Figure 11.18.1.

Prepare the MAb and MAb-conjugated resin24. Isotype each of the MAbs using a commercially available isotyping kit (or see UNIT

11.3).

Knowledge of the isotype is helpful for designing the purification of the MAb. The cellculture supernatant from freezing the hybridomas can be used for this assay.

25. Prepare ascites fluid from the hybridoma (UNIT 11.10).

Because some research institutes have placed restrictions on the production of ascites fluidin mice, alternative methods of large-scale production of MAbs should be considered (UNIT

OD

0.5

1.0

1.5

2.0

AS

NaCl

AS/EG

NaCl/EG

AS/PG

NaCl/pg

0.0

0.50.450.40.350.30.250.20.150.10.050

Salt concentration (M)

Figure 11.18.1 Optimization of eluting conditions using the ELISA-elution assay. Two differentsalts at several concentrations were tested with two different polyols. Sodium chloride (NaCl) andammonium sulfate (AS) were used at the indicated concentrations. Ethylene glycol (EG) andpropylene glycol (PG) were used at 40% (v/v).


11.18.4


11.10). The scale of the MAb production should be such that ∼5 to 10 mg of purified MAbcan be produced.

26. Purify mouse IgG2a and IgG2b on staphylococcal protein A–agarose (UNIT 11.11).

Mouse IgG1 does not bind well to protein A, but can be purified to ∼80% purity bychromatography on DE-52 cellulose. Add saturated (at 4°C) ammonium sulfate solutionto achieve a 45% ammonium sulfate cut of the ascites fluid. Stir on ice bath ∼30 min.Centrifuge to collect the precipitate, dissolve the precipitate in 1/2 volume of 50 mM Tris⋅Cl,pH 7.0, and 25 mM NaCl, and dialyze overnight against this buffer. Centrifuge to clarify.Equilibrate a 5-ml column of DE-52 in the same Tris buffer. Apply the material to the DE-52column and collect the flowthrough fractions. Most mouse IgG1 MAbs (and some of theIgG2a and IgG2b MAbs) flow through DE-52 under these conditions.

27. Determine the amount of MAb (in mg) by taking an absorbance reading at 280 nmand applying the following formula:

Absorbance × dilution factor/1.38 × volume of MAb solution (ml)

28. Calculate the amount of MAb needed for the conjugation, using an estimate of 2.5mg MAb per ml of swollen resin.

Slightly lower amounts can be used, but increasing the amount of antibody will notnecessarily increase the capacity of the resin. Hybridomas vary in the amount of MAb thatis produced in the ascites fluid. This is usually in the range of 1 to 10 mg/ml with mosthybridomas producing between 1 to 3 mg/ml. Thus ∼5 to 10 ml of ascites fluid will generallysuffice for preparation of a useful amount of MAb-conjugated resin.

29. Calculate the amount of cyanogen bromide-activated Sepharose needed to preparethe desired amount of resin.

Other matrices with other coupling chemistries can be used. However, because cyanogen-bromide–activated Sepharose performs well and is relatively inexpensive, the protocol isdescribed for this resin. 1 g of dried Sepharose swells to ∼3.5 ml of resin. Thus, 1 g wouldrequire 8.75 mg of MAb. The volumes described below are useful for 3.5 ml of Sepharose.

30. Dialyze (APPENDIX 3C) the appropriate amount of purified MAb against bicarbonatecoupling buffer overnight at 4°C and centrifuge 15 min at 6000 × g, 4°C, to clarify.

31. Swell the appropriate amount of cyanogen bromide-activated Sepharose in 1 mMHCl for ∼15 min at room temperature.

32. Wash the Sepharose on a 30-ml sintered glass filter with ∼200 ml of 1 mM HCl.Quickly wash the resin with ∼20 ml of bicarbonate coupling buffer and transfer tothe solution (∼10 ml) containing the MAb.

33. Mix the solution, end-over-end, on a laboratory rotator for 2 hr at room temperature.

34. Collect the resin on the sintered glass filter, reserving the flowthrough to determinethe amount of MAb coupled. Take the absorbance of the flowthrough at 280 nm andestimate the percentage of coupling, again assuming that 1 mg MAb/ml will give anabsorbance at 280 nm of 1.38.

35. Transfer the resin to a solution of 1 M ethanolamine, pH 8.3. For 3.5 ml of resin, use∼10 ml of ethanolamine solution.

36. Mix the solution, end-over-end, on a laboratory rotator for 2 hr at room temperature.

37. Collect the resin on the sintered glass filter and wash with ∼100 ml bicarbonatecoupling buffer.

38. Wash the resin on the sintered glass filter with ∼100 ml acetate washing buffer.


11.18.5

Immunology

39. Repeat steps 37 and 38 twice. Wash and resuspend in bicarbonate coupling buffer(∼10 ml).

40. Add 1/100 vol of 2% NaN3 and store at 4°C.

Purify target protein on MAb-conjugated resin41. Prepare the crude material from which the protein is to be purified by an appropriate

method to maintain stability of the protein.

Immunoaffinity chromatography (UNIT 10.11A) is generally one step in a protein purificationprotocol. Frequently, it is the only chromatography step necessary. The initial steps in thepurification procedure will vary with the nature of the material. Usually, some initial stepsmust be taken to remove nucleic acids to lower viscosity. In addition, generally somefractionation step might be used to reduce the volume of the material. See Chapter 10introduction for guidelines on how to structure a protein purification procedure. Thereferences contain protocols for preparing material from yeast (Edwards et al., 1990),wheat germ and calf thymus (Thompson et al., 1990), and bacteria (Thompson et al., 1992;Thompson and Burgess, 1994).

42. Pack a 10-ml disposable polypropylene column with the antibody-Sepharose (UNIT

10.11A). Apply the crude material to the MAb-conjugated resin.

Alternatively, this step can be performed in batch mode.

43. Wash the resin with the appropriate buffer containing ∼0.5 M salt but no polyol.

44. Move the resin to room temperature. If the material had been applied to the MAb-conjugated resin in a batch mode, pour the resin into a small column.

45. Elute the column with 1× TE buffer, pH 7.9, containing the salt/polyol.

Care should be taken to apply the polyol/salt solution gently to a well-packed column toavoid causing the resin to float due to the high density of the polyol/salt solution. Forreasons that are not clear, the polyol/salt elution is more effective at room temperature thanat 4°C. However, if lability of the protein is a concern, all steps up to the elution step canbe performed at 4°C. Fractions containing the protein can also be put in an ice bath asthey come off the column. Therefore, the protein is usually at room temperature for lessthan 30 min.

46. Perform a quick protein assay such as a Bradford assay (UNIT 10.1) and pool the peakfractions.

47. Dialyze the protein (APPENDIX 3C) against two changes of an appropriate buffer, for 2hr each at 4°C, to remove salt and polyol.

It is necessary to remove the salt and polyol to avoid interference in subsequent biologicalor biochemical assays. A buffer should be chosen that will maintain the stability and activityof the particular protein.

Regenerate and store MAb-conjugated resins48. Wash the resin with ∼10 ml of 1× TE buffer containing 2 M KSCN.

This can be performed in either batch or column mode, but the MAb-conjugated resinshould not be exposed to KSCN for longer than ∼20 min.

49. Wash the resin with 1× TE buffer containing 100 mM NaCl and 0.02% NaN3.

50. Store the MAb-resin at 4°C.

MAb-conjugated Sepharose can be stored up to ∼6 months at 4°C. After ∼6 months, someleaching of MAb from the resin can be detected. Most MAb-conjugated Sepharose prepa-rations can be used ∼10 times. After ∼10 uses, the column usually still works but morecontaminants seem to be recovered in the product.


11.18.6



Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, seeAPPENDIX 2; for suppliers, see APPENDIX 4.

Acetate washing bufferTo 900 ml H2O, add:13.6 g sodium acetate trihydrate29.22 g NaClAdjust to pH 4.0 at 23°C with glacial acetic acidAdd H2O to 1000 mlStore up to 6 months at room temperature

Bicarbonate coupling bufferTo 900 ml H2O, add:8.4 g NaHCO3

29.22 g NaClAdjust to pH 8.3 at 23°C with 1 N NaOHAdd H2O to 1000 mlPrepare fresh

Citrate buffer, 0.1 MSolution A: 29.4 g sodium citrate, dihydrate in 1000 ml H2OSolution B: 21.0 g citric acid, monohydrate in 1000 ml H2OAdd solution B to ∼500 ml of solution A until pH 5.0 is achieved. Filter sterilize andstore up to 1 year at 4°C.

EDTA, 0.2 M, pH 7.974.4 g disodium EDTA dihydrate700 ml H2OAdjust to pH 7.9 at 23°C with 10 M NaOHAdd H2O to 1000 ml

Ethanolamine, 1 M, pH 8.3To 80 ml of bicarbonate coupling buffer (see recipe), add 6 ml ethanolamine. AdjustpH to 8.3 at 23°C with 6 M HCl. Add coupling buffer to 100 ml. Prepare fresh.

Nonfat dry milk, 1%5 g nonfat dried milk50 ml 10× PBS (see recipe)450 ml H2OStore up to 1 week at 4°C

OPD substrate10 mg o-phenylenediamine (OPD)10 ml H2O10 ml 0.1 M citrate buffer (see recipe)20 µl 30% H2O2

Prepare fresh

PBST100 ml 10× PBS (see recipe)900 ml H2O1 ml Tween 20Store up to 2 months at room temperature


11.18.7

Immunology

Phosphate-buffered saline (PBS), pH 7.4, 10×To 900 ml H2O, add:80.0 g NaCl2.0 g KCl21.6 g Na2HPO4⋅7H2O2.0 g KH2PO4

Add H2O to 1000 mlStore up to 6 months at room temperature

TE buffer, pH 7.9, 10×25 ml 2 M Tris⋅Cl, pH 7.9 (see recipe)0.5 ml 0.2 M EDTA, pH 7.9 (see recipe)Add H2O to 100 ml with H2OStore up to 1 year at room temperature

Tris⋅Cl, 2 M, pH 7.9242.2 g Tris base500 ml H2OAdjust to pH 7.9 at 23°C with 6 M HClAdd H2O to 1000 ml

COMMENTARY

Background InformationThe development of practical methods for

generating MAbs has renewed interest in adapt-ing immunoaffinity chromatography tech-niques for protein purification. Most early stud-ies focused on low-affinity MAbs. However, inorder for a MAb to effectively capture an anti-gen from a dilute solution it must be a high-af-finity antibody. Most high-affinity antigen-an-tibody interactions are difficult to disrupt, re-quiring conditions that are generally denaturingto proteins, such as extremes of pH or 6 M urea.About 5% to 10% of mouse MAbs are polyol-responsive antibodies (Thompson et al., 1992)which release the antigen in the presence of acombination of low-molecular-weight polyoland a nonchaotropic salt. In addition to beinggentle and protein-stabilizing, these reagentsare very inexpensive and can be used economi-cally on a large scale.

This procedure can be used to purify labile,multi-subunit enzymes, such as eukaryoticRNA polymerase II (Edwards et al., 1990;Thompson et al., 1990) and bacterial RNApolymerase (Thompson et al., 1992; Marshaket al., 1996). In fact, the immunoaffinity chro-matography step was the critical purificationstep for the eventual crystallization of the yeastRNA polymerase II (Cramer et al., 2000). It hasalso been used for purification of eukaryotictranscription factors expressed in bacteria(Thompson and Burgess, 1994, 1999).

In some cases, it is desirable to follow theimmunoaffinity step with a high-resolutionion-exchange step to remove any minor con-taminants. Sepharose has some nonspecificbinding capacity, and some batches ofSepharose seem to have higher nonspecificbinding than others.

Using the ELISA-elution assay, it is possibleto screen for polyol responsiveness at the mas-ter-well stage of hybridoma production. Thus,if the goal is to isolate this type of MAb,valuable time can be saved by identifying theMAb early in the process. In addition, theELISA-elution assay can also be used toquickly screen different polyol/salt combina-tions and to help optimize eluting conditions.Some polyol-responsive MAbs respond to avariety of polyol/salt combinations (Thompsonet al., 1990, 1992). However, some MAbs aremore limited in their response (Thompson andBurgess, 1999). It is worth noting that polyol-responsive MAbs are not limited to a particularsub-class of mouse IgG.

Critical Parameters andTroubleshooting

One of the most critical parameters of thisprocedure is that the epitope for the MAb mustbe accessible in solution. The ELISA-elutionassay can generate some false positives. Thatis, a MAb might appear to be polyol-responsivein this assay, but it does not capture the antigenfrom solution. This is probably due to distortion


11.18.8


of the antigen on the polystyrene surface, re-sulting in exposure of epitopes that are notaccessible in solution. It is wise to examine theability of the MAb to remove the antigen fromsolution early in the identification procedure.This can be done with a simple immunodeple-tion assay using cell culture supernatant(Thompson and Burgess, 1994).

When a MAb is screened from ascites fluidor other concentrated source of antibody, caremust be taken to ensure that the MAb concen-tration is in the linear range. Otherwise a falsenegative might result due to overloading of theMAb. Thus, a 50% reduction might not benoticed, resulting in a false negative.

Another critical parameter is that the elutionof the antigen from the immobilized MAb ismuch more efficient at room temperature thanat 4°C. However, with careful planning it ispossible to limit the time that the protein isexposed to the higher temperature.

High levels of reducing agents in buffers canresult in reduction of the disulfides in the MAb.When the antigen is eluted with the polyol/salt,some immunoglobulin chains can be detectedin the protein product. To avoid this problem,the material should be applied in buffer thatdoes not contain reductant. In most cases, re-ductant can be added to each fraction as itcomes off the column. In cases where reducingagent is required for enzyme stability, the 0.05to 0.1 mM DTT can be used in the columnbuffers, but this might result in decreasing thelife of the column and recovery of some MAbchains in the product.

Anticipated ResultsApproximately 5% to 10% of MAbs are

polyol-responsive; therefore, the probability ofisolating one from a standard fusion is quitehigh. In most cases, the immunoaffinity chro-matography step is the only chromatographystep necessary for purification of the protein,resulting in a more rapid purification proce-dure. Because the protein is eluted in salt anda polyol, the product is in a solution that isgenerally protein-stabilizing.

Time ConsiderationsThe ELISA-elution assay takes ∼0.5 days to

perform. However, when screening hybrido-

mas at the master-well stage, the cells need toincubate ∼1 day between the initial screen forspecific antibody and the ELISA-elution assay.Thus, this adds ∼2 days to the hybridoma pro-duction.

Production of large amount of MAb by as-cites fluid production usually takes 2 to 3weeks.

The purification and conjugation of theMAb takes 2 days. Because the MAb can startto leach off the resin upon extended storage, theresin should be used within 6 months of conju-gation.

Literature CitedCramer, P., Bushnell, D.A., Fu, J., Gnatt, A.L.,

Maier-Davis, B., Thompson, N.E., Burgess,R.R., Edwards, A.M., David, P.R., and Kornberg,R.D. 2000. Architecture of RNA polymerase IIand implications for the transcription mecha-nism. Science 288:640-649.

Edwards, A.M., Darst, S.A., Feaver, W.J.,Thompson, N.E., Burgess, R.R., and Kornberg,R.D. 1990. Purification and lipid-layer crystal-lization of yeast RNA polymerase II. Proc. Natl.Acad. Sci. U.S.A. 87:2122-2126.

Marshak, D.R., Kadonaga, J.T., Burgess, R.R.,Knuth, M.W., Brennan, W.A. Jr., and Lin, S.-H.1996. Strategies for Protein Purification andCharacterization. Cold Spring Harbor Labora-tory Press, Cold Spring Harbor, N.Y.

Thompson, N.E. and Burgess, R.R. 1994. Purifica-tion of recombinant human transcription factorIIB by immunoaffinity chromatography. ProteinExpress. Purif. 5:468-475.

Thompson, N.E. and Burgess, R.R. 1999. Immu-noaffinity purification of the RAP30 subunit ofhuman transcription factor IIF. Protein Express.Purif. 17:260-266.

Thompson, N.E., Aronson, D.B., and Burgess, R.R.1990. Purification of eukaryotic RNA polym-erase II by immunoaffinity chromatography. J.Biol. Chem. 265:7069-7077.

Thompson, N.E., Hager, D.A., and Burgess, R.R.1992. Isolation and characterization of a polyol-responsive monoclonal antibody useful for gen-tle purification of Escherichia coli RNA polym-erase. Biochemistry 31:7003-7008.

Contributed by Nancy E. Thompson and Richard R. BurgessUniversity of WisconsinMadison, Wisconsin


11.18.9

Immunology

Documents

CHAPTER 11 Immunologysun-lab.med.nyu.edu/files/sun-lab/attachments/CPMB.Ch.11.Immunology.pdf · in this book documents the importance of immunological methods to the puriﬁcation