JOURNAL OF BACTERIOLOGY, Oct. 1978, p. 324-3300021-9193/78/0136-0324$02.00/0Copyright © 1978 American Society for Microbiology

Vol. 136, No. 1

Printed in U.S.A.

Crossed Immunoelectrophoresis, in the Presence of Tween 20or Sodium Deoxycholate, of Purified Membrane Proteins from

Acholeplasma laidlawiiKARL-ERIK JOHANSSON* AND HENRI WROBLEWSKIt

The Membrane Group, Institute of Biochemistry, Biomedical Center, University of Uppsala, S- 751 23Uppsala, Sweden

Received for publication 25 July 1978

Five membrane proteins from Acholeplasma laidlawii have been previouslypurified on a large scale. These proteins have been used to establish the relation-ship between the precipitation lines obtained by crossed immunoelectrophoresisof solubilized cell membrane proteins from A. laidlawii in the presence of theneutral detergent Tween 20 or those obtained in the presence of the anionicdetergent sodium deoxycholate. This relationship, which was unambiguouslyestablished for four of the five proteins, was determined by tandem or "parallel"crossed immunoelectrophoresis of the sodium deoxycholate-solubilized mem-brane together with the purified proteins. Membranes from strain A of A.laidlawii were composed of proteins, which were immunologically related to andprobably identical to membrane proteins from strain B of this organism.

Crossed immunoelectrophoresis (11) is an ex-tremely useful technique for quantitative andqualitative analysis of proteins, and many appli-cations have recently been reviewed (2, 13).However, intrinsic membrane proteins, whichare in general insoluble in ordinary buffers, canonly be analyzed by this technique if a detergentis included in the agarose gel (3) (at least in thefirst direction).Acholeplasma laidlawii is a procaryote be-

longing to the mycoplasmas (order Mycoplas-matales), which are characterized by their per-manent lack of a cell wall (14, 16, 18). Thisproperty makes them very suitable for mem-brane studies. The proteins of intact myco-plasma cells (17), as well as of their cytoplasm(1) and membranes (1, 8, 9, 19-22; H. Wr6b-lewski, Ph.D. thesis, University of Rennes,Rennes, France, 1974), have been analyzed bycrossed immunoelectrophoresis. A. laidlawii (8,9, 20, 22) and other mycoplasmas (1, 17, 19-22)have been used in these studies. The neutraldetergents Triton X-100 (1, 17, 21), Tween 20 (8,9, 20), and Brij 58 (20) and the anionic detergentssodium deoxycholate (DOC) (19-22), sodiumlauroylsarcosinate (Sarkosyl; 20), and sodiumdodecyl sulfate (SDS; 20) have been utilized.Both Tween 20 and DOC have been found to

be very useful detergents for analysis of mem-

t Present address: Laboratoire de Biologie Cellulaire, Fa-culte des Sciences Biologiques, Complexe de Beaulieu, 35042Rennes Cedex, France.

brane proteins from A. laidlawii. However, thepatterns of precipitation lines obtained bycrossed immunoelectrophoresis are completelydifferent depending upon which of these twodetergents is used (e.g., see 8, 22). Several of themembrane proteins from A. laidlawii have beenisolated (7-9; K.-E. Johansson and H. Wr6b-lewski, Proc. Soc. Gen. Microbiol. 3:152, 1976)and partially characterized (K.-E. Johansson,Ph.D. thesis, University of Uppsala, Uppsala,Sweden, 1974). These proteins have been usedfor determination of the relationships betweenthe bands obtained by polyacrylamide gel elec-trophoresis in Tween 20raiidlieep itationlines obtained by crossed immunoelectropho-resis in the presence of the same detergent (8).The aims of the present investigation were (i)

to identify the precipitation lines obtained bycrossed immunoelectrophoresis in DOC, sincethis detergent in many cases can be complemen-tary to Tween 20, and (ii) to relate these to theband patterns obtained by SDS-gel electropho-resis and to the precipitation lines obtained bycrossed immunoelectrophoresis ofTween 20-sol-uble proteins. It is very important to establish arelationship between precipitation lines ob-tained by crossed immunoelectrophoresis underdifferent conditions and bands obtained by poly-acrylamide gel electrophoresis to rationalize thedifferent nomenclatures used for identical pro-teins (cf. the early work with membrane proteinsfrom the human erythrocyte). Such an estab-lished relationship is necessary to facilitate com-

324

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

ANALYSIS OF MEMBRANE PROTEINS FROM A. LAIDLAWII 325

parison of results from laboratories that usedifferent analytical methods.

MATERIALS AND METHODSCultivation of A. laidlawii and preparation of

plasma membranes. A culture ofA. laidlawii (strainB, PG 9) was supplied originally by S. Razin (Depart-ment of Clinical Microbiology, The Hebrew Univer-sity-Hadassah Medical School, Jerusalem, Israel). Thestrain has not been subsequently cloned. It has beenpassaged for several years in our laboratory (in Upps-ala) and, therefore, may differ from the original strainB. It will be referred to as strain B(ju). Both thecultivation of the organisms and the preparation ofmembranes were, with some minor modifications (8),performed as described by Razin et al. (15).

Solubilization of the plasma membrane andisolation of membrane proteins. Tween 20, apolyoxyethylene sorbitan monolaurate obtained fromAtlas Chemie GmbH, Essen, Germany, was used tosolubilize the membranes. The complete procedurehas been described previously (8). The purificationprocedure for membrane proteins from A. laidlawii isoutlined in Fig. 1. Isolation of the Tween 20-solublemembrane proteins from the supernatant fluid wasmainly based upon different electrophoretic proce-dures (7-9). The Tween 20-insoluble membrane pro-tein D12 was extracted from the Tween residue withDOC, which was obtained from E. Merck AG, Darms-tadt, Germany. The Tween residue was prepared byrepeatedly extracting 100 mg (dry weight) of mem-

branes with Tween 20 (6) and finally washing it with4 ml of water. The Tween residue was extracted with2 ml of 5% DOC in 0.1 M glycine-NaOH buffer (pH9.1) and centrifuged as described in the legend to Fig.1. A supernatant and a residue were obtained. If mem-branes were dissolved directly in DOC, however, noresidue was obtained. The DOC supernatant washighly enriched in protein D12, which was furtherpurified by chromatographic molecular sieving (gelfiltration) on Bio-Gel P-200 in the presence of 0.5%DOC in the above buffer. The fractions correspondingto the second peak in the chromatogram only con-tained protein D12, as judged from polyacrylamide gelelectrophoresis in SDS.Crossed immunoelectrophoresis. In crossed im-

munoelectrophoresis according to Laurell (11), theantigens (e.g., detergent-solubilized membrane pro-teins) are first separated by agarose gel electrophore-sis. The antigens are then forced into the antibody-containing gel by electrophoresis at a pH which isclose to the isoelectric point of the antibodies and ina direction perpendicular to the first-dimensional elec-trophoresis. Precipitation lines are formed againsteach protein in the sample to which the antiserumcontains the corresponding precipitating antibodies.A modified technique for crossed immunoelectro-

phoresis was used (4), and the experiments were per-formed in the presence of either Tween 20 (9) or DOC(21) to keep the membrane proteins soluble. Approx-imately 10 ug of proteins in the unfractionated deter-gent extracts and 1 ug of the purified proteins wereanalyzed. The concentration of the antiserum, which

2

Step 3

- DOC, dialysis Stops

FIG. 1. Isolation of membrane proteins from A. laidlawii. One volume of a suspension of membranes (35mg of protein and lipid per ml) was mixed with I volume of 5% Tween 20 in 0.1 M tris-(hydroxymethyl)aminomethane-hydrochloride buffer (pH 8.0) in the first step (6). The mixture was centrifugedat 180,000 x g for 1 h at 2°C. In the next step the proteins in the Tween supernatant were fractionated byagarose suspension electrophoresis (ASE) in the presence of 1% Tween 20 (7-9). The proteins in the fractionsfrom ASE were purified by preparative polyacrylamide gel electrophoresis (PPGE) and dextran gel electro-phoresis (DGE) in the absence ofdetergents (8) andASE in thepresence of1% Tween 20 and 6M urea (8). Thebulk of Tween 20 was removed from proteins T4. and T4b by PPGE in the absence of detergents (8). Tween 20-insoluble protein D12 was extracted from the Tween residue (6) with 5% DOC as described in the text andfurther purified by chromatographic molecular sieving on Bio-Gel P-200 in the presence of 0.5% DOC. DOCwas removed from protein D12 by extensive dialysis.

VOL. 136, 1978

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

326 JOHANSSON AND WROBLEWSKI

A

T4b-

FIG. 2. Crossed immunoelectrophoresis of theTween 20-soluble fraction of A. laidlawaii strainB(ju) membranes in thepresence ofTween 20 (A) andofmembranes from A. laidlawii strain A solubilizedwith DOC in the presence ofDOC (B). Both experi-ments were performed with homologous antisera.

was unfractionated, was 10 pl/cm2, and the thicknessof the agarose layer in the second dimension was 1.2mm.

Polyspecific antiserum against A. laidlawii mem-brane proteins was prepared in rabbits by immuniza-tion subcutaneously with a suspension of whole mem-branes from A. laidlawii strain B(ju) mixed withFreund incomplete adjuvant (Fig. 2A; 5) or with a

suspension of whole membranes from A. laidlawiistrain A (PG 18, ATCC 23206) mixed with completeadjuvant (Fig. 2B and 3; 19).

It can be shown by "parallel" or tandem crossedimmunoelectrophoresis (10) which precipitation linecorresponds to a certain protein, if this protein can bepurified. In these techniques, the purified protein isapplied in a well beside or in front of a well with the

unfractionated detergent extract (see Fig. 3). Parallelcrossed immunoelectrophoresis was used to identifyseveral components in the DOC crossed immunoelec-trophoretic patterns (see Fig. 3B, D, F, and H). Anal-ysis of protein D12 was done by a tandem crossedimmunoelectrophoresis (10), which is shown in Fig.3J.

RESULTS AND DISCUSSIONCrossed immunoelectrophoresis of A.

laidlawii membrane proteins in the pres-ence of Tween 20. Figure 2A shows crossedimmunoelectrophoresis of the Tween 20-solublefraction (the so-called Tween supernatant [6])of A. laidlawii cell membranes. The designa-tions previously suggested for these proteins (6,8, 9) are indicated on the figure. One advantageof using Tween 20 for solubilization of the cellmembrane from A. laidlawii was the selectivity(6), which can be utilized as a first step inpurification of the membrane proteins (7-9). An-other advantage of using Tween 20 (and otherneutral detergents) in crossed immunoelectro-phoresis was the good resolution of the antigensin the first direction (see below).

It was usually possible to detect 15 to 20precipitation lines when the Tween supernatantwas analyzed by crossed immunoelectropho-resis. The precipitation pattern does not reflectthe protein composition of the membrane, sinceonly 50% of total membrane protein can besolubilized in one step with Tween 20, and thedifferent proteins are not solubilized to the sameextent under the conditions used (6).Crossed immunoelectrophoresis ofmem-

brane proteins in the presence of DOC.Tween 20 should not be used for solubilizationif the aim is to study all proteins of the A.laidlawii cell membrane. In such experiments itis necessary to choose a detergent which solu-bilizes the membrane as completely as possible.For example, SDS (6) and DOC (22) are veryefficient in solubilizing cell membranes of A.laidlawii. Certain problems are associated withthe use of SDS, however, if the solubilized pro-teins are to be analyzed by crossed immunoelec-trophoresis, since the reaction between antigenand antibody is interfered with by this detergent(12, 20). DOC can be more readily used forcomplete solubilization of the A. laidlawii cellmembrane and for keeping the membrane pro-teins solubilized during crossed immunoelectro-phoresis.

Figure 2B illustrates the crossed immunoelec-trophoretic pattern obtained with aDOC extractof A. laidlawii membranes run in the presenceof DOC. A reference system for designation ofthe precipitation lines has been introduced byWr6blewski and Ratanasavanh (22). The tech-niques have now been improved, however, and

J. BACTERIOL.

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

ANALYSIS OF MEMBRANE PROTEINS FROM A. LAIDLAWII 327

I.. .-. .

I...

11........T.

#.i...

F:

o_..

f

sS Wi s<t.., 6drtBiffi sS.

gE.

''.Sx t

.1,.t F.e Ngf= :'.

'4a _t te \ .g,

{ *,

.f

FIG. 3. Crossed immunoelectrophoresis in the presence ofDOG of the isolated membrane proteins T' (A),T3 (C), T47 (E), Ti, (G), and D12 (I) from A. laidlawii and a mixture of the DOC-soluble fraction of themembrane andproteins T2 (B), T3 (D), T4a (F), T4b (H), and D12 (I), usingparallel electrophoresis (B, D, F, andH) or tandem electrophoresis (J) in the first dimension.

VOL. 136, 1978

-,:..':.;*:

;:

,< -.

'f.,.*s

;:.*..:

'......

:S

........ ..

.....4

*;: ..

@. .4... tS

I

l..,

8T3

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

328 JOHANSSON AND WROBLEWSKI

.H..6

T4b _ a

.4.

i.I

,f...'....

1.

ko,

f -i

.DIG

FIG. :-COntinulied

more precipitation lines can be seen in Fig. 2Bthan in previous experiments (22). A comparisonof Fig. 2A and B shows that the resolution ofthe antigens (membrane proteins) in the firstdirection was better in the Tween system (Fig.2A) than in the DOC system (Fig. 2B) (see 20).The resolution in the DOC system was, however,good enough for observation of approximatelythe same number of precipitation lines as in theTween system. The most intensely stained pre-cipitation lines in the two systems were T3 andno. 8, respectively. These precipitation lineswere shown to represent the same protein (seebelow).Identification of immunoprecipitation

lines in the DOC reference pattern bycrossed immunoelectrophoresis of purifiedmembrane proteins in the presence ofDOC.Figure 3 shows crossed immunoelectrophoreticanalysis of several purified A. laidlawii mem-

brane proteins in the presence of DOC. Thepurified proteins were analyzed alone in the left-hand series (panels A, C, E, G, and I). As can beseen from these panels, each protein samplecontained one minor contaminant, which was ofno importance, however, for the following exper-iments. Note that protein D12 (panel I) gave acomparatively weak precipitate. In the right-hand series (panels B, D, F, H, and J), thepurified proteins were analyzed in parallel witha sample of unfractionated DOC extract or bytandem crossed immunoelectrophoresis (seeMaterials and Methods). The areas of the im-munoprecipitation lines which increased in theright-hand series compared with the correspond-ing areas of precipitates in Fig. 2B are denotedwith arrows. The results are summarized in Ta-ble 1, which also shows which protein band inSDS-polyacrylamide gel electrophoresis corre-sponds to the precipitation line. The relation-

J. BACTERIOL.

N-1

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

ANALYSIS OF MEMBRANE PROTEINS FROM A. LAIDLAWII 329

TABLE 1. Designation of membrane proteins fromA. laidlawii in different analytical systems

Crossed immunoelectrophoresis with:SDS-gel electro-phoresis (6) band Tween 20 (8, 9) DOC' lines

lines

D5 T4a 4D6 T2 5D6 T4h 6Di, T3 8D,2 ti,?t

tib

a The present nomenclature has no connection withthe reference system previously used (22).

' Could not be determined since protein D12 formeda weak precipitate in the presence of DOC.'The protein corresponding to precipitation line tlb

has not yet been identified in other systems, since ithas only been prepared so far in small amounts (8).

ships between bands in SDS-gel electrophoresisand precipitation lines in crossed immunoelec-trophoresis in the presence of Tween 20 hasearlier been established by using the purifiedproteins (8).

It has recently been shown that proteins ti8and D12 are identical. Since this protein was notvery soluble in Tween 20, the best yield wasobtained by isolation from the Tween residue,where it has earlier been defined as protein D12(6). The purified protein D12 forms a compara-tively intense precipitation line when analyzedby crossed immunoelectrophoresis in the pres-ence of Tween 20 (Johansson and Wroblewski,Proc. Soc. Gen. Microbiol. 3:152, 1976). WhenD12 was analyzed in DOC, however, a weakprecipitate was formed (see Fig. 3I and J). Itwas, therefore, impossible to correlate with cer-tainty precipitation line tia (protein D12) withany of the precipitates in the DOC referencesystem by tandem crossed immunoelectropho-resis, which should, in principle, give a doublepeak (10). The reaction between protein D12 andits antibodies was apparently inhibited to someextent by DOC.Comparison ofTween 20 and DOC. Tween

20 can be used in crossed immunoelectrophoresisof A. laidlawii membranes when it is importantto have good resolution of the antigens in thefirst direction, e.g., to follow the purification ofthe Tween 20-soluble membrane proteins of thisorganism (8, 9). Other neutral detergents (e.g.,Triton X-100) gave similar patterns in crossedimmunoelectrophoresis (20). DOC can be usedwhen it is important to have quantitative solu-bilization of all the membrane proteins of A.laidlawii or other mycoplasmas, e.g., for taxo-nomic purposes (20) or to obtain a precipitationpattern which as far as possible reflects the

complete protein composition of the membrane.It must be remembered, however, that to deter-mine the quantitative protein composition ofmycoplasmal membranes by crossed immunoe-lectrophoresis, one must obtain standard curvesfor every protein, since the area under a precip-itation line depends on both the antibody andthe antigen concentrations.Properties of the antiserum. The antise-

rum, which was used in the experiments shownin Fig. 3, was raised against A. laidlawii strainA, whereas the membrane proteins, used as an-tigens, were prepared from A. laidlawii strainB(ju). These experiments show that the fiveisolated membrane proteins from strain B(ju)have counterparts that are immunologically re-lated to those in the membrane of strain A.When membranes from strains A and B(ju) wereanalyzed by crossed immunoelectrophoresiswith antiserum raised against strain A, differ-ences in the precipitation patterns were ob-served. Preliminary findings indicate that themain proteins may exist in different proportionsin the two membranes and that minor proteinsmay be absent in one and present in the othermembrane.

ACKNOWLEDGMENTSWe thank Stellan Hjerten and Cyril Smyth for constructive

criticism and Irja Blomqvist and Anne-Marie Touzalin forexpert technical assistance.

This investigation was supported by grants from the Foun-dation of Magnus Bergwall, the Foundation of Ollie and ElofEricsson, the Foundation of 0. E. and Edla Johansson, theFoundation of Helge Ax:son Johnson, The Swedish NaturalScience Research Council, and the von Kantzow Foundationand by the Centre National de la Recherche Scientifique (LAno. 256, Contrat C.N.R.S.-Universite).

LITERATURE CITED

1. Alexander, A. G., and G. E. Kenny. 1977. Characteri-zation of membrane and cytoplasmic antigens of My-coplasma arginini by two-dimensional (crossed) im-munoelectrophoresis. Infect. Immun. 15:313-321.

2. Bjerrum, 0. J. 1977. Immunochemical investigation ofmembrane proteins. A methodological survey with em-phasis placed on immunoprecipitation in gels. Biochim.Biophys. Acta 472:135-195.

3. Bjerrum, 0. J., and P. Lundahl. 1973. Detergent-con-taining gels for immunological studies of solubilizederythrocyte membrane components. Scand. J. Immu-nol. 2(Suppl. 1):139-143.

4. Clarke, H. G. M., and T. A. Freeman. 1967. A quanti-tative immunoelectrophoresis method (Laurell electro-phoresis). Protides Biol. Fluids Proc. Colloq. 14:503-509.

5. Harboe, N., and A. Ingild. 1973. Immunization, isolationof immunoglobulins, estimation of antibody titre.Scand. J. Immunol. 2(Suppl. 1):161-164.

6. Hjerten, S., and K.-E. Johanason. 1972. Selective sol-ubilization with Tween 20 of membrane proteins fromAcholeplasma laidlawii. Biochim. Biophys. Acta288:312-325.

7. Johaneson, K.-E. 1974. Fractionation of membrane pro-teins from Acholeplasma laidlawii by preparative aga-rose-suspension electrophoresis. Protides Biol. Fluids

VOL. 136, 1978

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

330 JOHANSSON AND WROBLEWSKI

Proc. Colloq. 21:151-156.8. Johansson, K.-E., I. Blomqvist, and S. Hjerten. 1975.

Purification of membrane proteins from Acholeplasmalaidlawii by agarose suspension electrophoresis inTween 20 and polyacrylamide and dextran gel electro-phoresis in detergent-free media. J. Biol. Chem.250:2463-2469.

9. Johansson, K.-E., and S. Hjerten. 1974. Localization ofthe Tween 20-soluble membrane proteins of Achole-plasma laidlawii by crossed immunoelectrophoresis. J.Mol. Biol. 86:341-348.

10. Kr0ll, J. 1973. Tandem-crossed immunoelectrophoresis.Scand. J. Immunol. 2(Suppl. 1):57-59.

11. Laurell, C.-B. 1965. Antigen-antibody crossed immunoe-lectrophoresis. Anal. Biochem. 10:358-361.

12. Nielsen, C. S., and 0. J. Bjerrum. 1975. Immunoelec-trophoretical analysis of sodium dodecylsulfate-treatedproteins. Scand. J. Immunol. 4(Suppl. 2):73-80.

13. Owen, P., and C. J. Smyth. 1977. Enzyme analysis byquantitative immunoelectrophoresis, p. 147-202. In M.R. J. Salton (ed.), Immunochemistry of enzymes andtheir antibodies. John Wiley & Sons, Inc., New York.

14. Razin, S. 1975. The mycoplasma membrane, p. 257-312.In D. A. Cadenhead, J. F. Danielli, and M. D. Rosenberg(ed.), Progress in surface and membrane science. Aca-demic Press Inc., New York.

15. Razin, S., H. J. Morowitz, and T. M. Terry. 1965.Membrane subunits ofMycoplasma laidlawii and their

assembly to membranelike structures. Proc. Natl. Acad.Sci. U.S.A. 54:219-225.

16. -Smith, P. F. 1971. The biology of mycoplasmas. AcademicPress Inc., New York.

17. Thirkill, C. E., and G. E. Kenny. 1974. Serologicalcomparison of five arginine-utilizing Mycoplasma spe-cies by two-dimensional immunoelectrophoresis. Infect.Immun. 10:624-632.

18. Tourtellotte, M. E. 1972. Mycoplasma membranes: struc-ture and function, p. 439-470. In C. F. Fox and A. D.Keith (ed.), Membrane molecular biology. Sinauer As-sociates, Stamford, Conn.

19. Wr6blewski, H. 1975. Dissolution selective de proteinesde la membrane de Spiroplasma citri par le desoxycho-late de sodium. Biochimie 57:1095-1098.

20. Wr6blewski, H., K.-E. Johansson, and R. Burlot.1977. Crossed immunoelectrophoresis of membraneproteins from Acholeplasma laidlawii and Spiro-plasma citri. Int. J. Syst. Bacteriol. 27:97-103.

21. Wroblewski, H., K.-E. Johansson, and S. Hjerten.1977. Purification and characterization of spiralin, themain protein of the Spiroplasma citri membrane.Biochim. Biophys. Acta 465:275-289.

22. Wr6blewski, H., and D. Ratanasavanh. 1976. Etudepar immunoelectrophorese bidimensionnelle de la com-position antigenique de la membrane de quelquessouches de mycoplasmes. Can. J. Microbiol.22:1048-1053.

J. BACTERIOL.

on Decem

ber 24, 2020 by guesthttp://jb.asm

.org/D

ownloaded from