THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 250, No. 14, Issue of July 25, pp. 5375-5385, 1975

Printed in U.S.A.

Two-Dimensional Gel Analysis of Soluble Proteins

CHARACTERIZATION OF GUINEA PIG EXOCRINE PANCREATIC PROTEINS

(Received for publication, November 25, 1974)

GEORGE A. SCHEELE

From the Laboratory of Cell Biology, The Rockefeller University, New York, New York 10021

SUMMARY

A two-dimensional gel technique using slab gel isoelectric focusing in the first dimension and sodium dodecyl sulfate gradient gel electrophoresis in the second dimension has been developed for the separation of soluble proteins larger than 10,000 daltons. The technique is sensitive to 0.6 pg of protein and recovery of radiolabeled proteins averages 90 %. Analysis of secretory protein from the guinea pig exocrine pancreas shows the presence of 19 distinct high molecular weight pro- teins. Each of these proteins has been characterized by iso- electric point, molecular weight, and proportionate mass. Thirteen of the 19 proteins have been identified by actual or potential enzymatic activity, accounting for 96% of the protein mass resolved by the two-dimensional gels.

Exocrine pancreatic proteins have been studied in the past by a number of laboratories (l-7, 10-16) using classical methods for separation of proteins. In most of these studies exocrine proteins were obtained by sulfuric acid extraction of tissue or by direct cannulation of the pancreatic duct, temporarily stored at -20” after deionization and lyophilization, and analyzed by column chromatographic methods. While these studies have yielded important information about proteins in general and exocrine pancreatic proteins in particular, this approach has several disadvantages for further investigation of pancreatic proteins at the cell biological level: (a) secretory proteins are extracted from tissue and stored under nonphysiological conditions, (b) large amounts of starting material are required, (c) separation of pro- teins on columns is frequently incomplete, (d) proteins eluted from columns are recovered in relatively dilute solutions requir- ing subsequent concentration procedures, and (e) the time and degree of manipulation necessary to obtain adequate separation of exocrine proteins is burdensome.

Recently, exocrine pancreatic proteins of the guinea pig have been fractionated by Tartakoff et al. (17) using column isoelectric focusing and sodium dodecyl sulfate gel electrophoresis. Secretory proteins used in these fractionation procedures were obtained under physiological conditions using in vitro pancreatic lobules as described by Scheele and Palade (18). Eight proteins (amylase, trypsinogen, chymotrypsinogen, RNase, and two forms each of procarboxypeptidase A and 13) were isolated and identified by actual or potential enzymatic activity after fractionation on

focusing columns and the relative RF values for these proteins were then determined on Na dodecyl-S041 gels. IYa dodecyl-SO4 gel electrophoresis revealed two additional proteins not identified on focusing columns. However, complete separation of proteins was not obtained by either procedure. In isoelectric focusing amylase, ribonuclease, trypsinogen, and chymotrypsinogen focused closely together at pH 9.3 to 9.6. In Na dodecyl-SO4 gel electrophoresis the four forms of procarboxypeptidase A and B ran closely together in the 45,000 to 47,000 molecular weight range and chymotrypsinogen and one unidentified protein ran closely together in the 26,000 to 28,000 range.

In order to obtain complete separation of all secretory proteins, a two-dimensional gel technique was developed to be used on mixtures of soluble proteins. This technique involves slab gel isoelectric focusing in 5% polyacrylamide in the first dimension and sodium dodecyl sulfate gel electrophoresis in gradient (10 to 2Oo/,) polyacrylamide in the second dimension. In the first dimen- sion proteins are separated by charge; in the second dimension, by molecular weight.

EXPERIMENTAL PROCEDURE

Materials

Acrylamide, N,N’-methylenebisacrylamide, and N,N,N’,N’- tetramethylethylenediamine were purchased from Eastman Ko- dak, Rochester, N. Y. Riboflavin, ammonium persulfate, lauryl sulfate (sodium dodecyl sulfate), trizma base, glycine, Triton X-114, bromphenol blue, iodoacetamide, benzoyl-alanine-methyl ester, yeast (torula) RNA grade VI, and tripalmitin were obtained from Sigma Chemical Co. Porcine a-amylase (AA), bovine DNase I (D), bovine chymotrypsinogen A (WCGC OFA), bovine tryp- sinonen ITG 1DBj. bovine RNase A (ROGB). and 1voDhilized

I ,

trypiin (TRL 1 GA, 180 pm/mg) were odtainedfrom Worthington Biochemical Corp., Freehold, N. J. Coomassie brilliant blue, sucrose (ultrapure), carbachol diisopropylfluorophosphate, hip- puryl-arginine, hippuryl-phenylalanine, benzoyl-tyrosine-ethyl ester, and tosyl-arginine-methyl ester were from Schwarz/Mann. Ampholines (pH 3.5 to 10, pH 9 to 11, pH 3 t,o 6, pH 5 to 7) were from LKB. I%, amino-acids, reconstituted from an algal protein hgdrolgsate, 1 mCi/ml, Catalog No. 2133.11, were purchased from Schwa;z/Mann Radiochemicals. Bovine plasma albumin (crystal- lized) was obtained from Armour Pharmaceutical Co. Benzamidine HCl &as from Aldrich Chemical Co. Hydrogen peroxide was from Fisher. Xylene was from Mallinckrodtj. Whatman No. 3MM chromatography paper and glass fiber paper, GF/A (2.4 cm), were from W. and R. Balston Inc., England. Dithiothreitol was from

1 The abbreviations used are: Na dodecyl-Sod, sodium dodecyl sulfate; bisacrylamide, N,N’-methylenebisacrylamide; TEMED, N, N, N’, N’-tetramethylethylenediamine; carbachol, carbamyl- choline chloride.

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5376

Nutritional Biochemical Co. Sephadex G-25 was from Pharmacia Fine Chemicals. Inc. ll-14ClTrinalmitan was from Aonlied Sci- . __ __ ences Laboratories. Soluble starch powder was from Amend Drug and Chemical Co. Porcine enterokinase was a kind gift from Drs. Suzanne Maroux and Jacques Baratti, Centre Nationale de la Recherche Scientifique, Marseille, France.

Methods

Extraction of Secretory Protein

Secretory proteins are extracted by physiological methods. Pan- creatic lobules are prepared from male albino guinea pigs (Rocke- feller University colony), weighing 450 to 500 g, by the method of Scheele and Palade (18). Carbamvlcholine (10-S M) is used to stimulate the discharge of proteins‘from pancreatic lobules to the incubation medium (Krebs-Ringer bicarbonate + glucose + amino acids) as described by Scheele and Palade (18). Three-hour incubation of lobules from one pancreas gland in 10 ml of incuba- tion medium results in a solution of secretory protein approxi- mately 2 to 3 mg/ml in concentration. The medium then is centri- fuged at 105,000 X g for 30 min at 3” to remove small pieces of con- nective tissue, and the supernatant is stored in 1.5.ml Eppendorf plastic vials at -70” following rapid freezing in liquid nitrogen. Samples can be stored with or without the protease inhibitors, 1 mM diisopropylfluorophosphate (19) and 1 mM benzamidine-HCl (20). No att,empt is made to deionize or lyophilize samples. In addition, the secreted proteins can be radiolabeled by the addition of an algal hydrolyzate mixture of 15 ‘4C-amino-acids to the incu- bation medium (5 pCi/ml) in the absence of cold amino acids (17). No attempt is made to remove labeled amino acids unless recovery studies are done with radiolabeled proteins. In this case the se- creted radiolabeled proteins are passed through a Sephadex G-25 column equilibrated with 50 mM NaCl-10 mM Tris, pH 7.0.

Two-Dimensional Gel Technique

Slab Gel Isoelectric Focusing (First Dimension)-Isoelectric focusing of secretory protein is carried out in a 5yo polyacrylamide slab gel measuring 12 X 15 cm. The container for this gel is made of one standard glass plate (6.5 X 7.5 X >{s inch) and one standard plastic plate (6.5 X 7.0 X s/4 inch) separated by three 1.33.mm thick plastic shims placed at the two sides and bottom when the container is standing in the vertical position. The final dimensions of the slab gel are determined by the positions of the shims be- tween the two standard plates. The container is held together by large binder clips (clamps) and leakage is avoided by sealing the edges with 10% polyacrylamide. The sealing solution (30 ml of acrylamide/bisacrylamide (30~,/0.8~,), 100 ~1 of TEMED, and 60 ml of water) is kept readily accessible in a loo-ml stoppered cylin- der on the workbench and polymerizes rapidly (10 min) by the addition of 100 ~1 of 10% ammonium persulfate to 10 ml of the sealing solution. The gel mixture contains 5 ml of acrylamide/ bisacrylamide (30~o/0.8~o), 10 ~1 of TEMED, 0.75 ml of Ampholine (pH 3.5 to lo), 0.25 ml of Ampholine (pH 9 to ll), 0.10 ml of Am- pholine (pH 3 to 6)) 0.05 ml of Ampholine (pH 5 to 7), 6 ml of 60% sucrose, and water to 30.ml total volume. Sucrose is added for two reasons: (a) to increase the density of the focusing slab gel which facilitates the application of the focusing strip to the second di- mension gel and (b) to increase the viscosity of the focusing me- dium which inhibits drift in the pH gradient toward the cathode due to electroendosmosis.2 Polymerization is initiated by the addi- tion of 9 drops of a saturated solution of riboflavin and 0.075 ml of 10% ammonium persulfate, and the gel mixture is immediately delivered into the gel container (positioned vertically) by gravity through 0.040 internal diameter Solveflex plastic tubing (Tech- nicon Corp., Tarrytown, N. Y.). Slow delivery prevents formation of air bubbles on the nonwettable surface of the plastic plate. Polymerization, assisted by the presence of a 5uorescent light set 2 inches from the gel, occurs in 60 min at 24 or 3”. After polymeriza- tion the gel is placed in the horizontal position with the plastic plate on top and this plate is carefully removed (using thin spatu- las to pry between the side plastic shims and the top plastic plate). This somewhat difficult step, successful approximately 90% of the time, is facilitated by the hydrophobic surface of the plastic plate and by the removal of the plate as early after polymerization as

2 See LKB Application Note, March 29,1973, p. 12.

possible. Following removal of the plastic plate, the three plastic shims and the outside sealing acrylamide are removed, leaving only the focusing gel adhered to the underlying glass plate. At this point the gel can be used immediately or set on the 8” cooling plate and used 1 to 2 hours later.

Samples of protein are applied to the focusing slab gel by one of two methods. In the first method the sample, usually 30 to 50 ~1, is absorbed to a square (1 X 1 cm) of glass fiber paper (Whatman GF/A) and the paper square is placed on the surface of the gel. Glass fiber paper was used since preliminary trials with Whatman 3MM chromatography paper showed significant adherence of amylase to the paper. Since isoelectric focusing is an equilibrium procedure, the protein sample can be applied to any part of the surface of the gel although as shown below (cf. Figs. 6 and 7) it should not be exposed to extremely high pH. Eight samples sep- arated by 0.5.cm intervals can be applied to the surface of a slab gel 12.5 cm wide. In the second method, the sample is added to the gel mixture and directly polymerized into the slab gel, either within a narrow segment of the gel or spread diffusely throughout the gel. However, the protein sample should not come into contact with the paper wicks containing electrode solutions. In order to do this, the gel mixture is made up into three parts, only one part of which contains the protein sample. The slab gel then is polymer- ized in three stages or alternatively the three gel mixtures can be polymerized at the same time if differing sucrose concentrations are used to layer one mixture above the other. In this way the protein sample is delivered to the middle segment of the gel and the size of t,his segment is determined by the volume of the gel mixture to which the sample is added.

While both methods of sample application are successful, each has certain advantages. Absorption of the sample to glass fiber paper allows simultaneous comparison of different samples in isoelectric focusing, is convenient for use with small sample volumes (10 to 50 pg), and shows banding of proteins during move- ment toward isoelectric points (similar to banding during electro- phoresis). Direct polymerization of the sample into the gel matrix ensures 106% recovery in the focusing gel, is ideal for preparative procedures, allows for concentration of protein from dilute solu- tions (24 out of 30 ml of gel solution is water), and is preferred for the determination of isoelectric points since banding of proteins will only occur at these equilibrium points.

Electrical current is applied to the focusing slab gel through electrode solutions, 2.55% phosphoric acid at the positive elec- trode, and 0.3 N NaOH at the negative electrode. Contact is made between the focusing slab gel and the electrode solution reservoirs through paper wicks, each wick consisting of three layers of What- man No. 3MM paper (13 X 20 cm) and overlapping the ends of the gel by 1 cm. Focusing is conducted with the gel slab sitting on a Savant cooling plate, cooled to 5-10” with a Lauda constant T” water circulator. Focusing is initiated with constant current of 5 ma or less delivered by a Buchler model 3-1155 power supply. Voltage, usually 100 to 125 volts in the beginning, increases slowly at first and more rapidly later, to reach approximately 500 volts over 2.5 to 3.0 hours. At 506 volts the power source is switched to constant voltage, 500 volts, and the focusing procedure cont,inued for an additional 3 hours. Total focusing time is 6 hours.

At the conclusion of the focusing procedure, the paper wicks are removed from the gel, and the pH gradient is measured at 0.5. to 1.0.cm intervals using a surface pH electrode (glass electrode ES11727-1 for slab gels; reference electrode ES3731, Ingold Elec- trodes Inc., Lexington, Mass.). Measurements can be facilitated by putting a piece of graph paper between the gel glass plate and the metal cooling surface, thereby maintaining the low temperature of the now focused gel. The gel can be subsequently manipulated in several ways : (a) fixation and staining in a solution containing 57, sulfosalicylic acid, 5yo trichloroacetic acid, 30% methanol, and 0.012% Coomassie blue, and subsequently destaining in 30% methanol/lO’% acetic acid; (b) cutting out a strip containing the focused proteins and applying it to the second dimension, de- scribed below (a companion strip can be fixed, stained, and de- stained for later reference); (c) fractionation of the gel into 15 segments (l-cm each) for enzymatic screening after homogeniza- tion of gel segments and elution of proteins (cf. “Results”). Side and center longitudinal strips are removed first for fixation, stain- ing, and destaining, providing a later reference for the precise positions of protein bands; and (d) should it be necessary to cut out individual bands, the side and center reference strips can be fixed,

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5377

stained, and destained while the rest of the gel is covered with Saran wrap and stored frozen at -20”. Later the gel is thawed at 3” (15 min) and the segment with the desired protein is removed. All cutting of gels is done by hand using a straight edge razor blade and an underlying piece of graph paper for guidance.

Sodium Dodecyl Sulfate Gradient Gel Electrophoresis (Second Dimension)-Na dodecyl-SOa gel electrophoresis is done according to the method of Maize1 (21) with the following modifications: (a) the running gel is made as a polyacrylamide gradient, 10 to 20% from top to bottom; (b) the proteins to be separated are loaded in situ in a 0.5- to 1.5-cm wide isoelectric focusing strip placed horizontally on top of the sodium dodecyl sulfate gel, and (c) the dimensions of the gel apparatus are larger, the two glass plates containing the gel (run in the vertical position) being 10.5 X 14 X 36 inch with a l-inch cutout at the top of the inside plate. Stock solutions are: acrylamide/bisacrylamide (30c/o/0.8c~o) ; 2.0 bt Tris- HCl, pH 8.8; 0.5 M Tris-HCl, pH 6.8; 607, sucrose; 107, Na dodecyl- SO,; and lOy’, ammonium persulfate. The container for this gel is made up of the two glass plates mentioned, separated by plastic shims slightly thicker than those used for the focusing gel (1.60 mm), in a manner similar to that of the previous gel. The 10 to 2Oc/, polyacrylamide gradient is constructed in the running gel by using a hkavy solution (40 ml of acrylamide/bisacrylamide; 12 ml of 2 M Tris. nH 8.8: 25 ~1 of TEMED: 6 ml of 60% sucrose: 0.6 ml of 1O7o Na*dodec$l-Sba) and a light’solution (26ml acrylamide/bis- acrylamide; 12 ml of 2 M Tris, pH 8.8; 25 ~1 of TEMEl>; 0.6 ml of 10% Na dodecyl-S04, and water up to 60 ml), initiating polymeri- zation of both solutions by the separate addition of 150 ~1 of 10% ammonium persulfate to each and introduction of these solutions into the gel container (sitting vertical) through a relatively large gradient maker. The container is filled to a mark 3.5 cm below the cutout level in one of the glass plates. One or two milliliters of iso- butanol layered on top of the acrylamide solution preserves the horizontal interface at the gel edges. Polymerization takes 60 min. Once the running gel is polymerized, indicated by the formation of a third interface at the top of the gel (air/isobutyl alcohol, isobutyl alcohol/water, and water/acrylamide), the stacking solution (12 ml of acrylamide/bisacrylamide (3Oc/,/O.8%) ; 6 ml of 0.5 M Tris, pH 6.8,25 ~1 of TEMEI); 0.5 ml of 10yfi Na dodecyl-Sod; water up to 50 ml, plus 0.2 ml of 10% ammonium persulfate) is poured in, the “slot” forms (molds which project down into the gel 1.5 cm, excluding the acrylamide solution where slots are desired) are applied, and clamps are placed to put the two glass plates in close contact with these plastic molds during the 30 min required for stacking gel polymerization. Since glass plates tend to bow out from the pressure of the acrylamide solutions, the ap- plication of the plastic molds and clamps is also necessary while the previous gradient running gel is polymerizing. Following polymerization of the stacking gel, the final steps in preparation of the gel arc: (a) removal of the “slot” forms, (b) removal of the bottom shim (assisted by the lubricating effect of several drops of water), and (c) sealing the two glass plates with gel to the gel ap- paratus by using a wide strip of parafilm, acrylamide sealing solu- tion, and two large clamps on each side.

Application of Isoelectric Focused Strip to Xa Dodccyl-SOa Gradi- east Gel-The 0.5- to 1.5.cm wide focused strip cut by razor blade from the first dimension slab gel is immersed in a 1% Na dodecyl- Sod, 3% acrylamide solution consisting of 10 ml of acrylamidel bisacrylamide (3Oo/c/O.S~~); 12 ml of 0.5 hf Tris, pH 6.8; 75 ~1 of TEMED: 10 ml of 10% Na dodecvl-Sod: and water to 100 ml (mixed thoroughly) ; pills 3.0 ml of 107;’ ammonium persulfate (mixed again). The strip is immediately lifted out with a 6-inch plastic ruler and transferred to the Na dodecyl-SOa gel apparatus, positioned so that the strip can first be laid on the inner surface of the full glass plate. The strip then will slip into the slot if a few droas of the 1% Na dodecvl-SO&-3% acrvlamide oolvmerizinn

l” Li . , ” ” ~I solution are added to the transferred strip and the adjacent empty slot with the apparatus tilted slightly (-20”). Positioning of the strip is facilitated by the blunt edges of two spatulas. If done properly, there should be no air bubbles and polymerization of the isoelectric focusing strip in the horizontal slot should occur within 10 to 15 min. The top and bottom reservoirs then are filled with sufficient electrode buffer, 6 g of Tris, 28.8 g of glycine, and 10 ml of 10yti Na dodecyl-SO4 per liter, to make electrical contact with the gel, and the system is run at 30 to 35 ma for 16 bows. The bromphenol blue tracking dye runs off the gel at ~12 hollrs. Fol- lowing electrophoresis, gels are stained for 6 hours in 0.06% Coo-

massie blue/507, methanol/lO% acetic acid, and destained by re- peated washings with 30% methanol, 1Oojc acetic acid.

Focused strips fixed, stained, and destained as described earlier can also be run in the second dimension, providing the st.rip is soaked for 1 hour in a relatively large volume (100 ml) of 1% Na dodecyl-SOa/0.05 in Tris, pH 6.8 (cf. Figs. G and 7). This allows inspection of banding patterns in focusing gels prior to the further separation of these proteins in the second dimension, removes all ampholytes, and is convenient during preliminary stlldies when running conditions for gels are being optimized and ampholyte concentrations and ranges are being explored.

Preparation of Pawreatic Secretion and Molecular W’eight Stand- ards for Aya Dodccyl-SO1 Gradiwt Gel Elcctrophoresis-Samples consist of 25 ~1 of -1 mg/ml of protein solution, 5 ~1 of 25c/c Na dodecyl-Sod, 5 ~1 of 0.1% bromphenol blue, and 15 ~1 of GOc/o su- crose. When full reduction and alkylation are desired, 5 ~1 of freshly prepared 0.1 M dithiothreitol arc added at room tempcra- ture for 30 min, followed by 10 ~1 of freshly prepared 0.1 M iodo- acetamide for an additional 30 min. Samples are delivered to gel slots in 5- to 50.~1 (usually 25.~1) quantities. Molecular weight standards used are porcine amylase and bovine plasma albumin, DNase, chymotrypsinogen, trypsinogen, and IlNase from Worth- ington Biochemical Corp.

Radioautography of Radiolabelcd Proteirzs

Gels containing 14C-radiolabeled proteins are covered with Saran wrap and dried down on Whatman No. 3MM paper lmdcr vacuum. Dried gels are then exposed to DuPont chronex II x-ray film in the dark for 1 to 3 weeks, following which the film is developed.

Quantiatio?l of Radiolabeled Proteins

Gel bands or gel spots containing radiolabclcd proteins are removed by razor blade or Bard Parker dissection blade, trans- ferred to counting vials and dissolved within 2 days in l.Oml of 300/, H,Oy at GO”. Tenmilliliters of scintillation solution (257s Triton X-114. 75% xvlene. v/v. and 8% 2-(4’.t-butvlDhcnvl‘)-5-(4”.bi- phen$yl):i,3,“4oxihiaioie (22)) ai; addled and “thi viais’are vigor- ously shaken, set at room temperature for 1 day in order to dis- charge chemical illuminescence, and then counted in a Beckman model 350 liquid scintillation counter. In the case of proteins la- beled with l4C, the degree of quenching was uniform regardless of the size of the gel piece and the efficiency of counting was 89%.

ElutioiL of Secretory Proteins from Isoelectric Focu.siltg Gel Fractions

Gel segments were homogenized in a 5-ml solution of 0.1 M

Tris (pH 8.0)/0.1 by KCl/0.02 M CaClJ20 &ml of bovine plasma albumin with 10 strokes of a Teflon homogenizer (0.004 to 0.006- inch clearance). Gel homogenates were transferred to 15.ml ccn- trifuae tubes. shaken raDidlv (180 oscillations/min) at 3” for 12 hours, and thkn centrifuged at 6,000 rpm for ld min: Supernatant volumes were typically 2.5 to 3.0 ml and when measured by en- zymatic analysis, efficiency of elution was 35 to 557,.

Idetztifkation of Pancreatic Proteins by Elzryme Activities

All measurements were made under conditions giving a linear relationship between measured activity and protein concentra- tion. Procedures used for the activation of the guinea pig zymo- gens trypsinogen, chymotrypsinogen, and the procarboxypepti- dases A and B, and the assay of the gtlinea pig enzymes amylase, ribonuclease, trypsin, chymotrypsin, and the carboxypeptidases A and B are summarized by Scheele and Palade (18).

Lipase-The method of Kaplan (23) was used with [1-‘“C]tripal- mitin (1.5 X 105 dpm/ml) assubstrate. Prior to use, contaminating labeled free fatty acids were removed from sltbstrate by extracting three times with 0.1 N NaOH in ethylene glycol. lleactions were carried out at 37” for 20 min and activity is expressed as counts per min of free fatty acids extracted into alkaline solvent. The back- ground activity was usually 200 cpm.

Procla,stase-Activation was initiated by the addition of 10 ~1 of trypsin (10 pg) and carried out at 3 or 30” for 5 min to 24 hours. Preliminary studies on guinea pig pancreatic secretion showed that: (a) without added trypsin there was no activation of the zymogen, (b) with trypsin activation occurred immediately at 3 or 30” and remained stable for at least 24 hours, and (c) under the

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5378

conditions tested, elastase activity was directly proportional to the amount of protein present. Elastase activity was measured by the hydrolysis of benxoyl-alanine-methyl ester at 254 nm and 23”. The assay mixture cdntains 1.0 mM substrate; 0.1 M Tris-HCl, DH 8.0. and 0.1 M KC1 (24). Preliminarv studies showed that (a) henzoyi-alanine-methyl‘ester was not ‘hydrolyzed by bovine ‘0; guinea pig trypsin or chymotrypsin (10 to 100 &ml) and (b) guinea pig secretion contained 3.1 units potential elastase activity per mg of proelastase.

RESULTS

Slab Gel Isoelectric Focusing (First Dimension)-Separation of pancreatic exocrine proteins was studied using isoelectric focusing in polyacrylamide slab gels. Fig. 1 shows the banding pattern of secretory protein polymerized directly into the acid portion of the gel and focused for 6 hours. There are 12 prominent bands and several faint bands. The gel is bisected since a l-cm strip was removed and run in second dimension (not shown). The pH gradient profile is displayed immediately below and aligned to demonstrate the ease with which isoelectric points are deter- mined. The bands are labeled as subsequently identified by enzymatic assay.

12

10

I 8 a

a, 0-1 6

4

2

2 4 6 8 10 12 14

Length of gel (cm)

FIG. 1. Isoelectric focusing in a polyacrylamide slab gel of guinea pig pancreatic exocrine proteins. Gel contained 1% Ampho- line (pH 3.5 to lo), 0.16% Ampholine (pH 9 to ll), 0.132% Ampho- line (pH 3 to 6), 0.066% Ampholine (pH 5 to 7), and 12a/$ sucrose in 50/, polyacrylamide. Secretory protein (0.5 mg) was polymerized into a narrow segment at the acid end of the gel. Running time, 6 hours. Bands are labeled as subsequently identified by actual or potential enzymatic activity. Banding pH under these conditions define isoelectric points.

The focusing procedure results in a small degree of gel dis- tortion usually at the acidic end of the gel (cf. Figs. 1 and 4). IXstortions in this area are greater at the sides than in the center, and when present are evidenced by rippling in the gel surface which is easily seen at the end of the run. Distortions are minimized when focusing current is maintained below 5 ma.

Sodium Dodecyl Sulfate Gradient Gel Electrophoresis (Second Dimension)-Separation of pancreatic exocrine proteins was studied using Na dodecyLS0, gel electrophoresis in a 10 to 20% polyacrylamide gradient. The two-dimensional gel technique described in this paper requires that the proteins be run in their nonreduced state to preserve enzymatic activity during the focusing procedure. However the banding pattern of nonreduced pancreatic proteins in Na dodecyl-SO4 gradient gel electrophoresis is considerably different from that of reduced and alkylated proteins (Fig. 2), indicating that these proteins are relatively abundant in disulfide bonds. While there is better separation (12 versus 10 bands) and better spread of proteins in the nonreduced state, log molecular weight correlates better with RF values of reduced and alkylated proteins (Fig. 2a) than with the Rp values of nonreduced proteins (Fig. 2b). For example, in the nonreduced state, trypsinogen migrates more rapidly than expected and bovine plasma albumin does so to the extent that its relative mobility with that of lipase 1 is reversed. However, except for trypsinogen the RF values of nonreduced pancreatic proteins can be used as a rough guide to molecular weight values (cf. Table I).

Two-dimensional Gel Analysis-The possibility of combining isoelectric focusing and Na dodecyl-SO4 gradient gel electro- phoresis to achieve further separation of pancreatic proteins was studied. Transfer of focusing strips into appropriate slots in the electrophoresis gel was facilitated by immersion of strips in 1% Na dodecyl-Sod. It was not necessary to remove ampholytes by prolonged soaking since they are small, less than 1000 daltons, and are electrophoresed ahead of proteins. The mobility of strongly basic ampholytes is retarded but still sufficient not to interfere with proteins larger than 10,000 to 12,000 daltons. Although it was sufficient to place the strip in the appropriate slot without polymerization, polymerization did retard lateral diffusion of proteins passing across the interface formed by the strip and the Na dodecyl-SO1 stacking gel. The presence of an acrylamide gradient in the Na dodecyl-SO4 running gel was critical, allowing for continuous stacking of proteins into con- centrated spots. A small amount of lateral diffusion occurred during electrophoresis, particularly when the load was large (cf. Fig. 4).

Fig. 3 shows a two-dimensional gel of radiolabeled guinea pig pancreatic proteins, unreduced, accompanied by the banding patterns of pancreatic secretion and protein standards (porcine amylase, and bovine DNase, chymotrypsinogen, trypsinogen, trypsin, and RNase) in one-dimension Na dodecyl-SO4 gradient gel electrophoresis. Below the Coomassie blue staining pattern is the radioautogram developed after 19 days’ exposure. Proteins are run in their unreduced state. A companion focusing strip (fixed, stained, and destained) is shown immediately above for reference. Ampholytes have run off the gel except for strongly basic ones, which stain heavily at the bottom of the gel. No Coomassie blue bands remain in the focusing strip. Twenty protein spots are stained with Coomassie blue, each conveniently separated from the others. All but one (spot 19) appear in the radioautogram as radiolabeled proteins and are therefore true secretory proteins. Spot 19 is bovine plasma albumin which was added to prevent adherence of secretory proteins to glass or

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Reduced and Alkyloted Nonreduced

o b

Bovine plasma albumin\

4Lipase I ‘Bovine plasma albumin I-Amylose

)-Procorboxypeptidases A 8 B

I-Proelastose I + 2

-Chymotrypsinogen

FIG. 2. Sodium dodecyl sulfate electrophoresis in a polyacryl- amide gradient gel of guinea pig pancreatic exocrine proteins: a, reduced (10 mM dithiothreitol) and alkylated (20 mM iodoaceta- mide); and 6, nonreduced. Polyacrylamide gradient, 10 to 20%. Running time, 16 hours. Protein load, 5Opg each. Bands are labeled as subsequently identified by two-dimensional gel analysis. Ba, semilog plot of molecular weight versus RF value for reduced and alkylated exocrine proteins. Standards, similarly reduced and alkylated, are: bovine plasma albumin, A; porcine amylase, V; bovine chymotrypsinogen A, 0; bovine trypsinogen, 0; and bovine RNase A,O . Bb, semilog plot of molecular weight versus RF value for nonreduced exocrine proteins. Standards, similarly nonreduced, are the same as !?a.

FIG. 3. Two-dimensional gel analysis of W-radiolabeled guinea pig pancreatic exocrine proteins with accompanying one-dimen- sional Na dodeoyl-SOA gel analysis of exocrine proteins and protein standards, corresponding radioautogram, and companion isoelec- tric focusing strip. Secretory protein (100 pg) absorbed to glass fiber paper (1 X 1 cm) and applied to acid end of isoelectric focus- ing slab gel (arrow). Focusing time, 2 hours; maximum volts, 1000 (gel slightly underfocused under these conditions). Focusing strip soaked in 1% Na dodecyl-S04, 0.05 M Tris-HCl, pH 6.8, for 30 min, then applied to Na dodecyl-SOa gradient gel (polyacrylamide, 10 to 20%) without polymerization. Protein standard loads, 25 rg each; one-dimensional pancreatic proteins, 50 fig. Na dodecylSO4 gel running time, 16 hours. Second dimension spots, stained with Coomassie blue, are numbered from 1 to $0. Radioautogram was exposed 19 days.

plastic surfaces (18). Its isoelectric point is 4.8, its apparent molecular weight is 68,000, and it frequently appears in hetero- geneous form. Focusing, run for 2 hours at 500 to 1,000 volts in this case, was not complete as evidenced by the trails of protein associated with spots 3, 5, 6, 7, 16, 17, and 18, those having moved furthest from the glass fiber paper positioned at the acidic end of the focusing gel (see arrow). However, incomplete focusing here offers two advantages: the trails indicate molecular weight isolines, and spots 3, 5, and 7 (subsequently identified as RNase, trypsinogen, and chymotrypsinogen) which, under complete focusing conditions have the same isoelectric point (pH 8.7), migrate as separate entities to that point. Fig. 4 shows a two- dimensional gel obtained following complete focusing (6 hours) in

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5380

FIG. 4. Two-dimensional gel analysis of guinea pig pancreatic exocrine proteins with companion isoelectric focusing strip. Secre- tory protein (60 pg) absorbed to glass fiber paper (1 X 1 cm) and applied to acid end of isoelectric focusing slab gel (arrow). Focus- ing complete, time 6 hours (maximum volts, 500; maximum ma, 5). Focusing strip soaked in 1% Na dodecyl-Sod-0.05 M Tris-HCl, pH 6.8, for 0.5 min, then applied to Na dodecyl-SO& gradient gel (polyacrylamide, 10 to 20%) by polymerization with 3yo acryl- amide. Na dodecyl-SO1 gel running time, 16 hours. Second dimen- sion spots, fixed with 10% acetic acid-30yo methanol and stained with Coomassie blue, 0.06~0, are numbered from 2 to BO.

the first dimension and no protein trails are seen. Protein 19 (bovine plasma albumin) appears less apparent and more heterogeneous since, in this case, 20 pg/ml instead of 200 pg/ml were added to the incubation medium.

Separation of proteins in two dimensions reveals instances where proteins have similar or identical RF values in either single dimension analysis. For instance, in single dimension Na dodecyl- SO4 gel electrophoresis (Fig. 3), this occurs among the following groups of proteins: (a) proteins 1, 2, 3, and 4, (5) proteins 7 and 12, (c) proteins 8, 9, 10, and 11, and (d) proteins 13, 14, and 15. In single dimension slab gel isoelectric focusing (Fig. 3), this occurs among: (a) proteins 12, 14, 19, and 20, (5) proteins 11 and 15, (c) proteins 2 and 8, and (d) proteins 5, 17, and 18.

Enzymatic IdentQication of Proteins-Enzymatic activity or potential activity of secretory proteins was determined by direct assay of fractions derived from focusing gels (see “Methods”). Fig. 5 shows three enzymatic activities and five potential ac- tivities (after activation) found in 15 fractions derived from a gel loaded with 2 mg of secretory protein and focused 6 hours. The center reference strip (fixed, stained, destained, and dried) is placed at the top. No protease activity was found in any of the fractions prior to activation of zymogens. Each zymogen was

Actual and Potentlot Enzymatic Activities DetermIned In Isoelectric Focusing Slab Gel

i

FIG. 5. Actual and potential enzymatic activities determined in isoelectric focusing slab gel. Secretory protein (2 mg) applied to acid end of focusing slab gel (13 X 15 cm) (arrow). Focusing time, 6 hours. Side (1.0 cm) and center (0.5 cm) reference strips were fixed, stained, and destained. The remaining slab gel was divided into 15 l.O-cm fractions and proteins were eluted by homogeniza- tion of each fraction in 5 ml of 0.1 M KC1/0.02 M CaCIZ/O.l M Tris- HCl (pH 8.0)/20 fig/ml of bovine plasma albumin and subsequent vigorous shaking overnight at 3”. Proteins were recovered in supernatant after low speed centrifugation of acrylamide pieces. The center reference strip is shown at the top, aligned for easy reference to enzyme histograms. For each of the zymogens: 0, enzyme activity prior to activation; 0, activity after activation. Tos-Arg-OMe, tosyl-arginine-methyl ester; Bz-Tyr-OEt, benzoyl- tyrosine-ethyl ester; Bz-AZa-OMe, benzoyl-alanine-methyl ester; hipp-Arg, hippuryl-arginine; hipp-Phe, hippuryl-phenylalanine.

activated separately under conditions shown to be optimal for that guinea pig zymogen (cf. Ref. 18 for trypsinogen, chymo- trypsinogen, and procarboxypeptidases A and B and “Methods” for proelastase).

Enzyme activities allowed specific identification of protein bands in isoelectric focusing gels and where focusing bands con- tained a single protein, identities could be translated to two-

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5381

dimensional spots. Where focusing bands contained more than one protein, additional evidence was required to distinguish identities of spots in two-dimensional gels. Thus, the heavy band in focusing fraction 13 is identified by enzymatic activities as a mixture of trypsinogen, chymotrypsinogen 2, and RNase. Evidence derived from considerations of molecular weight and comigration with protein standards (Fig. 3) indicates that spot 7 is chymotrypsinogen 2, spot 5 is trypsinogen, and spot 3 or 4 is RNase. The band in focusing fraction 12 is identified as amylase and on second dimension it results in the appearance of two spots, 17 and 18, which comigrate with the two bands obtained from porcine amylase. The band at the interface of focusing fractions 11 and 12 is identified as lipase 2 and corresponds to spot 16 on the two-dimensional gel. The faint band in focusing fraction 11 is unidentified and corresponds to spot 2 on the two-dimensional gel. The faint band in focusing fraction 10 is unidentified and corresponds to spot 9 on the two-dimensional gel. There are two bands in focusing fraction 9, identified as procarboxypeptidase I3 and proelastase. Molecular weight considerations identify spot 15 in the two-dimensional gel as the former and spot 11 as the latter. The faint band in focusing fraction 8 remains unidentified and corresponds to spot 1 in the two-dimensional gel. The bands appearing in focusing fractions 4, 5, and 6 are slightly distorted and must be considered as a group. By enzymatic activity they are identified as chymotrypsinogen 1, procarboxypeptidase A, and lipase 1. Molecular weight and isoelectric point considera- tions (Fig. 3 and Ref. 17) identify spot 12 as chymotrypsinogen 13 and spots 13 and 14 as two forms of procarboxypeptidase A. I<y elimination, spot 20 is identified as lipase 1. Finally, the band in focusing fraction 14 is unidentified and corresponds to spot 6 in two-dimensional gel.

Further studies were done to isolate and identify protein 20, presumed to be lipase 1 above. Secretory protein was applied to the basic end of a focusing gel and incompletely focused 3.5 hours. Protein 20 being larger in molecular weight migrated more slowly towards its isoelectric point and was intercepted alone under these conditions. Assays of fractions derived from this focusing gel revealed lipase activity coincident with the intercepted band. The complete isolation of this protein and its identification as protein 20 was demonstrated by two-dimensional gel analysis.

The secretory protein mixture used here to identify protein bands and spots by enzymatic activity did not contain protein 10 on two-dimensional analysis. Since it was likely that this protein along with protein 11 might be a form of proelastasc>, further studies were done on the secretory protein mixture containing both proteins, 10 and 11. The balld patttrn of this secretion is identical to that seen in Figs. 1 and 4. Each of the bands, repre- senting proteins 10 and 11, was individually cut out and the rest of the isoelectric focusing gel was fractionated. There was no

hydrolysis of benzoyl-alaninc-methyl ester in any of the fractions prior to activation with trypsin. Following the addition of trypsin (10 pg/300 ~1 of clution buffer) at 3” there was immediate elastase activity in each of the fractions corresponding to the two proteins of interest (9.1 and 8.1 X 1OP units, respectively). Proteins 10 and 11 also had a very small degree of chymotrypsin activity, 4.8 and 9.2 x lop3 units, respectively. Accordingly, spots 10 and 11 on the two-dimensional gel are identified as two forms of proelastase.

Recovery of Radiolabeled Proteins in Two-dimensional Gel Analysis-Usiq the two-dimensional gel analysis described, recovery of proteins applied to focusing gels by both direct polymerization into the gel and absorption onto glass fiber paper was studied. Radiolabeled proteins (5 x lo5 cpm/mg of protein)

were used to provide a means of quantitation. Labeled amino acids were removed by filtration through Eephadex G-25. Electrophoresis of radiolabeled proteins on Na dodecyl-SOa gradient gels (one dimension) showed recovery of 88.5y0 (85% in one case and 92y0 in another) of labeled protein, indicating that 10 to 127, of proteins are lost either by migration off the Na dodecyl-SO4 gel due to their low molecular weight, or during fixation and staining. Fixation and staining does not cause serious loss of proteins since recovery in gels neither fixed nor stained was no better than those fixed and stained. Two-dimen- sional gels, regardless of method of sample application and fixation technique (10% acetic acid/50% methanol or 10% sulfosalicylic acid/5yc trichloroacetic acid) showed recovery of 80% of radiolabeled protein, comparing favorably to 88% found in one-dimensional Na dodecyl-SO4 gels. Furthermore, radio- labeled proteins appeared only in Coomassie blue stained spots since recovery from gels assayed for stained spots was equal to that of gels comprehensively sliced and counted. Manipulations required in the transfer of proteins from the first to the second dimension may result in the additional 8yc loss. When the sample was directly polymerized into the gel matrix, 5% of the radio- labeled proteins remained in the focusing strip. When glass fiber paper was used for sample application, 2.5% of the radioactivity remained in the paper square and 3% remained in the focusing strip. These amounts of radioactivity probably represent labeled amino acids incorporated into molecules, possibly pcptidoglgcans (17), unable to migrate out of 5O/c polyacrylamide due to size or inability to bind sufficient Na dodccyl-SO+ There was no selec- tive loss of proteins on glass fiber paper since secretory protein profiles obtained during recovery experiments were the same using either method of sample application.

Jlodificafions in Isoalecfric Points Due to Sample Loading- When the sample of secretory protein was polymerized through- out the gel and focused for 10 hours to insure complete migration of proteins to their isoelectric points, several proteins, amylase, chymotrypsinogen 2, trypsinogen, and ribonucleasc, showed two isoelectric points. One point corresponded to the pI1 reported in this paper while the other corresponded to a higher pH, similar for each of the four proteins (pH = 9.4). The two isoelectric points of these proteins are easily visualized in the two-dimeli- sional grl pattern derived from this study (Fig. 6). Each pair of spots (18-18’, 17.17’, ‘i-7’, 5-5’, 3/4-3’/4’) is connected by a narrow band of stained protein. Ko differences arc RCCI~ in the migration of neutral or acidic proteins. Since there is no change in molecular weight, the occurrence of two isoelectric points could not be due to dimcrization caused by oxidation of sulfhydrgl groups and formation of intermolecular disulfide bridges or to significant proteolytic degradation. However, since autoactiva- tion of trypsinogen and activation of chymotrypsinogen involve the removal of only a few amino acid residues, this possibility was seriously considered. Trypsin and chymotrypsin activity were determined under these conditions on isoelectric focusing gel fractions both before and after activation with enterokinase and trypsin, respectively. There was no activity of either protcase prior to activation. Following activation, large amounts of trypsin and chymotrypsin activity appeared at the two isoelectric points of each of the zymogcns. In addition, in several other esperiments, modifications in isoelectric points occurred in the presence of 1 rnhl diisopropylfluorophosphate which was previously distributed throughout the focusing gel. Thus the appearance of additional isoelectric points for these proteins is due neither to activation nor denaturation of proteins.

Conversion of proteins from one isoelectric point to the other

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5382

7a

FIG. 6 (Zejt). Two-dimensional gel analysis of guinea pig pan- creatic exocrine proteins. Sample, 2 mgof protein, was polymerized throughout isoelectric focusing slab gel. Focusing time, 10 hours. Isoelectric focusine eel was fixed in 5% trichloroacetic acid/5% -- ,- sulfosalicylic acid/20a/, methanol, stained with 0.01297~ Coomass’ie blue, and destained with 30y0 methanol/lOyO acetic acid. Fixed, stained, and destained isoelectric focusing strip was applied to Na dodecyl-SO4 gradient gel after a l-hour soak in 1% Na dodecyl- SOJO. M Tris-HCl, pH 6.8. Na dodecyl-SO4 gel running time, 16 hours. Second dimension spots were fixed, stained, and destained as described under “Exnerimental Procedure.” Artist’s rendition of the isoelectric focusing strip is shown above the two-dimen- sional gel.

FIG. 7 (right). Two-dimensional gel analysis of guinea pig pan-

does not take place during the focusing procedure. Identical samples of secretory protein, 75 pg, were focused in the same gel, one sample loaded at the acid end and the other loaded at the basic end. Focusing was carried out for 20 hours. The two- dimensional pattern derived from the sample loaded at the acid end, the “acid pattern” (Fig. 7c), showed the familiar pattern reported here. Loading the sample of proteins in the center of the gel at neutrality shows a pattern identical to the acid pattern. The two-dimensional pattern derived from the sample loaded at the basic end, the “basic pattern” (Fig. 7b), showed the set of higher isoelectric points seen in Fig. 6. The acid pattern and the basic pattern appear physically closer together (see Fig. 7a, and compare the orientation of protein 6 in Figs. 6 and 7, b and c) because of the more prominent “cathode shift” (progressive diminution of the pH gradient at the basic end of the gel with focusing timez) at 20 hours than at 10 hours. Despite the pro- longed duration of focusing, no conversion occurs from one pattern to the other. The basic pattern under these conditions also shows modifications occurring in the neutral proteins: (a) the two forms of the proelastase each result in a smear indicating the appearance of charge heterogeneity (Fig. 7b) ; and (b) procarboxy- peptidase B now shows two isoelectric points, a major one at pH 7.0 and a minor one at pH 6.6 (Fig. 7, a and b). No changes occur in the acidic proteins, which are not shown in Fig. 7 a, b, and c.

creatic exocrine proteins. 7b shows the two-dimensional pattern resulting from sample loading (75 rg of protein) at the basic end of the isoelectric focusing gel. 7c shows the two-dimensional pattern resulting from sample loading (75pg of protein) at the acidic end of the same isoelectric focusing slab gel. Corresponding isoelectric focusing strips are shown at the top of each two-dimensional pattern. 7u shows a side-by-side comparison of the isoelectric focusing strips, b and c, taken from the same slab gel. Isoelectric focusing strips were fixed, stained, and destained prior to soaking in 1% Na dodecyl-SOJO. M Tris-HCl, pH 6.8 (1 hour) and ap- plication to the Na dodecyl-SO4 gradient gel. Na dodecyl-SOa gel running time, 16 hours. Second dimension spots were fixed, stained, and destained as described under “Experimental Procedure.”

These results indicate that modifications in isoelectric points are not due to aggregates of proteins (intra- or interspecies) at one isoelectric point and movement of the aggregate to the other isoelectric point during the focusing procedure.

The basic pattern was converted to the acidic pattern when the basic protein band (pH 9.4) was removed from the focusing slab gel, the polyacrylamide homogenized, and the protein eluted into 100 mM Tris-HCl buffer, pH 6.8, polymerized into the acid end of an isoelectric focusing gel, and refocused. Each of the four proteins, amylase, chymotrypsinogen 2, trypsinogen and ribo- nuclcase, migrated to the isoelectric points seen in the acidic pattern. This study suggests that modifications in isoelectric points are not due to chemical modifications involving covalent bonds.

Characteristics of Guinea Pig Pancreatic Exocrine Proteins- The 20 spots separated by the two-dimensional gel analysis and their physical characteristics (molecular weight, isoelectric points, enzymatic function, and proportionate mass) are sum- marized in Table I. Molecular weights were estimated from proteins run in Na dodecyl-SO4 gradient gels under two condi- tions, nonreduced and reduced and alkylated. As discussed previously, the latter gives more accurate estimates. Isoelectric points were reproducible from gel to gel, never varying more than 0.2 pH units. Proportionate mass was determined by the per cent,

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5383

TABLE I Characteristics of guinea pig exocrine pancreatic proteins

spot numbe

Enzyme or zymogen

1

2

3 Ribonuclease 4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

20

Trypsinogen

Chymotrypsinogen 2

Proelastase 2

Proelastase 1 I

Chymotrypsinogen 1

Procarboxypeptidase Al

Procarboxypeptidase As

Procarboxypeptidase B

Lipase 2

Amylase

Amylase

Lipase 1

-

0.2

0.3

0.7

0.4

33.0

2.0

16.4

0.7

8.0

1.7

3.5

8.8

8.8

3.4

3.6

8.5

a Per cent distribution of radioactivity among individual secre- tory proteins of pancreatic lobules continuously labeled with a

mixture of 15 ‘4C-amino-acids and discharged into physiological medium during 3 hours of carbachol (10m5 M) stimulation.

* Estimated from semilog plot of molecular weight versus RF values for nonreduced proteins (Fig. 2b).

c Estimated from semilog plot of molecular weight versus Rp value for reduced and alkylated proteins (Fig. 2a).

distribution of radioactivity among individual secretory proteins of pancreatic lobules continuously labeled with a mixture of 14C- amino-acids and discharged into physiological medium during three hours of 10-j M carbachol stimulation (17, 25). Under these conditions, radioactivity is a good estimate of relative protein mass.

(30), however, insoluble proteins have been studied necessitating the presence of urea in both gel dimensions.

DISCUSSION

Two-dimensional gel analysis for the separation of complex mixtures of proteins dates back to 1964. Early studies (26, 27) showed increased resolution in the separation of plasma proteins using two-dimensional gel electrophoresis. Two studies were published showing the potential usefulness of tube gel isoelectric focusing (pH 3 to 10) in the first dimension and slab gel electro- phoresis (pH 8.9) in the second dimension for the separation of plasma proteins (28) and potato proteins (29). In 1970, Kalt- Schmidt and Wittmann (30) published a two-dimensional gel technique (bidirectional electrophoresis at pH 8.6 in the presence of 6 M urea in the first dimension and electrophoresis at pH 4.6 in the presence of 6 M urea in the second dimension) which proved useful for the separation and further study of insoluble Escherichia coli ribosomal proteins. Since then, this technique and modifications thereof have been used by a number of in- vestigators for the study of complex mixtures of insoluble proteins found in prokaryotic ribosomes (31, 32), eukaryotic ribosomes (33-35), non-histone nuclear fractions (36, 37), and inner mito- chondrial membranes (38). Modifications have included use of isoelectric focusing in the first dimension to separate proteins by charge (37, 38) and use of Na dodecyl-SO4 gel electrophoresis in the second dimension to separate proteins by molecular weight (31-33, 36-38). In every case since Kaltschmidt and Wittmann

The procedure presented here was developed for separation of soluble proteins, such as t.hose found in pancreatic secretion. It consists of slab gel isoelectric focusing in the first dimension and Na dodecyl-SO4 gradient gel electrophoresis in the second dimen- sion. The absence of urea in the first dimension has two ad- vantages: (a) proteins are separated according to native charge, and (b) enzymatic ident.ification can be made of separated pro- teins. Insertion of gradient acrylamide (10 to 2Oc/,) in the second dimension allows for better resolution of separated proteins. The procedure has been shown to be effective in the complete separa- tion of all high molecular weight exocrine pancreatic proteins (larger than 10,000); sensitive to 0.6 pg of protein/spot (some proteins to 0.3 pg) ; rapid, requiring 24 hours, and to yield good recovery (- 80 “/c) of radiolabeled proteins.

No attempt has been made to include proteins smaller than 10,000 molecular weight in the two-dimensional gel analysis since separation of these proteins in Na dodecyl-SO4 gels is not propor- tional to log molecular weight and since the presence of ampho- lytes interferes with the staining of these proteins in the two- dimensional gel. Consequently, the secretory trgpsin inhibitor (39), colipase (40), and possibly other unidentified proteins less than 10,000 are lost in the two-dimensional analysis of secretory proteins. However, from recovery studies done on radiolabeled secretory proteins separated in Na dodecyl-SO4 gels (cf. “Re- sults”), proteins smaller than 10,000 molecular weight can be expected to account for no more than 10 to 12cj, of the secretory

protein mass. If recovery studies include only those proteins analyzed by the two-dimensional gel technique (proteins larger than lO,OOO), then recovery figures average 90yG.

-

Molecular weight

b - c

14,000 13,000

14,000 13,000

14,000 13,000

14,600 13,600

18,750 24,400

21,800 24,900

25,000 25,850

28,200 -

27,500 -

28,700 28,000

28,200 28,000

25,000 25,850

45,100 45,000

45,300 45,000

46,000 47,700

47,600 49,700

51,000 52,000

54,000 53,000

64,700 66,000

-

I

-

[soelectric pointd

6.4

7.8

8.7

8.8

8.7

>9.3

8.7

7.8

7.6

7.5

6.9

4.8

4.6

4.8

6.6

8.1

8.4

5.0

d Average of three determinations for each protein.

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5384

Differences in isoelectric points have been found for a number of secretory proteins (amylase, chymotrypsinogen 2, trypsinogen, ribonuclease, proelastases 1 and 2, and procarbosypeptidase 13) depending on whether these proteins arc loaded at the acidic or basic end of the isoelectric focusing gel. The isoelectric points of these proteins resulting from sample loading at the acidic end of the gel are the ones reported in this paper. The isoelectric points resulting from sample loading at the basic erId of the gel arc similar to the ones reported by Tartakoff ef al. (17) using 72.hour column isoelectric focusing. Studies reported in this paper indicate that modifications in the isoelectric points of these proteins are not due to the following possibilities: (a) oxidation of sulfhydryl groups with formation of intermolecular &sulfide bonds, (b) proteolytic degradation, (c) autoactivation of tryp- sinogen or activation of chymotrypsinogen 2, (d) formation of intermolecular aggregates at one isoelectric point and movement of aggregates to a second isoelectric point during the focusing procedure, or (e) chemical modification of proteins involving covalent bonds. It would appear from these studies that modifica- tions in isoelectric points which occur when secretory proteins are loaded at the basic end of the gel (pH 9.4 or higher) result from changes in configuration of protein molecules related to extreme pH. Changes in configuration would result in different charge expressions of these molecules and therefore different isoelectric points. Differences in charge expression might occur solely as a monomolecular change or might involve the interaction of molecules either of the same type (dimers, trimcrs, etc.) as is known to occur with bovine chymotrypsinogen above pH 8.0 (41), or different types (aggregates of amylase, chymotryp- sinogen, trypsinogen, and RNase). Since in the basic pattern each of these proteins has closely similar isoelectric points, about pH = 9.4, aggregates of the latter type may well exist. No further studies have been done to elucidate the precise chemical changes responsible for modification of charge expressions which occur when secretory proteins are exposed to pH 9.4 or higher. Since the focusing pattern resulting from sample loading at the acidic end is the same as that loaded near neutrality and since under physiological conditions these proteins are not exposed to pH 9.4 or higher, the isoelectric points reported here would seem to be those related to the native molecules.

The development of this technique using secretory protein from the guinea pig exocrine pancreas has allowed the majority of these proteins to be characterized by isoelectric point, apparent molecular weight, and proportionate mass (using radiolabeling techniques). Within the range of adequate separation of proteins by Na dodecyl-SO4 gel electrophoresis (proteins larger than 10,000 daltons) 19 protein spots have been found. Thirteen of these proteins have been identified by enzymatic or potential enzymatic activity, accounting for 96% of the high molecular weight protein mass. Multiple molecular forms have been de- tected for many of the zymogens and enzymes. Primarily on the basis of charge separation two forms have been found for chymo- trypsinogen, proelastase, and procarboxypeptidase A. On the basis of size separation two forms of amylase have been found. These forms are detected only when a polyacrylamide gradient is present in the Na dodecyl-SO4 gel. The small difference in molec- ular weight, estimated to be 1,000, may be due to differing amounts of attached carbohydrate. On the basis of both charge and size separation, two forms of lipase have been found. A comparison of amino acid analysis data from both molecules will be necessary to determine if the relatively large difference in apparent size (approximately 17,000) is due to amino acid or carbohydrate content or both. One form of procarboxypeptidase

A cofocuses (isoelectric point = 4.8) with chymotrypsinogen 1 and may exist in aggregate form similar to that seen between bovine procarboxypeptidase A and chymotrypsinogen C (42). No attempts were made to measure DNase or prophospholipase activity in this study. Two forms of prophospholipase, consisting of 123 and 130 amino acid residues, have been isolated from porcine pancreas (43). It is likely that spots 1 and 2 in the two- dimensional pattern of guinea pig pancreatic proteins are pro- phospholipases.

Since 1958 csocrine pancreatic proteins have been studied in detail in six mammalian species: ox, pig, human, dog, rat, and guinea pig (1-17). Using the two-dimensional gel analysis pre- sented here, the high molecular weight exocrine pancreatic proteins of the guinea pig have been described in greater detail than has been possible for those of the other species using column chromatographic and immunoelectrophoretic methods. The two- dimensional gel technique is effective in the complete separation of high molecular weight cxocrine proteins of the pancreas, which is not the case with the other fractionation methods. It is rapid, inexpensive, and relatively easy to perform. It should prove useful not only in the further study of cxocrine pancreatic proteins but in the study of other soluble proteins as well.

Acknowledgments-The author wishes to thank Dr. Giinter Rlobel for his advice, consultation, and support during these studies; Drs. Philip Siekevitz and Stanford Moore for their critical review of the manuscript; and Gertrude Willems for her translation of the references written in French.

REFERENCES

1. NORTHRUP, J. H., KUNITZ, M., .~ND HERRIOT, R. M. (1948) Crystallzne Enzymes, Columbia University Press, New York

2. FISHER, E. H., .~ND BERNFELD, P. (1948) Helv. Chim. Acta 31, 1831-1839

3. KELLER, P. J., COMEN, E., AND NEURBTH, II. (1958) J. Biol. Chem. 233,344-349

4. GREENE, L. J., HIRS, C. H. W., AND P~L.~DE, G. E. (1963) J. Biol. Chem. 238, 205442070

5. PUIGSERVER, A., AND DESNUELIX, P. (1971) Biochim. Biophys. Acta 236, 499-502

6. BECKER, It. R., H.ILBROOK, J. L., .\ND HIRS, C. H. W. (1973) J. Bicl. Chem. 246, 7826-7832

7. MARCNIS-MOUREN, G. (19F5) Bull. Sot. Chim. Biol. 47, 2207- 2217

8. UILIEL, J., AND AVRAMEAS, S. (1965) Biochemistry 4, 1740-1749 9. AVR~~ML~S, S., AND URIXL, J. (1965) Biochemistry 4, 175G1757

10. UR~M. M.. AND LAMY. F. (1969) Biochim. Biophys. Acta 194, 102-111 ’

11. DESNUELLI.:, P., GRITECOS, D., CH~YRLES, M., PEANASKY, R., BaR.4TTI. J.. AND ROVERY. ibf. (1970) in Structure Function Relation$hib in Proteolytk Enzymes (DESNUELLE, P., AND NEURATH, H., eds) pp. 21-33, Munksgaard, Copenhagen

12. PLUMMER. T. H.. JR.. BND S~RDA. L. (1973) J. Biol. Chem. 248, 7885-7869 ’ ’

13. KELLER, P. J., AND ALLAN, B. J. (1967) J. Biol. Chem. 242, 281-287

14. CLEMENTI:,, F., DE C~RO, A., AND FIGARELLO, C. (1972) Eur. J. Biochem. 31, 186-193

15. REB~UD, J. P., PAS~RO, L., AND DESNUELLE, P. (1964) Bio- them. Biophys. Res. Commun. 17, 347-351

16. BRADSHAW, W. S., AND RUTTER, W. J. (1972) Biochemistry 11, 1517-1528

17. TARTAKOFF. A., GREENE, L. J., AND PAL.~DE, G. E. (1974) J. Biol. Ciem: 249,742&7431

18. SCHEISLE, G. A., AND P~L~DE, G. E. (1975) J. Biol. Chem. 260, 266&2670

19. MORGAN, P. H., ROBINSON, N. C., WALSH, K. A., AND NEU- RBTH, H. (1972) Proc. Natl. Acad. Sci. U. S. A. 69,3312-3316

20. MARES-GUIA, M., AND SHAW, E. (1965) J. Biol. Chem. 240, 1579-1585

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5385

21. MAIZEL, J. V. (1971) Methods Viral. 6, 179-246 22. ANDERSON, L. E., AND MCCLURE, W. 0. (1973) Anal. Biochem.

61,173-i79 23. KAPLAN. A. (1970) Anal. Biochem. 33.218-225 24. KAPLAN, H., AND DUG~S, H. (1969) Biochem. Biophys. Res.

Commun. 34, 681-685 25. TARTAKOFF, A., JAMIESON, J. D., SCHEICLE, G. A., AND P,\L.+DE,

G. E. (1975) J. Biol. Chem. 260,2671-2677 26. RAYMOND, S. (1964) Ann. N. Y. Acad. Sci. 121,350-365 27. MARGOLIS, J., AND KENRICK, K. G. (1969) Nature 221, 1056-

1057 28. DALE, G., AND LATNICR, A. L. (1969) Clin. Chim. Acta 24,61-68 29. MACKO, V., AND STEGEMANN, H. (1969) 2. Physiol. Chem. 360,

917-919 30. KALTSCHMIDT, E., AND WITTMANN, H. G. (1970) Anal. Bio-

them. 36, 401-412 31. GEISSER, M., TISCHIZNDORF, G. W., ST~FFLER, G., AND WITT-

MANN, H. G. (1973) Mol. Gen. Genet. 127,111-128 32. SUBR~M:INI~N, A. It. (1974) Eur. J. Biochem. 46,541-546 33. MARTINI, 0. H. W., AND GOULD, I-I. J. (1971) J. Mol. Biol.

62, 403-405

34. DELAUNAY, J., MATHIEU, C., AND SCHAPIRA, G. (1972) Eur. J. Biochem. 31, 561-564

35. SHERTON, C. C., AND WOOL, I. G. (1972) J. Biol. Chem. 247, 4460-4467

36. ORRICK, L. It., OLSON, M. 0. J., AND BUSCH, H. (1973) l’roc. Natl.‘Acad. kci. U. k’. A. 70, i316-1320

37. MACGILLIVRAY. A. J.. .ZND RICKWOOD. D. (1974) Eur. J. Bio- them. 41, 181L190 ’

, ~ ,

38. CHIGNELL, D. A., AND WINGFIELD, P. T. (1974) Fed. Proc. 33, 1283 abstr.

39. Br\RTl’:LT. D. C.. AND GREENE. L. J. (1971) J. Biol. Chem. 246, 2218-2i29

40. M.\YLI~~. M. F.. CHARLES. M.. GXHE. C.. AND DESNUELLI:.. P. (1971)’ Biochim. Bzophy’s. Acta 229,‘288-289

41. OSBORNE, J., AND STXINER, R. F. (1972) Arch. Biochem. Bio- phys. 162, 849-855

42. BROM~N, J. It., Cox, D. J., GREENSHIELDS, 11. N., W:ILSH, K. A., Y.IMAS~KI, M., .IND NIXJR.ITH, H. (1961) Proc. Null. Acad. Sci. U. S. A. 47, 1554-1560

43. NIEUI~ENHUIZISN, W., STISENBERGH, P., .ZND DEHUS, G. H. (1973) Eur. J. Biochem. 40, l-7

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from

G A Scheeleexocrine pancreatic proteins.

Two-dimensional gel analysis of soluble proteins. Charaterization of guinea pig

1975, 250:5375-5385.J. Biol. Chem. 

  http://www.jbc.org/content/250/14/5375Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/250/14/5375.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on August 18, 2020

http://ww

w.jbc.org/

Dow

nloaded from