ANALYTICAL BIOCHEMISTRY 135, 128- 133 ( 1983)

An Electrophoretic Method for the Identification of Poly-y-glutamyl Chain Lengths in Folates and Related Compounds

JORGE ZORZOPULOS, BENJAMIN RAJ, AND LLOYD M. KOZLOFF

Department of Microbiology, University of California, San Francisco, San Francisco, California 94143

Received June 13, 1983

An electrophoretic method for the identification and separation of folyl polyglutamates of different chain lengths and of the corresponding p-aminobenzoyl polyglutamate compounds has been developed. These compounds have been separated using electrophoresis in a 40% poly- acrylamide gel using a higher voltage and other modifications of the standard polyacrylamide gel electrophoresis procedures used to separate larger polypeptides. Good separation has been obtained on folates containing up to 12 ghttamyl residues. Further, this method has been used to investigate the nature of the products formed by the y-glutamyl carboxypeptidases from hog kidney and bovine liver.

KEY WORDS: folyl polyglutamates; paminobenzoyl polyglutamates; y-glutamyl carboxypep tidases; exopeptidases; polyghrtamates; folates.

The multiplicity of folate structures found in cells has been estimated by Baugh et al. (1) to exceed 140 members considering both sub- stituents on the pteridine ring and the variable addition of more glutamate residues. Recent discoveries (2) of folates containing from 12 to 14 glutamate residues in T4 bacteriophage- infected Esherichia coli, as well as renewed interest in the formation, breakdown, and function of folyl and related polyglutamates, has focused attention on the analytical iden- tification of these compounds. The variable substituents on the pteridine groups have made simple analytical methods such as Se- phadex or DEAE chromatography difficult to interpret. The current preferred method for polyglutamate analysis has involved the re- moval of the pteridine group by strong oxi- dation and/or reduction followed by the anal- ysis of the paminobenzoyl polyglutamate compounds on either DEAE (3) or by HPLC (4).

This paper reports initial results in adapting polyacrylamide gel electrophoresis for rapidly identifying and separating undegraded folyl polyglutamates as well as paminobenzoyl polyglutamates. The effect of various pteridine

substitutions on mobilities in this system has not been investigated. This analytical procs dure seems especially useful in rapidly char- acterizing various natural occurring y-gluta- my1 carboxypeptidases as either an exo- or endopeptidase.

MATERIALS AND METHODS

Gel electrophoresis. The slab gel electro- phoresis was carried out by a modification of the method described by Laemmli (5). The stock solution of acrylamide contained 60% acrylamide and 1.8% bisacrylamide. Twenty milliliters of this stock solution was mixed with 3.75 ml of 3 M Tris-HC1, pH 8.55, 6.28 ml water, 0.015 ml TEMED’ and 0.15 ml 10% ammonium persulfate. This solution was poured into the apparatus with the slot-former in place. No stacking gel was used to avoid lateral diffusion of the small molecules that were being analyzed. After the gel was poly- merized the slot-former was removed and the wells were washed several times by flushing

’ Abbreviations used: TEMED, N,N,N’,W-tetramethyl- ethylenediamine; SDS, sodium dodecyl sulfate; BSA, bo- vine serum albumin; pte, pteridine; glu, glutamate.

0003-2697/83 $3.00 Copyright 0 1983 by Academic Press. Inc. All rights of reproduction in any form reserved.

128

ELECTROPHORETIC ANALYSIS OF FOLYL POLYGLUTAMATES 129

them with water using a small pipet. The water was removed from the wells and 0.03- to 0.06- ml samples were placed in the wells. The sam- ples were prepared by boiling for 3-5 min a mixture of 0.05 ml of the solution being an- alyzed with 0.01 ml of a buffer containing 0.125 M Tris-HCl, pH 7.2,4% SDS, 40% glyc- erol, and 2 X lo-‘% bromphenol blue. After the samples were placed in the wells, running buffer was carefully layered on top using a syringe with plastic tubing at the tip of the needle. The running buffer at about pH 8.8 contained 1.5 g Tris base and 7.2 g glycine per liter. The gels were run at a constant volt- age of 300 V. The run was stoppsd when the bromphenol blue had run only l/3 of the total length of the gel. To avoid overheating, cooling water was circulated through the apparatus. Normally, good separation was obtained in 2 h, the migration was linear over this time, and standards in different lanes ran identically. Occasionally, to separate a large number of components, the run was carried out for 3.5 h. Under these circumstances, the migration was still linear within one lane, but there was some “arcing” between several lanes (see Fig. 4). It should be noted that the standard gel electrophoresis procedures for separating pro- teins uses a much lower gel concentration and the voltage rarely is raised above 100 V for a 4- to 5-h run or 30 V for an overnight run. After a run with polyglutamates, the gels were dried on a sheet of filter paper and heated at 80°C under vacuum. When radioactive com- pounds were used, the dried gels were exposed to Kodak X-omat AR film. Folic acid com- pounds ranging in amounts from 0.2 to 0.5 nmol could also be visualized on the dried gels by illuminating them with a uv lamp (254 nm). For uv detection, very low concentrations of bromphenol blue were used to avoid in- terference in detecting the folate compounds.

Enzyme preparation and enzymatic assays. Hog kidney y-glutamyl carboxypeptidase was partially purified from a dried hog kidney ex- tract (DIFCO) by ammonium sulfate precip- itation and Sephadex chromatography (6). A crude extract of bovine liver was obtained as

described by Silink et al. (7). Typical car- boxypeptidase assays were performed mixing 10 ~1 of labeled substrate (3000 cpm, sp act 2.3 X lo6 cpm/pmol), 5 ~1 of 1 mM glutamic acid, 30 ~1 of 0.2 M acetate buffer, pH 4.5 (unless otherwise stated), and 30 ~1 of the en- zyme preparation (8). The incubation was carried out for 2 h at 35 “C. Controls in which the enzyme preparation was replaced with a BSA solution containing the same amount of protein were always run. The reaction was stopped by boiling the samples for 5 min. Then the samples were centrifuged for 5 min in a Beckman microfuge. An aliquot of the su- pematant solution was analyzed by gel elec- trophoresis as indicated above.

Chemicals. The synthetic folyl polygluta- mate compounds used in standards in this work were furnished by C. M. Baugh (pteglur , pteglu 22, and pteglu’) and C. L. Krumdeick (pteglulZ). These compounds contained 14C in their C-terminal glutamate residue. The con- ditions for the extraction and cleavage of the folyl polyglutamates to p-aminobenzoyl poly- glutamates from Tcinfected E. coli B (labeled with 14C in their p-aminobenzoyl residue) and their chromatographic analysis were as de- scribed before (2). The other chemicals used were of analytical grade.

RESULTS

Separation of Folyl Polyglutamates

Figures IA and B show the separation of 14C-labeled folyl polyglutamates of different polyglutamate-chain length in two separate runs. Figure 2 shows the relation between the number of glutamate residues of the different folyl polyglutamates and the distance from the top of the gel. A constant voltage of 300 V was used and the run was completed in about 3 h. It can be seen in Fig. 1 that radio- active-degradation products were present in the folyl polyglutamate solutions analyzed. This was especially true for the pteglu’ while the pteglu3, pteglulz, and pteglu, (the newest preparation) had lesser amounts of breakdown products. Some of these preparations had been

130 ZORZOPULOS, RAJ, AND KOZLOFF

12-

:1

glu-

FIG. 1. Radioautography after electrophoresis of “C-labeled folyl polyglutamates with different polyglu- tamabchain lengths, as well as free glutamic acid. The electrophoresis was carried out in 40% polyacrylamide gels as described under Materials and Methods. The number on the left of each gel indicates the number of glutamic acid residues present in the polyglutamate portion of the molecule. The unnumbered spots indicate radioactive contaminants in the preparations.

prepared some years ago and stored in the dark at -20°C but folate compounds are well known for their instability. Since these com- pounds were labeled in the terminal glutamic acid residue, the degradation products most probably are a variety of polyglutamates. Free glutamic acid moved faster and was well sep-

I I I

4.0 4.5 5.0 5.5 6.0 7.0

Distance (cm)

FIG. 2. Relationship between the number of glutamic acid residues present in the polyglutamate portion of the different folyl polyglutamates and the distance migrated from the top of the gel. Electrophoresis was carried out as described under Materials and Methods. Free glutamic acid, which is also shown, runs with the electrophoretic front.

arated from all of the folyl polyglutamates an- alyzed (Fig. 1).

Separation of Products of Carboxypeptidase Action

Figure 3 shows the reaction products pro- duced by two y-glutamyl carboxypeptidases, one from hog kidney (Fig. 3A) and one from bovine liver (Fig. 3B), acting on either pteglu3, pteglu,, or pteglulz. The incubation was car- ried out at the optimum pH, and the main, if not the sole, product of the reaction in all cases migrated together with free glutamic acid. Different products were observed when the labeled pteglu was incubated with the crude hog kidney preparation for 2 h at pH’s between 4.0 and 8.0 (Fig. 4). In this experi- ment, during the electrophoresis, there was some upward curving of migration of the compounds in the gel in the center lanes due to the high voltage over the 3.5 hours in the run. But it is apparent that a number of labeled glutamyl compounds with electrophoretic mobilities intermediate between the substrate pteglu12 and free glutamic acid were formed.

ELECTROPHORETIC ANALYSIS OF FOLYL POLYGLUTAMATES

(A) (B)

a b c a b c

FIG. 3. Reaction products of the action of hog kidney y-glutamyl carboxypeptidase (A) or bovine hepatic y-glutamyl carboxypeptidase on three different folyl polyglutamates containing 3, 7, or 12 glutamic acid residues, respectively. (a) Shows the substrates incubated in the absence of enzyme; (b) the “‘C-labeled products of the carboxypeptidase action after 2 h of incubation under optimal conditions, and (c) the position of free i4C-labeled glutamic acid.

12-

glu-

FIG. 4. Reaction products produced by the action of peparation of hog kidney y-glutamyl carboxypeptidase on pteglu containing 14C in terminal glutamic acid residue. In this electrophoretic run, which went for 3.5 h at 300 V, there was some “arcing” in the migration of the free glutamic acid. Lane (a) shows the migration of the original pteglu and of free glutamic; (b) the reaction products after incubation at pH 7.0; (c) at pH 6.0; (d) at pH 5.0; (e) at pH 5.0; and lane (f) at pH 4.0.

131

Up to 11 separate labeled bands are visible in lane (b), pH 7.0, and lane (c), pH 6.0. The multitude of products suggests that there were two different carboxypeptidases in the partially purified hog kidney preparation able to cleave the different y-glutamyl bands. The major carboxypeptidase functioned as an exopepti- dase (pH 4-5.5) to liberate free glutamate while the other carboxypeptidase functioned as a random endopeptidase (at pH 5.7-7.0) causing the release of all the possible labeled polyglutamyl compounds such as glu-y-glu and glu-y-glu-y-glu.

Separation of p-Aminobenzoyl Polyglutamates

Figure 5A shows the analysis of [ 14Clp-ami- nobenzoyl polyglutamates obtained from E. coli B infected with a T4D bacteriophage am- ber mutant (26-) separated by the DEAE chromatographic procedure (2). Samples from some of the peak fractions obtained from the DEAE column run were subsequently ana- lyzed by the acrylamide electrophoretic method and the results are shown in Fig. 5B.

DISCUSSION

The use of stiff polyacylamide slab gels run at high voltage offers considerable advantages in rapidly separating polyglutamate com-

132 ZORZOPULOS, RAJ, AND KOZLOFF

20 40 60 80 100 120 140 160 180 200

Fraction Number

7-

6-

5-

4-

3-

FIG. 5. (A) DEAE-cellulose chromatcgraphic analysis of “C-labeled paminobenzoyl polyglutamates obtained from bacteriophage T4D (26- amber mutant)-infected E. coli B (2). (B) Gel electrophoretic analysis of the same compounds separated by the DEAE-cellulose chromatographic procedure in A. Samples from the peak fraction in A were run for 2 h at 300 V.

pound ranging in molecular size from about tidases on folyl polyglutamates, or (iii) prod- 2000 Da (pteglu12) to about 260-280 Da (pa- ucts of folyl polyglutamate synthetase. In baglu, or even glu-y-glu). Furthermore, only studying both the carboxypeptidase- or syn- very small samples are needed. As yet, we thetase-reaction products, there is no variation have not studied the mobility of compounds in substituents on the pteridine ring and gel containing various substituents on the pteri- electrophoretic analysis can be readily inter- dine ring but it is apparent that the ability to preted. The basis for separation of these com- rapidly analyze multiple samples will be of pounds by gel electrophoresis is not certain considerable use in looking at (i) undegraded and may involve both the charge and gel fil- or degraded folates extracted from cells, (ii) tration due to the radius of rotation of these products of action of y-glutamyl carboxypep molecules. The successful gel electrophoretic

ELECTROPHORETIC ANALYSIS OF FOLYL POLYGLUTAMATES 133

separation of these polyglutamate compounds suggests that the analysis of other classes of small molecules may similarly be possible.

ACKNOWLEDGMENT

This work was supported by Public Health Research Grant Number AI 28370, from the National Institute of Allergy and Infectious Diseases.

REFERENCES

I. Baugh, C. M., Braverman, E., and Nair, M. G. (1974) Biochemistry 13, 4952-496 1.

2. Nakamura, K., and Kozloff, L. M. (1978) Biochim. Biophys. Acta. 540, 313-319.

3. Baugh, C. M., May, L., Braverman, E., and Nair, M. G. (1979) in Chemistry and Biology of Pteri- dines (Kisliuk, R. L. and Brown, G. M., eds.), pp. 2 19-224, Elsevier/North-Holland, Amsterdam/ New York.

4. Shane, B., Bognor, A. L., Goldfarb, R. D., and LeBowitz, J. H. (1983) J. Bacterial. X3,316-325.

5. Laemmli, U. K. (1970) Nature (London) 227, 680- 685.

6. Kozloff, L. M., and Lute, M. ( 1965) J. Mol. Biol. 12, 780-792.

7. Silink, M., Reddel, R., Bethel, M., and Rowe, P. B. (1975) J. Biol. Chem. 250, 5982-5994.

8. Kozloff, L. M., and Lute, M. (198 1) J. Viral. 40,645- 656.