Immunochemical Specificity of Lactate Dehydrogenase-XLactate DehydrogenaseX 351 a 345 89 1011 FIG....

4
Proceedings of the National Academy of Sciences Vol. 68, No. 2, pp. 349-352, February 1971 Immunochemical Specificity of Lactate Dehydrogenase-X ERWIN GOLDBERG Department of Biological Sciences, Northwestern University, Evanston, Illinois 60201 Communicated by Clement L. Markert, November 17, 1970 ABSTRACT Sperm-producing testes of mammals and some birds contain a unique form of lactate dehydrogenase (EC 1.1.1.27) which is a tetramer composed of C subunits, rather than the A and B subunits of lactate dehydrogenase isozymes 1-5. Immunochemical evidence is presented which confirms the hypothesis that the C subunit is encoded in a separate gene from the A and B polypeptides. The evolutionary relationships of these genes are discussed. Studies by many investigators have clearly established that the five major types of lactate dehydrogenase (LDH; EC 1.1.1.27) in mammals are formed by the random combination of two different subunits into tetramers. Each subunit (A and B) is the product of a distinct gene locus (1). There is now substantial evidence that the isozyme LDH-X of the sperm cell is also a tetramer and that it is composed of a third poly- peptide type (C), which is encoded in a separate gene (2-5). Apparently this gene is active only during the primary sperinatocyte stage of the spermatogenic process and is in- active in all other cells of the organism (6, 7). Immunological procedures have proved to be potent tools for establishing subunit homologies (8) and for examining the evolution of lactate dehydrogenases (9). The selective in- hibiting and precipitating activities of antibodies have been used in the present investigation to demonstrate that the C subunit of LDH-X is indeed distinct from the A and B poly- peptide, thus confirming its independent genetic control, and further, to show that this isozyme from several vertebrate species has a common evolutionary origin. METHODS LDH-X was purified from mouse testes by a combination of ammonium sulfate fractionation and DEAE-Sephadex chromatography. The most active fractions from the column were combined and concentrated to contain approximately 760 international units of LDH-X activity per ml. This unit is defined as that amount of enzyme which will oxidize 1 Mmol of NADH2/min at 250C. While this material contained only the X isozyme of LDH, other contaminating proteins with different electrophoretic mobility were present. There- fore, as a final step in this purification a 2-ml sample, con- taining approximately 20 mg of protein, was subjected to electrophoresis on a preparative polyacrylamide slab-gel in the Ortec pulsed constant power system. The LDH-X band was identified by nitroblue tetrazolium staining as previously described (10) sliced from the gel, and homogenized in 0.05 M phosphate buffer at pH 7.4 that contained 0.85% NaCl. The gel slurry was shown to be homogeneous by disc gel electrophoresis with the protein band, stained by amido black, and the LDH-X band having coincident electrophoretic mobility. 71% of the enzyme activity applied to the gel was recovered. Approximately 4500 units of LDH-X was injected into three albino female rabbits. Each rabbit received three in- jections, of 1-ml each, into a footpad and the dorsal thoracic muscles. 2 weeks later, each doe was given a single injection containing 600 units of LDH-X. As Weintraub and Raymond (11) have shown, the acrylamide serves as an adjuvant in this procedure. Blood was withdrawn from the heart at weekly intervals, commencing on the 15th day, for a period of 2 months. In subsequent work, crystalline LDH-X in Freund's adjuvant was used as the provoking antigen for antiserum production. Antibody specificity and titer was established by the Ouchterlony diffusion method applied as recommended by Stollar and Levine (12), by the microcomplement fixation method of Wasserman and Levine (13), and by measuring inhibition and precipitation of enzyme activity. From the latter measurements, equivalence points for the reaction be- tween antibody and antiserum were calculated (14). Finally the procedure combining selective precipitation by antibodies followed by electrophoretic resolution of the unprecipitated isozyme was as described by Markert and Holmes (14). This method provides a sensitive measure of antibody specificity in a mixture of isozymes. Suitable controls with normal rabbit serum were performed. RESULTS Immunodiffusion experiments were performed in an agar gel containing a center well for antiserum and four test wells. A single, intense precipitin line was formed against purified LDH-X. This merged with the precipitin line with crude testis extract in the adjacent well. The fact that no spur was formed indicates a "reaction of identity" (12). However, a second, faint precipitin line was observed with the crude extract. Only the intense precipitin lines stained for LDH when the nitroblue tetrazolium reaction mixture was added to the plates. Lack of staining of the faint band reflected the presence of antibodies to impurities in the immunizing anti- gen. No precipitin reaction could be detected with homo- geneous preparations of mouse LDH-1 and LDH-5 in the other wells on the agar plate. Double diffusion analyses of antisera produced by rabbits injected with crystalline LDH-X indicated a high probability of immunochemical homogeneity, since a single precipitin line of identity was obtained from LDH-X and from crude extracts of testis. In no case was a reaction observed with homogeneous preparations of mouse LDH-1 and LDH-5 at several concentrations of antiserum and of enzyme. 349 Abbreviation: LDH, lactate dehydrogenase. Downloaded by guest on August 27, 2021

Transcript of Immunochemical Specificity of Lactate Dehydrogenase-XLactate DehydrogenaseX 351 a 345 89 1011 FIG....

Page 1: Immunochemical Specificity of Lactate Dehydrogenase-XLactate DehydrogenaseX 351 a 345 89 1011 FIG. 3. Disc gels showing the effect of antiserum to LDH-Xonisozymes of various mammalian

Proceedings of the National Academy of SciencesVol. 68, No. 2, pp. 349-352, February 1971

Immunochemical Specificity of Lactate Dehydrogenase-X

ERWIN GOLDBERG

Department of Biological Sciences, Northwestern University, Evanston, Illinois 60201

Communicated by Clement L. Markert, November 17, 1970

ABSTRACT Sperm-producing testes of mammals andsome birds contain a unique form of lactate dehydrogenase(EC 1.1.1.27) which is a tetramer composed of C subunits,rather than the A and B subunits of lactate dehydrogenaseisozymes 1-5. Immunochemical evidence is presentedwhich confirms the hypothesis that the C subunit isencoded in a separate gene from the A and B polypeptides.The evolutionary relationships of these genes are discussed.

Studies by many investigators have clearly established thatthe five major types of lactate dehydrogenase (LDH; EC1.1.1.27) in mammals are formed by the random combinationof two different subunits into tetramers. Each subunit (Aand B) is the product of a distinct gene locus (1). There is nowsubstantial evidence that the isozyme LDH-X of the spermcell is also a tetramer and that it is composed of a third poly-peptide type (C), which is encoded in a separate gene (2-5).Apparently this gene is active only during the primarysperinatocyte stage of the spermatogenic process and is in-active in all other cells of the organism (6, 7).

Immunological procedures have proved to be potent toolsfor establishing subunit homologies (8) and for examining theevolution of lactate dehydrogenases (9). The selective in-hibiting and precipitating activities of antibodies have beenused in the present investigation to demonstrate that the Csubunit of LDH-X is indeed distinct from the A and B poly-peptide, thus confirming its independent genetic control, andfurther, to show that this isozyme from several vertebratespecies has a common evolutionary origin.

METHODSLDH-X was purified from mouse testes by a combination ofammonium sulfate fractionation and DEAE-Sephadexchromatography. The most active fractions from the columnwere combined and concentrated to contain approximately760 international units of LDH-X activity per ml. This unitis defined as that amount of enzyme which will oxidize 1Mmol of NADH2/min at 250C. While this material containedonly the X isozyme of LDH, other contaminating proteinswith different electrophoretic mobility were present. There-fore, as a final step in this purification a 2-ml sample, con-taining approximately 20 mg of protein, was subjected toelectrophoresis on a preparative polyacrylamide slab-gel inthe Ortec pulsed constant power system. The LDH-X bandwas identified by nitroblue tetrazolium staining as previouslydescribed (10) sliced from the gel, and homogenized in 0.05M phosphate buffer at pH 7.4 that contained 0.85% NaCl.The gel slurry was shown to be homogeneous by disc gelelectrophoresis with the protein band, stained by amido black,

and the LDH-X band having coincident electrophoreticmobility. 71% of the enzyme activity applied to the gel wasrecovered.Approximately 4500 units of LDH-X was injected into

three albino female rabbits. Each rabbit received three in-jections, of 1-ml each, into a footpad and the dorsal thoracicmuscles. 2 weeks later, each doe was given a single injectioncontaining 600 units of LDH-X. As Weintraub and Raymond(11) have shown, the acrylamide serves as an adjuvant in thisprocedure. Blood was withdrawn from the heart at weeklyintervals, commencing on the 15th day, for a period of 2months. In subsequent work, crystalline LDH-X in Freund'sadjuvant was used as the provoking antigen for antiserumproduction. Antibody specificity and titer was established bythe Ouchterlony diffusion method applied as recommendedby Stollar and Levine (12), by the microcomplement fixationmethod of Wasserman and Levine (13), and by measuringinhibition and precipitation of enzyme activity. From thelatter measurements, equivalence points for the reaction be-tween antibody and antiserum were calculated (14). Finallythe procedure combining selective precipitation by antibodiesfollowed by electrophoretic resolution of the unprecipitatedisozyme was as described by Markert and Holmes (14). Thismethod provides a sensitive measure of antibody specificityin a mixture of isozymes. Suitable controls with normal rabbitserum were performed.

RESULTSImmunodiffusion experiments were performed in an agar gelcontaining a center well for antiserum and four test wells. Asingle, intense precipitin line was formed against purifiedLDH-X. This merged with the precipitin line with crudetestis extract in the adjacent well. The fact that no spur wasformed indicates a "reaction of identity" (12). However, asecond, faint precipitin line was observed with the crudeextract. Only the intense precipitin lines stained for LDHwhen the nitroblue tetrazolium reaction mixture was added tothe plates. Lack of staining of the faint band reflected thepresence of antibodies to impurities in the immunizing anti-gen. No precipitin reaction could be detected with homo-geneous preparations of mouse LDH-1 and LDH-5 in theother wells on the agar plate. Double diffusion analyses ofantisera produced by rabbits injected with crystalline LDH-Xindicated a high probability of immunochemical homogeneity,since a single precipitin line of identity was obtained fromLDH-X and from crude extracts of testis. In no case was areaction observed with homogeneous preparations of mouseLDH-1 and LDH-5 at several concentrations of antiserumand of enzyme.

349

Abbreviation: LDH, lactate dehydrogenase.

Dow

nloa

ded

by g

uest

on

Aug

ust 2

7, 2

021

Page 2: Immunochemical Specificity of Lactate Dehydrogenase-XLactate DehydrogenaseX 351 a 345 89 1011 FIG. 3. Disc gels showing the effect of antiserum to LDH-Xonisozymes of various mammalian

350 Genetics: E. Goldberg

100

0JI-0z0

z

50 W

0

4 EQ. PT. (177 lU. LDH-X/ML)

t EQ. PT. (170 lU. LDH-X/ML)

0

A AA A A A Av A

5 10 15 20 1000 2000pi ANTISERA/ml ENZYME (2.5 1.U./ml)

FIG. 1. Inactivation of LDH-X by homologous antiseraagainst LDH-X; O---O, inhibition of LDH-X by antisera;*- , precipitation of LDH-X by antisera; A A, attemptedinactivation of nonhomologous LDH (LDH-1 and LDH-5).

Similarly, there is no cross reaction by anti-LDH-X serumwith LDH-1 and LDH-5 in enzyme inhibition measurements.These experiments measured both the immediate enzyme-inhibiting capacity of the antiserum and the residual activityafter maximal precipitation of enzyme by antibodies. In-creasing quantities of antisera to LDH-X were mixed with afixed amount of homologous or heterologous enzyme activity(2.5 units of enzyme activity in the reaction mixture) for 15min and the residual enzyme activity was determined spec-trophotometrically. Then the antiserum-isozyme mixtureswere incubated at 250C for 1 hr and at 4VC overnight, centri-fuged at 100,000 X g for 30 min, and the supernatants weretested for LDH activity. From these data it appears that theantisera used contain primarily enzyme-inactivating anti-bodies. The equivalence points (Fig. 1) are virtually the samebefore and after removal of the enzyme-antibody precipitate.There are apparently no protective antibodies produced (14)since no enzyme activity remains after maximum treatmentwith antisera. No cross-reaction with A or B subunits couldbe detected even at a 100-fold increase in antiserum concen-tration. The equivalence points calculated (Fig. 1) are about30-times higher than those obtained with antisera producedfrom animals receiving LDH-X in acrylamide. Antiserumto the pure, crystalline LDH-X was used in all subsequentwork.

Antiserum was mixed with extracts of heart tissue and oftestes and then these preparations were subjected to elec-trophoresis on polyacrylamide slab gels or disc gels. The dataobtained are a striking demonstration of the specificity ofthe antiserum for LDH-X (Fig. 2), which is selectively re-moved from the testis homogenate while the other bands ofLDH are unaffected. That no cross reaction of antiserumwith A or B subunits occurs is even more evident from theheart extract pattern. Mouse heart muscle contains the usualfive somatic-tissue LDH isozymes and, typically, most of theenzyme activity is associated with the more anodal forms(i.e., LDH-1, 2, and 3). There is neither a qualitative norquantitative change in the isozyme pattern in the presenceof antiserum to LDH-X (Fig. 2). Furthermore, these datastrongly support the conclusion that the A4 homotetramer,which has an electrophoretic mobility coincident with LDH-X, is absent from mature testes and, more importantly, that

B-C subunit combinations do not occur in vivo (15, 16).When the BC3 and B2C2 tetramers, produced by in vitro dis-sociation-recombination of subunits (5) according to theprocedure of Markert (17), were treated with antiserum, theywere also selectively removed.

This procedure of antiserum treatment and electrophoresisis extremely useful in demonstrating homologies among LDHloci (8, 14, 18). The antiserum was tested in this manneragainst preparations of human spermatozoa and of testesfrom rat, rabbit, hamster, guinea pig (Fig. 3), and pigeon(Fig. 4). LDH-X from rats and humans migrates on poly-acrylamide disc gels between LDH-3 and 4, and LDH fromthe rabbit migrates between LDH4 and 5. Clearly, the anti-serum specifically and selectively removes LDH-X. Inhamster tissues, as in the mouse, LDH-X and LDH-5 possessthe same net charge and show coincident electrophoreticmobility. There is obviously only the C4 homotetrameroccupying this zone on the gel after electrophoresis of ham-ster-testis extract.Guinea pig testes contain three X-bands rather than two

as previously reported (19, 20). Two of these bands are readilydiscernible on the basis of electrophoretic mobility andmigrate between LDH-3 and 4. The third is cathodal to,and barely separated from, LDH4 on disc gels, and thuscould be easily overlooked except when the gel containing it iscompared with an antiserum-treated preparation (Fig. 3).While the existence of a third gene locus coding for t',e Cpolypeptide accounts for a single "band X" in testes, it ismore difficult to explain the occurrence of multiple forms ofLDH-X. Zinkham et al. (6) observed that the C locus waspolymorphic in some pigeons so that both C and C' allelesdirected the synthesis of five forms of LDH-X. Both thenumber of LDH bands and their relative mobilities wouldseem to support an argument against the formation of

4.

1 2 34 56 78 9 10FIG. 2. Immunochemical specificity of mouse LDH isozymes.

Abbreviations used in this legend: AS, rabbit antiserum to LDH-X; CS, rabbit control serum; T, testes extract; H, heart extract.The tissue extracts and serum were mixed in ratios listed belowand treated as described in the text. Channel 1, T: AS = 1:1;channel 2, T: AS = 1:2; channel 3 and 5, T control, LD)H-Xdesignated by closed circle; channel 4, AS; channel 6, T: CS =1: 1; channel 7, T: CS = 1:2; channel 8, CS; channel 9, H con-trol; channel 10, H: AS = 1:4.

Proc. Nat. Acad. Sci. USA

Dow

nloa

ded

by g

uest

on

Aug

ust 2

7, 2

021

Page 3: Immunochemical Specificity of Lactate Dehydrogenase-XLactate DehydrogenaseX 351 a 345 89 1011 FIG. 3. Disc gels showing the effect of antiserum to LDH-Xonisozymes of various mammalian

Lactate Dehydrogenase X 351

a

34 5 8 9 10 11FIG. 3. Disc gels showing the effect of antiserum to LDH-X on isozymes of various mammalian testes extracts. The tissue extracts

and antisera were mixed 1:1 in each case and treated as described in the text. Closed circles designate LDH-X, open circles mark theLDH isozymes in rabbit serum. Gels numbered 2, 5, 7, 9, and 11 show the isozyme pattern after antiserum treatment of the appropriateextract. Gel 3 is a diaphragm extract. All of the extracts were adjusted to contain the same total LDH activity prior to electrophoresis.

1, 2, human; 3-5, hamster; 6, 7, guinea pig; 8, 9, rabbit; 10, 11, rat.

C-C' as well as B-C and A-C heterotetramers in the guineapig testes. As an alternative, in animals with more than one

"X band," there may be an epigenetic modification of the Csubunit which alters the net charge of the tetramer, perhapsby combination with a small molecule, analogous, for ex-

ample, to the association of alkaline phosphatase isozymeswith varying amounts of sialic acid (21).The pigeon used in these experiments was apparently

heterozygous at the C locus (6) since five X bands were re-

moved by antiserum (Fig. 4). Under these conditions, Aand B subunit-containing tetramers were not affected. Con-firmation of this result was obtained by incubating a gel in a

reaction mixture in which 0.5 M a-hydroxyvalerate was sub-stituted for lactate so that only the five forms of LDH-Xshowed enzymatic activity (6).The antibody-antigen reactions were quantitatively

measured by micro-complement fixation analyses (13).Fig. 5 shows that mouse LDH-X reacted with antiserum ata concentration of 1:7500. The degrees of cross-reaction withhomogeneous preparations of LDH-1 and LDH-5 were ex-

pressed quantitatively as the index of dissimilarity or im-munological distance (ID); that is, the relative concentrationof antiserum required to produce a complement fixation curve

whose peak was as high as that given by the homologousLDH (22). Values of 75 were obtained with both LDH-1 andLDH-5. Thus, there would appear to be little structuralsimilarity between mouse LDH-X as compared to eitherLDH-1 or,5.

DISCUSSION

The evidence presented here clearly establishes that LDH-Xis immunologically distinct from LDH-1 and LDH-5, and

confirms the concept that the C subunit is encoded in a sepa-rate gene in mammals as well as in pigeons. These resultsrule out the suggestion by Stambaugh and Buckley (23),based on an assumed octameric structure for LDH, that theC subunit is in fact a hybrid subunit involving the same

+

1 2 3 4 5 6 7 8FIG. 4. Pigeon LDH isozyme patterns. Abbreviations used

are the same as in Fig. 2. Channel 1, AS; channel 2, T alone;channel 3, T:AS = 1:2; channel 4, T:AS = 1:4; channel 5,H alone; channel 6, H:AS = 1:2; channel 7, H:AS = 1:4;channel 8, H: AS = 1:8. There are five bands of LDH-X inthe testis preparation, all of which appear to be equally affectedby AS.

12

Vol. 68, 1971

i .:

W t: :. .;. "

Dow

nloa

ded

by g

uest

on

Aug

ust 2

7, 2

021

Page 4: Immunochemical Specificity of Lactate Dehydrogenase-XLactate DehydrogenaseX 351 a 345 89 1011 FIG. 3. Disc gels showing the effect of antiserum to LDH-Xonisozymes of various mammalian

352 Genetics: E. Goldberg

awx

U-zww-J

0U

50F

0

0.16 0.32 0.64 1.28 2.56

ENZYME UNITS ADDED

FIG. 5. Micro-complement fixation. Antiserum at a dilutionof 1:7500 was tested against a crystalline preparation of LDH-X,and homogeneous preparations of mouse LDH-1 and LDH-5.A curve coincident with that of LDH-X was obtained with crudetestis extract, thus confirming the purity of the antigen andspecificity of antiserum.

genes as the A and B polypeptides. This would require thatthe antiserum to LDH-X cross-react with isozymes containingB subunits.The data obtained allow the conclusion that the C sub-

units in various mammals and in the pigeon are homologouspolypeptides which are probably derived from a common

ancestral gene. Presumably, the different genes for LDHsynthesis evolved from a single ancestral gene which encodeda polypeptide with LDH activity. This gene duplicated toform the A and B genes, which then diverged to code forproducts that were immunochemically and kinetically distinct(14). The C gene must also have evolved as a consequence of aduplication event. Whether this involved the common evolu-tionary progenitor of the A and B genes, or one of these genesat a later time, is uncertain. The micro-complement fixationdata reveal that LDH-X is as unrelated to LDH-1 as it is toLDH-5. Nevertheless, all of these enzymes must have some

structural features in common since they catalyze the same

reaction and their subunits can combine, at least in vitro.At the present time it seems reasonable to propose that theprimary duplication event(s) produced three genes (A, B,and C) and that the C locus came under control of a regulator(gene?) so that its expression was limited to the primaryspermatocyte either by activation restricted to this cell typeor repression in all other cells. The finding that LDH-X is

indeed a unique protein of the sperm cell may be of consider-able practical significance by providing a new approach tocontraception in the male. Experiments are in progress todetermine whether interference with LDH-X synthesis oractivity will affect spermatogenesis or the fertilizing capacityof spermatozoa.

The technical assistance of Susan Handy, Jerrold Lerum, andKailas Patel is gratefully acknowledged. I am indebted to Dr. A.Bahn for advice on immunization procedures. The research uponwhich this publication is based was performed pursuant toContract No. NIH-69-2162 with the National Institutes ofHealth, Department of Health, Education, and Welfare.

1. Markert, C. L., Hereditary, Developmental and Immuno-logic Aspects of Kidney Disease, ed. J. Metcoff (Evanston, North-western University Press, 1962).

2. Blanco, A., and W. H. Zinkham, Science, 139, 601 (1963).3. Goldberg, E., Science, 139, 602 (1963).4. Blanco, A., W. H. Zinkham, and L. Kupchyk, J. Exp.

Zool., 156, 137 (1964).5. Zinkham, W. H., H. Isensee, and J. H. Renwick, Science,

164, 185 (1969).6. Zinkham, W. H., A. Blanco, and L. J. Clowry, Ann.

N.Y. Acad. Sci., 121, 571 (1964).7. Goldberg, E., and C. Hawtrey, J. Exp. Zool., 164, 309

(1968).8. Holmes, R. S., and C. L. Markert, Proc. Nat. Acad. Sci.

USA, 64, 205 (1969).9. Wilson, A. C., N. 0. Kaplan, L. Levine, A. Pesce, M.

Reichlin, and W. S. Allison, Fed. Proc., 23, 1258 (1964).10. Goldberg, E., Ann. N.Y. Acad. Sci., 121, 560 (1964).11. Weintraub, M., and S. Raymond, Science, 142, 1677

(1963).12. Stollar, D., and L. Levine, Methods in Enzymology, ed.

S. P. Colowick and N. 0. Kaplan (New York, Academic Press,1963), vol. VI.

13. Wasserman, E., and L. Levine, J. Immunol., 87, 290(1960).

14. Markert, C. L., and R. S. Holmes, J. Exp. Zool., 171, 85(1969).

15. Goldberg, E., Arch. Biochem. Biophys., 109, 134 (1965).16. Goldberg, E., and C. Hawtrey, J. Exp. Zool., 167, 411

(1968).17. Markert, C. L., Science, 140, 1329 (1963).18. Whitt, G. S., Science, 166, 1156 (1969).19. Zinkham, W. H., A. Blanco, and L. Kupchyk, Science,

142, 1303 (1963).20. Blackshaw, A. W., and J. I. Samisoni, Aus. J. Biol. Sci.,

19, 841 (1966).21. Robinson, J. C., and J. E. Pierce, Nature (London), 204,

472 (1964).22. Sarich, V. M., and A. C. Wilson, Science, 158, 1200

(1967).23. Stambaugh, R., and J. Buckley, J. Biol. Chem., 242,

4053 (1967).

LDH-X

LDH-5

LDH- I

Proc. Nat. Acad. Sci. USA

Dow

nloa

ded

by g

uest

on

Aug

ust 2

7, 2

021