38 Biochimica et Biophysica Acta, 671 (1981) 38-43 Elsevier/North-Holland Biomedical Press

BBA 38770

STABILITY OF LACTATE DEHYDROGENASE

II *. HYBRIDS AND GEOMETRIC ISOMERS

JOACHIM MULLER and CORNELIA KLEIN

lnstitut fiir Organische Chemic, Biochemie und lsotopenforschung der Universitiit Stuttgart, Pfaffenwaldring 55, D 7000 Stuttgart 80 (F.R.G.)

(Received May 7th, 1981)

Key words: Lactate dehydrogenase; Geometric isomer; Thermal stability; Denaturation

Hybrids of lactate dehydrogenases from pig heart and muscle and from chicken heart and pig heart were obtained by the freeze-thaw method [ 1,2]. Ion-exchange chromatography of the resulting mixtures of hybrids yielded un- usual elution patterns, i.e., the 2 + 2 hybrids P P (H2M2 and I-~2H~) were eluted in two separate peaks. These sub- forms were concluded to result from partial resolution of the three geometric isomers. The hybrids of chicken heart and pig heart lactate dehydrogenase showed three distinct levels of stability. The characteristic temperatures of denaturation were 61.5°C for H4 e, HCHP 3 and HCH2PI; 71°C for HCH2PlI and HCH p and 76.5°C for H4 c. The resis- tance towards thermal denaturation thus seemed to be governed by the least stable dimer within the tetrameric enzyme. The arrangement of stabilities of the dimers was in excellent agreement with the number of additional ion pairs between Arg241 (chicken) and Asps7 (chicken and pig) [3] within the Q-contact areas. The rate-deter- mining step of thermal denaturation of lactate dehydrogenase was concluded to comprise the distortion or disso- ciation of one of two Q-contacts of the tetramer.

Introduction

The thermal stabilities of the isoenzymes of lactate dehydrogenase from a variety of species [4 -6 ] or of artificially produced intra- and inter-species hybrids [7 -9 ] have been studied at many different labora- tories. The results, however, were obtained under widely varying conditions and they have been fitted

* Part I of this series is Ref. 3. Abbreviations: the isoenzymes of lactate dehydrogenase (L- lactate : NAD ÷ oxidoreductase, EC 1.1.1.27) from pig heart and pi~ muscle were abbreviated as H4 P, H3PM P, HPM2 P, HPM~ and M~ to characterize their composition from pig heart (H r) and pig muscle (M P) subunits. The hybrids from chicken heart and pig heart, which are purely heart-type enzymes, were characterized according to their content of chicken heart (H C) and pig heart (H P) subunits. The 2 + 2 hybrids

P P C P H2M 2 and H2H 2 were further subdivided into I and II based on their elution from the anion-exchange column.

to completely different models of thermal denatura- tion [7,8]. Since then, the X-ray structure [10,11] and sequence [12] have been published, but a reinter- pretation of thermal.denaturation studies has not been done.

Additional information was obtained in our labo- ratory by modifications of lactate dehydrogenases with imidic esters [3,13,14]. These data indicated the existence of four additional ion pairs within the en- zyme of chicken heart [3], which was shown to be exceptionally stable [15]. Since these ion pairs were identified as linking the Q-related subunits of the tetramer [3], the distortion of the Q-contacts was concluded to be the essential step in thermal denatu. ration. The study of hybrids of chicken and pig heart lactate dehydrogenase was expected to allow for further insight into the importance of the Q-contacts, since species with four, three, two, one and no addi- tional bridges should be formed. These hybrids have

0005-2795/0000-0000/$02.50 © 1981 Elsevier/North-Holland Biomedical Press

been investigated earlier [7], but because the data were fitted to a model with zero-order denaturation kinetics, the conclusions are questionable.

In contrast to the chicken heart/pig heart en- zymes, where the most significant difference was ob- served within the Q-contacts [3], these contacts were found to be nearly identical for the pair pig heart/pig

,'1 h e a r t t v _, m u s c l e ' ~ , muscle (with the exception of aer18s /x-e-las ) • Within the P- and R-contacts and the monomers themselves many more amino acid exchanges were observed and relevant information about the impor- tance of this complex pattern of exchanges should be obtained from the study of hybrids. Here again, earlier investigations were available [4-6,8,9], but they con- rained many inconsistencies. Mainly, the detailed investigations of Refs. 8 and 9 were in contrast to earlier investigations [4-6] where 'intermediate sta- bility of the hybrids' was observed. Unfortunately these investigations [8,9] did not cover the hybrids H3PM p and HPM3 P.

Methods

Enzymatic activity of lactate dehydrogenases was determined at 0.5 mM pyruvate and 0.3 mM NADH in 67 mM phosphate buffer, pH 7.2, at 25°C in a final volume of 1 ml [16]. The stable part of the pig heart and chicken heart enzymes was isolated by partial heat denaturation and affinity chromatography on Cibacron Blue Sepharose CL-6B as described in Ref. 3. Commercial pig muscle enzyme was used without further purification. Specific activities of 410 U/rag (chicken heart and pig heart) and 520 U/mg (pig muscle, sub-optimal pyruvate concentrations) were observed.

Protein concentration was based on E ~ o = 13.8 cm -t [17].

Measurement of thermal inactivation was done as described in Ref. 3.

Electrophoresis on cellulose acetate foils was car- ded out as previously described [3]. The buffer was 30 mM barbital/NaOH, pH 8.6, and 200 V were ap- plied for 1 h at room temperature. Lactate dehydro-

* Information concerning amino acid residues within the con- tacts was gained from comparison of the X-ray structure of dogfish muscle lactate dehydrogenase [10,11] with the sequence as published in Ref. 12.

39

genase was specifically stained as usual [ 18]. Discontinuous polyacrylamide gel electrophoresis

was performed according to Ref. 19. Hybridization of lactate dehydrogenases was

achieved by the freeze-thaw method [1,2]: 0.2 mg/ml concentrations (total concentration of both enzymes in 0.1 M phosphate 1 M NaC1, pH 7.0) were subjected to rapid freezing in a ---40°C cooling bath and were thawed in a 25°C water bath.

Ion-exchange chromatography on DEAE-Sepha- rose CL-6B was used for separation of the hybrids. The mixtures were applied to the preequilibrated column (4 × 27 cm) in 10 mM phosphate/30 mM NaC1, pH 7.2, and eluted with a linear gradient of 1.25 1 each of 10 mM phosphate buffer, pH 7.2, con- taining 30 mM and 250 mM NaC1, respectively. Frac- tions of 9 ml were collected.

Materials were obtained from the same sources as in part I of this series [3].

Results

Freezing and thawing of mixtures of lactate dehy- drogenase was used to produce hybrids, as described earlier [1,2]. Since the pig heart enzyme was less stable ** under these conditions than the enzyme from chicken heart, a 2-fold excess of the pig enzyme (6000 U, 15 mg) over the chicken heart enzyme (3000 U, 7.5 mg) was used. For preparation of pig heart/pig muscle hybrids equal amounts (33 mg each) of both pure forms were applied. Since both the chicken heart and the pig muscle enzyme are nearly uncharged at neutral pH, the electrophoretic iden- tification and chromatographic separation of the respective hybrids with the highly charged pig heart lactate dehydrogenase could be done in the same way. The results, too, were nearly identical (except for different peak areas because of different amounts and ratios of enzymes) and Figs. 1 and 2 apply as well to the isoenzymes of pig heart/pig muscle.

Fig. 1 shows the separation of the hybrids on the anion exchanger DEAE-Sepharose CL-6B.

In contrast to earlier publications, six peaks were

** The stability depends on the equilibrium of dissociation and reassociation and thus it is dependent on concentra- tion. The pig heart enzyme was less stable at 0.1 mg/ml but stable at 0.133 mg/ml.

40

2O

D

? 300

200

E -1OO

u

0 5 0 100 150 2 0 0 2 5 0 3130

Fraction no

Fig. 1. Ion-exchange chromatography of the hybridization mixture of chicken heart and pig heart lactate dehydrogenase on DEAE-Sepharose CL-6B. X X, enzymatic activity;

, concentration of sodium chloride. Hybridization was achieved by freezing and thawing of a mixture of 3000 U chicken heart and 6000 U pig heart lactate dehydrogenase in 0.1 M phosphate/1 M NaC1, pH 7.0. The hybridization mix- ture was loaded to the column (4 X 27 cm) in 10 mM phos- phate/30 mM NaC1, pH 7.2. Elution was achieved by apply- ing a linear gradient of 1.25 1 each of 10 mM phosphate/30 mM NaC1, pH 7.2, and 10 mM phosphate/250 mM NaC1, pH 7.2. Fractions of 9 ml were collected. The hybrids were named according to their composition from chicken heart (H C) and pig heart (H P) subunits as determined by their net charge (cL electrophoresis in Fig. 2).

resolved. The composition of the respective hybrids was determined from cellulose acetate- (Fig. 2) and disc-gel electrophoresis (not shown), which were in full agreement with each other.

According to these results the net charges of the enzymes had to be arranged in the order H4 c < HCH v < HCHPI = HCH~II < HCH3 v < H P and M4 v < HPM3 v < u~n~I = H~MzPII < H3VM P < H4 P. As in cellu- lose acetate electrophoresis, no separation of HCH~I/ HCHPII and v p v v H2M2I/H2M2II was achieved by disc-gel electrophoresis. The abbreviations were chosen to cahracterize the composition from chicken heart (H c) and pig heart (H P) or pig muscle (M v) and pig heart (H P) subunits. This characterization was based on the assumption of linear addition of the subunit charges to the net charge of the tetramer.

The thermal denaturation of two of the isolated chicken heart/pig heart hybrids is shown in Fig. 3. For both hybrids, the plot does not correspond to a simple first-order kinetic but to the decay of (at least) two active forms of greatly different stability. The amount of the labile form varied from 10 to 50% for different hybrids and the pure homotetramers iso- lated by chromatography exhibited labile activity, too, even if the stable forms were isolated previously

• ~ • ~ H~ • • C P

I . . . . H2H2I • •

• 6 -. --He4 Ill

• - _ _

o • . - -

" ~:: " - - HC~I+II (mixture)

• ~ HCHPII

l ÷

Fig. 2. Electrophoresis of pooled and concentrated fractions from ion-exchange chromatography on cellulose acetate foils. 0.05 #1 each of solutions cofitaining about 100 U/ml were placed at the start position and subjected to electrophoresis at 200 V for 1 h in 30 mM barbitalfNaOH, pH 8.6. The staining for enzymatic activity is shown. H C ... H P, pooled and concentrated fractions from chromatography (cf. Fig. 1).

r-~

_>

~3 Qe

2 . 0 ×

1

?o

0 4

I 0 5

×

I I I 10 15 20

t :mm

Fig. 3. Thermal inactivation of the hybrids HCH P and HCH2PI at 65°C. o, HCHP; ×, HCH2PI. Measurement was done by incubation of 0.1 mg/ml of enzyme in 67 mM phosphate buffer, pH 7.2, as described earlier [3]. Since the labile part of activity could be eliminated by partial thermal inactivation and affinity chromatography on Cibacron Blue Sepharose (of. Ref. 3), the half-life was taken from the straight section of the plot, as indicated by the solid line.

[3] and used for hybridization. As with pure pig heart and chicken heart enzymes, the stable part of the enzymes could be isolated by partial heat inac- tivation and affinity chromatography as described earlier [3]. This treatment did not lead to contamina- tion of the pure hybrids with other isoenzymes and raised the specific activities to 400 U/mg. This ob- servation was in agreement with the lack of rehy- bridization by partial thermal inactivation reported earlier [8,9]. The 'purified' hybrids now showed first- order kinetics of thermal inactivation and their half- lives were equal to the half-lives as determined from the straight section of the plots, as indicated by the solid line in Fig. 3. Therefore this half-life was con- sidered to be significant for comparison of the hy- brids. In any way the close similarity of the thermal stabilities of the hybrids HCH P and HCH~I is evident from Fig. 3.

Measurements of half-lives were carried out at two

41

TABLE I

CHARACTERISTIC TEMPERATURES OF DENATURA- TION OF CHICKEN HEART (HC)/PIG HEART (H P) HY- BRIDS OF LACTATE DEHYDROGENASE

The temperature for a half-life of 20 min is given. Half-lives were taken from measurements as indicated in Fig. 3 at two or three different temperatures. Characteristic temperatures were determined from Arrhenius plots of log ( t l /2 ) against the reciprocal absolute temperature.

Hybrid Characteristic temperatures (°C)

HP4 61.5

HCH P 61.7 2 2

HCHPI 61.3 2 2

H~I~I I 71.1

HCH 70.6

H4C 76.5

or three different temperatures. 'Characteristic dena- turation temperatures' corresponding to a half-life of 20 min then were determined from Arrhenius plots of log (tl/2) against the reciprocal absolute temperature. From the slopes of the plots an acivation energy of 500 kJ/mol was determined. Table I summarizes the results for the chicken heart/pig heart series. The hy- brids exhibit three distinct levels of stability at 61.5, 71 and 76.5°C, respectively. The results were not as clear-cut in the case of the pig heart/pig muscle isoen- zymes. The characteristic temperatures were deter- mined as 56°C for M~, 61.5°C for H4 e and 54-55°C for hybrids. Since the confidence limit of the deter- minations was about -+I°C, even the conclusion of equal stability of all isoenzymes except HP4 seemed possible. The measurements are in contrast to earlier publications stating intermediate stability of the hybrids [4-6] and they tend to support the measure- ments of Refs. 8 and 9, where a little destabilization of P P H2M2 (not separated into the subforms), as com- pared to M4 P, was found.

Discussion

Resolution of six peaks from hybridization mix- tures of lactate dehydrogenases is possible with the use of anion-exchange chromatography. In addition

42

to the known separation of the five isoenzymes, the (H2H2 and 2 2) are split peaks of the 2 + 2 hybrids c P HPMPX

into two well-separated subforms. This improvement in resolution may be due to the use of the ion-ex- changer DEAE-Sepharose CL-6B, instead of DEAE- cellulose [9] or DEAE-Sephadex [7], or to the use of a very high exchanger/protein ratio. No separation of the subforms is possible * by cellulose acetate or disc- gel electrophoresis and therefore the separation on the anion-exchanger cannot be due to a difference in net charge but only to differences in the distribution of charges on the surface of the hybrids.

The later eluting subforms (H2CHPII and HPM2PII) may be enabled to interact better with the exchanger by concentration of their charges on one side of the molecule. The premise for different distributions of charges is the existence of different arrangements of the subunits within the 2 + 2 hybrid-tetramer.

Indeed, the so-called 'geometric isomers' are well known from theoretic considerations [20]. Three geometric isomers are possible, since the identical monomers may interact via the P- or Q- or R-contacts [21]. The separation of only two of the three pos- sible isomers may be explained in two alternative ways: either by formation of only two subforms during hybridization or by inefficiency of the ion- exchanger in separating two of the three subforms from each other. A plausible model for answering this question and further support for the identification of the geometric isomers can be obtained from the ther- mal stabilities of the chicken heart/pig heart hybrids as follows.

The existence of three distinct levels of stability among the hybrids clearly indicates the distortion of three different structures within the rate-limiting step of denaturation. This is not compatible with the rate-determining denaturation of one monomer fol- lowed by rapid destruction of the whole structure [8], nor with a concerted transition of the tetramer, since in the first model the least stable monomer would determine the stability of the molecule, whereas the second model would predict an interme- diate behaviour of the hybrids. The most straightfor- ward explanation of the data is that the stability of the tetramer is determined by the least stable dimer.

* In our hands the separation of two subforms by disc-gel electrophoresis reported earlier [21] did not work.

The rate-limiting step thus is the destruction of one dimer and in our model this step is followed by rapid changes leading to the final completely inactivated state, whereas an earlier model [7] postulates the survival of the more stable dimer. The cited model, however, is based on the assumption of zero-order kinetics of denaturation.

The question of whether these dimers are bridged by the P-, Q- or R-contacts is answered by the earlier paper of this series [3]. The exceptional stability of the chicken heart enzyme is explained by the forma- tion of additional ion pairs bridging Arg241 and Asps7 across the Q-axis, whereas these ion pairs are weak or absent between Lys241 and Asps7 in pig heart lactate dehydrogenase. Thus HC-H c dimers con- tain two bridges within the Q-contact, HC-H P dimers only one and HP-H P dimers none. The three levels of stability of the chichen heart/pig heart hybrids thus are explained as follows: H P, HCH~ and H2CHPI con- tain the least stable HP-H P dimer; H2CH2PII and HCH l' are destroyed by the rate-limiting destruction of one HC-H P dimer and H4 c is the most stable since it con- tains only HC-H c dimers. HCHPI thus is concluded to contain only one geometric isomer with bridging of the identical subunits across the Q-axis, whereas HCHPII may be composed of the other two forms with interaction of the identical subunits by the P- and R.contacts, respectively. These two geometric isomers do not exhibit different stability, since they contain the same Q-bridged dimers.

Our model of thermal denaturation of lactate dehydrogenase is thus the following. The rate-deter- mining step is the distortion or dissociation of one Q-contact within the tetramer leaving the other Q-contact undisturbed. This step is not preceded by any reversible dissociation or association, since this would lead to rehybridization and/or to non-first- order kinetics. To allow for an extensive change within the Q-contact, however, the P- and R-contacts and the monomers themselves must change in con- formation too, but these changes are not observed as differences in thermal stability, since they occur within and between the closely similar chicken and pig heart subunits. The much greater differences between pig heart and pig muscle subunits which are located within the P.contacts, the monomers them- selves and especially within the R-contacts, but not between the Q-related subunits, do not give rise to

similarily important effects and thus cannot improve

the present view. The nature of the labile forms, which evidently are produced by the process of dissociation and reassociation, remains to be clarified.

References

1 Freier, E.F. and Bridges, R.A. (1964)Biochem. Biophys. Res. Commun. 17,335-340

2 Chilson, O.P., CosteUo, L.A. and Kaplan, N.O. (1965) Biochemistry 4, 271-281

3 Miiller, J. (1981) Biochim. Biophys. Acta 669, 210-215 4 Fondy, T.P., Pesce, A., Freedberg, I., Stolzenbach, F. and

Kaplan, N.O. (1964) Biochemistry 3, 522-530 5 Wachsmuth, E.D. and Pfleiderer, G. (1963) Biochem. Z.

336, 545-556 60kabe, K., Hayakawa, T., Hamada, M. and Koike, M.

(1968) Biochemistry 7, 79-90 7 Saito, M. (1972) Biochim. Biophys. Acta 258, 17-26 8 Siidi, J. (1970) FEBS Symp. 18 (1968), 169-176 9 Khan, M.G. and Siidi, J. (1968) Acta Biochim. Biophys.

Acad. Sci. Hung. 3,409-420 10 Rossmann, M.F., Adams, M.J., Buehner, M., Ford, G.C.,

Hackert, M.L., Liljas, A., Rao, S.T., Banaszak, L.J., Hill, E., Tsernoglou, D. and Webb, L. (1973) J. Mol. Biol. 76, 533

43

11 Holbrook, J.J., Liljas, A., Steindel, S.J. and Rossmann, M.G. (1975) in The Enzymes (Boyer, P.D., ed.), Vol. 11, pp. 191-292, Academic Press, New York

12 Eventoff, W., Rossmann, M.G., Taylor, S.S., Torff, H.-J., Meyer, H., Keil, W. and Kiltz, H.-H. (1977) Proc. Natl. Acad. Sci. USA 74, 2677-2681

13 Tuengler, P. and Pfleiderer, G. (1977) Biochim. Biophys. Acta 484, 1-8

14 Kapmeyer, W., Keil, W., Kiltz, H.-H., Meyer, J. and Pflei- derer, G. (1977) Hoppe-Seyler's Z. Physiol. Chem. 358, 39-46

15 Wilson, A.C., Kaplan, N.O., Levine, L., Pesce, A., Reich- lin, M. and Allison, W.S. (1964) Fed. Proc. 23, 1258- 1266

16 Pfleiderer, G., Holbrook, J.J., Zaki, L. Jeckel, D. (1968) FEBS Lett. 1,129-132

17 Pfleiderer, G. and Jeckel, D. (1957) Biochem. Z. 329, 370

18 Markert, C.L. and Massaro, E.J. (1966) Arch. Biochem. Biophys. 115,417-426

19 Maurer, H.R. (1971) Disk-Electrophoresis No. 1, p. 44, W. de Gruyter, Berlin

20 Monod, J., Wyman, J. and Changeux, J.P. (1965) J. Mol. Biol. 12, 88

21 Levitzki, A. (1972) FEBS Lett. 24, 301-304