Dissociation of mitochondrial malate dehydrogenase into active soluble subunits

Biochimica et Biophysica Acta 1478 (2000) 248^256 www.elsevier.com/locate/bba

Dissociation of mitochondrial malate dehydrogenase into active soluble subunits Shruti Shaw a , Ryan Geyer b , Gerald M. Alter

c;

*

a

Electronic Data Systems, 4100 Springboro Pike, Dayton, OH 45439, USA Biomedical Sciences Ph.D. Program, Wright State University, Dayton, OH 45435, USA Department of Biochemistry and Molecular Biology, Wright State University, Dayton, OH 45435, USA b

c

Received 11 November 1999; received in revised form 8 February 2000; accepted 10 February 2000

Abstract Gel exclusion chromatographic studies demonstrate that cytosolic and mitochondrial malate dehydrogenases (cMDH and mMDH) dissociate into subunits in the presence of 0.1% of the non-ionic detergent Triton X-100 (TX-100). The presence of cofactor and catalytically competent cofactor^substrate pairs does not protect mMDH against this dissociation. In contrast, cMDH dimers resist dissociation in the presence of either addition. Since steady state kinetic studies indicate both enzymes are fully active in the presence of 0.1% TX-100, we conclude that quaternary structure is not a kinetically important feature of mMDH structure and cooperativity does not account for mMDH kinetic anomalies. In contrast, cooperativity is a reasonable explanation for cMDH kinetic properties even in the presence of 0.1% TX-100, since this enzyme's subunits associate in the presence of active site ligands. The existence of fully active mMDH subunits raises the possibility that this species rather than the dimer may be a constituent of proposed multi-enzyme complexes of the mitochondrion. Preliminary chromatographic experiments involving gently disrupted mitochondria have found MDH activity in differently sized complexes, all with molecular weights larger than the mMDH dimer but smaller than complexes anticipated for multienzyme complexes involving other enzymes and the mMDH dimer. ß 2000 Elsevier Science B.V. All rights reserved. Keywords: Malate dehydrogenase; Cooperativity; Multi-enzyme complex; Triton X-100

1. Introduction Malate dehydrogenase (MDH) is a ubiquitous enzyme which exists in at least two isozymic forms in all eukaryotes. These enzymes are essential to cellular

* Corresponding author. Fax: +1-937-775-3730; E-mail: [email protected]

metabolism. The mitochondrial enzyme (mMDH) is a central component of the citric acid cycle and the aspartate^malate shuttle [1]. The cytoplasmic enzyme (cMDH) is a cytoplasmic participant in the same shuttle, in gluconeogenesis and in fatty acid synthesis [2]. The aggregation state of these enzymes and the signi¢cance of these aggregates to mMDH and cMDH physiological function have been a topic of sustained interest. Previous reports have concluded that cMDH is active only as a dimer and undergoes little or no dissociation except under denaturing conditions [2].

0167-4838 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 3 8 ( 0 0 ) 0 0 0 3 3 - 9

BBAPRO 36116 11-5-00

S. Shaw et al. / Biochimica et Biophysica Acta 1478 (2000) 248^256

Reports suggesting that the enzyme is cooperative are consistent with an intimate correlation between activity and dimerization [3,4]. The situation does not seem as clear in the case of the mitochondrial enzyme. The aggregation state of this enzyme has been examined extensively. It crystallizes with four subunits per asymmetric subunit [5,6], seems to be predominantly dimeric in solution [7] and may be involved in mitochondrial multi-enzyme complexes [8,9]. Investigators have reported that this dimeric enzyme dissociates in solutions when protein concentrations are low [10,11], or at acidic pHs [10,12, 13] or at high pressure [14]. Others have reported a salt dependence of mMDH dimerization and activity [15]. Still other investigators have reported negligible dissociation at or around neutral pH [7, 16,17]. Reports have linked the aggregation state of mMDH with its catalytic activity. Harrison's group [12,18] reports that cofactor binding and enzymatic activity are linked with monomer association to form homo-dimers. Several investigators report producing mMDH subunits and characterizing them with respect to their catalytic activity. Rather drastic conditions, such as treatment with denaturants, insure subunit dissociation but cause extensive loss of enzymatic activity by unraveling tertiary structure [19]. However, by covalent cross-linking to solid supports, mMDH monomers have been produced, and their kinetic properties have been determined. These modi¢ed monomers have activities that are similar to the native enzyme [20,21]. This is consistent with kinetic properties of hetero-dimers comprised of a native and an inactive subunit having many of the kinetic features of the native enzyme [22]. Reports of organized multi-enzyme complexes involving mMDH emphasize the importance of determining the size of the smallest soluble and active mMDH and cMDH species. Here, we report dissociation of MDHs into soluble subunits in the presence of the non-ionic detergent, Triton X-100 (TX-100). Cofactor and substrate cause cMDH but not mMDH to form an active dimer while in the presence of TX-100. The presence of active mMDH monomers in the absence of chemical modi¢cations suggests that this could be the minimal mMDH component in multi-enzyme complexes of the mitochondrion.

249

2. Materials and methods 2.1. Materials mMDH (lot # 127892) was purchased from Boehringer Mannheim. CL-6B blue Sepharose, Sephacryl100 and Sephacryl-200 were obtained form Amersham Pharmacia Biotech while substrates and cofactors (OAA, malate, NAD and NADH) were purchased from Sigma chemical company. All other chemicals were of reagent grade or better. All water used in this study was from a Milli Q (Millipore Corp., Danvers, MA, USA) reverse osmosis system. 2.2. Assay of mMDH and cMDH MDHs were routinely assayed in the direction of OAA reduction. The standard assay mixture for the mitochondrial enzyme contained 50 mM sodium phosphate bu¡er, 0.25 mM NADH and 0.2 mM OAA, pH 7.5, 24³C. cMDH was assayed in solutions containing 0.25 mM NADH and 1 mM OAA in 50 mM HEPES, pH 8.0. In both cMDH and mMDH assays, NADH utilization was monitored by measuring the change in absorbance per unit time at 340 nm in a Cary 219 spectrophotometer. NADH concentrations were calculated from absorbance measurements using an NADH extinction coe¤cient of 6300 M31 cm31 [23]. Standard assay conditions were routinely used to monitor chromatographic column e¥uents regardless of whether columns were run in the presence or absence of TX-100. When detailed kinetic studies were performed in the presence or absence of TX-100, concentrations of substrate, cofactor and TX-100 that were used are noted in the ¢gure legends. Units of speci¢c activity for both enzymes were Wmol of substrate used per minute per mg of enzyme. 2.3. Puri¢cation of porcine heart mMDH and cMDH Commercial mMDH was dialyzed against a 50 mM phosphate bu¡er, pH 7.5, and then loaded on a CL-6B blue Sepharose chromatography column. The column was washed with two column volumes of 5 mM phosphate bu¡er, pH 6.2, and the MDH eluted with 100 WM NADH, in the same bu¡er. Porcine cMDH was puri¢ed as previously described [4].

BBAPRO 36116 11-5-00

250

S. Shaw et al. / Biochimica et Biophysica Acta 1478 (2000) 248^256

Both enzyme preparations were characterized by single bands in SDS^PAGE analysis. 2.4. Preparation of cytosolic and mitochondrial fractions from bovine heart The methods of Azzone and coworkers [24] were used to prepare cytosolic and mitochondrial fractions from beef heart extract. Brie£y, heart ventricular muscle tissue was used for all preparations and mince was prepared by grinding the tissue in 250 mM sucrose bu¡er (2 mM EDTA, 10 mM Tris^Cl, pH 7.4), 1:2 weight to volume. The mince was suspended in the same bu¡er, 1:2 weight to volume, and strained through cheese cloth. For cytoplasmic preparations, the suspension was sonicated brie£y (2^15 s) to disrupt cell membrane and centrifuged for 10 min at 17 000Ug to remove organelles and cell debris. The supernatant was used without further fractionation. For mitochondrial preparations, strained mince was blended twice for 9 s at high speeds spaced by a 25 s interval. The suspension was centrifuged at 700Ug for 20 min to sediment cell debris but not mitochondria. The supernatant was then centrifuged at 35 000Ug for 15 min at 20³C to pellet the mitochondria. The resuspended pellet is the `heavy mitochondria' fraction and was used without further puri¢cation. To verify that mitochondrial matrix preparations were free of cMDH and visa versa, 50 Wl of preparations was incubated with 350 Wl of 1.3 mM 4,4Pbis(dimethylamino)diphenylcarbinol (BDCOH) for 30 min at room temperature and then assayed for activity. BDCOH is a covalent inhibitor which e¤ciently inactivates mMDH but not cMDH [25]. As expected, BCDOH profoundly inhibited MDH activity of mitochondrial preparations but did not e¡ect activities of cytoplasmic preparations. 2.5. Gel exclusion column chromatography All types of samples were chromatographed on a Sephacryl-200 column (Pharmacia), 80 cm long and 1.7 cm in diameter. Preparations were also chromatographed on a Sephacryl-100 column, 170 cm long and 1.7 cm in diameter. Both columns were standardized using commercial catalase, alcohol dehydrogenase, cMDH, albumin, ovalbumin and trypsin, as

well as ferritin, to identify the columns' void volume. Standards were dissolved in the appropriate chromatographic bu¡er and chromatographed separately. Routinely, when chromatographic bu¡ers were changed, columns were washed with ¢ve column volumes of the new bu¡er to assure complete equilibration. When column bu¡er changes involved introducing or removing TX-100, columns were recalibrated using the standards just mentioned. Routinely, columns were loaded with 1 ml of a 0.6 mg/ml MDH solution when chromatographing a pure protein. When cytosolic or mitochondrial preparations were chromatographed, 1 ml of the appropriate preparation was layered on the column and subsequently eluted. Recalibration of columns after chromatography of crude cytosol and mitochondrial preparations demonstrated that these preparations did not change column properties. 2.6. Circular dichroism (CD) spectra of native mMDH CD spectra of mMDH (400 Wg/ml) in standard bu¡er were obtained at 23³C in the presence and absence of 0.1% hydrogenated TX-100 (hTX-100), using a JASCO J500A spectro-polarimeter. The spectral range examined was 190^270 nm. 3. Results 3.1. Chromatography of puri¢ed mMDH and cMDH Puri¢ed preparations of mMDH and cMDH were subjected to gel exclusion chromatography in the presence and absence of 0.1% TX-100. Elution pro¢les were characterized by single symmetrical activity peaks indicative of single enzyme species (Fig. 1) (cMDH not shown). MDH elution volumes varied depending on the sample preincubation and column elution bu¡er. Molecular weights corresponding to these peaks were estimated based on the elution volume of peak fractions. Triplicate determinations under each of several chromatographic conditions were used to quantify the elution behavior of MDH. Results of a number of chromatographic experiments are summarized in Table 1. These results show that in standard preincubation and elution

BBAPRO 36116 11-5-00

S. Shaw et al. / Biochimica et Biophysica Acta 1478 (2000) 248^256

Fig. 1. Representative Sephacryl-100 elution pro¢le of puri¢ed mMDH in the presence (b) and absence (R) of 0.1% TX-100. Samples were preincubated (overnight) in the same bu¡er used to elute the column. Other components in these bu¡ers were 50 mM phosphate and 1 mM EDTA, pH 7.5. Lines in this ¢gure represent a ¢t of experimental data (points) to a Lorentzian function. Activities of eluant fractions were determined as described in Section 2.

bu¡ers (50 mM phosphate, pH 7.5, 1 mM EDTA), cMDH and mMDH elute as 70 and 66 kDa species, respectively, consistent with both being active dimers. There is little evidence for any other active species. However, adding TX-100 to these bu¡ers caused cMDH activity to elute as a 47 kDa species (Table 1) while mMDH eluted as a 33 kDa species (Fig. 1, Table 1). Addition of cofactor (0.8 mM NADH) or cofactor plus substrate (2 mM OAA) caused the cMDH activity to again elute as a

Fig. 2. Representative OAA dependence of mMDH activity in the presence (b) and absence (F) of 0.1% TX-100. In these assays, NADH concentrations were ¢xed at 0.2 mM. Other conditions were 50 mM phosphate, pH 7.5 and 24³C.

251

Fig. 3. Representative NADH concentration dependence of mMDH assayed in the presence (b) and in the absence (F) of 0.1% TX-100. In these assays, NADH concentrations were held constant at 0.2 mM. Other bu¡er components were 50 mM phosphate, pH 7.5. The temperature was maintained at 24³C. Lines are no-linear least squares ¢ts of the experimental data (points) to a model that assumes normal Michaelis^Menten kinetics and independent binding of substrate to an inhibitory site.

70 kDa species. In contrast, neither cofactor (0.8 mM NADH) nor cofactor and substrate (0.8 mM NADH, 2 mM OAA) caused mMDH to elute as a 66 kDa peak; rather, a 33 kDa peak was observed (Table 1). Further, mMDH preincubated in 0.1% TX-100-containing bu¡er chromatographed as 33 kDa species even when eluted with bu¡er lacking TX-100 (Table 1).

Fig. 4. Representative Eadie^Hofstee plot of the OAA dependence of mMDH activity assayed in the presence (b) and in the absence (F) of 0.1% TX-100. In these assays, the NADH concentrations were held constant at 0.2 mM. Other bu¡er components were 50 mM phosphate, pH 7.5. The temperature was maintained at 24³C. Lines are linear regression lines calculated from each data set (points).

BBAPRO 36116 11-5-00

252

S. Shaw et al. / Biochimica et Biophysica Acta 1478 (2000) 248^256

Fig. 5. Representative Eadie^Hofstee plot of the OAA dependence of cMDH activity assayed in the presence (b) and absence (R) of 0.1% TX-100. In these assays, NADH concentrations were 0.25 mM. Assays were performed in 50 mM phosphate bu¡er, pH 7.5 at 24³C.

3.2. Conformational analysis The conformation of mMDH was monitored by CD spectroscopy. The spectrum of the enzyme was the same in bu¡ers containing 0.1% hTX-100 and in bu¡ers de¢cient in this detergent in the wavelength range form 190 to 270 nm (data not shown). hTX100 was used in this assay because, in contrast to TX-100, it is spectrally transparent in the wavelength range we examined. Yet, hTX-100 has structural and detergent properties very similar to TX-100 [26].

Fig. 6. Representative Sephacryl-200 chromatographic pro¢le of mitochondrial matrix prepared with 10 s of sonication of heavy mitochondria and assayed as described in Section 2. Relative MDH activity (R) and absorbance at 280 nm (b) are shown. The solid line in this ¢gure represents a ¢t of experimental data (points) to Lorentzian functions. The ¢tting process assumed that the elution pro¢le represented two unresolved eluant peaks. The dashed line represents a non-parametric ¢t of absorbance data points.

3.3. Kinetic characterization of MDHs Since speci¢c activities of mMDH were similar in the presence and the absence of 0.1% TX-100, detailed kinetic analyses were performed under the same conditions of pH as the column chromatographic experiments. It is apparent from both OAA (Fig. 2) and NADH (Fig. 3) concentration depen-

Table 1 Chromatographic estimates of MDH molecular weights Sample

Bu¡er additionsa

Molecular weight (kDa)

cMDH cMDH cMDH mMDH mMDH mMDH mMDH mMDH Cytosolic fractiond Matrix fractiond

None 0.1% TX-100 0.1% TX-100, 0.8 mM NADH, 2.0 mM OAA None 0.1% TX-100 0.1% TX-100 preincubationc 0.1% TX-100, 0.8 mM NADH 0.1% TX-100, 0.8 mM NADH, 2.0 mM OAA Nonee Nonee

Matrix fraction

0.1% TX-100

70 þ 1.6 70 þ 1.6 47 þ 1.9 66 þ 1.8 33 þ 1.8 33 þ 1.7 33 þ 2.1 33 þ 2.1 69 þ 3.1 109 þ 1.8 80 þ 1.8 33 þ 1.9

a

Signi¢canceb

0.011 0.007 0.007 0.007 0.007 0.003 0.007 0.007

Additions are to bu¡ers containing 50 mM phosphate, pH 7.5. P values are for comparisons of the MDH molecular weights in the particular preparation with the dimeric mitochondrial (66 kDa) or cytoplasmic (70 kDa) enzymes. c Additions were made to the preincubation bu¡er but not the elution bu¡er. d Prepared as described in Section 2. e Bu¡er contains 250 mM sucrose, 2 mM EDTA, 10 mM Tris^Cl, pH 7.4. b

BBAPRO 36116 11-5-00

S. Shaw et al. / Biochimica et Biophysica Acta 1478 (2000) 248^256

253

Table 2 Kinetic constants describing MDH activity in the presence and absence of 0.1% TX-100a Enzyme

0.1% TX-100 addition

Variable substrate

KbM

VcM

KdD

mMDH

No

No

NADH OAA NADH OAA OAA OAA

1755 þ 119 1438 þ 112 2370 þ 135 2190 þ 150 1010 þ 90e 400 þ 70f 920 þ 110e 350 þ 120f

ND 740 þ 60 ND 760 þ 40

Yes

89 þ 9 22 þ 2 49 þ 5 15 þ 3 28 þ 3e 4 þ 2f 34 þ 5e 4 þ 2f

Yes cMDH

a

Assays performed in 50 mM phosphate bu¡ers, pH 7.5 Units are WM. Error is standard error of estimate. c Units are Wmol/min, mg. Error is S.E.M. of estimate. d Units are WM. Error is S.E.M. of estimate. e Estimated from the limiting slopes and ordinate intercepts of plots like those in Fig. 5 as abscissa values approach zero. f Estimated from the limiting slopes and ordinate intercepts of plots like those in Fig. 5 as abscissa values approach in¢nity. b

dences of mMDH activity that the qualitative characteristics of mMDH kinetics do not change in the presence or absence of 0.1% TX-100. Eadie^Hofstee plots (Fig. 4) were used to derive kinetic constants describing mMDH kinetics at low to medium NADH concentrations and at low OAA concentrations. When analyzing a broad range of OAA concentrations, KM and VM values from Eadie^Hofstee analyses were used as initial parameter values for non-linear ¢tting of the entire OAA range examined. These variations were ¢t to a mechanism in which the inhibition phase of the plots (Fig. 2) was assumed to re£ect substrate binding to a second, independent site that completely inhibits mMDH activity. NADH dependences of mMDH activity did not extend into an inhibitory phase, so this feature was not included when modeling this concentration dependency. The lines in Figs. 2 and 3 represent the best ¢t of the model to experimental data. Constants derived from kinetic analyses are summarized in Table 2. Eadie^Hofstee plots of the OAA dependence of cMDH activity (Fig. 5) show that qualitative characteristics of this plot are not signi¢cantly changed in the presence of 0.1% TX-100. OAA concentration dependences are similar to those previously reported from this laboratory [4]. 3.4. Chromatography of cytosolic and mitochondrial matrix cell fractions Matrix from heavy mitochondria fractions and cy-

tosol preparations was chromatographed to estimate the size of macromolecules and/or macromolecular complexes containing MDH activity. When gently prepared cytosol (Section 2) was chromatographed, cMDH activity reproducibly eluted as a single species with an apparent molecular weight of 69 þ 3.0 kDa (Table 1). More severe methods of preparation (sonication for as much as 15 s) did not change the elution pro¢le. Mild disruption of heavy mitochondria by sonication (10 s) or by overnight treatment with 0.1% TX-100 (250 mM sucrose, 2 mM EDTA, 10 mM Tris^HCl, 0.1% TX-100) reproducibly resulted in a broad elution pro¢le suggesting two or more unresolved peaks (Fig. 6) and a single symmetrical elution peak, respectively. As indicated in Table 1, the broad peak has maxima consistent with 109 and 80 kDa species, respectively, while mMDH activity elutes as a 33 kDa species in the presence of 0.1% TX-100 (Table 1). 4. Discussion Here, we report that mMDH readily dissociates into its subunits in the presence of 0.1% of the non-ionic detergent TX-100. This conclusion is based on gel exclusion chromatography results which indicate that the enzyme's activity chromatographs as a 33 kDa species in the presence of 0.1% TX-100 (Fig. 1). Further, the presence of neither cofactor (NADH) nor a productive cofactor^substrate pair (NADH+

BBAPRO 36116 11-5-00

254

S. Shaw et al. / Biochimica et Biophysica Acta 1478 (2000) 248^256

OAA) induces subunit reassociation (Table 1). The kinetic parameters that quantitatively describe mMDH activity are perturbed by the presence of 0.1% TX-100. In fact, KM s, VM s and kcat /KM s describing mMDH activity are somewhat better in the presence of the non-ionic detergent (Table 2). However, qualitative aspects of mMDH activity pro¢les are not changed by these dissociation conditions. We conclude that this TX-100 treatment produces active subunits which display the kinetic characteristics of the native enzyme. This behavior con¢rms and extends previous reports that active subunits of mMDH with kinetic properties like the native enzyme can be produced by covalent immobilization on solid support [20,21]. However, results reported here are free of uncertainties regarding the e¡ects of chemical modi¢cations and di¤culties in quantifying speci¢c activities of solid state enzymes. In light of our results, and previous reports, it is very unlikely that subunit^subunit interactions play a role in mMDH kinetic activities. Rather, these activities must re£ect the structural properties of each individual subunit. CD measurements which detect no perturbations of mMDH secondary structure with the addition of hTX-100 suggest that subunit association/dissociation does not change the tertiary structure of mMDH subunits. This is quite consistent with activity being largely una¡ected by subunit dissociation. We also conclude that cMDH does not form active monomers under conditions that we have explored here. Though a 47 kDa species is produced by treatment with 0.1% TX-100, activity assays are performed under conditions in which cMDH dimers reform (no TX-100; 0.25 mM NADH; 0.2 mM OAA, pH 7.5 and 24³C). Therefore, it is likely that the 47 kDa species reverts to the dimer during assays. The molecular constitution of the 47 kDa species is unclear. Its molecular weight could indicate a large complex, perhaps a subunit^detergent complex or alternatively, a partially unfolded subunit. We cannot distinguish among these possibilities. It is clear, however, that incubation of cMDH with cofactor or substrate plus cofactor results in the dimeric form of the enzyme even in the presence of 0.1% TX-100. This species has the same kinetic properties as the native enzyme, including negative cooperativity. Results reported here do imply that cofactor binding or

cofactor plus substrate binding alters the subunit's interface conformation enhancing subunit^subunit interactions. This is certainly consistent with cMDH being a cooperative enzyme [4]. In contrast, the inability of active site ligands to perturb TX-100dependent dissociation of mMDH is consistent with these ligands not e¡ecting the enzyme's subunit interface. This is certainly consistent with mMDH being a non-cooperative enzyme. It is also interesting to note that mMDH subunits probably bind TX-100 more avidly than cMDH. We arrive at this conclusion since chromatographing an mMDH^TX-100 complex on S200 or S100 Sephacryl columns does not remove su¤cient TX-100 to allow mMDH to dimerize, while TX-100-dependent dissociation of cMDH is reversed simply by addition of active site-bind ligands. Di¡erences in the e¡ect of TX-100 on cMDH and mMDH may re£ect di¡erences in the surface properties of these enzymes. Gleason and coworkers [6] compared surface properties of cMDH and mMDH and concluded that mMDH's surface is more hydrophobic. This could rationalize the greater TX-100 susceptibility of mMDH dimers. Alternatively, comparing subunit interfaces of these two proteins [27,28], it is clear that they both are predominantly hydrophobic. However, they di¡er in that the mMDH interface has an excess of formal positive charge relative to cMDH. This may weaken mMDH subunit interactions relative to cMDH and make them more susceptible to TX-100 insertion. Preliminary results for column chromatography of mitochondrial matrix and cytosol preparations suggest that MDH activity may exist in several species within heart cells. Though MDH activity in cytosolic preparations chromatographs as if it were a dimer, MDH activity in mitochondrial matrix preparations prepared by brief sonication chromatographed as two or more species (Table 1). None of the apparent sizes of these species (80 to about 110 kDa, Table 1) corresponds to the size of the metabolon recently proposed by Srere and coworkers [9]. However, the results reported here raise an intriguing possibility that subunits of MDH, which might dissociate from dimers, could interact with hydrophobic portions of other enzymes. Therefore, MDH subunits might be one of the components in multi-enzyme complexes. Results form TX-100-treated prepara-

BBAPRO 36116 11-5-00

S. Shaw et al. / Biochimica et Biophysica Acta 1478 (2000) 248^256

tions of mitochondrial matrix (Table 1) may indicate that this detergent releases the mMDH subunit from such a complex much as it does from the dimeric enzyme. In summary, we report that mMDH and cMDH may be dissociated into subunits in the presence of 0.1% TX-100. However, only mMDH remains a subunit in the presence of active site ligands. These subunits have essentially native mMDH activity, indicating that quaternary structure is not a requisite for activity. This is in apparent contrast to cMDH, whose subunits reassociate in the presence of active site ligands. Finally, our results suggest that mMDH subunits may be the basic constituent of mMDH activity in in vivo multi-enzyme complexes. Acknowledgements The authors thankfully acknowledge the support of the Biomedical Sciences Ph.D. Program of Wright State for support of R.G., and the ¢nancial support provided by the Ohio Valley A¤liate of the American Heart Association for these studies. References [1] R.A. Musrati, M. Kollarove, N. Mernik, D. Mikulasova, Malate dehydrogease: distribution, function and properties, Gen. Physiol. Biophys. 17 (1998) 193^210. [2] L.J. Banaszak and R.A. Bradshaw, Malate dehydrogenase, in: P.D. Boyer (Ed.), The Enzymes, Vol. XI, Academic Press, New York, pp. 369^396. [3] C.T. Zimmerle, P.P. Tung, G.M. Alter, Ligand-induced asymmetry between active sites of cytoplasmic malate dehydrogenase: a chemical modi¢cation study, Biochemistry 26 (1987) 8535^8541. [4] C.T. Zimmerle, G.M. Alter, Cooperativity in the mechanism of malate dehydrogenase, Biochemistry 32 (1993) 12743^ 12748. [5] M.S. Weininger, L.J. Banaszak, Mitochondrial malate dehydrogenase; Crystallographic properties of the pig heart enzyme, J. Mol. Biol. 119 (1978) 443^449. [6] W.B. Gleason, Z. Fu, J. Birktoft, L.J. Banaszak, Re¢ned crystal structure of mitochondrial malate dehydrogenase from porcine heart and the consensus structure for dicarboxylic acid oxidoreductases, Biochemistry 33 (1994) 2078^2088. [7] S.A. Sanchez, T.L. Hazlett, J.E. Brunet, D.M. Jameson, Aggregation states of mitochondrial malate dehydrogenase, Protein Sci. 7 (1998) 2184^2189.

255

[8] P.A. Srere, B. Sumegi, A.D. Sherry, Organizational aspects of the citric acid cycle, Biochem. Soc. Symp. 54 (1987) 173^ 178. [9] C. Velot, M.B. Mixo, M. Teig, P.A. Srere, Model of a quinary structure between Krebs TCA cycle enzymes: a model for the metabolon, Biochemistry 36 (1997) 14271^14276. [10] D.C. Wood, C.T. Hodges, J.H. Harrison, The relation of the pH and concentration-dependent dissociation of porcine heart mitochondrial malate dehydrogenase, Biochem. Biophys. Res. Commun. 82 (1978) 943^950. [11] J.D. Shore, S.K. Chakrabarti, Subunit dissociation of mitochondrial malate dehydrogenase, Biochemistry 15 (1976) 875^879. [12] C.T. Hodges, J.C. Wiggin, J.H. Harrison, Investigation of the relation of the pH-dependent dissociation of malate dehydrogenase to modi¢cation of the enzyme by N-ethylmaleimide, J. Biol. Chem. 252 (1977) 6038^6041. [13] R. Jaenicke, C. Noth, Structure-function relationship of mitochondrial malate dehydrogenase at high dilution and in multicomponent systems, Biol. Chem. Hoppe Seyler 368 (1987) 871^878. [14] N.L. Klyachko, P.A. Levashov, A.V. Levashov, C. Balny, Pressure regulation of malic dehydrogenase in reversed micelles, Biochem. Biophys. Res. Commun. 254 (1999) 685^ 688. [15] W.A. Jensen, J.M. Armstron, J. De Giorgio, M.T. Hearn, Stability studies on pig heart mitochondrial malate dehydrogenase: the e¡ect of salts and amino acids, Biochem. Biophys. Acta 1296 (1996) 23^34. [16] C. Frieden, J. Honegger, H.R. Gilbert, Malate dehydrogenases. The lack of evidence for dissociation of the dimeric enzyme in kinetic analyses, J. Biol. Chem. 253 (1978) 816^ 820. [17] D.M. Jameson, V. Thomas, D.M. Zhou, Time-resolved £uorescence studies on NADH bound to mitochondrial malate dehydrogenase, Biochem. Biophys. Acta 994 (1989) 187^ 190. [18] D.M. Bleile, R.A. Schulz, J.H. Harrison, E.M. Gregory, Investigation of the subunit interactions in malate dehydrogenase, J. Biol. Chem. 252 (1977) 755^758. [19] D.C. Wood, S.R. Jurgensen, J.C. Geesin, J.H. Harrison, Subunit interactions in mitochondrial malate dehydrogenase. Kinetics and mechanism of reassociation, J. Biol. Chem. 256 (1981) 2377^2382. [20] S.R. Jurgensen, D.C. Wood, J.C. Mahler, J.H. Harrison, The immobilization of mitochondrial malate dehydrogenase on Sepharose beads and the demonstration of catalytically active subunits, J. Biol. Chem. 256 (1981) 2383^2388. [21] G. DuVal, H.E. Swaisgood, H.R. Horton, Some kinetic characteristics of immobilized protomers and native dimers of mitochondrial malate dehydrogenase: an examination of the enzyme mechanism, Biochemistry 24 (1985) 2067^2072. [22] A.J. McEvily, T.R. Mullinax, D.R. Dulin, J.H. Harrison, Regulation of mitochondrial malate dehydrogenase: kinetic modulation independent of subunit interaction, Arch. Biochem. Biophys. 238 (1985) 229^236.

BBAPRO 36116 11-5-00

256

S. Shaw et al. / Biochimica et Biophysica Acta 1478 (2000) 248^256

[23] R.B. McComb, L.W. Bond, R.W. Burnett, R.C. Keech, G.N. Bowers, Determination of molar absorptivity of NADH, Clin. Chem. 22 (1976) 141^150. [24] G.F. Azzone, R. Colonna, B. Ziche, Preparation of bovine heart mitochondria in high yield, Methods Enzymol. 55 (1979) 46^50. [25] B.A. Humphries, M.S. Rohrbach, J.H. Harrison, 4,4P-Bis dimethylamino diphenylcarbinol: a new reagent for selective chemical modi¢cation. Interaction with porcine malate dehydrogenase, Biochem. Biophys. Res. Commun. 50 (1973) 493^499.

[26] G.E. Tiller, T.J. Mueller, M.E. Dockter, W.G. Struve, Hydrogenation of Triton X-100 eliminates its £uorescence and ultraviolet light absorption while preserving its detergent properties, Anal. Biochem. 141 (1984) 262^266. [27] J.J. Birktoft, R.A. Bradshaw, L.J. Banaszak, Structure of porcine heart cytoplasmic malate dehydrogenase: combining X-ray di¡raction and chemical sequence data in structural studies, Biochemistry 26 (1987) 2722^2734. [28] S.L. Roderick, L.J. Banaszak, The three-dimensional structure of porcine heart mitochondrial malate dehydrogenase at î resolution, J. Biol. Chem. 261 (1986) 9461^9464. 3.0-A

BBAPRO 36116 11-5-00