ARCHIVES

Vol.

OF BIOCHEMISTRY

290, No. 1, October,

AND

BIOPHYSICS

pp. 21-26,

1991

The Effect of Ligand Binding on the Proteolytic Pattern of Methylmalonate Semialdehyde Dehydrogenase’ Natalia Y. Kedishvili,2 Department 635 Barnhill

Received

Kirill

M. Popov, and Robert A. Harris

Biochemistry and Molecular Biology, Indiana Drive, Indianapolis, Indiana 46202-5122

of

February

25, 1991, and in revised

form

May

University

Methylmalonate semialdehyde dehydrogenase (EC 1.2.1.27) (MMSDH)3 was recently purified from rat liver i This work was supported by grants from the U.S. Public Health Services (NIH DK 40441). the Diabetes Research and Training Center of Indiana University School of Medicine (AM 20542), and the Grace M. Showalter Residuary Trust. The work was also supported by fellowships (N.Y.K. and K.M.P.) from the American Heart Association, Indiana Affiliate, Inc. s To whom correspondence and requests for reprints should be addressed. 3 Abbreviations used: MMSDH, methylmalonate semialdehyde dehydrogenase; ADH, alcohol dehydrogenase; ALDH, aldehyde dehydrogenase; SDS-PAGE, polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate; DTT, dithiothreitol; DB, digestion buffer.

Copyright All

rights

$3.00 19 1991 by Academic Press, of reproduction in any form

of

Medicine,

24, 1991

Native rat liver methylmalonate semialdehyde dehydrogenase was proteolyzed by lysylendopeptidase C, chymotrypsin, and trypsin to generate different cleavage fragments of molecular masses: 50,8,55,44,39,53,45, and 40 kDa. A proteolytic cleavage map of MMSDH was constructed based on sequencing data and a comparison of appearance and degradation rates of the different protein fragments as shown by SDS-PAGE. NAD+ was highly effective as a protector against proteolysis in both the N-terminal and the C-terminal parts of the intact enzyme. NADH did not efficiently protect the intact enzyme: however, it stabilized proteolytic fragment L50 from further degradation. This suggests that the NAD+-binding domain is not destroyed by cleavage of the N-terminal part of MMSDH. CoA had no effect on the proteolytic cleavage patterns of MMSDH. However, CoA esters reduced the protective effect of NAD+ with an order of effectiveness of acetyl-CoA > propionyl-CoA > butyrylCoA. p-Nitrophenyl acetate, substrate for esterase activity by the enzyme, partially prevented the protective effect of NAD+ against proteolysis. These results suggest that S-acylation of the enzyme prevents a stabilizing conformational change induced in MMSDH by NAD+ binding. 8 1991 Academic Press, Inc.

0003.9861/91

School

in order to define the distal portion of valine catabolism and related pathways in mammals (1). The monomer molecular mass, determined by SDS-PAGE, was 58 kDa. The native molecular mass, estimated by gel filtration, was 250 kDa, suggesting a tetrameric structure. The purified enzyme was active with malonate semialdehyde and consumed both stereoisomers of methylmalonate semialdehyde, implicating a single semialdehyde dehydrogenase in the catabolism of valine, thymine, and compounds catabolized by way of @-alanine. This enzyme is not a simple aldehyde dehydrogenase because the reaction is CoA-dependent, the products being acetyl-CoA and propionyl-CoA from malonate semialdehyde and methylmalonate semialdehyde, respectively. The formation of butyryl-CoA from ethylmalonate semialdehyde, although expected, was not found (1). Although some kinetic studies have been carried out (l), little is known yet about the structure, function, and catalytic mechanism of MMSDH. Rat liver MMSDH was recently shown to possess the ability to hydrolyze p-nitrophenyl acetate with concomitant formation of an Sacyl enzyme (Popov, K. M., Kedishvili, N. Y., and Harris, R. A., in preparation). Deacylation of the S-acyl enzyme intermediate was the rate-limiting step of the esterase reaction, and the activity was regulated by pyridine nucleotides (NAD+, NADH) and CoA. Kinetic studies suggest that coenzyme binding in the active site of MMSDH may induce a conformational change of the enzyme that affects the accessibility of an enzyme-thioester intermediate for deacylation. For the above reasons, we decided to study the influence of various components of the MMSDH reaction on the structure of rat liver methylmalonate semialdehyde dehydrogenase by the method of limited proteolysis. MATERIALS

AND

Materials. NAD+, alcohol dehydrogenase MO). Trypsin (treated

METHODS NADH, p-nitrophenyl acetate, CoA esters, and (ADH) were from Sigma Chemical (St. Louis, with L-1-tosylamido-2-phenylethyl chloromethyl 21

Inc. reserved.

22

KEDISHVILI,

POPOV,

ketone) and chymotrypsin (treated with N”-tosyl-L-lysine chloromethyl ketone) were from Worthington Diagnostic Systems (Freehold, NJ). Lysylendopeptidase C was a product of Calbiochem (San Diego, CA). PVDF membranes were from Millipore (Bedford, MA). Sources of other chemicals used were given previously (1). Rat liver MMSDH was prepared and stored Preparation of MMSDH. as previously described (1) with modifications given below. Frozen liver (300 g), maintained at -70°C was allowed to partially thaw at 4°C then was homogenized at the high speed setting of a Waring blender for 1 min in 4 vol of buffer A (20 mM ammonium acetate, pH 7.5 at 4’C, 0.1 mM EDTA, 2 mM DTT, 0.5 mM NAD+) supplemented with protease inhibitors (1). The suspension was further homogenized in five portions with a Polytron PT 10 homogenizer at a setting of 4 for 1 min. The pH was adjusted to 7.5 at 4’C. Insoluble material was removed by centrifugation at 100,OOOgfor 60 min. The clear supernatant was carefully decanted, the pH adjusted to 6.5 at 4’C with acetic acid, and the extract mixed with 600 ml of CM-Sepharose equilibrated with buffer A, pH 6.5, at 4°C. The slurry was stirred gently for 30 min and then unbound material (containing the MMSDH) was removed by filtration. The CMSepharose was washed twice with 1 vol of buffer A. Filtrates were combined, adjusted to pH 7.0 at 4°C with NH,OH, and mixed with DEAESephacel(800 ml) equilibrated with buffer A, pH 7.0, at 4°C. The slurry was mixed for 30 min and unbound material containing MMSDH was removed by filtration. The DEAE-Sephacel was washed three times with buffer A. All washes were combined, and the pH was adjusted to 7.5 at 4°C with NH,OH. This extract was applied at a flow rate of 60-80 ml/ h to an hydroxyapatite column (2.5 X 20 cm) equilibrated with buffer B (25 mM potassium phosphate, pH 7.5, 0.1 mM EDTA, 2 mM DTT, 0.5 mM NAD’, 10% glycerol). MMSDH was eluted with a linear gradient of potassium phosphate (total volume 500 ml) from 100 to 300 mM prepared in buffer B. The enzyme solution was concentrated to a volume of lo-20 ml under N2 pressure with a YM-10 membrane and applied at a flow rate of 50 ml/h to a Sephacryl S-300 column (2.5 X 95 cm) equilibrated with buffer A (pH 7.5 at 4°C). Fractions containing MMSDH were pooled, pH adjusted to 6.0 at 4’C with acetic acid, and the extract applied to a S-Sepharose fast flow column (1.5 X 10 cm) equilibrated with buffer A (pH 6.0) with 10% glycerol. In the presence of NAD+, MMSDH does not bind to S-Sepharose and elutes in the void volume, whereas most other proteins remain bound. The enzyme obtained was concentrated on a phenyl-Sepharose column, dialyzed against buffer C (10 mM TrisHCl, pH 8.0, 0.1 mM EDTA, 2 mM DTT, 0.5 mM NAD+), aliquoted, and stored as described previously (1). Ten milligrams of the enzyme protein can be purified from 100 g of rat liver with a specific activity of 7-9 units (pmol/min) per milligram of protein measured with malonate semialdehyde as substrate. Enzyme activity was routinely measured by following the reduction of NAD+ at 340 nm with a cocktail consisting of 30 mM sodium pyrophosphate, pH 8, with HCl at room temperature, 2 mM dithiothreitol, 2 mM NAD+, 0.5 mM CoA, and 0.5 mM malonate semialdehyde. Reactions were initiated with enzyme. In some studies, enzyme activity was measured by a coupled assay based on the generation of methylmalonate semialdehyde from L-0.hydroxyisobutyrate by /3-hydroxyisobutyrate dehydrogenase (1). Apoenzyme preparation. MMSDH at a concentration of 1-2 mg/ml in buffer C was supplemented with (NH&SO, to 2 M, incubated 15 min at room temperature, and applied at a flow rate of 1 ml/h to a Sephadex G-25 fast flow column equilibrated with 15 mM potassium phosphate, pH 7.8,0.1 mM EDTA, and 1 mM DTT. The enzyme obtained was used directly for proteolysis studies. This procedure efficiently removed NAD+ from the enzyme. To measure the residual amount of NAD+ in MMSDH, 1 mg of enzyme was precipitated with 6% (w/v final concentration) perchloric acid, the extract neutralized with potassium hydroxide, and NAD+ measured with an enzymatic end-point assay (6). Less than 0.05 mol of NAD+ per mole of enzyme was found. Proteolysis. MMSDH at a final concentration of 50 pg/ml in a buffer 10 mM in potassium phosphate, pH 7.5, 0.1 mM in EDTA, and 0.1 mM

AND

HARRIS

in DTT (DB, digestion buffer) was treated with the indicated proteases for specific lengths of time. The incubation temperature was 3O’C. For the study of the effects of ligands, MMSDH in DB was preincubated with saturating concentrations of individual ligands for 3-5 min on ice and then treated with indicated proteases for 5-60 min at 3OOC.Activity measurements were done immediately as described above. Samples for SDS-PAGE analysis were boiled for 5 min with buffer containing 2% SDS, 5% mercaptoethanol, and 10% glycerol (final concentrations) and kept on ice until all time points were taken. Digestion with trypsin was stopped with trypsin inhibitor (10X excess over trypsin concentration) in some experiments. Gel electrophoresis and densitometry. Proteolytic fragments were analyzed by SDS-PAGE with 12 and 4% acrylamide in the running and stacking gels, respectively (7). Gels were stained with silver as described by Oakley et al. (8) and photographed. Gel scanning was carried out with a Hoeffer 350 densitometer. A linear dependence of MMSDH loaded and the amount determined by densitometry was confirmed. N-Terminal sequence analysis. Proteolytic fragments separated by SDS-PAGE were blotted onto PVDF membranes and stained with Coomassie R-250. Peptide bands were cut from the membrane and the peptides sequenced on an Applied Biosystems 477A protein sequencer as described previously (1).

RESULTS

Digestion of Substrate-Free

MMSDH

Proteolytic patterns of MMSDH digested with three different residue-specific endopeptidases-lysylendopeptidase C, chymotrypsin, and trypsin-are shown in Fig. 1 (A, B, and C, respectively). Lysylendopeptidase C generated increasing amounts of two major fragments of 50 (L50) and 8 kDa (LB) during the 60 min of incubation, while the amount of the native 58-kDa MMSDH decreased. With chymotrypsin a 55-kDa (C55) product appeared after 5 min of incubation. Its amount increased significantly during the first 30 min and then decreased while products of lower molecular weights (44 and 39 kDa; C44 and C39, respectively) appeared. The trypsinolysis picture was more complicated. A 53-kDa product (T53)

B

A

- L50

C

-c55

-T53

-c44 -c39

1;g

- L8

FIG. 1. Limited proteolysis patterns of MMSDH with different proteases. (A) Lysylendopeptidase C, (B) chymotrypsin, (C) trypsin. Protease:MMSDH ratio 1:300; time points 5, 30, and 60 min; 30°C. The low molecular weight band present in all lanes of C is due to the presence of trypsin inhibitor.

RAT

LIVER

METHYLMALONATE

,ssss

I MMSDH

,NMNLYS

I

I

L50

L8

,ssss

1

,ssss

I , NMNLY S

c55

I

T53

1

T45

I

T40

FIG. 2. Proteolytic cleavage map of MMSDH. Scheme is based on data given in Fig. 1. N-terminal sequence of MMSDH: SSSSVPTVKLFINGKFVQ; N-terminal sequence of L50: NMNLYSYRLPLGVCAGIAPFNFPAG.

appeared at the fifth minute of incubation and was further increased at 30 min but decreased after 60 min. A 45kDa (T45) fragment was apparent after 30 min but was then lost with the appearance of a 40-kDa (T40) fragment. Other minor proteolytic bands could be seen on the gels but no attempt was made to follow them. The native MMSDH and proteolytic fragments L50, C&5, T53, and T40 were transferred to a PVDF membrane by electroblotting and subjected to N-terminal amino acid sequencing. The analysis indicated that the amino terminal parts of the native enzyme and proteolytic fragments C55 and T53 were identical, suggesting that the cleavage sites of both chymotrypsin and trypsin are located in the very end of the C-terminus of MMSDH (Fig. 2). The N-terminal sequences of peptides L50 and T40 were also identical but different from the N-terminus of the native MMSDH. Thus, if lysylendopeptidase C cleaves MMSDH only in its N-terminal part, trypsin must first attack the C-terminus and subsequently the N-terminal part of MMSDH. The loss of MMSDH activity resulting from the cleavage of MMSDH with lysylendopeptidase C is shown in Fig. 3. The data clearly demonstrate that the decrease of MMSDH activity correlates with the decrease in quantity of the 5%kDa band corresponding to the protein of the native MMSDH. The same correlation was observed for chymotrypsinolysis of MMSDH (data not shown). Thus, cleavage in both the N-terminal part of MMSDH (lysylendopeptidase C) and the C-terminal part leads to the loss of enzyme activity. Ligand-Bound

MMSDH

Digestion

The proteolytic patterns peptidase C, chymotrypsin,

of MMSDH and trypsin

with lysylendoin the presence

SEMIALDEHYDE

23

DEHYDROGENASE

of various components of MMSDH reaction (NAD’, NADH, CoA, and propionyl-CoA) are shown in Fig. 4. The primary substrates for MMSDH activity-malonate semialdehyde and methylmalonate semialdehyde-are unstable and, therefore, were not used in these experiments. All ligands were used at saturating conditions as previously determined. The same fragmentation as that found with the MMSDH alone was observed with each endopeptidase. However, the time courses of proteolysis were altered. These alterations are specific for MMSDH, i.e., not due to direct effects of the compounds on the proteases, since no such changes were observed in identical experiments using bovine serum albumin in place of MMSDH (data not shown). Furthermore, the patterns of fragmentation were not altered when the proteolysis was done at 4 and 25°C. When NAD+ was present (Fig. 4, lanes 3,4), the amount of 58-kDa band corresponding to native MMSDH remained almost unchanged during 60 min of incubation with three different proteases. In contrast, significant reduction in the amount of 58-kDa band was apparent when the enzyme was incubated with these proteases without NADf (Fig. 4, lanes 1, 2). Thus, NAD+ very effectively protects MMSDH against proteolysis. With NADH the enzyme was cleaved slower than the no addition condition but faster than when NAD’ was present (Fig. 4, lanes 5, 6). CoA (Fig. 4, lanes 7, 8) and propionyl-CoA (Fig. 4, lanes 9,10) had almost no effect on the proteolytic pattern of MMSDH when added individually.

0’0

30

6

TIME (Mid FIG. 3. Effect of lysylendopeptidase C on MMSDH activity and mass of intact MMSDH. (0) Percentage of remaining MMSDH activity; (0) MMSDH mass. Protease:MMSDH ratio 1:300, 30°C. Specific activity of MMSDH 7.8 units/mg protein; 0.5 pg of enzyme protein used for activity measurement. Each point represents the average of two independent experiments. Residual 5%kDa protein was quantitated as described under Materials and Methods. The amount of enzyme activity loss over the same period of time in the absence of proteolysis was less than 5%.

24

KEDISHVILI.

POPOV,

As shown in previous work (l), the best substrate for MMSDH is malonate semialdehyde and the next best is methylmalonate semialdehyde. Ethylmalonate semialdehyde is either not a substrate or a very poor substrate for MMSDH. The CoA esters that the enzyme makes from these compounds (acetyl-CoA, propionyl-Cob, and butyryl-CoA, respectively) were studied with respect to their relative effects, with and without NAD+, on the proteolysis of native MMSDH. The enzyme was preincubated for 510 min with 1 mM NAD+ on ice and then CoA or the CoA-esters were added and the reaction followed for 60 min. The results were similar for all three proteases; therefore, only the data obtained with trypsin are discussed. The proteolytic pattern of native MMSDH with NAD+ plus CoA (Fig. 5A, lanes 3,4) was the same as with NAD+ alone (Fig. 5A, lanes 1, 2), indicating no effect of CoA on the protection exerted by NAD+. In contrast, acetyl-CoA completely blocked the protective effect of NAD+ (Fig. 5A, lanes 5, 6). Propionyl-CoA was less effective (Fig. 5A, lanes 7, 8) and butyryl-CoA even less effective in blocking the protective effect of NAD+ against proteolysis (Fig. 5A, lanes 9, 10). MMSDH has recently been found to have esterase activity toward p-nitrophenyl acetate (Popov, K. M., Kedishvili, N. Y., and Harris, R. A., in preparation). An S-acyl enzyme (thioester) is formed as an intermediate of the reaction catalyzed by MMSDH. It was of interest, A

AND

HARRIS 0

A

-T ,T ‘T

123

4

5

6

7

8

9

10

-c

53 45 40

1

2

55

34

FIG. 5. (A) Tryptic cleavage of MMSDH in the presence of NAD+ and different CoA esters. Trypsin:MMSDH 1:300,3O”C; time points 20 and 60 min. Ligands (1 mM) added: 1,2-NAD+; 3,4-NAD+ and CoA; 5, 6-NADf and acetyl-CoA; 7, 8-NAD’ and propionyl-CoA, 9, loNAD’ and butyryl-CoA. (B) Chymotryptic cleavage of MMSDH in the presence of NAD+ and p-nitrophenyl acetate. Chymotrypsin:MMSDH ratio 1:30, 30°C; lanes l-3-NAD+ (1 mM) and p-nitrophenyl acetate (1 mM): 5, 30, and 60 min; lane 4-NAD+ only with 60 min incubation.

therefore, to use p-nitrophenyl acetate to promote the formation of the S-acyl enzyme form of MMSDH and then determine whether this influences the protective effect of NAD+ on the proteolytic susceptibility of MMSDH. The enzyme was preincubated with p-nitrophenylacetate for 5-10 min at room temperature, i.e., conditions shown in other studies to result in the stoichiometric acetylation of the enzyme. NAD+ was then added and 5 min later proteolysis initiated with the addition of one of the proteases. Figure 5B gives the proteolytic pattern of S-acetyl MMSDH produced by chymotrypsin in the presence of NAD+. The pattern of degradation of the 58-kDa band was very similar to that obtained for MMSDH enzyme in the presence of NAD+ and acetyl-CoA or propionyl-CoA. Thus, acetylation of the enzyme with p-nitrophenylacetate has the same effect as acetyl-CoA and propionyl-CoA on the protective effect exerted by NAD+. Similar results were obtained with trypsin. DISCUSSION

Structural

-T

1

2

3

4

5

6

7

8

53

910

FIG. 4. Proteolytic patterns of MMSDH in the presence of different ligands. (A) Lysylendopeptidase C, (B) chymotrypsin, (C) trypsin. Protease:MMSDH ratio 1:30, 30°C; time points 30 and 60 min. Ligands (1 mM) added: 1, Z-none; 3,4-NAD+; 5,6-NADH; 7,8-CoA; 9, lopropionyl-CoA. In A, lanes l-6 and lanes 7-10 for lysylendopeptidase C correspond to different gels but identical experiments.

Organization

Digestion of rat liver MMSDH with lysylendopeptidase C generates mainly one large proteolytic fragment of molecular mass 50 kDa and a small peptide of 8 kDa (L50 and L8, respectively). Chymotrypsin generates three fragments: 55, 44, and 39 kDa (C55, C44, C39). Tryptic digestion produces fragments of 53, 45, and 40 kDa (T53, T45, T40). On the basis of a combination of peptide sequence analysis and relative fragment sizes, a proteolytic map for MMSDH can be constructed. Although the T45 protein and L8 peptide were not sequenced, this is not necessary because it can be deduced how T45 is formed, and L8 likely derives from the same

RAT

LIVER

METHYLMALONATE

cleavage as L50 and represents the N-terminal part of MMSDH. To determine the position of T45, it is proposed that all tryptic fragments bear a precursor-product relationship and that each appears as a result of a single split from its precursor. The simplest explanation is that T40 is produced from T53 as a result of two sequential splits with the intermediate formation of T45. One of these splits must occur in the N-terminal part of T53 because the N-terminal sequences of T53 and T40 differ. T40 has an N-terminal sequence identical to that of L50. This means that T40 lacks an 8-kDa peptide from the N-terminus exactly like L50. The difference in molecular weight between T53 and T45 is 8 kDa, so apparently the first split occurs at an N-terminal cleavage site common for trypsin and lysylendopeptidase (after lysine). The

ALDH: MMSDH:

SEMIALDEHYDE

second split then occurs in the C-terminal part of T45 and produces T40. An interesting observation is that in spite of similar substrate specificities, trypsin and lysylendopeptidase C cleave MMSDH in different parts of the molecule. The simplest explanation is that an arginine residue which creates a cleavage site for trypsin but not lysylendopeptidase C must be present in the C-terminal part of MMSDH. From the sequencing data available at this time we have also found that the N-terminal sequence of proteolytic products L50 and T40 has 64% homology with the sequence of rat and bovine liver mitochondrial aldehyde dehydrogenases (ALDH) in the region of Cys-162 (5). The other amino acids of this domain in MMSDH are conservative substitutions:

162 I -Pro-Val-Gly-Val-Cys-Gly-Gln-IleIle -Pro-TrpAsn-Phe-ProI I I I I I I I -Pro-Leu-Gly-Val-Cys-Ala-Gly-Ile-Ala-Pro-Phe-Asn-Phe-Pro-

The only cleavage site for lysylendopeptidase C is located close to this region and cleavage at this site leads to the loss of enzyme activity. This sequence represents a conserved, strongly hydrophobic segment from Pro-158 to Met-174 of ALDH (9). At present, however, there is no direct evidence for a function of Cys-162 in the catalytic mechanism of ALDH. Since proteolytic cleavage occurred in good yield at similar sites in ALDH and MMSDH, this region may play an important role in stabilization of enzyme structure (9-11). Our study indicates that the middle portion of MMSDH is quite resistant to proteolysis, suggesting that this region may form a stable tertiary structure. The proteolytic study of ligand-bound MMSDH showed that the L50 fragment was stabilized by NADH and partially by propionyl-CoA from further degradation. This indicates that the pyridine nucleotide binding domain was not affected by the loss of the 8-kDa N-terminal peptide of MMSDH. The T53 fragment which lacks the C-terminal 5 kDa peptide of intact MMSDH was not protected by pyridine nucleotides, suggesting either than the NADH binding site has been lost from T53 or that the subsequent site of tryptic cleavage was not protected by NADH binding. The C55 fragment has a C-terminus longer than T53 by 2 kDa and was slightly protected by NADH. To confirm that C55 must bind pyridine nucleotides, we preincubated MMSDH alone with chymotrypsin for 5 min, then added NAD+, and followed the pattern of proteolysis for the next 55 min. The rate of C55 cleavage in the presence of

25

DEHYDROGENASE

I

NAD+ was slower than in the no NAD+ control (data not shown). Thus, the data are consistent with the hypothesis that the pyridine nucleotide binding domain is located close to the C-terminal end of MMSDH, whereas the catalytic domain likely includes the N-terminus of MMSDH. Considerably more work will be necessary to confirm/ refuse this hypothesis. Digestion of Ligand Bound MMSDH. NAD+ is a highly effective protector of MMSDH against cleavage by various proteases. CoA has no effect on the proteolytic patterns of MMSDH. However, CoA esters added to NAD+-protected MMSDH reverse the protective effect of NAD+. The efficiency of this action decreases from acetyl-CoA to butyryl-CoA, probably because the increasing size of the side chains of these compounds affects their binding in the active site of MMSDH. S-Acetyl MMSDH is formed as a relatively stable intermediate when MMSDH is incubated with p-nitrophenyl acetate (Popov, K. M., Kedishvili, N. Y., and Harris, R. A., in preparation). The acetylated form of the enzyme shows the same pattern of proteolysis in the presence of NAD’ as the NAD+-bound enzyme in the presence of CoA esters. Thus, the reduced protection of MMSDH by NAD+ in the presence of CoA esters is likely the consequence of the formation of an acylated form of the enzyme.

The results of this study suggest that NAD+ binding induces a conformational state of MMSDH more stable

26

KEDISHVILI,

POPOV,

to proteolysis than that of MMSDH alone. Taking into account the antagonistic effects of p-nitrophenyl acetate versus NAD+ on the susceptibility of MMSDH to proteolysis, we conclude that the S-acylated enzyme differs from the nonacylated enzyme in its ability to undergo a conformational change in response to NAD’ binding. Work currently in progress on the catalytic mechanism of this enzyme should help clarify this point.

AND

HARRIS

4. Sidhu, R. S., and Blair, A. H. (1975) J. Biol. Chem. 250, 78947898. 5. Farres, J., Guan, K.-L., and Weiner, H. (1989) Eur. J. Biochem.

180,67-74. 6. Klingenberg, M. (1974) in Methods of Enzymatic Analysis, (Bergmeyer, H. U., Ed.), Vol. 4, p. 2048, Academic Press, New York. 7. Laemmli, U. K. (1970) Nature (London) 227, 680-685. 8. Oakley, B. R., Kirsch, D. R., and Morris, N. R. (1980) Anal. Biochem.

105,361-363. 9. Hempel, J., Kaiser, R., and Jornvall,

REFERENCES 1. Goodwin,

P. M., Davis, E. J., and Harris, (1989) J. Bid. Chem. 264, 14,965-14,971.

G. W., Rougraff,

2. Park, J. H., Meriwether, B. P., Clodfelder, L. W. (1961) J. Biol. Chem. 236, 136-141. 3. Feldman,

272.

R. I., and Weiner,

H. (1985) Eur. J. Biochem.

153,13-28. R. A.

P., and Cunningham,

H. (1972) J. Biol. Chem. 247,

267-

10. van Bahr-Lindstrom, H., Hempel, J., and Jornvall, H. (1984) Eur. 141,37-42. J.&o&em. 11. Hempel, J., van Bahr-Lindstrom, H., and Jornvall, H. (1984) Eur. J. Biochem. 141,21-35. 12. Rossman, M. G., Liljas, A., Branden, G.-I., and Banaszak, L. J. (1975) in The Enzymes (Bayer, P., Ed.), Vol. 11, p. 64, Academic Press, New York.