Action of surfactants on porcine heart malate dehydrogenase isoenzymes and a simple method for the differential assay of these isoenzymes

Bioehimica et Biophysica Acta

884 (1986) 109-118

109

Elsevier BBA 22515

Action of sudactants on porcine heart malate dehydrogenase isoenzymes and a simple method for the differential assay of these isoenzymes Keith Smith * and Trichur K. Sundaram Department of Biochemistry and Applied Molecular Biology, Unioersityof Manchester Institute of Science and Technology, Manchester, M60 1QD (U.K.)

(Received29 May 1986)

Key words: Malate dehydrogenase; Surfactant action; Isoenzyme inactivation; Enzyme assay; (Porcine heart)

The cationic surfactant, cetyl (hexadecyl) trimethylammonium bromide (CTAB), completely inactivates porcine heart cytoplasmic malate dehydrogenase (L-malate:NAD + oxidoreductase, EC 1.1.1.37) at concentrations (of sudactant) which do not affect the activity of the mitochondrial isoenzyme. These concentrations are close to, or higher than, the critical mieelle concentration of CTAB. An increase in the ionic strength of the medium significantly retards the CTAB-indnced inactivation of the cytoplasmic enzyme. The enzyme is also markedly protected against CTAB inactivation by NADH; L-mulate on its own has no effect but a combination of NADH and L-malate affords greater protection than NADH alone. The CTAB inactivation is not reversed by dilution of the sudactant. The highly selective action of CTAB on the two malate dehydrogenases, which correlates well with their electrostatic charges, has been exploited for a simple and reliable differential assay of these isoenzymes. The anionic sudactant, sodium dodecyl sulphate (SDS), at concentrations well below the critical micelle concentration, inactivates both isoenzymes, but the mitochondrial enzyme is significantly more sensitive than its cytoplasmic counterpart. There is thus some correlation, though not as strong as with CTAB, between SDS inactivation and the charges of the two malate dehydrogenases. An increase in ionic strength has opposite effects on the two isoenzymes: the mitochondrlal enzyme becomes more resistant and the cytoplasmic enzyme less so. Both isoenzymes are rendered more resistant to SDS by the inclusion of NADH. Inactivation of the enzymes caused by short exposure to SDS is largely reversed by dilution of the detergent, but longer exposure leads to progressive irreversible loss of activity. NADH very effectively protects the isoenzymes against irreversible inactivation. It is likely that a reversible phase of inactivation precedes an irreversible phase and that in the former phase SDS acts competitively with NADH. Both malate dehydrogenases possess considerable resistance to the nonionic detergent, Triton X-100. Introduction The study of the action of surfactants on enzymes has been considered important for an un* Present address: Royal College of Surgeons of England. Dental Research Unit, West Hill, Downe Orpington, BR6 7JJ, U.K. Correspondence address: Dr. Trichur Sundaram, Department of Biochemistryand Applied Molecular Biology,Universityof Manchester Institute of Science and Technology,Manchester, M60 1QD, U.K.

derstanding of the interaction between proteins and lipids in biological membranes. Despite the plethora of literature on protein-surfactant interactions, their intimate details such as the nature of the binding forces involved are not completely clear, although ionic and hydrophobic bondings have both been implicated [1]. It appears, however, that cationic surfactants can have clearly distinct effects on the activity and structure of proteins from the effects produced by anionic

0304-4165/86/$03.50 © 1986 ElsevierSciencePublishers B.V. (Biomedical Division)

110 surfactants [1-3]. This is illustrated by the interesting finding of Sanford et al. [1] that the surfactant-induced inactivation of porcine lactate dehydrogenases correlated with the electrostatic charges on the ionic surfactants and the isoenzymes; cationic surfactants preferentially inactivated isoenzymes with a net negative charge and anionic surfactants preferentially inactivated isoenzymes having a net positive charge. We have carried out a further test of such specificity of protein-surfactant interaction by studying the relative susceptibilities of the two porcine heart malate dehydrogenase (L-malate: N A D ÷ oxidoreductase, EC 1.1.1.37) isoenzymes to ionic and nonionic surfactants. Eukaryotic cells generally possess two malate dehydrogenases: a cytoplasmic enzyme that is a component of the malate-aspartate shuttle and a mitochondrial enzyme that functions in this shuttle and in the tricarboxylic acid cycle [4,5]. Although similar in size and structure [6], the mitochondrial and cytoplasmic isoenzymes of porcine heart widely differ in charge density, having isoelectric points of 10 and 5.1, respectively [7]. This system was therefore eminently suitable for an appraisal of the specificity of surfactant action mentioned above. We present the results of such a study and a simple method, arising from it, for the differential assay of the two malate dehydrogenase isoenzymes. Materials and Methods

Materials. N A D H , N A D ÷ and porcine heart cytoplasmic malate dehydrogenase were obtained from Boehringer Corporation; oxaloacetic acid from Calbiochem-Behring Corporation; L-malate, Triton X-100 and porcine heart acetone powder from Sigma London Chemical Company Ltd.; cetyl (hexadecyl) trimethylammonium bromide (CTAB) and sodium dodecyl sulphate (SDS), specially pure, from BDH Chemicals Ltd. Porcine heart mitochondrial malate dehydrogenase was isolated from the heart acetone powder as described previously [8]. Malate dehydrogenase assay. The assay was carried out at 25 ° C in a system which, in addition to enzyme, contained 60 mM sodium/potassium phosphate (pH 7.5), 0.14 mM N A D H and 0.3 mM oxaloacetic acid; the rate of decrease in ab-

sorbance at 340 nm was measured in a recording spectrophotometer. Action of surfactants on malate dehydrogenase. This was studied by two methods. Method 1. The enzyme (0.08 unit) was incubated at 25°C with buffer, surfactant at the appropriate concentration and any desired supplement in a total volume of 0.98 ml in a cuvette for 30 s. The residual enzyme activity was then measured after the addition of 10 #1 14 mM N A D H and 10 /~1 33 mM oxaloacetic acid. The monitoring of the absorbance at 340 nm started within 10 s after the mixing of N A D H and oxaloacetate with the incubated system containing enzyme and surfactant and continued for at least 1 min. Method 2. A higher concentration of enzyme (0.8 unit) was incubated with buffer, surfactant and supplement in a total volume of 1 ml at 25°C and at appropriate time intervals 40 ttl portions were removed and added to 0.96 ml of a mixture containing phosphate buffer, N A D H and oxaloacetic acid for the assay of the residual enzyme activity. Suitable controls (i.e., without surfactant) confirmed that there was no loss of enzyme activity that was not surfactant-dependent in these experiments. The residual enzyme activity following the enzyme-surfactant interaction was calculated from the rate of decrease in absorbance at 340 nm determined by visual fit of the best straight line in the initial part (i.e., within 60 s of the start of the enzyme reaction) of the recorder trace. Results

Action of the cationic detergent, CTAB, on the mitochondrial and cytoplasmic malate dehydrogenases The effect of CTAB on the activity of the two malate dehydrogenases is shown in Fig. 1. Under the conditions employed (60 mM sodium/potassium phosphate buffer, pH 7.5), the cytoplasmic enzyme was strongly inhibited by CTAB, the inhibition being almost complete at a surfactant concentration of 0.5 raM. By contrast, the mitochondrial enzyme was virtually unaffected by CTAB at concentrations of at least up to 1 mM (data shown only up to 0.7 mM). The effect of ionic strength on the CTAB-induced inactivation

111

I00

>_ 80 ~ <

60 4O

I-._1

"n."'

20

.=

0"1

0"2 CTAB

0'3

0"4

05

!

I

0"7

06

p h o s p h a t e b u f f e r d i m i n i s h e d the extent of the s u r f a c t a n t - i n d u c e d inactivation. Similar results were seen ( d a t a n o t shown) when 10 m M p h o s p h a t e buffer was used in the i n c u b a t i o n m i x t u r e a n d the ionic strength was increased with a s u p p l e m e n t of NaCI. This c o n f i r m s that the effect o b s e r v e d in Fig. 2 was caused b y an increase in ionic strength rather t h a n specifically b y an increase in p h o s p h a t e ion c o n c e n t r a t i o n . Inclusion of 0.14 m M N A D H d u r i n g the i n c u b a t i o n of the e n z y m e with C T A B a f f o r d e d some p r o t e c t i o n against inactivation (Fig. 3), b u t L-malate (1 m M ) a l o n e d i d n o t have a n y effect.

CONCENTRATION (mM)

Fig. 1. Action of CTAB on porcine malate dehydrogenase isoenzymes. The enzymes were each incubated in 60 mM sodium/potassium phosphate buffer (pH 7.5) at 25°C for 30 s with CTAB at the concentrations indicated and then assayed for activity by addition of NADH and oxaloacetate (see Method 1). Mitochondrial malate dehydrogenase (e); cytoplasmic malate dehydrogenase (O). of the c y t o p l a s m i c i s o e n z y m e was investigated b y v a r y i n g the m o l a r i t y of the p h o s p h a t e b u f f e r in the b u f f e r - s u r f a c t a n t i n c u b a t i o n mixture. T h e ina c t i v a t i o n profiles thus o b t a i n e d are p r e s e n t e d in Fig. 2. A n increase in the c o n c e n t r a t i o n of the

Irreversibifity of the inactivation of cytoplasmic malate dehydrogenase by CTAB T h e C T A B - i n d u c e d i n a c t i v a t i o n of the cytop l a s m i c isoenzyme was tested for reversibility as follows. I n c u b a t i o n of the e n z y m e (0.08 unit) with 0.2 m M C T A B in 60 m M p h o s p h a t e buffer resulted in the loss of approx. 55% of the original e n z y m e activity (Fig. 1). W h e n a 4 0 / x l p o r t i o n of this mixture was d i l u t e d 25-fold in a 1 ml assay system, there was n o r e a p p e a r a n c e of the lost e n z y m e activity for at least 10 min. T h e C T A B c o n c e n t r a t i o n of 8 # M after the 25-fold d i l u t i o n was too low to have a n y effect on the enzyme

I00

I00

80

,~8o >k--

I'-

.~ 60

I'-

p¢O

ta 40 _>

_>

I'-

ta

..J, 20 IE 0

I

I

0.1

(>2

CTAB

I

0"3 04 O~ CONCENTRATION (mM)

O6

Fig. 2. Effect of phosphate buffer concentration on the inactivation of cytoplasmic malate dehydrogenase by CTAB. The enzyme was incubated in phosphate buffer (pH 7.5) at 25°C for 30 s with CTAB at the concentrations indicated and then assayed for activity as described in Fig. 1. Buffer concentrations: 10 mM (e); 30 mM (A) and 60 mM (O).

0 0.I

0.2

0-3

0'4

0-5

0.6

CTAB CONCENTRATION , (raM)

Fig. 3. Effect of NADH on the CTAB-induced inactivation of cytoplasmic malate dehydrogenase. The enzyme was incubated in 60 mM phosphate buffer (pH 7.5) at 25°C for 30 s with CTAB (O) at the concentrations indicated or with CTAB plus 0.14 mM NADH (e) and then assayed for activity as described in Fig. 1.

112

activity (Fig. 1). The inactivation of the cytoplasmic enzyme by CTAB was thus apparently irreversible. In the above experiment the 25-fold dilution reduced the enzyme concentration to a level at which the accuracy of the assay was in some doubt. A different protocol was therefore used to verify the apparent irreversibility of the CTAB effect. The amount of enzyme incubated with 0.2 mM CTAB in 60 mM phosphate buffer was increased to 0.8 unit/ml. After incubation for 30 s at 25°C, 40 #l of this mixture was assayed for enzyme activity in a 1 ml system. As shown in Fig. 4, a 50% inactivation was observed in this experiment, despite the 25-fold dilution of the CTAB. This degree of inactivation is in good agreement with the 55% inactivation seen when the assay was performed without a significant dilution of CTAB (Fig. 1). This result then confirms that the inactivation of the cytoplasmic malate dehydrogenase by CTAB is irreversible by dilution of the surfactant. In Fig. 4 are also presented the results that were obtained when the enzyme was incubated with CTAB for longer periods. The inactivation continued steadily for about 6 min and thereafter proceeded at a slow rate. As observed during the shorter incubation (Fig. 1), inclusion of N A D H also protected the enzyme against CTAB inactivation during these longer incubations. Although L-malate alone did not afford any protection, a combination of N A D H and L-malate was appreciably more effective than N A D H alone (Fig. 4). With a supplement of 0.56 mM N A D H and 1 mM L-malate, CTAB (0.2 mM) only caused a 20% loss in enzyme activity in 24 rain as against a loss of over 95% without the supplement (Fig. 4).

Differential assay of cytoplasmic and mitochondrial malate dehydrogenases The preferential inactivation of the cytoplasmic enzyme by CTAB, an interesting and hitherto unreported observation, was exploited to develop a method for the differential assay of the cytoplasmic and mitochondrial malate dehydrogenases in a mixture of the two isoenzymes. An extract of pig heart acetone powder in 50 mM s o d i u m / potassium phosphate buffer (pH 7) was prepared as reported earlier [8] and assayed in 60 mM

I00

60

4o

20

0

I

I

4.

1

I

I

|

I

8 12 EXPOSURE TIME

I

16 (rain)

!

20

24

Fig. 4. Activity of cytoplasmic malate dehydrogenase after incubation with CTAB and subsequent dilution of the surfactant. Malate dehydrogenase (0.8 unit) was incubated with 0.2 m M CTAB and the supplement indicated in 60 mM sodium/potassium phosphate buffer (pH 7.5) at 25°C in a total volume of 1 ml. At appropriate time intervals 40 #1 portions were assayed for enzyme activity in a 1 ml system (see Method 2). CTAB (0); CTAB+0.14 mM N A D H ([3); CTAB +0.14 mM N A D H + I mM L-malate (Ill); CTAB+0.56 mM N A D H (zx); CTAB + 0.56 mM N A D H + 1 mM L-malate (A).

phosphate buffer (pH 7.5) for the total (cytoplasmic plus mitochondrial) malate dehydrogenase activity. A second assay was carried out after incubation of the extract with 0.6 mM CTAB for 30 s at 25°C as described in Fig. 1. Under these conditions the cytoplasmic enzyme was selectively and completely inhibited, whereas the mitochondrial enzyme was unaffected (see Fig. 1). From the two assays it was estimated that the mitochondrial isoenzyme accounted for 83% of the malate dehydrogenase activity in the extract. A check of this differential assay was made as follows. The extract, containing approx. 5 m g / m l of protein and 125 units/ml of total enzyme activity, was diluted 5-fold with distilled water. A portion of the diluted extract containing 500 units of activity was passed through a column of DEAE-cellulose (2 × 3 cm; 9.5 ml) equilibrated in 10 mM sodium/potassium phosphate buffer (pH 7). Under these conditions the cytoplasmic enzyme is selectively adsorbed to the anion-exchanger [9]. The column was washed with 30 ml of the equilibration buffer. The malate dehydro-

113 genase activity in the combined effluent (initial effluent plus column washings) constituted 80% of the activity applied to the column and was confirmed to be due to the mitochondrial isoenzyme by electrophoresis in polyacrylamide gel and staining of the gel for malate dehydrogenase activity [8,10]. Under the conditions of electrophoresis employed, the mitochondrial and cytoplasmic isoenzymes are clearly separated and distinguished. The adsorbed malate dehydrogenase was eluted with 30 m M phosphate buffer and found to consist entirely of the cytoplasmic enzyme by polyacrylamide gel electrophoresis and staining of the gel for enzyme activity. The excellent agreement between the results obtained from the selective inhibition by CTAB and the selective adsorption on DEAE-cellulose of the cytoplasmic enzyme establishes the validity of the differential assay of the two isoenzymes using CTAB. Clearly this assay is much simpler and quicker than the one using DEAE-cellulose.

Stabifity of mitochondrial and cytoplasmic malate dehydrogenase isoenzymes to the anionic detergent, SDS The inactivation of the two isoenzymes at various concentrations of SDS is depicted in Fig. 5.

Although both enzymes were inactivated by SDS, the cytoplasmic isoenzyme was significantly more resistant. An increase in the concentration of phosphate buffer had markedly divergent effects on the two isoenzymes; the inactivation of the mitochondrial enzyme was retarded, whereas that of the cytoplasmic enzyme was enhanced (Figs. 6 and 7). These effects were the result of an increase in ionic strength rather than due specifically to a higher phosphate concentration, since similar inactivation profiles were observed for each isoenzyme irrespective of whether the phosphate concentration was raised (to 30 m M or 60 mM from 10 mM) or NaC1 was added (to 10 m M phosphate buffer) (data not shown). Inclusion of N A D H slowed down the inactivation of both isoenzymes (Figs. 8 and 9), but L-malate by itself had no effect.

Reversibifity of the inactivation of mitochondrial malate dehydrogenase by SDS The inactivation of the mitochondrial enzyme by SDS could be reversed in several ways. Incubation of the enzyme for 30 s with 0.1 m M SDS in 10 m M phosphate buffer (pH 7.5) resulted in 85% loss of activity (see Fig. 6). When, immediately after the treatment with SDS, NaCI was added to

I00

IOO

80 >-

>1-

0

4O

,,, 40 _>

>_ I--

k-

20

0

d z0

0.1

0.2 SDS

0.~

0'4

-

O~

=

0.6

CONCENTRATION (raM)

Fig. 5. Inactivation of porcine malate dehydrogenase isoenzymes by SDS. The enzymes were each incubated in 10 mM sodium/potassium phosphate buffer (pH 7.5) at 25oC for 30 s with SDS at the concentrations indicated and then assayed for activity by addition of NADH and oxaloacetate (see Method 1). Mitochondrial malate dehydrogenase (11); cytoplasmic malate dehydrogenase (e).

o

i 0.1

0.2 SDS

0.3

0-4

CONCENTRATION (raM)

Fig. 6. Effect of phosphate buffer concentration on the inactivation of mitochondrial malate dehydrogenase by SDS. The enzyme was incubated in phosphate buffer (pH 7.5) at 25°C for 30 s with SDS at the concentrations indicated and then assayed for activity as described in Fig. 5. Buffer concentrations: 10 m M (11); 30 m M (&); and 60 mM (©).

114

I00

I00

8o

8O

>F-

I'-" ,< w

>I-

6o

"~

60

uu >

40

Io

4O

_>

I.2 ,,, a,

_1

20

~u 20

l¢

0

0"1 SDS

Fig. 7. Effect of phosphate buffer concentration on the inactivation of cytoplasmic malate dehydrogenase by SDS. Experimental details were similar to those given in Fig. 6. Buffer concentrations: 10 mM (0); 30 mM (A); and 60 mM (m).

raise the ionic strength to a level similar to that of 60 mM phosphate buffer, the extent of enzyme inactivation immediately dropped to 50%. The addition of the same amount of NaC1 to a control I00

80

(>1

0

(>2 0"3 (>4 0~5 CONCENTRATION (raM)

0-2

0"3

0-4

05

SOS CONCENTRATION (raM) Fig. 8. Effect of N A D H on the inactivation of mitochondrial malate dehydrogenase by SDS. The enzyme was incubated in 60 m M phosphate buffer (pH 7.5) at 25°C for 30 s with SDS (©) at the concentrations indicated or with SDS plus 0.14 mM N A D H (O) and then assayed for activity as described in Fig. 5.

0.1

0.2 0"3 0"4 0"5 SDS CONCENTRATION (raM)

Fig. 9. Effect of N A D H on the inactivation of cytoplasmic malate dehydrogenase by SDS. Experimental details were similar to those given in Fig. 8. SDS (n); SDS+0.14 mM NADH

(o).

system containing 10 mM phosphate buffer (pH 7.5), enzyme, 0.14 mM N A D H and 0.3 mM oxaloacetic acid caused a 20% enhancement of enzyme activity. Allowing for this activation, NaC1 still produced a significant reversal of the SDS inactivation, which could be attributed to an increase in ionic strength. It is worth noting that the inactivation of the mitochondrial malate dehydrogenase by SDS was retarded at high ionic strength (see above) and that the level to which NaCl restored the enzyme activity following the SDS inactivation was comparable to the activity that remained after the treatment of the enzyme with 0.1 mM SDS for 30 s in 60 mM phosphate buffer (pH 7.5) (see Fig. 6). The SDS inactivation could also be reversed by removal of the surfactant by dialysis. Thus when a sample in which the enzyme was treated with 0.1 mM SDS for 1 min in 10 mM phosphate buffer (over 90% inactivation occurred as a result) was dialyzed for 1 h at 4°C against numerous changes of phosphate buffer, 85% of the original activity was regained. A third way to reverse the SDS inactivation was to dilute out the surfactant, as illustrated by the following experiment. Mitochondrial malate dehydrogenase (0.8 unit) was incubated in 1 ml 60 mM phosphate

115 buffer (pH 7.5) containing 0.15 m M SDS for 1 min at 25°C. This treatment almost completely inactivated the enzyme. When a 40/~1 portion of this mixture was diluted in a 1 ml assay system, about 95% of the original activity reappeared (Fig. 10), suggesting that the 25-fold dilution largely reversed the SDS inactivation. As the duration of the SDS treatment was prolonged, the inactivation became less reversible by dilution. Thus the degree of irreversible inactivation was about 43% after 10 min treatment, 68% after 20 min treatment and 78% after 30 min treatment (Fig. 10). The inclusion of 0.14 m M N A D H almost fully restored the reversibility even after 30 min treatment with SDS. L-Malate, however, had no effect either by itself or in conjunction with N A D H (Fig. 10).

Reversibility of the inactivation of cytoplasmic malate dehydrogenase by SDS The SDS-induced inactivation of the cytoplasmic enzyme was reversible by dilution of the detergent. The enzyme was completely inactivated within 1 min when exposed to 0.6 m M SDS in 60 m M phosphate buffer ( p H 7.5). U p o n dilution of this mixture 25-fold in the assay system, over 90% of the original enzyme activity was recovered (Fig.

100

so

<

I

2

I

I

I

6

I

I0

EXPOSURE

I

I

I

14 TIME

I

18

I

I

22

I

I

26

I

I

30

(min)

Fig. 10. Reversibility of SDS inactivation of mitochondrial malate dehydrogenase. The enzyme (0.8 unit) was incubated at 25°C in 1 ml of 60 mM phosphate buffer (pH 7.5) containing 0.15 mM SDS and the supplement indicated. At appropriate time intervals 40/~l portions were assayed for enzyme activity in a 1 ml system (see Method 2). SDS_+I mM L-malate (O); SDS+0.035 mM NADH (i); SDS+0.035 mM NADH+I mM L-malate(12); SDS+0.14 mM NADH (A).

I00-

.L

~'- ~

:..

±

.

•

~,

_J

~ ao I

0

2

I

I

6

I

I

I

I0 EXPOSURE

I

I

14 TIME

I

18

I

I

22

I

I

26

I

I

30

(rain)

Fig. 11. Reversibilityof SDS inactivation of cytoplasmicmalate dehydrogenase. Experimentaldetails were similar to those given in Fig. 10, the SDS concentration during the incubation of the enzyme with the surfactant being 0.6 mM. SDS + 1 mM Lmalate (0); SDS+0.014 mM NADH+I mM L-malate (12); SDS + 0.028 mM NADH (A).

11). Longer exposure to SDS resulted in some irreversible inactivation as shown in Fig. 11. Comparison of these results with those obtained with the mitochondrial enzyme (Fig. 10) reveals that the cytoplasmic isoenzyme is more resistant to irreversible inactivation than its mitochondrial counterpart. The greater stability of the cytoplasmic enzyme to SDS was also seen in earlier experiments (Fig. 5). Somewhat surprisingly, the activity recovered upon dilution was virtually constant for durations of SDS treatment of 10-30 min (Fig. 11). As with the mitochondrial malate dehydrogenase, N A D H , but not L-malate, strongly protected the cytoplasmic isoenzyme against irreversible inactivation by SDS (Fig. 11). The SDS inactivation of the cytoplasmic malate dehydrogenase was not reversed, but was accentuated, when, after partial inactivation, the ionic strength was raised by the addition of NaC1. This contrasts with the effect seen with the mitochondrial enzyme but is consistent with the increased sensitivity of the cytoplasmic enzyme to SDS at high ionic strength observed earlier.

Action of the nonionic detergent Triton X-IO0 on the cytoplasmic and mitochondrial malate dehydrogenases The two isoenzymes were each incubated in 1 ml phosphate buffer (10, 30 or 60 mM), p H 7.5,

116 containing 2% ( w / v ) Triton X-100 for 10 min at 25°C, and the activity was determined after the addition of N A D H (0.14 #mol) and oxaloacetic acid (0.3 /~mol). No loss of activity of either enzyme was seen following treatment with the detergent. Discussion The interaction of the nonionic surfactant Triton X-100 with proteins is thought to be mainly hydrophobic in nature and to involve hydrophobic domains of the proteins [11]. The considerable stability of the cytoplasmic and mitochondrial malate dehydrogenases to this surfactant suggests that the inactivation of these isoenzymes by the ionic detergents CTAB and SDS is unlikely to be caused by hydrophobic interactions alone. On the other hand, there is strong support for the notion that ionic interactions play an important role in the inactivation; the cationic surfactant CTAB selectively inactivates the negatively charged cytoplasmic enzyme (pI = 5.1) and the anionic surfactant SDS preferentially, but not completely selectively, inactivates the positively charged mitochondrial enzyme (pI = 10). There are other previous reports of differential action of anionic and cationic surfactants on proteins [1-3]. For example, Sanford et al. [1] demonstrated that lactate dehydrogenase isoenzyme A 4, having a net positive charge at pH 7.4, is rapidly inactivated by anionic surfactants at that pH but is only minimally inactivated by cationic detergents, whereas lactate dehydrogenase isoenzyme B4 with a net negative charge at pH 7.4 is highly sensitive to cationic surfactants and relatively insensitive to anionic surfactants. We have shown that, under appropriate conditions, it is possible to achieve complete inhibition of the cytoplasmic malate dehydrogenase by CTAB without any effect on the activity of the mitochondrial enzyme. This rather remarkable and absolute selectivity of CTAB action forms the basis of a new method for the differential assay of the two malate dehydrogenase isoenzymes. The critical micelle concentration of CTAB in 30 mM sodium phosphate buffer (pH 7.4) at 25°C has been determined to be 0.1 m M [1]. Our experimental conditions (see Fig. 2), especially when 30

mM phosphate buffer was used, closely approximated to the conditions of the critical micelle concentration determination. The results presented in Fig. 2, and in particular the striking increase in the degree of inactivation above a CTAB concentration of 0.1 mM, may therefore be interpreted to mean that CTAB predominantly acts in the micellar form to inactivate the porcine cytoplasmic malate dehydrogenase. There are reports that anionic surfactants bind cooperatively to several proteins at or above their critical micelle concentration [12-14]. Moreover it has been suggested that surfactant ions bind to sites on the protein molecule and unfold it, that further binding of surfactant occurs, often cooperatively, and that the protein is denatured as a result [12,15-18]. It is likely that CTAB inactivates cytoplasmic malate dehydrogenase by a similar mechanism. The diminution of the inactivation at high ionic strength can be explained in terms of a weakened electrostatic interaction between surfactant and enzyme. The binding of N A D H alone or in conjunction with L-malate induces considerable conformational changes in the active site region of malate dehydrogenase [4,6]. The protective effect of N A D H and of N A D H plus L-malate (Figs. 3 and 4) on the cytoplasmic malate dehydrogenase against CTAB may imply that the surfactant binds, at least in part, at or near the active site or that the native structure of the enzyme is stabilized by the substrate(s) as a result of the substrate-induced conformational change masking potential surfactant-binding sites. The observation that L-malate has a protective effect in combination with N A D H but not on its own can be rationalized by the finding in the mammalian malate dehydrogenase system that L-malate can bind to the enzyme only after the binding of the nucleotide coenzyme [4]. The irreversibility of CTAB inactivation upon removal of the surfactant by dilution follows the trend established with other enzymes [1]. It is interesting that both malate dehydrogenase isoenzymes are readily inactivated by SDS (Figs. 5-7) at concentrations well below 0.9 mM, the critical micelle concentration in 30 mM sodium phosphate buffer (pH 7.4) at 25°C [1]. This is a departure from the pattern observed with other proteins such as porcine lactate dehydrogenase, which is inactivated at or above the critical micelle

117

concentration [1]. Therefore cooperative binding of SDS to the malate dehydrogenases may not be a prerequisite for the inactivation observed by us (cf. the inactivation of cytoplasmic malate dehydrogenase by CTAB). High ionic strength has opposite effects on the sensitivity of the two isoenzymes to SDS; the mitochondrial enzyme becomes more resistant, and the cytoplasmic enzyme more susceptible, to SDS inactivation (Figs. 6 and 7). The stabilization of the former isoenzyme at high ionic strength could be due to attenuated electrostatic interaction between the positively charged enzyme molecule and the negatively charged surfactant. The greater sensitivity of the cytoplasmic enzyme at high ionic strength can be attributed to an effect on the critical micelle concentration of SDS, which is lowered by an increase in ionic strength [1,13,19]. Alternatively or additionally, high ionic strength possibly favours hydrophobic interactions, which may be prominent in the binding of the negatively charged SDS to the negatively charged cytoplasmic malate dehydrogenase. It is noteworthy that the SDS-induced inactivation of both malate dehydrogenases, following brief exposure to the surfactant, is almost totally reversible by dilution (Figs. 10 and 11). Longer treatment with the detergent causes some irreversible inactivation. Presumably, therefore, an initial reversible phase of inactivation is followed by an irreversible phase. Reversal of SDS inactivation by simple dilution is a rather unusual feature of these malate dehydrogenases. In other systems where reversal has been reported the SDS-inactivated protein was first treated with a denaturant like guanidinium chloride [20]. A kinetic analysis of the inhibition of several enzymes by SDS at sub-critical micelle concentration levels appears to demonstrate that, on the dimeric dehydrogenases, malate dehydrogenase and isocitrate dehydrogenase from pig heart, the surfactant acts as a competitive inhibitor with respect to the nucleotide coenzyme [21]. We may therefore speculate that, in the first reversible phase of the inactivation, SDS acts as a competitive inhibitor against NADH. The binding sites for the coenzyme are structurally very similar in the mitochondrial and cytoplasmic malate dehydrogenases [6]. The adenosine group in the coen-

zyme binds in a shallow crevice that is lined by predominantly hydrophobic residues, whilst the nicotinamide group binds in a cavity that is hydrophobic on one side and hydrophilic on the other (substrate-binding) side. It is possible that SDS, which has a long hydrophobic chain, competes to bind to the coenzyme domain. The ability of NADH to protect the enzymes efficiently against SDS inactivation (Figs. 10 and 11) supports this view. Presumably the second irreversible phase of inactivation is distinct from the first reversible phase and entails the irreversible binding of surfactant to sites on the protein, which may be outside the active site region, and NADH is able to provide virtually complete protection against irreversible inactivation during prolonged treatment with SDS (see Figs. 10 and 11) by blocking the possibly obligatory initial reversible phase.

Acknowledgement We are grateful to the Royal Society of Great Britain for support through equipment grants. References 1 Sanford, K.J., Meyer, D.J., Mathison, M.J. and Figueras, J. (1981) Biochemistry 20, 3207-3214 2 Blinkhorn, C. and Jones, M.N. (1973) Biochem. J. 135, 547-549 3 Wojtczak, L. and Nalecz, M.H. (1979) Eur. J. Biochem. 94, 99-107 4 Banaszak, L.J. and Bradshaw, R.A. (1975) in the Enzymes, 3rd Edn. (Boyer, P.D., ed.), Vol. 11, pp. 369-396, Academic Press, New York 5 Crow, K.E., Braggins, T.J., Batt, R.D. and Hardman, M.J. (1982) J. Biol. Chem. 257, 14217~14225 6 Birktoft, J.J., Fernley, R.T., Bradshaw, R.A. and Banaszak, L.J. (1982) Proc. Natl. Acad. Sci. USA 79, 6166-6170 7 Malamud, D. and Drysdale, J.W. (1978) Anal. Biochem. 86, 620-647 8 Smith, K. and Sundaram, T.K. (1983) Biosci. Rep. 3, 1035-1043 9 Gerding, R.K. and Wolfe, R.G. (1969) J. Biol. Chem. 244, 1164-1171 10 Smith, K., Sundaram, T.K., Kernick, M. and Wilkinson, A.E. (1982) Biochim. Biophys. Acta 708, 17-25 11 Helenius, A. and Simons, K. (1972) J. Biol. Chem. 247, 3656-3661 12 Jones, M.N. and Wilkinson, A. (1976) Biochem. J. 153, 713-718 13 Makino, S. and Niki, R. (1977) Biochim. Biophys. Acta 495, 99-109

118 14 Satake, I. and Yang, J.T. (1976) Biopolymers 15, 2263-2275 15 Nelson, C.A.A. (1971) J. Biol. Chem. 246, 3895-3901 16 Jones, M.N., Skinner, H.A., Tripping, E. and Wilkinson, A. (1973) Biochem. J. 135, 231-236 17 Stauffer, C.E. and Treptow, R.S. (1973) Biochim. Biophys. Acta 295, 457-466 18 Steinhardt, J., Scott, J.R. and Birdi, K.S. (1977) Biochemistry 16, 718-725

19 Helenius, A., McCaslin, D.R., Fries, E. and Tanford, C. (1979) Methods Enzymol. 56, 734-749 20 Liao, T.-H. (1975) J. Biol. Chem. 250, 3831-3836 21 Vincenzini, M.T., Favilli, F., Treves, C. and Vanni, P. (1982) Life Sci. 31,463-470