Comparative inhibition patterns of adenylate kinases from mammals, bird, fish and microorganisms

Comp. Biochem. PhysioL Vol. 107B, No. 3, pp. 489-494, 1994

Pergamon

© 1994 Elsevier Science Lid Printed in Great Britain. All rights r--~erved 0305-0491/94 $6.00 + 0.00

Comparative inhibition patterns of adenylate kinases from mammals, bird, fish and microorganisms Anita Williams, Joseph P. Taulane and Percy J. Russell Department of Biology, 0601 University of California, San Diego, La Jolla, CA 92093, U.S.A. The Ss inhibitions of AKs from six different sources were studied in mammals, birds, fish, and a microorganism. All AKs tested were inhibited by S,. Except for carp, all inhibited AKs from those tested were reactivated by DTT. Inhibitions of AKs by other hydrophobic inhibitors, NEM, butanol and ethanol were also studied. The inhibitions by Ss suggest that the hydrophobic pockets in the AKs cover a wide phylogenetic range. All inhibitions by Ss are reactivated by DTT. Unlike the inhibitions by Ss, the characteristics of inhibitions by the other hydrophobic inhibitors differed among the AK sources tested and none was the irreversible type. The data suggest that no covalent bonds were formed with NEM. Similarly, the ability to reactivate the inhibitions by DTF differed among the AK sources. The possibility that the hydrophobic domains in the AKs may serve as part of an enzyme activity control mechanism is discussed. Key words: Inhibition; Adenylate kinase; Mammals; Birds; Fish; Microorganisms.

Comp. Biochem. PhysioL 107B, 489-494, 1994.

Introduction

The adenylate kinases (AKs) are ubiquitous enzymes found in all living cells and in subcellular compartments. The functions of AKs in the cellular metabolism have been proposed as important in the control of energy charge (Atkinson and Walton, 1967) and oxidative phosphorylation (Veuthey and Stucki, 1987). A comparison was made among the AKs from various mammalian muscle sources; the amino acid sequences and three-dimensional structures were very similar (Reuner et al., 1988). It has been shown that the properties of AK isozymes can have very different physical and chemical properties and that there is a high degree of organ specificity among animals (Russell et ai., 1974). It was surprising to find that Sa, considered a novel inhibitor of RMAK (Conner Correspondenceto: PercyJ. Russell,Department of Biology, 0601 Universityof California, San Diego, La Jolla, CA 92093, U.S.A. Tel.: (619)534-4184;Fax: (619)534-5293. Abbreviations--AK, adenylate kinase; DTT, dithiothreitol; RMAK, rabbit muscle adenylate kinase; PMAK, porcine muscle adenylate kinase; CMAK, chicken muscle adenylatekinase; YFAK, yellowfintuna adenylate kinase. Received 7 June 1993; accepted 7 July 1993.

and Russell, 1983), inhibits AKs from a wide variety of sources--mammals, birds, fish and microorganisms. Previous studies (Conner and Russell, 1983; Russell et al., 1984) showed that RMAK was very sensitive to inhibition by $8, which was reactivated by sulfhydryl compounds. It was proposed that these reactions might be a mechanism for controlling the activity of this enzyme (Conner and Russell, 1983). This study determines how universal inhibition of AKs by $8 is from different sources. Materials and M e t h o d s Materials Rabbit skeletal muscle adenylate kinase (RMAK, EC 2.7.4.3) is the commercial preparation from Sigma Chemical Co. (St Louis, MO) designated "Myokinase, grade III from rabbit (Lepus cuniculus) muscle." When subjected to isoelectric focusing, the RMAK showed a single isozyme with an isoelectric point value near 9 and gave a single protein band by acrylamide electrophoresis with the silver-staining method (Morrissey, 1981). We

489

490

Anita Williams

used the commercial RMAK without further purification. The specific activity was in the range of 2000EU/mg under our assay conditions. Adenylate kinases from Bacillus stearothermophilus (bacterial AK), chicken (Gallus domesticus) muscle (CMAK), and porcine (Sus scrofa) muscle (PMAK) were from Sigma. Yellowfin tuna (Neothunnus agentivittatus, YFAK) and carp (Cyprinus carpio) skeletal muscle adenylate kinase (carp AK) were provided by Dr Joe W. Conner, Biomedicinal Chemistry Department, School of Pharmacy, University of Maryland at Baltimore. The following chemicals were obtained from Sigma: N-ethylmaleimide, NEM; dithiothreitol, DTT. A K assay. We measured AK activity, AMP + MgATP = ADP + MgADP, according to Adam (1965), and modified elsewhere (Russell et al., 1974). The 1-ml assay mixture was 20 mM potassium phosphate buffer, pH 7.0; 0.3 mM phosphoenolpyruvate; 0.4 mM NADH; 8.0mM AMP and ATP each; and 8.0 mM MgCI 2. Sufficient amounts of lactate dehydrogenase (LDH, EC 1.1.1.27) and pyruvate kinase (PK, EC 2.7.1.40) were added so that the coupling system was not rate-limiting. Reactions were initiated by the addition of AK. All initial reaction rates were determined by measuring the decreased absorbance of NADH at 340 nm with time. The molar absorptivity value of 6220 was used to convert the change of NADH absorbance to micromoles of

et al.

product formed. One enzyme unit (EU) of activity is the formation of 1 mole of ADP/min at 25°C. Inhibition assays. The conditions for the inhibition of RMAK by $8, NEM, and the alcohols were as follows, unless stated otherwise. The RMAK, in 100mM potassium phosphate buffer, pH 8.0, was incubated with the inhibitor under study for 0.5-1.0 hr and the activity remaining was then determined. For the concentrations of enzymes and inhibitors used in these studies, it has been determined that no additional significant inhibitions occurred after 0.5 hr of incubation. Protein determinations. The RMAK concentrations were determined spectrophotometrically using the absorbancy value E °j'/* = 0.53 at 279 nm (GornaU et al., 1949). CD measurements. The CD measurements were carried out on a modified Cary 61 spectropolarimeter (Goodman et al., 1985) as described elsewhere (Russell et al., 1990). The CD spectra were obtained by a signal averaging 10 scans, using a 0.1 mm cell. Each experiment had a baseline for that specific condition minus RMAK. All CD spectra were deconvoluted by the method of Compton and Johnson (1986, 1986), as detailed previously (Russell et al., 1990). The buffer was a 10 mM potassium phosphate at pH 8.0. The RMAK was 200/zg/ml always. The percents of the secondary structures were determined as described previously (Compton and Johnson, 1986; Brahms and Brahms, 1980).

Table 1. Specificity of $8 inhibition of AK from various sources Tissue

Enzyme*

Inhibition1" (%)

Skeletal muscle Red muscle Intestinal muscle Skeletal muscle Intestinal muscle Stomach muscle Liver Skeletal muscle Heart Muscle Muscle Brain Muscle Muscle

100 100 0 100 0 0 0 0 0 0 100 0 0 0 100 100 100 0

Organism Yellow tuna

B. stearothermophilus

--

C. r n y c o d e r m a M . cerevisiae

---

YFAK AK AK RMAK AK AK AK PK LDH LDH PGK CK CK HK GIyK AK GlyK HK

--

AK

Rabbit

0

hexokinase (EC2.7.1.1); GIyK, glycerol kinase (EC 2.7.1.30); and PGK, phosphoglycerate kinase (EC 2.7.2.3). tOne to 2EU/ml of enzyme activity in 0.I M KPO4 buffer, pH7, was incubated for 1 hr at 25°C in 97/zM $8. The level of $8 is about four times in excess for 100% inhibition of RMAK. The activity remaining was then determined. *Abbreviations--HK,

Inhibition of adenylate kinases Table 2. 50% inhibition by S,, NEM, BuOH and EtOH with AK from various sources* $8 (~M)

NEM (mM)

BuOH (M)

EtOH (M)

RMAK 2.0 0.2 0.52 5.2 PMAK 3.6 0.2 0.25 5.0 CMAK 3.0 0.02 0.33 5.0 YFAK 17 0.06 0.76 3.8 Carp AK 140 0.07 0.97 4.9 Bacterial AK 400 1.0 none 13.0 *Estimations of 50% inhibition derived from titrations of inhibitor against approximately 20 EU/ml.

Results and Discussion Specificity of Ss inhibition The specificity of the inhibitions of AK by Ss paralleled those reported for other sulfhydryl reagents; the AK-1 (muscle type) were inhibited and the AK-2 (liver or mitochondrial types) were not inhibited (Russell et al., 1974). Because the inhibition of AK by Ss was novel, we investigated the AKs derived from a variety of sources. In addition, other kinases were studied to determine if the inhibitions were restricted to AK. Glycerokinases from two microorganisms were also inhibited by Ss. These results are given in Table 1. We extended the study to include the specificity of Ss inhibition and other hydrophobic inhibitors, NEM, butanol (BuOH) and ethanol (EtOH), among AKs from various species.

Comparison of hydrophobic inhibitors of AK from various sources Characteristics of Ss inhibition. Table 2 shows that all AKs tested appear to be sensitive to hydrophobic inhibitors with the exception of carp and bacterial AK which are resistant to Ss and BuOH inhibition. Since Ss, NEM, and the alcohols represent such different chemical groups, we investigated whether the inhibitions were reversible. It should also be noted that the effectiveness of the inhibitors is in the order of their hydrophobic character--Ss > NEM > butanol > ethanol. The amino acid compositions and sequences of AKs from a variety of sources are known and the changes among species are considered conservative. The studies show conservative features of the hydrophobic domains of the AKs (Schulz, 1987). All AK sources we studied-mammals, birds, fish, and bacteria--are inhibited by the highly hydrophobic Ss. Among the fish AKs, carp AK was distinctly different from that of tuna. Carp AK was the most resistant to inhibition by Ss among the animals tested. The bacterial AK tested required high concentrations of Ss compared to AKs from other sources. The sensitivity differences probably reflect composition differences of hydrophobic

491

domains (Schulz, 1987). The other hydrophobic inhibitors, NEM and the alcohols, also showed different potencies among AKs from different sources. M o s t Ss inhibitions o f A K s have the feature o f reversible inhibition p a t t e r n s when titrated against S 8. T h e Y F A K inhibition by S 8 shows an irreversible p a t t e r n b u t is reactivated by D T T . F i g u r e 1 is a typical p a t t e r n o f reversible inhibitions, c h a r a c t e r i z e d by rectilinear convergence at the origin (Segal, 1975); this p a t t e r n is c o m m o n for inhibitions b y Sa, N E M a n d the alcohols. Sensitivity differences a m o n g the A K s suggest i n h i b i t o r site differences.

Characteristics of NEM inhibition. Generally, NEM is considered an irreversible inhibitor. Thioether bonds formed with NEM and cysteinyl groups are not broken by DTT like disulfide bonds. Among the AKs, NEM is considered a hydrophobic inhibitor because it does not give an irreversible pattern of inhibition with all the AKs we tested. When titrated against NEM, the patterns are similar to those obtained for Ss and the alcohols. In addition, the RMAK inhibition is reactivated by DTT. Chicken and fish are more sensitive to NEM inhibition than mammals, and bacterial AK was the least sensitive. Among the AKs from mammalian sources, the greater the hydrophobicity of the inhibitor the more sensitive the AK was to the inhibitor. We propose that inhibition by NEM is not a covalent interaction, but a hydrophobic inhibition. Bacterial AK may be an exception since it requires at least five times the concentration of NEM, about 1 mM, to inhibit to the same degree as AKs from other sources. High NEM concentrations may indicate secondary inhibition with sulfhydryl group interactions in the case of bacterial AK. Non-covalent linked 40

u

10

0

" "

0

0.25

O.SO RMAK, nmcles

0.75

1.00

Fig. 1. The reversible character of PMAK with butanol inhibition. This is the pattern of convergence at the origin indicating a reversible inhibition (Segal, 1975). The symbols for the concentrations of butanol are as follows: (l-q), none; (A), 0.57 M; (A), 0.76 M; and ( , ) , 0.86 M.

492

Anita Williams et al. Table 3. Reactivation of inhibited* AKs from various sources by 10 mM DTT Ss/DTT NEM/DTT BuOH/DTT EtOH/DTT ControUi Enzymet (% Roe) (% Roe) (% Roe) (% Roe) DTT RMAK 92 47 59 93 112 PMAK 93 100 70 29 122 CMAK 100 0 0 63 104 YFAK 91 0 0 0 105 Carp AK 0 0 0 103 109 Bacterial AK 98 0 NA 108 112 *The average inhibition, from a minimum of three runs, ranged between 70 and 95% inhibition. tThe average AK concentration was 20 EU/ml. :~DTT controls were equal to, or greater than, the original controls without DTT.

inhibition by N E M has been previously reported (Lin and Seltzer, 1981). Characteristics o f butanol and ethanol inhibitions. The bacterial A K is not inhibited by water with saturated b u t a n o l - - a concentration of approximately 0.9M. By contrast, all animal A K s tested show a marked sensitivity to butanol inhibition when compared to inhibition by ethanol. In Table 2, we show that R M A K has a higher specificity for butanol than for ethanol as an inhibitor. A characterstic of hydrophobic inhibitions of R M A K is reactivation by DTT; this characteristic was examined in our A K s from various sources.

Reactivation of hydrophobic inhibitions by D T T with A K from various sources We tested a variety of A K sources to determine whether the reactivation by D T T obtained for all of these hydrophobic inhibitions. Table 3 shows that all A K sources inhibited by Ss were reactivated by D T T , except for carp. Chicken, fish and bacterial A K inhibitions by N E M and butanol are not reactivated by DTT. All ethanol inhibitions of AKs, except Y F A K , were reactivated by DTT. Previous C D studies (Russell et al., 1990) showed that D T T caused conformational changes. We suggest that the mechanisms of inhibitions observed are also the result of conformational changes. Since heat inactivation is generally considered to be the result of conformational changes, we investigated the effect of D T T on heat inactivation.

Reactivation of heat-denatured A K from various sources Table 4 shows that heat-inactivated A K s from all sources are partially reactivated by DTT. Though not shown here, glutathione and cysteine are as effective as D T T in reactivating A K s inhibited by heat or by the hydrophobic inhibitors. We followed the R M A K conformational changes associated with heat inactivation by C D methodology. R M A K was the only enzyme pure enough for C D methodology. Table 5 shows the losses of R M A K activity with time at 100°C and the changes in the secondary structures, estimated from C D spectra ( G o o d m a n et al., 1985; C o m p t o n and Johnson, 1986; C o m p t o n and Johnson, 1986). Figure 2 shows that after 2 min heating time, there are no large changes in the C D spectrum. On the other hand, with increasing losses of activity, there are significant changes in percentages of the s-helix and fl-pleated sheet regions. Increased inactivation shows a decrease content of the s-helix regions and an increase in flpleated sheet regions. These changes are the opposite of those reported previously when the activity of R M A K is enhanced by the addition of D T T (Russell et al., 1990). Our hypothesis is that the hydrophobic inhibitors promote a conformational change (Russell et al., 1990; McDonald and Cohn, 1975) that causes a loss of activity and the D T T causes a conformational change that restores the activity. We studied the effects of D T T on the reactivation of activity after heat denaturation. All heat-inactivated A K s from all sources

Table 4. Reactivation of heat-denatured adenylate kinase with DTr Percent recovery with l0 mM DTr* YFAK CARP Bacterial RMAK PMAK CMAK AK AK AK Heat DTTt 72 61 71 37 29 54 Control DTT 107 121 104 99 104 102 *After the heating period, the AK solutions were made 10 mM DTT, incubated at 25°C for 1 hr and the activity then measured. tHeating 20 EU/ml at 100°C in 10 mM potassium phosphate buffer, pH 8.0, for 1.5 min caused complete loss of activity.

Inhibition of adenylate kinases

493

Table 5. RMAK heat-denatured activityand CD structure at various times Heating time at Activity a-Helix B-Pleated Unordered 100°C* remaining regions sheets regions (min) (EU/ml) (%) (%) (%) 0 109 25 45 30 2 81 20 55 25 4 61 15 60 25 9 48 10 60 30 *Except for the times, the heating conditions are the same as in Table 4. are reactivated by DTT. The extent of the reactivation varied with the source. This shows that all the AKs, including carp AK, are capable of being reactivated by sulfhydryl groups. The reactivation of fish A K was about half that of the mammalian sources and bacterial reactivations were somewhere between these extremes. Inhibition by Ss appears to be a common feature of AK. The reactivation of $8 by D T T also appears to be a c o m m o n feature of the A K s tested. Carp was a clear exception. Unlike the muscle A K s of known composition (Schulz, 1987), carp A K contains only one cysteinyl group (Reuner et al., 1988). D T T only reactivates A K activity lost by N E M inhibitions of the mammalian sources. The reversible inhibitor patterns obtained from the titrations o f enzyme against N E M suggest that either the N E M - A K complexes are more stable outside the m a m malian species or that the interaction of the sulfhydryl groups with the reactivating site is weaker. The fish enzymes are not reactivated by D T T when N E M or butanol are the inhibitors. We propose a general mechanism of inhibition for these chemically diverse hydrophobic inhibitors. Our hypothesis is that hydrophobic inhibitors, such as those investigated in this paper, interact with the hydrophobic domains of A K (Reuner et al., 1988; McDonald and

25-

-'1 ~

190

,

I

200

,

I

w

210

220

J

I

!

230

240

nm Fig. 2. The effect of heat and time on R M A K CD spectra.

The conditions of measurements and the calculations are given under Materials and Methods. The RMAK samples were the same as those given in Table 4. The symbols represent the following heating conditions: ( ), 0 min; (O), 2 rain; (0) 4 min; and (A), 9 min.

Cohn, 1975) and promote conformational changes that inactivate AK. The different inhibition profiles suggest inhibitor site differences among the AKs. Using R M A K as a model, Table 6 shows that butanol, the least hydrophobic of the inhibitors, caused the most change in the g-helix and the //-pleated sheets. Also, there is no correlation between the degree of change in conformation of inhibited R M A K and the extent of reactivation by DTT. For example, butanol and N E M show the similar extents of D T T reactivation yet the conformational changes estimated from C D data are quite different. Similarly, the conformational changes induced by Ss and N E M are the same but the extent of reactivation by D T T is complete for Ss and half as much for N E M . Heat denaturation also causes changes in the secondary protein structures concomitant with an activity loss that is largely reactivated by DTT. These observations suggest that changes in conformation by various inhibitors are complex and are not similar. Support for the hypothesis that hydrophobic inhibitors exert their effect by promoting conformational changes has been presented. The A K s denatured by heat show significant reactivation of activity by the addition of DTT. Except for carp AK, all show reactivation of activity from Ss inhibition. Inhibition profiles, reactivations by DTT, and CD analysis of the secondary structures under various conditions and various species, are consistent with the general hypothesis. We recognize that details of individual mechanisms of these hydrophobic inhibitors may differ with respect to sites of Table 6. Effect of inhibitors on the percent g-helix of RMAK* Reactivated by I0 mM ]/-Pleated Inhibition DTT g-Helix sheets Inhibitor (%) (%) (%) (%) None 0 110 17 54 7/~M Ss 89 100 14 54 0.5 mM NEM 83 47 15 55 0.8 M butanol 90 58 3 77 Heart 100 72 12 62 *The measurements and concentrations are the same as given in Materials and Methods. tHeating conditions are similar to Table 4.

494

Anita Williams et al.

interaction and the conformational changes promoted by individual inhibitors. The inhibitions and reactivations shown in this study show that the activity of a small enzyme may be modified by low molecular weight compounds in a manner similar to allosteric modifications. Acknowledgements--We express our appreciation and gratitude to Dr M. Goodman of the Chemistry Department, University of California, San Diego for the use of the CD apparatus and his assistance in the interpretation of the CD data.

References Adam H. (1965) In Methods of Enzymatic Analysis (Edited by Bergmeyer H. U.), pp. 573-577. Academic Press, New York. Atkinson D. E. and Walton G. M. (1967) Adenosine triphosphate conservation in metabolic regulation. J. biol. Chem. 72, 248-254. Brahms S. and Brahms J. (1980) Determination of protein secondary structure in solution by vacuum ultraviolet circular dichroism. J. molec. Biol. 138, 149-178. Compton L. A. and Johnson W. C. Jr (1986) Analysis of protein circular dichroism spectra for secondary structure using a simple matrix multiplication. Analyt. Biochem. 155, 155-167. Compton L. A. and Johnson W. C. Jr (1986) Analysis of protein CD spectra for secondary structure using a simple matrix multiplication. Biophys. J. Abstr. 49, 494a. Conner J. and Russell P. J. (1983) Elemental sulfur: a novel inhibitor of adenylate kinase. Biochem. biophys. Res. Commun. 113, 348-352. Goodman M., Venkatachalapathi Y. V., Mammi S. and Katakai R. (1985) A guest-host approach to oligodepsi-

peptide structure. Proc. Int. Symp. Biomol. Struct. Interactions, Suppl. J. Biosci. 8, 223-238. Gornall A. G., BardawiU C. J. and David M. M. (1949) Determination of serum proteins by means of biuret reaction. J. biol. Chem. 177, 751-766. Lin M. and Seltzer S. (1981) Maleylaceton cis-trans isomerase: formation of a N-ethylmaleimide-labeled enzyme only during the slow phase of biphasic inhibition reaction. FEBS Lett. 124, 169-172. McDonald G. G. and Cohn M, (1975) Proton magnetic resonance spectra of porcine muscle adenylate kinase and substrate complexes. J. biol. Chem. 250, 6947-6954. Morrissey J. (1981) Silver stain for proteins in polyacrylamide gels: a modified procedure with enhanced uniform sensitivity. Analyt. Biochem. 117, 307-310. Reuner C., Hable M., Wilmanns M., Kiefer E., Schiltz E. and Schulz G. E. (1988) Amino acid sequence and three-dimensional structure of cytosolic adenylate kinase from carp muscle. Protein Seq. Data Anal. 1, 335-343. Russell P. J. Jr, Horenstein J. M., Goins L., Jones D. and Laver M. (1974) Adenylate kinase in human tissues--I. Organ specificity of adenylate kinase isoenzymes. J. biol. Chem. 249, 874-1879. Russell P. J. Jr, Conner J. and Sisson S. (1984) Sulfur specifically inhibits adenylate kinase in assays for creatine kinase. Clin. Chem. 30, 1555-1557. Russell P. J. Jr, Chinn E., Williams A., David-DiMarino C., Taulane J. P. and Lopez R. (1990) Evidence for conformers of rabbit muscle adenylate kinase. J. biol. Chem. 265, 11,804-11,809. Schulz G. E. (1987) Structural and functional relationships in the adenylate kinase family. CoM Spring Harb. Symp. quant. Biol. 52, 429-439. Segal I. H. (1975) Enzyme Kinetics, pp. 127-128, 293-296. John Wiley & Sons, New York. Veuthey A. and Stucki J. (1987) The adenylate kinase reaction acts as a frequency filter towards fluctuations of ATP utilization in the cell. Biophys. Chem. 26, 19-28.