66 Q 1992 Elsevier

BBALIP

Science

BiocAimlcu CI BiO~Ihysicu Arlu, I I24 (I 492) h6-70 Publishers B.V. All rights reserved OOOs-27hO/92/$OS.O0

53831

Competitive inhibition of ~ipo~ytic enacts. VI. inhibition of two human phos~ho~ipas~s A, by acy~a~ino phospho~ipi~ anaiogues L. van den Berg, P.A. Franken, H.M. Verheij, R. Dijkman and G.H. de Haas Lkpartmenr

of Enzymoiogr). cmd Protein

Engimxring,

(Received

Key words:

Competitive

inhibition;

Platelet

CBLE,

State Unicwsity

9 September

phospholipase

ofUtwchr, Utrwhr (Nethrrlundsf

1991)

A,; Pancreatic

phospholipase

A,; (Human)

The competitive inhibition of human pancreatic and a mutant human platelet phospholipase A, (PLA,) was investigated using acylamino phospholipid analogues, which are potent competitive inhibitors of porcine pancreatic PLA, [De Haas et al. (1990) Biochim. Biophys. Acta 1046,249-2571. Both the mutant platelet PLA, and the human pancreatic PLA, are effectively inhibited by these compounds. The enzyme from platelets is most strongly inhibited by compounds with a negatively charged phosphoglycol headgroup. Compounds with a neutral phosphocholine headgroup are only weak inhibitors, whereas an inhibitor with a phosphoethanolamine headgroup shows an intermediate inhibitory capacity. The pIatelet PLA, is most effectively inhibited by negatively charged inhibitors having a relatively short (four or more carbon atoms) alkylchain on position one and a acylamino chain of I4 carbon atoms on position two. For the pancreatic enzyme an inhibitor with a phosphoethanolamine headgroup was more effective than inhibitors with either a phosphochoiine or a phosphogIyco1 headgroup. The chainlength preference of the pancreatic enzyme resembles that of the plateiet PLA,. The Iargest discrimination in inhibition between the human platelet and the human pancreatic PLA, is obtained with inhibitor with a negatively charged phosphoglycol headgroup, an alkyd chain of four carbon atoms on position one and a long acylamino chain of 14-16 carbon atoms on position two. Because the platelet PLA, is thought to have several biological functions, specific inhibitors of this enzyme could have important implications in the design of pharmaceutically interesting compounds.

Introduction

Phospholipases A, belong to a class of lipolytic enzymes which are widespread in nature. The most extensively investigated member of this group is the extracellular enzyme from porcine pancreas which has a role in the digestion of dietary phospholipids. During the last few years much attention has been shifted towards the isolation and characterization of mammalian non-pancreatic PLA,s. These enzymes, which are found in almost every mammalian cell, are thought to have several biologically interesting functions. An important function of these PLA2s is the liberation from membrane phospholipids of arachidonic acid. This acid is the precursor for biologically active compounds like prostaglandins and leukotrienes [1,21, and thus

Correspondence: L. van den Berg, Department of Enzymology and Protein Engineering, CBLE, State University of Utrecht, Padualaan 8, 3584 CH Utrecht, The Netherlands.

have PLA,s been implicated in the inflammatory response [3]. Until now, two classes of non-pancreatic PLA,s have been described in the literature; extracellular PLA,s having a molecular mass of 14 kDa and a cytosolic PLA, with a molecular mass of 110 kDa [4]. A 14 kDa PLA, has been purified from human synovial fluid [5], and its sequence has been shown to be identical to the PLA, associated with human platelets [6]. Because of its possible biological functions, there is a widespread pharmacological interest to design potent inhibitors which are (specifically) directed against this enzyme. In a recent paper [71 it was shown that certain acylamino phospholipid analogues are very potent inhibitors of the porcine pancreatic PLA,. Recently Ransac et al. [Sl tested these inhibitors on phospholipases from different sources and the authors concluded that the inhibition was quite different for the various PLA,s. Thus it would be interesting not only to investigate if these compounds are also able to inhibit the human platelet PLA2, but also to test whether

67 such inhibitors can discriminate between this non-pancreatic enzyme and the human pancreatic PLA,. In this paper we describe the inhibition of the MSL mutant human platelet PLA, [9] and of human pancreatic PLA, by acylamino phospholipid analogues. Materials

and Methods

Substrates and Inhibitors (RI-l,2 didodecanoyl-glycero-3-phosphocholine (diC,,PC) was prepared as described by Cuber0 Robles and van den Berg [lo]. The general synthesis and purification of the inhibitors used in this study has been described by Dijkman et al. [ill. The ethanolamine derivative of compound 14 was synthesized according to Eibl and Nicksh [12]. All inhibitors were synthesized in chirally pure (RI- and (S)-configurations, except the analogues lO,ll, and 12, which were obtained and tested in racemic form. [‘HINMR spectroscopy was used for confirmation of the chemical structures [ll]. In Fig. 1 the general structures of the inhibitors used in this study are given. Phospholipases A, The M8L mutant human platelet PLA, was expressed in Escherichia coli and isolated as described by Franken et al. [9]. The human pancreatic PLA, was isolated according to Verheij et al. [13].

C,,H23-C-NHY

\H !H

2--o-L--o-choline

(a)

Om

Assays PLA, activity was routinely determined in the pH stat at pH 8.0. Conditions were: 5 mM Tris, 10 mM Ca2+, and 150 mM NaCl at 25°C. A mixed micellar solution of CR)-1,2 didodecanoyl glycero-3-phosphocholine (3 mM) in sodium-taurodeoxycholate (TDOC; 3 mM, obtained from Sigma) was used throughout. In this assay-system the platelet and the pancreatic enzyme have specific activities of 125 and 10 U . mgg ‘, respectively. Inhibition was determined by measuring the enzymatic velocity as a function of the mole fraction LY[ = I/(1 + S)]. The sum of substrate and inhibitor concentrations was kept constant at 3 mM. Under these conditions all enzyme is present in the lipid-water interface, to give a maximal enzymatic velocity (data not shown). Competitive inhibition at interfaces An equation for the steady-state enzymatic velocity under bulk conditions in a mixed micellar system containing detergent (D), inhibitor (I) and substrate (S) has been derived by Ransac et al. [14]. Furthermore, introducing a ratio of steady-state velocities CR,) and the mole fraction (Y (= l/(1 + S)), these authors derived the following expression: G K:

--1 R,=l+o

=1+(Y.z

Plotting R, as a function of (Y yields a straight line with slope Z, the inhibitory power of the molecule tested. For good inhibitors the value of Z can be regarded as the ratio K,*/K,*. Also, the mole fraction for 50% inhibition (a,,> can be determined: as,, = l/(2 + Z>. The advantage of using as0 compared to the commonly used IC,, value is situated in the substrateconcentration independency of the former [14]. Results and Discussion

@I

CY

@_~_NH~~~~

p

/choline

cH,-O-Y-O-

glycol

b-

(cl

Iethanolamine

Fig. 1. General structures of the acylamino phospholipid analogues that were used in this study. Choline, -CH, -CH, -N(CH,);, glycol stands for: -CH,-CH,-OH, ethanolamine stands for: -CH,CH,-NH;. Different substitutions are indicated by the letters X, Y and Q; for the character of these substitutions see the text and the legends to Fig. 2 and Tables I and II.

The first analogues that we investigated, had a long (11 or 12 carbon atoms) alkyl chain attached via functional groups of differing polarity to position one, and an acylamino chain with a constant length of 12 carbon atoms on position two (Fig. lA1. For these compounds the inhibitory power Z was determined and the Z-values of these phosphocholine derivatives are shown in Fig. 2. It is clear that these compounds weakly inhibit both enzymes: the affinity of the enzyme for the inhibitors is 3-10 times higher than the affinity of the enzyme for the substrate. When we compare these values with the Z-values ranging from 40 to 116 which were obtained with the porcine pancreatic PLA, [7], it is evident that the latter enzyme is much more effec-

68

a-

2

6-

3 N

4-

compound

number

Fig. 2. Z-values of CR)-dodecylamino-2-deoxyphosphocholine analogues with the general structure given in Fig. lA, measured on the human platelet (black bars) and the human pancreatic PLA, (blank bars). The character of the substitution X is as follows: compound 1. X = -NH-CO-(CH21,,,-CH3: compound 2. X = -O-CO-tCH,),,,CH 3; compound 3. X = -0-(CH2), , -CH,; compound 4, X =-S(CH,),,-CH,.

tively inhibited by these compounds. Comparing the two human enzymes, it is clear that both PLA,s are inhibited to about the same extent, with the exception of the ester inhibitor 2 which is a 20-fold weaker inhibitor for the pancreatic PLA, than for the platelet PLA?. An explanation for this discrimination is not easily given. A general property of these inhibitors is that the inhibitory power Z increases with decreasing polarity of the functional group on position one, indicating that hydrophobic interactions play an important role in the inhibitor-enzyme contacts. The same conclusion was drawn from the results with the porcine pancreatic PLA z [7]. The apparent importance of the local hydrophobicity led us to examine the influence of the size and overall hydrophobicity of the alkyl chain on position one. This was done by varying the length of the alkyl chain, while the length of the acylamino chain on position two was kept constant at 12 carbon atoms (Fig. 1B). In addition, the influence of the polar headgroup on the inhibition was tested. The results are presented in Table I, from which it can be seen that for the human platelet PLA, the negatively charged phosphoglycol derivatives are very good inhibitors. It can also be seen that these phosphoglycol derivatives inhibit the platelet PLA, a factor 30-60 times better than the neutral phosphocholine compounds. Although the platelet PLA, is a very basic enzyme, the very high isoelectric point (IEP > 10.5) alone is not sufficient to explain this behavior. The neutral porcine pancreatic PLAz (IEP = 6.5) also shows a large preference for

phosphoglycol inhibitors 171, suggesting that the preference for phosphoglycol derivatives is based upon a more specific interaction of the headgroup with the active sites of both enzymes. The data in Table I also show that the inhibitors with short alkyl chains (two or three carbon atoms) on position one give low Z-values, while alkyl chains of four carbon atoms or longer give rise to larger Z-values which are relatively independent of the precise length of the alkyl chain. In contrast it has been shown that the inhibition of the porcine pancreatic PLA, shows a clear optimum for the alkyl chain length at the position one. For this reason it has been concluded that the porcine pancreatic PLA, has a well defined binding pocket for this alkyl chain [7]. Our data suggest that in the human platelet PLA, the binding pocket for the alkyl chain at position one is less well defined. For the human pancreatic PLA2, as for the platelet PLA,, the Z-values for inhibitors with alkyl chains of four or more carbon atoms on position one seem to be relatively independent of the exact length of the alkyl chain. The phosphocholine inhibitors all display similar low Z-values of about 10-20. Again, the negatively charged phosphoglycol compounds are better inhibitors, although the preference for glycol over choline derivatives is smaller than for the platelet enzyme (a factor 20 or less for the pancreatic enzyme compared to a factor 30-60 for the platelet enzyme). If we compare the Z-values of the phosphoglycol derivatives obtained for both enzymes (Table I), it is striking that two human PLA,s are inhibited to such different extents by these compounds. The phosphoglycol derivatives of compounds 9, 11 and 12, for example, inhibit the human platelet PLA, 10-15 times stronger than the human pancreatic PLA,. TABLE

I

Inhibitory power Z of 2-(R)-dodecylamino-2-droxyphosphocholinr / glycol analogue3 measured on the human platelet and the human pancreatic PLA 7 The inhibitors

have the general

Inhibitor No.

Identity

5 6 7 8 9 10 I1 12

H

of Y

CH, CH2-CH, (CHz)z-CH, tCH,),-CH, (CH,),-CH, (CH2),-CH, (CH21q-CH,

structure

given in Fig. IB.

Z-values platelet PLA z

Z-values pancreatic

PC

PC

PG

0.0 n.d. n.d. n.d. 13 20 IX 9

0.0 nd. n.d. n.d IO3 I40 x9

0.6 2.1 7.x 13 30 18 20 25

PC 7.7 nd. n.d. n.d. 1 300 .540 i1 1000 ‘I 1660 ‘I

PLA,

I x0

n.d. = not determined. a The Z-values of inhibitors Nos. 10, 11 and 12 are the (corrected)values for the pure (R&isomers, by multiplying the measured values by a factor two. PC stands for the phosphocholine, and PG for the phosphoglycol analogue, respectively.

69 So far, the best inhibitor seems to be the glycol derivative of compound 9, which has the largest specificity for the platelet PLA, as compared to the pancreatic enzyme. In addition this inhibitor can be easily obtained in chirally pure form starting from optically active norleucine. Therefore we decided to use inhibitors with an alkyl chain of four carbon atoms on position one and with an acylamino chain of variable length on position two in order to investigate if the specificity of inhibition could be increased (Fig. 1C). The results of the inhibition studies with this series of compounds are shown in Table II. From this table it is obvious that the platelet PLA, is optimally inhibited by inhibitors with an acylamino chain of 14 carbon atoms, which points to the presence of a binding pocket of limited size in the enzyme. Based on the X-ray structure of the porcine pancreatic PLA ,-inhibitor complex Thunnissen et al. [15] concluded that there is a binding pocket for the two position that can accomodate about 14 carbon atoms. The authors also suggested that the human platelet PLA, has a similar binding pocket. At the other hand, Wery et al. [16] concluded from the X-ray structure of the human platelet PLA, that this binding pocket might be narrower than in the pancreatic PLA,. As is the case with the inhibitors discussed before, compounds with a negatively charged phosphoglycol headgroup are much better inhibitors for the platelet PLA, than compounds with a phosphocholine headgroup. In addition, the ethanolamine derivative of compound 14 was tested as an inhibitor of the platelet enzyme. This inhibitor appeared to have a Z-value of 400, which means that it occupies an intermediate position between negatively charged glycol derivatives and neutral choline derivatives. For the human pancreatic PLA,, the data of Table II show that the choline compounds are rather weak inhibitors, which all display similar Z-values. The glycol inhibitors, however, interact more strongly with the enzyme, giving the largest Z-value for a compound with an acylamino chain of 14 carbon atoms, as was found

TABLE

for the platelet PLA,. Striking is the fact that the ethanolamine derivative of inhibitor 14 clearly is the most potent inhibitor for the human pancreatic PLA,, with a Z-value of 248. From the data it is clear that the glycol derivatives of the compounds with the general formula presented in Fig. 1C are better inhibitors for the platelet PLA z than for the human pancreatic PLA 2. To check the influence of the stereospecificity of the inhibitors, the (S&isomer of the best inhibitor, compound 15, was used. It appears that both the platelet (Z-value = 3.1) and the pancreas PLA, (Z-value = 0) are hardly or not inhibited by this compound, pointing to a stereospecific way of binding in the active centre of the PLA,s. A similar stereospecific inhibition has been reported for porcine pancreatic PLA, [7]. In summary, this study clearly shows that acylamino phospholipid analogues, which were shown to be potent competitive inhibitors of the porcine pancreatic PLA, 171, are also able to inhibit PLA,s of human origin. Compounds with phosphocholine and phosphoethanolamine headgroups are relatively weak inhibitors that inhibit platelet PLA, and pancreatic PLA, to about the same extent. Inhibitors with an ester function on position one seem to be quite able to discriminate between the platelet and the pancreatic PLA, but the absolute Z values are low (cf. Fig. 2). Most inhibitors show a preference for the mutant PLA, from human platelets, as compared to the human pancreatic PLA,. The former enzyme is most effectively and specifically inhibited by a compound with a negatively charged phosphoglycol headgroup, a short alkyl chain of four carbon atoms on position one, and a long acylamino chain of 14 or 16 carbon atoms on position two (c.f. Tables I and II). The platelet PLA, is inhibited by such compounds 15-25 times more effectively than the human pancreatic PLA,, showing that these compounds are able to inhibit the human platelet PLA, in a specific way. In conclusion, the inhibitors which were shown to be rather selective inhibitors of the mutant human platelet PLA, relative to the human pancreatic PLA, can be regarded as potentially

II

Inhibitory power Z The inhibitors

of 2-CR)-acylamino

have the general

phospholipid analogues, measured on the human platelet and the human pancreatic PLA,

structure

given in Fig. 1C

Inhibitor No.

Identity

13 14 15 16 17

(CHz)h-CH, (CHzkCH3 (CH,),,-CH, (CH,),,-CH, (CH2),-CH=CH-(CH*),

n.d. = not determined.

of Q

Z-values platelet PLA,

Z-values pancreatic

PLA,

PC

PC

PC

PC

32 30 42 21 n.d.

80 1300 2 200 1560 1200

16 13 10 I3

51 103 130 63

n.d.

85

70 interesting pharmacological compounds. Whether or not these analogues also selectively inhibit the platelet PLA, in vivo remains to be answered. References Irvine, R.F. (1982) Biochem. J. 204, 3-16. Dennis, E.A. (1987) Drug Dev. Res. 10, 205-220. Pruzanski, W., Vadas, P., Kim, J., Jacobs. H. and Stefanski, E. (1989) J. Rheumatol. 15, 791-794. Kramer. R.M., Roberts, E.F., Manetta, J. and Putnam. J.E. (19911 J. Biol. Chem. 266, 5268-5272. Seilhamer, J.J., Pruzanski, W., Vadas, P., Miller, J.A., Kloss, J. and Johnson, L.K. (1989) J. Biol. Chem. 264, 5335-533X. Kramer. R.M., Hession, C., Johansen, B.. Hayes, G., MC Gray, P., Pizchang Chow, E., Tizard, R. and Pepinski, R.B. (1989) J. Biol. Chem. 264, 5768-5775. De Haas, G.H., Dijkman, R., Ransac. S. and Verger, R. (19901 Biochim. Biophys. Acta 1046, 240-257. Ransac. S., Aarsman, A.J., van den Bosch, H.. Gancet, C., de Haas. G.H. and Verger, R., Eur. J. Biochem., in press.

Y Franken,

IO

11 12 13 14 15

16

P., Van den Berg, L., Huang, J., Gunyuzlu, P., Lugtigheid, R.B., Verheij, H.M. and De Haas, G.H. (1991) Eur. J. Biochem., in press. Cuber0 Robles. E. and Van den Berg, D. (1969) Biochim. Biophys. Acta 187, 520-526. Dijkman, R.. Dekker, N. and De Haas, G.H. (1990) Biochim. Biophys. Acta 1043. 67774. Eibl, H. and Nicksch, A. (197X) Chem. Phys. Lipids 22, I-X. Verheij, H.M., Westerman, J., Sternby, B. and De Haas. G.H. (1983) Biochim. Biophys. Acta 747, 93-99. Ransac, S., Rivibre, C.. Soulie, J.M., Gancet. C., Verger, R. and De Haas, G.H. (1090) Biochim. Biophys. Acta 1043. 57-66. Thunnissen, M.M.G.M., AB, E., Kalk, K.H., Drenth, J., Dijkstra. B.W., Kuipers, O.P.. Dijkman, R., de Haas. G.H. and Verheij, H.M. (1990) Nature 347. 6X9-691. Wery, J.P., Schevitz, R.W., Clawson, D.K., Bobbitt, J.L., Dow, E.R., Gamboa, G.. Goodson Jr., T.. Hermann, R.B., Kramer, R.M.. McClure, D.B., Mihelich, E.D.. Putnam. J.E.. Sharo. J.D.. Stark, D.H., Teater. C.. Warrick, M.W. and Jones: N.D.‘il991) Nature 352. 79-82.

Competitive inhibition of lipolytic enzymes. VI. Inhibition of two human phospholipases A2 by acylamino phospholipid analogues.

The competitive inhibition of human pancreatic and a mutant human platelet phospholipase A2 (PLA2) was investigated using acylamino phospholipid analo...
525KB Sizes 0 Downloads 0 Views