Novel inhibitors of prolyl endopeptidase: effects of stereochemistry.

Prolyl endopeptidase is a serine protease that specifically cleaves peptides on the carboxyl side of proline residues. We have show that racemic 2-(2-formyl-pyrrolidine-1-carbonyl)-2,3-dihydro-indole-1-carboxylic acid benzyl ester (IP), racemic trans-2-(2-formyl-pyrrolidine-1-carbonyl-1-cyclohexane-carboxylic acid benzyl ester (cis-CP) are slow binding inhibitors of mouse brain prolyl endopeptidase with Ki values of 0.35, 2.4, and 3 nM, respectively. In order to determine whether IP and trans/cis-CP are stereoselective in their inhibition profile, five stereoisomers were synthesized and tested for inhibition. Kinetic analysis indicates that the 2S, 2'S-isomer is necessary for inhibition by racemic IP. trans/cis-CP also requires S-stereochemistry on two of its three chiral centres; the third can be either R or S. This suggests that our novel, non-peptide inhibitors bind at the same site as peptide inhibitors which require L-configuration at the P1 and P2 binding pockets.

Beryllium competitively inhibits brain myo-inositol monophosphatase, but unlike lithium does not enhance agonist-induced inositol phosphate accumulation.

Despite limiting side-effects, lithium is the drug of choice for the treatment of bipolar depression. Its action may be due, in part, to its ability to dampen phosphatidylinositol turnover by inhibiting myo-inositol monophosphatase. Beryllium has been identified as a potent inhibitor of partially purified myo-inositol monophosphatase isolated from rat brain (Ki = 150 nM), bovine brain (Ki = 35 nM), and from the human neuroblastoma cell line SK-N-SH (Ki = 85 nM). It is over three orders of magnitude more potent than LiCl (Ki = 0.5-1.2 mM). Kinetic analysis reveals that beryllium is a competitive inhibitor of myo-inositol monophosphatase, in contrast with lithium which is an uncompetitive inhibitor. Inhibition of exogenous [3H]inositol phosphate hydrolysis by beryllium (IC50 = 250-300 nM) was observed to the same maximal extent as that seen with lithium in permeabilized SK-N-SH cells, reflecting inhibition of cellular myo-inositol monophosphatase. However, in contrast with that observed with lithium, agonist-induced accumulation of inositol phosphate was not observed with beryllium in permeabilized and non-permeabilized SK-N-SH cells and in rat brain slices. Similar results were obtained in permeabilized SK-N-SH cells when GTP-gamma-S was used as an alternative stimulator of inositol phosphate accumulation. The disparity in the actions of beryllium and lithium suggest that either (1) selective inhibition of myo-inositol monophosphatase does not completely explain the action of lithium on the phosphatidylinositol cycle, or (2) that uncompetitive inhibition of myo-inositol monophosphatase is a necessary requirement to observe functional lithium mimetic activity.

Slow tight-binding inhibition of prolyl endopeptidase by benzyloxycarbonyl-prolyl-prolinal.

Prolyl endopeptidase is a serine proteinase that specifically cleaves peptides on the carboxy side of proline residues. Wilk & Orlowski [(1983) J. Neurochem. 41, 69-75] have shown that benzyloxycarbonyl-prolyl-prolinal (Z-prolyl-prolinal) is a potent inhibitor of prolyl endopeptidase. We show that Z-prolyl-prolinal is a slow-binding inhibitor of mouse brain prolyl endopeptidase with Ki 0.35 +/- 0.05 nM. Kinetic analysis indicates that the mechanism is a simple, but slow, reversible equilibrium between free and bound enzyme (E + I in equilibrium EI) with rate constants for association (kon) and dissociation (koff) of 1.6 X 10(5) M-1.s-1 and approx. 4 X 10(-5) s-1 respectively. Slow-binding inhibition is dependent on the presence of the aldehyde group since the alcohol (Z-prolyl-prolinol) is a rapid and 50,000-fold poorer inhibitor (Ki 19 microM). Prolyl endopeptidase from human brain is also inhibited by Z-prolyl-prolinal with kinetics similar to those of the mouse brain enzyme.