Acetylenic mechanism-based inhibitors of cholesterol side chain cleavage by cytochrome P-450scc.

The following acetylenic steroids appear to be the first reported mechanism-based inhibitors of cytochrome P-450scc: 20-(1-propynyl)-5-pregnen-3 beta, 20 alpha-diol, 20-(1-hexynyl)-5-pregnen-3 beta, 20 alpha-diol, and 20-(1,5-hexadiynyl)5-pregnen-3 beta, 20 alpha-diol. Oxygen and NADPH are required for enzymatic oxidation and all three steroids yield pregnenolone as a major product. Incubation of P-450scc with 20-(1,5-hexadiynyl)-5-pregnen-3 beta, 20 alpha-diol under turnover conditions completely inactivates the enzyme with a half-time of 11 min. The partition ratio for inactivation by the steroid was determined to be about 6 molecules of the steroid processed per molecule of P-450scc inactivated.

Compounds containing acetylenic, olefinic, or allylic functional groups seem to act as mechanism-based inhibitors of the phenobarbitol-induced liver microsomal P-450 isozymes (1)(2)(3)(4)(5)(6). We were encouraged by the findings of Ortiz de Montellano and his co-workers, who showed that acetylenes and olefins yield covalent heme adducts when acted on by the microsomal P-450 oxygenation system (4, [7][8][9], to design mechanism-based inhibitors of P-450,,, which oxidatively cleaves cholesterol to pregnenolone (10). In the present study, steroid derivatives with acetylenic side chains were prepared in an effort to devise a substrate of P-450,, which would generate a reactive species in the active site (1, ll), thus leading to mechanism-based or suicide inhibition of the enzyme (Scheme 1). These are potentially valuable for mechanistic studies of the enzyme as well as for physiological and clinical investigations in which specific chemical ablation of steroidogenesis could be of interest (12). Here, we report that several acetylenic steroids are excellent mechanism-based inhibitors of P-450,,,, although they appear to inactivate the enzyme in a manner distinct from the action of acetylenes on the microsomal enzyme.

Materials
The cytochrome P-450.,, (9 nmol of P-450/mg of protein), adrenodoxin, and adrenodoxin reductase were isolated and assayed as * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "uduertisement" in accordance with 18    Chromutography and Moss Spectroscopy-Reverse-phase high performance liquid chromatography was performed on a Waters Associates pBondapak CIS column. The applied sample was 1 ml of a saturated methanol solution. Elution was performed with methanol/ water (80:20, v/v) and the flow rate was 2 ml/min. Analytical GC2 was performed on an instrument equipped with flame ionization detectors and a glass column (2 mm, inner diameter, X 3 feet) packed with 2% SP-2100 on Supelcoport (80-100 mesh). The temperature gradient was from 210 to 300 "C a t 30 "C/min, and nitrogen gas was used as the carrier at 20 pounds. inch-'. Mass spectroscopy was performed on a Hewlett-Packard gas chromatograph-quadrupole mass spectrometer model 5992A equipped with 2% SP-2100 column (2 mm, inner diameter, x 4 feet), operated similarly to the analytical GC, with helium as carrier gas a t 30 ml/min. Enzyme Incubations and Steroid Isolation-The reconstitution system was typically 0.7 p M in P-450,, 4 FM in adrenodoxin, 0.2 p~ in adrenodoxin reductase, 16 p M in NADPH, 10 mM in MgCl,, 50 mM in MOPS, pH 7.4, 3 mM in glucose-6-P and contained 1 unit each of catalase and glucose-6-phosphate dehydrogenase/ml. Incubations were at 20 "C and reactions were started by addition of either adrenodoxin reductase or the steroid. For TLC, GC, and GC-MS analyses, the steroid products were extracted from the reaction mixture with a small column containing C,, reverse-phase gel which had been washed with methanol and equilibrated with water. The reaction mixtures were passed through the columns, washed with water, and the steroids were eluted with methanol into small test tubes. The methanol was evaporated to dryness in a vacuum desiccator.
The recovery was determined in trial cases with added [7-3H]pregnenolone standards carried through the same process. These recoveries were all in excess of 90%.
3&20cu-diol ( 4 ) to cytochrome P-450.,,, produced predominately type I1 (18)  for NADPH and oxygen are clearly seen. The change in the Soret absorbance during action of P-450 on the steroid could be titrated, with a 10-fold molar excess of steroid 4 being sufficient to titrate the system (Fig. 2). The partition ratio3 ' The ratio of the total inhibitor molecules processed to the number of inactivation events (Partition Ratio = kCat/kinact). is 5.9 f 0.5 min with 45 PM (10-fold excess) steroid. We also measured the integrity of the chromophore in the enzyme by testing its ability to form the CO complex (in the presence of excess dithionite (18)) which has a maximal absorbance a t 448 nm. A first order time-dependent decay of complexforming ability was observed with a half-life ( tlh) of 10.3 +: 0.5 min for steroid 4 (Fig. 4). For steroids 2 and 3, the rate of decrease in the absorbance at 448 nm returned to that of the control level before the ability of the enzyme to form the Fe(I1). CO complex was completely lost. Upon further addition of 2 or 3, the rate of absorbance loss a t 448 nm was not altered.
Silica gel thin layer chromatography and GC-MS revealed that steroid 1 does not yield pregnenolone or hydroxylated steroids when acted on by P-450 but that action of the enzyme on steroids 2, 3, and 4 yields pregnenolone as the major product. Oxygen and NADPH were required for turnover of the acetylenic steroids since no other steroids besides the starting materials were found in the absence of these two components.

DISCUSSION
The results of these studies indicate that the interaction of cytochrome P-450,,, with steroid 4 exhibits the characteristics of suicide substrate inhibition. The rates of decrease in both the side chain cleavage activity of 2001-hydroxycholesterol and the Soret absorbance and the loss of the ability of the enzyme to form the characteristic Fe(I1). CO complex are identical within the experimental error, with a tlh of 11 min ( k of 0.063 rnin"). Each time-dependent process exhibits first order kinetics for a t least three half-lives. These results indicate that the processing of this steroid (and, presumably, 2 and 3 as well) is accompanied by inactivation of the enzyme. The requirement for the complete three-enzyme system, in addition to NADPH and oxygen, suggests that P-450,,, catalyzes a t least one monooxygenation reaction with the acetylenic steroids prior to inactivation. Furthermore, the lack of protection by added sulfhydryl reagents suggests that the enzyme-generated inactivator does not escape from the active site.
One product of P-450.,, action on steroids 2, 3, and 4 is pregnenolone. The time dependence of pregnenolone formation during turnover of steroid 3 and 4 (Fig. 5) shows that the steroids are indeed substrates of the enzyme. The half-life of the decay of the pregnenolone production rate of approximately 7 min for steroid 4 is also in good agreement with the enzyme inactivation kinetics. To our surprise, the terminal acetylenic steroid 1 was neither an inhibitor nor a substrate.
We propose here (Scheme 1) that the inactivation of P -450.,, by the acetylenic steroids proceeds by formation of a reactive oxirene or oxirene-related species. Previous work (Refs. in Ref. 19 We thus propose that alkylation of an amino acid residue a t the active site of P-450,,,, rather than derivatization of the porphyrin, is likely to be responsible for the inactivation we have documented by the kinetic experiments presented here. In conclusion, steroid 4 (and, preliminarily, steroids 2 and 3 ) has been characterized as a potent mechanism-based inhibitor of cytochrome P-450,,,. From measurements of the amount of steroid 4 remaining after enzyme inactivation and the Soret absorbance decrease, the partition ratio has been estimated to be 6. This is in agreement with an estimate of the partition ratio derived from a ratio of the turnover number of P-450,,, with cholesterol (k,,, P 1 min") to the inactivation rate constant with 4 (kinact s 0.1 min-'). This value is an order of magnitude lower than the ratio of 200 determined for microsomal P-450 inactivation by allylisopropyl acetamide (24). Thus, the mechanism and the relative effectiveness of the suicide inhibition of hepatic microsomal and adrenal mitochondrial P-450s by alkynes appear to be interestingly diverse.