An investigation into fungal metabolism of halogenated and related steroids

Fungal metabolism of halogenated and related steroids was investigated. The fungi Aspergillus niger ATCC 9142, Curvularia lunata NRRL 2380 and Rhizopus stolonifer ATCC6227b were studied in this regard. 2l-Fluoro-, 2l-chloro, 2l-bromo- and 2l-methyl-pregn-4-ene-3,20diones were prepared and incubated with ~ niger (a C-2l-hydroxylator) in order to observe the effect of the C-2l substituent on the metabolism of these substrates. In all four cases, the C-2l substituent prevented any significant metabolism of these substrates. llB-Fluoropregn-4-ene-3,20-dione was prepared and incubated with C. lunata (an llB-hydroxylator) and ~ stolonifer (an lla-hydroxylator). With ~ lunata, the ll-fluoro- substituent prevent hydroxylation at the 11 position, but diverted it to a site remote from the fluorine atom. In contrast, with ~ stolonifer the llB-fluoro- substituent, although slowing the apparent rate of hydroxylation, did not prevent its occurrence at the 11a- position. llB-Hydroxypregn-4-ene-3,20-dione was also incubated with R. stolonifer. The llB-hydroxy-;group did not appear to have any significant effect on hydroxylation at the lla- position. The incubation of a substrate, unsaturated at a favoured site of hydroxylation with Rhizopus arrhizus ATCC 11145 provided a complex mixture of products; among them were both the a and S epoxides. The formation of these products is rationalized as arising because of the lack of regio- and stereospecificity of the hydroxylase enzyme(s) involved.

The 21-dehydroxylation of corticosteroids: evidence for an enol intermediate and the hydroxylation of steroids by fungal cyto-chrome P450: evidence for a stepwise mechanism

Two enzyme mechanisms were examined: the 21-dehydroxylation of corticosteroids by the anaerobe Eubacterium l en tum, and the hydroxylation of steroids by fungal cytochrome P450. Deuterium labelling techniques were used to study the enzymic dehydroxylation. Corticosteroids doubly labelled (2H) at the C-21 position were incubated with a culture of Eubacterium lentum. It was found that t he enzymic dehydroxylation proceeded with the loss of one 2H f rom C-21 per molecule of substrate. The kinetic isotope ef fect f or the reaction was found to be k~kD = 2. 28. These results suggest that enzyme/substr ate binding in this case may proceed via t he enol form of the substrate. Also , it appears that this binding is, at least in part, the rate determining step of t he reaction. The hydroxylation of steroids by fungal cytochrome P450 was examined by means of a product study. Steroids with a double bond at the A8 (9), ~( lO ), or ~ (ll) position were synthesized. These steroids were then incubated with fungal strains known to use a cytochrome P450 monooxygenase to hydroxylate at positions allylic to these doubl e bonds. The products formed in these incubations indicated that the double bonds had migrated during allylic hydroxylat ion. This suggests that a carbon centred radical or ion may be an intermediate i n the cytochrome P450 cat alytic cycle.

The investigation of the reduction mechanism of the 5a-reductase enzyme of Penicillium decumbens /

The 5a-reductase of Penicillium decumbens ATCC 10436 was used as a model for the mammalian enzyme to investigate the mechanism of reduction of testosterone to 5adihydrotestosterone . The purpose of this study was to search for specific 5a-reductase inhibitors which antagonize prostate cancer . In a whole-cell biotransformation mode, this organism reduced testosterone (1) to 5a-dihydrosteroids (8) and 5aandrostane- 3, 17-dione (9) in yields of 28% and 37% respectively. Control experiments have shown that 5aandrostane- 3, 17-dione (9) can be produced from the corresponding alcohol (8) in a subsequent reaction separate from that catalysed by the 5a-reductase enzyme . Androst-4- ene-3, 17-dione (2) is reduced to give only (9) with a recovery of 80% The stereochemistry of the reduction was determined by 500 MHz ^H NMR analysis of the products resulting from the deuterium labelled substrates. The results were obtained by an analysis of the NOE difference spectra, double-quantum filtered phase sensitive COSY 2-D spectra, and ^^c-Ir 2-D shift correlation spectra of deuterium labelled products. According to the unambiguous assignment of the signals due to H-4a and H-4Ii in 5a-dihydro steroids, the NMR data show clearly that addition of hydrogen to the 4{5)K bond has occurred in a trans manner at positions 413 and 5a. To Study the reduction mechanism of this enzyme, several substrates were prepared as following; 3-methyleneandrost-4-en- 17fi-ol(3), androst-4-en-17i5-ol(5) , androst-4-en-3ii, 17fi-diol (6) and 4, 5ii-epoxyandrostane-3, 17-dione (7) . Results suggest that this enzyme system requires an oxygen atom at the 3-position of the steroid in order to bind the substrate. Furthermore, the mechanism of this 5a-reductase may proceed via direct addition of hydrogen at the 4,5 position without involvement of a carbonyl group as an intermediate.

An investigation into some mechanistic aspects of estrogen biosynthesis

The mechanistic aspects of the 19-hydroxy1ation and aromatization of androgens were investigated. Fungal, bacterial and mammalian enzymatic activities were studied in this regard . The fungus Pell i cular~ fi1amentosa metabolized androst-4-ene-3 , 17-dione to the corresponding 110<' , 11 f and 14 0( hydroxylated derivatives. No ~19- hydroxylated products were isolated, although this transformation was previously observed for the C21-steroids . The intestinal bacterium Clostridi um paraputrific~ had been reported to aromatize androsten-4-ene-3,17-dione. In the present study, however, only the ring A reduced products , 17(3 - hydroxy-5f -andro8tane- 3-one and 5f-androstane-3,17-dione , were recovered . Human placental microsomes contain substantial aromatase activity and were employed in an effort to elucidate some of the mechanistic details of aromatization. Selectively deuterated steroidal substrates were employed as a probe in order to distinguish b'!tween certain of the mechanisms proposed for aromatization . Retention of deuterium at C4 and C6 was observed. It was concluded that no free intermediates allowing for loss of hydrogen from either of these two positions are implicated in this process . The involvement of a Schiff base enzyme-sup strate complex in aromatization was examined using the substrate 17f - hydroxyandrost-4-ene-3-one- 3_ 1BO. Since no loss of label was ob~erved, the implication of a Schiff base was discounted . Mixed label1ir~ studies were performed in order to determine if hydroxylation at C19 is a rate-determining process in aromatization . Isotope effects of 2 .1 and 1.7 were determined for the conversion of 17f - hydroxyandrost-4-ene-J-one-19,19,19-dJ and -19-dl respectively to estrogens. It was concluded from this that 19-hydroxylation is at l east a partially rate-determinjng process in aromatization. A homoenb~ation mechanism for 19-hydroxylation was not supported by the data obtained in this s tudy. In vitro 1JC NMR monitoring using l7f-hydroxyandrost-4-ene-Jone- 19-l3C was found not to be a successful approach in the study of steroid transformations, owing in part t o their low solubility in the incubation medium.

Reactions of steroidal epoxides with strong organic bases and an investigation into the syntheses of some labelled pregnane derivatives

Reactions of 5,6- and 4,5-epoxycholestane derivatives with strong bases were investigated. Epoxidation of 3a-acetoxycholest-5-ene also gave a new compound along with the anticipated epoxides. Interconversions of the latter were observed. Some possible mechanisms of its formation and rearrangements have been pIioposed. No reaction was observed with any of the 5,6- and 4,5-steroidal epoxides employed in the present study, using potassium tertiary butoxide under refluxing conditions. n-Butyllithium reacted only with 5,6-epoxycholestanes bearing a ketal moiety at the C3 carbon. Opening of the ketal group was observed with n-butyllithium in the case of a ~-epoxide. The reaction was also investigated in the absence of epoxide functionality. A possible mechanism for the opening of ketal group has been proposed. Lithium diethylamide (LDEA) was found effective in rearranging 5,6- and 4,5-epoxides to their ~orresponding allylic alcohols. These rearrangements presumably proceed via syn-eliminations, however the possibility of a corresponding anti-elimination has not been eliminated. A substituent effect of various functional groups (R = H, OH, OCH2CH20) at C3 has-been observed on product distribution in the LDEApromoted rearrangements of the corresponding epoxides. No reaction of these epoxides was observed with lithium diisopropylamide (LDA) • In the second part of the project, several attempts were made towards the sYRthesis of deoxycorticoste~one~17,2l,2l~d3' a compound desirable for the 2l-dehydroxylation studies of deoxycorticosterone. Several routes were investigated, and some deuterium labelled pregnane derivatives were prepared in this regard. Microbial 21-hydroxylation of progesteronel7,21,21,2l- d4 by ~ niger led to loss of deuterium from C21 of the product. An effort was made to hydroxylate progesterone microbially under neutral condtions.

The 21-dehydroxglation of corticosteroids of Eubacterium lentum and the dehalogenation of 4-chlorobenzoic acid by 4-chlorobenzoate dehalogenase

Two enzyme mechanisms were investigated: the 21-dehydroxylation of corticosteroids by Eubacterium lentum and the dehalogenation of 4-chlorobenzoic acid by Pseudomonas sp. CBS 3. , Chemical and enzymic methods of reduction of 21-oxo steroids were used to generate C-21-d1 compounds of tetrahydrodeoxycorticosterone, with both predominant stereochemistries. It was found that during the dehydroxylation the pro-S hydrogen at the C-21 position was lost preferentially. This suggests that the enzyme removes the pro-S hydrogen during binding to the active site as the ene-diol. To study the hydrolytic replacement of chlorine by hydroxyl , p-chlorobenzoic acid-d4 was prepared and sent to Germany for an ~ncubation with an enzyme preparation of 4-Chlorobenzo~te Dehalogenase. Results suggests the possible loss of deuterium during the conversion of p-chlorobenzoate to p-hydroxybenzoate, from all four ring positions. Many methods of preparing the control compound p-hydroxybenzoic acid-d4 were investigated.