Chemical suppression of steroidogenesis.

A large number of chemicals are known to interfere with steroidogenesis in the adrenal cortex and other tissues. Many xenobiotics inhibit steroid hormone production as a result of interactions with cytochrome P-450-containing hydroxylases in adrenal mitochondria or microsomes. For example, metyrapone, a compound used clinically in the evaluation of pituitary-adrenocortical function, binds to various cytochromes P-450 in the adrenal, preventing the interactions of steroid substrates with the enzymes and inhibiting steroidogenesis. The mineralocorticoid antagonist, spironolactone, and its major circulating metabolite, canrenone, also competitively interact with adrenal steroid hydroxylases. In addition, spironolactone is converted by adrenal microsomes to an unknown metabolite which promotes the destruction of cytochromes P-450, decreasing the activities of steroid hydroxylases. Carbon tetrachloride is similarly "activated" by adrenal microsomal mixed function oxidases resulting in a decline in steroidogenic enzyme activity. Carbon tetrachloride (in the presence of NADPH) initiates lipid peroxidation in adrenal microsomes but its toxic effects on steroid hydroxylases are fully demonstrable when lipid peroxidation is inhibited by EDTA. A number of heavy metals, including cadmium, also inhibit adrenal steroid hydroxylases. When incubated with adrenal microsomes, cadmium does not affect cytochrome P-450 levels but decreases basal and substrate stimulated NADPH-cytochrome P-450 reductase activity. Although inhibitory effects of many chemicals on steroidogenesis have been described, the toxicological significance as well as definitive mechanisms of action have in most cases yet to be determined.

A large number of chemicals are known to interfere with steroidogenesis in the adrenal cortex and other tissues. Many xenobiotics inhibit steroid hormone production as a result of interactions with cytochrome P450-containing hydroxylases in adrenal mitochondria or microsomes. For example, metyrapone, a compound used clinically in the evaluation of pituitary-adrenocortical function, binds to various cytochromes P450 in the adrenal, preventing the interactions of steroid substrates with the enzymes and inhibiting steroidogenesis. The mineralocorticoid antagonist, spironolactone, and its major circulating metabolite, canrenone, also competitively interact with adrenal steroid hydroxylases. In addition, spironolactone is converted by adrenal microsomes to an unknown metabolite which promotes the destruction of cytochromes P-450, decreasing the activities of steroid hydroxylases. Carbon tetrachloride is similarly "activated" by adrenal microsomal mixed function oxidases resulting in a decline in steroidogenic enzyme activity. Carbon tetrachloride (in the presence of NADPH) initiates lipid peroxidation in adrenal microsomes but its toxic effects on steroid hydroxylases are fully demonstrable when lipid peroxidation is inhibited by EDTA. A number of heavy metals, including cadmium, also inhibit adrenal steroid hydroxylases. When incubated with adrenal microsomes, cadmium does not affect cytochrome P-450 levels but decreases basal and substrate stimulated NADPH-cytochrome P-450 reductase activity. Although inhibitory effects of many chemicals on steroidogenesis have been described, the toxicological significance as well as definitive mechanisms of action have in most cases yet to be determined.
There are numerous substances which are known to interfere with the production of steroid hormones. The types of compounds range from therapeutic and diagnostic agents to environmental toxins and chemical carcinogens. In addressing the topic of chemical suppression of steroidogenesis I shall not attempt to catalog all of those compounds. Instead, this presentation will concentrate on a few different classes of chemical agents with the objective of illustrating several mechanisms through which chemicals can affect steroid hormone production. For the sake of simplicity and uniformity, only the effects of chemicals on steroidogenesis in the adrenal cortex will be discussed although in many cases similar effects are also manifested in other steroid-producing tissues. Furthermore, in keeping with the theme of this conference, greater emphasis will be placed on those compounds whose effects on steroidogenesis are more in the realm of toxicology than on those drugs which have been developed specifically to inhibit steroidogenesis.

April 1981
Before discussing the actions of chemicals on steroidogenesis, it is important to introduce some background material on the types of enzymatic reactions involved in the production of steroid hormones so that the actions of chemicals on these enzymes can be more fully appreciated. Figure 1 illustrates a portion of the steroidogenic pathway which exists in the adrenal cortex. Some of the pathway overlaps with that in other steroidogenic tissues. However, production of glucocorticoids, such as cortisol, and of mineralocorticoids, like aldosterone, is specific for the adrenal. Of particular interest are the role of cholesterol as the common precursor to all steroid hormones and the number of hydroxylation reactions involved in the steroidogenic pathway. For example, note the 111, 18, 17a, and 21-hydroxylases as well as the cholesterol sidechain cleavage reaction which includes 2 hydroxylation steps (1,2). Thus, hydroxylases are critically important in steroidogenesis and as will be seen, represent the sites of action of many chemicals which influence steroidogenesis.
The adrenal steroid hydroxylases are typical mixed function oxidases which contain the heme-119 protein, cytochrome P-450. They are similar to those hepatic enzymes involved in the oxidative metabolism of drugs, carcinogens and other xenobiotics (3,4). All require NADPH and molecular 02 as cofactors and are membrane-bound enzymes. Some are located in mitochondria and others in the endoplasmic reticulum of the cell (1-4). The mitochondrial and microsomal enzymes differ somewhat in their composition but for the purpose of this discussion, all can be thought of as two functional units, the cytochrome P-450 to which the steroid substrate binds, and a reductase component which provides electrons derived from NADPH to the cytochrome P-450. Once the substrate binds to the cytochrome P-450, the reductase provides electrons to reduce the substrate-cytochrome complex and also to promote activation of oxygen for incorporation into the steroid nucleus. The specific site of hydroxylation of the steroid as well as the substrate specificity of the enzyme is governed by the cytochrome P-450 component. The number of different cytochromes P-450 involved in steroidogenesis has yet to be determined, but it is possible, perhaps even likely, that each hydroxylase contains its own distinct cytochrome P-450 moiety. Both the reductase and cytochrome P-450 components of the enzyme are essential for hydroxylase activity.
A large number of the chemicals which have been employed clinically as inhibitors of steroidogenesis interact with various cytochromes P450 in steroidproducing tissues and inhibit the binding of normal steroid substrates, thereby decreasing hydroxylase activities (5). One of the best known and most thoroughly studied of this group of compounds is the pyridine derivative, metyrapone (SU-4885), which is widely used in the evaluation of pituitaryadrenocortical function. By inhibiting cortisol pro-120 duction, metyrapone should elicit an increase in pituitary ACTH secretion and thereby serves as a test of pituitary secretory reserve (5). When metyrapone is added to adrenal mitochondrial preparations, it produces a characteristic difference spectrum, indicative of its binding to cytochromes P-450 (6). The spectrum is characterized by an increase in absorbance at about 425 nm and a decrease at about 400 nm and is known as a type II spectral change. The size of the spectrum increases with increasing concentrations of metyrapone until its binding sites on the enzyme are saturated. In the presence of metyrapone the size of the spectral change resulting from substrate interactions with adrenal mitochondrial cytochromes P-450 is diminished (5, 6), illustrating the mechanism of action of the drug on steroid hydroxylation reactions.
Although originally proposed to be a relatively specific ll,B-hydroxylase antagonist, metyrapone has been found to inhibit various other steroid hydroxylation reactions, including 18-and 19-hydroxylation and cholesterol sidechain cleavage (5,7). Each of these enzymes is located in mitochondria and the results of several early studies suggested that adrenal microsomal hydroxylases were not affected by metyrapone. Neither 21-hydroxylase nor 17a-hydroxylase activity was inhibited by metyrapone in rat or bovine adrenals (8,9). However, the specificity of the actions of metyrapone appears to be species-dependent (10). In the rat adrenal ( Fig. 2), for example, concentrations of metyrapone which produced inhibition of mitochondrial 11,-hydroxylation 1Wad no effects on microsomal 21-hydroxylase activity. In the guinea pig ( Fig. 3), by contrast, metyrapone had similar effects on ll3-and 21-hydroxylation. Observations such as these indicate that care must be taken when using compounds like metryapone for experimental purposes since the specificity of their actions is not as great as some earlier studies suggested and is influenced by a number of variables.  April 1981 the mineralocorticoid antagonist, spironolactone (SL). SL is used as a diuretic and in the treatment of some types of hypertension. Several clinical observations indicated that SL might interfere with adrenal steroidogenesis in man (11,12) and prompted a number of studies on its effects in experimental animals (13-1 7). Following the administration of SL (25 mg/kg/day for 3 days) to guinea pigs there was a decline in cortisol production by adrenal sections in vitro (Table 1) (15). The deficit was apparent whether or not the adrenals were incubated with ACTH to stimulate steroidogenesis. Indirect effects of SL on the adrenal cortex resulting from inhibition of ACTH secretion were excluded by comparing the actions of SL with those of cortisone, a potent inhibitor of pituitary ACTH secretion (15). Once the inhibitory effect of SL on adrenocortical function was established, further studies were carried out to determine its mechanism of action as well as the role of canrenone, the major circulating metabolite of SL, in the actions of SL on the adrenal.
Either SL or conrenone administration to guinea pigs decreased the activities of several steroid hydroxylases in both adrenal mitochondria and microsomes ( Table 2) (16). The effects of SL on the microsomal enzymes were greater than those of canrenone. The decline in steroidogenic activity could be attributed in part to the direct interactions of SL and canrenone with steroid hydroxylases. Both compounds produced type I difference spectra in adrenal mitochondria and microsomes, indicative of binding to cytochrome(s) P-450, and interfered with the binding of steroid substrates to cytochromes P-450 (16). As a result, SL and canrenone were direct inhibitors ofmitochondrial and microsomal steroid hydroxylation in vitro (Figs. 4 and 5). Each compound produced a concentration-dependent inhibition of mitochondrial 11-hydroxylation and microsomal 21-hydroxylation. The effects ofcanrenone on both enzymes were greater than those of SL. Thus, the relative potencies in vitro were opposite those obtained after in vivo administration of SL and canrenone (Table 2). However, direct competition for binding to cytochromes P-450 provided only part of the explanation for the actions of SL on steroidogenesis. SL treatment also resulted in a decline in the amount of cytochromes P-450 in adrenal mitochondria and microsomes (Table 3). Canrenone, in contrast, had no effect on cytochrome P-450 concentrations, which may account for the greater potency of SL in vivo on steroidogenic enzymes. The effect of SL on adrenal cytochrome P-450 levels could also be demonstrated in vitro under the appropriate experimental conditions (14,16). When 11B -HYDROXYLATION  adrenal microsomes were incubated with SL plus NADPH, a decline in cytochrome P-450 levels resulted (Fig. 6). Omission of either SL or NADPH from the incubation medium eliminated the effect. In contrast, incubation with canrenone, with or without NADPH, did not affect cytochrome P-450 content. The decline in cytochrome P-450 levels caused by SL plus NADPH was accompanied by a fall in microsomal heme content. The specificity of action of SL on cytochrome P-450 is indicated by the absence of effects on the other major adrenal microsomal hemeprotein, cytochrome b5 (Fig. 6). These and other observations indicate that SL is  converted by the adrenal cortex to a metabolite which promotes the destruction of adrenal cytochrome(s) P-450. In subsequent studies (17) it was found that only analogs of SL with the sulfur group at the C-7 position decrease cytochrome P-450 levels. Thus, SL appears to interfere with steroidogenesis through two related but distinct mechanisms: (1) SL and/or one or more of its metabolites compete(s) with normal steroid substrates for binding sites on cytochrome P-450, and (2) SL is converted to a toxic metabolite by adrenal enzymes, resulting in the destruction of cytochromes P-450 and loss of hydroxylase activity. The identity of the active metabolite is presently unknown and is now being pursued in several laboratories. Another compound which requires conversion to April 1981 an active metabolite before its toxic effects are manifested is carbon tetrachloride (CC14) (18). The hepatotoxic effects of CC14 are well known and have been extensively studied for many years. Less well known is that CC14 poisoning also produces adrenal necrosis in man as well as in some experimental animals (19). Until recently, little was known about the specific actions of CC14 on the adrenal cortex. When CC14 was given to guinea pigs and the animals killed 48 hr later, changes in adrenal microsomal steroid hydroxylases were found (Table  4). Both 17aand 21-hydroxylase activities in ad enal microsomes were decreased by CC14 treatment but the mitochondrial ll,-hydroxylase was unaffected. Similarly, microsomal but not mitochondrial cytochrome P-450 levels were decreased by CC14. It is reasonably well established that in the liver, the toxic effects of CC14 depend upon its conversion by cytochrome P-450-containing enzymes to the highly reactive trichloromethyl radical (18). However, the nature of the subsequent events in CC14 toxicity is somewhat controversial; the controversy is centered about the involvement of lipid peroxidation. The two prevalent hypotheses for the mechanism of action of CC14 include the conversion of CC14 to the trichloromethyl radical. However, one hypothesis advocates a direct attack of the radical on cellular macromolecules, including cytochromes P-450, resulting in the destruction of enzymes. The other hypothesis proposes an obligatory role for lipid peroxidation in the actions of CCl4; that is, an attack on the unsaturated fatty acids of membrane phospholipids by the trichloromethyl radical, resulting in hydroperoxide fonnation and membrane disruption and secondarily causing a loss of cytochrome P-450 and associated enzyme activities.
In examining the effects of CC14 on adrenal steroidogenic enzymes, we attempted to determine iflipid peroxidation was involved. Lipid peroxidation activity was assayed as the formation of malon- aldehyde, one of the breakdown products of hydroperoxides. In several preliminary experiments we found that the effects of CC14 on adrenal microsomal enzymes were readily demonstrable in vitro (Fig. 7). When CC14 was incubated with adrenal microsomes plus NADPH, a decline in the rates of steroid 17a-and 21-hydroxylation resulted. The decline in enzyme activity was dependent upon the duration of incubation and required the presence of NADPH, suggesting that CC14 was converted to an active metabolite by adrenal microsomes. Incubation with CC14 plus NADPH also produced a decrease in adrenal microsomal cytochrome P-450 content and, as previously found in the liver, CC14 7a -Hydroxylase 21-Hydroxylose . Effects of preincubation of guinea pig adrenal microsomes with carbon tetrachloride (CC14) on steroid hydroxylation and lipid peroxidation. Microsomes (2.0 mg protein/ml) were incubated in 1.15% KCI-0.05M Tris-HCI (pH 7.4) for 60 min at 3700 under air. NADPH (0.4mM) and/or CC14 (5 pl/ml) were added prior to the start of incubation. Lipid peroxidation was measured as the rate of production of malonaldehyde (MDA). 124 plus NADPH stimulated lipid peroxidation in adrenal microsomes.
To determine if lipid peroxidation was essential for the toxic effects of CC14 on adrenal steroidogenic enzymes, similar incubations were carried out in the presence and absence of EDTA, a well known inhibitor of lipid peroxidation (Fig. 8). The lipid peroxidation initiated by CC14 plus NADPH was almost completely inhibited when 1.0 mM EDTA was also present in the incubation medium. Nonetheless the decreased in 17a-and 21-hydroxylase activities produced by NADPH plus CC14, as well as the decline in cytochrome P-450 levels, were unaffected by the EDTA. The time course for the decreases in 17aand 21-hydroxylase activities during incubation of adrenal microsomes with CC14 plus NADPH was also unaffected by EDTA (Fig.  9). These observations indicate that CC14 can indeed stimulate lipid peroxidation in adrenal microsomes but that the lipid peroxidation is not obligatory for the toxic effects of CC14 on adrenal microsomal enzymes.
Although we have not been able to demonstrate any cause and effect relationship between lipid peroxidation and CC14 toxicity, the significance of lipid peroxidation in relation to steroidogenic enzyme activity remains an interesting question. It is known that lipid peroxication has deleterious effects on the membranous components of cells (18) and steroid hydroxylases are membrane-bound enzymes. However, relatively little is known about lipid peroxidation in steroid-producing tissues. The relationship between lipid peroxidation and steroidogenesis may provide fertile ground for future research since some of the factors known to promote lipid peroxidation, such as ascorbate, iron, and NADPH, are found in relatively high concentrations in steroidogenic tissues like the adrenal cortex.
The compounds discussed thus far all exert effects on steroidogenesis as a result of actions on the cytochrome P450 portion of steroid hydroxylases.
Environmental Health Perspectives  21), and recently we found that cadmium administration to guinea pigs produced substantial decreases in adrenal steroid hydroxylases as well.
The actions of cadmium on adrenal microsomal steroid metabolism are also demonstrable in vitro. Incubation of adrenal microsomes with Cd at 37°C produced concentration-dependent decreases in steroid 21-hydroxylase activity (Fig. 10). Similar changes in microsomal 17a-hydroxylase and mitochondrial 11-hydroxylase activities were also observed. The decreases in 21-hydroxylase activity were proportional to the length of incubation and were temperature-dependent. Incubations at 0°C had no effects on enzyme activity. The time-course for the changes in 21-hydroxylase activity in the presence of 50,uM cadmium is shown in Figure 11. Incubation for 60 min produced almost complete loss of enzyme activity. Despite the large decrease in the rate of steroid metabolism, there was no change in cytochrome P-450 content in the microsomes. In addition, cadmium did not affect the interaction of steroid substrates with cytochromes P-450, as indicated by the absence of change in the size of the type I spectral change produced by 17a-hydroxyprogesterone, a substrate for 21-hydroxylation. However, cadmium produced decreases in NADPH-cytochrome P-450 reductase activity which were proportionately similar to the decreases in 21-hydroxylase activity. Thus, cadmium appears to exert its inhibitory effects on steroidogenesis via selective actions on the reduc-  Fig. 10. Cytochrome P-450 was measured as the dithionite reduced CO complex and NADPH-cytochrome P-450 reductase activity was determined in the presence of 17-hydroxyprogesterone (5 x 10-5M), a substrate for 21-hydroxylation. The type I spectral change was obtained by addition of 5 x 10M 17-hydroxyprogesterone to adrenal microsomal suspensions. tase portion of the steroid hydroxylase and not on the cytochrome P-450. Preliminary studies indicate that other metals such as zinc and lead act in a similar manner. The compounds which have been discussed in this presentation are but a few of the many chemicals that can affect steroidogenesis. The toxicological significance of these agents as well as definitive mechanisms of action have in most cases yet to be established but represent areas of ongoing investigation in a number of laboratories. However, it should be borne in mind that there also remain many gaps in our knowledge of the fundamental mechanisms in the regulation of steroid hormone production. Only with a more complete understanding of the processes involved in steroidogenesis are we likely to be able to fully comprehend the actions of chemicals on this important metabolic pathway.