Hormonal and developmental regulation of expression of the hepatic microsomal steroid 16 alpha-hydroxylase cytochrome P-450 apoprotein in the rat.

The hormonal regulation of the sexually differentiated cytochrome P-450 isozyme which catalyzes 16 alpha-hydroxylation of testosterone and 4-androstene-3,17-dione in male rat liver (P-450(16) alpha) was investigated. Estradiol valerate injection of male rats caused a decrease in P-450(16) alpha levels to almost the female level, while methyltrienolone injection had the reverse effect in female animals. Hypophysectomy abolished the sex difference in P-450(16) alpha levels. Human growth hormone infusion into male rats, mimicking the female pattern of growth hormone secretion, caused a feminization of P-450(16) alpha levels. The same effect was also seen in hypophysectomized rats of both sexes. In contrast, a different administration schedule involving 12 h injections of human growth hormone, mimicking the male pattern of growth hormone secretion, caused a masculinization of P-450(16) alpha levels in hypophysectomized rats, at a daily dose which causes feminization when given by infusion. Thus, the level of expression of P-450(16) alpha in the liver is dependent on the temporal pattern of blood growth hormone levels. While infusion of rat growth hormone into male rats also feminized the P-450(16) alpha levels, infusion of ovine prolactin had no effect. Ontogenic studies showed that the developmental pattern of P-450(16) alpha expression in the liver coincided with the known pattern of development of the sexual differentiation of hepatic steroid 16 alpha-hydroxylase activity and of the diurnal pattern of growth hormone secretion.

The hormonal regulation of the sexually differentiated cytochrome P-450 isozyme which catalyzes 16ahydroxylation of testosterone and 4-androstene-3,17dione in male rat liver (P-45O1&) was investigated. Estradiol valerate injection of male rats caused a decrease in P-45016, levels to almost the female level, while methyltrienolone injection had the reverse effect in female animals. Hypophysectomy abolished the sex difference in P-45016, levels. Human growth hormone infusion into male rats, mimicking the female pattern of growth hormone secretion, caused a feminization of P-45016, levels. The same effect was also seen in hypophysectomized rats of both, sexes. In contrast, a different administration schedule involving 12 h injections of human growth hormone, mimicking the male pattern of growth hormone secretion, caused a masculinization of P-45016, levels in hypophysectomized rats, at a daily dose which causes feminization when given by infusion. Thus, the level of expression of P-45016, in the liver is dependent on the temporal pattern of blood growth hormone levels. While infusion of rat growth hormone into male rats also feminized the P-45016, levels, infusion of ovine prolactin had no effect. Ontogenic studies showed that the developmental pattern of P-45016, expression in the liver coincided with the known pattern of development of the sexual differentiation of hepatic steroid 16a-hydroxylase activity and of the diurnal pattern of growth hormone secretion.
The steroid 16a-hydroxylase activity of rat liver microsomes is one of a number of sexually differentiated liver functions which are under a unique type of hormonal regulation (1, 2). The differential expression of these functions is controlled at two levels: the sexually dimorphic, diurnal variation in blood growth hormone levels appears to be responsible for the regulation of the specific liver functions, while the presence of androgens in the neonatal period imprints the *This work was supported by Grants 13X-2819 and 03V-6456 from the Swedish Medical Research Council. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. hypothalamus to generate a male profile of GH1 secretion, and hence to express male-specific liver functions in the adult phase. This profile is maintained by circulating androgens in adult life.
We, and others, have shown that androgen imprinting, resulting in male levels of 16a-hydroxylase activity, is achieved by regulation of the level of the specific P-45Ols, apoprotein in the microsomes (3,4). Thus, levels of P-45016, are approximately 20-fold higher in mature male rats than in mature female rats or in adult males which were castrated in the neonatal period. In order to investigate further the mechanisms of gonadal and pituitary regulation of this enzyme in the rat, we have studied the role of pituitary hormones on the regulation of P-45016, apoprotein levels in rat liver. In addition, the ontogenesis of P-45Ol6, expression in rats of both sexes was investigated. In doing so, we also wished to test the suitability of P-45016, as a model protein for the study of mechanisms of GH action in the liver.

EXPERIMENTAL PROCEDURES
The preparation of hepatic microsomes, the purification of P-45016,, and preparation of specific, immunoabsorbed rabbit anti-P-4501k IgG have been described previously (4). Specific levels of P-45016~ in microsomes were quantitated by a modification of the method of Towbin et al. as described in Ref. 4. The proteins, separated by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate and blotted on nitrocellulose filters, were incubated at a primary antibody concentration of 20 pg of IgG/ml. The antigenantibody complex was detected using '251-Protein A (2-10 pCi/mg, New England Nuclear) followed by autoradiography. The radioactivity in the bands was then counted. The radioactivity was proportional to the amount of purified or microsomal protein applied on the gel within the range of 50-500 ng and 5-50 wg, respectively.
Continuous infusion of hormones was achieved by administration from Alzet osmotic minipumps (Alza Corp., Palo Alto, CA) implanted subcutaneously on the back of the animals. Minipumps with filling volumes of 225 and 2000 pl, and pumping rates of 1 and 10 pl/h at 37 "C, respectively, were used. Control animals were implanted with a piece of silicone tubing of similar size. Unless otherwise stated, hormones were dissolved in physiological saline for administration to the rats. Where necessary, a small amount of NaOH was added to dissolve GH, and the pH of the solution was readjusted to 7.4.
The major phenobarbital-inducible P-450 isozyme also has steroid 16a-hydroxylase activity but is present at extremely low levels in uninduced rats and is not sexually differentiated. 11895 and ovine prolactin (P-S-14, 31 units/mg) were kindly supplied by the National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases. Sprague-Dawley rats were used throughout this study. Hypophysectomized rats were obtained from Mdllegards Avelslaboratorium, Skensved, Denmark.
All results are expressed as the mean f S.E. for each group of animals. One-way analysis of variance was used to test for significant differences among the means. The level of significance was set at p < 0.05.

RESULTS AND DISCUSSION
The expression of the hepatic 16a-hydroxylase apoprotein in adult male rats is imprinted by androgen in the neonatal period (3, 4), and it is known that the male level of 16ahydroxylase activity is also maintained by circulating androgen in the adult animal (1,9). However, when administered to adult rats, exogenous sex steroids can masculinize or feminize the hepatic pattern of steroid metabolism (1,10,11). Therefore, we investigated the effects of androgen and estrogen administration on P-45016, levels in the adult rat. As observed in Table I, injection of the potent synthetic androgen methyltrienolone to female rats raised the P-45016, content of hepatic microsomes by approximately 60-fold, to 61% of the male value. Conversely, a single injection of estradiol valerate in male rats caused a 96% decrease in the P-450160r level to a value which was only slightly higher than the female value. These striking effects of sex steroids are similar to those observed for 16a-hydroxylase activity (1, lo), which are only seen in the presence of an intact pituitary (1,11).
Since evidence is accumulating that the episodic pattern of pituitary growth hormone secretion is the major determining factor in regulation of sexually differentiated liver function (1, 12), we studied the involvement of GH in the regulation of the microsomal P-45016, levels. The episodic pattern of plasma GH levels differs markedly between the sexes. Male rats show regular peaks of high GH levels every 3-4 h, interspersed by troughs with almost null levels (13). In contrast, females have peaks only at irregular intervals, and the trough values are higher than in the male. This sex difference in GH secretory pattern is seen in adult animals only, the sexual differentiation beginning at puberty (13). Thus, by infusing GH continuously by way of osmotic minipumps, we are able to mimic the female pattern of GH levels. Alternatively, we can try to simulate the male pattern of GH levels by injecting GH intermittently. Hypophysectomy of male rats caused a decrease, and of female rats an increase, in P-45016, levels to values intermediate to those of the intact males and females (Fig. 1). Thus, the sex difference in P-45016, was TABLE I Effects of androgen and estrogen administration on p-45ol6, levels in intact adult rats Estradiol valerate was administered as a single intramuscular injection of 125 pg to male rats, which were killed 14 days later. Methyltrienolone (R1881) was given subcutaneously once daily for 14 days, 125 pg in 0.1 ml of propylene glycol, to female rats. Control rats received vehicle only. The rats were killed at the age of 10 weeks, 14 days after initiation of hormone treatment. Each group contained 3 rats.  Effects of hGH infusion and hypophysectomy on hepatic microsomal levels of P-45OIe, in the rat. hGH was infused at a rate of 5 pg/h for 7 days to intact or hypophysectomized ( H z ) male and female rats. Hx rats were operated on at 8 weeks, and all animals were 10 weeks old at initiation of the treatment. After 7 days' treatment, microsomes were prepared from the three different animals in each group, and P-45016, was quantified. a, significantly different from intact male value; b, significantly different from intact female value.

Effects of intermittent hGH administration on P-45OI6, levels in
hypophysectomized male rats hGH was injected subcutaneously at 12-h intervals (8 a.m. and 8 p.m.) for 7 days into Hx male rats. The rats were 4 weeks old a t the beginning of the experiment, and had been operated on 2 weeks previously. After 7 days' treatment, microsomes were prepared, and P-45OI6, levels were determined. The number of animals is given in Darentheses. abolished by hypophysectomy (Fig. 1). Continuous infusion of hGH to intact male rats caused a decrease in P-45016, levels to a value which was not significantly different from the female level (Fig. 1). A similar effect was observed in hypophysectomized rats of both sexes, although the magnitude of effect was slightly less in the hypophysectomized females. These effects of hypophysectomy and GH infusion are identical with those observed for 16a-hydroxylase activity (12,14) and suggest an essential role of GH in regulation of the sexually differentiated p-45ol6, isozyme. Thus, administration of GH in a manner designed to give a female-type pattern of blood GH levels caused a feminization of p-45ol6, levels in the liver. We then tested the effects of intermittent GH administration, ie. giving a male-type pattern of blood GH levels, on the P-45016, levels in the livers of Hx rats. Administration of hGH, 50 pg/day, to Hx male rats by 12-h, subcutaneous injections caused a 3.4-fold increase in their P-45016, levels (Table 11). This is the same magnitude as the difference between Hx and intact male rats (Fig. l), indicating that GH injection masculinized the P-45016a levels and reversed the effect of hypophysectomy.

Group of animals
Taken together, the results presented in Fig. 1 and Table I1 strongly support the hypothesis that the up-and down-regulation of P-45Ol6, by GH is dependent on the plasma pattern of GH levels. The slightly lower daily dose of GH used in the injection experiment was not the reason for the difference in effects, since the same low dose administered to Hx males by infusion causes a feminization (not shown). It should be noted that a given dose of GH causes the same rate of growth in rats, irrespective of whether it is administered by infusion or by 12 h injections (15).
hGH can bind to both PRL and GH receptors in rat liver (16). It was therefore essential to determine whether the observed effects of hGH on P-45016, levels were due to a somatogenic or lactogenic action. Fig. 2 shows the effects of infusion of rGH, which has no lactogenic activity in the rat, and oPRL, a specific lactogen in the rat, on P-45016, levels in intact male rats. In addition, we studied the effect of administration of an identical daily dose of rGH by subcutaneous injection every 12 h. Whereas oPRL infusion had no effect on P-45016, levels, infusion of rGH caused a 75% decrease, and hGH a 94% decrease. This favors the idea that the effects of GH are achieved by an action on somatogenic receptors. The lower potency of the rat hormone compared to the human preparation may be explained by a reduced stability of rGH in aqueous solution at physiological pH (17), effectively resulting in a lower dose of active hormone reaching the liver. In contrast to the effect of rGH infusion, 12-h injections of the hormone at the same daiiy dose had no effect on p-45ol6, levels (Fig. 2), again emphasizing the importance of continuous blood GH levels, i.e. a female-type pattern of pituitary GH secretion, for a feminization of p-45ol6, levels in the liver.
The sex difference in the pattern of plasma GH levels is observed first at puberty, and is developed fully in the sexually mature animal (12,13). The sex difference in steroid 16ahydroxylating activity also follows the same developmental pattern, with the male activities increasing at puberty while female values remain low (3,18). In addition, Maeda et al. (19) have demonstrated a similar ontogenesis of "P-450 male" apoprotein levels. Therefore, we investigated the development of P-45016, expression in liver microsomes of male and female rats from the ages of 6 days to 8 months (Fig. 3). P-45016, levels are very low in female rats of all ages. The levels are also low in males until the age of puberty. At 35 days, just after the onset of puberty, the levels are about 46% of the levels in the intact male. The male level of expression is fully developed at 54 days and is maintained until the animal is at least 8 months old. Thus, the development of P-45O16, expression in rat liver coincides with the development of microsomal steroid 16a-hydroxylase activity and of the male pattern of pituitary GH secretion.
Some variation is observed in the absolute levels of P-45Ol6, in microsomes of adult male rats, as measured by the Western blot immunoassay. This variation is puzzling, but as we have noted previously (4), the Western blot technique may not be suited to the measurement of absolute contents of proteins in microsomes. However, we feel that the assay is satisfactory for the measurement of relative amounts of P-45OI6, in microsomal samples assayed at the same time, based on our observations that the relative amounts of P-45016, among samples are constant if the same samples are measured on repeated occasions (results not shown).
In conclusion, our results support the hypothesis that the episodic plasma pattern of GH levels is the ultimate determining factor in the expression of the sexually differentiated P-45016isozyme and that the hormone acts via somatogenic receptors. In addition, the data shows that the P-45Ol6, level in the liver is regulated by steroid hormones and by GH, in precisely the same manner as has been shown for steroid 16ahydroxylase activity, providing further, strong evidence that p-45ol6, is the enzyme which is responsible for that activity in rat liver microsomes. The mechanism(s) by which the liver can recognize the male and female pattern of GH secretion, and respond with either expression or nonexpression of the p-45ol6, apoprotein is still unclear.

Hormonal
Regulation of P-45OI6,