Evidence from dwarf rats that growth hormone may not regulate the sexual differentiation of liver cytochrome P450 enzymes and

Transcription

1 Pro. Nail. Aad. Si. USA Vol. 88, pp , June 1991 Biohemistry videne from dwarf rats that growth hormone may not regulate the sexual differentiation of liver ytohrome P45 enzymes and steroid 5a-redutase PTR BULLOK*, BRIAN GMZIK*, DONALD JOHNSONt, PAUL THOMASt, AND ANDRW PARKINSON* *Department of Pharmaology, Toxiology and Therapeutis, enter for nvironmental and Oupational Health, and tdepartment of Gyneology and Obstetris, University of Kansas Medial enter, Kansas ity, KS 6613; and tdepartment of hemial Biology and Pharmaognosy, Rutgers University ollege of Pharmay, Pisataway, NJ 8855 ommuniated by Ronald W. stabrook, Marh 18, 1991 (reeived for review Deember 29, 199) ABSTRAT Differenes in the pattern of growth hormone (GH) seretion in mature rats (i.e., "ontinuous" seretion in females versus "pulsatile" seretion in males) are thought to be the underlying ause of sex-dependent differenes in a subpopulation of liver mirosomal P45 enzymes and steroid 5aredutase. A new strain of dwarf rats (NIMR/AS) has reently been shown to have low or undetetable levels of irulating GH due to a seletive defet in pituitary GH synthesis. We have measured the levels and/or ativity of ila1 (P45a), 11A2 (P45m), li11 (P45h), I112 (P45i), IIIA2 (a P45p isozyme), and steroid 5a-redutase in liver mirosomes from male and female dwarf rats, to test the hypothesis that the expression of these sexually dimorphi enzymes is regulated by GH. In mature rats, the levels of liver mirosomal IIA2, II1l, and IIIA2 were higher in male than in female dwarf rats, whereas the levels of ativity of IIA1, II12, and steroid 5a-redutase were greater in female than in male dwarf rats. These sex differenes resulted from age-related hanges in either male dwarf rats (i.e., an inrease in II11 and 11A2 and a derease in IIA1) or female dwarf rats (i.e., an inrease in II12 and 5a-redutase and a derease in IIIA2). The magnitudes of these sex-dependent, age-related hanges were essentially indistinguishable from those observed in normal rats. These unexpeted results suggest that GH is not the pituitary fator responsible for regulating the levels of sexually dimorphi, steroid-metabolizing enzymes in rat liver. Alternatively, it is possible that these enzymes are regulated by extremely low levels of GH. In either ase, the urrent model of how steroidmetabolizing enzymes are regulated in rats must be revised to aount for the normal sexual differentiation of these enzymes in dwarf rats. Liver mirosomes from male and female rats ontain different amounts of steroid-metabolizing enzymes, suh as A4-3- ketosteroid Sa-redutase (steroid 5a-redutase) and several forms of ytohrome P45, inluding IIA1, IIA2, IIll, II12, and IIIA2 (1-5). These sex differenes are evident in mature but not immature rats and are largely the result of a postpubertal expression or suppression of the genes enoding these enzymes. For example, the rate of steroid 5a-redution and the 15,8-hydroxylation of 5a-androstane-3a,17,8-diol 3,17-disulfate inreases after puberty in female but not male rats (6-9). The latter ativity is atalyzed by II12, whih is onsidered to be a female-speifi P45 enzyme. In ontrast, the levels of IIA2 and II11 inrease postpubertally in male but not female rats, and both ofthese enzymes are onsidered male-speifi enzymes (7-14). The postpubertal expression of II11 is assoiated with a male-speifi, age-dependent inrease in the oxidation of testosterone to 2a- and 16ahydroxytestosterone and androstenedione. (3, 4). Sex differ- The publiation osts of this artile were defrayed in part by page harge payment. This artile must therefore be hereby marked "advertisement" in aordane with 18 U.S solely to indiate this fat. enes in the adult levels of IIA1 (female > male) and IIIA2 (male > > female) result from a developmental suppression of these enzymes. After weaning and throughout puberty, the levels of IIA1 deline in male rats to a greater extent than in female rats, and this deline is assoiated with a deline in testosterone 7a-hydroxylase ativity (7, 12-17). onversely, the levels of IIIA2 deline markedly in female but not male rats, and this deline is assoiated with a dramati postpubertal derease in the rate of testosterone 2,8-, 6,3-, and 15/3-hydroxylation (7, 16-18). Differenes in the pattern of somatotropin (growth hormone, GH) seretion in mature rats (i.e., "ontinuous" seretion in females versus "pulsatile" seretion in males) are thought to be the underlying ause of sex differenes in liver mirosomal ytohrome P45 and steroid 5a-redutase (1-3). Sex differenes in liver steroid metabolism are abolished by hypophysetomy. onstant infusion of human GH to hypophysetomized rats, whih mimis the relatively ontinuous seretion of GH in female rats (19), inreases the levels of II12 and 5a-redutase, whih auses a feminization of liver steroid metabolism (6, 2, 21). In ontrast, periodi infusion of GH to hypophysetomized rats, whih mimis the pulsatile pattern of GH seretion in male rats (19), inreases the levels of II11 and auses a masulinization of liver steroid metabolism (1, 21, 22). Synthesis of II12 an also be indued in ultured primary hepatoytes from male rats by exposure to human or bovine GH (23). harlton et al. (24) have established a olony of GHdefiient (dwarf) rats. These dwarf rats (derived from an NIMR/AS strain) are homozygous for an autosomal reessive mutation that redues pituitary GH levels to -5% of normal, whih redues irulating GH to barely detetable levels (<5 ng/ml) (24-27). Homozygous dwarf rats have normal or slightly elevated levels of other pituitary hormones, suh as prolatin, thyrotropin, follitropin, lutropin, and ortiotropin, indiating that dwarfism results from a stable, heritable, and seletive defet in GH synthesis and storage (24, 27). We have examined the levels and/or ativity of a subpopulation of P45 enzymes and steroid 5a-redutase in liver mirosomes from homozygous dwarf rats with virtually undetetable irulating GH. Unexpetedly, our results demonstrate that steroid 5a-redutase and IIA1, IIA2, II11, II12, and IIIA2 are expressed in a sex- and agedependent manner in dwarf rats and that the magnitude of the sex differenes observed in mature dwarf rats is omparable to that observed in normal rats. Abbreviation: GH, growth hormone. To whom reprint requests should be addressed. %In previous publiations, we have used the nomenlature system of Ryan and Levin (4), in whih P45a, -b, -e, -h, -i, -m, and -p orrespond to lial, IIB1, 1IB2, li11, II12, IIA2, and IIIAL. P45 I1IA2 is a onstitutively expressed, developmentally regulated isozyme of IIIAM (4, 5). 5227

2 5228 Biohemistry: Bullok et al. Pro. Natl. Aad. Si. USA 88 (1991) MATRIALS AND MTHODS Maintenane and Breeding of Dwarf Rats. The dwarf rats used in this study have been desribed in detail (24). Although the GH defiieny was first identified in Lewis rats, it was suessfully introdued into NIMR/AS rats at Mill Hill and Oxford, ngland. Four females and two males from this latter strain of dwarf rats were obtained through Harlan-Sprague- Dawley and provided the base for our olony. The rats were allowed free aess to Purina Rodent how (no. 58) and water and were maintained at 23 ± 1 with a 14-hr light/ 1-hr dark yle. Female rats displayed normal 4-day estrous yles. Pups were born on day 23 of pregnany, and the average litter size was 7.5 ±.6 (n = 12). Rats were weaned when 25 days old. Protein Purifiation and Antibody Prodution. P45 IIA1, IIA2, IIBi, IIB2, II11, and IIIA1 and NADPH-ytohrome P45 redutase were purified as desribed (4, 12, 28). Polylonal antibodies against these enzymes were raised in rabbits and were subjeted to immunoabsorption hromatography to remove antibodies that rossreated with P45 enzymes in other gene families (29). The immunoabsorbed antibodies were not monospeifi but reognized two or more distint enzymes belonging to the same P45 gene subfamily (29). A mouse monolonal antibody (H7) speifi for II11 was prepared as desribed (3). Immunostaining of Western blots was quantified by sanning densitometry (Shimadzu S-9). Steroid Oxidation and 5a-Redution. The pathways of testosterone oxidation atalyzed by liver mirosomes were determined by HPL (12, 16). onditions to measure the 5a-redution of testosterone were similar to those for measuring testosterone oxidation exept that liver mirosomes were inubated at 23 for 2 min with antibody against NADPH-ytohrome P45 redutase (5 mg of IgG per mg of mirosomal protein) to inhibit ytohrome P45. The 15,8-hydroxylation of Sa-androstane-3a,17,B-diol 3,17- disulfate was determined as desribed (9), exept that unreated substrate and produt were resolved by HPL. Substrate was kindly provided by Wayne Levin (Hoffmann-La Rohe). RSULTS The mature dwarf rats used in this study weighed approximately half as muh as normal rats. harlton et al. (24) have shown that the irregular but frequent pulses of GH harateristi of normal female rats are abolished in female dwarf rats, and likewise the prominent 3-hourly pulses of GH typial of normal male rats are abolished in male dwarf rats. We onfirmed that serum levels of GH were onsistently low (2.9 ±.5 ng/ml) in dwarf rats, as were the pituitary levels of GH. Pituitary levels of GH in female ( ,ug/mg) and male (5.6-15,ug/mg) dwarf rats were the same as those reported by harlton et al. (24). Liver mirosomes were prepared from immature (3-weekold) and mature (3- to 4-month-old) female and male dwarf rats (n = 4 or 5 per group). ytohrome P45 levels were similar to those previously reported for normal rats, whereas ytohrome b5 levels were slightly higher than normal (Fig. 1; refs. 12, 16, 31, and 32). In ontrast, NADPH-ytohrome redutase ativity was 1-5% less than that reported for normal rats (32). Liver mirosomes from mature male dwarf rats atalyzed the fastest rate of ytohrome redution, and this refleted higher levels of NADPH-ytohrome P45 redutase (Fig. 2). Low levels of NADPH-ytohrome P45 redutase in liver mirosomes from normal pre-weanling rats have been reported (17). Thyroid Hormone Status of Dwarf Rats. The thyroid hormone status of dwarf rats was assessed beause liver mi- n I O 1.4 L. ) th >.6,,._ 1 _w = - _~ -, o O ;; - o *a a m x 3 I.r ' 2 _o m r 'a 8) - 1 8) (A 5 8) D m. X 1... I Z ~L O w~ &U 12 U* ytohrome P-45 1 J ytohrome b5 li NADPH-yoohrome Redutase!F~ Immature Immature Mature Mature lo i8 F6 24 k 2 FIG. 1. ytohrome P45 and b5 levels, NADPH-ytohrome redutase ativity, and testosterone oxidation by liver mirosomes from NIMR/AS dwarf rats. A, androstenedione formation. Values are mean ± S. - 3 n ċ. ( o._ 25 o o - 2 Uv o.2 )L. - t5 8- O x 'a. (L Ur.. S N. a AF O. Vw

3 Biohemistry: Bullok et al. rosomes from dwarf rats have slightly elevated levels of ytohrome b5 and diminished levels of NADPH-ytohrome P45 redutase, whih is a harateristi of hypothyroid rats (31). Serum thyroxine and triiodothyronine in dwarf rats were within normal ranges (Table 1). Previous studies have shown that dwarf rats have normal levels of serum or pituitary thyrotropin, follitropin, lutropin, ortiotropin, prolatin, and testosterone (24, 27). Testosterone Oxidation. In Fig. 1, the metabolites of testosterone generated by liver mirosomes are arranged aording to the atalyti ativity of IIA1/2 (6a-, 7a-, and 15ahydroxylation), II11/IIB1 (2a-, 16a-, and 16f3-hydroxylation, and 17-oxidation to androstenedione), and IIIA1/2 (213-, 61-, and 15f3-hydroxylation). However, the only reliable indiators of the atalyti ativity of IIA1, II11, and IIB1 are testosterone 7a-, 2a-, and 16p3-hydroxylation, respetively. All pathways of testosterone oxidation were stimulated 2-14% when liver mirosomes (.5 mg/ml) were inubated with purified NADPH-ytohrome P45 redutase (.5 nmol/ml) (results not shown). Stimulation was observed regardless of the soure of liver mirosomes; hene, the qualitative aspets of the results shown in Fig. 1 were not influened by the slightly lower-than-normal levels of NADPH-ytohrome P45 redutase in liver mirosomes from dwarf rats. Regulation of P45 II11 (P45h). In normal rats, II11 is present only in mature male rats, and its expression is thought to be dependent on a pulsatile pattern of GH seretion (1-3, 1, 2-22). We predited that II11 would not be expressed Antibody IMMUNOBLOT P45s Against Reognized NADPH ytohrome P45 Redutase IIAl (P45a) l (P45n) IIA2 (P45m) IIA1 (P45a) II 1 (P45h) _ I11 (P45h) HiAl (P45p) I_ li I_ 51 kda lila 5 kda lila Std Std I I I Standard Immature Immature Mature Mature Standard P-45 Levels Immature Mature (pmol/mg protein) IIAl IIA I IIIA11/ Total P FIG. 2. Immunoblots of liver mirosomes (1,ug of protein) from NIMR/AS dwarf rats probed with antibodies against NADPHytohrome P45 redutase or various P45 enzymes. Lanes marked "Std" ontained purified NADPH-ytohrome P45 redutase (.1 pmol), IIAl (.25 pmol) and IIA2 (.25 pmol), II11 (2 pmol), or IIIAM (1 pmol). Pro. Natl. Aad. Si. USA 88 (1991) 5229 Table 1. Serum levels of thyroid hormones in NIMR/AS dwarf rats and normal Holtzman rats Triiodothyronine, Rat strain and Thyroxine, /Lg/dl ng/dl age status Dwarf immature 3.3 ± ±.7 92 ± 6 14 ± 12 Dwarf mature 3.4 ± ±.6 97 ± ± 1 Normal mature* 3.3 ± ± ± 18 Values are mean ± SD (n = 3-7). *For mature male Sprague-Dawley rats, normal levels of serum thyroxine and triiodothyronine are 3-7 tug/dl and 25-1 ng/dl, respetively (33). in the GH-defiient dwarf rats. However, II11 was speifially expressed in adult male dwarf rats, just as it is in normal rats (Figs. 1 and 2). Testosterone 2a-hydroxylation, whih is atalyzed speifially by II11, and II11 as shown by reativity with monolonal antibody H7 were essentially undetetable in liver mirosomes from immature dwarf rats and from mature female dwarf rats. The levels and ativity of II11 in liver mirosomes from mature male dwarf rats were omparable to or slightly greater than values reported for normal mature male rats (7, 16, 17). Regulation of P45 IIA1 (P45a). GH appears to exert opposing effets on IIA1 levels depending on its pattern of seretion. A pulsatile pattern is thought to suppress IIA1 levels, whereas a ontinuous pattern of GH seretion is thought to elevate IIA1 levels (34). onsequently, in normal rats, the levels of IIA1 are high in immature rats; they remain high in female rats but deline after weaning and throughout puberty in male rats (7, 12, 17, 34). We predited that IIA1 would remain at pre-weaning levels in both male and female dwarf rats. However, IIA1 was suppressed in mature male dwarf rats, but not female dwarf rats, just as in normal rats (Fig. 2). hanges in the levels of IIA1 were paralleled by hanges in the rate of formation of 6a- and 7a-hydroxytestosterone (Fig. 1). The levels and atalyti ativity of IIA1 were omparable to values for liver mirosomes from normal rats (13, 17). Regulation of P45 IIA2 (P45m). In normal rats, both IIA2 and II11 are male-speifi proteins, and their postpubertal expression is neonatally imprinted by testosterone (11). Despite these similarities, IIA2 and II11 are regulated quite differently. xpression of II11 is thought to be dependent on regular surges of GH seretion in male rats (1-3). In ontrast, the fator that stimulates IIA2 synthesis after puberty has not been identified (11). However, GH is thought to antagonize this fator by suppressing the synthesis of IIA2, beause the levels of IIA2 inrease in both male and female rats following hypophysetomy (11). In female rats, the ontinuous pattern of GH seretion ompletely suppresses IIA2 synthesis. In male rats, the suppressive effet of GH is presumably relaxed during the periodi troughs in irulating GH levels. Based on this model of IIA2 regulation, we antiipated exeptionally high levels of IIA2 in mature dwarf rats of both sexes, beause GH-defiient rats would be unable to suppress the postpubertal synthesis of IIA2. However, IIA2 was expressed only in mature male dwarf rats, just as it is in normal rats (Fig. 2). IIA2 was deteted on immunoblots probed with antibody against IIA1 beause IIA1 and IIA2 are 88% idential in amino aid sequene (14). The levels of immunoreative IIA2 in liver mirosomes from mature male dwarf rats were omparable to their normal ounterparts (12). Testosterone 15a-hydroxylase ativity was a poor index of IIA2 in liver mirosomes from dwarf rats (Fig. 1), just as it is in normal rats (13). Regulation of P45 I1A2. In normal rats, the levels of IIIA2 are high in immature rats; they remain high in mature male rats but deline abruptly and markedly after puberty in

4 523 Biohemistry: Bullok et al. female rats (7, 17, 18, 35). GH is thought to suppress the levels of IIIA2, and a ontinuous pattern of GH seretion exerts a greater suppressive effet than a pulsatile pattern of seretion (possibly beause the low trough levels provide periodi relief from GH suppression) (11, 18, 2, 36, 37). Based on this model of IIIA2 regulation, we antiipated high levels of IIIA2 in immature and mature dwarf rats of both sexes, beause these GH-defiient rats would be unable to suppress the expression of IIIA2. However, a 51-kDa protein reognized by anti-iiial was suppressed in mature female dwarf rats,just as it is in normal rats (Fig. 2). Based on reports by Gonzalez et al. (35) and Waxman et al. (37), this 51-kDa protein is most likely IIIA2. However, the antibody used reognizes at least two 51-kDa proteins (i.e., IIIA1 and IIIA2); hene, the identity of the 51-kDa IIIA protein in dwarf rats is not ertain. The putative IIIA2 protein was not suppressed in mature male rats, whih resulted in a marked sex differene (male > > female) in testosterone 2X3-, 6X3-, and 15f3-hydroxylase ativity (Fig. 1). Although the regulation of IIIA2 was qualitatively similar between normal and dwarf rats, the absolute levels and atalyti ativity of IIIA2 in liver mirosomes from dwarf rats were -5% of the orresponding values reported for normal rats (7, 17). The reason for this differene is unknown. Antibody against IIIA1 reognizes a 5-kDa IIIA1 protein (28), whih may be idential to P purified by Nagata et al. (38). This 5-kDa protein was not expressed in preweanling dwarf rats and was expressed only in mature male dwarf rats (Fig. 2), whih suggests that the 5-kDa IIIA protein is regulated in dwarf rats as it is in normal rats. Regulation of P45 II12 (P45i) and Steroid 5a-Redutase. P45 II12, whih atalyzes the 15p-hydroxylation of Saandrostane-3a,17f-diol 3,17-disulfate (9), and Sa-redutase are onsidered female-speifi enzymes (1-3). The high levels of these enzymes in liver mirosomes from mature female rats are thought to be dependent on a ontinuous pattern of GH seretion (1-3, 6, 2, 23). Aordingly, we predited that high levels of II12 and Sa-redutase would not be expressed in the GH-defiient dwarf rats. However, Fig. 3 indiates that high II12 ativity and high 5a-redutase ativity are expressed in adult female dwarf rats, just as they are in normal rats. In mature dwarf rats, the ativities of II12 and Saredutase were at least 1 times greater in liver mirosomes from female rats ompared with those from male rats. The ativities of II12 and Sa-redutase in liver mirosomes from _.3 25 :i.2-6 ~ x U ~. X LA.5 1 _Y Immature Immature Mature Mature j3 ') - x n W =- (o (a) - lw FIG. 3. Testosterone 5a-redution and 5a-androstane-3,17-diol 3,17-disulfate 15f3-hydroxylation by liver mirosomes from NIMR/ AS dwarf rats. Values are mean S. (Inset) An autoradiogram of a TL plate used to separate ['4]testosterone (T) from 5adihydrotestosterone (DHT) and androstenedione (A). D, dwarf rats; N, normal Sprague-Dawley rats; F, female; M, male. Pro. Natl. Aad. Si. USA 88 (1991) mature female dwarf rats were omparable to their normal ounterparts. Regulation of ITB1 (P45b). The major phenobarbitalinduible P45 enzyme, IIB1, is not normally present in liver mirosomes from untreated rats (4, 15). In mature rats, the synthesis of IIB1 is thought to be suppressed by GH, beause the levels of IIB1 inrease following hypophysetomy (39). Aordingly, we antiipated high onstitutive levels of IIB1 in liver mirosomes from GH-defiient dwarf rats. Liver mirosomes from dwarf rats had barely detetable IIB1 (results not shown) and were no more ative than their normal ounterparts at atalyzing the 16.3-hydroxylation of testosterone, whih is a reliable indiator of IIB1 (Fig. 1). DISUSSION The results show that GH-defiient dwarf rats display the same sex-speifi expression of liver steroid-metabolizing enzymes as normal rats. These results were unexpeted beause sexspeifi differenes in the pattern of GH seretion are thought to be responsible for the sexual differentiation of liver steroidmetabolizing enzymes (1-3). There are three possible explanations for the normal sexual differentiation of steroidmetabolizing enzymes in dwarf rats: (i) ontrary to urrent dogma, GH is not the feminizing/masulinizing hormone sereted by the pituitary; (ii) GH is the feminizing/ masulinizing hormone sereted by the pituitary, but extremely low levels of irulating GH are suffiient to regulate the expression of liver steroid-metabolizing enzymes; (iii) our urrent theory of liver enzyme regulation by GH is orret, but a "bakup" system ensures the normal sexual differentiation of steroid-metabolizing enzymes in the virtual absene of GH. This third possibility implies that sexual differentiation of liver steroid-metabolizing enzymes is of paramount importane, a onept that laks substantive support. If GH is not the pituitary fator responsible for regulating steroid-metabolizing enzymes in rat liver, two questions arise. First, how an we explain the large body of ompelling evidene impliating GH as the bona fide feminizing/ masulinizing fator? One possibility is that previous investigators have usually examined the effets of human or bovine GH, not rat GH, on rat steroid-metabolizing enzymes. Although human and bovine GH exert somatogeni effets in rats, it is oneivable that these foreign hormones also mimi or modulate the effets of the pituitary fator that atually regulates liver steroid-metabolizing enzymes. This possibility is not without preedent, beause human GH mimis the effets of GH and prolatin in rats. Furthermore, steroidmetabolizing enzymes and prolatin reeptor levels an be ompletely feminized or masulinized with relatively small amounts of human or bovine GH, whereas larger amounts of rat GH are usually required to ahieve even partial feminization or masulinization (1, 2, 4). However, this differene might reflet the greater stability of human and bovine GH ompared with rat GH (41). The seond question is: If not GH, what pituitary fator is ultimately responsible for regulating steroid-metabolizing enzymes in rat liver? Mode et al. (41) isolated the feminizing/ masulinizing fator from rat pituitary and demonstrated its apparent identity with GH. However, the pituitary fator was not purified to homogeneity, and Mode et al. (41) reognized the possibility that a ontaminant in the GH preparation might atually be the feminizing/masulinizing fator. An explanation that would aount for many of these observations is that the feminizing/masulinizing fator is similar but not idential to GH. Rat pituitary appears to ontain several eletrophoretially distint forms of GH with different growth-promoting and immunohemial properties (42). Jeffrey et al. (43) reported that liver arboni anhydrase III is sexually differentiated in dwarf rats (male > > female), just as it is in normal rats. The levels of arboni anhydrase III

5 were masulinized by intermittent treatment of normal female rats with somatostatin, but not somatoliberin, both of whih ause a pulsatile pattern of GH release. Interestingly, intermittent treatment of female dwarf rats with somatostatin, but not somatoliberin, also masulinized liver arboni anhydrase III levels. Norstedt et al. (44) observed a partial feminization of liver steroid-metabolizing enzymes and prolatin reeptors after dereasing somatostatin in male rats. onversely, inreasing somatostatin in female rats aused a partial masulinization of prolatin reeptor levels (44). These results suggest that trough levels of GH, rather than peak levels, are important for masulinizing liver enzymes. However, another interpretation is that arboni anhydrase III, steroid-metabolizing enzymes, and/or prolatin reeptors are regulated by diret effets of somatostatin on the liver. If GH is the pituitary hormone responsible for regulating steroid-metabolizing enzymes in rat liver, the unexpeted results obtained with dwarf rats neessitate a revision of our urrent model. The dwarf rats used in this study are not absolutely defiient in GH, and although stunted they grow larger than hypophysetomized rats (24). It is possible, therefore, that steroid-metabolizing enzymes an be regulated by exeedingly low levels of GH, perhaps beause these effets are mediated by high-affinity GH reeptors in rat liver. The possibility that low levels of GH are suffiient to regulate liver steroid-metabolizing enzymes has been proposed previously. Waxman et al. (45) demonstrated that liver mirosomal II12 and steroid 5a-redutase were expressed normally in mature female rats treated neonatally with monosodium glutamate at 4 g/kg, whih abolished irulating GH levels (<2 ng/ml). Similar treatment of male rats abolished irulating GH and the postpubertal expression of liver mirosomal II11, IIA2, and IIIA2, whih supports a role for GH in the expression of these male-speifi enzymes. However, Shapiro et al. (46) demonstrated that treatment of neonatal male rats with monosodium glutamate at only 2 g/kg did not abolish the expression of II11, IIA2, and IIIA2 in adult male rats, and this treatment diminished but did not abolish the pulsatile pattern of GH seretion. Neonatal treatment of rats with monosodium glutamate destroys 8-9% of the neurons in the aruate nuleus of the hypothalamus, whih disrupts several endorine systems. In ontrast, the NIMR/AS dwarf rats appear to have a seletive defet in GH synthesis (24-27). Therefore, dwarf rats have the potential Biohemistry: Bullok et al. to provide a simple model to address the question of whether extremely low levels of GH an regulate liver steroid-metabolizing enzymes. The dwarf rats will also be invaluable in identifying the feminizing/masulinizing fator, in the event that GH is not the pituitary hormone responsible for regulating steroid-metabolizing enzymes in rat liver. This work was supported by Grants S3765 and GM3744 from the National Institutes of Health (NIH). B.G. was supported by NIH Training Grant S779, and A.P. is the reipient of NIH Researh areer Development Award S Zaphiropoulos, P. G., Mode, A., Norstedt, G. & Gustafsson, J.-A. (199) Trends Pharmaol. Si. 1, Skett, P. (1988) Pharmaol. Ther. 38, Waxman, D. J. (1988) Biohem. Pharmaol. 37, Ryan, D.. & Levin, W. (1989) Pharmaol. Ther. 45, Gonzalez, F. J. (1989) Pharmaol. Rev. 4, Zaphiropoulos, P. G., Mode, A., Strom, A., Moller,., Fernandez,. & Gustafsson, J.-A. (1988) Pro. Nati. Aad. Si. USA 85, Waxman, D. J., Dannan, G. A. & Guengerih, F. P. (1985) Biohemistry 24, Kamataki, T., Maeda, K., Yamazoe, Y., Nagai, T. & Kato, R. (1983) Arh. Biohem. Biophys. 225, Ryan, D.., Dixon, R., vans, R. H., Ramanathan, L., Thomas, Pro. Natl. Aad. Si. USA 88 (1991) 5231 P.., Wood, A. W. & Levin, W. (1984) Arh. Biohem. Biophys. 233, Morgan,. T., MaGeoh,. & Gustafsson, J.-A. (1985) J. Biol. hem. 26, Waxman, D. J., LeBlan, G. A., Morrissey, J. L., Staunton, J. & Lapenson, D. P. (1988) J. Biol. hem. 263, Arlotto, M. P., Greenway, D. J. & Parkinson, A. (1989) Arh. Biohem. Biophys. 27, Arlotto, M. P. & Parkinson, A. (1989) Arh. Biohem. Biophys. 27, Matsunaga, T., Nagata, K., Holsztynska,. J., Lapenson, D. P., Smith, A., Kato, R., Gelboin, H. V., Waxman, D. J. & Gonzalez, F. J. (1988) J. Biol. hem. 263, Thomas, P. T., Reik, L. M., Ryan, D.. & Levin, W. (1981) J. Biol. hem. 256, Sonderfan, A. J., Arlotto, M. P., Dutton, D. R., MMillen, S. K. & Parkinson, A. (1987) Arh. Biohem. Biophys. 255, berhart, D.. & Parkinson, A. (199) Toxiologist 1, 123 (abstr.). 18. Yamazoe, Y., Murayama, N., Shimada, M., Yamauhi, K., Nagata, K., Imaoka, S., Funae, Y. & Kato, R. (1988) J. Biohem. (Tokyo) 14, Jansson, J.-O., den, S. & Isaksson,. (1985) ndor. Rev. 6, Mode, A., Norstedt, G., Simi, B., neroth, P. & Gustafsson, J.-A. (1981) ndorinology 18, Mode, A., Wiersma-Larsson,., Strom, A., Zaphiropoulos, P. G. & Gustafsson, J.-A. (1989) J. ndorinol. 12, Yamazoe, Y., Shimada, M., Kamataki, T. & Kato, R. (1986) Jpn. J. Pharmaol. 42, Guzelian, P. S., Li, D., Shuetz,. G., Thomas, P., Levin, W., Mode, A. & Gustafsson, J.-A. (1988) Pro. Natl. Aad. Si. USA 85, harlton, H. M., lark, R. G., Robinson, I.. A. F., Porter Goff, A.., ox, B. S., Bugnon,. & Bloh, B. A. (1988) J. ndorinol. 119, Skottner, A., lark, R. G., Fryklund, L. & Robinson, I.. A. F. (1989) ndorinology 124, armigna, D. F. & Robinson, I.. A. F. (199) J. ndorinol. 127, Bartlett, J. M., harlton, H. M., Robinson, I.. A. F. & Nieshlag,. (199) J. ndorinol. 126, Halvorson, M., Greenway, D., berhart, D., Fitzgerald, K. & Parkinson, A. (199) Arh. Biohem. Biophys. 277, Gemzik, B. & Parkinson, A. (1991) Methods nzymol., in press. 3. Reik, L. M., Maines, S. L., Ryan, D.., Levin, W., Bandiera, S. & Thomas, P.. (1987) J. Immunol. Methods 1, Rozman, K., Hazelton, G., Klaassen,. D., Arlotto, M. & Parkinson, A. (1985) Toxiology 37, Gorski, J. R., Arlotto, M. P., Klaassen,. D. & Parkinson, A. (1985) arinogenesis 6, Depaolo, L. V. & Masoro,. J. (1989) in The linial hemistry of Laboratory Animals, eds. Loeb, W. F. & Quimby, F. W. (Pergamon, New York), pp Waxman, D. J., Morrissey, J. J. & LeBlan, G. A. (1989) ndorinology 124, Gonzalez, F. J., Song, B. J. & Hardwik, J. P. (1986) Mol. ell. Biol. 6, Waxman, D. J., Morrissey, J. J. & LeBlan, G. A. (1989) Mol. Pharmaol. 35, Waxman, D. J., Ram, P. A., Notani, G., LeBlan, G. A., Alberta, J. A., Morrissey, J. J. & Sundseth, S. S. (199) Mol. ndorinol. 4, Nagata, K., Gonzalez, F. J., Yamazoe, Y. & Kato, R. (199) J. Biohem. (Tokyo) 17, Yamazoe, Y., Shimada, M., Murayama, N. & Kato, R. (1987) J. Biol. hem. 262, Norstedt, G., Mode, A., neroth, P. & Gustafsson, J.-A. (1981) ndorinology 18, Mode, A., Norstedt, G., neroth, P. & Gustafsson, J.-A. (1983) ndorinology 113, Russell, S. M., Vodian, M. A., Hughes, J. P. & Nioll,. S. (1978) Life Si. 23, Jeffrey, S., arter, N. D., lark, R. G. & Robinson, I.. A. F. (199) Biohem. J. 266, Norstedt, G., Mode, A., Hokfelt, T., neroth, P., lde, R., Ferland, L., Labrie, F. & Gustafsson, J.-A. (1983) ndorinology 112, Waxman, D. J., Morrissey, J. J., MaLeod, J. N. & Shapiro, B. H. (199) ndorinology 126, Shapiro, B. H., MaLeod, J. N., Pampori, N. A., Mo Tssey,J. J., Lapenson, D. P. & Waxman, D. J. (1989) ndorinology 125,