Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Although the average daily concentration of plasma growth hormone in males and females are similar, circulating profiles of the hormone in rats, mice, humans, and other species have been shown to be sexually dimorphic (Shapiro et al., 1995). In the case of the rat, males secrete growth hormone in episodic bursts (~200-300 ng/ml of plasma) every 3 to 4 h. Between the peaks, growth hormone levels are undetectable. In females, the hormone pulses are more frequent and irregular and are of lower magnitude than those in males, whereas the interpulse concentrations of growth hormone are always measurable (Legraverend et al., 1992; Shapiro et al., 1995). These sexual differences in the circulating growth hormone profiles, and not growth hormone concentrations per se, are responsible for observed sexual dimorphisms ranging from body growth to the expression of hepatic enzymes (Jansson et al., 1985). In this regard, rat, as well as murine liver, each contain at least a dozen sex-dependent isoforms of P450¹ that are regulated by the sexually dimorphic profiles of circulating growth hormone (Legraverend et al., 1992; Waxman, 1992; Shapiro et al., 1995). Expression of some isoforms may be restricted to one sex (i.e., sex-specific), whereas others could be expressed in both males and females, albeit at consistently different levels (i.e., sex-predominant). In the rat, P450 responses to growth hormone regulation are almost as variable as the number of growth hormone-dependent isoforms. Within the sex-dependent circulating growth hormone profiles are numerous intrinsic "signals", both inductive and repressive. Some isoforms are induced or suppressed by discerning the length and/or concentration of growth hormone in the interpulse periods; others respond to pulse frequencies or amplitudes. Still others recognize the mean circulating concentrations of growth hormone; and some are regulated by a combination of these signals (Pampori and Shapiro, 1996, 1999; Agrawal and Shapiro, 2000, 2001).

In senescent mammals, including humans, there is a reduction in the daily output of pituitary growth hormone expressed by a decrease in the plasma pulse amplitudes (Ho and Hoffman, 1993; Xu and Sonntag, 1996). Also associated with aging is a decline in hepatic drug-metabolizing capacity characterized by a reduced expression of selective sex-dependent P450 isoforms (Imaoka et al., 1991; Dawling and Crome, 1996) that could be correlated to changes in the circulating growth hormone profiles. In contrast, little is known about drug metabolism and growth hormone secretory profiles during the period between young adulthood and old age (i.e., middle age). That is, are the profound changes in drug metabolism and growth hormone secretory profiles observed in senescence anticipated in middle age?

In the present study, we correlate selective changes in the plasma growth hormone profiles of middle-aged male and female rats with expression levels of the male-specific isoforms CYP2C11 and CYP2C13 and female-predominant CYP2C7 and examine aging-associated changes in the releasability and effectiveness of growth hormone-releasing factor (GRF).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals were housed in the University of Pennsylvania Laboratory Animal Resources facility (Philadelphia, PA), under the supervision of certified Laboratory Animal Medicine veterinarians, and were treated according to a research protocol approved by the University's Institutional Animal Care and Use Committee. Housing conditions and breeding and treatment protocols have been reported (Shapiro et al., 1989). Basically, newborn Sprague-Dawley rats [Crl:CD(SD)BR] were treated with monosodium glutamate (MSG) (4 mg/g b.wt.; Sigma Chemical Co., St. Louis, MO) on alternate days, starting within 24 h of birth for a total of five s.c. injections. Vehicle controls received an equivalent amount of 1.97 M NaCl diluent (12 µl/g b.wt.). Rats were weaned at 25 days of age and, thereafter, were maintained under conventional "clean" housing conditions in filter-top cages until hormone studies commenced at approximately 3 months and 1 year of age.

Repetitive blood samples (10 µl) were obtained at 15-min intervals from unrestrained, unstressed, and completely conscious rats outfitted with our mobile catheterization apparatus (MacLeod and Shapiro, 1988; Pampori et al., 1991). Eight-hour plasma growth hormone profiles were determined by using a radioimmunoassay with a sensitivity of 2 to 3 ng/ml. Procedural details and statistical validation of the assay have been reported elsewhere (Shapiro et al., 1989).

Seven to ten days after serial blood collections for the determination of circulating growth hormone profiles, 3- and 12-month-old vehicle-treated and 12-month-old MSG-treated rats were administered the dopamine--hydroxylase inhibitor sodium diethyldithiocarbamate trihydrate (DDC; Aldrich Chemical Co., Inc., Milwaukee, WI), which has been reported to block norepinephrine synthesis and GRF-dependent growth hormone secretion for up to 12 h (Negro-Vilar et al., 1979; Katakami et al., 1981; McCormick et al., 1985). The drug, dissolved in 0.9% NaCl at a dose of 500 mg/kg b.wt. (0.2 ml/100 g b.wt.), was injected via the intra-atrial catheter used for serial blood collections. Two hours after DDC treatment, each rat received the growth hormone secretagogue clonidine-HCl (0.2 mg/kg b.wt. dissolved in phosphate-buffered saline, 0.1 ml/100 g b.wt.; Sigma Chemical Co., St. Louis, MO). Two hours later, the 12-month-old rats were also treated with a higher dose of clonidine (2.0 mg/ml/kg b.wt.). Three hours after the last dose of clonidine, the rats were treated with human GRF (Sigma/RBI, Natick, MA). GRF was injected, as were all drugs, via the indwelling intra-atrial catheter, dissolved in phosphate-buffered saline at a dose of 3 µg/kg b.wt. (0.1 ml/100 g b.wt.). Serial blood samples were obtained for growth hormone determination at 15-min intervals, from 7:00 AM, when DDC was administered, until 4:00 PM, 2 h after GRF treatment. There was always a 5- to 6-min hiatus between the time of drug administration and blood collection. Behavioral effects of the neuroleptics were carefully noted throughout the 9-h period.

Three- and twelve-month-old vehicle-treated female and male rats were decapitated, and the livers quickly were removed and processed for total RNA, as previously reported (Pampori and Shapiro, 1996). The RNA was analyzed by Northern blotting using oligonucleotide probes specific for CYP2C11, CYP2C7, and CYP2C13 (Waxman, 1991). Evidence that RNA was equally loaded and transferred was obtained by the equivalent intensity of ethidium bromide staining of 18S- and 28S-rRNA bands (Schuetz et al., 1990). Furthermore, the rat 18S-rRNA oligonucleotide probe was used as a control to verify the consistency and integrity of RNA loading (Ramsden et al., 1993). Quantitation of the mRNA by laser densitometry of the X-ray films was kept within the linear range as established by slot-blot hybridizations and normalized to the 18S-rRNA signals in each lane and to two control samples repeatedly run on every blot.

Data were subjected to analysis of variance, and differences among pairs of means were determined using Student's t statistics and the Bonferroni procedure for multiple comparisons.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Expression Levels of Hepatic CYP2C11, CYP2C7, and CYP2C13 in Middle-Aged Rats. Between the ages of 3 and 12 months, male rats exhibited a modest 30% decline in hepatic CYP2C11 mRNA levels (100 ± 4 and 71 ± 15%, respectively). As expected, no female liver expressed the male-specific CYP2C11 isoform. During the same age period, female rats exhibited a 40% decline in female-predominant hepatic CYP2C7 mRNA (100 ± 6 and 61 ± 12%, respectively), whereas males, which expressed half the female level of CYP2C7, exhibited an ~50% reduction in middle age (54 ± 8 and 28 ± 3, respectively). CYP2C13 mRNA expression was limited to the 3-month-old males; no other group expressed detectable levels of the transcript. All values were normalized to the 3-month-old concentrations, assigned a value of 100%, and expressed as a mean ± S.D. (n  4). Statistical differences between values at 3 and 12 months were P < 0.01 for both sexes.

Sexually Dimorphic Circulating Growth Hormone Profiles. The dramatic sexual dimorphism in the ultradian profiles of circulating growth hormone observed in middle-aged rats was comparable, although not identical, to that seen in young rats (Fig. 1). In female rats of both ages, growth hormone was released in a continuous pattern. Frequent low-amplitude pulses (as compared with males) of the hormone resulted in peaks averaging around 90 to 100 ng/ml, followed by short-lived interpulse periods. However, there was an age-associated reduction in the peak frequency and interpulse hormone concentration in the middle-aged females, which resulted in significantly longer lasting interpulses and an ~35% decline in the mean growth hormone concentration (Table 1). Growth hormone secretion in the males was more episodic than in females. In both young and middle-aged males, the hormone was released in pulses about every 3.5 h, resulting in half the number of daily peaks as secreted in females but at twice their amplitudes (Fig. 1). The major difference in the masculine profile between the 3- and 12-month-old rats was an age-associated decline in the pulse amplitudes that resulted in an ~40% lower mean growth hormone concentration in the older males (Table 1). Moreover, although the interpeak hormone levels in the young rats were consistently undetectable, they were generally measurable, albeit at very low levels, in middle-aged males.²

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Circulating growth hormone profiles determined from individual, undisturbed, catheterized 3-month-old (young) and 1-year-old (middle-aged) female and male rats. Serial blood samples were collected at 15-min intervals for 8 h consecutively. Similar findings were obtained from at least three additional animals in each treatment group.

Response to Growth Hormone Secretagogues. The effectiveness of the dopamine--hydroxylase inhibitor DDC in blocking norepinephrine-dependent GRF secretion was apparent from the absence of spontaneous growth hormone pulses, clearly depicted during the first 2 h after DDC treatment (Fig. 2). Since low-dose clonidine was ineffective in stimulating growth hormone secretion in the middle-aged rats, the growth hormone-blocking effects of DDC were observable for an additional 2 h. Moreover, the absence of any spontaneous growth hormone secretory pulses in the young rats of both sexes from 12:30 PM to 4:00 PM (9 h after DDC administration) effectively demonstrates the long-lasting potency of the inhibitor. There appeared to be a small, but significant (P < 0.01), sexual difference in the effectiveness of DDC in that plasma growth hormone concentrations of exposed males were below assay sensitivity, whereas hormone levels averaged 4 to 6 ng/ml in comparably treated females (Table 2).

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Plasma growth hormone responses to growth hormone secretagogues in young (~3 months old) vehicle-treated and middle-aged (~12 months old) vehicle-treated and MSG-treated (~12 months old) female and male rats. Seven to ten days after serial blood collection for determination of plasma growth hormone profiles (see Fig. 1) rats were intravenously injected with DDC (500 mg/kg b.wt.). Two hours later, rats received low-dose (0.2 mg/kg b.wt.) clonidine. After an additional 2 h, only the middle-aged vehicle- and MSG-treated rats received a higher dose (2.0 mg/kg b.wt.) of clonidine. Three hours after the last clonidine treatment, the rats were intravenously injected with GRF (3 µg/kg b.wt.). Serial blood samples were obtained at 15-min intervals from 7:00 AM, when the DDC was administered, until 4:00 PM. Growth hormone values from 3:00 PM to 4:00 PM were undetectable in all animals and have been omitted from the figures.

Drug Induced Behavioral Effects. In spite of the notable languor exhibited by neonatally MSG-treated rats (Shapiro et al., 1989), intravenous administration of DDC induced a similar and almost immediate sedation in all animals, ranging from deep sleep and unresponsive to physical stimuli to an interrupted sleep pattern with awkward ambulation during consciousness. Moreover, all the rats exhibited blepharospasms. In general, subsequent treatment with low-dose clonidine (0.2 mg/kg b.wt.) eliminated the blepharospasms but increased the level of sedation and psychomotor inhibition in most animals. A 10-fold higher dose of clonidine administered to middle-aged rats 2 h after the initial clonidine treatment had no additional behavioral effects other than a slight abatement in the level of sedation and lethargy in all animals. Three hours after treatment with clonidine, intravenous administration of 3 µg of GRF/kg b.wt. produced an immediate frenzied hyperactivity in almost all the animals that lasted up to 2 h. Inexplicably, about one-third of the male and female MSG-treated rats exhibited a pronounced lethargy after the GRF treatment. There appeared to be no relationship, however, between the levels of growth hormone released by GRF and the behavioral effects.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Sex differences in growth hormone secretory profiles (see the Introduction) appear around puberty and persist in this general pattern throughout life. Although exhibiting sexually dimorphic profiles, plasma growth hormone concentrations during the peripubertal period in female and male rats (Agrawal and Shapiro, 1995, 1996) and in girls and boys (Rose et al., 1991; Ho and Hoffman, 1993) are about 2 to 3 times greater than that observed in young adults. The increased hormone levels during puberty are due to an increase in the amount of growth hormone secreted per burst (i.e., peak) without any change in the pulse frequency or interpulse hormone concentration. During senescence, both rats (Xu and Sonntag, 1996) and humans (Ho and Hoffman, 1993) experience a dramatic decline in growth hormone secretion resulting from a decrease in pulse amplitudes. Much less is known, however, about the secretory patterns of the hormone in middle age. Judging from the few human studies in the literature, mean circulating growth hormone concentrations decline approximately 30% from about 20 to 50 years of age (Ho and Hoffman, 1993). In agreement with the human studies, our findings using 1-year-old male and female rats indicate a similar percent decrease in the mean plasma growth hormone concentrations in middle-aged rats of both sexes in what remain characteristically female and male profiles. In the case of males, rats and humans [the nearly exclusive gender examined in most reports (Ho and Hoffman, 1993)], we found that the decline in mean plasma growth hormone is due solely to a decrease in the amount of hormone secreted per pulse without any change in pulse duration or frequency. In contrast, reduced mean plasma growth hormone concentrations in year-old female rats were not due to a decline in the amount of hormone secreted in each pulse. Rather, growth hormone profiles in middle-aged female rats were characterized by a decrease in pulse frequencies, causing a commensurate decrease in the amount of hormone secreted daily.

Although not noted in human studies, we observed measurable growth hormone concentrations during the interpulse interval (i.e., nadir) in middle-aged male rats and a reduced concentration and lengthening (~2-fold) of the interpulse period in similarly aged females compared with young adults. These changes are rather subtle and could understandably go unnoticed. For example, the masculine growth hormone profile in young adult rats is characterized by undetectable (0 to <3 ng/ml of plasma) interpulses, whereas we observed just detectable (~3 ng/ml) interpulse hormone concentrations in middle-aged males. Since the expression of hepatic P450 isoforms are normally regulated by subtle changes in the sexually dimorphic growth hormone profiles (Pampori and Shapiro, 1996, 1999; Agrawal and Shapiro, 2000, 2001), these slight changes in middle-aged rats could portend the senescence-associated decline in P450 expression.

Masculine expression levels of male-specific CYP2C11, representing >50% of the total P450 in male rat liver (Morgan et al., 1985), is solely dependent upon growth hormone-devoid interpulses lasting at least 2 to 2.5 h (Agrawal and Shapiro, 2001). An interpulse hormone concentration of just 1 ng/ml of plasma can significantly suppress CYP2C11 mRNA, protein, and catalytic activity levels (Pampori and Shapiro, 1996, 1999). In agreement, we see a correlation between the presence of extremely low, but measurable, interpulse growth hormone concentrations in middle-aged male rats with a 30% reduction in CYP2C11 expression.³

Another male-specific isoform, CYP2C13, is optimally expressed when exposed to the masculine episodic growth hormone profile or under conditions of no hormone (i.e., hypophysectomy), whereas the feminine growth hormone profile completely suppresses CYP2C13 (Legraverend et al., 1992; Shapiro et al., 1995). In fact, a feminine profile of continuous growth hormone secretion at <3% of normal (i.e., <0.6 ng/ml of plasma) effectively suppresses all CYP2C13 expression (Pampori and Shapiro, 1996). Thus, as a result of the measurable hormone concentrations secreted during the interpulse periods, the CYP2C13 discriminator "sees" the growth hormone profile in the middle-aged males as a continuous (i.e., feminized) profile that fully suppresses expression of the isoform. The feminine profile in both the young and middle-aged females is sufficient to inhibit CYP2C13 expression in these rats.

In contrast to male-specific CYP2C11 and CYP2C13, female-predominant CYP2C7 expression is regulated by very different intrinsic signals in the sexually dimorphic growth hormone profiles. CYP2C7 expression is dependent upon the feminine growth hormone profile and is completely suppressed in the hypophysectomized rat. However, exposure to the masculine profile allows for expression of the isoform at 25 to 40% normal female levels (Legraverend et al., 1992; Shapiro et al., 1995). The CYP2C7 discriminator in both female and male liver appears to be rather scrupulous, requiring exposure to the precise physiologic gender-dependent growth hormone profiles for normal expression. Nominal changes in the profiles (e.g., pulse or interpulse concentrations or frequencies) produce commensurate reductions in CYP2C7 expression levels (Pampori and Shapiro, 1996, 1999; Agrawal and Shapiro, 2000, 2001) explaining the age-associated decline in CYP2C7 mRNA in both middle-aged females and males.

Subtle changes in the sex-dependent growth hormone secretory profiles of middle-aged rats (in turn responsible for aging-associated declines in P450 expression) may result from changes in the sensitivity of the aging hypothalamic-pituitary axis to growth hormone regulation. Accordingly, we examined the responsiveness of female, male, and arcuate nucleus-lesioned (i.e., MSG-treated) middle-aged rats to hypothalamic acting (clonidine) and pituitary-acting (GRF) growth hormone secretagogues. As noted above, the vast number of articles investigating the activities of growth hormone secretagogues have been confined to young male rats.⁴ However, in the few studies in which the sexually dimorphic effects of clonidine and GRF were compared, it appears, as we have found, that in young rats, both secretagogues stimulate greater levels of growth hormone release in male rather than female rats (Conway et al., 1989; Maiter et al., 1991), with clonidine producing a far more dramatic dimorphism. The greater effectiveness of clonidine in males may be explained by higher ₂-noradrenergic receptor binding (Johnson et al., 1988) and GRF levels (Corder et al., 1990; Hasegawa et al., 1992) in male hypothalami, both mediators of clonidine action (Eriksson et al., 1982; Conway et al., 1990). Higher baseline pituitary growth hormone reserves in males (Chihara et al., 1979; Critchlow et al., 1986) may explain the greater responsiveness of males to GRF. Whether or not these explanations for sex differences in clonidine and GRF actions are applicable to aging animals, the dimorphism persists in middle-aged rats, with males secreting about twice as much growth hormone as females when exposed to the secretagogues.

However, just as we have found small, but P450-sensitive, changes in the sex-dependent circulating growth hormone profiles of middle-aged rats, we have also observed subtle alterations in growth hormone response to clonidine and GRF stimulation in these older animals. We have found in agreement with others that young male (Durand et al., 1977; Katakami et al., 1981; Bluet-Pajot et al., 1982) and female rats (Negro-Vilar et al., 1979) of ~12 weeks of age respond immediately to 0.2 mg/kg b.wt. clonidine by secreting physiologic-like growth hormone pulses. In contrast, the middle-aged rats were unresponsive to this dose. A 10-fold increase in the dose of the secretagogue to 2.0 mg/kg b.wt. did produce normal-like, gender-dependent growth hormone pulses in the middle-aged rats, but unlike young rats, the response was highly delayed, requiring ~40 min in females and ~80 min in males to observe growth hormone release. In contrast, response to GRF was fairly similar in young and middle-aged rats. In both age groups, the same dose of GRF induced an almost immediate release of pituitary growth hormone, although the magnitude of response may have been somewhat greater in the younger animals (Miki et al., 1984; Pinilla et al., 1990). Thus, changes in the sexually dimorphic plasma growth hormone profiles observed in middle age may be explained by a reduced ability of the hypothalamus to secrete GRF and, to a lesser extent, by a possible reduction in pituitary responsiveness to GRF. In this regard, although hypothalamic secretory levels of GRF are dramatically reduced in senescent male rats (Sonntag et al., 1981; Meites, 1988), pituitary responsiveness to the hypothalamic factor is basically unaltered in old rats (Wehrenberg and Ling, 1983; Sonntag and Gough, 1988) or aged men (Pavlov et al., 1986; Lang et al., 1987).

Although it may seem reasonable to assume that the subnormal response of pituitary growth hormone to clonidine in the middle-aged rats is a result of insufficient GRF secretion, we tested this hypothesis using MSG-treated rats. Neonatal exposure to MSG produces permanent lesions in the arcuate nucleus, characterized by very low secretory concentrations of GRF, which is in turn responsible for the barely detectable to undetectable plasma growth hormone levels in both sexes (Shapiro et al., 1989, 1995). We found that growth hormone secretion in middle-aged MSG-treated female rats was completely unresponsive to clonidine administration, whereas similarly treated males responded (~120 min after the clonidine injection) with a small (25% of control) secretory pulse of growth hormone. These findings indicate that clonidine induces growth hormone secretion by stimulating hypothalamic release of GRF and that the reduced response of our control middle-aged male and female rats to the secretagogues can be explained as an aging-induced decline in GRF secretion.

In contrast to the clonidine findings, the subnormal response of the MSG-treated rats to GRF does not necessarily indicate a reduced sensitivity of pituitary somatotropes to GRF regulation in middle age. Rather, the GRF deficiency in male and female MSG-treated rats results in a pituitary growth hormone content of only 30 to 40% of normal (Shapiro et al., 1986), limiting the maximum response to exogenous GRF (as we observed) to an expected 30 to 40% of normal. Interestingly, although neonatal exposure to MSG produced the same inhibition in growth hormone and GRF secretion in both males and females, sexually dimorphic responses to clonidine and GRF administration persisted.

In conclusion, middle-aged rats secrete growth hormone in sexually dimorphic profiles that are similar but not identical to that of young rats. Some of the aging-associated subtle changes include elevated interpulse growth hormone concentrations and reduced pulse amplitudes in the male profile and a decrease in pulse frequencies and interpulse hormone levels in the female profile. These changes in the secretory growth hormone profiles may be explained, in part, by an aging-induced decrease in GRF release. Since the dozen or so sex-dependent isoforms of P450 in the rat are each regulated by different "signaling elements" in the sexually dimorphic growth hormone profile, the subtle changes in the profiles of middle-aged rats could explain the accompanying decline in CYP2C11, CYP2C7, and CYP2C13 expression. Indeed, changes in the plasma growth hormone profiles and P450 expression levels observed during middle age may be the forerunners of similar, but far more profound, changes occurring during senescence.