Main

GH secretion and its neuroendocrine regulation undergo significant changes during ontogeny. An important and incompletely understood phenomenon is the hypersecretion of GH during the perinatal period in several species including the human and the rat(15). Hypersecretion of GH in neonatal rats is in part due to maternal factors, as maternal deprivation results in a significant decrease in plasma GH levels(6, 7). Milk-borne factors(5, 8) and mother-offspring interaction(9, 10) have been implicated as the mediators of maternally modulated GH secretion. Episodic secretion of GH is absent in suckling rats(11); in adult rats, the episodic pattern is sexually distinct(12). In male rats, episodic bursts of GH secretion occur more regularly but less frequently, and are of higher amplitude than in female rats. The trough (“basal”) levels of GH are lower in males than in females, which is attributed to hypothalamic secretion of somatostatin (SRIF)(13). The episodic bursts are mediated by GHRH and a concomitant decrease in SRIF(14). Pentobarbital anesthesia eliminates episodic bursts and increases basal levels of GH in adult male rats, presumably by disconnecting both stimulatory and inhibitory hypothalamic influences from pituitary GH secretion(15). In 5-d-old rat pups not separated from their mothers, pentobarbital suppresses GH secretion(16); pentobarbital has essentially the same effect in maternally deprived 2-d-old pups (our unpublished observation). The absence of an episodic secretory pattern and the GH-suppressing effect of pentobarbital suggest that in the neonatal rat GH secretion is under a tonic hypothalamic stimulatory control, in sharp contrast with the phasic hypothalamic influence seen in adults. This tonic hypothalamic stimulation and a decreased pituitary responsiveness to SRIF(17, 18) may contribute to neonatal hypersecretion of GH.

In the classical model of the hypothalamic regulation of GH secretion, the final integrative pathway consists of two neuropeptides: GHRH and SRIF. Several reports have recently indicated that other mechanisms that involve neither of these peptides have an important regulatory influence on GH secretion in neonatal rats. GH release evoked by N-methyl aspartic acid, neural stimuli associated with nursing (defined as the active maternal component of suckling), and GHRP-6 stimulate GH secretion at least in part by liberating GRFs distinct from GHRH(1921). Candidates for these alternative GRFs include γ-aminobutyric acid, TRH, and galanin. Pituitary responsiveness to these putative GRFs is high during the early postpartum period but declines with age in rats(2224).

GHRP-6, hexarelin, and related peptides are synthetic GH-releasing drugs unrelated to the structure of GHRH(25, 26). These drugs are effective after either oral or intranasal administration, making them attractive alternatives to GH, which is active only when injected. Long-term intranasal administration of hexarelin is an effective means to stimulate the GH-IGF-I axis and to promote statural growth in short children(27). An additional benefit of GHRPs is that they enhance episodic secretion of GH, thereby attaining a more physiologic profile of circulating GH than achieved by injections of GH(28). GHRP-6 has a dual mechanism of action: it acts directly upon the pituitary gland and, more important, acts indirectly by liberating a hitherto unspecified GRF distinct from GHRH(29), probably from the hypothalamus(30). In the present study, tests were made to determine whether GH secretion induced by GHRP-6 or nursing was mediated by TRH in neonatal rats.

METHODS

Animals

All animal studies were approved by the Institutional Animal Care and Use Committee. Adult female Sprague-Dawley rats were purchased from Harlan Laboratories, Inc. (Indianapolis, IN) and maintained on a 12-h light: 12-h dark schedule (lights on at 0600 h) in a temperature-controlled vivarium(20-22°C, relative humidity 40-60%). The animals were bred, housed individually from about d 19 of gestation, and checked for parturition twice a day. The day of birth was designated d 0 of postnatal life for the pups. A standard rodent diet and tap water were available to the mothers ad libitum. Two-day-old pups from litters containing 8-16 pups were used in the experiments. A total of 401 pups representing 36 litters were used; the male to female ratio was 208:193, i.e., 51.9% of the population was male. Male to female ratios are shown for each experiment in the legends to the figures.

Experimental Procedure

Peptides. GHRP-6 and TRH were purchased from Bachem (Torrance, CA) and Peninsula Laboratories (Belmont, CA), respectively.

In vitro experiments. Pituitary hormone responses to GHRP-6 and TRH were tested in vitro using methods described earlier(31) with modifications. Two-day-old rat pups were removed from their mothers and decapitated, and pituitary glands were collected in culture medium. The culture medium was composed of DMEM (the formulation containing 1 g/L glucose, 110 mg/L sodium pyruvate, and 584 mg/L L-glutamine; Life Technologies, Inc., Gaithersburg, MD) supplemented with 10 mM N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid, 100 U/mL penicillin, 100 μg/mL streptomycin (Life Technologies), 20μg/mL L-ascorbic acid, and 0.1% RIA-grade BSA (Sigma Chemical Co., St. Louis, MO). Male and female pituitaries were handled separately and placed individually in the wells of 24-well culture dishes (Corning, New York). To stabilize basal secretion, each gland was incubated in 1 mL of medium for 3 h at 37°C in a humidified 95% air/5% CO2 incubator. After the stabilization period, each well was rinsed twice with 1 mL of DMEM, and the glands were further incubated in 1 mL/well DMEM for 1 h to determine the basal secretion rate for each individual gland. The conditioned medium was collected, the wells were rinsed twice with 1 mL of DMEM, and the glands were exposed to the test substances (GHRP-6 and/or TRH) diluted in DMEM at the concentrations indicated in the figures. The glands were again incubated for 1 h, and the conditioned medium was collected from each well. Care was taken to assign approximately equal numbers of male and female pituitary glands to each treatment group. The stimulating agents (DMEM and all dilutions of GHRP-6 and/or TRH) and the samples were stored at -20°C until assayed for GH and PRL.

In vivo experiments. To test the effects of GHRP-6 and TRH in 2-d-old pups, the pups were separated from their mothers for 6 h. During separation, the pups were kept in a cage filled with wood chips and were covered with a paper towel to reduce heat loss. The pups were not fed during separation. The pups were numbered and weighed (precision: 0.1 g) at the time of separation. In every experiment, each litter was divided into treatment groups to assure the use of littermate controls. In addition, an even distribution of males and females among the treatment groups was attempted in each litter. The peptides were injected intraperitoneally in physiologic saline solution at the volume of 1 μL/g of body weight. Each pup was decapitated 15 min later, and trunk blood was collected. After clotting overnight at 4°C, serum was collected and stored at -20°C until assayed for GH and PRL. The interaction between GHRP-6 and TRH was evaluated by injecting the premixed peptides as a single injection in comparison with injections of saline, GHRP-6, or TRH alone. The interaction between either of the two peptides and nursing was evaluated by injecting separated (6 h) and nonseparated (nursed) littermate pups with saline or the peptide.

Hormone Assays

GH assay. Rat GH was detected by RIA using materials generously provided by Dr. A. F. Parlow and the National Hormone and Pituitary Program, NIDDK (Bethesda, MD) as described elsewhere(20). Rat GH RP-2 was used as a standard reference preparation. Concentrations of GH in pup sera and conditioned medium samples were determined in 5 μL/tube and 30μL/tube, respectively. All samples from each experiment were measured in the same RIA to avoid interassay variability. The assays were analyzed using the AssayZap program (Biosoft, Cambridge, UK) and a Power Macintosh 6100/66 personal computer (Apple Computer Inc., Cupertino, CA). The intraassay and interassay coefficients of variation were below 5 and 7%, respectively, in the entire range of the assay.

Prolactin assay. Biologically active PRL was analyzed by the Nb2 lymphoma assay. This assay is more sensitive than RIA; thus, it enabled the measurement of PRL in individual serum samples of neonatal rats even after measuring GH in the same samples. All samples from each experiment were measured in the same assay to avoid interassay variability. The culture and assay conditions were the same as described earlier(32) with the following modifications. After depriving Nb2 cells of FCS for 24 h, the cells were dispensed in the wells of flat-bottomed 96-well tissue culture plates (Corning) at a density of 5000 cells/well in 80 μL/well of FCS-free medium [Fisher's medium (Life Technologies), pH 7.6, supplemented with 10% defined horse serum (HyClone, Logan, UT), 100 U/mL penicillin, 100 μg/mL streptomycin (Life Technologies), and 0.1 mM 2-mercaptoethanol (Sigma Chemical Co.)]. Serially diluted standard PRL (B-6; NHPP, NIDDK) and a 1:4 dilution of each serum sample obtained from 2-d-old pups were prepared in FCS-free Nb2 assay medium. Conditioned DMEM samples were used undiluted.

Twenty microliters of the standard or sample were added to each well. Standards and samples were analyzed in quadruplicate. The cells were incubated for 48 h at 37°C in a humidified 95% air/5% CO2 incubator. PRL-induced cell proliferation was assessed by conversion of MTT (Sigma Chemical Co.) into formazan. This procedure is also known as eluted stain assay(3335). MTT was dissolved in 25 mM PBS (pH 7.6) at a concentration of 5.5 mg/mL, sterile-filtered, and stored refrigerated in the dark for up to 12 mo. At the end of the 48-h incubation, 10 μL of MTT solution was added to each well, and the plates were incubated for another 2 h in the CO2 incubator. Color development was stopped, and the formazan product was solubilized by adding 100 μL/well 0.1 M acetate buffer (pH 4.7) containing 20% (wt/vol) SDS and 50% (vol/vol)N,N-dimethylformamide(36). After the addition of the solubilizing agent, the plates were wrapped in plastic foil and incubated in the dark at room temperature for 2 d. Absorbances were determined with an automatic ELISA plate reader at a wavelength of 595 nm.

The data were analyzed with the AssayZap program. Based on 10 assays (each involving 6-12 plates), the detection limit of the assay was 50 pg/mL (in 20μL/well undiluted sample volume; i.e. 1 pg/well), the effective dose at 20% of the maximal stimulus (ED20) was 0.17 ng/mL, the ED50 was 1.245 ng/mL, and the ED80 was 9.77 ng/mL. The interassay CVs were 18.4% at ED20, 12.9% at ED50 and 18.1% at ED80. The average intraassay interplate CVs were 23.1% at ED20, 15.0% at ED50 and 34.5% at ED80. The intraassay intraplate CVs were calculated from the calibration points closest to the above effective doses i.e., 0.2 ng/mL (ED20), 1.60 ng/mL (ED50), and 12.8 ng/mL (ED80). The average intraassay intraplate CVs were, 16.1, 13.7, and 34.3%, respectively. Based on 78 assay plates, the accuracy values of the assay at these calibration points (expressed as recoveries of the nominal concentration) were 95.0% (CV = 7.4%), 107.9% (CV = 9.2%), and 102.2%(CV = 27.7%), respectively. These assay parameters represent an improvement over the traditional Nb2 assay in which cell proliferation was assessed by incorporation of [3H]thymidine(32).

Statistical Analysis

Statistical evaluation of the data were performed using the SuperANOVA and StatView programs (Abacus Concepts, Inc., Berkeley, CA) and a Power Macintosh 6100/66 personal computer. The data are shown as mean + SEM (n), where n is the number of individual pituitary glands or pups. The number of litters (N) used in each experiment is given separately. To establish a dose-response relationship, data were analyzed by one-way ANOVA followed by Student-Newman-Keuls and by Bonferroni-Dunn post hoc tests. When two variables were used in an experiment (e.g. nursing and a drug), the data were analyzed by two-way ANOVA. Differences were considered statistically significant whenever p values were≤0.05.

RESULTS

In vitro experiments. GHRP-6 and TRH stimulated GH secretion of pituitary glands obtained from 2-d-old pups. GHRP-6 evoked a maximal GH response at the dose of 10 nM; further increases in the dose of GHRP-6 up to a maximum of 10 μM yielded no additional increase in GH secretion. In contrast, the GH response to TRH was biphasic. The GH response to 0.1 to 100 nM TRH increased in a dose-dependent manner, reaching a maximum at 100 nM. Exposure to 1 μM TRH resulted in a smaller GH response compared with the effect of 100 nM TRH, although compared with the basal secretion of GH, 1μM TRH still induced a statistically significant increase. GHRP-6 stimulated secretion of biologically active PRL only at the highest dose tested (10 μM), whereas TRH was an effective stimulus at a concentrations of 1 or 10 nM. The PRL response declined at doses of TRH exceeding 10 nM (Figs. 1 and 2).

Figure 1
figure 1

Hormone secretion in vitro in response to a concentration range of GHRP-6 (1 nM to 10 μM). Culture media (DMEM) were conditioned two consecutive times for 1 h each with whole pituitary glands obtained from 2-d-old rats (number of litters, N = 4; number of pups, n = 41; male to female ratio, 19:22). The number at the bottom of each column represents the number of data points (n),i.e. individual pituitary glands in the experimental group. The relative responses were calculated by comparing hormone secretion during the second incubation (exposure to test substances) with hormone secretion during the first incubation (basal secretion rate). GH secretion (upper panel): The effect of GHRP-6 was highly significant (one factor ANOVA: p = 0.0001). The basal secretion rate of GH was 16.9 ± 1.1 ng/mL/h (mean± SEM, n = 41). PRL secretion (lower panel): The effect of GHRP-6 was significant (one factor ANOVA: p = 0.0036). The basal secretion rate of PRL was 0.78 ± 0.13 ng/mL/h (n = 41). For both panels, the asterisks indicate statistically significant increases(*p ≤ 0.05, **p ≤ 0.01) compared with the DMEM control as assessed with Bonferroni-Dunn and Student-Newman-Keuls tests.

Figure 2
figure 2

Hormone secretion in vitro in response to a concentration range of TRH (0.1 nM to 1 μM). Pituitary glands obtained from 2-d-old rats were tested in vitro (N = 3, n = 30, male to female ratio = 14:16). GH secretion (upper panel): The effect of TRH was highly significant (one factor ANOVA: p = 0.0001). The basal secretion rate of GH was 6.9 ± 0.6 ng/mL/h (n = 30). PRL secretion (lower panel): The effect of TRH was significant (one factor ANOVA:p = 0.0052). The basal secretion rate of PRL was 0.59 ± 0.11 ng/mL/h (n = 30). For both panels, the asterisks indicate statistically significant increases (*p ≤ 0.05, **p≤ 0.01) compared with the DMEM control as assessed with Bonferroni-Dunn and Student-Newman-Keuls tests.

When tested in a separate experiment at their maximally effective GH-releasing doses (100 nM for each peptide), TRH and GHRP-6 induced similar(compared with medium alone) approximately 5-fold increases in GH secretion (Fig. 3). Treatment with a combination of the maximally effective doses of GHRP-6 and TRH resulted in a GH response comparable to that evoked by either treatment alone. In this experiment, the PRL-releasing effect of 100 nM TRH reached statistical significance. This TRH-induced release of PRL was not altered by the presence of GHRP-6 (Fig. 3).

Figure 3
figure 3

Hormone secretion in vitro in response to treatment with a combination of TRH and GHRP-6 (100 nM each). Pituitary glands obtained from 2-d-old rats were tested in vitro (N = 3,n = 35, male to female ratio = 15:20). GH secretion (upper panel): The effects of TRH and GHRP-6 were highly significant (two factor ANOVA, factor 1 (TRH): p = 0.0028, §; factor 2 (GHRP-6): p = 0.0012, ^). The significant interaction between the two factors (p = 0.0339, *) indicated that the combined effect of the peptides was significantly lower than the sum of the effects of separate treatments. The basal secretion rate of GH was 18.2 ± 2.1 ng/mL/h (n = 35). PRL secretion (lower panel): The effect of TRH was highly significant (two factor ANOVA, factor 1 (TRH): p = 0.0002, †). GHRP-6 exerted no effect on PRL secretion [factor 2 (GHRP-6): p = 0.6398]. There was no significant interaction between the two factors (p = 0.6629,#). The basal secretion rate of PRL was 0.26 ± 0.03 ng/mL/h(n = 26). Note the difference between the upper and lower panels in the number of pituitaries included in the calculations. In this experiment the basal secretion rate of nine pituitary glands were below the detection limit of the PRL bioassay.

Neither GHRP-6 nor TRH displayed detectable cross-reactivity in RIA for rat GH. None of the peptides induced proliferation of Nb2 cells at any of the concentrations introduced with the samples into the PRL bioassay (not shown). No statistically significant sexual differences were found in basal or GHRP-6- or TRH-induced secretion of either GH or PRL.

In vivo experiments. GHRP-6 (doses of 18.75-600 ng/g) stimulated GH secretion in a dose-dependent manner in 2-d-old rats separated from their mothers for 6 h. GHRP-6 was maximally effective at a dose of 300 ng/g. PRL-like bioactivity in the same serum samples was not affected (Fig. 4). TRH (doses of 62.5 pg/g to 10 ng/g) increased the serum levels of both immunoreactive GH and bioactive PRL in a dose-dependent manner; statistically significant increases in serum GH and PRL were seen at of 1 and 10 ng/g (Fig. 5). In contrast to their effects in vitro, at their maximally effective doses (300 ng/g GHRP-6 and 10 ng/g TRH) the GH response induced by GHRP-6 (an approximately 5-fold increase) exceeded the GH response induced by TRH (an approximately 2-fold increase). In contrast to the in vitro experiments, treatment with a combination of the maximally effective doses of GHRP-6 and TRH resulted in significantly higher serum levels of GH than either treatment alone. The combination of the two peptides produced a synergistic effect, as the levels of GH were significantly greater than the sum of the levels produced by each peptide alone. GHRP-6 did not alter the TRH-induced release of PRL (Fig. 6), a result similar to that obtained in thein vitro experiments.

Figure 4
figure 4

Hormone secretion in vivo in response to a dose range of GHRP-6 (18.75 ng/g to 600 ng/g of body weight). Two-day-old rats were separated from their mothers for 6 h, injected intraperitoneally with the peptide dissolved/diluted in physiologic saline (volume dose of 1 μL/g), and decapitated 15 min later. Hormone concentrations were measured in serum. The number at the bottom of each column represents the number of data pointsi.e., individual pups. Five litters were used in the experiment(N = 5, n = 57, male to female ratio = 30:27). GH levels(upper panel): The effect of GHRP-6 was highly significant (one factor ANOVA:p = 0.0001). The asterisks indicate statistically significant increases (*p ≤ 0.05, **p ≤ 0.01) compared with the saline-injected control as assessed with Bonferroni-Dunn and Student-Newman-Keuls tests. PRL levels (lower panel): The effect of GHRP-6 was not significant (one factor ANOVA: p = 0.4623).

Figure 5
figure 5

Hormone secretion in vivo in response to a dose range of TRH (62.5 pg/g to 10 ng/g of body weight). Separated 2-d-old rats were injected intraperitoneally with the peptide and decapitated 15 min later (N = 5, n = 60, male to female ratio = 35:25). Hormone concentrations were measured in serum. GH levels (upper panel): The effect of TRH was highly significant (one factor ANOVA: p = 0.0006). PRL levels (lower panel): The effect of TRH was significant (one factor ANOVA:p = 0.0115). For both panels, the asterisks indicate statistically significant increases (*p ≤ 0.05, **p ≤ 0.01) compared with saline-injected controls as assessed with Bonferroni-Dunn and Student-Newman-Keuls tests.

Figure 6
figure 6

Hormone secretion in vivo in response to treatment with a combination of TRH and GHRP-6 (10 ng/g and 300 ng/g of body weight, respectively). Separated 2-d-old rats were injected intraperitoneally with the peptide and decapitated 15 min later (N = 5, n = 52, male to female ratio = 25:27). Hormone concentrations were measured in serum. GH secretion (upper panel): The effects of TRH and GHRP-6 were highly significant (two factor ANOVA, factor 1 (TRH): p = 0.0005, §; factor 2 (GHRP-6): p = 0.0001, ^). The significant interaction between the two factors (p = 0.0293, *) indicated that the combined effect of the peptides was significantly higher than the sum of the effects of separate treatments, indicating a synergistic effect. PRL secretion (lower panel): The effect of TRH was highly significant (two factor ANOVA, factor 1(TRH): p = 0.0001, †). GHRP-6 exerted no effect on PRL secretion (factor 2 (GHRP-6): p = 0.7629). There was no significant interaction between the two factors (p = 0.9054, #).

Nursing was an effective stimulus for GH secretion; it was approximately as effective as TRH. Serum levels of PRL increased only marginally upon nursing. The combined effect of nursing either with GHRP-6 or with TRH was additive (Figs. 7 and 8). No statistically significant sexual differences were found either in basal or GHRP-6- or TRH-induced secretion of either GH or PRL.

Figure 7
figure 7

Hormone secretion in vivo in response to a combination of nursing and GHRP-6 (300 ng/g of body weight). To evaluate the effect of nursing, 2-d-old rats were either allowed to remain with their mothers (pups being nursed) or separated from their mothers for 6 h (nonnursed controls). The pups were injected intraperitoneally with GHRP-6, and decapitated 15 min later (N = 6, n = 66, male to female ratio = 37:29). Hormone concentrations were measured in serum. GH secretion(upper panel): The effects of nursing and GHRP-6 were highly significant (two factor ANOVA, factor 1 (nursing): p = 0.0026, §; factor 2(GHRP-6): p = 0.0001, ^). The interaction between the two factors was nonsignificant (p = 0.6524, *), indicating that the combined effect of the peptides was not significantly different from the sum of the effects of separate treatments, i.e., the combined treatment resulted in an additive effect. PRL secretion (lower panel): The effect of nursing was significant (two factor ANOVA, factor 1 (nursing): p = 0.0476, †). GHRP-6 exerted no effect on PRL secretion (factor 2(GHRP-6): p = 0.7653). There was no significant interaction between the two factors (p = 0.5296, #).

Figure 8
figure 8

Hormone secretion in vivo in response to a combination of nursing and TRH (10 ng/g of body weight). To evaluate the effect of nursing, 2-d-old rats were either allowed to remain with their mothers (pups being nursed) or separated from their mothers for 6 h (nonnursed controls). The pups were injected intraperitoneally with TRH and decapitated 15 min later (N = 5, n = 60, male to female ratio = 33:27). Hormone concentrations were measured in serum. GH secretion (upper panel): The effects of nursing and TRH were highly significant (two factor ANOVA, factor 1 (nursing): p = 0.0001, §; factor 2 (TRH):p = 0.0007, ^). The interaction between the two factors was nonsignificant (p = 0.7037, *), indicating that the combined effect of the peptides was not significantly different from the sum of the effects of separate treatments, i.e. the combined treatment resulted in an additive effect. PRL secretion (lower panel): The effect of nursing did not reach statistical significance (two factor ANOVA, factor 1 (nursing):p = 0.1379). The effect of TRH was highly significant (factor 2(TRH): p = 0.0001, †). There was no significant interaction between the two factors (p = 0.8273, #).

DISCUSSION

Locatelli et al.(21) recently reported that, in infant rats, GHRP-6 had only a slight and transient effect on pituitary GH secretion in vitro, but exerted a robust effect on GH secretion in vivo. Therefore, due to the lack of a significant direct pituitary action, the infant rat appeared to be a promising model for studying the indirect mechanism of action of GHRP-6. However, the present observations contradict the findings of Locatelli et al.(21). In the present study, GHRP-6 was a potent GH secretagogue in 2-d-old rat pituitary glands in vitro and was approximately equipotent with TRH, resulting in an increase of approximately 200-400% over the basal secretion rate of GH. The discrepancy between the two reports might be explained by a general unresponsiveness of pituitary cultures to GH secretagogues in the report of Locatelli et al.(21): the positive control in their experiments, a challenge of the pituitary cells with 10 nM human GHRH-44, resulted in an approximately 25-30% increase over basal secretion rate, a response definitely below the severalfold increase that was to be expected(23). The present findings suggest that GHRP-6 stimulates GH secretion in 2-d-old rats both by acting directly upon the pituitary gland and indirectly via the liberation of a hypothalamic GRF, as is the case in adult rats.

GHRP-6 stimulates both GH and PRL secretion in normal and acromegalic human subjects(37). In acromegalic patients, the mode of action of GHRP-6 is similar to that of TRH(38). Therefore, TRH appeared to be good candidate for the alternative GRF liberated by GHRP-6. However, in 2-d-old rats, GHRP-6 in vivo selectively increased serum GH levels without affecting PRL, whereas TRH increased serum levels of both GH and PRL. In addition, GHRP-6 and TRH exerted a synergistic effect on GH secretion. These findings indicate that in 2-d-old rats the action of GHRP-6 in vivo is not mediated by TRH.

In 2-d-old rats in vitro, GHRP-6 (up to 1 μM) stimulated only GH but not PRL secretion, whereas TRH stimulated the release of both GH and PRL. This observation suggests that GHRP-6 and TRH act upon different pituitary receptors; GHRP-6 receptors are preferentially expressed by somatotropes, whereas TRH receptors are similarly expressed by both somatotropes and mammotropes. The finding that an in vitro combined treatment with the maximally effective doses of GHRP-6 and TRH resulted in a GH response comparable to that evoked by either treatment alone suggests that the two peptides share a common intracellular signaling system. It is well established that the TRH receptor belongs to the family of single polypeptide putative seven-transmembrane element receptors(39). The TRH receptors are coupled in a pertussis toxin-insensitive fashion to the stimulation of inositol phosphate and diacylglycerol production(40, 41); TRH-induced pituitary hormone secretion is mediated by an increase in intracellular calcium and activation of protein kinase C(42, 43). Recent evidence indicates that GHRP-6 also stimulates phosphatidylinositol turnover and activates protein kinase C(44). Moreover, GHRP-6, the related peptide hexarelin, and two non-peptide compounds (MK-0677 and L-692,429) have been reported to bind competitively to a single, G protein-coupled receptor in porcine pituitary homogenates(45). It must be noted that a combined treatment of four different human somatotroph adenoma cell cultures with GHRP-6 and TRH resulted in a GH secretion significantly higher than with either treatment alone(46). The maximally effective dose of GHRP-6 and TRH was the same for the human adenomas as for the neonatal rat pituitary glands. Renner et al.(46) have described the combined effect as additive, although this conclusion is supported neither by their statistical approach(one-way ANOVA with a post hoc test) nor by the figures which indicate that the response to the combined treatment was less than the sum of either treatment alone. A less than additive effect supports the view that the responses to GHRP-6 and TRH are at least in part mediated by the same second messenger system.

The difference between the in vitro and in vivo interactions of GHRP-6 and TRH (i.e., similar mechanism of actionin vitro, synergistic action in vivo) provides further evidence that TRH is not the mediator of the indirect GH-releasing action of GHRP-6. These data suggest that the GRF liberated by GHRP-6 in vivo not only acts upon receptors distinct from the receptors of TRH but, at least in part, appears to activate a different postreceptor signaling pathway.

The neural stimuli associated with nursing are known to induce GH secretion by liberating a GRF distinct from GHRH(20). TRH has been suggested to serve as the alternative GRF based on experiments that showed that TRH was an effective inducer of GH secretion only if the pups were separated from their mothers, but TRH was ineffective if the pups were being nursed(8). In contrast, the present results indicate that the effects of TRH and nursing were additive. Moreover, whereas TRH was a highly effective stimulus for both GH and PRL, nursing preferentially stimulated GH secretion and had only a marginal effect on PRL. These findings suggest that the GRF liberated by nursing is distinct from TRH. The finding that the effects of nursing and GHRP-6 were additive suggests that the GRF liberated by GHRP-6 is distinct from the one that mediates the nursing-induced release of GH.

PRL was measured by an Nb2 lymphoma proliferation bioassay in these experiments because the less sensitive RIA was not suitable for detection of PRL in individual samples. The Nb2 lymphoma assay is relatively specific for PRL; other than lactogenic hormones (PRL, placental lactogens, and human but not rat (GH) only IL-7 and to a lesser degree IL-2 are known to stimulate proliferation of the Nb2 cells(4749). PRL displays a remarkable molecular heterogeneity(50). The bioactivity (B) and immunoreactivity (I) of PRL variants depend on the bioassay system and the antibodies used in the assay, respectively. At any point of time, a “cocktail” of PRL variants is present in the circulation that may result in subtle differences inB/I ratios. Systemically administered TRH induced PRL secretion in adult rats either as detected by Nb2 assay or by RIA. In addition, TRH treatment increased the B/I ratio of PRL(51). The B/I ratio of PRL in 2-d-old rats is high and decreases during ontogeny(49). It is not known if stimulation by TRH increases the B/I ratio of PRL in neonatal rats. TRH has been reported to induce secretion of immunoreactive PRL from rat pituitary glands obtained on d 19 of fetal life or on d l postpartum(52). This observation is consistent with a release of not only bioactive but also immunoreactive PRL in the present experiments. Even though PRL was measured only in a bioassay, the interpretation of the data based on the difference in PRL levels after TRH and GHRP-6 treatment is unambiguous.

GHRP-6 and related peptides have proven to be highly potent inducers of GH secretion in all species studied. The present findings indicate that both the pituitary and the hypothalamic sites of action of GHRP-6 are functional in the rat as early as the 2nd d postpartum. The natural ligand of the GHRP-6 receptor is still unknown. However, it is expected to be an important regulator of GH secretion. The present results suggest that GHRP-6 and nursing liberate different GRFs, both of which are distinct from TRH. GHRH, TRH, at least two alternative GRFs, and probably the natural ligand of the GHRP-6 receptor may all be involved in the regulation of GH secretion in neonatal rats. Thus, the classical model of the hypothalamic regulation of GH needs to be expanded to include alternative GRFs.