Characterization and Regulation of the Neonatal Growth Hormone Surge

Abstract

High neonatal growth hormone (GH) secretion has been described in several species. However, the neuroendocrine mechanisms behind this surge remain unknown. Thus, the pattern of postnatal GH secretion was investigated in mice and rats. Blood GH levels were very high on postnatal day (P)1 and progressively decreased until near zero by P17 in C57BL/6 mice without sex differences. This pattern was similar to that observed in rats, except that female rats showed higher GH levels on P1 than males. In comparison, follicle-stimulating hormone exhibited higher secretion in females during the first 3 weeks of life. Hypothalamic Sst mRNA and somatostatin neuroendocrine terminals in the median eminence were higher in P20/P21 mice than in newborns. Knockout mice for GH-releasing hormone (GHRH) receptor showed no GH surge, whereas knockdown mice for the Sst gene displayed increased neonatal GH peak. Leptin deficiency caused only minor effects on early-life GH secretion. GH receptor ablation in neurons or the entire body did not affect neonatal GH secretion, but the subsequent reduction in blood GH levels was attenuated or prevented by these genetic manipulations, respectively. This phenotype was also observed in knockout mice for the insulin-like growth factor-1 (IGF-1) receptor in GHRH neurons. Moreover, glucose-induced hyperglycemia overstimulated GH secretion in neonatal mice. In conclusion, GH surge in the first days of life is not regulated by negative feedback loops. However, neonatal GH secretion requires GHRH receptor, and is modulated by somatostatin and blood glucose levels, suggesting that this surge is controlled by hypothalamic-pituitary communication.

The influence of maternal or placental hormones on fetal hypothalamic-pituitary axes is well-established (1). During late pregnancy, maternal or placental hormones suppress several neuroendocrine axes in the fetal organism through negative feedback loops (1, 2). However, the abrupt cessation of this influence after birth is believed to trigger a significant secretion of several pituitary-derived hormones in the newborn (1, 3). These hormonal surges are likely essential in the development of the endocrine system. They are thought to set the stage for the subsequent hormonal changes during puberty and adulthood, thereby underlining the importance of understanding the mechanisms involved in this physiological process (1).

In humans, there is a transitory activation of the hypothalamic-pituitary-gonadal axis in the first months of life, known as minipuberty (3, 4). Minipuberty presents notable sex differences. For example, boys show a higher luteinizing hormone (LH) secretion than follicle-stimulating hormone (FSH). Both LH and FSH become very low after 6 months of life. In girls, FSH secretion is higher than LH, and FSH levels can be elevated during the initial years of life (1, 3, 4). The gonadotropin surges during minipuberty stimulate the secretion of gonadal hormones, such as estrogens in girls and testosterone in boys (1, 3, 4). After minipuberty, the secretion of several pituitary hormones becomes suppressed until being resumed at the beginning of puberty (3, 5).

While the physiological importance of minipuberty is not entirely understood, it is clear that hormone secretion during this phase has far-reaching implications. It influences numerous features, including development, behavior, cognition, and hormone production. For example, hormones secreted during the neonatal/minipuberty period regulate the development of reproductive organs (4). In males, hormone secretion in the first months of life stimulates testicular and penile growth, as well as the proliferation and differentiation of gonadal cells (4). Minipuberty hormones possibly act in the reproductive organs of females, including the mammary gland and uterus (4). Minipuberty can also influence linear growth and body composition, and alterations in these hormones are a possible cause of hypogonadism (1, 4). Finally, there is evidence associating minipuberty hormones with cognitive development and regulation of behavior later in life (3).

Not only gonadotropins but growth hormone (GH) is also highly secreted in the neonatal period. Very high circulating GH levels are observed within the first 48 hours of postnatal life in humans, followed by a progressive decrease over the weeks (6-8). GH secretion on the first day of life has a pulsatile pattern characterized by high-amplitude and high-frequency pulsatile secretion without changes in hormone half-life (8). The postnatal GH surge is also observed in mice and rats, with peak levels on the first day of life and a progressive decline in serum GH levels over the first few weeks (9, 10).

Several variables and factors possibly regulate postnatal GH surge. Human studies have shown that plasma GH levels are higher in premature infants than full-term babies (7, 11). Postnatal GH secretion is likely regulated by dopamine signaling since dopamine infusion in newborn babies (postnatal age 10-30 hours) promptly blunts GH secretion (12). Dopamine withdrawal increases GH levels by 3-fold within 2 hours (12). A previous study also suggested that leptin signaling may regulate postnatal GH secretion in mice (9). In this regard, mice carrying somatotrope-specific leptin receptor (LepR) ablation present dysregulation of the GH secretion during the first 2 weeks of life, with higher serum GH levels in some ages [postnatal day (P)5] but reduced levels in others (P8 in females and P10 in males) (9). Despite these studies, the neuroendocrine mechanisms behind the control of the neonatal GH surge remain largely unknown. Thus, in the current study, our initial objective was to describe the pattern of postnatal GH surge in mice and rats and compare it with the secretion of other pituitary hormones. Subsequently, we investigated possible mechanisms that control the neonatal GH surge in mice.

Material and Methods

Animals

This study used Wistar rats; wild-type C57BL/6J mice (The Jackson Laboratory, Bar Harbor, ME; RRID: IMSR_JAX:000664); Lep^ob/ob mice (The Jackson Laboratory; RRID: IMSR_JAX:000632), and their respective lean littermates (Lep^ob/+); Ghrhr^lit/lit mice (The Jackson Laboratory; RRID: IMSR_JAX:000533), and their respective control littermates (Ghrhr^lit/+); and Ghr^KO/KO mice (13). Growth hormone receptor (GHR) ablation in the brain was achieved by breeding Nestin^cre/+ mice (The Jackson Laboratory; RRID: IMSR_JAX:003771) with Ghr^flox/flox animals (14). Inactivation of the insulin-like growth factor-1 (IGF-1) receptor (IGF1R) in growth hormone-releasing hormone (GHRH) neurons was accomplished by breeding Igf1r^flox/flox mice (The Jackson Laboratory; RRID: IMSR_JAX:012251) with Ghrh^Cre/+ mice (The Jackson Laboratory; RRID: IMSR_JAX:031096). In these experiments, conditional knockout mice were homozygous for the LoxP-flanked allele and carried the Cre transgene in heterozygosity, whereas control mice did not have the Cre allele. Knockdown mice for the Sst gene (SST^KD) were produced by generating animals homozygous for the Sst^Cre mutation, which display a robust reduction in somatostatin (SST) expression as formerly described (15). To visualize GHRH-expressing neurons and projections, Ghrh^Cre/+ mice were crossed with Rosa26^{CAG-LoxPSTOPLoxP-tdTomato} mouse (The Jackson Laboratory; RRID: IMSR_JAX:007909) or Rosa26^{CAG-LoxPSTOPLoxP-eGFP-L10A} mouse (16). The soma and axons of SST-expressing cells were visualized in mice carrying Sst^Flp (The Jackson Laboratory; RRID: IMSR_JAX:031629) and Rosa26^{CAG-FrtSTOPFrt-tdTomato} (The Jackson Laboratory; RRID: IMSR_JAX:032864) alleles. The mutations were confirmed through polymerase chain reaction (PCR) using the DNA extracted from samples obtained from the tail tip (REDExtract-N-Amp™ Tissue PCR Kit, MilliporeSigma, St. Louis, MO, USA). Previous studies confirmed the specificity and efficacy of the genetic deletions in Nestin^ΔGHR (17, 18) and GHRH^ΔIGF1R (19) mice. All mice were in the C57BL/6J background, and the experiments were performed in males and females. The experimental procedures were approved by the Ethics Committee on the Use of Animals of the Institute of Biomedical Sciences, University of São Paulo, and the Ethics Committee on the Use of Experimental Animals of the Universidade Federal de Minas Gerais.

Blood Collection

Adult (2- to 5-month-old) females were bred with sexually experienced males. After the pregnancy was detected, females were single-housed, and afterward, the date of birth was monitored daily. Serial blood collections began in the pups approximately 24 hours after the birth (P1). Blood samples (2 µL each) were collected from the tail tip of mice and rats 3 times per week until the animals were 3 weeks old. Blood samples were immediately transferred to a tube containing phosphate-buffered saline (PBS) with 0.05% Tween-20 (30X dilution), frozen on dry ice, and stored at −80 °C. After each blood collection, a soft fingertip pressure was applied to the tail tip to stop bleeding, and then the pups were returned to their respective dams. In a single cohort of C57BL/6J mice, trunk blood was collected after decapitation in previously anesthetized mice with isoflurane. In this experiment, blood was collected 2 hours after birth (P0) or approximately 24 hours after birth (P1). Another group of C57BL/6J mice on P1 was separated from the dams for 1 hour, followed by subcutaneous injection of D-Glucose (2 g/g of body weight) or 2-deoxy-D-glucose (2DG; 0.5 g/g of body weight; Sigma-Aldrich). Blood samples were collected immediately before the injections (baseline) and 45 minutes later to measure GH and glucose concentrations. The control group received the same manipulation and a subcutaneous injection of vehicle solution (saline). Glycemia was assessed using a portable glucose meter (Contour Plus, Ascensia Diabetes Care, Bekasi, Indonesia).

Hormone Assessment

Blood GH, LH, FSH, and prolactin levels were assessed using sensitive sandwich enzyme-linked immunosorbent assays (ELISAs). An antibody list summarizes all antibodies used (Table 1). Briefly, 50 µL of monkey anti-rat GH antibody (1:50 000; National Institute of Diabetes and Digestive and Kidney Diseases–National Hormone and Pituitary Program [NIDDK-NHPP], Cat# AFP411S; RRID: AB_2665564), mouse anti-bovine LH beta subunit antibody (1:2500; Pablo Ross, UC Davis, Cat# 518B7; RRID: AB_2756886), guinea pig anti-mouse FSH antibody (1:20 000; NIDDK-NHPP, Cat# AFP1760191, RRID: AB_2665512), or guinea pig anti-rat prolactin antibody (1:1500; NIDDK-NHPP, Cat# AFP65191; RRID: AB_2756841) were used to coat 96-well high-binding plates (9018, Corning, Kennebunk, ME) overnight at 4 °C. After decanting the coating antibody, wells were incubated with 200 µL of blocking buffer (5% skim milk powder in PBS-T) for 2 hours at room temperature (RT). The standard curves of each ELISA consisted of a 2-fold serial dilution of mouse recombinant GH (NIDDK-NHPP, Cat# AFP-10783B), mouse recombinant LH (NIDDK-NHPP, Cat# AFP-5306A), rat recombinant FSH (NIDDK-NHPP, Cat# AFP4621B) or mouse prolactin reference preparation (NIDDK-NHPP, Cat# AFP-6476C) in 0.2% bovine serum albumin PBS-T. The wells were incubated with 50 µL of samples at 1:30 dilution for 24 hours at RT. The plate was washed in PBS-T, and the wells were incubated with 50 µL of rabbit anti-rat GH antibody (1:100 000; NIDDK-NHPP, Cat# AFP5672099, RRID: AB_2721132), rabbit anti-rat LH antibody (1:40 000; NIDDK-NHPP, Cat# AFP240580Rb; RRID: AB_2665533), rabbit anti-rat FSH antibody (1:6250; NIDDK-NHPP, Cat# AFPC0972881; RRID: AB_2687903) or rabbit anti-rat prolactin antibody (1:70 000; NIDDK-NHPP, Cat# AFP131581570; RRID: AB_2722653) diluted in blocking buffer for 24 hours at 4 °C. After washing in PBS-T, wells were incubated with 50 μL of horseradish peroxidase-conjugated goat anti-rabbit IgG antibody (Sigma-Aldrich; Cat# A9169; RRID: AB_258434) diluted in 50% PBS, 50% blocking buffer at 1:30 000 for 90 minutes at RT. After a final wash in PBS-T, wells were incubated with 100 μL of 2 mg/mL o-phenylenediamine dihydrochloride (P1526, Sigma-Aldrich) in citrate-phosphate buffer (pH 5.0) containing 0.02% hydrogen peroxide for 45 minutes at RT. The reaction was stopped with 50 μL of 3 M HCl. The absorbance was determined at 490 nm with the Epoch microplate reader (Biotek, Winooski, VT), and the wavelength of 650 nm was used for background correction. The hormone levels in the samples were calculated by interpolating the optical density of the samples against a nonlinear regression of the standard curve. The lower limits of detection and the intra-assay and inter-assay coefficients of variation of each ELISA were calculated and reported in previous publications (19-24).

Table 1.

Antibody list

Antigen	Species	Finality	Source	RRID	Dilution
Rat GH	Monkey	ELISA/capture	NIDDK-NHPP, Cat# AFP411S	AB_2665564	1:50 000
Bovine LH beta subunit	Mouse	ELISA/capture	Pablo Ross, UC Davis, Cat# 518B7	AB_2756886	1:2500
Mouse FSH	Guinea pig	ELISA/capture	NIDDK-NHPP, Cat# AFP1760191	AB_2665512	1:20 000
Rat prolactin	Guinea pig	ELISA/capture	NIDDK-NHPP, Cat# AFP65191	AB_2756841	1:1500
Rat GH	Rabbit	ELISA/detection	NIDDK-NHPP, Cat# AFP5672099	AB_2721132	1:100 000
Rat LH	Rabbit	ELISA/detection	NIDDK-NHPP, Cat# AFP240580Rb	AB_2665533	1:40 000
Rat FSH	Rabbit	ELISA/detection	NIDDK-NHPP, Cat# AFPC0972881	AB_2687903	1:6250
Rat prolactin	Rabbit	ELISA/detection	NIDDK-NHPP, Cat# AFP131581570	AB_2722653	1:70 000
Rabbit IgG	Goat	ELISA/HRP-conjugated antibody	Sigma-Aldrich, Cat# A9169	AB_258434	1:30 000
c-Fos	Rabbit	IHC/primary antibody	MerckMillipore, Cat# Ab5	AB_2314043	1:10 000
Rabbit IgG	Goat	IHC/AlexaFluor594-conjugated antibody	Thermo Fisher Scientific, Cat# A-11037	AB_2534095	1:500

Abbreviations: ELISA, enzyme-linked immunosorbent assay; FSH, follicle-stimulating hormone; GH, growth hormone; HRP, horseradish peroxidase; IHC, immunohistochemistry; LH, luteinizing hormone; NIDDK-NHPP, National Institute of Diabetes and Digestive and Kidney Diseases–National Hormone and Pituitary Program.

Hypothalamic Gene Expression

The entire hypothalamus of C57BL/6 mice on P1 and P20 was collected and stored at −80 °C. RNA was extracted using TRIzol (Invitrogen, Carlsbad, CA). Subsequently, total RNA was incubated in DNase I RNase-free (MilliporeSigma, St. Louis, MO, USA), followed by cDNA synthesis using 2 µg of total RNA, SuperScript II Reverse Transcriptase (Invitrogen), and random primers p(dN)6 (MilliporeSigma). Real-time PCR was performed using the 7500TM Real-Time PCR System (Applied Biosystems, Warrington, UK), SYBR Green Gene Expression PCR Master Mix (Applied Biosystems), and specific primers for target genes: Actb (forward: gctccggcatgtgcaaag; reverse: catcacaccctggtgccta), Ghrh (forward: tatgcccggaaagtgatccag; reverse: atccttgggaatccctgcaaga), Ppia (forward: tatctgcactgccaagactgagt; reverse: cttcttgctggtcttgccattcc) and Sst (forward: ctgtcctgccgtctccagt; reverse: ctgcagaaactgacggagtct). Data were normalized to the geometric average of Actb and Ppia, and the relative quantification of mRNA was calculated by 2^−ΔΔCt.

Brain Histology

To visualize GHRH or SST-expressing neurons and axons, 2- and 21-day-old GHRH and SST-reporter mice were anesthetized with isoflurane and transcardially perfused with saline, followed by 10% formalin fixative solution (approximately 100 mL). To assess cFos expression in GHRH neurons, GHRH-reporter mice at P1 were separated from the dams for 1 hour, followed by subcutaneous injection of saline, D-Glucose (2 g/g of body weight) or 2DG (0.5 g/g of body weight). These mice were perfused 90 minutes after the injections. Brains were collected and postfixed in the same fixative containing 30% sucrose overnight at 4 °C. Subsequently, brains were cut into 30-µm thick sections using a cryostat (Leica Instruments GmbH, Nussloch, Germany), and the sections were mounted onto gelatin-coated slides. The cFos staining was performed in the slides using anti-c-Fos primary antibody (1:10,000, MerckMillipore, Temecula, CA, Cat# Ab5; RRID: AB_2314043) and Alexa Fluor⁵⁹⁴-conjugated secondary antibody (1:500, Thermo Fisher Scientific, Eugene, OR, Cat# A-11037; RRID: AB_2534095). The visualization of reporter proteins did not require any enhancement of staining. Then, brain sections were coverslipped with Fluoromount G mounting medium (Electron Microscopic Sciences, Hatfield, PA). The photomicrographs were obtained using a Zeiss Axiocam 512 color camera adapted to an Axioimager A1 microscope (Zeiss, Munich, Germany).

Statistical Analysis

Statistical analyses and graphs were generated using the Prism software (version 8.4.3; GraphPad, San Diego, CA). According to each experiment, two-way repeated measures analysis of variance (ANOVA), one-way ANOVA, and paired or unpaired two-tailed Student t tests were used to compare the experimental groups. Bonferroni's multiple comparisons test was used to identify the differences between the groups after the one-way or two-way ANOVA. All results were expressed as mean ± standard error of the mean, and only P values < .05 were considered statistically significant.

Results

Unlike Gonadotropins, Postnatal GH Secretion Does Not Present Marked Sex Differences

Serial blood collections were performed in C57BL/6 mice from day 1 of life (P1) until P21 to assess GH, LH, FSH, and prolactin levels. Very high blood GH levels were detected in P1, followed by a progressive decrease until near zero at around P17 (Fig. 1A). No differences in blood GH levels were observed between male and female mice in the first 3 weeks of life (Fig. 1A and 1E). In contrast, female mice exhibited increased blood LH and FSH levels compared to male mice in the first 3 weeks of life (Fig. 1B-1C and 1E). Both male and female mice showed very low postnatal prolactin levels without sex differences (Fig. 1D-1E). Since serial blood collections started the day after the pups were born, we compared blood GH levels immediately after birth (P0) and approximately 24 hours later (P1). No differences in blood GH levels were observed between P0 and P1 (Fig. 1F), indicating a stable and very high GH secretion during the first 24 hours of life.

Figure 1.

Unlike gonadotropins, postnatal GH secretion does not present a sex difference in mice. A-D, Blood GH, LH, FSH, and prolactin levels in male and female C57BL/6J mice from postnatal day 1 until postnatal day 19 or 21. E, Mean hormone levels in the first 3 weeks of life in male (n = 18) and female (n = 6) C57BL/6J mice. F, Blood GH levels in C57BL/6J mice immediately after birth (P0; n = 6-9/sex) and approximately 24 hours later (P1; n = 8-16/sex). *P < .05; ****P < .0001.

Serial blood collections were also performed in male and female Wistar rats to evaluate possible species-specific differences in the postnatal secretion of anterior pituitary hormones (Fig. 2A-2E). Unlike mice, female rats exhibited higher blood GH levels at P1 than males (Fig. 2A). However, no differences in blood GH levels were observed between males and females after P1, leading to similar mean GH levels between the sexes (Fig. 2A and 2E). LH and prolactin concentrations in the blood were higher at P1 and decreased after the first days of life without sex differences (Fig. 2B and 2D and 2E). FSH secretion in rats exhibited a similar pattern to mice (Fig. 2C). In this regard, female rats showed higher blood FSH levels during the second and third weeks of life compared to males, and this difference was no longer observed at P19 (Fig. 2C).

Figure 2.

Postnatal secretion of anterior pituitary hormones in rats. A-D, Blood GH, LH, FSH, and prolactin levels in male and female Wistar rats from postnatal day 1 until postnatal day 21. E, Mean hormone levels in the first 3 weeks of life in male (n = 6-9) and female (n = 6-9) Wistar rats. *P < .05.

Hypothalamic Sst mRNA Expression and SST Innervation of the Median Eminence Increase During the First Weeks of Life

To gain insights into the mechanisms regulating the postnatal GH surge in C57BL/6 mice, hypothalamic gene expression was compared between the P1 and P20 periods, in which GH secretion is very high and low, respectively. A slight but significant increase in hypothalamic expression of Ghrh mRNA was observed in P20 compared to P1 (Fig. 3). Notably, a 3-fold increase in Sst mRNA levels was observed in P20 mice compared to P1 (Fig. 3). These differences were observed in male and female mice (Fig. 3).

Figure 3.

Hypothalamic Sst mRNA levels increased 3 times on postnatal day 20 compared to postnatal day 1. Quantitative PCR analyzing the expression of Ghrh and Sst mRNA in the hypothalamus of C57BL/6J mice on postnatal day 20 (P20; males: n = 3; females: n = 6) compared to postnatal day 1 (P1; males: n = 4; females: n = 6). **P < .01; ****P < .0001.

To investigate whether changes in hypothalamic gene expression were associated with neuroendocrine terminals that reach the median eminence (ME), histological experiments were performed in P2 and P21 to visualize GHRH and SST terminals innervating the ME. Using GHRH and SST-reporter mice, we observed an intense GHRH fiber distribution in the external layer of the ME in P2 (Fig. 4A), suggesting that the GHRH-ME neurocircuitry is already formed in neonatal mice. GHRH fiber distribution in the ME remained similarly strong in P21 (Fig. 4B). In contrast, the intensity of axonal projections from SST neurons to the external layer of the ME was much weaker in P2 (5.63 ± 0.70 a.u.; n = 5) when compared to P21 (31.90 ± 0.86 a.u.; t₍₆₎ = 23.15, P < .0001; n = 3; Fig. 4C-4D). This difference was evident despite the abundant number of SST neurons in the periventricular and paraventricular hypothalamic nuclei of P2 mice (insight in Fig. 4C), which are the source of SST fibers in the ME (25, 26). Thus, increases in hypothalamic SST expression and SST fiber density in the ME are key aspects accompanying the progressive decrease in GH secretion over the first weeks of life in mice.

Figure 4.

Projections of GHRH and SST neurons to the median eminence (ME). A-D, Epifluorescence photomicrographs showing GHRH- and SST-expressing neurons and projections using tdTomato-reporter mice at postnatal day 2 (P2) or postnatal day 21 (P21). Notice the projections of GHRH and SST neuroendocrine terminals to the ME. The insight in C shows cell bodies of SST-expressing neurons in the periventricular (PV) and paraventricular nucleus of the hypothalamus (PVH) in the same animal.

Abbreviations: 3V, third ventricle; ARH, arcuate nucleus of the hypothalamus. Scale bar = 100 µm.

Postnatal GH Secretion Requires GHRH Signaling and Is Regulated by SST

To investigate the potential roles of GHRH and SST in regulating postnatal GH secretion, blood GH levels were assessed in Ghrhr^lit/lit mice. Since there is no sex difference in postnatal GH secretion in mice (Fig. 1), the results obtained in males and females were initially combined (Fig. 5A). GHRH receptor deficiency completely abolished the postnatal GH surge (Fig. 5A). Similar results were obtained when male and female mice data were analyzed separately (Fig. 5B-5C). To evaluate whether SST neurotransmission also regulates postnatal GH surge, SST knockdown (SST^KD) mice were generated, as previously described (15). As determined by real-time PCR, SST^KD mice had a 90% reduction in hypothalamic Sst mRNA expression (0.10 ± 0.01 a.u.; n = 3) compared to control mice (1.00 ± 0.05 a.u.; n = 4; t₍₅₎ = 30.08, P < .0001). SST^KD mice showed increased blood GH levels in the first days of life (Fig. 5D-5F). In addition, the area under the curve of blood GH levels during the first 3 weeks of life was higher in SST^KD mice than in control animals (Fig. 5D). Thus, postnatal GH secretion requires GHRH signaling. Additionally, hypothalamic SST modulates the magnitude of postnatal GH surge, especially in the first days of life.

Figure 5.

Postnatal GH secretion requires GHRH signaling and is regulated by SST. A, Blood GH levels and the area under the curve in Ghrhr-deficient (Ghrhr^lit/lit; n = 9) and control littermates (Ghrhr^lit/+; n = 14) from postnatal day 1 until postnatal day 21. The data from males and females were combined. B-C, Blood GH levels were shown separately in male (n = 1-6/group) and female (n = 8/group) Ghrhr^lit/lit and Ghrhr^lit/+ mice. D, Blood GH levels and the area under the curve in SST knockdown mice (SST^KD; n = 18) and control mice (n = 16) from postnatal day 1 until postnatal day 21. The data from males and females were combined. E-F, Blood GH levels were shown separately in male (n = 10-15/group) and female (n = 3-8/group) control and SST^KD mice. *P < .05; ***P < .001; ****P < .0001.

Postnatal GH Secretion Does Not Require Leptin

To gain further insights into the mechanisms regulating postnatal GH secretion, we investigated the potential role of leptin since a previous study suggested that LepR expression in somatotropic cells modulates postnatal GH secretion in mice (9). Thus, the postnatal GH surge was compared between leptin-deficient (Lep^ob/ob) mice and lean littermates (Fig. 6A-6C). Postnatal GH secretion did not differ between Lep^ob/+ and Lep^ob/ob mice, except for increased GH levels in Lep^ob/ob males on P6 compared to controls (Fig. 6A-6C). Thus, leptin does not play a major role in regulating blood GH levels in neonatal mice.

Figure 6.

Postnatal GH secretion does not require leptin action. A, Blood GH levels and the area under the curve in leptin-deficient (Lep^ob/ob; n = 11) and lean littermates (Lep^ob/+; n = 12) from postnatal day 1 until postnatal day 22. The data from males and females were combined. B-C, Blood GH levels were shown separately in male (n = 6-8/group) and female (n = 3-6/group) Lep^ob/ob and Lep^ob/+ mice. *P < .05.

Hypothalamic GHR and IGF1R Signaling Regulates the Magnitude of Postnatal GH Secretion

GHR ablation in the brain impairs GH negative feedback, increasing GH secretion and body growth (21). Similarly, IGF1R inactivation only in GHRH neurons causes gigantism, increased pulsatile GH secretion, and elevated plasma IGF-1 levels in adult mice, indicating that IGF-1 also feeds back into the hypothalamus to control the activity of the GH/IGF-1 axis (19). Some authors have suggested that neonatal GH surge is driven by the abrupt absence of maternal GH or IGF-1, causing the activation of negative feedback loops in newborns (6, 27). Thus, we investigated whether GH or IGF-1 signaling regulates postnatal GH secretion. GHR ablation in the brain (Nestin^ΔGHR mice) did not affect neonatal GH secretion (first days of life) but significantly increased GH secretion in the second and third weeks of life compared to control animals (Fig. 7A-7C). Whole-body Ghr^ko/ko mice also showed similar GH secretion in the first days of life compared to controls (Fig. 7A-7C); however, the complete absence of GHR signaling prevented the progressive reduction in blood GH levels observed during the first weeks of life (Fig. 7A-7C). When GHRH^ΔIGF1R mice were investigated, these animals exhibited increased postnatal GH secretion compared to control mice as determined by the area under the curve of blood GH levels in the first 3 weeks of life (Fig. 7D-7F). Thus, both GHR and IGF1R signaling regulates postnatal GH secretion in mice, although these mutations did not affect the peak of GH secretion in the first days of life.

Figure 7.

GHR and IGF1R signaling regulates the magnitude of postnatal GH secretion. A, Blood GH levels and the area under the curve in mice carrying GHR ablation in the brain (Nestin^ΔGHR; n = 6) or the whole body (Ghr^KO/KO; n = 7), and control mice (n = 9) from postnatal day 1 until postnatal day 21. The data from males and females were combined. B-C, Blood GH levels were shown separately in male (n = 3-5/group) and female (n = 2-4/group) Nestin^ΔGHR, Ghr^KO/KO, and control mice. D, Blood GH levels and the area under the curve in mice carrying IGF1R ablation in GHRH-expressing cells (GHRH^ΔIGF1R; n = 8) and control littermates (n = 14) from postnatal day 1 until postnatal day 21. The data from males and females were combined. E-F, Blood GH levels were shown separately in male (n = 5-8/group) and female (n = 3-6/group) GHRH^ΔIGF1R and control mice. * P < .05; **** P < .0001.

Hyperglycemia Overstimulates GH Secretion in Neonatal Mice

Our previous results suggest that GH secretion on the first day of life is not regulated by negative feedback from GH and IGF-1. This led us to explore alternative mechanisms controlling GH peak on P1. A classical study by Cornblath et al (7) provides evidence that changes in glycemia regulate GH secretion in the first days of life in humans. Thus, we examined GH levels in P1 mice following an acute infusion of glucose or 2DG, a glucopenic drug that triggers a counter-regulatory response (28-30). As anticipated, both glucose and 2DG infusions led to a significant rise in blood glucose levels in P1 mice, with glucose infusion inducing a higher increase in glycemia than 2DG (Fig. 8A). Intriguingly, control (saline) injection led to a significant rise in blood GH levels compared to baseline (Fig. 8B). Both glucose and 2DG injections also triggered substantial increases in blood GH levels compared to baseline (Fig. 8C-8D). However, only glucose injection resulted in a relative increase in blood GH levels compared to saline- or 2DG-injected mice (Fig. 8E).

Figure 8.

Hyperglycemia overstimulates GH secretion in neonatal mice without altering the expression of cFos protein in ARH^GHRH neurons. A, Blood glucose levels 45 minutes after an acute injection of saline, glucose, or 2-deoxy-D-glucose (2DG) in mice at postnatal day 1 (n = 8-11/group). B-D, Blood GH levels at baseline and 45 minutes after an acute injection of saline, glucose, or 2DG in mice at postnatal day 1 (n = 8-10/group). E, Relative increase in blood GH levels compared to baseline. F-H, Representative photomicrographs showing the colocalization between cFos (nuclear labeling) and GHRH (visualized via eGFP reporter protein) in the arcuate nucleus of the hypothalamus (ARH) 90 minutes after an acute injection of saline, glucose, or 2DG in mice at postnatal day 1. The arrows indicate examples of double-labeled neurons shown in higher magnification in the insights.

Abbreviation: 3V, third ventricle. Scale bar = 50 µm. *P < .05; **P < .01; ***P < .001; ****P < .0001.

GHRH neurons in the arcuate nucleus of the hypothalamus (ARH) are known to be glucose-sensing cells, and their activity is increased when glucose availability is decreased or after 2DG stimulus (29, 30). Thus, we evaluated the cFos expression, a marker of neuronal activation, in ARH^GHRH neurons after acute injections of saline (Fig. 8F), glucose (Fig. 8G), or 2DG (Fig. 8H) in P1 mice. Few ARH^GHRH neurons expressing cFos were observed in control (1.5 ± 0.0 cells/section; n = 3), glucose (1.0 ± 0.3 cells/section; n = 5), and 2DG-injected mice (2.2 ± 0.6 cells/section; n = 4), without significant differences between the groups (Fig. 8F-8H). Thus, glucose-induced hyperglycemia overstimulates GH secretion in neonatal mice without altering the expression of cFos protein in ARH^GHRH neurons.

Discussion

In the present manuscript, we confirmed and expanded the findings of human and animal studies (7-12), demonstrating that GH is highly secreted in neonatal rats and mice and presents a progressive decline during the first weeks of life. Unlike gonadotropins, postnatal GH secretion shows minor sex differences, detected only in P1 rats. Subsequently, we used different mouse models and approaches to demonstrate that postnatal GH secretion (i) requires GHRH action; (ii) is regulated by hypothalamic SST; (iii) does not require leptin; (iv) is modulated by GH and IGF-1 negative feedback but not in the first days after birth; and (v) is regulated by blood glucose levels.

Compared to newborns (P1), Ghrh mRNA expression is increased in P20 mice. This increase likely reflects the development of hypothalamic neurons during the first weeks of life (31-33). Nonetheless, GHRH neuron terminals in the ME are abundantly found in newborn mice, suggesting that the capacity of hypothalamic GHRH to regulate GH secretion is functional in the very early stages of postnatal life. Furthermore, evidence indicates that neonatal rats are more sensitive to the stimulatory effect of GHRH than post-weaning or adult animals (34, 35). This increased responsiveness is also observed in neonatal pituitary experiments in vitro (36). The completely blunted postnatal GH surge in GHRH receptor-deficient (Ghrhr^lit/lit) mice further confirmed the importance of GHRH signaling in controlling early-life GH secretion. Although we cannot rule out that postnatal GH secretion may also be controlled at the pituitary level, our findings indicate that early-life GH secretion is mainly regulated by hypothalamic-pituitary communication, primarily via GHRH release.

SST inhibits GH secretion (37-39). Not only was Sst mRNA expression robustly increased in the hypothalamus of P20 mice compared to newborns, but a robust increase in SST neuron terminals was observed in the ME, coinciding with the decrease in GH secretion at the end of the third week of life. SST peptide content also increases from birth until reaching its peak by the fourth week of life in rats (10). Importantly, hypothalamic SST content shows a significant negative correlation with serum GH concentration during the first weeks of life in rats (10). SST^KD mice were used to test the hypothesis that SST regulates postnatal GH secretion. A limitation of this model is that SST was not completed ablated, and the 10% remaining SST expression may be sufficient to prevent significant neuroendocrine consequences (15). Despite the residual SST expression, SST^KD mice showed increased postnatal GH secretion. Since GHRH is required for postnatal GH secretion, the early-life effects of SST on GH secretion are likely mediated by the regulation of the activity of GHRH neurons or the response of somatotroph cells to GHRH, which are mechanisms also observed in adult animals (38, 39). It is important to note that former studies described reduced expression of SST receptors in the pituitary gland of newborn rats, followed by a robust increase during the first weeks of life (40, 41). Thus, our findings and previous studies suggest a significant role of SST in controlling GH secretion during the first weeks of life, especially in the first days of life.

Somatotroph cells express the LepR. LepR ablation using a GH-Cre mouse line reduces pituitary GH content and serum GH concentration and causes obesity in adult mice (42, 43). Nonetheless, a previous study has shown that postnatal GH secretion was not prevented in mice carrying LepR deletion in somatotroph cells (9). Additionally, fasting-induced changes in GH secretion are leptin-independent (44). Notably, a postnatal leptin surge has been described in mice and rats, occurring mainly from the end of the first week of life until the beginning of the third week (45-47). Thus, the postnatal leptin surge does not coincide with the peak of neonatal GH secretion. Using leptin-deficient mice, we confirmed these early observations (9), indicating none or, at most, a minor role of leptin in regulating postnatal GH surge. Pubertal and adult Lep^ob/ob mice show reduced GH secretion (48). However, this effect is likely secondary to their obese phenotype. Importantly, no differences in body composition are observed between Lep^ob/ob mice and lean controls until the second week of life (49), further supporting the normal postnatal GH peak observed in Lep^ob/ob mice. Decreased GH secretion is also observed in nonmutant obese individuals and animals oversecreting leptin (50-52). Hyperinsulinemia (52-54) and increased free fatty acid flux (51) have been postulated as potential mechanisms linking obesity with reduced GH secretion.

Numerous redundant negative feedback loops regulate GH secretion (38, 39). In this sense, both GH and IGF-1 can act either in hypothalamic neurons (17, 19, 21, 22, 55) or in the pituitary gland (56, 57) to control the pattern of GH secretion. Several authors have suggested that the cessation of the influence of maternal/placental hormones relieves negative feedback mechanisms, leading to a high secretion of pituitary hormones during the early postnatal period (1, 3). The same principle has been postulated to explain the neonatal GH surge (6, 11). In contrast to this idea, we show that interfering with GH and IGF-1 negative feedback did not alter GH peak in the first days of life. Thus, the early peak of GH secretion in neonatal mice is not regulated by negative feedback loops. However, the absence of such feedback induced significant increases in the area under the curve of blood GH levels in the first 3 weeks of life. The capacity of GH to stimulate hepatic IGF-1 production begins to be observed from the third week of life (58). Thus, despite the immaturity of the somatotropic axis, GH and IGF-1 are able to feed back into the hypothalamus and control GH secretion in the first weeks of life. The more substantial alterations in GH secretion observed in Ghr^KO/KO mice, compared to Nestin^ΔGHR or GHRH^ΔIGF1R mice, suggest the participation of extrahypothalamic tissues in this control, possibly via the pituitary gland.

GH secretion is robustly increased in situations of metabolic stress (59-61), including hypoglycemia (28, 62). While somatotrophs do not express glucokinase (63), a protein typically found in glucose-sensing cells, glucokinase is abundantly expressed by ARH^GHRH neurons (29). Accordingly, the activity of ARH^GHRH neurons is decreased in response to increases in extracellular glucose concentration (29). Conversely, 2DG injection induces cFos expression and the activation of ARH^GHRH neurons in adult mice (30). However, 2DG did not induce cFos expression in ARH^GHRH neurons of neonatal mice nor amplified GH secretion beyond the increases observed in saline-injected newborns. Paradoxically, glucose infusion augmented GH secretion in P1 mice. Although these findings are the opposite of those in adults, glucose-induced hyperglycemia also stimulates GH secretion in full-term babies (7). Therefore, our data are in accordance with the few human studies investigating this aspect (7). The causes for this discrepancy between newborns and adults are unknown. Speculatively, it may be related to changes in the expression of ion channels in hypothalamic neurons during the first weeks of life. Baquero et al (64) have shown that ARH neurons expressing the neuropeptide Y (NPY) are depolarized by leptin in P13 to P15 mice. In contrast, leptin starts to induce hyperpolarization in these cells in P21 to P23 mice (64). The striking switch of leptin's effect on the activity of ARH^NPY neurons is explained by the acquisition of functional ATP-sensitive potassium (K_ATP) channels during the periweaning period, whose expression was absent earlier (64). Since K_ATP channels are critical proteins involved in the ability of cells to respond to nutrient (glucose) availability (65, 66), a similar switch might occur in ARH^GHRH neurons during development. Future studies are desired to investigate this possibility.

The increase in GH secretion in saline-injected newborns compared to baseline is intriguing. We predict maternal or food deprivation, changes in body temperature, or stress as possible causes for the increased GH secretion observed in the 45-minute interval between baseline and subsequent blood collection after saline injection. Two hours of maternal separation decreases serum GH concentration in neonatal rats, whereas 30 minutes of suckling restores GH levels (67). Thus, neither maternal nor food deprivation seems to explain the increased GH secretion observed in our experiment. Another study showed that exposing 2-day-old rat pups to 37 °C increases serum GH concentration compared to pups housed at 22 °C for 6 hours (68). Moreover, several neural mechanisms were proposed to regulate postnatal GH secretion, including those mediated by dopamine (12), acetylcholine (69, 70), serotonin (69, 70), α2-agonist clonidine (69, 70), and thyrotropin-releasing hormone (71). The release of these neurotransmitters may be altered in stressful situations, like the subcutaneous injections employed in our experimental animals. Finally, ghrelin could act as a GH secretagogue in the postnatal period. However, several studies demonstrated that although newborns have considerable levels of total and acyl-ghrelin in their circulation, no correlation was observed between GH and ghrelin concentrations (72-75), suggesting that neonatal GH peak is not driven by ghrelin.

In conclusion, our findings provide novel and relevant information about the neuroendocrine mechanisms regulating postnatal GH surge (Fig. 9). Our results indicate that negative feedback loops do not passively regulate GH surge in the first days of life. Instead, it is likely that the hypothalamus intentionally drives GH secretion in neonatal animals via the balance between GHRH and SST secretion. Thus, what could be the physiological importance of GH in the first days of life? Although Ghrhr^lit/lit mice and other GH or GHR deficiency models are viable, the lack of GHR signaling in early life may cause several consequences. For example, the lipolytic effect of GH may provide essential nutrients to maintain cell metabolism immediately after birth, which is a moment when nutrient supply becomes acutely limited. Since GH increases blood glucose levels by inducing insulin resistance and stimulating hepatic glucose production (59-61), neonatal GH peak may represent a neuroendocrine mechanism to prevent hypoglycemia in newborns. GH-induced glucose production and mobilization of free fatty acids may also support brown adipose tissue thermogenesis to avoid hypothermia. Finally, GH-responsive cells are abundantly found in numerous brain areas (76, 77). Thus, GH could act as a neurotrophic factor regulating the development and maturation of several neuronal populations (60). Accordingly, GH- or GHR-deficiency impairs the formation of axonal projections of ARH neurons (78, 79). Thus, future studies should be carried out to investigate the physiological importance of the neonatal GH surge and the consequences of its absence.

Figure 9.

Schematic diagram summarizing the timeline of the postnatal regulation of GH secretion. Note the respective ages at which these mechanisms controlling GH secretion are observed. The mouse images from postnatal days 0 and 8, respectively, were extracted from The Jackson Laboratory®.

Abbreviations

2DG

2-deoxy-D-glucose

ANOVA

analysis of variance

ARH

arcuate nucleus of the hypothalamus

ELISA

enzyme-linked immunosorbent assay

FSH

follicle-stimulating hormone

GH

growth hormone

GHR

growth hormone receptor

GHRH

growth hormone-releasing hormone

IGF-1

insulin-like growth factor-1

IGF1R

insulin-like growth factor-1 receptor

LepR

leptin receptor

LH

luteinizing hormone

ME

median eminence

NIDDK-NHPP

National Institute of Diabetes and Digestive and Kidney Diseases–National Hormone and Pituitary Program

P#

postnatal day#

PBS

phosphate-buffered saline

PCR

polymerase chain reaction

RT

room temperature

SST

somatostatin

Funding

This work was supported by the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP-Brazil; grant number: 2020/01318-8 to J.D.J., 2021/03316-5 to D.O.G., 2023/11833-5 to L.M.M.S. and 2024/03067-3 to S.J.R.R.), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq/Brazil, 306024/2023-3 and 404393/2023-3 to J.D.J.), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES/Brazil; Finance Code 001), Fundacao de Amparo a Pesquisa do Estado de Minas Gerais (FAPEMIG; grant number: APQ-02538-21, APQ-04900-22) and National Institutes of Health (NIA grant number: R01AG059779 to J.J.K. and E.O.L.).

Disclosures

R. E. Szawka is an Editorial Board Member for Endocrinology and played no role in the Journal's evaluation of the manuscript.

Data Availability

The data supporting this study's findings are available from the corresponding author upon reasonable request.