Pubertal- and Stress-Dependent Changes in Cellular Activation and Expression of Excitatory Amino Acid Receptor Subunits in the Paraventricular Nucleus of the Hypothalamus in Male and Female Rats

The hypothalamic-pituitary-adrenal (HPA) axis is the primary neuroendocrine mediator of the hormonal stress response. Upon experiencing a stressor, neurons in the paraventricular nucleus (PVN) of the hypothalamus are activated and secrete corticotropin-releasing hormone (CRH) into the hypophyseal portal system, which in turn stimulates the anterior pituitary gland to release adrenocorticotropic hormone (ACTH). ACTH then leads to the production and secretion of the glucocorticoids from the adrenal cortex, namely, cortisol in primates and corticosterone in many rodent species [1]. In the short term, exposure to stress-induced elevations of glucocorticoids leads to many adaptive responses, including enhanced cognitive function and the mobilization of needed energy. However, prolonged or more chronic exposure to these hormones can lead to maladaptive responses, such as disrupted emotional function and altered metabolism [2, 3]. Thus, factors that modulate stress-induced HPA reactivity can have significant effects on mental and physical health and well-being.

Pubertal maturation is associated with changes in the function of the HPA axis [4‒8]. Specifically, compared to adults, prepubertal male rats show more prolonged stress-induced ACTH and corticosterone responses following a variety of stressors, including foot shock, ether inhalation, and restraint stress [9‒11]. We have also shown that prepubertal female rats mount a significantly protracted hormonal stress response compared to adult females following restraint stress [12, 13]. It is unclear what the ramifications are for this greater hormonal stress reactivity prior to puberty, but it may contribute to the higher levels of stress-related dysfunction often noted during a stressful adolescence [14‒16]. Also unclear are the neuroendocrine mechanisms mediating these age-dependent changes in hormonal stress reactivity. However, it has been noted that the PVN exhibits significantly greater stress-induced cellular activation, as measured by FOS immunohistochemistry, in prepubertal males compared to adult males [17‒20]. Therefore, these data suggest that greater stress-induced activation of the PVN may mediate the protracted hormonal stress response observed prior to pubertal maturation, at least in males.

Stress-induced activation of the HPA axis is in part mediated by glutamatergic inputs to the PVN from a variety of forebrain areas, including cell groups in the hypothalamus [21, 22]. Thus, the protracted hormonal response observed in prepubertal males and females compared to their adult counterparts may be partially mediated by a greater sensitivity to the excitatory glutamatergic inputs to the PVN prior to puberty. To further investigate the role of the PVN in mediating pubertal changes in hormonal stress reactivity, two experiments were conducted. The first experiment examined whether the PVN of female rats showed similar age-related differences in cellular activation patterns following a single session of restraint stress, as has been previously observed in males [17‒20]. The second experiment explored whether age or stress exposure in males and females altered the expression of ionotropic glutamate receptor subunits known to show relatively high expression levels in the PVN, namely, Grin1, Grin2a, and Grin2b (NMDA subunits); Gria1 and Gria2 (AMPA subunits); and Grik1 and Grik2 (kainate subunits) [23‒25]. Given the heightened stress-induced hormonal response often observed prior to puberty in males and females, we predict greater stress-induced FOS responses and higher expression levels of ionotropic glutamate receptor subunits in the PVN of prepubertal males and females compared to their adult counterparts.

Two experiments were conducted. Thirty-six male and 36 female Sprague Dawley rats were obtained from our breeding colony at Columbia University and used for each experiment. All animals were weaned at 21 days of age. Animals were housed two per cage in clear polycarbonate cages (22 × 28 × 38 cm) that contained wood chip bedding. Cagemates were the same age and sex but derived from different litters, and no more than two siblings were used in each experimental group. All animals were maintained on a 12-h light-dark schedule (lights on at 07:00 h), and the temperature in the room was maintained at 21 ± 2°C. Standard rodent chow (LabDiet 5012; PMI Nutrition International, LLC, Brentwood, MO, USA) and water were available ad lib. All procedures were carried out in accordance with the guidelines established by the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee (IACUC) of Columbia University.

In experiment 1, prepubertal (30 days of age) and adult (70 days of age) male and female animals were perfused either before (basal), immediately after (0 min), or 40 min following a 30-min session of restraint stress (n = 6 per age, sex, and time point). The estrous cycle was not tracked in the adult females. At the time of perfusion, animals were weighed and administered an overdose of ketamine (80 mg/kg; i.p.) and xylazine (5 mg/kg; i.p.). Following deep anesthetization, animals were transcardially perfused with 0.9% heparinized saline followed by 4% paraformaldehyde in 0.1 m phosphate buffer (PB). Brains were then postfixed for 24 h in 4% paraformaldehyde, cryoprotected in 20% sucrose in 0.1 m PB for 24 h, snap-frozen on powdered dry ice, and stored at −20°C until sectioning (see below). Restraint stress was administered by placing the animals in the prone position in wire mesh restrainers, sized so that animals at different developmental stages were equally restrained. The animals in the 40-min group were returned to their home cages until tissues were collected at the 40-min post-stress time point. All animals were killed between 1100 and 1400 h during their circadian nadir of HPA activity.

In experiment 2, prepubertal (30 days of age) and adult (70 days of age) male and female animals were exposed to 30 min of restraint stress as described above. However, for this experiment, animals were killed by rapid decapitation using a guillotine, and a blood sample was taken either before (basal), immediately after (0 min), or 40 min following restraint stress (n = 6 per age, sex, and time point). Similar to experiment 1, the estrous cycle was not tracked in the adult females. Brains were removed, snap-frozen on dry ice, and stored at −80°C until further processing for tissue punching and qRT-PCR (see below).

Brains were sectioned on a cryostat in the coronal plane at a 40-μm thickness. Sections were placed in cryoprotectant solution of 1 part 20% sucrose in 0.1 m PB and 1 part ethylene glycol. Sections were stored in the cryoprotectant solution at −20°C until processed for histology. For each animal, every 8th section through the hypothalamus was mounted and Nissl-stained to ensure the tissue sections processed for FOS immunohistochemistry contained the PVN. Two free-floating sections separated by 120 μm that contained the PVN were rinsed three times in 0.1 m PB and then incubated in 0.05% H₂O₂ in 0.1 m phosphate buffered saline (PBS) for 5 min. Sections were then rinsed 3 times for 5 min each in 0.1 m PB with 0.1% triton X-100 (PBT) and then blocked in 2% normal goat serum in the PBT solution for 1 h at room temperature (RT). Sections were incubated for 48 h at 4°C in FOS anti-rabbit primary antibody (1:20,000; Abcam; Waltham, MA) in blocking solution. Sections were then rinsed three times for 5 min each in PBT and then incubated in a secondary antibody (1:200; goat-anti-rabbit IgG; Vector Laboratories; Burlingame, CA) in PBT for 1 h at RT. Sections were then rinsed three times for 5 min each in PBT and then incubated in avidin-biotin horseradish peroxidase complex (1:250; VECTASTAIN Elite ABC Kit; Vector Laboratories) in PBT for 1 h at RT. Tissues were then rinsed three times in PBS for 5 min each and then exposed to nickel-enhanced 3,3′-diaminobenzidine in a 3N sodium acetate buffer (Millipore-Sigma, St. Louis, MO) containing 0.05% H₂O₂. To stop the 3,3′-diaminobenzidine reaction, sections were washed once for 1 min in PBS solution, followed by three 5-min washes in PBS solution. Tissues were mounted onto Fisherbrand Plus slides (Fisher Scientific, Pittsburg, PA). After air drying, sections were dehydrated in an ascending series of ethanol solutions (70%, 95%, and 100%), cleared in xylene, and coverslipped with DPX mounting medium (Millipore-Sigma). FOS-positive cells in the parvocellular region of the PVN (bregma −2.40 to −2.76 mm [26]) were counted from images captured under a Zeiss 200 m Axiovert light microscope (Carl Zeiss, Thornwood, NY) using a ×20 objective. Using a superimposed ocular grid of 40,000 µm² on each image, bilateral counts were made for each section and averaged to compute the mean number of FOS-positive cells (Fig. 1).

For qRT-PCR, methods have been described previously [27]. Briefly, to measure NMDA, AMPA, and kainate receptor mRNA expression, specifically the Grin1, Grin2a, Grin2b, Gria1, Gria2, Grik1, and Grik 2 receptor subunits, brains were mounted on a cryostat and punched using a 0.5-mm diameter sample corer tool (Fine Science Tools, Foster City, CA, USA). Bilateral punches were made of the PVN, corresponding to plate 42 in a standard rat brain atlas [26], then placed in 50 μL of RNAlater solution (Millipore-Sigma), stored at 4°C for 24 h, and then transferred to −20°C until further processing.

To prepare tissue, punches were homogenized in RNAzol (Molecular Research Center, Cincinnati, OH) and stored at −80°C until processing. RNA isolation was in accordance with the manufacturer’s protocol. First-strand cDNA synthesis was in accordance with New England Biolabs MmULV protocols (New England Biolabs, Ipswitch, MA) with random priming. We used predesigned and pre-validated TaqMan assays from Applied Biosystems (Life Technologies, Norwalk, CT) run in multiplex with the housekeeping gene Eukaryotic 18S rRNA, 18 s. For each plate and assay, gene expression was calculated based on a standard curve included on each plate. We used a CFX384 RT-qPCR system (Bio-Rad, Hercules, CA) and associated software for all gene expression profiling. Table 1 includes the gene symbol, the gene name, and TaqMan assay Id (Table 1).

Plasma corticosterone concentrations were measured using a commercially available kit (MP Biomedicals, Solon, OH). Samples were run in duplicate, and the values were averaged. The coefficient of variation and the lower limit of detectability were 6.4% and 25.38 ng/mL, respectively.

For both experiments 1 and 2, data were analyzed using two-way ANOVAs, with age and time point as the variables of analysis. Significant effects were further analyzed with Tukey’s honestly significant difference post hoc tests. It is important to note that though the experimental designs were identical, data analyses for males and females were conducted separately as these male and female subjects were derived from different cohorts of rats. In experiment 1, one basal adult male and one 40-min post-stress time point adult female were excluded from all analyses due to poor tissue fixation. In experiment 2, due to poor sample quality with wide variability within triplicate values, some of the groups ultimately had an n = 5. All data are presented as the mean ± SEM. Differences were considered significant when p < 0.05. All statistical analyses were performed using GraphPad Prism, version 8.3.0 (GraphPad Software Inc., San Diego, CA).

In males, a two-way ANOVA revealed a significant main effect of age on the number of FOS-positive cells in the PVN. Specifically, prepubertal males had a significantly greater number of FOS-positive cells compared to adults (F(1, 29) = 14.69, p < 0.05). A main effect of time point in males was also observed (F(2, 29) = 19.35, p < 0.05), such that at the 0- and 40-min post-stress time points, the number of FOS-positive cells was greater than under the basal, non-stressed condition (Fig. 2a). However, there was no statistically significant interaction between age and time. Thus, these data suggest that exposure to stress resulted in increased cellular activation of the PVN, with prepubertal males overall showing greater activation compared to adults.

In females, a main effect of age was also found, with adult females exhibiting a significantly greater number of FOS-positive cells compared to prepubertal females (F(1, 28) = 5.299, p < 0.05). Therefore, the main effect of age in females was the opposite of that found in males. However, similarly to males, the two-way ANOVA revealed a significant main effect of time point on FOS-positive cell expression in the PVN of females (F(2, 28) = 29.97, p < 0.05). Specifically, both prepubertal and adult females displayed a significantly greater number of FOS-positive cells immediately after the 30-min session of restraint stress and 40 min after the stressor had been terminated compared to the basal time point (Fig. 2b). Hence, these data suggest that stress increases cellular activation in the PVN, but unlike in the males, the adult females are showing greater activation overall compared to the prepubertal females.

There was a significant interaction between age and time point on male corticosterone levels (F(2, 30) = 5.14, p < 0.05), such that prepubertal males exhibited significantly higher plasma corticosterone concentrations compared to adult males 40 min following the 30-min session of restraint stress (Fig. 3a). There was also a significant interaction between age and time point on corticosterone levels in females (F(2, 30) = 10.38, p < 0.05), with adult females at the 0-min time point exhibiting significantly greater levels of plasma corticosterone than the prepubertal females at the 0-min time point Fig. 3b).

For males, main effects of age and stress time points were noted for Grin1 (F(1, 27) = 6.53 and F(2, 27) = 3.59, respectively, p < 0.05) and Grin2b (F(1, 26) = 10.48 and F(1, 26) = 5.04, respectively, p < 0.05). Specifically, adult males had higher expression of these two subunits compared to prepubertal males, and expression levels were lowest at the 40-min post-stress time point, independent of age (Fig. 4a, c). There were no significant main effects or interaction for Grin2a (Fig. 4b). For females, there were significant interactions between age and stress time point for all three subunits (F(2, 30) = 4.34, F(2, 29) = 4.64, and F(2, 30) = 3.46 for Grin1, Grin2a, and Grin2a, respectively, p < 0.05), with prepubertal females at the 40-min post-stress time point having higher expression levels than the prepubertal females at either the basal or 0-min time points (p < 0.05; Fig. 4d–f).

For males, main effects of age and stress time point were found for Gria1 (F(1, 25) = 12.37 and F(2, 25) = 4.67, respectively, p < 0.05). Specifically, adult males had higher expression levels compared to prepubertal males, and expression levels were lowest at the 40-min post-stress time point, independent of age (Fig. 5a). For Gria2, there was a main effect of age (F(1, 26) = 16.63, p < 0.05), such that adult males had higher expression levels compared to prepubertal males (Fig. 5b). For females, there were no significant main effects or interaction for Gria1 (Fig. 5c). For Gria2, there was a significant interaction between age and stress time point (F(2, 29) = 3.39, p < 0.05), with adult females at the basal time point having higher expression levels than the prepubertal females at the basal time point (Fig. 5d).

For males, there were only main effects of age for both Grik1 and Grik2, such that adult males had higher expression levels compared to prepubertal males (F(1, 25) = 9.62 and F(1, 27) = 4.60, respectively, p < 0.05; Fig. 6a, b). For females, there was a significant interaction between age and stress time point for Grik1 (F(2, 30) = 3.78, p < 0.05), with prepubertal females at the 40-min post-stress time point having higher expression levels than the prepubertal females at either the basal or 0-min time points (p < 0.05; Fig. 6c). For Grik2, there was only a main effect of age, such that adult females had higher expression levels compared to prepubertal females (F(1, 29) = 5.54, p < 0.05; Fig. 6d).

Collectively, these experiments show that pubertal maturation and stress affect activation of the PVN and the expression of ionotropic glutamate receptor subunits in the PVN. It also appears that these effects of age and stress on the PVN are different in males and females. In the context of stress-induced PVN activation, our current data agree with that previously published in males; in that, the PVN of prepubertal males has been reported to show greater FOS responses following a 30-min session of restraint stress compared to adult males [17‒20]. In females, the main effect of age, with adult females having a greater number of FOS-positive cells compared to prepubertal females, was unexpected, given the previously reported protracted hormonal stress response in prepubertal compared to adult females [12, 13]. An earlier study showed that restraint stress increased the number of FOS-positive cells in PVN to a similar degree in prepubertal and adult female rats [17]. However, it should be noted that this earlier study used adult rats at an earlier age than used here (i.e., 60 days of age) and only assessed the FOS response immediately after termination of a 30-min session of restraint [17]. Independent of the differences in design and outcome of this previous study and our current results, together, these studies indicate that the stress-induced FOS response in the PVN in prepubertal and adult animals is different in males and females. Though collected from different cohorts of female subjects, the greater hormonal stress reactivity measured in the adult compared to the prepubertal females (experiment 2) is parallel to the greater FOS response in the adult females (experiment 1). Though it is unclear why we do not observe a greater hormonal response in the prepubertal females, as has been previously reported [12, 13], it is possible that a significant portion of our adult female rats were in proestrus, when hormonal stress reactivity is highest [28], and thus potentially obscuring a pubertal difference in hormonal reactivity. Regardless, future experiments will need to further investigate this relationship between changes in hormonal stress reactivity and cellular activation in the PVN in females. It will also be important for future experiments to investigate whether CRH-containing neurons within the PVN display differential activation in prepubertal compared to adult females, as previously reported in males [18].

Similar to the prediction above regarding cellular activation in the PVN, given the greater hormonal stress response often reported in prepubertal compared to adult rats, we had predicted that the PVN of prepubertal animals would have higher expression levels of NMDA, AMPA, and kainate receptor subunits compared to their adult counterparts. However, our results do not support this notion. In fact, adult males had higher levels of Grin1, Grin2b, Gria1, Gria2, Grik1, and Grik2 expression in the PVN compared to prepubertal males. Thus, there appears to be no clear association between the expression levels of these excitatory amino acid receptor subunits in the PVN and the greater hormonal stress reactivity observed in these prepubertal compared to adult males. In females, expression levels of all these subunits were largely similar in prepubertal and adult subjects, except Grik2, which showed higher expression levels in the PVN of adult females, and hence may be related to the greater hormonal stress reactivity reported in these adult females. It should be noted that stress did elevate the expression of the NMDA and Grik1 subunits in prepubertal females, so unlike what was observed in males, stress-induced regulation of these subunits appears to be greater in females prior to pubertal maturation.

Exposure to various chronic stressors in adulthood has been shown to modulate the expression of the ionotropic glutamate receptor subunits in brain areas such as the hippocampus, hypothalamus, and ventral tegmental area [25, 29‒31]. Moreover, though a few reports have indicated transient increases in AMPA receptor binding and increased NMDAR1 expression in the hypothalamus of female rats entering puberty [32, 33], to our knowledge, this is the first report on how acute stress and pubertal maturation affect the expression of these glutamate receptor subunits in the PVN of male and female rats. Unfortunately, as our samples were obtained through tissue punching, it is unclear whether these changes in gene expression are within particular subdivisions of the PVN, such as the parvocellular or magnocellular regions, or within particular cell types, such as CRH neurons. Future experiments could address these questions using alternative experimental approaches, such as RNAscope that would help quantify mRNA expression and provide cell-specific resolution of transcript.

It should be noted that no direct statistical comparisons were made between the sexes for these receptor subunits in the PVN, but similar to the FOS results, it appears that the expression of these genes is affected by age and stress differently in males and females. Specifically, we show that adult males tend to have higher expression levels of NMDA, AMPA, and kainate subunits in the PVN than prepubertal males, while females do not show this age difference, except for the higher expression levels of Grik2 in adult females. Moreover, stress reduces the expression of NDMA and AMPA receptor subunits in males, while stress increases the expression of the NMDA and Girk1 subunits in prepubertal females. Though it is unclear what may mediate these differences between males and females, these data indicate that the effects of stress and age on the expression of many of these ionotropic glutamate receptor subunits are further influenced by the sex of the individual. Gonadal hormones have been shown to alter the transcription of these ionotropic receptors and their subunits in a variety of brain regions [34‒37], including the hypothalamus [38, 39]. In adult female rats, ovariectomy leads to decreased NR2b and GluR1 levels in the hippocampus and striatum, respectively [34, 37], while estradiol treatments in adult female rats and dihydrotestosterone treatments in adult males increase NMDA receptor binding in the hippocampus [35, 36]. Moreover, progesterone has been noted to reduce kainate receptor subunit expression in the hypothalamus of ovariectomized adult females, while estradiol and progesterone increase GluR1 levels in the hypothalamus of ovariectomized females. Thus, the differential exposure to androgens, estrogens, and progestins between our males and females in these prepubertal and adult stages of development may have contributed to some of the age- and stress-related differences noted here.

In conclusion, our data indicate greater stress-induced cellular activation in the PVN of adult females compared to prepubertal females, while males show the opposite patterns, with prepubertal males showing greater activation than adult males. Moreover, contrary to our original prediction, we found that prepubertal males and females show similar or reduced expression levels of various ionotropic glutamate receptor subunits in the PVN compared to their adult counterparts, particularly prior to stress. Our current data, therefore, suggest additional mechanism(s) may be involved in mediating changes in hormonal stress reactivity exhibited during pubertal development, and these mechanisms may be further modulated by sex. For instance, prepubertal animals may demonstrate less sensitivity to the inhibitory GABAergic inputs to the PVN, a mechanism known to help terminate the stress-induced HPA response [40]. Regardless, continued study is clearly needed to further our understanding of the mechanisms underlying age- and sex-dependent changes in HPA function and the contributions these changes may have on both emotional and physical health.

All procedures were carried out in accordance with the guidelines established by the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee (IACUC) of Columbia University (D16-00003).

The authors have no conflicts of interest to declare.

This work was supported in part by a Faculty Research Grant from Barnard College (R.D.R.) and NIH RO1 MH115049 (K.G.B.) and MH115914 (K.G.B.).

C.P., J.O., and R.D.R. designed these experiments, and C.P., J.O., S.C., K. G.B., and R.D.R. conducted these experiments and contributed to the writing and editing of the manuscript.

All data generated and analyzed during these studies are included in this article. Further inquiries can be directed to the corresponding author.