(3α,5α)3-Hydroxypregnan-20-one (3α,5α-THP) Regulation of the HPA Axis in the Context of Different Stressors and Sex

Corticotropin-releasing factor (CRF) regulates the stress response in the hypothalamus and modulates neurotransmission across the brain through CRF receptors. Acute stress increases hypothalamic CRF and the GABAergic neurosteroid (3α,5α)3-hydroxypregnan-20-one (3α,5α-THP). We previously showed that 3α,5α-THP regulation of CRF is sex and brain region dependent. In this study, we investigated 3α,5α-THP regulation of stress-induced hypothalamic CRF, CRF receptor type 1 (CRFR1), CRF binding protein (CRFBP), pro-opiomelanocortin (POMC), and glucocorticoid receptor (GR) by western blot and circulating corticosterone (CORT) by enzyme-linked immunosorbent assay (ELISA) in male and female Sprague Dawley rats. Tissue was collected after rats were injected with 3α,5α-THP (15 mg/kg, IP) or vehicle 15 min prior to 30 min of restraint stress (RS), or 10 min of forced swim stress (FSS) and 20 min recovery. The initial exposure to a stress stimulus increased circulating CORT levels in both males and females, but 3α,5α-THP attenuated the CORT response only in females after RS. 3α,5α-THP reduced GR levels in male and females, but differently between stressors. 3α,5α-THP decreased the CRF stress response after FSS in males and females, but after RS, only in female rats. 3α,5α-THP reduced the CRFR1, CRFBP, and POMC increases after RS and FSS in males, but in females only after FSS. Our results showed different stress responses following different types of stressors: 3α,5α-THP regulated the HPA axis at different levels, depending on sex.


Introduction
The activation of the hypothalamic-pituitary-adrenal (HPA) axis in response to stress stimuli is crucial for an organism. The stress response cascade is essential for the survival and dysregulation of HPA axis activity, and has been observed in several neuropsychiatric disorders, including drug addiction [1]. Corticotropin-release factor (CRF) is the main activator of the stress response. Identified by Vale in 1981 [2], CRF is a 41 amino-acid peptide that is well preserved across species. Released from the paraventricular nucleus (PVN) of hypothalamus, CRF reaches the anterior pituitary through the portal vessels and stimulates the transcription of the cDNA for pro-opiomelanacortin (POMC), and the subsequent synthesis and release of the adrenocorticotrophic hormone (ACTH) and β-endorphins [3] to the bloodstream. From the anterior pituitary, ACTH travels in the systemic circulation and reaches the zona fasciculata of the adrenal gland cortex, activating the synthesis and release of corticosteroids, mainly glucocorticoids and mineral-corticoids, such as corticosterone in rats (CORT) or cortisol in humans. Once corticosterone is released into the bloodstream, it is able to regulate its own production through the activation of glucocorticoid (GR) and of 77 male and 80 female Sprague-Dawley rats were used to perform the experiments in this study. Given that the main progesterone peak in the rat estrous cycle occurs in the early evening of proestrus and returns to basal levels by the morning of estrous [22,23], to avoid fluctuations in endogenous 3α,5α-THP due by estrus cycle all experiments were conducted between 8:00-11:30 a.m. All procedures were performed in accordance with guidelines approved by the Institutional Animal Care and Use Committee at the University of North Carolina at Chapel Hill (IACUC approval number: 21-118).

3α,5α-THP Administration
3α,5α-THP (Steraloids Inc., Newport, RT, USA, #P3800) was dissolved in hydroxypropyl-βcyclodextrin (45% w/v in water) at a concentration of 7.5 mg/mL. Solutions were prepared the day before the experiment and kept stirring at 4 • C overnight. Animals were randomly divided in two groups and, according to their weight, all rats received an IP injection of 3α,5α-THP (15 mg/kg) or an equivalent volume of vehicle (VEH). This dose was chosen based on our previous data, showing inhibition of CRF expression in male rats in specific brain regions [20,24]. This dose is known to have an anxiolytic and anti-convulsant effect, [25,26], but no hypnotic effect [27]. 15 min after VEH or 3α,5α-THP administration, rats were submitted to (a) restraint stress or (b) forced swim stress. 45 min after 3α,5α-THP administration, rats were euthanized by decapitation, brains and blood were collected, and brains were immediately frozen at −80 • C until the assay. Data from VEH groups are described as the baseline measure in experiments.

Stress Paradigms 2.3.1. Restraint Stress
Animals were randomly assigned to control (no restraint stress, NRS) and stressed (restraint stress, RS) group, then 15 min after VEH or 3α,5α-THP administration rats were placed in plastic DecapiCone tubes (Braintree Scientific, Inc., Braintree, MA, USA, #DCL120) and subjected to restraint stress for 30 min in another room. The animals were secured inside the cones at the tail. A hole in the head of the cones allowed rats to breathe. Animals were not able to move or turn around. During the restraint, the animals had no access to food and water. Non-restraint rats (control group) remained in the home cages until behavioral experiments started. Rats were euthanized by decapitation immediately after the end of restraint stress, correspondent to 45 min after the VEH or 3α,5α-THP administration.

Forced Swim Stress
The forced swim test was performed in a clear plastic cylinder (45.7 cm H × 20 cm diameter; Stoelting Co., 620 Wheat Ln, Wood Dale, IL, USA, #60160), filled with water (25 ± 1 • C) depth~30 cm, so that rats could not touch the bottom of the cylinder with the tail. Animals were randomly assigned to control (no forced swim stress, NFSS) and stressed (forced swim stress, FSS) group. 15 min after VEH or 3α,5α-THP administration, rats were placed into the cylinder filled with water at the appropriate temperature and allowed to explore the environment. The test lasted for 10 min, then the animals were dried with absorbent towels and returned in the cage for 20 min. Cylinders were cleaned and water was changed after each animal. Rats were euthanized by decapitation 45 min after the VEH or 3α,5α-THP administration.

Immunoblotting
The brains were dissected using a brain block and the hypothalamus was homogenized and sonicated in ice-cold CelLytic MT lysis buffer (Sigma-Aldrich, Saint Louis, MO, USA, #C3228) with 1× HALT protease and phosphatase inhibitor (Thermo Fisher Scientific, 81 Wyman Street Waltham, MA, USA, #1861281). The tissue was left on ice for 30 min and then centrifugated at 14,000× g for 30 min at 4 • C. The supernatant was transferred to new tubes, used immediately or stored at −80 • C. Total protein was determined by the bicinchoninic acid assay (BCA, Thermo Fisher Scientific #23228, #1859078). The proteins (40 µg/lane) were denatured at 95 • C for 5 min in LDS sample buffer (Thermo Fisher Scientific #NP0007) and sample reducing agent (Thermo Fisher Scientific #NP0009) and were resolved by NuPAGE™ 10% Bis-Tris Midi Protein Gel (Thermo Fisher Scientific #WG1202, #WG1203) electrophoresis at 125 V for 10 min and 165 V for the rest of the running and transferred to polyvinylidene fluoride membranes (iBlot2 PVDF regular stacks, Thermo Fisher Scientific #IB24001) using the iBlot 2 Dry Blotting System (Thermo Fisher Scientific).
The blots were blocked with 5% Blotting-Grade Blocker (Bio-Rad Laboratories, 1000 Alfred Nobel Drive, Hercules, CA, USA, #1706404) in PBS-T (0.5% Tween-20) for 2 h (room temperature) and exposed to primary antibody overnight (4 • C), followed by horseradish peroxidase-labeled secondary antibodies for 1 h (room temperature). PBS-T (0.5%) was used to wash the blot 3 times (15 min each, room temperature) after incubation with primary and secondary antibodies. Blots were stripped with a commercial stripping solution (Restore Western Blot stripping buffer, Thermo Fisher Scientific #21059), washed, and blocked again for 1 h, before being re-probed with different primary antibodies (3-5 times).

Antibodies
Antibodies were commercially obtained and used according to the manufacturer's instructions. Primary antibodies are listed in Table S1. Horseradish peroxidase-labeled secondary antibodies were anti-rabbit (Cell Signaling Technology, 3 Trask Ln, Danvers, MA, USA, #7074), anti-mouse (Cell Signaling Technology #7076), and anti-goat IgG (Thermo Fisher Scientific #A24452).

Corticosterone Assay
After decapitation, blood was collected in glass tubes (BD, 1 Becton Drive Franklin Lakes, NJ, USA, BD Vacutainer ® Serum tubes #366430), kept on ice until the end of the experiment and immediately centrifugated at 1750× g for 15 min at 4 • C. Serum was transferred to Eppendorf Flex 2 mL tubes and frozen at −80 • C until the assay. Serum was assayed with ELISA kit for corticosterone (MP Biomedicals, 9 Goddard Irvine, CA, USA, #07DE-9922) following the manufacturer's instructions. Results are expressed as nanograms/milliliter (ng/mL).

Statistical Analysis
All immunoblotting results are presented as % change vs VEH no stress ± SEM. Corticosterone data are presented as mean ± SEM in ng/mL. Statistical analyses were conducted by two-way ANOVA and significant interactions were followed up by Tukey Honest Significant Differences (HSD) test. All data were analyzed with GraphPad Prism 9 (GraphPad Software, 2236 Avenida De La Playa La Jolla, CA, USA). A value of p < 0.05 was considered statistically significant.

3α,5α-THP Partially Prevents the Corticosterone Stress Response after Restraint Stress in Female, but Not in Male Rats
As corticosterone (CORT) is a common parameter to evaluate the stress response, we measured serum levels after restraint stress or forced swim stress, in male and female rats. As expected, circulating CORT levels increased in males and females after restraint stress or forced swim stress ( Figure 1). Restraint stress increased CORT levels by 595% in male and 295% in female rats, respectively (Males restraint stress:  (1,29) = 82.97, p < 0.0001; Interaction between factors F(1,29) = 0.0023, p = 0.962). Consistent with previous literature, 3α,5α-THP had no effect on circulating CORT levels in the absence of either stress exposure. Surprising, 3α,5α-THP failed to attenuate the stress-induced CORT escalation following either restraint stress or forced swim stress in male rats. However, in female rats, 3α,5α-THP administration slightly decreased the CORT stress response following restraint stress, but not following forced swim stress.
(GraphPad Software, 2236 Avenida De La Playa La Jolla, CA, USA ). A value of p < 0.05 was considered statistically significant.

Glucocorticoid Receptor (GR) Regulation by 3α,5α-THP Differs Based on Type of Stressor, but Not Sex
Given 3α,5α-THP did not affect CORT circulating levels, we investigated the possibility that 3α,5α-THP exerted its action at a different level of the stress response in the hypothalamus. Through a negative feedback loop, CORT is able to regulate its own production binding GRs in different brain regions, including the hypothalamus [4]. Our results show that stress and 3α,5α-THP regulation of GR in the hypothalamus differ based on the stress stimuli.
In both male and female animals, restraint stress increased hypothalamic GR expression ( Figure 2). GR expression was increased by 29% in males and 41% in females. 3α,5α-THP reduced hypothalamic GR expression following restraint stress by 37% in males, but not in females.

3α,5α-THP Diminished CRF Stress-Induced Response after Forced Swim Stress, but Only in Female Rats after Restraint Stress
It is well known that stress stimulus increases in hypothalamic CRF levels, leading a cascade of events called stress response. To confirm the activation of the HPA axis, we measured CRF expression in male and female rats, following restraint or forced swim stress.

3α,5α-THP Regulation of CRFR1 Is Both Stressor and Sex-Dependent
Similar to CORT, CRF regulates its own production via binding of its receptors (CRFR1/CRFR2) at the hypothalamic level. CRFR1 mediates the activation of the HPA axis [11,28] and its activation induces anxiety-like behavior [29]. To understand 3α,5α-THP effects on the stress response, we measured CRFR1 expression in the hypothalamus of male and female rats, following restraint or forced swim stress.
CRFR1 response to stress differed between stressors, in both male and female rats. In male animals, restraint stress induced an increase by 60% in hypothalamic CRFR1 expression and 3α,5α-THP prevented CRFR1 rise after stress (

3α,5α-THP Effect on CRF Binding Protein (CRFBP) Differs between Stressors and Sex
CRF effects are mediated and controlled not only by CRF receptors, but also by the CRF-binding protein (CRFBP). CRFBP binds CRF with high affinity and sequesters it away from the receptors, reducing CRF activity [30]. Given CRFBP is an important component of the stress response, we verified the possibility that 3α,5α-THP regulates CRFBP levels in male and female rats, following restraint stress or forced swim stress.

3α,5α-THP Decreased Stress-Induced Hypothalamic POMC Levels in Male, but Not in Female Rats
CRF stimulates the synthesis and release of POMC, precursor of ACTH, from the anterior pituitary. Although POMC exerts its main effect in the pituitary, the POMC gene is found in different brain regions, including hypothalamus [31]. We explored the possibility that 3α,5α-THP regulates POMC levels, in male and female rats, following restraint stress or forced swim stress.

Discussion
We previously discovered that 3α,5α-THP regulation of CRF expression is sex and region-specific at baseline levels [20]. The aim of this work was to investigate the effect of 3α,5α-THP on the HPA axis stress response following two different types of stressors: a psychological stress, such as restraint stress, and a physical stress, such as forced swim stress. Our results show different stress responses following different type of stressors and that 3α,5α-THP regulates the HPA axis at various different levels, depending on sex (see Table 1 for a summary of the results). Abbreviations: VNRS = VEH no restraint stress; VRS = VEH restraint stress; VNFSS = VEH no forced swim stress; VFSS = VEH forced swim stress. Significant effect was found using Two-way ANOVA, followed by Tukey HSD test: **** p < 0.0001, *** p < 0.001, ** p < 0.01, * p < 0.05. Data are express in % vs. correspondent VEH groups.

Circulating Corticosterone
Exposure to acute stress rapidly increases 3α,5α-THP levels in both plasma and brain [14,15]. Acute stress-induced increase of 3α,5α-THP is considered part of the HPA axis negative feedback mechanism, to promote homeostasis by reducing HPA axis activation due to 3α,5α-THP action on GABAergic neurotransmission [32,33]. Additionally, 3α,5α-THP influences the genomic pathway by downregulating gene expression of key-components of HPA axis stress response. In fact, it has been reported that 3α,5α-THP (and other GABAergic neuro-steroids) regulate HPA axis activation by reducing corticosterone levels, CRF mRNA transcript and/or ACTH levels in male rodents [13,[17][18][19].
As expected, our results showed that serum corticosterone levels increased in both male and female rats, following restraint stress or forced swim stress. Surprisingly, 3α,5α-THP failed to prevent the corticosterone stress response, in male animals, regardless of the type of stressor. This paradoxical effect is quite surprising and cannot be explained by sex or type of stressor, since we evaluated the same stressors as previously reported in male rats. In our previous work [20], we showed that 3α,5α-THP levels were increased in both serum and brain regions, including hypothalamus, at the same time point as the present study (45 min after the initial 3α,5α-THP injection). For this reason, we assumed that the dose and timing used in this study were appropriate to detect an effect on the stress response. However, given the different duration of the stress exposure (30 min for the restraint stress vs. 10 min for the forced swim stress) and the recovery time (no recovery time after restraint stress vs. 20 min recovery after forced swim stress), we cannot exclude the possibility that differences between stressors were influenced by those factors. Future studies are needed to clarify this point.
The contrasting findings between our results and previous data that demonstrated the ability of GABAergic neuro-steroids to inhibit the corticosterone response to stress are not unique [34]. For example, other studies showed that administration of 3α,5α-THP can exacerbate the corticosterone response following acute stress in mice [35]. We speculate that the differences in 3α,5α-THP effects on corticosterone levels could be due to differences in the 3α,5α-THP dose administered or the timing chosen to collect the samples (e.g., plasma, brain tissues, etc.). In fact, the protocols used in previous studies employed different stressors, 3α,5α-THP doses, and collection times. For example, Patchev et al., 1994 [19] activated the HPA axis of Wistar rats with I.C.V. administration of CRF 0.5 µg/rat following subcutaneous administration of 1 mg of 3α,5α-THP; Patchev et al., 1996 [18] injected Wistar rats I.P. with a dose of 50 µg/kg of 3α,5α-THP and 30 min later exposed the animals to airpuff stressor; Sarkar et al., 2011 [13] used C57BL/6 or Gabrd−/− mice and direct injected THDOC into the PVN. However, it is also possible that unknown variables contribute to the lack of replication of previous reports.
In contrast, 3α,5α-THP treatment prior to restraint stress partially reduced the stressinduced corticosterone response in the females. Because this sex difference was unique to the restraint stress procedure, it is unlikely to be explained by estrus cycle effects, but rather may involve effects of psychological stress that are unique to females. Upon further evaluation, we found circulating corticosterone levels under control conditions were higher in female than male rats, consistent with previous data that showed female rats in resting conditions displayed higher serum corticosterone concentrations [36,37], depending of the stage of the estrous cycle [38]. We did not monitor the estrous phase in the female rats used for this study, to avoid the increase in corticosterone and 3α,5α-THP levels due to the stress of the monitoring procedure. Indeed, the female animals used were unlikely to have bee sampled during the estrus phase, based on their elevated corticosterone levels, which are known to be equivalent to males during this phase [38]. Nevertheless, we cannot completely exclude a potential role of the estrous cycle in our results.
Consistent with previous data, 3α,5α-THP did not affect serum corticosterone levels in resting conditions. Following restraint stress, serum corticosterone levels increased in both male and female rats; in male rats this increase was greater than in females, probably due to their lower resting corticosterone levels. Similarly, exposure to forced swim stress increased circulating corticosterone levels to a similar magnitude in both male and female groups. Interestingly, the magnitude of the increase in corticosterone following forced swim stress exposure is similar between males and females, but male rats displayed a lower percentage of increase following swim stress than restraint stress. This difference between the two types of stressors has also been observed in mice [39], confirming a different activation of the HPA axis following different stress exposures. Furthermore, 3α,5α-THP failed to attenuate the corticosterone stress response following forced swim stress in both males and females.

Hypothalamic CRF
Despite the lack of effect on corticosterone levels, the final product of the stress response, our data showed that 3α,5α-THP regulated other important markers of the stress response in the hypothalamus in a sex-and stress stimulus-dependent manner. Restraint stress affected several hypothalamic components of the stress response in a different manner than forced swim stress. For example, exposure to restraint stress activated the HPA axis by increasing the expression of GRs, CRF, and POMC in male animals. In contrast, forced swim stress did not induce changes in GRs and CRFR1 in male or female rats, or alter CRFBP or POMC levels in female rats.
In our previous study [20] we showed a sexual dimorphism in 3α,5α-THP regulation of hypothalamic CRF signals under resting conditions. Consistent with previous data [19], we observed that 3α,5α-THP had no effect in female rats but significantly decreased CRF mRNA (but not peptide expression) in male rats. These new data support our previous observation, confirming a different 3α,5α-THP regulation of HPA axis depend upon sex. As expected, the exposure to both restraint stress and forced swim stress raised hypothalamic CRF content. Despite the similar stress response, 3α,5α-THP reduced the stress-induced increase in hypothalamic CRF following swim stress in both male and female animals. Moreover, 3α,5α-THP decreased the CRF stress response following restraint stress in female rats. However, this effect was not observed after restraint stress in male rats, where 3α,5α-THP failed to attenuate the CRF stress increase, but instead enhanced CRF. These results suggest differences in the regulation of HPA axis by 3α,5α-THP depending on the type of stressor, but this effect is also sex-specific.

CRF Binding Protein (CRFBP)
CRF activity is correlated with the concentration of the binding protein, CRFBP, the carrier that transports CRF through the vessels and sequesters it away from the receptors [6]. Our data showed that both sex and the type of stress influenced hypothalamic CRFBP expression. While restraint stress reduced CRFBP, forced swim stress increased the levels of this protein in male rats. These diverse effects suggest a stress-stimulus-dependent HPA axis activation; 3α,5α-THP did not alter the reduced CRFBP levels in male rats following restraint stress. This result could be linked to the higher CRF levels in males following restraint stress since the stress-induced decrease in CRFBP levels could lead to an increase of CRF availability and activity. In contrast, following forced swim stress, 3α,5α-THP attenuated the CRFBP stress-induced response, in male rats. These results suggest, once again, that 3α,5α-THP action is dependent on the type of stress.
Confirming the sex differences in the activation and regulation of HPA axis, in female rats we did not detect any change in CRFBP levels following restraint stress or forced swim stress. Further studies are necessary to clarify the role of CRFBP in the female stress response.

CRF Receptor 1 (CRFR1)
Another important component of the stress response is represented by the CRF receptors. In this study we analyzed only the CRFR1 type, that exhibit ubiquitous distribution in the brain. CRFR1 has been well studied in neuropsychiatric disorders such depression, anxiety, and addiction in the attempt to reduce the overexpression of CRF and the hyper-activity of HPA axis observed in these disorders [40]. Preclinical studies showed positive results using a CRFR1 antagonist [41,42], but unfortunately in human studies this approach did not appear to have efficacy as a monotherapy for neuropsychiatric disorders [43]. Our data showed that hypothalamic CRFR1 expression is influenced by the type of stress in male animals. While restraint stress induced an increase in CRFR1 levels, forced swim stress did not modify CRF receptors. However, 3α,5α-THP decreased CRFR1 expression following both restraint stress and forced swim stress in male rats. It is possible that the lack of 3α,5α-THP effect on CRF levels observed in male rats is compensated by the decrease in stress-induced CRFR1 following restraint stress. This reduction in CRF receptors could be a 3α,5α-THP-mediated indirect mechanism to reduce CRF signaling; although CRF levels are still high, the decrease in the abundance of CRFR1 leads to a decrease in CRF signaling.
It is very interesting that neither restraint stress nor forced swim stress altered CRFR1 expression in female animals, and 3α,5α-THP did not affect CRFR1 levels following restraint stress. It appears that the absence of 3α,5α-THP effect following restraint stress is related to the direct effect of 3α,5α-THP on CRF expression in the absence of stress in female rats. In contrast, 3α,5α-THP reduced CRFR1 concentration in the absence of and following forced swim stress. This discrepancy is clearly attributable to the type of stress.

Glucocorticoid Receptors (GRs)
It has been demonstrated that GRs have a dual mechanism of action in the regulation of the stress response: a rapid regulation lead by membrane receptors and a genomic mechanism lead by nuclear receptors [44]. In this work, we measured the total fraction of hypothalamic GRs, so we were not able to discriminate between membrane and nuclear receptors. However, our data support the theory that the increase in hypothalamic GRmediated signaling is an essential mechanism for the activation of the stress response. The 3α,5α-THP treatment prevented the stress-induced increase in GR expression following restraint stress in male rats. Reducing the binding substrate of corticosterone, it is likely that 3α,5α-THP is indirectly reducing corticosterone effects. Interestingly, the decrease induced by 3α,5α-THP in males was higher than females following restraint stress; this effect could be related to the higher corticosterone response observed in the males vs. the females. While restraint stress induced an increase in GRs levels in males, forced swim stress did not affect hypothalamic GR levels in both male and female animals. The 3α,5α-THP-induced downregulation in GRs observed in male rats, following restraint stress, may represent an indirect mechanism to reduce corticosterone action through its receptors. Given that corticosterone binds both GRs and MRs, further studies need to determine the effect of 3α,5α-THP on MRs.
The changes observed in GR expression (and CRFR1) after 3α,5α-THP administration, suggest that 3α,5α-THP may have a genomic mechanism that influences receptor expression, independent from the well-known 3α,5α-THP GABAergic action. Even though there is no evidence that 3α,5α-THP directly binds or activates GRs [45], in vitro studies showed that 3α,5α-THP and other pregnenolone derivates inhibited, in a dose-dependent manner, GR-mediated gene transcription [46]. 3α,5α-THP might act at downstream levels, interacting with various protein kinases involved in the regulation of GR function, in a similar way as antidepressant drugs [47,48]. It has also been demonstrated in rats that the ability of 3α,5α-THP to prevent the effect of restraint stress was reduced by blocking GRs with RU486 (a progesterone/glucocorticoid receptor antagonist) [49]. Moreover, it has been shown that 3α,5α-THP, after conversion to 3α,5α-DHP, can activate progesterone receptor-mediated gene transcription [50]. Additionally, 3α,5α-THP is able to activate another nuclear receptor known as pregnane xenobiotic receptor (PXR) [51]. PXR regulates cholesterol metabolism through the cholesterol-binding translocator protein (TSPO), essential for the transport of cholesterol into the mitochondria, where steroidogenesis starts, suggesting that 3α,5α-THP can also regulate its own production [52,53]. Previous studies demonstrated that the activation of PXR increased corticosterone levels with no effect on ACTH secretion [54]. Furthermore, a reciprocal regulation between nuclear receptors (such as nurr1 and nurr77) and CRF signaling was found [55]. Finally, 3α,5α-THP may act on other nuclear receptors, regulating the CRFR1 or GR-mediated gene transcription involved in the stress response. Further studies are needed to address these possibilities.

Pro-Opiomelanocortin (POMC)
Increasing CRF following a stress stimulus leads to the synthesis and release of ACTH from the pituitary. POMC is the precursor of ACTH, an essential hormone in the cascade of stress response. Our data showed a strong sex difference in the regulation of this precursor, depending also on the type of stressor. In male rats, both restraint and forced swim stress induced an increase in hypothalamic POMC levels, and 3α,5α-THP attenuated the stress-induced response in male rats, following both type of stressors. Restraint stress also increased POMC levels in female rats, but 3α,5α-THP failed to reduce this stress-dependent effect. In contrast, forced swim stress did not change hypothalamic POMC concentration in females. However, in female animals, 3α,5α-THP reduced POMC levels in absence of stress.
Despite the lack of 3α,5α-THP effect on corticosterone levels, our data demonstrated the ability of this neuro-steroid to modify other important components of the HPA axis following a stress exposure. These 3α,5α-THP effects are dependent on sex and the type of stressor the animals faced. Corticosterone and other glucocorticoids are well-known since the 1970s for their immunosuppressive and anti-inflammatory proprieties. However, in the past decades, several studies proved that glucocorticoids have also a pro-inflammatory impact on immune system [56,57]. Previous data from our lab [24,58] showed that 3α,5α-THP inhibits toll-like receptor pathways, reducing the production of pro-inflammatory chemokines and cytokines. Stress can activate the inflammatory response in the brain and peripherally [59]. Acute stress activates microglia and increases pro-inflammatory cytokine production [60], as observed in neuropsychiatric disorders such as depression [61], anxiety [62], schizophrenia [63], and autism spectrum disorders [64]. Finally, 3α,5α-THP might be controlling HPA axis stress-induced response through the immune system, reducing inflammation caused by stress. This idea will also require further study.

Conclusions
3α,5α-THP plays an essential regulatory role on the neuroendocrine modifications induced by the stress response. Our data show various different levels of HPA axis regulation by 3α,5α-THP, depending on the type of stressor and sex. These results provide a more complete understanding of HPA axis biomarkers involved in stress responses and the role that 3α,5α-THP plays in stress regulation. Furthermore, these data show the importance of sex as biological variable in the regulation of the HPA axis, given the differences in steroidogenesis and treatment responses previously observed [65]. Moreover, different types of stressors produce different activation of various HPA axis components, and, accordingly, different 3α,5α-THP effects. Further studies are required to determine the role of 3α,5α-THP in the regulation of aberrant HPA axis activity. The recent evidence in clinical studies, using the 3α,5α-THP formulation, brexanolone, and its FDA approval for the treatment of post-partum depression [66], underscore the therapeutic importance of neuro-steroid signaling in the brain [67]. Stimulation of 3α,5α-THP biosynthesis through development of selective neuro-steroidogenic drugs or precursor administration may also be effective for the treatment of stress-related disorders.