Hypogonadism induced by surgical stress and brain trauma is reversed by human chorionic gonadotropin in male rats: A potential therapy for surgical and TBI‐induced hypogonadism?

Abstract Introduction Hypogonadotropic hypogonadism (HH) is an almost universal, yet underappreciated, endocrinological complication of traumatic brain injury (TBI). The goal of this study was to determine whether the developmental hormone human chorionic gonadotropin (hCG) treatment could reverse HH induced by a TBI. Methods Plasma samples were collected at post‐surgery/post‐injury (PSD/PID) days ‐10, 1, 11, 19 and 29 from male Sprague‐Dawley rats (5‐ to 6‐month‐old) that had undergone a Sham surgery (craniectomy alone) or CCI injury (craniectomy + bilateral moderate‐to‐severe CCI injury) and treatment with saline or hCG (400 IU/kg; i.m.) every other day. Results Both Sham and CCI injury significantly decreased circulating testosterone (T), 11‐deoxycorticosterone (11‐DOC) and corticosterone concentrations to a similar extent (79.1% vs. 80.0%; 46.6% vs. 48.4%; 56.2% vs. 32.5%; respectively) by PSD/PID 1. hCG treatment returned circulating T to baseline concentrations by PSD/PID 1 (8.9 ± 1.5 ng/ml and 8.3 ± 1.9 ng/ml; respectively) and was maintained through PSD/PID 29. hCG treatment significantly, but transiently, increased circulating progesterone (P4) ~3‐fold (30.2 ± 10.5 ng/ml and 24.2 ± 5.8 ng/ml) above that of baseline concentrations on PSD 1 and PID 1, respectively. hCG treatment did not reverse hypoadrenalism following either procedure. Conclusions Together, these data indicate that (1) craniectomy is sufficient to induce persistent hypogonadism and hypoadrenalism, (2) hCG can reverse hypogonadism induced by a craniectomy or craniectomy +CCI injury, suggesting that (3) craniectomy and CCI injury induce a persistent hypogonadism by decreasing hypothalamic and/or pituitary function rather than testicular function in male rats. The potential role of hCG as a cheap, safe and readily available treatment for reversing surgery or TBI‐induced hypogonadism is discussed.


| INTRODUC TI ON
Traumatic brain injury (TBI) is a major public health problem 1 due to the relatively high incidence rate (106 per 100,000 globally), 2 and the lack of effective treatments. The incidence of TBI in males is 3 times that of females, but normalizes to 1:1 by age 65. 3 The consequences of a TBI can include functional (eg decreased cognitive performance), psychopathological (eg post-traumatic stress disorder), neuroanatomical (eg cystic infarcts, neurodegeneration) and biochemical (eg inflammation) changes.
An underappreciated endocrinological complication of TBI is hypogonadotropic hypogonadism (HH). TBI can markedly suppress pituitary gonadotropin secretion and gonadal sex steroid production. [4][5][6][7][8][9][10][11][12][13][14][15][16] Such hypothalamic-pituitary-gonadal (HPG) axis hormones have welldescribed roles in the formation and maintenance of brain structure and cognitive function (reviewed in Hara et al. 17 ). Our early studies utilizing human embryonic stem cells (hESC) as a model of early embryogenesis 18 identified human chorionic gonadotropin (hCG), a trophoblastic gonadotropin hormone, as modulating the expression and processing of the amyloidβ precursor protein (AβPP 19,20 ). Although this protein has well-defined neurogenic properties, [21][22][23][24][25][26] our data suggested AβPP also had important developmental functions during early human embryogenesis (prior to the formation of neural precursor cells). We subsequently determined that hCG signals via the luteinizing hormone/chorionic gonadotropin receptor (LHCGR) on hESC to promote the growth and development of the pre-implantation embryo, including the formation of the 3 germ layers in the morula and its development into a blastocyst. 27,28 hCG mediated these effects via the upregulation of steroidogenesis (P 4 synthesis); P 4 signalling was found to be obligatory for both embryoid body (aka morula) and neuroectodermal rosette (aka primitive neural tube) formation.
In addition to hCG's role in early embryonic and brain development, 27 hCG and its adult homolog luteinizing hormone (LH) promote neuronal proliferation, 29 while hCG and sex steroids regulate adult neuronal differentiation (ie neuritogenesis, spine density, synaptogenesis). 30,31 Conversely, age-related or age-induced reproductive endocrine dyscrasia has a negative impact on cognitive function, 32 an effect that can be slowed or halted by the partial rebalancing of the HPG axis with sex steroid supplementation 33,34 or gonadotropin-releasing hormone (GnRH) superagonist and antagonist treatment. 35 HH could therefore greatly compromise cognitive performance and neuroregeneration following a TBI. Importantly, hCG has been shown to improve functional recovery following spinal cord injury in rats, 30,36,37 and decrease ischaemic brain injury in adult and neonatal rodent models of stroke. [38][39][40] Aside from these neuroprotectant properties, hCG also promotes neurite sprouting and neuron survival. 40 TBI-induced HH is thought to result from damage to the hypothalamus or pituitary 41,42 and/or stress-induced cortisol-mediated suppression of the HPG axis. 43,44 While prevalence rates for HH vary widely, likely due to the severity of the injury, location and type of injury, time of screening and design of the study, there is increasing consensus that even mild TBIs can induce HH and that severe TBIs induce persistent HH. 6,14,[45][46][47] This silent condition goes mostly undiagnosed and therefore untreated. Based on the pleiotropic properties of hCG, in this study we tested whether hCG treatment could reverse HH induced by a commonly used TBI model (controlled cortical impact; CCI). We find that hCG treatment reverses HH induced by either a craniectomy alone, or a craniectomy and penetrating CCI injury, in adult male rats.

| Subjects
Male Sprague-Dawley rats (n = 58, 5 to 6 months old) were acquired from Harlan Laboratories Inc. (Madison, WI) and acclimated to the environment over 2 days. Rats were then weighed and handled for no less Hospital approved the procedures used in this study, and the research was conducted in an AAALAC-approved facility. Experimenters were blinded as to the identity of the animals throughout injections, blood collections, and body weight and hormone data analyses.

| Surgeries
All surgical procedures were carried out under isoflurane gas anaesthesia (5% for induction; 1.5%-3.0% for maintenance, craniectomy ~15-to 30-min duration; craniectomy +CCI injury ~25-to 45-min duration). An anaesthesia chamber was used for induction, and a nose cone was used for maintenance.

| Controlled cortical impact and sham surgeries
Anaesthetized rats were mounted in a Kopf stereotaxic device (Model 900), where the animal's head was held in place by nontraumatic ear bars and a bite bar. Anaesthesia was maintained by nose cone while the head was shaved and sterilized with 70% ethanol and Betadine™ (Purdue Products L.P.) antiseptic solution.
Throughout surgery, anaesthesia levels were monitored closely and K E Y W O R D S human chorionic gonadotropin, hypoadrenalism, hypogonadism, RU-486, testosterone, traumatic brain injury were frequently adjusted as needed, based on heart rate, respiration rate and oxygen saturation. A homeothermic blanket control unit (Harvard Apparatus, Holliston, MA) was used to monitor body temperature and to prevent hypothermia throughout surgery.
Under aseptic conditions, the cranium and its bony landmarks including bregma (β) and lambda (λ) were exposed by making a midline incision along the scalp into the skin and fascia covering the skull. A 6-mm-diameter craniectomy was centred on the midline at 2.5 mm anterior to β. The cortical impact was made at 2.5 mm anterior to β over the midline of the medial frontal cortex with an Impact One™ Stereotaxic CCI instrument (Leica), using a 5 mm impactor (bit size), travelling at 2.25 m/s (velocity), extending 3 mm below the cortical surface (impact depth) for 100 ms (dwell time). Sham-injured groups received the same surgical procedures up to and including craniectomy but no CCI injury. After surgery, the rats were placed on a heating pad, monitored closely and upon awakening were tested 30 min later for righting reflex to assess any immediate effects of craniectomy or CCI injury on righting ability, and then returned to their home cages.

| Experimental design
Once out of quarantine all rats were weighed and handled for one week. The final body weights at the end of this week were (1) ranked from highest to lowest (as a function of age) and then (2) used to assign each rat to a surgery/treatment group in a counterbalanced manner (ie using the ABBA method).

| Experiment 2
Rats were assigned to the following groups: Sham + saline + RU-486

| Blood collection and hormone analyses
Rats were anaesthetized (between 9:00 a.m.-12:00 noon) and their tails placed in a 200-ml beaker filled with warm water (≤44°C) for 5 min. The tail was cleaned with 70% alcohol, the minimal amount of the tail tip snipped with a blade, and/or the wound reopened by removal of the scab for subsequent bleeds, and ~1 ml of whole blood was collected directly into EDTA tubes at baseline (post-injury day (PID)−10) and at PID, 1, 11, 19 and 29. Blood collected did not exceed 1% of body weight every 2-week period. Animals were injected with Lactate Ringers solution (5 ml) for fluid resuscitation. At the terminal bleed (day 29), blood also was collected via heart puncture. Collected

| Statistical analysis
A mixed factorial analysis of variance (ANOVA) for repeated measures was performed on the weight, behavioural, hormonal and gross lesion data (GraphPad Prism, v.7; GraphPad Software, Inc.). Post hoc analyses were performed using the Tukey multiple comparison test.

| hCG reverses craniectomy and CCI-induced hypogonadism and attenuates hypoadrenalism
hCG treatment of animals that underwent a craniectomy (Sham surgery) or craniectomy plus CCI injury (CCI group) significantly increased circulating concentrations of T and P 4 back to baseline concentrations by PSD/PID 1 ( Figure 1A Figure 1C). hCG treatment had no effect on increasing corticosterone concentrations in Sham animals at any time point, but did increase circulating corticosterone in the CCI animals on PID 1 and 11 ( Figure 1D). Together, these results demonstrate that hCG can reverse hypogonadism induced by a craniectomy or a craniectomy +CCI injury, but has lesser effect on reversing hypoadrenalism.

| Relationships between circulating steroid concentrations before and after sham surgery, CCI injury and hCG treatment
Correlation analyses demonstrated strong positive correlations in baseline plasma samples between P 4 with androstenedione (r = .84, p < .01), androstenedione with its metabolite T (r = .82, p < .01) and with corticosterone and its precursor 11-DOC (r = .89, p < .001; Table 1). Sham injury obviated the significant correlations between sex steroids, but not corticosterone and its precursor 11-DOC (r = 0.98, p < .001, Table 2). hCG treatment of Sham animals was sufficient in restoring the strong positive relationship between T and androstenedione (r = .76, p < .05), but not between P 4 with androstenedione. hCG treatment induced two additional positive correlations between androstenedione with 11-DOC (r = .84, p < .01) and F I G U R E 2 RU-486 treatment attenuates hCG-induced reversal of circulating testosterone plasma concentrations. Plasma concentrations (mean ± SEM) of T (A), P 4 (B), 11-DOC (C) and corticosterone (D) in ng/ml on PID −10, 1, 11, 19 and 29 for the following groups: RU-486: Sham +vehicle (n = 5), RU-486: Sham +hCG (n = 5), RU-486: CCI +saline (n = 5) and RU-486: CCI +hCG (n = 5). Data were analysed using 2-way repeated-measures ANOVA; post hoc analyses were performed using the Tukey multiple comparison test (p < .05; letters indicate differences between treatment groups and pre-and post-injury days). Differences between RU-486-induced changes in plasma hormones between treatment groups in Figures 1 and 2  androstenedione with corticosterone (r = .72, p < .05; Table 2). These results suggest that Sham surgery alone is sufficient to disrupt the relationship between sex steroid metabolism, a relationship that is partially reversed with hCG treatment.
CCI injury, like sham injury, resulted in positive relationships between P 4 with corticosterone (r = .58, p < .01) and 11-DOC (r = .68, p < .001), between corticosterone and its precursor 11-DOC (r = .91, p < .001; Table 3), as well as the loss of significant correlations between sex steroids. Unlike sham injury, CCI injury resulted in a strong correlation between androstenedione with its metabolite T

| DISCUSS ION
We demonstrate for the first time that intraperitoneal injection of hCG is effective in reversing hypogonadism ( Figure 1A,B) and attenuating hypoadrenalism ( Figure 1C,D) following a craniectomy or craniectomy +CCI injury in young adult male rats. Both craniectomy and craniectomy +CCI injury promoted corticosteroid production in favour of sex steroid (P 4 ) production, a relationship that was reversed with hCG treatment (Figure 1; Tables 1-3). hCG's ability to increase sex steroid plasma concentrations following a craniectomy and following a moderate-to-severe brain injury supports its potential as a treatment for TBI-induced hypogonadism.
It is important to note that while hCG has potential to reverse hypogonadism and promote neurogenesis and cognitive recovery, an increase in circulating hCG/LH concentrations as a result of ovariectomy 53-58 or treatment 59,60 has been shown to impair cognition in rodents, while lowering LH or blocking LHCGR signalling is protective of memory in rodents 53,54,[56][57][58][60][61][62][63] and humans. 35 Conversely, one study has demonstrated that intracerebroventricular hCG delivery after OVX rescued dendritic spine density and spatial memory. 64 The general negative impact of LH/hCG on cognitive performance appears to be dependent upon the ratio of gonadotropins to sex steroids since situations where gonadotropins and sex steroids are in balance such as during the adult reproductive period 65 are periods of normal cognitive performance and do not involve dyotic signalling. [32][33][34] This is illustrated by the findings that interventions that reverse dyotic signalling such as sex steroid supplementation of ovariectomized animals (see references above), GnRH agonist suppression of gonadotropins in post-menopausal women 35 and caloric restriction (eg Ref. 66), either reverse or halt cognitive decline. Therefore, one might predict that functioning gonads are essential for hCG to promote cognitive recovery from a TBI, as we have found in intact male rats 67 (unpublished data). Thus, hCG treatment might be expected to be most beneficial in pre-menopausal and pre-andropausal individuals, while those further along the postreproduction spectrum might benefit most from a combination therapy of hCG supplemented with appropriate sex steroids.

| Causes of craniectomy and CCI injury induced hypogonadism and hypoadrenalism
The induction of hypogonadism and hypoadrenalism in young male rats following a craniectomy and a craniectomy +CCI injury (reduction in plasma concentrations of P 4 , T, 11-DOC and corticosterone;

| hCG treatment for reversing hypopituitarism
Our results in craniectomized and craniectomized plus CCI-injured rats demonstrate that post-surgery and post-injury male rats retain the capacity to synthesize and secrete T (Figure 1). The reversal of hypothalamic/pituitary function in animals induced by a TBI, craniectomy and/or isoflurane anaesthesia indicates the utility of hCG for reversing hypogonadism and hypoadrenalism in these conditions. hCG treatment comes with the advantage of not only increasing neurotropic hCG/LH, but also increasing the dozens of gonadal sex steroid and protein hormones that regulate normal brain structure and function. 32 hCG has been shown to increase T production in aged male rats. 97,98 hCG is a safe, cheap, FDA-approved treatment for hypogonadism in men (chronically), infertility in men and women, and to promote the descent of testicles in young boys with cryptorchidism. 99 In this context, hCG treatment has recently been shown to be effective in (1) raising plasma T concentrations in healthy men with chronic spinal cord injury, and this was not significantly different from hCG's elevation of plasma T in able-bodied male control subjects, 100 (2) protecting the rodent adult 37,38 and neonatal brain from hypoxicischaemic cellular degeneration in vivo and inhibiting glutamatedependent excitotoxic or necrotic neuronal cell death in vitro 40 ; and (3) increasing ERK phosphorylation, neurite outgrowth and rescuing ovariectomy-induced spatial memory deficits in C57Bl/6J mice. 64 In addition, hCG also partially attenuated hypoadrenalism in male rats.
Although there are few studies that have assessed the impact of hCG on regulating adrenal steroid production, hCG has been demonstrated to increase follicular fluid concentrations of 11-DOC, but not corticosterone, 101 while LHβ overexpressing female mice have enlarged adrenals, increased LHCGR expression and a 14-fold elevation in serum corticosterone. 102 In this latter study, the authors proposed that enhanced ovarian oestrogen synthesis causes increased secretion of prolactin, which elevates LHCGR expression in the mouse adrenal cortex, leading to elevated, LH-dependent, corticosterone production. 102 Continuous exposure to hCG is, however, known to suppress the expression of LHCGR via the down-regulation of mRNA (eg Ref. 103,104). To circumvent the down-regulation of the receptor, in our study hCG was administered in the form of Pregnyl every other day, as is used clinically. 105 Since initial phase half-life of urinary-derived Pregnyl is between 5.6 and 11 h (https://www. merck.ca/), the 48 h between doses appears sufficient to maintain LHCGR expression, as circulating concentrations of sex steroids ( Figure 1A,B) were sustained over the 29-day experiment.

| RU-486 impact on plasma steroid concentrations
Elevations in circulating corticosterone observed in our study following RU-486 treatment are consistent with elevations in corticosterone in the male rats 106 and cortisol, corticotropin or adrenocorticotropic hormone that is observed in human men, 107,108 women, 108,109 and non-human primates (Macaca fascicularis 110 ).
Blocking glucocorticoid and P 4 signalling using RU-486 had little effect on sex steroid changes induced by craniectomy (uninjured) or CCI injury, but significantly diminished the decline in 11-DOC and corticosterone concentrations in craniectomy (uninjured) but not CCI-injured animals (Figure 2A-D), indicating that blocking glucocorticoid (and perhaps P 4 ) receptor signalling partially prevents the suppression of 11-DOC and corticosterone (either by limiting stressinduced suppression of 11-DOC/corticosterone or by elevating their synthesis).

| LIMITATI ON S OF THE S TUDY
The suppression of circulating sex steroids by the Sham surgery procedures was unexpected, and the "Sham CCI" group cannot be considered a normal "Control" group in the usual sense of the term.
Nonetheless, the fact that stress-induced HH can be reversed with hCG administration is important from the perspective of a potential treatment option for veterans returning from combat with stress-or TBI-induced HH. 6,14,[45][46][47]  Research, UW-Madison, NIH UL1T000427, provided support for the steroid assays.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

AUTH O R S ' CO NTR I B UTI O N S
RIG, KH, RR, MW, QB, ADJ, IMA, GF and CSA performed surgeries, controlled cortical impact injuries and collected blood. AM and TEZ performed blood hormone analyses. RIG, SVM and CSA performed data and statistical analyses. CSA, SVM and RIG conceived the study.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.