HIV-1 Tat Protein Promotes Neuroendocrine Dysfunction Concurrent with the Potentiation of Oxycodone’s Psychomotor Effects in Female Mice

Human immunodeficiency virus (HIV) is associated with neuroendocrine dysfunction which may contribute to co-morbid stress-sensitive disorders. The hypothalamic-pituitary-adrenal (HPA) or -gonadal (HPG) axes are perturbed in up to 50% of HIV patients. The mechanisms are not known, but we have found the HIV-1 trans-activator of transcription (Tat) protein to recapitulate the clinical phenotype in male mice. We hypothesized that HPA and/or HPG dysregulation contributes to Tat-mediated interactions with oxycodone, an opioid often prescribed to HIV patients, in females. Female mice that conditionally-expressed the Tat1–86 protein [Tat(+) mice] or their counterparts that did not [Tat(−) control mice] were exposed to forced swim stress (or not) and behaviorally-assessed for motor and anxiety-like behavior. Some mice had glucocorticoid receptors (GR) or corticotropin-releasing factor receptors (CRF-R) pharmacologically inhibited. Some mice were ovariectomized (OVX). As seen previously in males, Tat elevated basal corticosterone levels and potentiated oxycodone’s psychomotor activity in females. Unlike males, females did not demonstrate adrenal insufficiency and oxycodone potentiation was not regulated by GRs or CRF-Rs. Rather OVX attenuated Tat/oxycodone interactions. Either Tat or oxycodone increased anxiety-like behavior and their combination increased hypothalamic allopregnanolone. OVX increased basal hypothalamic allopregnanolone and obviated Tat or oxycodone-mediated fluctuations. Together, these data provide further evidence for Tat-mediated dysregulation of the HPA axis and reveal the importance of HPG axis regulation in females. HPA/HPG disruption may contribute vulnerability to affective and substance use disorders.


Introduction
Human immunodeficiency virus type 1 (HIV-1) infection remains a significant public health concern with~1.1 million people living with HIV in the U.S. The widespread use of combined antiretroviral therapeutics (cART) has largely increased life expectancy among HIV-infected patients; however, the inability of cART to target latent central nervous system (CNS) reservoirs including microglia and astrocytes [1], likely contributes to the persistence of central viremia and neurological symptomatology [2,3]. Approximately 50% of the HIV + population contend with a constellation of neurological symptoms (collectively referred to as "neuroHIV") that include affective disorders (i.e., increase generalized anxiety and major depression), psychomotor deficits and cognitive impairments [4][5][6]. CNS complications also include neuropathic pain resulting in 8-52% of HIV-1 patients being prescribed opioids [7][8][9][10][11][12] which may interact with HIV-1 proteins, complicating cycle (lights off at 09:00 h) with ad libitum access to food and water. Briefly, mice expressed the Tat 1-86 protein via conditional, GFAP-relegated expression under regulation of a reverse tetracycline-controlled transactivating (rtTA) transcription factor. We and others have previously confirmed mRNA expression of tat in the brain and spinal cord of these mice [44][45][46] and include confirmation in the present tissues via quantitative real-time PCR (see Supplemental Data, Figure S1). Tat(−) control mice lacked the tat transgene but did express the rtTA transcription factor [47]. While rtTA gene expression is notably 'leaky' [48], Tat(−) controls lack the tat transgene making them an optimal comparison group. Moreover, we have previously demonstrated that young-adult Tat(+) mice do not demonstrate tat-mediated behavioral interactions with drugs of abuse until doxycycline-induction [23] and include a saline-administered, un-induced control group herein (see Supplemental Data, Figure S2).

Surgical Manipulation
A subset of mice (n = 32) underwent bilateral ovariectomy under isoflurane anesthesia (2.5-4.0%) per previous methods [49]. Post-surgery, mice were transferred to a clean cage with food, water, and access to acetaminophen (2 mg/mL) for a period of 96 h. Weight, neurological status, surgical site, and food consumption were monitored daily [50]. Two mice failed to recover and were thus excluded from the study. Mice were allowed 7 days of recovery to allow for endogenous hormone washout prior to additional pharmacological manipulations.

Determination of Estrous Cycle Phase
Estrous cycle was determined via vaginal lavage conducted daily at 09:00 h. Cytology was assessed to determine the estrous cycle phase as modified from previous descriptions [43,55]. Mice were tested when in the proestrous or diestrous phase of their estrous cycle in order to control for hormonal variations that might influence behavioral responses. In the current study, no significant differences were observed between mice on diestrus I (a.k.a. metestrus) or diestrus II; as such, these groups were combined.

Behavioral Assays 2.5.1. Forced Swim Stress
To activate the HPA stress axis, mice were exposed to a forced swim stimulus as previously described [42]. Briefly, mice were placed in a container filled with room temperature water (approximately 22 • C) and permitted to swim for 15 min. At the end of the swim stressor, mice were dried and returned to their home cages. Following swim stress mice were assessed in a behavioral battery comprised of an open field and a light-dark transition test (described below).
hicle injection (to account for potential injection stress). All mice received either acute oxycodone or vehicle saline 15 min prior to assessment in the open field and light-dark transition tests. Notably, mice that received vehicle, antalarmin, or RU-486 were administered a final dose 30 min prior to behavioral testing. To assess the restoration of the HPA axis response, all mice were sacrificed approximately 2 h after the initiation of behavior testing (a time when circulating stress steroids are resolving) as opposed to prior observations that were conducted at 30 min post-testing (a time of peak stress steroid response [42]). had transgene expression induced via doxycycline (30 mg/kg, i.p., once daily for 5 days with 2 days for washout; n = 8-10/group). Some mice were pretreated with either vehicle. antalarmin, RU-486 or were ovariectomized (OVX). On the day of testing, proestrous, diestrous, or OVX mice were exposed to a forced swim stress (or not) and administered saline or oxycodone 15 min prior to assessment in an open field and light-dark transition test. (B) Distance (m) travelled in an open field among (B) non-stressed mice, (C) stressed mice and (D) mice that were pretreated with antalarmin or RU-486 or were OVX. † indicates a main effect for oxycodone-administered mice to differ from saline-administered controls. § indicates an interaction wherein oxycodone-administered Tat(+) mice differ from respective Tat(−) controls and saline-administered controls. # indicates a main effect of estrous cycle wherein diestrous mice differ from proestrous mice. ‡ indicates an Tat(−) controls] had transgene expression induced via doxycycline (30 mg/kg, i.p., once daily for 5 days with 2 days for washout; n = 8-10/group). Some mice were pretreated with either vehicle. antalarmin, RU-486 or were ovariectomized (OVX). On the day of testing, proestrous, diestrous, or OVX mice were exposed to a forced swim stress (or not) and administered saline or oxycodone 15 min prior to assessment in an open field and light-dark transition test. (B) Distance (m) travelled in an open field among (B) non-stressed mice, (C) stressed mice and (D) mice that were pretreated with antalarmin or RU-486 or were OVX. † indicates a main effect for oxycodone-administered mice to differ from saline-administered controls. § indicates an interaction wherein oxycodone-administered Tat(+) mice differ from respective Tat(−) controls and saline-administered controls. # indicates a main effect of estrous cycle wherein diestrous mice differ from proestrous mice. ‡ indicates an interaction wherein oxycodone-administered Tat(+) OVX mice differ from all other oxycodone-administered Tat(+) groups, p < 0.05.

Open Field Test
The open field test was used to assess spontaneous motor behavior as previously described [42,43]. Briefly, mice were placed gently in a corner of a square-shaped open field box (40 × 40 × 35 cm 3 ; Stoelting Co., Wood Dale, IL, USA) with a brightly-lit center (inner 20 × 20 cm 2 ) and allowed to behave for 5 min. Dependent measures included overall distance travelled (in meters) and mean velocity (m/s) as indices of spontaneous motor behavior.

Light-Dark Transition Test
Following testing in the open field, mice were assessed for anxiety-like behavior in the light-dark transition test. Given the psychostimulatory effects of opioids, the light-dark test was preferred given that anxiety-like indices are less confounded by manipulations that affect motor behavior [56]. The light-dark apparatus consisted of two compartments, one brightly-lit (20 × 20 × 35 cm 3 ) and the other dark and enclosed (20 × 20 × 35 cm 3 ; Stoelting Co., Wood Dale, IL, USA). Briefly, mice were placed in a corner of the brightly-lit side of the apparatus and permitted to behave for 5 min. The latency to enter the dark chamber and the total time spent in the light zone were considered indices of anti-anxiety-like behavior.
2.6. Biochemical Assays 2.6.1. Tissue Collection Following behavior testing, mice underwent cervical dislocation followed by rapid decapitation. Whole brains and trunk blood were collected. Brains were immediately flashfrozen on dry ice. Blood was collected in a chilled aliquot tube and serum was separated via Viruses 2021, 13, 813 5 of 20 centrifugation at 13,500× g. Tissues and serum were stored at −80 • C. Hypothalamus was later grossly dissected on ice as previously described [43] and frozen at −80 • C until assay.

Enzyme-Linked Immunosorbant Assay (ELISA)
Serum steroid extraction was achieved via ether-snap freezing. Briefly, 1 mL of anhydrous ethyl ether (ice-cold) was added to serum samples which were vortexed and snap frozen using a dry ice/acetone mixture as described previously [43]. Ether supernatant was collected and evaporated to dryness overnight in a chemical fume hood. Crystalline steroids were reconstituted 5× (for estradiol), 25× (for progesterone) or 50× (for corticosterone) their initial volume in assay extraction buffer (Neogen Life Sciences, Lexington, KY, USA).

Ultra-Performance Liquid Chromatography (UPLC)-Mass Spectrometry (MS)
For UPLC-MS/MS, charcoal-stripped tissue (brain tissue derived from Tat-tg mice) was utilized to prepare both the calibration curve and quality control samples for analysis. A simple protein precipitation method was used for steroid extraction. Samples were homogenized (100 µL of PBS pH 7.4) and precipitated with 100 µL of acetonitrile followed by vortexing (2 min) and centrifugation (10 min at 14,000 rpm). After centrifugation, the supernatant solution was mixed with 50 µL of derivatizing solution (20 mg/mL of 2-hydrazinopyridine solution prepared in 0.5% trifluoroacetic acid ethanol solution) and incubated at 60 • C for 1 h. Following incubation, 20 µL of the internal standard solution (1 µg/mL) was added and vortex mixed. For sample analysis, aliquots of 2 µL were injected onto the UPLC-MS/MS instrument.

Procedure
All manipulations (apart from genotype and cycle phase) were randomly assigned. HIV-1 Tat was conditionally induced in female adult transgenic mice via doxycycline administration for 5 days (i.p.). Given that the anti-inflammatory effects of doxycycline could potentially mask some Tat-mediated effects, two days of doxycycline washout were included prior to testing [57]. Estrous cycle was determined daily, and mice were assessed when in the proestrous or diestrous phases. Given that effects of Tat are found to be stable for at least 3 weeks [45,58], mice were behaviorally assessed no more than 14 days from Tat induction ( Figure 1A,B). Some mice underwent forced swim stress for 15 min (stressed paradigm) or not (non-stressed paradigm). Some mice were pretreated with vehicle, the CRF-R antagonist, antalarmin, or the GR antagonist, RU-486. Some mice were ovariectomized to remove the primary source of gonadal hormones and administered a daily vehicle injection (to account for potential injection stress). All mice received either acute oxycodone or vehicle saline 15 min prior to assessment in the open field and lightdark transition tests. Notably, mice that received vehicle, antalarmin, or RU-486 were administered a final dose 30 min prior to behavioral testing. To assess the restoration of the HPA axis response, all mice were sacrificed approximately 2 h after the initiation of behavior testing (a time when circulating stress steroids are resolving) as opposed to prior observations that were conducted at 30 min post-testing (a time of peak stress steroid response [42]).

Statistical Analyses
Biochemical endpoints (i.e., circulating and central steroid measures) were initially utilized as covariates in multivariate analyses of variance (MANOVA) but were not found to explain significant variance in primary dependent measures. Thus, behavioral and biochemical dependent measures were assessed via separate two to three-way analyses of variance (ANOVA) with estrous cycle phase (proestrous or diestrous), pretreatment (vehicle, oxycodone, antalarmin, RU-486, or OVX), and/or genotype [Tat(−) or Tat(+)] as between-subject factors. Group differences following main effects were determined via Fisher's Protected Least Significant Difference post hoc tests. Significant interactions were delineated via simple main effects and main effect contrasts with α controlled for family-wise error. All analyses were considered significant when p < 0.05.

HIV-1 Tat and Oxycodone-Mediated Psychostimulation Is Moderated by Stress and Estrous Cycle
In order to assess combined HIV-1 Tat interactions with oxycodone, Tat was induced via systemic administration of doxycycline for 5 days (with two days of washout; Figure 1A). Estrous cycle was assessed daily over the next 12 days and mice were behaviorally tested when in the proestrus or diestrus phase of their estrous cycle (whichever came first). A 15 min forced swim was used to activate the HPA axis in the stressed paradigm (or not in the non-stressed paradigm). On the day of testing, all mice received saline or oxycodone (3 mg/kg, i.p.) 15 min prior to behavioral assessment ( Figure 1A).

Gonadal Steroids Are Necessary for Tat to Potentiate Oxycodone-Mediated Psychostimulation
In order to determine the importance of the HPA and HPG axes in Tat-potentiated psychomotor and/or anxiety-like behavior, HPA axis receptor sites were pharmacologically blocked and circulating gonadal hormones were surgically attenuated. To achieve this, some mice were administered vehicle, antalarmin (CRF-R blocker), or RU-486 (GR blocker) concurrent with Tat induction for seven days. Some mice were OVX to remove the primary source of circulating gonadal hormones. Gonadally-intact mice were tested when in proestrus. All mice received an acute injection of saline or oxycodone (3 mg/kg) 15 min prior to behavior testing ( Figure 1A).
In summary, OVX attenuated Tat's capacity to potentiate oxycodone-mediated psychostimulation. Neither CRF nor GR blockade influenced these effects, further supporting the influence of HPG factors in female mice.
Among swim stress-exposed mice, circulating corticosterone was significantly greater in oxycodone-administered mice [F(1,70) = 6.30, p < 0.05] (see †, Figure 2C  In summary, OVX prompted greater circulating corticosterone among Tat(+) mice compared to Tat(−) controls, the endocrine phenotype associated with an attenuated psychomotor response to oxycodone. While CRF-R or GR blockade altered total glucocorticoid levels, no differences between Tat(−) and Tat(+) mice were observed. These data support the notion that reinstatement of HPA responsivity may involve HPG manipulation and may rescue the Tat-mediated psychomotor response to oxycodone.
In summary, OVX prompted greater circulating corticosterone among Tat(+) mice compared to Tat(−) controls, the endocrine phenotype associated with an attenuated psychomotor response to oxycodone. While CRF-R or GR blockade altered total glucocorticoid levels, no differences between Tat(−) and Tat(+) mice were observed. These data support the notion that reinstatement of HPA responsivity may involve HPG manipulation and may rescue the Tat-mediated psychomotor response to oxycodone.
In summary, hypothalamic allopregnanolone was greater among diestrous, compared to proestrous, mice and OVX apparently increased basal allopregnanolone content and obviated oxycodone-or Tat-mediated fluctuations.  Figure 1 had allopregnanolone content (ng/g) assessed in the hypothalamus in (A) non-stressed, (B) stressed, and (C) OVX mice (n = 8/group). # indicates a main effect for diestrous mice to differ from proestrous mice. § indicates an interaction wherein the denoted group differs from Tat(−) or Tat(+) proestrous mice.ˆindicates an interaction wherein the denoted group differs from oxycodoneadministered, Tat(+) diestrous mice, p < 0.05.
In summary, hypothalamic allopregnanolone was greater among diestrous, compared to proestrous, mice and OVX apparently increased basal allopregnanolone content and obviated oxycodone-or Tat-mediated fluctuations.

Discussion
The hypotheses that the HPA and/or HPG axes contribute to Tat-and oxycodonemediated psychomotor and affective behavior were upheld; albeit the HPA phenotype produced by Tat in females and the mechanisms contributing to these effects differed from those previously observed in males. Exposure to Tat activated the HPA stress axis, recapitulating the clinical phenotype of elevated basal cortisol seen in up to 46% of HIV + patients [26,28,[59][60][61]. Moreover, either Tat expression and/or oxycodone exposure, increased anxiety-like behavior in the light-dark transition test. These data recapitulate our prior findings in male Tat-tg mice [42]; however, important sex differences were also revealed. Females exposed to Tat did not demonstrate adrenal insufficiency, nor did pharmacological blockade of the HPA feedback loop attenuate Tat's capacity to potentiate oxycodone's psychomotor effects. One potential mechanism underlying these sex differences may involve the contribution of HPA-mediating gonadal steroid hormones. Pregnane steroids fluctuate to greater levels in females, compared to males and may help offset HPA insults. In support, we observed a full attenuation of Tat's capacity to potentiate oxycodone's motor effects when mice were OVX, which was accompanied by a compensatory increase in adrenal glucocorticoid. Moreover, neurosteroids such as allopregnanolone are well-characterized to play a compensatory role to curtail stress-induced HPA-activation [62][63][64]. We assessed the hypothalamic levels of allopregnanolone and observed that OVX or combined Tat and oxycodone exposure promoted an increase in endogenous hypothalamic allopregnanolone. This may serve as a central adaptive response to stress similar to our prior findings in males exposed to Tat and morphine [23]. Together, these data suggest that Tat can dysregulate glucocorticoid neuroendocrine function and that endogenous neurosteroids like allopregnanolone may be upregulated in response. These data extend findings to implicate dysregulated HPA/HPG axes in the susceptibility to neuropsychiatric complications, opioid effects, and potential abuse liability.
Sex/gender differences in HIV are understudied, contributing to some discrepancies within the extant literature. While some studies find HIV + women to have an improved immunological response to cART [65,66] and slower viral progression [67] compared to HIV + men, studies assessing neuroHIV outcomes reveal women to be more vulnerable to HIV-associated neurocognitive disorders [68][69][70]. However, neuroHIV vulnerability may differ based on the dimensions assessed. While women are more prone to affective disorders in the general population, large clinical studies find HIV + women to have a lower prevalence of major depression or any anxiety disorder [71,72] compared to HIV + men. Unfortunately, these studies are often not stratified by gender and many are not controlled by route of HIV acquisition or endocrine factors, adding heterogeneity to findings. Some studies find no gender differences in HIV-related affect [73,74]. The potential mechanisms that underlie these discrepancies in cognitive vs. affective gender differences, may involve vulnerability to stress. Domains that involve HPA axis mediation, such as generalized anxiety and depressive symptomatology, may favor women [72,75,76]. HPA axis function provides a biological basis for sex differences in stress-related psychiatric disorders [77] and may interact with HPG factors. Gonadal hormones in premenopausal women contribute to the central responsivity of the stress response [78]. In animal models, females also have a particularly robust neuroendocrine response to stress [79][80][81][82][83] and demonstrate increased hypothalamic release of CRF and downstream circulating glucocorticoids [79,83,84]. Dysregulation of HPG and HPA axes may promote neurological behavioral deficits [37,85,86]. In support, we found stressed mice to be more vulnerable to Tat/oxycodone behavioral interactions when in the diestrous phase of their cycle (higher estradiol:progesterone ratio). These data are consistent with higher estradiol levels promoting psychostimulant response to drugs of abuse in rodent models [87][88][89][90].
Psychomotor locomotion is a robust assay to quantify the response of opioids in rodents [91] and may inform abuse liability. Opioid prescriptions for HIV + patients are rising [7][8][9] and opioids exert long-term capacity to promote neuroendocrine dysfunction in people [92,93]. Given the potential vulnerability that neuroendocrine dysregulation confers to substance use and affective disorders [94][95][96][97], it is important to understand the effects that HIV and opioids exert on the HPA and HPG axes. We and others find that HIV-1 Tat facilitates the rewarding and sensitizing effects of opioids [42,43,98,99]. Indeed, HPA dysregulation confers vulnerability to substance use disorders [100]. Together, these data support the notion that maintenance of neuroendocrine function may help counteract HIV-related opioid abuse liability.
There are several mechanisms that may underlie endocrine abnormalities observed in HIV patients. Hypercortisolemia is reported in a considerable proportion of HIV + patients [61,101,102]. Tat may be a contributor given that the current animal model recapitulates this phenotype. There may be several mechanisms for the hypercortisolemia observed. Tat has been proposed to promote GR transcription partly via its capacity to drive positive transcription elongation factor-b accumulation on GR-responsive promot-ers [103]. As well, Tat (alone or in conjunction with other proinflammatory HIV proteins) may promote GR activation indirectly via cytokine activation. Elevated levels of IL-2 and IL-4 decrease the affinity of the GR for cortisol and lead to a glucocorticoid resistant state [104]. As well, the β isoform of the GR, which inhibits the active α isoform of the receptor, may be increased by proinflammatory cytokines, thereby reducing GR signaling and promoting an increase in adrenal glucocorticoid [105][106][107]. Importantly, affective neu-roHIV symptoms such as depression are associated with impaired GR signaling in HIV + women [108]. Another aspect of HPA dysregulation observed in HIV patients is adrenal insufficiency (observed in up to 46% patients [24][25][26][27][28]61]); however, we did not observe this in Tat-exposed females. Adrenal insufficiency based on sex as biological variable has not been well characterized in the HIV clinical population due to its multifactorial nature [24,61,[109][110][111]. There may be several explanations for the lack of adrenal insufficiency in females including lower CRF-R internalization, thereby increasing sensitivity to CRF [112]. Other reasons may include lower GR receptor density and GR translocation at the level of the hypothalamus in females, thereby reducing negative feedback [113,114]. Additionally, male mice demonstrate a robust increase in hypothalamic, GR-mediated negative feedback compared to females [114], which may increase their vulnerability to HPA insults. Females may also have higher corticosterone binding globulin, reducing bioavailable corticosterone and potentially the reserve for negative HPA feedback [115]. Lastly, both circulating and central pregnane steroids fluctuate to greater levels in females compared to males and may confer protection.
The HPA axis is modulated by non-traditional-acting steroid hormones that are produced de novo in the brain (i.e., neurosteroids [62]). Neurosteroids, such as allopregnanolone, are rapidly produced in response to stress and reinstate HPA-axis homeostasis, partly via potent positive allosteric modulation of inhibitory GABA A receptors [116][117][118][119]. The paraventricular nucleus of the hypothalamus (PVN) is rich in GABAergic neurons [120] and GABA A receptor expression [121]. We have previously observed Tat to promote the selective loss of GABAergic interneurons within the hippocampus, particularly those that are nNOS + , SST + or PV + [122]. Increased AlloP content in hypothalamus may compensate for inhibitory deficits promoted by the loss of GABAergic interneurons. AlloP's capacity to prevent upregulation of CRF mRNA in the PVN following adrenalectomy, attenuates CRF release in response to stress and inhibits the firing of PVN CRF neurons [121,[123][124][125]. Moreover, downstream ACTH release and subsequent corticosterone secretion in rodents are attenuated by neurosteroids [62,123,124,[126][127][128]. These effects are dynamic; while neurosteroids dampen the HPA response, a dysregulated HPA axis may alter neurosteroidogenesis [129][130][131][132]. We have found that HIV-1 Tat expression in mice increases neurosteroidogenesis of pregnenolone and 5α-reduced steroid metabolites, including allopregnanolone [23]. These demonstrations of upregulated allopregnanolone content in the brain are similar to those observed following central trauma and may indicate a central adaptive response to stress. In support, CNS allopregnanolone is increased following traumatic brain injury, spinal cord injury, ischemic stroke and in neurodegenerative disease models where it exerts a neuroprotective functions (reviewed in [133]). Herein, we find that OVX mice have an increase in basal, hypothalamic AlloP content concurrent with a reduction of oxycodone's psychomotor effects, supporting its role to maintain homeostasis. Beyond its capacity to modulate GABA A receptors [134], allopregnanolone is an inhibitor of L-type calcium channels [135,136] which may also attenuate Tat-mediated excitotoxicity [23,137]. Greater lability and fluctuations of pregnane steroids, such as allopregnanolone, in females may confer resilience to HPA insults such as those produced by Tat exposure.
It must be noted that Tat is not the only HIV-1 protein that may be expressed in the CNS. Other proteins including gp120 and viral protein R (VPR) may act alone or in concert with Tat to influence the HPA axis. In support, HIV-1 gp120 is observed to increase pituitary ACTH as well as circulating corticosterone [138]. Moreover, exogenous gp120 stimulates hypothalamic CRH and arginine vasopressin secretion in rats [139]. HIV-1 VPR promotes glucocorticoid hypersensitivity and was found to bind GRs directly and potentiate their signaling, in part by acting as a GR co-factor [140][141][142]. Effects of VPR to promote GR transcription may coincide with Tat's potential capacity to co-activate GRs [61]. Future studies should assess the endocrine interactions between these proteins.

Conclusions
Together, these data demonstrate the capacity for HIV-1 Tat to dysregulate the HPA stress axis and promote vulnerability to the psychostimulant effects of oxycodone. The HPG axis likely plays an important regulatory role in these effects. Tat and oxycodone exposure altered steroid hormone production in circulation and neurosteroid formation in the hypothalamus. These effects occurred concurrent with increased psychomotor and anxiety-like behavior. Thus, maintenance of the neuroendocrine axes may exert beneficial effects on combined HIV-1 and oxycodone mediated pathology.
Supplementary Materials: The following are available online at https://www.mdpi.com/xxx/s1, Figure S1: Fold changes of tat mRNA expression using Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR), Figure S2: Motor behavioral measures acquired in open field, Table S1: Primers used for qRT-PCR, Table S2: Summary of findings for primary dependent measures.