Cocaine Self-Administration Influences Central Nervous System Immune Responses in Male HIV-1 Transgenic Rats

Cocaine use increases the neurotoxic severity of human immunodeficiency virus-1 (HIV-1) infection and the development of HIV-associated neurocognitive disorders (HAND). Among the studied cellular mechanisms promoting neurotoxicity in HIV-1 and cocaine use, central nervous system (CNS) immunity, such as neuroimmune signaling and reduced antiviral activity, are risk determinants; however, concrete evidence remains elusive. In the present study, we tested the hypothesis that cocaine self-administration by transgenic HIV-1 (HIV-1Tg) rats promotes CNS inflammation. To test this hypothesis, we measured cytokine, chemokine, and growth factor protein levels in the frontal cortex (fCTX) and caudal striatum (cSTR). Our results demonstrated that cocaine self-administration significantly increased fCTX inflammation in HIV-1Tg rats, but not in the cSTR. Accordingly, we postulate that cocaine synergizes with HIV-1 proteins to increase neuroinflammation in a region-selective manner, including the fCTX. Given the fCTX role in cognition, this interaction may contribute to the hyperimmunity and reduced antiviral activity associated with cocaine-mediated enhancement of HAND.


Introduction
The development of human immunodeficiency virus (HIV)-associated neurocognitive disorders (HAND) remains a clinical burden among HIV-1 infected individuals [1]. Combined antiretroviral therapies (cART) do not completely mitigate the impact of HIV-1 on the central nervous system (CNS) [2]. There is a high prevalence of HIV-1 comorbidities, including drug addiction, that heighten the risk for HIV-1 infections and the acceleration of HIV/AIDS [3][4][5][6][7]. For example, cocaine can target immune cells such as macrophages and lymphocytes to impair the host immune response and enhance HIV viral replication [8][9][10]. Cocaine presumably enhances the development of HAND via increased neuroimmune signaling and CNS inflammation [11]; however, the combined impact of cocaine and HIV-1 on heightened immune responses remains unclear.
Despite the control of HIV-1 replication during cART, chronic inflammation persists [12] and is associated with disease progression and neurocognitive disorders in people infected with HIV [13,14]. Levels of inflammatory cytokines including interleukin (IL)-6, IL-1β, tumor necrosis factor (TNF)-α and interferon (IFN)-γ remain elevated in Rat brain tissues, harvested from male HIV-1 Tg (n = 15) and (WT) F344 (n = 12) rats, were obtained from a specimen repository in the laboratory of Dr. T. Celeste Napier (Rush University Medical Center, Chicago IL). Surgical and self-administration procedures generally followed previously published protocols [51,52]. In brief, rats were purchased from Envigo Laboratories (Indianapolis, IN, USA) and housed in genotype-and treatmentsimilar pairs; food and water were provided ad libitum. Rats were implanted with custommade silastic catheters (0.3 mm ID × 0.64 mm OD; Dow Corning Co., Midland, MI, USA) inserted into the right jugular vein. The distal end of the catheters extended subcutaneously over the mid-scapular region and exited through a metal guide cannula (22 gauge; Plastics One Inc., Roanoke, VA, USA) anchored to a subcutaneously implanted vinyl mesh. The duration of post-surgery recovery lasted at least 7 days, during which catheters were flushed daily with 0.1-0.2 mL sterile saline to maintain patency. Self-administration took place in ventilated, sound-attenuating operant chambers equipped with two 'nose-poke' holes, a stimulus light above each hole, an audio tone generator, and a house light (Med-Associates, St. Albans, VT, USA). Operant sessions were conducted 2 h/day for a total of 14 days on a fixed-ratio 1 (FR1) schedule of reinforcement. A nose-poke in the active hole resulted in a 6 sec infusion of cocaine (1.0 mg/kg/0.1 mL), delivery of an audio tone, and illumination of the stimulus light. Nose-pokes in the inactive hole had no programmed consequence. Control rats were yoked to a cocaine counterpart of the same genotype, receiving a non-contingent infusion of saline (1.0 mL) each time their counterpart selfadministered cocaine. For saline-yoked rats, nose-pokes in either hole were recorded but had no programmed consequence. Brain tissues were harvested one day after concluding the operant task, fast-frozen on dry ice, and stored at −80 • C. Rats were handled in accordance with the procedures established in the Guide for the Care and Use of Laboratory Animals (National Research Council, Washington DC, USA) as approved by the Rush University Institutional Animal Care and Use Committee.

Dataset
A total of 27 rats were used for this study. From each rat, fCTX and caudal STR were isolated. Cytokine expressions were measured from brain regions that included the fCTX and cSTR of 15 HIV-1 Tg rats and 12 or 9 WT rats, respectively. Multiplex immunoassay results that were out of range (not within the linear range of the standard curve) were removed from the dataset prior to analysis.

Quantification of Brain Innate Immune Proteins by BioPlex
Cytokine levels in rat brain protein lysate were measured using the Bio-Plex Pro Rat Cytokine Group I Panel 23-plex assay (BioRad, Cat # 12005641, Philadelphia, PA, USA) following the manufacturer's instructions and as previously described [53]. Briefly, Cells 2022, 11, 2405 4 of 12 rat brain lysates were diluted 1:4 in Bio-Plex sample diluent (containing BSA to a final concentration of 0.5%). Standards were reconstituted and coupled beads were prepared following manufacturer's instructions. Approximately 50 µL of coupled beads was added to each well in a 96-well plate, then washed prior to adding 50 µL of standard and samples (both assayed in duplicate) to the appropriate well in a 96-well plate. The plate was incubated and washed following manufacturers' instructions, then read using a Bio-Plex 200 system. Innate immune protein concentrations were normalized to total protein as measured by Precision Red Advanced Protein Assay (Cytoskeleton, Inc.). Assay sensitivity and limit of detection (pg/mL) for each target: G-CSF (0.

Statistical Analyses
Cumulative cocaine intake was compared between groups using a two-sample t-test. Protein expression levels measured in tissue lysates were not normally distributed; we therefore performed a log 2 transformation for all variables prior to performing t-tests and linear regression analysis. In addition, prior to performing t-tests, the data were stratified by rat model. Two-sample t-tests with unequal variance were used to compare the difference in mean expression levels of each cytokine in HIV-1 Tg rats that self-administered cocaine versus saline-yoked rats or WT rats with cocaine self-administration versus saline-yoked rats. Bonferroni correction was used for multiple testing correction of p-values obtained from the t-tests. The expression levels of each cytokine/chemokine, stratified by rat model (WT vs. HIV-1 Tg ), were described using mean and standard deviation. Additionally, we performed linear regression analysis to evaluate main and interaction effects of treatmentcocaine versus saline (reference) and genotype, HIV-1 Tg versus WT (reference), on cytokine and chemokine expression levels. All tests were conducted in R (version 3.6.1) and an alpha level of <0.05 defined statistical significance and were two-tailed.

Cocaine Self-Administration in WT and HIV-1 Tg Rats
The average cumulative cocaine intake in the WT rats and HIV-1 Tg rats was 112.2 ± 13.6 mg/kg and 121.3 ± 13.6 mg/kg, respectively. There was no difference between genotypes with respect to cocaine intake (p = 0.65), which is consistent with our prior reports on cocaine [51] and methamphetamine [54] self-administration in these rats.

Table 1. Frontal cortex (fCTX) cytokine level summary statistics stratified by rat model WT vs.
HIV-1 Tg . Student's t-tests with unequal variance were used to compare the difference in mean expression levels (and SD) of each cytokine, chemokine, or growth factor in WT (saline-yoked (n = 7) and cocaine SA (n = 8)) and HIV-1 Tg (saline-yoked (n = 6) and cocaine SA (n = 6)) in the fCTX. The 95% CI, overall p value and cumulative replicates are given in the table. Bold cytokines, chemokines, or growth factors indicate targets that were significantly increased in HIV-1 Tg rats compared to WT. Interaction effects were observed. Specifically, we adjusted for HIV-1 protein, and compared HIV-1 Tg saline with HIV-1Tg cocaine, given that we found neither WT cocaine nor HIV-1 Tg saline (after adjustment) to independently result in statistically significant differences in cytokine expression when compared to WT saline. We observed that cocaine synergized with HIV-1 to increase expression (Table 2)  (interaction effect: p = 0.046), MIP-3α (interaction effect: p = 0.021), and TNF-α (interaction effect: p = 0.019). Neither WT cocaine nor HIV-1 Tg saline (after adjustment) independently resulted in statistically significant differences in expression of each of these cytokines when compared to WT saline. Table 2. Linear regression results for frontal cortex (fCTX) cytokine, chemokine, or growth factor levels with interaction effect. Linear regression analysis was performed to evaluate main and interaction effects of treatment-cocaine versus saline (reference) and genotype, HIV−1 Tg versus WT (referent) on cytokine, chemokine, or growth factor levels in the fCTX. Bold type indicates targets with significant interactions.

Discussion
The contribution of inflammation in the CNS likely plays a significant role in the development of HAND, which is exacerbated during drug addiction [3,4]. Findings from clinical studies are not always consistent likely reflecting the heterogeneity of patient populations and differences in use history. To help control these factors, we used an HIV−1 rodent model and implemented operant self−administration protocols to elucidate the impact of cocaine self−administration on HIV−1 related neuroinflammation. The HIV−1 Tg rat is a noninfectious rodent model of HIV infection that allows for chronic lifelong exposure to viral proteins including Tat and gp120 and develops characteristic immune deficiencies [52,[55][56][57]. This may pose some limitations to this model, given that it does not recapitulate cART−induced suppression of viral gene expression as seen with treated HIV−infected patients. We demonstrated that cocaine−self−administration negatively impacted particular factors for HIV−1 related CNS inflammation, within brain regions that are involved in the development of HAND. These outcomes point to mechanisms of hyperinflammation and reduced antiviral activity that may occur during HIV−1 infection and cocaine comorbidity.
In the absence of cocaine, we did not find any differences in inflammatory markers between the WT and HIV−1 Tg groups. This was an unexpected outcome, as lifelong exposure to viral proteins in HIV−1 Tg rats are associated with neurological and behavioral deficits that are characteristic of human HIV−1 infections, and because viral proteins can activate host immune responses [58,59]. Repunte−Canonigo et al. reported increased expression of astrocyte and microglia activation markers, GFAP and Iba1, in the brains of HIV−1 Tg rats compared to WT controls, which indicates activation of neuroinflammatory processes [60]. However, they did not find pro−inflammatory genes to be differentially expressed except for the antiviral gene, interferon stimulated gene (ISG)15 [60]. Gene expression levels of the chemokine monocyte chemoattract protein−1 (Mcp−1/Ccl2) are significantly increased in the hippocampus of HIV−1 Tg rats compared to WT, but no significant differences in gene expression levels of Il−1β, NF−kB or Tnf were detected [61]. Reid et al. showed an age−dependent loss in reactive phenotypes of microglia and astrocytes in the STR and corpus callosum of HIV−1 Tg compared to WT controls [62], but these authors did not measure inflammatory responses. Our findings are also consistent with reports demonstrating no differences in cytokine/chemokine levels measured for STR and hippocampal tissue lysates of 3−month−old and 9−month−old male WT and HIV−1 Tg rats [63]. Thus, changes in cytokine levels in the brains of WT and/or HIV−1 Tg rats following cocaine self−administration strongly implicate neuroimmune activation following cocaine exposure.
In humans, cocaine increases the onset and severity of HAND (reviewed in [64]). This likely reflects the impact of cocaine on CNS inflammation, given that chronic low levels of neuroinflammation exacerbate neurotoxicity [65,66]. Levels of pro−inflammatory cytokines, IL−1β, TNF−α and IL−6, are elevated in humans with cocaine use disorder [26,[67][68][69]. In rats non−contingently exposed to cocaine (i.p.), elevated levels of IL−1β are measured in the pfCTX and nucleus accumbens [27]. In rats that self−administered cocaine, IL−1β and TNFα are increased in the ventral tegmental area (VTA) [70]. Another study demonstrated that inflammatory markers, including IL−1β, IL−1α, IL−2, IL−4, IL−6, IL−10, IL−12, and IFNγ, are enhanced in the pfCTX, STR, and VTA of rats that self−administered cocaine, but only in the presence of traumatic brain injury [71]. Although we did not observe an effect of cocaine on fCTX or STR neuroinflammatory mediators in WT rats, we found that cocaine−self−administration resulted in a hyperimmune response in the fCTX of HIV−1 Tg rats, with higher levels occurring in 13 cytokines, chemokines, and growth factors. Our results indicate an interactive effect of HIV−1 toxic proteins (e.g., Tat, gp120) and cocaine on neuroinflammation. There are limited clinical and preclinical studies demonstrating the impact of cocaine and HIV−1 on the neuroimmune response; however, there is evidence that drug dependency enhances HIV−1 related cognitive impairments [72], which is negatively impacted by inflammation [73].
Taken together, our results demonstrate that in this HIV−1 Tg rat model of cocaine use disorder, HIV−1 proteins or cocaine alone are not enough to initiate a neuroinflammatory response in the fCTX or cSTR. However, in the fCTX, the hyper−neuroimmune response during combined exposure of cocaine and HIV−1 indicates that rather than an additive effect of cocaine on HIV−1 related neuroinflammation, there is a synergistic relationship between cocaine and HIV−1 on the neuroimmune response.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/cells11152405/s1, Figure S1: Cytokines with no significant difference in the fCTX of WT and HIV−1 Tg rats; Table S1. Caudal striatum (cSTR) cytokine level summary statistics stratified by rat model WT vs. HIV−1 Tg .