Human immunodeficiency virus (HIV) remains a major global public health issue, with 39.0 million (33.1-45.7 million) people living with HIV (PLWH) worldwide at the end of 2022, out of which 76% (65-89%) received combination antiretroviral therapy (cART) (1). The overall prevalence rate of HIV-1 infection is higher among adult women than men, with 15% more women living with HIV-1 relative to men in 2019 (2). Importantly, HIV-1 affects women differently than men, with greater immune activation observed in women despite lower overall viral load (3). Given the historical underrepresentation of females across human and animal models of HIV infection, the present study centers specifically on females to better characterize the effects of HIV and novel intervention strategies in this statistically underrepresented population.

While cART successfully reduces viral load in peripheral tissues and significantly improves life expectancy of PLWH (4–6), these treatments are largely incapable of effectively crossing the blood-brain barrier (BBB) to suppress replication of viral proteins, such as transactivator of transcription (Tat), which enters the central nervous system (CNS) within two weeks of infection both directly and through peripherally infected monocytes and macrophages (7–10). Secretion and synthesis of viral protein reservoirs persist in the CNS, where they alter the cellular environment, contributing to chronic neuropathy and HIV-associated neurocognitive disorders (HAND), even in patients actively undergoing treatment (7). Underlying HAND are several dysfunctional immunomodulatory mechanisms within the CNS, including persistent inflammation and immune activation (11–14), which ultimately lead to dendritic injury and damage to neurons (15–17). Notably, cART itself also contributes to both neuroinflammation and altered neuronal connectivity with prolonged exposure (18, 19). In preclinical animal studies, the HIV Tat transgenic mouse model is a well-established model for neuroHIV since their neuropathology and behavioral deficits tend to mirror those observed in cART-treated PLWH with HAND (20–22). These include structural abnormalities in neurons/dendrites, such as reduced spine density and changes in synaptic proteins (21–23), disrupted frontostriatal circuitry (24), and glial abnormalities including microglial activation and micro/astrogliosis (20, 21, 25). Further, these mice and related neuroHIV rodent models also develop changes in learning/memory, motor activity, and motivated behaviors (24, 26–30) relevant for cART-treated PLWH. As dopaminergic neurocircuitry is highly susceptible to disruption by HIV proteins, including HIV Tat (29, 31–34), motivational alterations remain problematic neurobehavioral manifestations in cART-treated PLWH, including apathy (35, 36) which parallels lack of motivation.

The endogenous cannabinoid (endocannabinoid) system is a critical line of defense against neurodegenerative and inflammatory conditions (37). It is well established that endocannabinoids, such as 2-arachidonoylglycerol (2-AG) and N-arachidonoylethanolamine (AEA) upregulate in certain disorders (e.g., Parkinson’s disease, Alzheimer’s disease, multiple sclerosis) and reduce or abolish unwanted effects of these disorders or slow their progression (38, 39). Recent work has demonstrated therapeutic potential of inhibiting monoacylglycerol lipase (MAGL), the primary enzyme that hydrolyzes 2-AG. 2-AG, is an endogenous ligand at neuronal cannabinoid type-1 receptors (CB₁R) and, importantly, also an agonist at cannabinoid type-2 receptors (CB₂R) in immune cells throughout the brain and body (40). Activation of CB₂R has shown protective properties against neuroinflammation and neurodegenerative disorders (41). MAGL-focused strategies are of great interest because targeting enzymatic breakdown is more beneficial relative to phytocannabinoids, as 2-AG tone is affected locally on-demand where its breakdown is dysregulated, thus reducing off-target effects relative to exogenous agonists (42). Because the downstream products of 2-AG hydrolysis, including arachidonic acid (AA), are themselves broken down into proinflammatory eicosanoid lipid mediators such as prostaglandins, hydroxyeicosatetraenoic acids (HETEs), and epoxyeicosatrienoic acids (EETs), inhibiting 2-AG breakdown also stands to reduce proinflammatory processes (43, 44).

Previous studies have demonstrated the neuroprotective properties of MAGL inhibition across numerous models of brain damage and inflammation (45–48), including suppressed astrocytic and microglial activation in models of Alzheimer’s disease (49, 50) and reduced HIV-1 envelope glycoprotein (gp120)-associated synapse loss and IL-1β (51). Further, as 2-AG plays a significant role in modulating mesolimbic dopamine release and associated behaviors (52), inhibiting 2-AG degradation by targeting MAGL has been shown to enhance dopamine signaling, reward-related behavior, and motivation with the MAGL inhibitor JZL184 (53, 54). While JZL184 targets 2-AG more potently and selectively (55) compared to earlier-generation MAGL inhibitors such as URB602 (56), studies have shown desensitization of CB₁R, which can contribute to physical dependency (57), as well as cross-reactivity with other serine hydrolases including α/β hydrolase domain (ABHD) (55). Newer-generation MAGL inhibitors such as MJN110 confer greater 2-AG selectivity with adequate potency to effectively inhibit MAGL in vivo at doses as low as 1 mg/kg (58). MJN110, in particular has shown antinociceptive, anti-inflammatory, and neurorestorative effects previously (59, 60). While MJN110 has also previously been shown to increase reward-directed behavior, this observation was specific to doses of 5 and 10 mg/kg (61). Given these findings, we were interested in characterizing whether Tat or lower-dose MJN110 may affect motivated behavior in the present model.

Thus, the present study used the HIV-1 Tat transgenic mouse model to investigate the chronic effects of MAGL inhibitor MJN110 (1 mg/kg) on reward-related motivated behavior. Immunohistochemical and lipid mediator analyses were conducted to assess potential anti-inflammatory and neuroprotective effects of chronic MJN110 exposure against Tat-induced toxicity on the CNS system, with focusing on the ventral striatum (nucleus accumbens) and infralimbic cortex. It is hypothesized that MJN110 will reduce Tat-driven dysregulation in motivated behavior and exert protective effects against Tat-driven neuroinflammation and neuronal injury.

While no behavioral effects were observed, the present study demonstrated that in the infralimbic cortex and nucleus accumbens, MJN110 successfully reduced Tat-induced dendritic injury and Tat-induced astrocyte-related neuroinflammation, as shown by decreasing Tat-induced upregulated astrocyte density or astrocytic IL-1β colocalization. Notably, Tat induced region-specific inflammatory effects such that astrocyte recruitment was increased in the infralimbic cortex, whereas instead, astrocytic IL-1β was increased in the nucleus accumbens. No effects were noted for Iba-1⁺ microglia density in either brain region, and only minor effects were noted for microglia morphology assessed in the nucleus accumbens. Further, selected HETE-related inflammatory lipid mediators in the striatum were downregulated by chronic MJN110 treatment.

The absence of behavioral differences, while contrary to hypothesized outcomes, may indeed be of some significance insofar as these results demonstrate MJN110 lacks cannabimimetic effects. Tetrahydrocannabinol (THC), a major CB₁R agonist component of cannabis, has been shown to increase dopamine signaling in the nucleus accumbens as well as the ventral tegmental area (75–77), with both regions being critical components of the mesocorticolimbic system involved in reward-related behaviors and salience processing. These THC-mediated dopamine alterations underlie commonly observed increases in food intake, often referred to as the “munchies” (78). While other studies of chronic THC use have contrastingly found decreases in dopamine synthesis and activity (79), it is important to note these results are largely driven by cannabis-dependent individuals with high-severity use (80, 81). The limited existing data available for MJN110 effects on motivated behavior demonstrate that acute treatment increases reward-directed response; however, this effect was only observed at doses five- and tenfold higher than the 1 mg/kg dose used in the present study (61).

The lack of behavioral alterations across groups may also be explained by compensatory mechanisms such as homeostatic scaling (82). Specifically, excitatory synapse loss and increased inhibitory tone have previously been observed, both of which function to protect neurons from increased N-methyl-D-aspartate (NMDA) receptor activity in the presence of Tat-induced excitotoxicity (26, 83). Previous work has also found immune tolerance driven by Tat when assessing microglia activity [65]. As these compensatory changes in excitability and immune response are evident in the context of chronic rather than acute Tat exposure (63, 84), it could be speculated that more robust behavioral alterations may have been observed earlier after Tat induction than the time period presently investigated. Immune tolerance may also contribute to the lack of Tat-induced effects on lipid metabolite expression observed presently; while Tat initially induces inflammation, potentially contributing to dendritic injury seen in vehicle-exposed Tat(+) mice, the immune system becomes desensitized to its effects after about three months of exposure (63). Reduction in lipid metabolite expression following MJN110 treatment may suggest the potential to modulate immune responses in the chronic phase of HIV-1-associated neuroinflammation wherein observable inflammatory insult is diminished, but immune priming remains present (85). While no differences were observed between Tat(+) and Tat(−) subjects, the lipid metabolites assessed may not be as sensitive to Tat exposure in chronic conditions. MJN110 may exert effects through other mechanisms (e.g., modulation of immune cell activity or regulation of downstream signaling pathways associated with neuroinflammation).

Another plausible alternative explanation for behavioral similarity across groups is the growing body of research demonstrating biological sex-specific effects of HIV-1 and constituent viral proteins, including Tat, on both peripheral and CNS outcomes (86). Homeostatic scaling through upregulation of inhibitory activity by γ-aminobutyric acid (GABA)ergic neurons in the hippocampus (26, 83) and the prefrontal cortex have been established in the context of Tat expression, with the latter particularly in females (87). In accordance with these neurophysiological data, studies focused on neurocognitive outcomes of infection have shown worsened behavioral profiles among infected females in domains of spatial memory, learning, and motor skills (88, 89). Data have also shown female-specific hypersensitivity to pain (90), slower recovery from nerve injury, and heightened inflammatory responses to local injury (91). As sex-specific findings regarding executive function have historically been inconsistent (92, 93) and relatively little has been established for motivational deficits, the addition of male subjects in the present work would have provided valuable insight into potential sex differences in these domains. In addition to between-sex variation, there may be within-sex variation due to estrous cycle. Previous studies employing similar models have demonstrated that HIV-1 affects motivation, including reduced response vigor and increased reward threshold (28, 94); however, female subjects were not included in those studies. Prior work has also shown decreased motivation for food rewards (92, 95), even under food-restricted conditions when in estrus, during which time attention is diverted preferentially to sexual motivation (95). It has been posited that this decrease in consumption during estrus occurs due to reductions in positive-feedback signals elicited by natural reinforcers (96), potentially a result of increased striatal dopamine turnover and decreased dopamine concentrations during this phase of the estrous cycle (97). Tracking estrous cycles during assessment or phase-locking assessments to one estrous stage would better elucidate the effects of hormones on motivated behavior and potentially reduce variability.

Though this study did not directly observe the effects of MJN110 against HAND in terms of behavioral outcomes, it provides insights into its potential for neuroprotection and as a modulator of neuroinflammatory responses in the frontostriatal circuitry. The MJN110 observed increase of Tat-induced downregulation of MAP2ab⁺ dendritic intensity and reduction of neuroinflammatory markers demonstrates promise for curbing the underlying inflammatory priming linked with HIV-1, potentially diminishing the emergence or persistence of additional behavioral sequelae associated with HAND. To better clarify the involvement of other cytokines in these effects, further quantification of TNF-α would be advantageous. Because TNF-α inhibits astrocytic glutamate uptake and increases reactive oxygen species (ROS) (98, 99), characterizing HIV-1-associated effects on this measure would enable elucidation of additional mechanisms by which these cells become dysfunctional in infection and may be restored with MJN110. Notably, biomarkers of oxidative stress are elevated in PLWH treated with highly active antiretroviral drugs (100). This increased oxidative stress contributes to neurotoxicity, HAND, and premature aging (100, 101). The observed upregulation of MAP2ab⁺ dendritic signal by MJN110 in Tat(+) mice and MJN110-induced reduction in ROS-derived 11-HETE in the present study suggests a mechanism apart from other enzymatic pathways by which this treatment may be neuroprotective, and collecting data for TNF-α would provide insight into whether the reduction in oxidative stress biomarkers is related to possible additional MJN110-driven immunoreactivity-suppressing effects. Assessing lipid mediator expression in the infralimbic cortex would also provide valuable data to determine whether the effects of Tat and MJN110 may be distinct from those observed in the striatum. As hippocampal lipid mediator expression analyses revealed Tat-induced upregulation of 11-HETE ( Supplementary Figure S1D ), additional region-specific effects may also be observed elsewhere. Given the differences in neuroinflammatory phenotypes between the infralimbic cortex and striatum in astrocyte density and colocalization with IL-1β, it remains possible that diverse cellular mechanisms are recruited to drive distinct inflammatory responses across regions.

In contrast to previous studies of MAGL inhibition, which have shown reductions in prostaglandin expression (102), no such alteration was observed in our analyses (see Figure 6 and Supplementary Figure S1 heatmap panel sections for COX pathway). However, the underlying difference may involve which specific MAGL inhibitor is administered, what dosage is used, or a combination of the two factors. MJN110 is a newer-generation drug relative to those used in prior work, with benefits over previous-generation drugs as described earlier, including fewer off-target effects. Despite the similarity in that prostaglandins and HETEs are both inflammatory metabolites of arachidonic acid (AA), their derived mechanisms differ in their metabolizing enzyme. Prostaglandins are derived from cyclooxygenase (COX), whereas HETEs are largely derived from lipoxygenases (LOX) and cytochrome P450 (CYP) enzymes (103). That MJN110 decreased HETEs without affecting prostaglandins suggests a more COX-independent pathway for MJN110’s profile of anti-inflammatory effects. This finding is important because while COX-targeting therapeutics such as non-steroidal anti-inflammatory drugs (NSAIDs) are effective against inflammation and pain, they can also increase the risk of adverse cardiovascular events (104–106). MJN110 confers anti-inflammatory effects while more specifically targeting pathways beneficial to cardiac, respiratory, and cerebrovascular health, including LOX and CYP (107, 108), with the exception of 5-HETE, which is derived from AA through the LOX system but metabolized through the COX system (109). Given predispositions of PLWH to cardiovascular disease and related complications (110, 111), this therapeutic strategy may improve quality of life across more domains relative to earlier-developed MAGL inhibitors, which interact with COX pathways. Of note, HIV-1 gp120 is more strongly linked to increases in signaling within the COX pathway, particularly through induction of COX-2 expression (112), so using subjects who express more constituent proteins of the HIV-1 viral genome may provide a more complete understanding of virus-associated inflammatory profiles as well as therapeutic potential of MJN110 across signaling pathways.

Within astrocytic mechanisms, assessing other factors, such as neurotrophin synthesis and hypertrophy, could further elucidate additional contrasting effects of Tat and MJN110 on astrogliosis. Astrocyte density was assessed presently as increased astrocyte counts are widely understood to indicate increased proliferation associated with astrogliosis (113); however, a limitation remains in the inability to distinguish between normal and activated astrocytes. There also remains the question of other cellular mechanisms potentially driving independent inflammatory responses. While number of microglia and microglia morphology was not altered by Tat and MJN110, microglia are recognized as a productive viral reservoir that contributes to IL-1β upregulation through stimulation of nucleotide-binding domain leucine-rich repeat-containing proteins (NLRs) (114). The present study could not further probe microglial activation-induced IL-1β upregulation due to technical difficulties with triple-immunolabeling, but these contributions to inflammatory tone remain important to consider.

While the expression and distribution of MAGL were not assessed in the current study, recently published work specifically examined MAGL expression in our transgenic mouse model, and demonstrated Tat induction does not alter MAGL expression (87). Based on these findings, it can reasonably be inferred that MAGL expression was not a confounding factor in the data observed presently. Additionally, a previous paper authored by our lab provides relevant insights into MJN110’s effects in the brain (62). This study, which utilized the same subject preparation methods, including dosage, frequency of MJN110 administration, and duration of Tat induction as the present study, investigated the effect of MJN110 on endocannabinoid levels in various brain regions and showed that treatment with MJN110 increased levels of 2-AG in multiple brain regions. These findings suggest that this MAGL inhibitor effectively modulated endocannabinoid signaling in the brain without baseline MAGL alteration driven by Tat. Nevertheless, to directly conclude that MJN110 effects are due to increases in 2-AG levels, mice should be administered a 2-AG synthesis blocker, such as DO34, similar to what has been published previously (60).

Several limitations of this study should be considered when applying current findings to the therapeutic potential of targeting MAGL in the context of HIV. First, despite the Tat transgenic mouse model being a well-established neuroHIV model, only one of the many viral proteins is expressed. It is known that viral proteins can interact and target various signaling pathways (115); thus they can modify CNS and behavior in different ways compared to a single viral protein. Second, the use of doxycycline for inducing Tat expression is not ideal, as doxycycline has neuroprotective effects on its own (116). Nevertheless, both Tat(–) and Tat(+) mouse groups received doxycycline chow throughout the study to control for this confound and minimize bias. Third, only females were used in the current study without monitoring estrous cycle. Future studies should monitor estrous cycle as hormones can alter motivated behavior associated with a food reward (94), as well as include male mice for comparison, especially due to the known sex-specific differences in immune response to the virus (117). Lastly, ART medication was not considered in the study, as the HIV Tat transgenic mouse model mimics individuals on ART because no virus replication/entry/integration is present, similar to what is seen in PLWH on ART with undetectable viral load. Nevertheless, it is known that cannabinoids and ART drugs are metabolized through the CYP450 system, thus leading to potential drug-drug interaction (118). Interestingly, no significant interactions have been reported between cannabinoids and HIV protease inhibitors in previous studies (119, 120), but additional pharmacokinetic studies remain necessary to increase our understanding of cannabinoid-ART interactions.

A wide body of evidence indicates while HIV-1 treatment has improved life expectancy and quality over the years, supplemental strategies and studies are needed to address persistent issues with patient health, shortcomings of cART, and historically problematic research design. Ultimately, understanding the specific virus-associated vulnerabilities across different subpopulations – and especially the therapeutic potential of novel adaptogenic compounds like MJN110, as demonstrated here – will enhance the foundation of knowledge upon which similar future studies are built and improve management strategies for HIV-1 and other inflammatory diseases.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was approved by University of North Carolina at Chapel Hill Institutional Animal Care and Use Committee. The study was conducted in accordance with the local legislation and institutional requirements.

AFL: Writing – review & editing, Writing – original draft, Visualization, Validation, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. BJY-S: Formal analysis, Data curation, Writing – review & editing. RK: Data curation, Writing – review & editing, Investigation. CAC: Investigation, Formal analysis, Writing – review & editing. IRJ: Methodology, Writing – review & editing. JM: Writing – original draft, Validation, Formal analysis, Data curation, Writing – review & editing, Resources. MJN: Writing – review & editing, Resources. BFC: Writing – review & editing, Resources. AHL: Writing – review & editing, Resources. BMI-J: Writing – review & editing, Resources. SF: Writing – review & editing, Writing – original draft, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization.