HIV and Cocaine exposure promote Tau phosphorylation through RSK-1 in a GSK3β-independent manner

10 HIV and cocaine are known to disrupt neuronal signaling and contribute to neurocognitive dysfunction, yet 11 the underlying molecular mechanisms are not clear . In this study, we delineate the underlying molecular 12 mechanism by which HIV and/or cocaine enhance Tau phosphorylation (p -Tau S396), a marker of Tau -13 mediated neuropathies. Furthermore, we elucidate how these two independent neuropathogenic factors , 14 cocaine and HIV , exploit distinct yet convergent signaling pathways to drive this pathological event. We 15 demonstrate that HIV robustly activates and upregulates RSK1, which functions upstream of AKT and 16 promotes Tau phosphorylation through an AKT -independent mechanism while simultaneously inactivating 17 GSK3β via serine -9 phosphorylation (p -GSK3β S9). However, cocaine not only activates RSK1 but also 18 strongly stimulates AKT1, resulting in sustained GSK3 β inhibition and persistent Tau phosphorylation. 19 Notably, Tau phosphorylation persists even under conditions of GSK3β inactivation in both HIV and cocaine 20 exposure, revealing a previously unrecognized GSK3 β-independent mechanism of Tau modification. 21 Collectively, these findings identify RSK1 as the primary mediator of Tau phosphorylation upon HIV and/or 22 cocaine exposure, and uncover a novel RSK1 -driven, GSK3 β-independent pathway contributing to 23 Tauopathy. Through a combination of immunofluorescence, immunoblotting, genetic knockout, and 24 overexpression approaches, we establish RSK1 as a central signaling hub linking the AKT-GSK3β pathway 25 to Tau phosphorylation. We demonstrate that RSK1 operates as a critical upstream regulator of AKT and 26 GSK3β signaling, playing dual roles, both activating AKT and suppressing GSK3β, thereby uncovering a novel 27 layer of pathways that regulates Tau phosphorylation. The reproducibility of these main signaling pathways 28 across SH-SY5Y neurons, mixed cell 3D spheroids, and human brain organoids underscores the robustness 29 and biological relevance of this mechanism. Collectively, these findings reveal mechanistic convergence of 30 HIV and cocaine on RSK1-dependent signaling and provide critical insight into how diverse neuropathic / 31 neuropathological factors remodel neuronal signaling to drive Tau-associated dysfunction. These findings 32 provide novel mechanistic insight into the molecular underpinnings of neuro -HIV and substance abuse 33 associated Tauopathy. By identifying RSK1 as a master regulator and demonstrating that Tau phosphorylation 34 can bypass GSK3β inhibition, our study advances understanding of signaling complexity and highlights new 35 opportunities for therapeutic intervention. Targeting RSK1 may represent a promising strategy to mitigate Tau 36 pathology, induced due to insoluble aggregates of phosphorylated Tau, a common factor promoting cognitive 37 decline not only in individuals with Alzheimer’s disease but also in those exposed to cocaine or/and infected 38 with HIV. 39 Significances 40 This study demonstrates that exposure to HIV and/or cocaine induces Tau phosphorylation at serine 396 41 (S396), a well -established marker of Tau pathology, and delineates how these two independent 42 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint neuropathogenic factors engage distinct yet convergent signaling pathways to drive this pathogenic event. 43 We show that HIV exposure drives robust RSK1 activation, positioning it upstream of AKT to promote Tau 44 phosphorylation via an AKT-independent mechanism, while concurrently suppressing GSK3β activity through 45 serine-9 phosphorylation. In contrast, cocaine, while only moderately activating RSK1, primarily enhances 46 AKT signaling, leading to sustained GSK3 β inhibition and increased Tau phosphorylation. Notably, Tau 47 phosphorylation persists even under conditions of GSK3β inactivation in both settings, revealing a previously 48 unrecognized, RSK1 -centered, GSK3 β-independent pathway of Tau modification. Overall, our findings 49 demonstrate that Tau phosphorylation in the context of HIV infection and cocaine exposure is a complex, 50 multi-layered regulatory process involving multiple signaling nodes. Importantly, we identify RSK1 as a central 51 integrative hub linking viral and substance -induced signaling to downstream Tau pathology. This work 52 advances our understanding of the molecular mechanisms underlying neuroHIV and substance abuse –53 associated neurodegeneration. Furthermore, it highlights RSK1 as a novel and promising therapeutic target 54 for mitigating Tauopathy in both cocaine-using and non-using people with HIV (PWH). 55 Highlighted points 56 • RSK1 acts as a central regulator of Tau phosphorylation, capable of driving this process through a GSK3β-57 independent mechanism. 58 • HIV promotes Tau phosphorylation primarily via robust upregulation and activation of RSK1, operating 59 largely independent of AKT1, while concurrently inducing GSK3β inactivation. 60 • Drugs of abuse, such as cocaine induces Tau phosphorylation through dual activation of AKT1 and RSK1, 61 alongside sustained inactivation of GSK3β. 62 • Tau phosphorylation persists despite GSK3β inhibition, revealing a complex AKT1 -RSK1 signaling axis 63 and underscoring the dominant role of GSK3β-independent mechanisms in Tau pathology following HIV 64 and cocaine exposure. 65 • HIV and cocaine engage distinct yet convergent signaling pathways that disrupt neuronal homeostasis 66 and drive tauopathy, providing mechanistic insight into neuroHIV and substance abuse -associated 67 neurodegeneration. 68 • RSK1 functions as a key upstream modulator of AKT and GSK3β pathways, positively regulating AKT 69 signaling while negatively regulating GSK3β activity. 70 • RSK1 emerges as a potential therapeutic target, offering new opportunities for intervention in HIV -71 associated neurocognitive disorders (HAND) and drug-induced neurodegeneration. 72 • Established and characterized H80 cells as a novel neuronal cell model and demonstrated their suitability 73 for studying neuron-specific signaling pathways, including Tau phosphorylation. 74 • The conserved and widespread nature of the signaling cascade driving Tau phosphorylation in response 75 to HIV and/or cocaine exposure was validated across multiple model systems, including both 2D neuronal 76 cell cultures and 3D systems such as human brain organoids and spheroids. 77 78 Strength of the Study 79 This original study provides novel mechanistic insight into how HIV and cocaine, two independent 80 neuropathological factors, converge and diverge on intracellular signaling pathways to regulate Tau 81 phosphorylation. By integrating immunofluorescence, immunoblotting, genetic knockout, and overexpression 82 approaches, we identif ied RSK1 as a master regulator of Tau phosphorylation. Importantly, we discovered 83 that HIV robustly upregulates and activates RSK1 to promote Tau phosphorylation through an AKT -84 independent route while simultaneously inactivating GSK3β. On the other hand, cocaine exerts a moderate 85 effect on RSK1 but strongly stimulates AKT to induce GSK3β inactivation and drive Tau phosphorylation. A 86 key strength of this work is the discovery that Tau phosphorylation persists despite GSK3 β inactivation, 87 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint revealing a complex, GSK3β-independent mechanism, involving RSK1 in Tau pathology. Moreover, our study, 88 for the first time, identify RSK1 as an upstream regulator of AKT-GSK3β signaling cascade, enhancing AKT 89 signaling while simultaneously inhibiting GSK3β activity, thereby underscoring the critical role of RSK1 in Tau 90 phosphorylation and associated illnesses, such as HAND and Alzheimer’s disease. Together, these findings 91 not only advance our understanding of the molecular underpinnings of neuroHIV and substance abuse 92 associated tauopathy but also highlight RSK1 as a promising therapeutic target for not only HIV and cocaine 93 induced neurotoxicity but also other neurodegenerative diseases, such as Alzheimer’s disease. Another key 94 strength of this study is the establishment and characterization of H80 cells as a novel neuronal model, 95 demonstrating their suitability for investigating neuron -specific signaling pathways, including Tau 96 phosphorylation. The combination of comparative signaling analysis, genetic perturbations, and integrative 97 mechanistic modeling makes this study both conceptually and technically novel, besides broadly relevant to 98 the fields of neurovirology, addiction neuroscience, neurodegeneration, and cognitive impairments. 99

100 Human immunodeficiency virus (HIV) infection remains a significant global health concern, with an estimated 101 38 million people currently living with the virus worldwide [1]. Although the introduction of combination 102 antiretroviral therapy (ART) has markedly improved life expectancy and viral suppression in people with HIV 103 (PWH), the burden of HIV -associated neurocognitive disorders (HAND) persists [2, 3] . HAND affects 104 approximately 30-50% of PWH, even in those achieving robust viral suppression on ART [4]. The etiology of 105 HAND is multifactorial and complex, involving persistent neuroinflammation, the activity of neurotoxic viral 106 proteins (e.g., Tat, gp120), and dysregulation of host signaling pathways that collectively disrupt synaptic 107 integrity and neuronal function [3]. Importantly, while neurons are the primary cells affected in HAND, glial 108 cell population in the central nervous system (CNS), are increasingly recognized as key mediators of HAND-109 related neuropathology [5, 6]. 110 The microtubule-associated protein Tau is a central regulator of neuronal function, ensuring the stability of 111 axonal microtubules and supporting efficient transport of cargo essential for synaptic activity and neuronal 112 survival [7, 8] . Under physiological conditions, Tau protein undergoes tightly regulated cycles of 113 phosphorylation and dephosphorylation that allow dynamic modulation of cytoskeletal structure. However, 114 disruption of this phosphorylation event (Tau hyperphosphorylation) gives rise to neurodegenerative disease 115 [7, 9, 10]. Aberrantly phosphorylated Tau exhibits diminished binding to microtubules, misfolds into abnormal 116 conformations, and progressively accumulate into insoluble neurofibrillary tangles [11, 12]. These inclusions 117 not only serve as histopathological hallmarks of Alzheimer’s disease (AD) and related tauopathies but also 118 correlate strongly with synaptic dysfunction, neuronal loss, and the severity of cognitive decline [13, 14]. 119 Multiple factors contribute to the pathological transformation of Tau, ultimately driving its involvement in AD 120 and related dementias [15, 16]. Emerging evidence suggests that HIV infection, even in individuals receiving 121 suppressive antiretroviral therapy, can disrupt the physiological regulation of Tau phosphorylation. Since 122 neurons do not express canonical HIV entry receptors such as CD4 and co-receptors CCR5 or CXCR4, they 123 are not infected by the virus [17, 18] . Nevertheless, neuron cells remain profoundly susceptible to the 124 downstream consequences of viral exposure. Instead of direct infection, neuronal injury arises predominantly 125 through indirect mechanisms, most notably the actions of soluble viral proteins , mainly Tat and gp120, and 126 released cytokines from infected glial or immune cells [19, 20] . These cytotoxic factors dysrupt cellular 127 homeostasis and interfere with host kinase -phosphatase signaling cascades, leading to dysregulated Tau 128 phosphorylation events that compromise cytoskeletal integrity, axonal transport, and synaptic stability 129 ultimately leading to HAND, and related tauopathies. 130 Cocaine, one of the most prevalent substances abused among PWH, is a well -established cofactor in the 131 progression of HAND [21, 22] . Cocaine abuse independently exacerbates neurodegenerative processes, 132 accelerates cognitive decline, and has been associated with increased susceptibility to HIV infection and 133 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint replication within the CNS [22-24]. Mechanistic studies suggest that cocaine induces oxidative stress, disrupts 134 blood-brain barrier integrity, activates glial cells, and modulates multiple signaling pathways, including those 135 governing inflammation and cell survival. Our recent studies further reveal that cocaine use enhances HIV 136 transcription by activating transcription factors such as NF-κB and MSK1, altering epigenetic modifications at 137 the long terminal HIV repeat (LTR) promoter [24, 25]. 138 Beyond transcriptional activation, we have shown that cocaine increases the susceptibility of CD4⁺ T cells to 139 HIV infection by augmenting key co-stimulatory signaling pathways , involving, NF -kB, NFAT and AP -1, 140 thereby creating a cellular state more favorable to viral entry and replication [25-27]. Additionally, we have 141 demonstrated that cocaine activates DNA-dependent protein kinase (DNA-PK) in both T cells and microglial 142 cells, which alleviates RNA polymerase II pausing at the LTR. This effect is mediated through phosphorylation 143 of TRIM28, a chromatin -associated repressor, thus enabling more efficient transcriptional elongation and 144 sustained viral gene expression [28, 29]. These molecular changes establish a favorable environment for 145 persistent HIV activity and may synergize with host signaling dysregulation to exacerbate neuropathology, 146 particularly within the central nervous system. 147 Neurodegeneration is the consequence of dysregulated intracellular signaling, in which kinases play a crucial 148 role [30]. In the healthy normal brain, a balance between kinases and phosphatases ensures proper regulation 149 of cytoskeletal dynamics, synaptic activity, and stress adaptation. Disruption can cause series of 150 phosphorylation events that promote neuronal dysfunction [31]. Therefore, several cellular pathways have 151 been involved in the regulation of Tau phosphorylation ultimately leading to neurodegeneration. However, 152 glycogen synthase kinase 3 beta (GSK3β), a serine/threonine kinase that directly phosphorylates Tau at 153 multiple pathological sites remain a major kinase [32, 33]. GSK3β activity is inhibited by phosphorylation at 154 serine 9 (Ser9), a modification typically mediated by the upstream kinase AKT (also known as protein kinase 155 B), a central node in cell survival, metabolism, and growth signaling [34]. Dysregulation of this AKT-GSK3β 156 axis has been consistently reported in models of tauopathy, and other neurodegenerative conditions, 157 underscoring its pathogenic significance [35, 36]. In addition to GSK3β, several other kinases have been 158 shown to phosphorylate Tau, including Cyclin-dependent kinase 5 ( CDK5), Extracellular signal-regulated 159 kinases ( ERK1/2), c-Jun N -terminal kinase ( JNK), p38 MAPK, Microtubule affinity -regulating kinases 160 (MARKs), AMP-activated protein kinase ( AMPK) [37, 38] , Protein kinase A ( PKA), Calcium/calmodulin-161 dependent kinase II (CaMKII) [39], and Protein kinase C (PKC), act on overlapping sets of phosphorylation 162 sites and often respond to cellular stress signals such as inflammation, oxidative damage, and excitotoxicity. 163 Dysregulation of the MAPK/ERK signaling pathway has been strongly associated with neurodegenerative 164 disorders, including AD, where abnormal kinase activity contributes to synaptic dysfunction, Tau 165 hyperphosphorylation, and neuronal loss [40, 41]. 166 Ribosomal S6 kinase 1 (RSK1, encoded by RPS6KA1) is known to be a key downstream effector of 167 MAPK/ERK pathway and plays important roles in regulating cell growth, survival, and gene expression [42, 168 43]. Despite its central role in MAPK/ERK signaling, RSK1 has not yet been systematically investigated in the 169 context of Tau phosphorylation or AD, and direct evidence linking its dysregulation to disease onset or 170 progression remains limited. This gap in knowledge prompted us to investigate this aspect in detail and define 171 if RSK1 is an underexplored contributor to the molecular mechanisms underlying Tauopathy responsible for 172 AD and related neurodegenerative conditions. Additionally, several of the aforementioned kinases have been 173 extensively studied in AD and other tauopathies, their specific contributions to Tau dysregulation during HAND 174 remain poorly understood. Viral proteins such as Tat and gp120 are known to disrupt intracellular signaling, 175 yet the precise mechanisms by which they interact with the kinase -phosphatase networks that regulate Tau 176 remain unclear. Elucidating the signaling pathways through which Tau pathology accelerates HAND is 177 therefore a critical research priority. Addressing this knowledge gap will not only advance our mechanistic 178 understanding of HAND pathogenesis but may also reveal convergent therapeutic targets relevant to both 179 classical and virally mediated neurodegenerative disorders, including AD and HAND. 180 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Selecting an appropriate model system is critical for studying neurodegenerative diseases, as it directly 181 influences the reliability, reproducibility, and translational relevance of the findings. While primary neurons 182 offer high physiological relevance, they are short-lived, fragile, and technically challenging to maintain under 183 in vitro conditions, highlighting the need for alternative neuron -like systems that are more experimentally 184 tractable yet retain key neuronal features [44]. Glioma cell lines provide a robust, practical, and experimentally 185 tractable platform to investigate mechanisms of neurodegeneration due to their neural origin, robust growth 186 properties, and retention of signaling pathways relevant to function and disease pathology [45, 46]. Unlike 187 primary neurons, which are post-mitotic and difficult to maintain long term, glioma cells readily expand in vitro, 188 enabling reproducible experiments and large-scale molecular and pharmacological studies [47]. Importantly, 189 these cells retain critical signaling pathways relevant to disease pathology, such as MAPK/ERK and PI3K/AKT 190 signaling, oxidative stress responses, and glial -neuronal interactions, all of which are central to the 191 progression of neurodegenerative disorders [48-50]. Since glial dysfunction and altered kinase signaling 192 contribute significantly to synaptic loss, protein aggregation, and neuronal death, glioma cells serve as a 193 practical surrogate model to dissect these mechanisms. Although glioma cells cannot fully replicate the 194 complexity of the CNS, especially neuronal and glial interactions in vivo, their tractability and physiological 195 relevance make them a useful and valuable model for mechanistic studies of neurodegeneration, as well as 196 testing the potential therapeutics. 197 In this study , using both 2D and 3D neuronal model systems, we delineate the molecular mechanisms 198 underlying Tau phosphorylation in response to HIV infection and/or cocaine exposure. We demonstrate that 199 HIV exposure robustly upregulates and activates RSK1, which in turn inactivates GSK3β through an AKT -200 independent mechanism. Activated RSK1 directly promotes Tau phosphorylation. On the other hand, cocaine 201 exposure not only induces RSK1 but also strongly activates AKT1, leading to GSK3β inactivation through 202 phosphorylation at serine 9 (p -GSK3β S9) in an AKT -dependent manner, thereby further enhancing Tau 203 phosphorylation. Notably, Tau phosphorylation persists even under conditions of GSK3β inhibition during both 204 HIV and cocaine exposure, indicating that Tau modification is primarily driven through an RSK1 -centered, 205 GSK3β-independent pathway. These findings highlight the pivotal role of RSK1 and underscore the 206 complexity of the signaling networks regulating Tau phosphorylation. Using complementary genetic and 207 pharmacological approaches, including CRISPR/Cas9 -mediated knockout, overexpression systems, and 208 selective kinase inhibition, we further establish that RSK1 functions upstream of both AKT activation and 209 GSK3β inactivation, exerting context -dependent effects on Tau phosphorylation. Collectively, our findings 210 highlight the kinase signaling crosstalk underlying HAND and cocaine-associated tauopathy, identifying RSK1 211 as a mechanistic hub and potential therapeutic target for neurotoxicity and HAND in PWH, including those 212 who use illicit substances, such as cocaine. 213 Running Title: - Signaling Crosstalk Underlying Tauopathy during HIV infection and Cocaine Abuse 214

- HIV, Cocaine, Tau phosphorylation, RSK1, AKT, GSK3β. 215

216 Cell Culture 217 H80 cells (originally obtained from the Darell Bigner Laboratory, Duke University) [51] were maintained in 218 DMEM/F12 medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin. Jurkat 219 T cells (human CD4 ⁺ T lymphocyte line; ATCC TIB ‑152), MT -4 cells and U937 cells were cultured in 220 RPMI‑1640 supplemented with 10% FBS, 1% penicillin–streptomycin, and 2 mM L‑glutamine. HEK293T cells, 221 microglial cells, and SH ‑SY5Y neuroblastoma cells were propagated in Dulbecco’s modified Eagle medium 222 (DMEM) containing 10% FBS, 1% penicillin –streptomycin, and 2 mM L ‑glutamine. All cell lines were 223 maintained at 37 °C in a humidified incubator with 5% CO2 and were used between passages 3 and 8. 224 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Inhibitor Treatments 225 H80 cells [51] were seeded and allowed to adhere overnight prior to treatment. Cells were incubated with 226 selective inhibitors targeting RSK1 or GSK3β, at a final concentration of 10 µM each. The RSK1 inhibitor BI-227 D1870 (Selleckchem, Cat. No. S2843), and GSK3β inhibitor CHIR99021 (Tocris Bioscience, Cat. No. 4423), 228 were prepared as stock solutions in Dimethyl sulfoxide (DMSO) and prepared working solutions at 10 mM. 229 Cells were treated with inhibitors or equivalent volumes of DMSO vehicle control for 24 hours at 37°C in a 230 humidified 5% CO₂ incubator. Following treatment, cells were either exposed to HIV or cocaine or both and 231 harvested for downstream analyses including Immunoblotting. All treatments were performed in experimental 232 triplicate or biological triplicate to ensure reproducibility. 233 Cocaine Treatment 234 Cells were treated with 10 µM cocaine hydrochloride. For acute exposure, treatments were applied for 235 durations ranging from 15 minutes up to 6 hours. Otherwise, specifically mentioned all the treatments are 236 done chronically. For chronic exposure, cells received two treatments each day randomly for 48 hours and at 237 least 30 min-3h prior cells harvesting. Control cells were treated with PBS or kept untreated. 238 HIV Virus Production and Infection 239 Jurkat cells or MT-4 cells were infected with replication -competent Human Immunodeficiency Virus Type 1 240 (strain 93/TH/051, R5- and X4-tropic virus) (NIH AIDS Reagent Program) by spinoculation at 1,200 × g for 2 241 h at 25°C in the presence of 8 µg/mL Polybrene. Following infection, cells were incubated for 48 h -72 h. 242 Supernatants containing HIV virions were harvested, cleared by low-speed centrifugation (500 × g, 10 min), 243 filtered through a 0.45 µm syringe filter, and stored at −80 °C. HIV production was confirmed by 244 immunoblotting for the HIV p24 capsid protein. 245 H80 Exposure to HIV 246 H80 were seeded and exposed to HIV-containing supernatant or in normal medium (control) by spinfection at 247 1,000 rpm for 2 h at room temperature (RT). The following day, cells underwent a second spinfection under 248 the same conditions and were subsequently transferred to 100 -mm dishes. After 48 h, cells were harvested 249 for protein analysis. Control H80 cells were processed identically with supernatant from uninfected Jurkat/MT-250 4 cells. 251 Lentiviral production and CRISPR/Cas9-mediated RPS6KA1 knockout 252 Lentiviral particles encoding Cas9 were produced by co‑transfecting HEK293T cells with either lentiCRISPR 253 v2 (Addgene #52961) or lentiCas9 ‑Blast (Addgene #52962), a gift from Feng Zhang [52], together with the 254 packaging plasmid psPAX2 (Addgene #12260), and the envelope plasmid pMD2.G (Addgene #12259), a gift 255 from Didier Trono using Lipofectamine 2000 (Thermo Fisher Scientific). Viral supernatants were harvested 256 48 h post‑transfection, clarified through a 0.45‑µm filter, and used immediately or stored at −80 °C. H80 cells 257 were transduced in the presence of 8 µg/mL polybrene and selected with puromycin (1–2 µg/mL)/ blasticidin 258 to generate stable Cas9 ‑expressing populations. To disrupt RPS6KA1, Cas9 ‑positive H80 cells were 259 subsequently transduced with lentiviral particles encoding one of three independent RPS6KA1 ‑targeting 260 sgRNAs (BRDN0001148481, Addgene #75499; BRDN0001145974, Addgene #75497; BRDN0001148103, 261 Addgene #75498), originally developed by John Doench and David Root [53]. These sgRNA lentiviruses were 262 produced in HEK293T cells using the same Lipofectamine‑based system. Following transduction, cells were 263 allowed to recover for 48 h and then selected with puromycin (1 –2 µg/mL) for 3 –5 days to obtain 264 RSK1‑knockout populations. 265 Overexpression in H80 Cells 266 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint For RSK1 overexpression, H80 cells were seeded at 60 –70% confluence in 60 mm plates and transiently 267 transfected with 2 µg of a CMV promoter -driven full-length human RSK1 expression plasmid ( pKH3-human 268 RSK1, Addgene cat no #13841, a gift from John Blenis [54]) or empty vector control using Lipofectamine 2000 269 (Thermo Fisher Scientific) according to the manufacturer’s instructions. Briefly, plasmid DNA and 270 Lipofectamine reagent were diluted separately in Opti-MEM (Gibco), combined, and incubated for 30 minutes 271 at RT before adding drop by drop to cells. Cells were incubated with the transfection complexes for 6 hours, 272 after which the medium was replaced with fresh growth medium. Protein lysates were harvested 48 hours 273 post-transfection using lysis buffer with protease and phosphatase inhibitors (Roche). Overexpression 274 efficiency was confirmed by immunoblotting with anti-RSK1 antibodies. 275 Spheroids formation 276 Three-dimensional spheroids were generated using 96 -well round -bottom Biofloat 3D cell culture plates 277 (Sarstedt, Cat. No. 83.3925.400), which provides a non -adhesive surface to promote uniform spheroid 278 formation. To prevent cell attachment, each well was pre -treated with 60 µL of Anti -Adherence Rinsing 279 Solution (AARS; Stemcell Technologies, Cat no #07010) and incubated under sterile conditions at RT for 24 h. 280 Following incubation, the AARS was aspirated and stored it for potential reuse, and wells were rinsed with 281 100 µL of phosphate -buffered saline (PBS) to remove residual solution. Prepared plates were either used 282 immediately or stored in sterile bags at 4 °C for up to two weeks. For spheroid assembly, a mixed cell 283 suspension containing H80 cells, SH-SY5Y neuroblastoma cells, and microglia (5,000 cells of each type) was 284 prepared in 150 µL of DMEM supplemented with 10% FBS and 1% penicillin–streptomycin. This suspension 285 was dispensed into each treated well, and plates were incubated at 37 °C in a humidified atmosphere with 286 5% CO₂ for 24 h to allow initial aggregation and spheroid formation. After 48 h of culture, 100 µL of medium 287 was carefully removed from each well and replaced with fresh medium containing HIV virus to initiate infection 288 or exposure. Spheroids were incubated for 5 h under the same conditions, after which the medium was 289 exchanged for fresh medium containing either cocaine or no treatment. Cocaine was administered twice daily 290 for 48 h to mimic repeated exposure. At the end of the treatment period, spheroids were harvested by pooling 291 24 spheroids into a single Falcon tube, representing one biological sample. A total of 96 spheroids were 292 collected, corresponding to four experimental groups: untreated control (24 spheroids), cocaine -treated (24 293 spheroids), HIV -exposed/infected (24 spheroids), and HIV -exposed/infected plus cocaine -treated (24 294 spheroids) (spheroid figure in Supplementary). Each pooled sample was washed with 1 mL PBS to remove 295 residual medium and treatment compounds, followed by addition of 80 µL of 1× passive lysis buffer (Promega 296 E1941) to facilitate cell lysis and protein extraction for Immunoblot analyses. 297 Generation of human cerebral organoids (hCOs) 298 Human cerebral organoids (hCOs) were generated from human induced pluripotent stem cells (hiPSCs) 299 following our previously established protocols, as described in detail in reference [55]. Briefly, human induced 300 pluripotent stem cells (hiPSCs), derived from dermal fibroblasts, were used to generate hCOs following 301 STEMdiff™ protocols (STEMCELL Technologies). The cells were plated in ultra-low attachment 96-well plates 302 at 11,000 cells per well and incubated for 24 hours to form embryoid bodies (EBs). As the EBs grew to 400–303 600 μm over about 5 days, they were transferred to 24 -well plates and cultured in induction medium for 48 304 hours. The EBs were then embedded in Matrigel and moved to 6-well plates with expansion medium, where 305 they developed neuroepithelial structures after 3 days. Finally, the organoids were matured on an orbital 306 shaker at 70 rpm in maturation medium at 37°C for an additional 40 days. 307 Immunoblotting 308 Total cell lysates were prepared using 1X Passive Lysis Buffer (Promega E1941) supplemented with protease 309 and phosphatase inhibitor cocktails (Roche), following the manufacturers’ instructions. Following cell 310 harvesting, lysates were incubated on ice for 30 minutes with intermittent vortexing for 30 seconds every 311 10 minutes to facilitate complete lysis. For the Spheroids and organoids, samples were lysed by mechanical 312 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint disruption with passive lysis buffer through repeated passage through a 200 µL pipette tip, followed by eight 313 cycles of rapid freeze–thaw in liquid nitrogen and a 37 °C water bath. The lysates were then incubated on ice. 314 After incubation, samples were centrifuged at maximum speed (≥14,000 × g) for 30 minutes at 4 °C to pellet 315 cell debris. The resulting supernatants were collected, and protein concentrations were determined using the 316 Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific). Protein concentration was normalized, and an 317 equal amount of protein was mixed with 5X Laemmle Sample buffer, heated to 95ºC for 10 min, and then 318 resolved by SDS -PAGE on a 9% or 10% or 12% gel at 120 volts until the dye reached the bottom. The 319 resolved proteins were then transferred to a nitrocellulose membrane (Amersham). Membranes were blocked 320 for 1 h at RT in 3% bovine serum albumin (BSA) in Tris-buffered saline containing 0.1% Tween-20 (TBS-T), 321 followed by overnight incubation at 4 °C with primary antibodies against phospho-RSK1 Ser380 (sc-136476), 322 phospho-RSK1 thr348 (sc-101770), phospho-p90RSK (Thr359/Ser363) (CST#9344), RSK1 (CST #9347), 323 RSK1/2/3 (CST #14813), phospho-AKT T308 (CST #4056), phospho-AKT S473 (CST #4060), AKT1 (CST 324 #2938), phospho-GSK3β S9 (CST #5558), GSK3β (CST #12456), phospho-Tau (CST #9632S), Tau (CST 325 #46687), MAP2 (17490-1-AP), GAPDH (sc-25778), and β-actin (Sigma-Aldrich A5316). After three washes 326 with 1X TBST, the blot was detected using the Odyssey infrared imaging system application software 3.0 (Li-327 cor Bioscience). 328 RNA Extraction and Quantitative PCR (qPCR) 329 Total RNA was extracted from H80 cells after 24 h of HIV exposure using the RNeasy Plus Mini Kit (Qiagen) 330 following the manufacturer’s protocol, ensuring elimination of genomic DNA contamination. RNA purity and 331 concentration were confirmed by nanodrop and RNA gel electrophoresis. Complementary DNA (cDNA) was 332 synthesized from 1 µg of total RNA using the High -Capacity cDNA Reverse Transcription Kit (Applied 333 Biosystems). Quantitative PCR was performed using SYBR Green on a QuantStudio 5 Real -Time PCR 334 system (Applied Biosystems) with gene-specific primers for IL-1β, TNF-α, RSK1, and GAPDH as an internal 335 control. Relative gene expression was quantified by the 2^−ΔΔCt method, normalizing target gene expression 336 to GAPDH and comparing HIV exposed to controls (exposed without HIV). All reactions were conducted in 337 technical triplicates across at least three biological replicates. 338 Immunofluorescence staining and imaging. 339 To characterize the H80 cells, H80 cells were cultured on sterile coverslips , which was initially treated with 340 PolyD Lysine and allowed to adhere overnight. Cells were fixed with 4% paraformaldehyde (PFA) in PBS for 341 30 minutes at RT, followed by permeabilization with 0. 25% Triton X-100 in PBS for 10 minutes at RT. After 342 permeabilization, cells were washed thrice with PBS and then incubated with a blocking solution containing 343 10% horse serum and 2% BSA in PBS for 60 minutes at RT to reduce non -specific binding. Subsequently, 344 cells were washed and then incubated overnight at 4°C with directly conjugated primary antibodies: anti-NeuN 345 Alexa Fluor 647 (Cat. No. 608453, BioLegend), p-Tau S396 (Phospho-Tau (Ser396) (PHF13) Mouse mAb 346 #9632)/anti-Tau phospho ser396 (BioLegend #829001) and MAP2 Polyclonal antibody (proteintech, 17490-347 1-AP). The following day, samples were washed three times with PBS, incubated with secondary for 1 h and 348 counterstained with Hoechst (300 nM in PBS) for 10 minutes at RT, followed by an additional three PBS 349 washes. Coverslips were mounted using Aqua -Mount mounting medium (Epredia, Cat. No. 13800) and 350 imaged using an EVOS M7000 Imaging System (Cat no. AMF7000) equipped with 20× and 40× oil immersion 351 objectives. 352 To investigate the cellular effects of cocaine and HIV exposure, exposed cells were fixed with 4% PFA in PBS 353 for 15 minutes at RT. After washing the cell thrice in PBS, Fixed cells were permeabilized with 0.25% Triton 354 X-100 in PBS for 10 minutes at RT, then incubated for 1 hour in blocking buffer composed of 10% horse serum 355 and 2% BSA in PBS. Primary antibodies against phosphorylated Tau (Phospho-Tau (Ser396) (PHF13) Mouse 356 mAb #9632), RSK-1 (CST #9333), p-GSK3β S9 (CST #5558), p-AKT S473 (CST #4060), AKT1 (CST #2938), 357 were incubated overnight at 4°C. The next day, cells were washed thoroughly to remove unbound antibodies 358 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint and subsequently incubated with species -specific, fluorophore -conjugated secondary antibodies for 45 359 minutes at RT in the dark. Nuclei were counterstained with Hoechst for 10 minutes at RT, followed by three 360 PBS washes. Coverslips were mounted using Aqua Mount (Epredia, cat. no. 13800), and samples were 361 imaged using an EVOS imaging system equipped with 10x, 20x and 40x objectives. 362 Flowcytometry 363 H80 cells were analyzed for surface expression of the HIV entry receptors CD4, CCR5, and CXCR4 using 364 multicolor flow cytometry. Cells were harvested, washed with PBS containing 2% FBS, and incubated with 365 fluorochrome‑conjugated monoclonal antibodies for 30 minutes at 4 °C in the dark. To assess co‑expression 366 of CD4 and CXCR4, cells were stained with APC anti ‑human CD4 (BioLegend, cat. no. 317416) and PE 367 anti‑human CD184 (CXCR4) (BioLegend, cat. no. 306505). For CD4 and CCR5 co ‑staining, cells were 368 incubated with PE anti ‑human CD4 (BioLegend, cat. no. 357403) and APC/Cyanine7 anti ‑human CD195 369 (CCR5) (BioLegend, cat. no. 359110). Following staining, cells were washed, resuspended in PBS, and 370 analyzed on a flow cytometer. Data acquisition and compensation were performed using standard instrument 371 settings, and analysis was conducted with FlowJo software (BD Biosciences). 372 Densitometry and Statistical Analysis 373 All experiments were performed with a minimum of three independent biological replicates and/or 374 experimental triplicates. Immunoblots were quantified using ImageJ (NIH , Version 1.53e). Band intensities 375 were normalized to β-actin or GAPDH or corresponding total protein and expressed as fold change relative 376 to controls. Data are shown as mean ± standard deviation (SD) from ≥3 independent experiments. Statistical 377 analyses were performed using GraphPad Prism v9 (Version 9.1.2). For comparisons between two groups 378 (e.g., control vs. RSK1 knockout or Ctrl vs. RSK1O/E), unpaired two-tailed Student’s T-tests were employed. 379 For experiments involving multiple conditions or time points, one -way or two -way analysis of variance 380 (ANOVA) followed by Dunnett’s multiple comparisons test was used to assess significance, with p < 0.05 381 considered statistically significant. 382

383 H80 Cells Exhibit Neuronal Characteristics as Evidenced by NeuN, MAP2, and Tau Expression 384 Given the considerable variability , as well as the growth and maintenance challenges associated with 385 commonly used neuronal cell lines such as SH-SY5Y [56], we sought to evaluate whether a glioma cell line, 386 H80 retains key neuronal characteristics, particularly relevant signaling pathways and susceptibility to 387 neurotoxicity. Our findings indicate that H80 cells exhibit features suitable for modeling neuron -specific 388 signaling events, including pathways involved in Tau phosphorylation. Importantly, H80 cells offer several 389 practical advantages over conventional neuronal models, including robust and reproducible growth, rapid 390 proliferation, low baseline cytotoxicity, and stable culture behavior, making them a reliable and efficient system 391 for mechanistic studies of neuronal signaling and neurotoxicity. The neuronal characteristics of the H80 glioma 392 cells [51] were confirmed by performing immunofluorescence staining using NeuN, a nuclear marker widely 393 recognized for its specificity to post -mitotic neurons. Both unstained and secondary -only controls were 394 included in parallel to validate antibody specificity and to exclude background artifacts. Our analysis revealed 395 robust nuclear NeuN immunoreactivity across the majority of H80 cells, providing clear evidence that H80 is 396 a neuronal cell line (Figure 1A). The expression of Tau, a microtubule-associated protein that plays a critical 397 role in axonal stability and is centrally implicated in tauopathies and other neurodegenerative processes , 398 further substantiates the neuronal identity of H80 cells (Supplementary Figure S1). The presence of Tau not 399 only reinforces the neuronal -like characteristics of H80 cells but also highlights their relevance as a model 400 system for studying Tau-associated signaling pathways and neurotoxicity. 401 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Furthermore, to demonstrate that H80 cells possess the molecular characteristics of differentiated neurons, 402 we focused on the expression of microtubule -associated protein 2 (MAP2). MAP2 is a neuron -specific 403 cytoskeletal protein that is predominantly localized to dendrites, where it plays a pivotal role in stabilizing 404 microtubules and maintaining neuronal morphology. MAP2 expression is therefore widely recognized as a 405 hallmark of neuronal differentiation and identity. We initially assessed MAP2 expressions in H80 cells using 406 immunofluorescence microscopy, with HEK293T cells serving as a non -neuronal reference (control), 407 Microglial cells (unknown or known MAP2 negative cell line) and SH-SY5Y cells serving as neuronal reference 408 (positive control or known to express MAP2). Notably, the expression of MAP2 was exclusively observed in 409 H80 cells (Figure 1B and supplementary S1) and also in positive control (SH-SY5Y), where the protein 410 displayed a distinct filamentous distribution throughout the cytoplasm, consistent with the structural 411 organization seen in neurons. As anticipated, HEK293T cells lacked detectable MAP2 signal whereas SHS5Y 412 has a strong MAP2 expression under identical staining conditions ( Figure 1B). The expression of MAP2 413 indicates that H80 cells, but not HEK293T cells and microglial cells, belongs to neuronal lineage. To further 414 corroborate these findings, we performed immunoblotting using total cellular lysates from H80 cells, HEK293T 415 cells, and microglial cells. Consistent with the immunofluorescence results, MAP2 protein was robustly 416 detected in H80 cell lysates, whereas it was undetectable in both HEK293T and microglial samples ( Figure 417 1C). Importantly, the absence of MAP2 expression in microglial cells, another brain-resident glial cells of non-418 neuronal lineage, underscores the neuronal specificity of this marker. Together, these complementary assays 419 provide convergent evidence that H80 cells exclusively express MAP2. The presence of MAP2 exclusively in 420 H80 cells, but not in two distinct non-neuronal cell types, strongly supports the conclusion that H80 cells is a 421 neuronal cell line that exhibits cytoskeletal and molecular features consistent with neuronal identity and 422 differentiation. 423 Therefore, the co-expression of NeuN, MAP2, and Tau provides convergent and robust evidence that H80 424 cells exhibit hallmark neuronal features (Figures 1A to C, and Supplementary Figure S1). Collectively, these 425 findings support the classification of H80 cells as a neuronal-like cell model. The presence of these canonical 426 neuronal markers not only confirms their neuronal identity but also highlights their suitability as a versatile 427 platform for investigating neuron -specific molecular mechanisms. In particular, H80 cells offer a valuable 428 system for studying signaling pathways involved in the regulation of Tau protein phosphorylation and activity, 429 as well as broader processes underlying neuronal function and neurotoxicity. 430 Since our study focuses on HIV induced neurotoxicity and neurons are not directly infected by HIV [17], we 431 next examined the expression of the key HIV entry receptors and co-receptors in H80 cells (Figure 1D). To 432 determine the expression profile of HIV entry receptors on H80 cells, we performed flow cytometric analysis 433 for CD4, CCR5, and CXCR4. Cells were co-stained with either CD4 (APC anti-human CD4 from BioLegend 434 cat no 317416) and CXCR4 (PE anti-human CD184 (CXCR4) Antibody from BioLegend cat no. 30 6505) or 435 CD4 (PE anti-human CD4 Antibody from BioLegend cat no. 357403 and CCR5 (APC/Cyanine7 anti-human 436 CD195 (CCR5) Antibody from BioLegend cat no. 359110). Our results demonstrated that H80 cells lack 437 detectable CD4 expression under basal conditions, indicating the absence of the primary receptor required 438 for productive HIV entry. Notably, approximately 20% of H80 cells expressed surface CXCR4, whereas CCR5 439 expression was undetectable. These findings suggest that while H80 cells are unlikely to support productive 440 HIV infection due to the absence of CD4, the presence of CXCR4 on a subset of cells may render them 441 responsive to HIV -associated proteins or signaling pathways, thereby contributing to HIV -induced 442 neurotoxicity. To ensure assay specificity and reliability, HEK293T cells were included as a negative control 443 and showed no detectable expression of CD4, CCR5, or CXCR4. In contrast, U937 cells served as a positive 444 control and exhibited robust basal expression of CD4 and both co -receptors. Collectively, these data further 445 support the suitability of H80 cells as a model to study HIV-mediated neurotoxic effects independent of direct 446 viral infection. 447 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Altogether, our findings demonstrate that H80 cells exhibit key neuronal characteristics, as evidenced by the 448 expression of canonical neuronal markers, including NeuN, MAP2, and Tau. In addition, receptor profiling 449 revealed that H80 cells lack detectable expression of CD4 and CCR5 but express the chemokine receptor 450 CXCR4 on a subset of cells. Previous studies, including those by Kaul et al., have shown that both CXCR4 451 and CCR5 can mediate HIV -associated neuronal injury, while CCR5 may also engage neuroprotective 452 signaling pathways [57]. The selective expression of CXCR4 in H80 cells is particularly noteworthy, as it 453 suggests that analogous to neurons, these cells may be responsive to HIV-associated proteins and signaling 454 events linked to CXCR4 engagement, despite the absence of productive viral entry. This receptor profile 455 closely aligns with current understanding that neuronal damage in NeuroHIV is largely mediated indirectly 456 through viral proteins and host signaling pathways rather than direct infection. 457 Based on these observations, we next sought to determine how HIV exposure influences neuronal stress 458 responses and neurotoxicity in this model, with a particular focus on Tau phosphorylation, a well-established 459 marker of tauopathy. Given the confirmed neuronal phenotype of H80 cells and their expression of CXCR4, 460 these cells provide a biologically relevant system to study HIV-induced neuronal dysfunction independent of 461 productive infection. Accordingly, we directly exposed H80 cells to HIV to investigate the underlying molecular 462 mechanisms driving HIV -associated neurotoxicity and Tau dysregulation, enabling us to dissect signaling 463 pathways that contribute to neurodegenerative processes in the context of NeuroHIV. 464 HIV exposure upregulates RSK1 expression 465 To investigate the signaling pathways underlying HIV -induced tauopathy, H80 cells were exposed to HIV 466 virions (strain 93/TH/051, R5- and X4-tropic virus, dual-tropic HIV-1). Because neurons lack the primary HIV 467 receptor (CD4), they are resistant to productive infection; thus, analogous to neurons, this model allows us to 468 specifically examine HIV -mediated signaling and neurotoxic effects independent of viral replication. HIV 469 virions were generated by infecting Jurkat T cells, and successful infection was confirmed by immunoblot 470 detection of the viral capsid protein p24 in infected cell lysates ( Figure 2A). Virus-containing supernatants 471 from either Jurkat or MT -4 cells were then collected and used to expose H80 cells using a two -round 472 spinfection (spinoculation) protocol, which enhances viral contact and ensures efficient exposure of neuronal 473 cells to viral particles. As a negative control, H80 cells were exposed to supernatants from uninfected 474 Jurkat/MT-4 cells (Figure 2A). Following exposure, H80 cells were harvested at 24 and 48 hours after the 475 second spinfection for downstream analyses. Total RNA and protein lysates were collected to assess changes 476 in intracellular signaling pathways and Tau phosphorylation status, enabling us to define the molecular 477 mechanisms by which HIV exposure induces neuronal stress and Tau dysregulation in this system. 478 To assess the impact of HIV exposure on the cellular transcriptional machinery and to confirm successful viral 479 exposure on H80 cells, we quantified mRNA expression levels of representative inflammatory and signaling 480 genes using qRT-PCR at 24 hours post-exposure. HIV-exposed H80 cells expressed a robust upregulation 481 of interleukin -1β (IL -1β) and tumor necrosis factor -alpha (TNF -α) transcripts, both of which are central 482 mediators of proinflammatory signaling cascades ( Figure 2 B). Consistent with previous reports 483 demonstrating that neurons can produce cytokines in response to HIV-associated stress [58], this pronounced 484 induction provides strong evidence of effective HIV exposure and activation of innate immune signaling in 485 H80 cells. In addition, we observed a modest but reproducible increase in RSK1 mRNA expression. Given 486 the established role of RSK1 as a downstream effector of MAPK signaling, this finding suggests early 487 activation of stress-responsive and pathogen-associated signaling pathways, complementing the observed 488 inflammatory response. Collectively, these transcriptional changes not only confirm ed the upregulation of 489 RSK1 upon HIV exposure but also underscore the functional reactivation of the cells, thereby validating the 490 functional responsiveness of H80 cells to HIV exposure. These results further strengthen the biological 491 relevance of our H80 neuronal model for dissecting the molecular mechanisms underlying HIV -induced 492 neuroinflammation, stress signaling, and Tau-associated pathology. 493 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint HIV exposure-driven RSK1 upregulation induces Tau phosphorylation 494 To investigate whether HIV exposure promotes pathological Tau modification, we first performed 495 immunofluorescence staining using an antibody specific for the phosphorylated Tau protein at Ser396 (p-Tau-496 Ser396), a site commonly associated with neurotoxicity and neurodegeneration. Compared with controls 497 (exposing without the virus) , HIV-exposed H80 cells displayed a marked increase in p -Tau-Ser396 signal, 498 suggesting that HIV exposure drives Tau phosphorylation (Figure 2 C). Notably, in addition to the 499 phosphorylation of Tau, HIV exposure markedly upregulates RSK1 (Supplementary Figure S 2). The 500 concurrent induction of RSK1 suggested a role of this kinase in promoting Tau phosphorylation. These 501 findings indicate that the effect of HIV exposure on Tau phosphorylation facilitated through the regulation of 502 RSK1, thereby highlighting a mechanistic link between RSK1 expression and Tau phosphorylation. 503 To substantiate these observations, we conducted immunoblot analysis of whole -cell lysates following HIV 504 exposure (under the same condition s). H80 cells were cultured in four independent dishes (two biological 505 replicates per condition), and whole cell lysates were collected 48 h after HIV exposure. Protein lysates from 506 each dish were prepared and quantified individually. Equivalent amounts of protein were resolved by 507 immunoblotting to assess RSK1, RSK1/2/3, phosphorylated Tau (p Tau S396), and Tau, using actin or total 508 protein as loading controls. Consistent with the immunofluorescence imaging results, immunoblotting 509 revealed a robust induction of p-Tau-Ser396 in HIV-exposed cells (lanes 3-4) compared to control (exposing 510 with supernatant from uninfected cells, lanes 1-2). In contrast, total Tau protein levels exhibited only a modest 511 increase, indicating that HIV exposure predominantly induces post -translational modification of Tau, rather 512 than significantly altering its overall expression. Notably, this increase in Tau phosphorylation was 513 accompanied by a pronounced upregulation of RSK1, as well as elevated levels of the RSK1/2/3 isoforms , 514 suggesting activation of the RSK signaling pathway in response to HIV exposure. These findings indicate that 515 RSK1 activation occurs in parallel with HIV-induced Tau phosphorylation and may serve as a critical molecular 516 link between viral exposure and Tau dysregulation in H80 cells ( Figure 2D ). Quantitative densitometric 517 analysis of the immunoblot signals, normalized to β actin and/or total protein, revealed a statistically significant 518 increase in the levels of both RSK1 and p -Tau Ser396 in HIV exposed H80 cells relative to controls/No HIV 519 (Figure 2E). To further assess whether H80 cells support productive HIV infection, lysates from HIV-exposed 520 and control (no HIV) conditions were analyzed by immunoblotting using an anti-HIV p24 antibody, with Jurkat 521 T cells included as positive (HIV -exposed) and negative (uninfected) controls ( Figure 2F). Consistent with 522 the absence of CD4 expression, p24 was not detected in H80 cell lysates, indicating that these cells do not 523 support productive HIV infection and are instead exposed to HIV virions without undergoing viral replication. 524 Importantly, despite the lack of productive infection, exposure to HIV virions was sufficient to induce robust 525 activation of RSK1 signaling in H80 cells, which correlated with increased Tau phosphorylation. Given the 526 well-established roles of RSK1 in nuclear gene regulation and cytoplasmic signaling crosstalk, these findings 527 suggest that HIV-induced activation of RSK1 serves as a key mediator linking viral exposure to downstream 528 neuronal stress responses. Collectively, our results support a model in which HIV virion exposure, 529 independent of productive infection, triggers RSK1 activation, leading to pathological Tau phosphorylation 530 and tauopathy-associated signaling, thereby establishing a mechanistic connection between HIV exposure 531 and neuronal dysfunction mediated through the RSK1 pathway. 532 Cocaine enhances Tau phosphorylation by upregulating and activating RSK1 533 To investigate whether cocaine contributes to Tau pathology, we chronically exposed H80 cells to cocaine 534 twice daily for 2 days and subsequently assessed Tau phosphorylation by immunofluorescence (Figure 3A). 535 Cocaine-exposed H80 cells exhibited a pronounced increase in phosphorylated Tau (p -Tau-Ser396) 536 compared with untreated controls, while the total abundance of Tau protein remained unchanged (Figure 3B). 537 These findings indicate that cocaine primarily promotes post -translational modification of Tau, rather than 538 significantly altering its overall expression. Given that our previous findings demonstrated that HIV exposure 539 upregulates RSK1 (Figure 2), we next examined whether cocaine -induced Tau phosphorylation is similarly 540 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint associated with upregulation in RSK1 expression. Immunofluorescence analysis revealed a significant 541 upregulation of RSK1 following chronic cocaine exposure ( Supplementary Figure S 2). Thus, parallel 542 increment in RSK1 expression and Tau phosphorylation suggested crucial role for RSK1 signaling in 543 mediating cocaine -induced Tau modification . These findings were further validated in an independent 544 experiment, confirming the reproducibility of RSK1 upregulation and enhanced Tau phosphorylation in 545 response to cocaine exposure in H80 cells (Figure 3C–3F). 546 To further solidify the data obtained through immunofluorescence analysis, we chronically treated H80 cells 547 alone or in combination with cocaine and HIV as shown in Figure 3A. The cell lysate was analyzed by 548 Immunoblotting. The i mmunoblotting of whole -cell lysates revealed that cocaine produced a modest but 549 reproducible increase in total RSK1 protein, accompanied by a n elevation in its phosphorylation at Thr 348, 550 Thr 359, S363 and S380, which marks functionally active form of RSK1. Given that specific posttranslational 551 modification of RSK1 was quantitatively correlated with the increase in p-Tau-Ser396, implicating RSK1 as a 552 mediator of cocaine-driven Tau modification (Figure 3C and 3D). Both cocaine and HIV significantly increased 553 p-Tau-Ser396 relative to controls, although the magnitude of the effect differed: HIV produced a robust 554 elevation in Tau phosphorylation, whereas cocaine induced a more modest increase. Notably, co-exposure to 555 cocaine and HIV did not result in strictly additive effects (but it is more on higher side), suggesting that these 556 stimuli converge on overlapping molecular pathways. Analysis of RSK1 activation under these conditions 557 revealed that HIV strongly enhanced both total RSK1 expression and its phosphorylation, including at S380, 558 Thr348, Thr359/ and S363 far exceeding the effects of cocaine alone. HIV exposure also led to a moderate 559 increase in Thr348 phosphorylation, further distinguishing its mode of RSK1 regulation from cocaine. 560 Importantly, the degree of RSK1 activation in each condition closely aligned with the extent of Tau 561 phosphorylation, strengthening the link between RSK1 signaling and Tau modification. Together, these results 562 establish that both cocaine and HIV drive activation of the RSK1 signaling axis in H80 cells, albeit with distinct 563 magnitudes and mechanistic profiles (Figure 3C and 3D). Cocaine modestly increases RSK1 expression and 564 Tau phosphorylation, whereas HIV produces a more potent and broad er activation of RSK1, resulting in 565 stronger downstream Tau modification. These findings provide mechanistic insight into how viral infection and 566 substance use converge on a shared signaling pathway to promote Tau pathology. 567 To delineate the immediate signaling responses triggered by cocaine and HIV, we examined the acute 568 activation of RSK1 in H80 neuronal cells following short term exposure to each stimulus. H80 cells were 569 cultured in eight independent dishes, providing two biological replicates for each of the four treatment 570 conditions (Control, cocaine, HIV, and cocaine + HIV). Cells were exposed acutely for 15 minutes ( Figure 571 3E), after which they were harvested and lysed. Protein lysates from each dish were prepared and quantified 572 individually, and equivalent amounts of total protein were resolved by immunoblotting to evaluate 573 phosphorylation of RSK1 at the activating sites Ser380 and Thr359/Ser363, alongside total RSK1 levels, 574 using actin or total protein as loading controls. Acute exposure to either cocaine or HIV elicited rapid and 575 robust activation of the RSK1 signaling pathway in H80 cells. Immunoblot analysis revealed marked increases 576 in RSK1 phosphorylation at both Ser380 and Thr359/Ser363 relative to controls ( Figure 3F and 3G ). This 577 activation was consistently observed across biological replicate lanes corresponding to each treatment group 578 (lanes 3–4, 5 –6, and 7 –8 compared with lanes 1 –2), demonstrating strong reproducibility of these rapid 579 phosphorylation events. Notably, the magnitude of RSK1 activation differed between stimuli. While cocaine 580 induced a clear and reproducible increase in phosphorylation at both regulatory sites (p-RSK-1 Thr359/S363 581 and S380), HIV exposure elicited a substantially stronger response, producing the highest levels of RSK1 582 activation among all acute treatment conditions. 583 Together, these findings establish that RSK1 is a rapidly responsive kinase activated within minutes of cocaine 584 or HIV exposure, and that the amplitude of this response is greater under HIV stimulation than under cocaine 585 alone. This further reinforces the pattern observed under chronic exposure conditions, underscoring the 586 consistency of cocaine and HIV ‑driven enhancement of RSK ‑1 activation across temporal paradigms , 587 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint reinforcing its central role as an early signaling mediator linking cocaine and HIV exposure to downstream 588 neuronal stress responses and Tau pathology. 589 HIV or Cocaine converge on GSK3β inhibition through Ser9 phosphorylation 590 GSK3β is a key regulator of neuronal signaling and a well -established mediator of Tau phosphorylation [59-591 63]. The enzymatic activity of GSK3β is primarily controlled through its phosphorylation at the serine-9 (Ser9) 592 residue, (p-GSK3β-Ser9), which serves as an inhibitory modification that limits substrate access and thereby 593 inhibits GSK3β kinase activity. Given the well-established role of GSK3β in mediating Tau pathology, we next 594 examined whether HIV and cocaine influence the regulation of this kinase. As shown in Figures 2 and 3 , 595 both HIV and cocaine exposure markedly enhanced Tau phosphorylation, raising the possibility that these 596 effects are mediated, at least in part, through altered GSK3β activity. To investigate further, we specifically 597 assessed the phosphorylation status of GSK3β at its inhibitory Ser9 residue, thereby determining whether 598 HIV and cocaine relieve the inhibitory regulation of GSK3β and contribute to the observed increase in Tau 599 phosphorylation. Using total cellular lysates from Figure 2A, we examined the phosphorylation status of 600 GSK3β at the inhibitory Ser9 site by immunoblotting. Surprisingly, o ur results demonstrated a marked 601 increase in Ser9 phosphorylation in response to HIV infection in Jurkat cells, while the levels of total GSK3β 602 remained unchanged (Figure 4A). This increase in inhibitory phosphorylation suggests that GSK3β becomes 603 inactivated upon HIV infection or under conditions of ongoing viral infection. 604 To determine whether HIV exposure directly modulates GSK3β activity, we exposed H80 cells for 15 min to 605 the supernatant derived from HIV-infected cells (Jurkat infected with HIV), while supernatant from uninfected 606 cell cultures (Jurkat uninfected with HIV) was used as a control, as shown in Figure 2A . After 15 min 607 exposure, the cells were harvested, and protein lysates were subjected to immunoblot analysis to evaluate 608 the phosphorylation status of GSK3β at Ser9 site, marking functionally inactive form of GSK3β. We observed 609 a pronounced increase in Ser9 phosphorylation of GSK3β in HIV-exposed cells compared to the control 610 (Figure 4B and 4C). Since phosphorylation at Ser9 is known to suppress GSK3β catalytic activity, this 611 increase strongly suggests that acute HIV exposure enhance posttranslational modification of GSK3β at Se9, 612 which functionally inactivates GSK3β. These findings indicate that HIV exposure can rapidly influence host 613 kinase signaling pathways. 614 Subsequently, we examined how acute exposure to cocaine and HIV modulates GSK3β signaling in H80 615 cells. To evaluate these rapid effects, cells were cultured in eight independent dishes, providing two biological 616 replicates per treatment condition, and exposed for 15 minutes to cocaine, HIV, or both. Following treatment, 617 cells were harvested and lysed, and protein lysates from each dish were prepared and quantified individually. 618 Equal amounts of total protein were subjected to immunoblot analysis to assess phosphorylation of GSK3β 619 at Ser9, with total protein levels serving as loading controls. Both cocaine and HIV independently elicited a 620 clear increase in Ser9 phosphorylation of GSK3β (Lane 3-4, lane 5-6 and lane 7-8 compared to lane 1 -2), 621 indicating enhanced inhibitory modification of the kinase (Figure 4E and 4F). Notably, the consistent elevation 622 of Ser9 phosphorylation in HIV exposed samples further confirms the reproducibility of HIV-mediated GSK3β 623 inactivation, as also observed in Figures 4B and 4C . These findings demonstrate that acute exposure to 624 either cocaine or HIV is sufficient to rapidly inactivate GSK3β, revealing a shared regulatory mechanism by 625 which both stimuli attenuate GSK3β activity in H80 cells. Importantly, this occurs despite the observed 626 increase in Tau phosphorylation, further supporting the notion that cocaine - and HIV -induced Tau 627 dysregulation proceeds through GSK3β-independent signaling pathways, likely involving alternative kinases 628 such as RSK1. 629 To further substantiate our finding that both HIV and cocaine are independently able to inactivate GSK3β, we 630 performed immunoblot analysis to assess inhibitory phosphorylation of GSK3β at Ser9. H80 cells were 631 chronically exposed for 48 hours to cocaine, HIV virions, or a combination of both. Under each condition, 632 cocaine alone, HIV alone, or combined HIV plus cocaine exposure, we observed a robust increase in Ser9 633 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint phosphorylation relative to untreated controls, whereas total GSK3β protein levels remained unchanged. 634 These results indicate that the effects of HIV and cocaine on GSK3β are mediated through post-translational 635 inhibitory modification, rather than changes in protein abundance. Notably, combined exposure to HIV and 636 cocaine also produced a clear increase in p -GSK3β Ser9 compared with control cells ( Figure 4G and H ), 637 confirming that both stimuli converge on functional inactivation of GSK3β. The consistency of this response 638 across all treatment groups strongly supports the conclusion that HIV and cocaine each suppress GSK3β 639 activity in H80 neuronal cells. This finding is particularly striking because GSK3β is widely recognized as a 640 major Tau kinase. Accordingly, if GSK3β is the main Tau kinase, inactivation of GSK3β would be expected 641 to reduce Tau phosphorylation. In contrast, we observed the opposite outcome: despite clear evidence of 642 GSK3β inactivation, Tau phosphorylation was markedly increased following exposure to HIV and/or cocaine 643 (Figures 2 and 3). This apparent paradox strongly suggests that Tau phosphorylation in this setting is driven 644 through an alternative pathway that is independent of GSK3β activity. Our data point towards RSK1 as a likely 645 upstream mediator of this effect. Indeed, both HIV and cocaine induced significant upregulation and activation 646 of RSK1, coinciding with enhanced Tau phosphorylation under conditions in which GSK3β remained 647 inactivated. These observations support a model in which RSK1-driven signaling bypasses the need for active 648 GSK3β and sustains pathological Tau phosphorylation, thereby promoting tauopathy -associated neuronal 649 stress responses. 650 Collectively, these findings demonstrate that HIV and cocaine independently converge on GSK3β inactivation 651 via Ser9 phosphorylation yet simultaneously promote Tau hyperphosphorylation through a distinct upstream 652 mechanism, most likely involving RSK1. This convergence on inhibitory GSK3β signaling, together with 653 activation of an alternative Tau -phosphorylating pathway, identifies a critical molecular axis by which viral 654 exposure and substance use disrupt neuronal signaling, ultimately contributing to Tau dysregulation and the 655 CNS impairments, including neuropathological processes associated with HAND. 656 Cocaine, but not HIV exposure, activates AKT 1 signaling through phosphorylation at Thr308 and 657 Ser473. 658 Phosphorylation of GSK3β at Ser9, a critical inhibitory modification, is tightly regulated by upstream kinases, 659 most notably AKT, which plays a central role in neuronal signaling cascades and survival pathways [34]. AKT-660 mediated phosphorylation of GSK3β at Ser9 serves as a key inhibitory checkpoint that suppresses GSK3β 661 catalytic activity and prevents excessive substrate phosphorylation. Given our findings that both HIV and 662 cocaine independently increase Ser9 phosphorylation on GSK3β, thereby promoting its inactivation, we next 663 sought to determine whether these are direct effect are mediated through activation of the AKT signaling 664 pathway. To investigate this, we examined the phosphorylation status of AKT at its regulatory sites that control 665 its functional activity . We hypothesized that stimulation of AKT activity would restrict GSK3β activity by 666 catalyzing its phosphorylation at Ser9 (p-GSK3β-Ser9). This approach allowed us to directly evaluate whether 667 HIV and cocaine converge upstream on AKT to regulate GSK3β activity and thus, contribute to Tau 668 hyperphosphorylation. 669 To determine the regulation of AKT signaling pathway upon HIV and cocaine exposure, we performed 670 immunofluorescence staining for phosphorylated AKT at Ser473 (p-AKT-Ser473), a well-established marker 671 of AKT activation and a critical modification required for full kinase activity. H80 cells were chronically exposed 672 for 2 days to cocaine, HIV virions, or both, after which AKT1 phosphorylation status was evaluated (Figure 673 5A). Cocaine -treated cells displayed a robust , reproducible and significant increase in p -AKT-Ser473 674 fluorescence compared with untreated controls, indicating that cocaine strongly activates AKT signaling 675 pathway (Figure 5B). In contrast, HIV exposure alone did not show any effect in p-AKT-Ser473 levels under 676 the same conditions (Supplementary Figure S3), suggesting that HIV does not directly induce AKT activation 677 in this context. Notably, total AKT protein levels were unaffected across all conditions, confirming that the 678 observed changes were attributable to post -translational regulation of phosphorylation rather than changes 679 in protein expression or stability. 680 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint To further validate our data obtained through immunofluorescence analyses and confirm the diverse impact 681 of cocaine and HIV on AKT signaling, we performed immunoblot analysis using cell lysates from H80 cells 682 following 2 days of chronic exposure to cocaine, HIV, or a combination of both and evaluate both the 683 phosphorylation sites of AKT. Immunoblotting revealed that cocaine treatment induces a significant increase 684 in phosphorylation of AKT at both Thr308 and Ser473 compared with untreated controls (Figure 5C). Since 685 phosphorylation at Thr308 and Ser473 are both essential for full activation of AKT, the simultaneous increase 686 in phosphorylation at these two regulatory sites strongly confirms that cocaine induces a robust activation of 687 the AKT signaling pathway. In contrast, HIV exposure alone did not alter phosphorylation at either site (further 688 validating our immunofluorescence results), demonstrating that HIV exposure alone does not enhance or 689 activate AKT signaling pathway. Notably, combined treatment with cocaine and HIV reproduced the increase 690 in phosphorylation pattern of AKT observed with cocaine alone, indicating that cocaine exerts a dominant 691 stimulus in activating AKT signaling, even in the presence of viral exposure. This suggests that cocaine 692 overrides any potential influence of HIV on this pathway. Quantitative densitometric analysis (Figure 5D) 693 provided further support, showing a significant increase in AKT phosphorylation at both Thr308 and Ser473 694 in cocaine alone and HIV+ cocaine-exposed cells, while HIV exposure alone had no measurable impact 695 relative to controls. Importantly, total AKT protein levels remained constant across all conditions, confirming 696 that the observed changes reflect post-translational modifications rather than at gene expression. Together, 697 these results establish that cocaine, but not HIV, selectively activates AKT signaling in H80 cells, with cocaine 698 driving strong phosphorylation of AKT at both activation sites and dominating over HIV when both stimuli are 699 present. 700 Collectively, these findings establish cocaine, but not HIV, as a potent activator of the AKT signaling pathway 701 in H80 cells. Mechanistically, cocaine promotes sustained AKT phosphorylation at Thr308 and Ser473, which 702 subsequently promotes inactivation of GSK3β by catalyzing its phosphorylation at Ser9 (p-GSK3β-Ser9). 703 Thus, cocaine promotes GSK3β inactivation (p-GSK3β-Ser9) both directly through AKT stimulation and also 704 via RSK1 activation. In contrast, HIV inactivates GSK3β exclusively through an AKT-independent 705 mechanisms, primarily through RSK1 signaling. 706 Thus, cocaine and HIV converge on a shared downstream effector, GSK3 β inactivation, but diverge in their 707 upstream regulatory pathways. Cocaine acts through an AKT-dependent pathway that also involves RSK1 708 stimulation, whereas HIV acts predominantly through an AKT-independent, RSK1-driven pathway. This dual 709 convergence and divergence highlight the complexity of signaling networks regulating Tau phosphorylation 710 and underscore how viral exposure and substance use engage distinct yet overlapping molecular pathways 711 to drive neurotoxicity and tauopathy. 712 HIV and Cocaine upregulate RSK1 to drive Tau phosphorylation through a GSK3 β-independent 713 mechanism 714 Our findings thus far indicate that HIV and cocaine share a common upstream signaling pathway through the 715 upregulation of RSK1 (Figure 2 and 3). However, their downstream signaling pathways diverge in a stimulus-716 specific manner. Notably, HIV does not activate the AKT signaling pathway ( Figure 5), whereas cocaine 717 exposure leads to robust AKT activation, as evidenced by increased phosphorylation at both key regulatory 718 sites, Thr308 and Ser473 ( Figure 5A-D). Despite this divergence, both stimuli ultimately converge on the 719 inactivation of GSK3 β, as demonstrated by increased inhibitory phosphorylation at Ser9 ( Figure 4). This 720 shared downstream effect establishes GSK3 β as a critical point of signaling convergence. Importantly, this 721 convergence results in a common pathological outcome, the enhanced Tau phosphorylation of Tau (Figure 2 722 and 3), a hallmark of neurodegenerative processes and tauopathy. 723 These observations suggest a model in which distinct upstream signaling pathways (AKT -dependent for 724 cocaine and AKT -independent for HIV) converge on shared downstream nodes, while simultaneously 725 engaging alternative kinase pathways, such as RSK1, to drive Tau phosphorylation. The persistence of Tau 726 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint phosphorylation despite GSK3 β inactivation further underscores the involvement of GSK3 β-independent 727 mechanisms, likely mediated by RSK1. Based on these findings, we next sought to delineate the relative 728 contributions and mechanistic interplay between RSK1 and GSK3 β in mediating Tau phosphorylation 729 following HIV and cocaine exposure. This approach aims to clarify how these signaling pathways integrate to 730 produce a shared neurotoxic phenotype, thereby providing deeper insight into the mechanistic basis of HIV- 731 and substance use-mediated tauopathy and neurodegeneration. 732 To validate our findings and further define the effects of cocaine and HIV on Tau phosphorylation, we 733 performed a comprehensive immunoblot analysis using lysates from H80 cells following 48 hours of chronic 734 exposure to cocaine, HIV, or their combination. Cells were seeded into 12 independent culture dishes across 735 ≥3 passages/days, yielding three biological replicates per condition (Control, cocaine, HIV, and cocaine + 736 HIV). After treatment, cells from each dish were harvested and lysed individually, and equal amounts of total 737 protein were subjected to immunoblot analysis. Consistent with our earlier observations ( Figures 2 and 3), 738 immunoblotting revealed that cocaine, HIV, and their combination each induced a significant increase in Tau 739 phosphorylation at Ser396 compared with untreated controls ( Figures 6A and 6B ). These results robustly 740 confirm and extend our previous findings, demonstrating that both stimuli, independently and in combination, 741 promote sustained Tau hyperphosphorylation under chronic exposure conditions. Phosphorylation of Tau at 742 Ser396 is widely recognized as a marker of pathological Tau, and the observed increase at this site strongly 743 confirms that cocaine exposure, HIV exposure, and their combination each induce Tau hyperphosphorylation. 744 In parallel, we assessed the activity of GSK3 β under these conditions. Interestingly, despite the increase in 745 Tau phosphorylation, GSK3 β was found to be functionally inactivated, as evidenced by enhanced 746 phosphorylation at its inhibitory Ser9 residue (Figures 6A and 6B). These findings demonstrate that both HIV 747 virion exposure and chronic cocaine treatment promote Tau phosphorylation while simultaneously restricting 748 GSK3β activity, indicating that Tau hyperphosphorylation occurs through a GSK3β-independent mechanism. 749 Importantly, although both stimuli converge on GSK3β inactivation, they do so via distinct upstream pathways. 750 HIV induces GSK3 β inactivation in an AKT -independent manner, whereas cocaine mediates this effect 751 through AKT activation, as reflected by increased phosphorylation at both Thr308 and Ser473. Together, these 752

highlight a critical mechanistic distinction in upstream signaling while reinforcing a shared downstream 753 outcome, Tau hyperphosphorylation despite GSK3 β inhibition, suggesting the involvement of alternative 754 kinases, such as RSK1, in driving Tau pathology. 755 Furthermore, to delineate the relative contributions of RSK1 and GSK3β to Tau phosphorylation in response 756 to exposures HIV and cocaine, we employed highly specific small molecular pharmacological inhibitors. Our 757 rationale was that if RSK1 is the common mediator of Tau phosphorylation induced by these stimuli, then its 758 inhibition should suppress this effect, whereas inhibition of GSK3β would not. H80 cells were pretreated for 759 24 hours with BI -D1870 (a selective RSK1 inhibitor) or CHIR -99021 (a highly specific GSK3 β inhibitor), 760 followed by exposure to HIV, cocaine, or both. After treatment , total protein lysates were analyzed by 761 immunoblotting to assess signaling and phosphorylation dynamics (Figure 6C and 6D). Consistent with our 762 earlier findings (Figures 2 and 3), both HIV and cocaine exposures resulted in robust upregulation of RSK1 763 activity compared to untreated controls (Figure 6 C and 6D). Importantly, pretreatment with BI -D1870 764 effectively suppressed RSK1 activation, whereas CHIR-99021 had no effect on RSK1 activity, indicating that 765 RSK1 functions independently of, and upstream from, GSK3β in this signaling cascade (Figure 6C and 6D). 766 We next examined the effect of these inhibitors on GSK3β activity. As shown previously (Figure 4), both HIV 767 and cocaine exposures led to inactivation of GSK3β, as evident from enhanced phosphorylation at S9 (Figure 768 6C and 6D). Notably, inhibition of RSK1 with BI-D1870 reduced Ser9 phosphorylation, suggesting restoration 769 of GSK3β activity and indicating that RSK1 contributes to GSK3β inactivation (Figure 6C and 6D). In contrast, 770 direct inhibition of GSK3β with CHIR-99021 resulted in sustained inactivation, confirming the specificity and 771 effectiveness of the inhibitors. To evaluate the downstream effects of these perturbations, we examined Tau 772 phosphorylation and found that both HIV and cocaine exposure markedly increased Tau phosphorylation, 773 consistent with RSK1 activation. (Figures 6C and 6D). 774 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Importantly, RSK1 inhibition with BI-D1870 markedly suppressed Tau phosphorylation, whereas inhibition of 775 GSK3β with CHIR -99021 did not alter Tau phosphorylation induced by HIV exposure or cocaine. These 776 findings demonstrate that Tau phosphorylation in this context is primarily mediated through an RSK1 -777 dependent, GSK3β-independent mechanism. Interestingly, prolonged treatment (≥24 hours) with BI -D1870 778 also resulted in a reduction in total RSK1 protein levels, while housekeeping controls (GAPDH) remained 779 unchanged. This suggests that sustained pharmacological inhibition may influence not only RSK1 activity but 780 also its protein stability or turnover. 781 Collectively, these results establish a hierarchical signaling relationship in which RSK1 acts upstream of 782 GSK3β and plays a central role in mediating Tau phosphorylation following HIV and cocaine exposure. 783 Furthermore, they highlight RSK1 as a critical therapeutic target, as its inhibition effectively attenuates Tau 784 pathology while also modulating downstream kinase signaling. We next investigated the temporal dynamics 785 of acute cocaine- and HIV-induced signaling in H80 cells to determine whether rapid inactivation of GSK3 β, 786 reflected by increased phosphorylation at the inhibitory Ser9 site, occurs in parallel with changes in Tau 787 phosphorylation at the pathological Ser396 residue. To assess time-dependent effects, cells were exposed to 788 cocaine or HIV for 1, 3, and 6 hours (Figure 6E). Protein lysates collected at each time point were analyzed 789 by immunoblotting for p -GSK3β Ser9, p-Tau Ser396, total Tau, and actin. Both cocaine and HIV induced a 790 sustained increase in GSK3β Ser9 phosphorylation at the 3- and 6-hour time points (lanes 3–4 vs. lane 1; 791 lanes 7–8 vs. lane 5), indicating persistent inhibition of GSK3β activity during acute exposure. Notably, this 792 inhibitory modification did not lead to a reduction in Tau phosphorylation. Instead, we observed a progressive 793 and robust increase in p -Tau Ser396 over time, demonstrating that Tau phosphorylation continues to 794 accumulate despite functional inactivation of GSK3 β. The simultaneous suppression of GSK3 β activity and 795 enhancement of Tau phosphorylation provides strong evidence for a GSK3β-independent mechanism of Tau 796 regulation under both cocaine and HIV exposure. These findings implicate alternative kinases, most 797 prominently RSK1, as key drivers of Tau phosphorylation at Ser396, even in the absence of active GSK3 β. 798 Collectively, these data demonstrate that acute cocaine and HIV exposures sustain Tau hyperphosphorylation 799 independently of GSK3β activity, highlighting RSK1 as a dominant upstream kinase in this process. These 800

are consistent with our chronic exposure studies, further reinforcing a model in which RSK1-dependent 801 signaling persistently drives Tau phosphorylation across temporal contexts. 802 RSK1 knockout impairs AKT signaling, activates GSK3β, and suppresses Tau phosphorylation 803 To further confirm the direct role of RSK1 in regulating Tau phosphorylation, we generated RSK1 knockout 804 (KO) H80 cells using CRISPR-Cas9. Immunoblot analysis confirmed efficient loss/reduction of RSK1 protein, 805 enabling us to investigate the downstream signaling pathways. Strikingly, RSK1 ablation led to a marked 806 reduction in Tau phosphorylation at Ser396 (p -Tau S396), while total Tau levels remained unchanged, 807 indicating that RSK1 regulates Tau primarily through post -translational mechanisms rather than 808 transcriptional or translational control. Quantitative analyses across independent experiments confirmed a 809 significant decrease in p-Tau S396 in RSK1 KO cells compared with controls (Figure 7A and B), establishing 810 that RSK1 is required for efficient phosphorylation of Tau at this pathological site. 811 Interestingly, RSK1 knockout also impacted GSK3β signaling. Specifically, loss of RSK1 resulted in a 812 reduction of inhibitory phosphorylation of GSK3β at Ser9, indicating reactivation of GSK3β kinase activity. 813 These findings demonstrate that RSK1 acts upstream of GSK3β and contributes to its inactivation, consistent 814 with our pharmacological inhibition data (Figure 6). Notably, however, reactivation of GSK3β did not restore 815 Tau phosphorylation at Ser396, strongly documenting that RSK1 drives site -specific Tau phosphorylation 816 independently of GSK3β (Figure 7). This observation highlights the complexity of Tau regulatory networks 817 and indicates that RSK1 is the main kinase controlling pathological Tau modification (Tau-S396), rather than 818 merely modulating canonical Tau kinases such as GSK3β. 819 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint We further evaluate the impact of RSK1 loss on AKT signaling pathway. RSK1 knockout resulted in severely 820 impaired AKT signaling pathway, as evidenced by a marked reduction in phosphorylation at both Thr308 and 821 Ser473. Notably, total AKT protein levels were also reduced, suggesting that RSK1 contributes not only to 822 AKT activation but also to AKT protein stability and/or abundance. These findings identify RSK1 as a positive 823 regulator of AKT signaling at both functional and protein stability levels (Figure 7C and D). Collectively, the 824 coordinated effects of RSK1 deletion, including suppression of Tau phosphorylation, reactivation of GSK3β, 825 and attenuation of AKT signaling, establish RSK1 as a central upstream regulator of interconnected kinase 826 networks governing Tau pathology. Importantly, the inability of GSK3β reactivation to rescue Tau 827 phosphorylation further underscores the primary role of RSK1 in mediating Tau-S396 phosphorylation. 828 Taken together, these results position RSK1 as a critical signaling hub integrating AKT and GSK3β pathways 829 to regulate Tau phosphorylation and identify it as a promising therapeutic target for Tauopathies, including 830 HIV-associated neurocognitive disorders (HAND) and cocaine-associated neurodegeneration. 831 RSK1 functions as an upstream regulator of AKT- GSK3β signaling cascade. 832 As detailed above, we identified RSK1 as a key upstream regulator of both AKT and GSK3β signaling in H80 833 cells. Loss (knock out) or pharmacological inhibition of RSK1 impaired AKT activation and simultaneously 834 reactivated GSK3β, as evidenced by reduced phosphorylation at its inhibitory Ser9 site. To further 835 substantiate this regulatory hierarchy, we next tested whether RSK1 overexpression produces the reciprocal 836 effects. Based on our prior findings, we hypothesized that elevated RSK1 levels would enhance AKT 837 activation while promoting GSK3β inactivation via increased Ser9 phosphorylation. 838 To evaluate the signaling consequences of RSK1 upregulation, H80 cells were transiently transfected with a 839 CMV-driven RSK1 expression construct, and lysates were collected 48 hours post -transfection for 840 immunoblot analysis. Overexpression was confirmed by a marked increase in total RSK1 protein, along with 841 elevated levels of RSK1/2/3 isoforms, validating robust induction of the RSK signaling axis ( Figure 8A ). 842 Notably, phosphorylation of RSK1 at Ser380, a key autophosphorylation site associated with catalytic 843 activation, was significantly increased. However, phosphorylation at Thr348, a critical activation loop residue 844 in the N -terminal kinase domain that is typically phosphorylated downstream of ERK signaling, remained 845 unchanged or even decreased relative to vector -transfected controls. The selective increase in pSer380 846 without a corresponding increase in pThr348 suggests that RSK1 overexpression alone does not result in full 847 enzymatic activation and may reflect a partial or ERK -independent activation state. This phosphorylation 848 pattern implies that RSK1 may be primed for signaling but is not fully engaged in substrate phosphorylation, 849 potentially limiting canonical downstream activity. 850 Consistent with our model, RSK1 overexpression resulted in robust activation of AKT, as demonstrated by 851 increased phosphorylation at both Thr308 and Ser473, the two critical regulatory sites required for full AKT 852 activity (Figure 8B and D). Importantly, total AKT protein levels remained unchanged, indicating that RSK1 853 enhances AKT signaling primarily through post -translational activation rather than changes in protein 854 abundance. 855 In parallel, RSK1 overexpression led to a pronounced increase in inhibitory phosphorylation of GSK3β at Ser9 856 (Figure 8C and D), confirming functional inactivation of this kinase. Total GSK3β levels remained constant, 857 further supporting that this effect reflects post-translational regulation. These findings reinforce the conclusion 858 that RSK1 negatively regulates GSK3β activity through phosphorylation-dependent inhibition, likely in part via 859 AKT activation. 860 Collectively, these results establish RSK1 as a central upstream modulator of the AKT-GSK3β signaling axis 861 in H80 cells. By simultaneously activating AKT and suppressing GSK3β, RSK1 integrates MAPK/RSK and 862 PI3K/AKT signaling pathways and creates a cellular environment conducive to pathological Tau 863 phosphorylation. This mechanistic link suggests that dysregulation of RSK1 could shift neuronal kinase 864 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint networks toward pathological Tau modification, with broader implications for survival, metabolism, and 865 neurodegenerative processes. This coordinated regulation highlights RSK1 as a critical signaling hub that 866 governs kinase network balance and promotes Tau dysregulation. Altogether, our data position RSK1 as a 867 key mechanistic driver and potential therapeutic target in conditions characterized by aberrant Tau 868 phosphorylation, including HIV-mediated neurotoxicity (HAND) and cocaine-induced neurodegeneration. 869 Cocaine- and HIV-induced signaling is conserved across 2D neuronal cultures, 3D spheroids, and 870 brain organoid model systems 871 To determine whether the signaling pathways induced by cocaine and HIV in H80 cells are conserved across 872 additional neuronal systems, we extended our investigation to SH -SY5Y neuroblastoma cells ( Figure 9B), 873 3D neuronal spheroids ( Figure 9A and 9C ), and human iPSC-derived brain organoids (Figure 9D). This 874 approach allowed us to evaluate the robustness and reproducibility of the identified signaling axis across 875 models of increasing biological complexity. 876 First, to assess reproducibility in an independent neuronal cell line, SH -SY5Y cells were exposed to HIV for 877 48 hours, followed by immunoblot analysis of key signaling markers. Chronic HIV exposure resulted in a 878 pronounced upregulation of RSK1, accompanied by increased inhibitory phosphorylation of GSK3β at Ser9, 879 while total GSK3β levels remained unchanged (Figure 9B; lanes 3–4 vs. 1–2). Notably, this was paralleled 880 by a significant increase in Tau phosphorylation at Ser396, demonstrating that HIV induces a coordinated 881 signaling response involving RSK1 activation, GSK3β inactivation, and Tau dysregulation. These findings 882 confirm that the signaling axis identified in H80 cells is reproducible in additional neuronal cell types. 883 To further examine whether these mechanisms are preserved in a more physiologically relevant 3D context, 884 we utilized a multicellular neuronal spheroid model. Spheroids were generated by co-culturing equal numbers 885 of H80 cells, microglia, and SH-SY5Y cells (15,000 cells total per spheroid), and subjected to control, cocaine, 886 HIV, or combined treatments. Following 48 -hour exposure, pooled spheroids from each condition were 887 analyzed by immunoblotting. Consistent with 2D models, both cocaine and HIV treatments induced robust 888 RSK1 upregulation, increased GSK3β Ser9 phosphorylation, and enhanced Tau phosphorylation at Ser396 889 (Figure 9C ). Importantly, the presence of microglia enabled productive HIV infection within the spheroid 890 system, further increasing physiological relevance. These coordinated molecular changes demonstrate that 891 the RSK1-GSK3β-Tau signaling axis is preserved within a multicellular 3D neuronal microenvironment. 892 We next evaluated whether these findings extend to higher -order neural systems using human cerebral 893 organoids (hCOs) derived from hiPSCs. Following exposure to cocaine, HIV, or both, organoids were 894 processed for immunoblot analysis. Consistent with results from both 2D cultures and spheroids, treated 895 organoids exhibited marked upregulation of RSK1, increased inhibitory phosphorylation of GSK3β at Ser9, 896 and sustained Tau phosphorylation at Ser396 (Figure 9D). Notably, Tau phosphorylation remained elevated 897 despite GSK3β inactivation, reinforcing the presence of a GSK3β -independent mechanism, likely mediated 898 by RSK1. These results confirm that the identified signaling pathway is conserved even in complex, human-899 relevant 3D brain models. 900 Together, these data demonstrate that the core signaling cascade , RSK1 activation/upregulation, GSK3β 901 inactivation, and pathological Tau phosphorylation , identified in H80 cells is highly reproducible across 902 neuronal systems of increasing complexity, including 3D cultures, multicellular spheroids, and brain 903 organoids. This consistency underscores the biological robustness and generalizability of this pathway and 904 highlights its relevance across diverse human-derived neural contexts, strengthening its potential significance 905 in HIV- and cocaine-associated neurodegeneration. 906 Overall, our investigation identifies RSK1 as a central signaling hub that integrates and coordinates multiple 907 kinase pathways governing Tau phosphorylation. Both HIV and cocaine robustly activate RSK1, which 908 eventually promotes the inactivation of GSK3β, establishing a convergent downstream signaling axis despite 909 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint distinct upstream regulatory mechanisms. Importantly, Tau phosphorylation persists even under conditions 910 of GSK3β inhibition, demonstrating that RSK1 drives pathological Tau modification through a GSK3β -911 independent mechanism. These findings establish RSK1 as an essential upstream regulator of 912 interconnected kinase networks that control site-specific Tau phosphorylation. 913 Notably, these signaling dynamics are consistently reproduced across multiple neuronal systems , including 914 2D cultures, 3D spheroids, and human brain organoids , underscoring the robustness, reproducibility, and 915 biological relevance of this pathway across diverse neural contexts. Collectively, these results support a 916 unified model in which RSK1 serves as the primary mediator linking HIV and cocaine exposure to Tau 917 dysregulation and neuronal stress responses. Beyond HAND and substance use -related neurotoxicity, this 918 mechanism has broader implications for Tau -driven neurodegenerative diseases, including Alzheimer’s 919 disease and related cognitive disorders. Thus, RSK1 emerges as a key mechanistic driver and a promising 920 therapeutic target for conditions characterized by aberrant Tau phosphorylation and neurodegeneration. 921 Distinct yet convergent signaling mechanisms by which HIV exposure and cocaine drive Tau 922 phosphorylation/ pathology 923 To summarize our findings, we propose the following model for HIV - and cocaine -induced Tau 924 phosphorylation (Figure 10). In this study, we demonstrate that exposure to HIV and cocaine leads to robust 925 Tau phosphorylation, and we delineate the distinct yet convergent molecular mechanisms underlying this 926 process. Although both stimuli ultimately induce Tau phosphorylation, the upstream signaling pathways they 927 engage are mechanistically distinct. In the context of HIV exposure, we observed a pronounced and sustained 928 upregulation and activation of RSK1. Activated RSK1 promotes Tau phosphorylation while simultaneously 929 inhibiting GSK3β activity through an AKT -independent mechanism. Consistent with this pathway, HIV 930 exposure resulted in a marked increase in Tau phosphorylation, identifying RSK1 as a dominant mediator of 931 HIV-driven Tau dysregulation. On the other hand, cocaine exposure engages a partially overlapping but 932 distinct signaling cascade. While cocaine induces modest activation of RSK1, it strongly stimulates AKT, as 933 evidenced by robust phosphorylation at Thr308 and Ser473, both required for full catalytic activation. 934 Activated AKT subsequently phosphorylates GSK3β at Ser9 (p -GSK3β-Ser9), leading to its functional 935 inactivation. Despite differences in upstream signaling intensity, cocaine also promotes Tau phosphorylation, 936 highlighting a mechanism that does not rely solely on the robust RSK1 activation observed in the case of HIV. 937 Notably, Tau phosphorylation persists even under conditions of GSK3β inactivation in both HIV- and cocaine-938 exposed systems. This finding reveals the existence of a GSK3β-independent mechanism of Tau modification 939 and establishes RSK1 as a key regulator of Tau phosphorylation under these conditions. The persistence of 940 Tau phosphorylation despite suppression of canonical GSK3β activity suggests the involvement of parallel or 941 compensatory signaling pathways that warrant further investigation. 942 Importantly, we identify RSK1 as a critical upstream regulator of both AKT and GSK3β signaling, exerting 943 positive control over AKT activation while negatively regulating GSK3β activity. This dual regulatory capacity 944 positions RSK1 as a central signaling hub integrating viral and drug-induced pathways that converge on Tau 945 pathology. Collectively, our findings provide mechanistic insight into how HIV and cocaine exposure, through 946 distinct yet convergent pathways, drive Tau dysregulation and contribute to neurotoxicity. From a therapeutic 947 perspective, these results highlight RSK1 as a promising target for intervention, offering a unifying framework 948 for mitigating tauopathy in neuroHIV, HAND, and cocaine-associated neurodegeneration. 949

950 Neuronal cell lines such as SH‑SY5Y are widely used experimental models, yet they have limitations that can 951 compromise experimental robustness [56]. Neurons cells are highly sensitive to culture conditions, exhibit 952 variable growth and differentiation rates, and frequently display batch ‑to‑batch and passage ‑dependent 953 heterogeneity [64]. Such instability poses significant challenges for studies requiring long ‑term culture or 954 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint consistent phenotypic behavior, particularly investigations focused on neurodegeneration, Tau ‑related 955 pathology, or kinase‑driven signaling mechanisms. These constraints highlight the need for neuronal models 956 that retain essential neuron ‑like properties while offering greater stability, reproducibility, and ease of 957 maintenance. More robust and tractable cell systems not only improve experimental consistency but also 958 enable higher ‑throughput analyses and more reliable interpretation of signaling mechanisms relevant to 959 neuropathies, tauopathies, and neurotoxic exposures. 960 In this study, we characterized H80 cells as a stable and experimentally tractable neuronal model and 961 employed them to investigate the molecular mechanisms by which HIV and cocaine promote Tau 962 phosphorylation and neurotoxicity. H80 cells were selected based on their robust proliferation, low cytotoxicity, 963 and stable culture performance. Through immunofluorescence, qPCR, and Western blot analyses, we 964 confirmed that H80 cells express key neuronal markers, including NeuN, MAP2, and Tau, consistent with a 965 mature neuronal phenotype (Figure 1A-C; Supplementary Figure S1). Notably, the expression of MAP2, an 966 axon-associated protein critical for neuronal architecture and implicated in neurodegenerative processes, 967 further supports the neuronal identity of H80 cells. Collectively, the presence of these well -established 968 neuronal markers demonstrates that H80 cells possess essential neuron-like features and extends their utility 969 beyond glioma research. Importantly, given our focus on HIV -induced neurotoxicity, we assessed whether 970 H80 cells express key HIV receptors and co-receptors. Our results show that H80 cells do not express CD4 971 or CCR5, the canonical receptor and major co -receptor for HIV entry, respectively. However, H80 does 972 express one of the HIV co -receptors, CXCR4, in approximately 20% of cells ( Figure 1D). This expression 973 profile is consistent with previous reports indicating that neurons lack CD4 but express chemokine receptors 974 such as CXCR4 and CCR5. Prior studies, including those by Kaul and colleagues, have demonstrated that 975 although due to the absence of HIV receptor, neurons are not productively infected by HIV, they remain highly 976 susceptible to HIV-induced injury mediated through chemokine receptors and downstream signaling cascade, 977 with CXCR4 serving as a key mediator of neurotoxic signaling in neurons [17, 57]. The absence of CCR5 and 978 the selective expression of CXCR4 in H80 cells therefore provide a unique and focused system to investigate 979 CXCR4-dependent mechanisms of HIV-associated neuronal stress. This receptor profile enables us to dissect 980 how HIV exposure perturbs neuronal signaling in the absence of CCR5 -mediated protective pathways, 981 thereby facilitating a clearer understanding of HIV-induced neurotoxicity. 982 Our study identifies a previously unrecognized mechanistic link between viral exposure and neuronal stress 983 pathways. Using an integrated approach combining transcriptional analysis, immunofluorescence, and 984 biochemical analyses, we demonstrate that exposure to HIV virions robustly activates inflammatory signaling 985 cascades and induces neurotoxic responses in H80 neuronal cells. Specifically, HIV exposure significantly 986 upregulates the transcripts of pro-inflammatory cytokines IL-1β and TNF-α (Figure 2B), revealing a previously 987 underappreciated neuron-intrinsic inflammatory response. These findings indicate that direct interaction with 988 viral particles is sufficient to trigger canonical neuroinflammatory programs in neuronal cells. Given that IL-1β 989 and TNF-α are well-established mediators of neuronal injury in both HIV-associated neurocognitive disorders 990 (HAND) and Alzheimer’s disease, our results highlight a shared inflammatory axis between virally induced 991 and classical neurodegenerative processes, characterized by proinflammatory cytokines and pathologic Tau 992 phosphorylation. Although prior studies have primarily attributed HIV-induced cytokine production to microglia 993 [65, 66] , emerging evidence suggests that neurons can also produce cytokines that modulate synaptic 994 function and central nervous system homeostasis, a fact further strengthened using our novel neuronal model 995 system. [67, 68]. The robust cytokine induction observed here further supports effective exposure of H80 cells 996 to HIV virions and underscores the capacity of neurons to directly engage in inflammatory signaling. 997 Importantly, our biochemical analyses reveal that HIV-exposed H80 cells exhibit concurrent increases in RSK1 998 protein levels and Tau phosphorylation at Ser396 (p -Tau S396), with quantitative immunoblotting 999 demonstrating a strong correlation between these events (Figures 2C-E). In contrast, total Tau protein levels 1000 were only modestly altered, indicating that HIV primarily drives post -translational modification of Tau rather 1001 than increasing its overall expression and abundance. These findings support a model in which HIV-induced 1002 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint RSK1 activation directly promotes pathogenic Tau phosphorylation. Given that phosphorylation at Ser396 is 1003 closely associated with Tau protein aggregation and neuronal toxicity, our data establishes RSK1 as a critical 1004 effector linking HIV exposure to early tauopathic signaling. The simultaneous induction of inflammatory 1005 cytokines and RSK1 further suggests coordinated activation of inflammatory and stress -responsive kinase 1006 pathways, creating a signaling environment that may initially be adaptive but ultimately becomes maladaptive. 1007 As a downstream effector of MAPK signaling, RSK1 plays key roles in transcriptional regulation and cellular 1008 stress responses; however, its sustained activation appears to drive pathological Tau modification. 1009 Collectively, these findings identify RSK1 as a central molecular node connecting HIV-induced inflammatory 1010 signaling to Tau pathology. By delineating this pathway, our study provides new mechanistic insight into how 1011 HIV exposure can initiate and accelerate neurodegenerative processes within the central nervous system. 1012 In addition to the effects induced by HIV, our data demonstrate that cocaine also independently activates the 1013 RSK1 signaling cascade to drive Tau phosphorylation, revealing a shared, previously unappreciated kinase 1014 dependency underlying both viral - and drug-induced neurotoxicity underscoring the convergence of drug s 1015 and HIV‑mediated stress responses on a shared kinase pathway. Chronic cocaine exposure triggered robust 1016 phosphorylation of RSK1 at key regulatory residues (Thr348, Thr359/Ser363, and Ser380), accompanied by 1017 a pronounced increase in Tau phosphorylation at Ser396, without any significant change in total Tau protein 1018 levels (Figures 3B–D). These findings establish that cocaine drives Tau pathology primarily through post -1019 translational mechanisms rather than altering Tau expression via transcriptional or translational regulation. 1020 Strikingly, RSK1 activation was both rapid and highly sensitive; even a brief 15 -minute exposure to cocaine 1021 was sufficient to induce multi -site phosphorylation ( Figures 3E –G). A comparable activation profile was 1022 observed following acute HIV exposure, positioning RSK1 as an immediate and convergent sensor of diverse 1023 neurotoxic stimuli. Functional interrogation of RSK1 in mediating Tau phosphorylation unequivocally 1024 establishes RSK1 as essential for Tau phosphorylation. Both pharmacological inhibition and genetic ablation 1025 of RSK1 completely abolished Tau -Ser396 phosphorylation induced by either HIV or cocaine ( Figure 6), 1026 demonstrating that RSK1 is not merely correlative but a required driver of this process. Notably, these findings 1027 add a new layer to the prevailing paradigm that GSK3β is the dominant Tau kinase, instead identifying RSK1 1028 as the principal effector of Tau phosphorylation under conditions of HIV and cocaine exposure. Despite distinct 1029 upstream dynamics, HIV and cocaine converge on a common downstream outcome, pathological Tau 1030 phosphorylation, through a shared RSK1 -centered signaling axis. HIV elicited more robust RSK1 activation 1031 than cocaine; however, both stimuli produced comparable levels of Tau phosphorylation, indicating that RSK1 1032 activity, rather than upstream signal intensity, dictates the pathological output. Importantly, combined HIV and 1033 cocaine exposure failed to produce additive or synergistic effects, instead reaching a plateau consistent with 1034 saturation of a shared signaling pathway. Collectively, these findings redefine the molecular framework of Tau 1035 dysregulation by establishing RSK1 as a central and dominant integrator of viral and drug -induced neuronal 1036 stress. By orchestrating Tau phosphorylation, RSK1 provides a common node through which diverse 1037 upstream perturbations converge on a common pathological outcome. The rapid activation of RSK1 following 1038 acute exposure further suggests that it may function as an early sensor of neuronal stress, initiating 1039 downstream signaling cascades that culminate in Tau pathology. The rapid activation kinetics further suggest 1040 that RSK1 functions as an early molecular sentinel that initiates downstream tauopathic cascades. The 1041 convergence of HIV and cocaine on this shared node provides a mechanistic explanation for the heightened 1042 vulnerability to neurodegeneration observed in individuals exposed to either insult, particularly in the context 1043 of neuroHIV, where comorbid stimulant use is widespread and associated with accelerated cognitive decline. 1044 The identification of RSK1 as a unifying mechanistic driver provides a compelling framework for understanding 1045 how these interactions may arise and highlights a promising therapeutic target for mitigating Tau pathology 1046 across diverse neurotoxic contexts. 1047 An unexpected and conceptually important observation emerged from our investigation, an independent 1048 exposure to either HIV or cocaine consistently resulted in inactivation of GSK3 β ( an increase in 1049 phosphorylation at S9), a kinase implicated in driving Tau phosphorylation and neurodegenerative pathology. 1050 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint This suppression of GSK3 β activity was evidenced by a robust increase in its inhibitory phosphorylation at 1051 Ser9, yet Tau remained persistently hyperphosphorylated under these same conditions (Figure 4 and Figure 1052 6). This strongly suggests that, Tau phosphorylation proceeds through GSK3 β‑independent mechanisms, 1053 thereby elevating, RSK-1. Our findings highlight RSK1 as the dominant kinase responsible for maintaining 1054 Tau phosphorylation when GSK3β is rendered inactive. Our data reveal a remarkably consistent pattern, both 1055 HIV and cocaine suppress GSK3β activity while Tau phosphorylation remains elevated (Figure 4 and Figure 1056 6). This divergence between upstream signaling and Tau modification further supports a model in which RSK1 1057 activation becomes the primary driver of Tau ‑Ser396 phosphorylation under HIV and cocaine exposure. 1058 Importantly, this signaling paradigm proved highly reproducible across multiple experimental conditions. We 1059 observed the same pattern of GSK3 β inactivation coupled with persistent Tau phosphorylation during acute 1060 (Figure 4C, 4D, 4E, 4F and 6E ) as well as chronic cocaine and HIV exposure (Figure 4G, 4H, 6A, 6B, 6C 1061 and 6D). HIV and cocaine suppress GSK3β and act through RSK1 to drive Tau phosphorylation, positioning 1062 RSK1 as a central mediator of their shared pathogenic effects. Altogether, our findings demonstrate that both 1063 cocaine and HIV decrease GSK3 β activity and instead engage RSK1 to drive Tau phosphorylation. These 1064

identify RSK1 as a central signaling node through which cocaine and HIV converge to promote shared 1065 pathogenic mechanisms. 1066 Further mechanistic dissection revealed a clear divergence in how cocaine and HIV regulate upstream kinase 1067 signaling. Cocaine, but not HIV, robustly activated the AKT pathway, as evidenced by increased 1068 phosphorylation at Thr308 and Ser473, two critical residues required for full catalytic activation ( Figure 5). 1069 This activation was accompanied by a corresponding increase in inhibitory phosphorylation of GSK3β at Ser9, 1070 confirming that cocaine suppresses GSK3 β through a canonical AKT-dependent pathway. In contrast, HIV 1071 exposure failed to induce measurable AKT activation, indicating that HIV -mediated inhibition of GSK3 β 1072 proceeds via an AKT -independent mechanism. Instead, HIV selectively upregulates and activates RSK1, 1073 which in turn drives GSK3 β inactivation. Thus, while both stimuli converge on GSK3 β inactivation and Tau 1074 hyperphosphorylation, they do so through distinct upstream routes, cocaine engaging both AKT and RSK1, 1075 and HIV relying predominantly on RSK1. These findings support a bifurcated signaling model in which cocaine 1076 and HIV converge on a shared downstream outcome, GSK3 β inhibition and Tau hyperphosphorylation, yet 1077 reach this endpoint through mechanistically distinct routes. Cocaine engages both AKT and RSK1, thereby 1078 broadly amplifying kinase signaling networks, whereas HIV bypasses AKT entirely and relies predominantly 1079 on RSK1 activation. This distinction provides important biological insight: cocaine simultaneously activates 1080 survival-associated (AKT) and stress -responsive (RSK1) pathways, while HIV exerts a more targeted yet 1081 potent effect through selective RSK1 induction. 1082 Pharmacological and genetic perturbation studies further establish RSK1 as the central regulator of this 1083 signaling architecture. Inhibition of RSK1 with BI -D1870 markedly reduced Tau -Ser396 phosphorylation 1084 despite restoration of GSK3β activity, demonstrating that RSK1, not GSK3β, is the primary kinase sustaining 1085 Tau phosphorylation under both cocaine and HIV exposure (Figure 6). Conversely, inhibition of GSK3β with 1086 CHIR-99021 had no effect on RSK1 activation, confirming that RSK1 operates upstream of GSK3 β in this 1087 hierarchy. These findings were further validated by CRISPR -Cas9-mediated knockout of RSK1, which 1088 abolished Tau phosphorylation, reactivated GSK3 β, and reduced both AKT phosphorylation and total AKT 1089 levels (Figure 7). Notably, the reduction in AKT abundance following RSK1 depletion suggests that RSK1 1090 contributes to AKT stabilization and activation, placing it at the apex of a coordinated kinase network. 1091 Collectively, these results define a unified signaling paradigm in which RSK1 functions as a central integrator 1092 linking HIV and cocaine exposure to pathological Tau hyperphosphorylation. This RSK1 -driven mechanism 1093 operates independently of GSK3β and, in the case of cocaine, is further reinforced by AKT activation, thereby 1094 integrating distinct upstream perturbations into a common pathological outcome. The convergence of viral 1095 and drug-induced signaling on RSK1 provides a mechanistic explanation for the heightened vulnerability to 1096 Tau-associated neurodegeneration observed in neuroHIV, particularly in the context of stimulant drug use. 1097 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Importantly, the demonstration that pharmacological inhibition of RSK1 reverses Tau hyperphosphorylation 1098 and restores kinase balance underscores its translational potential. These findings position RSK1 as a 1099 tractable and previously underappreciated therapeutic target for mitigating Tauopathy in HIV -associated 1100 neurocognitive disorders (HAND), as well as in broader neurodegenerative conditions such as Alzheimer’s 1101 disease. Future studies in preclinical models, including humanized mouse models, will be essential to 1102 determine whether targeting RSK1 can attenuate neuroinflammation, prevent Tau pathology, and ultimately 1103 slow or halt neurodegenerative progression. 1104 Prior studies have shown that the Tat protein of HIV can activate GSK3 β and contributes to Tat-mediated 1105 neurotoxicity [19, 69, 70] . However, in this study we examined neuronal responses to intact, replication -1106 competent HIV virions (HIV-1 strain 93/TH/051; dual-tropic, R5/X4), thereby capturing the integrated effects 1107 of the full viral particle without infection and replication. Notably, both acute and chronic exposure to these 1108 dual-tropic virions consistently resulted in functional inactivation of GSK3 β, as evidenced by increased 1109 phosphorylation at the inhibitory Ser9 site ( Figures 4 and 6). These observations likely reflect the complex 1110 composition of intact virions, which contain multiple structural and accessory proteins capable of exerting both 1111 activating and inhibitory influences on intracellular kinase networks, ultimately shifting the balance toward 1112 GSK3β suppression. Strikingly, cocaine exposure produced a similar biochemical signature, robust Ser9 1113 phosphorylation and inactivation of GSK3 β. This convergence suggests that HIV and cocaine, despite 1114 engaging distinct upstream signaling pathways, AKT-dependent in the case of cocaine and AKT-independent 1115 for HIV, ultimately suppress GSK3 β through a shared downstream mechanism. Our data identify RSK1 as 1116 the central integrator of this process, coordinating GSK3 β inhibition and sustaining Tau phosphorylation. 1117 These findings indicate that virion-mediated effects are not merely additive but instead converge on a common 1118 RSK1-driven intracellular signaling axis that governs neuronal stress responses and Tau pathology. 1119 Future studies will focus on defining the specific viral determinants within the intact virion that initiate RSK1 1120 activation, as well as identifying the neuronal receptors involved, including the potential role of CXCR4 -1121 mediated signaling. Elucidating the precise regulatory sites on RSK1 that mediate downstream suppression 1122 of GSK3 β, particularly those governing Ser9 -directed inhibitory phosphorylation, while simultaneously 1123 sustaining Tau phosphorylation at Ser396 will be critical for resolving the hierarchical organization of this 1124 signaling network. In parallel, comparative analyses of cocaine- and HIV-mediated pathways will be essential 1125 to determine whether these distinct stimuli converge on shared upstream sensors or utilize overlapping 1126 signaling modules. Such investigations will help establish whether a common molecular node integrates viral 1127 and environmental stressors to modulate neuronal vulnerability. 1128 Importantly, our findings identify RSK1 as a major effector of HIV -induced Tau phosphorylation at Ser396, 1129 establishing a direct mechanistic link between viral exposure and tauopathic processes ( Figure 2 ). HIV 1130 exposure activates inflammatory signaling cascades, leading to upregulation of RSK1 and subsequent 1131 phosphorylation of Tau at Ser396, a modification strongly associated with neurofibrillary tangle formation in 1132 Alzheimer’s disease and HAND. The concurrent induction of pro-inflammatory cytokines, including IL-1β and 1133 TNF-α, further implicates inflammatory stress as a key upstream driver of this pathway. Collectively, these 1134

support a model in which HIV -induced inflammatory and stress -responsive signaling converge on 1135 RSK1 to drive Tau pathology. This RSK1 -centered mechanism provides a unifying framework linking viral 1136 exposure, neuroinflammation, and neurodegeneration, and offers important insight into how HIV infection may 1137 initiate or accelerate Tau-mediated neuronal dysfunction within the CNS. 1138 In summary, our findings support a unified signaling network in which RSK1 functions as a central molecular 1139 node linking both HIV and cocaine exposure to Tau hyperphosphorylation. We show that Tau phosphorylation 1140 can be sustained through a GSK3 β independent mechanism under conditions of HIV and cocaine induced 1141 stress, extending current models of Tau regulation. In the context of cocaine exposure, our data indicates the 1142 presence of a dual axis signaling architecture in which RSK1 cooperates with AKT, integrating stress 1143 responsive and survival signaling pathways into a convergent downstream outcome. Collectively, these 1144 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint

indicate that RSK1 is not merely associated with Tau dysregulation but plays a functional role in 1145 mediating Tau phosphorylation in the setting of HIV and cocaine exposure. 1146 Through complementary genetic and pharmacological approaches, we demonstrate that RSK1 acts as a 1147 dominant and convergent driver of Tau pathology, redefining the hierarchical organization of kinase signaling 1148 networks that govern Tau modification. These findings fill a critical gap in our understanding of how diverse 1149 upstream insults, viral infection, and substance abuse , converge on shared molecular pathways to drive 1150 neurodegeneration. The significance of this work lies in its ability to bridge mechanistic and clinical 1151 observations. HIV infection and stimulant use are independently associated with accelerated cognitive 1152 decline, yet the molecular basis of their interaction has remained poorly understood. By identifying RSK1 as 1153 a unifying signaling hub, our study provides a mechanistic framework that explains how these factors 1154 synergistically promote Tauopathy in HIV -associated neurocognitive disorders (HAND) and related 1155 neurodegenerative conditions, including Alzheimer’s disease (AD). This conceptual advance establishes a 1156 new foundation for investigating the intersection of neuroHIV and substance abuse –associated 1157 neuropathology. 1158 Importantly, our findings have immediate translational implications. We demonstrate that pharmacological 1159 inhibition of RSK1 reverses Tau hyperphosphorylation and restores kinase homeostasis, identifying RSK1 as 1160 a tractable and high -value therapeutic target. Targeting RSK1 offers the potential to intercept pathogenic 1161 signaling cascades upstream of irreversible neuronal damage, representing a fundamentally new strategy for 1162 mitigating Tau-driven neurodegeneration. Future studies will focus on evaluating the therapeutic efficacy of 1163 RSK1 inhibition in physiologically relevant preclinical models, including humanized mouse systems, to 1164 determine whether targeting this pathway can attenuate neuroinflammation, suppress Tau pathology, and 1165 preserve neuronal function. Successful validation of this approach has the potential to transform therapeutic 1166 strategies for HAND and other Tau-associated disorders by targeting a shared and central molecular driver 1167 of disease. 1168

1169 The main limitation of the study is that while NeuN, MAP2, and Tau serve as well -established neuronal 1170 markers, future studies should incorporate additional proteins associated with synaptic activity and neuronal 1171 function, such as synaptophysin, neurofilament, and neuron -specific enolase (NSE), to further validate 1172 whether H80 cells exhibit fully functional neuronal behavior. H80 cells also require further characterization to 1173 determine whether they correspond to distinct neuronal lineages, including dopaminergic, glutamatergic, 1174 GABAergic, or cholinergic neurons. Moreover, it remains unclear whether neuronal protein expression in H80 1175 cells arises from intrinsic differentiation potential, genetic reprogramming, or adaptive responses to the tumor 1176 microenvironment. Elucidating these mechanisms will be essential for defining the broader biological 1177 significance of our observations. However, despite limitations, our study provides compelling evidence that 1178 H80 cells possess neuronal lineage features, as demonstrated by the expression of NeuN, MAP2, and Tau. 1179 These findings expand the characterization of H80 cells and underscore their potential as a hybrid model 1180 system at the intersection of glioma biology and neurodegenerative research. Nevertheless, some of the 1181 salient findings have been confirmed in well -established neuronal models, such as SH -SY5Y neuronal cell 1182 line, and 3D models, such as spheroid and organoids containing either neuronal cell line (SHSY5Y) or iPSCs-1183 derived neurons, respectively, which substantially enhance the rigor and robustness of the findings. 1184

1185 These findings collectively support a unified model in which RSK1 functions as a central signaling hub 1186 integrating diverse upstream perturbations into a common downstream outcome, pathological Tau 1187 phosphorylation. The convergence of HIV- and cocaine-induced signaling on RSK1 provides a mechanistic 1188 framework for understanding how viral infection and substance abuse jointly exacerbate neurodegenerative 1189 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint processes, particularly in the context of neuroHIV. Importantly, this RSK1 -driven mechanism appears to 1190 operate independently of GSK3 β, indicating the presence of an additional regulatory layer beyond the 1191 established role of GSK3β in Tau phosphorylation. In the context of cocaine exposure, this pathway is further 1192 reinforced through coordinated activation of AKT, supporting a dual -axis signaling network that integrates 1193 survival and stress -responsive pathways into a shared pathological outcome. This study provides direct 1194 evidence that RSK1 is not merely associated with but is functionally required for Tau phosphorylation in the 1195 setting of viral and substance -induced neuronal stress. By identifying RSK1 as a dominant and convergent 1196 driver of Tau pathology, our work redefines the regulatory hierarchy of Tau -directed kinase signaling and 1197 uncovers a critical, previously underappreciated mechanism underlying neurodegeneration. From a 1198 translational perspective, the identification of RSK1 as a dominant and druggable driver of Tau pathology has 1199 important implications. The ability of RSK1 inhibition to reverse Tau hyperphosphorylation and restore kinase 1200 balance suggests that targeting this pathway may offer a viable therapeutic strategy for mitigating Tauopathy 1201 in HIV -associated neurocognitive disorders, as well as in broader neurodegenerative conditions. More 1202 broadly, these results provide mechanistic clarity to longstanding clinical observations linking HIV infection 1203 and stimulant use with accelerated cognitive decline, and position RSK1 as a promising point of intervention 1204 for preventing or slowing neurodegeneration. Future studies will focus on evaluating whether pharmacologic 1205 inhibition of RSK1 in physiologically relevant preclinical models, including humanized mouse systems, can 1206 attenuate neuroinflammation, suppress Tau pathology, and preserve neuronal function. Such investigations 1207 will be critical for establishing the therapeutic viability of RSK1-targeted interventions and may ultimately pave 1208 the way for novel treatment strategies aimed at preventing or slowing neurodegeneration in HAND and related 1209 Tau-associated disorders. 1210

1211 We thank the AIDS Research and Reagent Program, Division of AIDS, National Institute of Allergy, and 1212 Infectious Diseases, US National Institutes of Health. We thank Dr. Jonathan Karn and his laboratory for 1213 providing the C20 human microglial cell line. We acknowledge the NIH HIV Reagent Program (Division of 1214 AIDS, NIAID, NIH) and Dr. Douglas Richman for providing MT 4 cells (ARP 120). This study utilized services 1215 offered by core facilities of Thomas Jefferson University (FACS and Imaging) and the Comprehensive 1216 NeuroHIV Center (CNHC) at Temple University Lewis Katz School of Medicine. Moreover, we would like to 1217 thank the Center for Translational Medicine, Thomas Jefferson University, including all staff members for their 1218 technical support and assistance in conducting the experiments for this study. 1219 Funding 1220 Research reported in this publication was supported by the National Institutes of Health under Award Number 1221 R01DA041746 and 1R21MH126998 -01A1 to M.T.; Institutional TJU grant (908107) to M.T. The content is 1222 solely the responsibility of the authors and does not necessarily represent the official views of the National 1223 Institutes of Health. 1224 Authors’ Contribution 1225 Conceptualization, A.L.S. and M.T.; methodology, A.L.S. and I.K.S.; software, A.L.S., I.K.S., and M.T.; 1226 validation, A.L.S., I.K.S., and M.T.; formal analysis, A.L.S., I.K.S., U.P.N., and M.T.; investigation, A.L.S., I.K.S., 1227 U.P .N., and M.T.; data curation, A.L.S., I.K.S., U.P.N., and M.T.; writing—original draft preparation and review, 1228 A.L.S., I.K.S., U.P.N., and M.T.; project supervision and funding acquisition, M.T.; all authors have read and 1229 approved the final version of the manuscript. 1230 Declaration of interests 1231 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint The authors declare no competing interests 1232 Ethics statement 1233 No ethical approval required 1234 Generative AI statement 1235 The authors declare that no generative AI was used in the creation of this manuscript 1236 Resource availability 1237 Lead contact 1238 Requests for further information and resources should be directed to and will be fulfilled by the lead contact, 1239 Mudit Tyagi ([email protected]). 1240

availability 1241 This study did not generate new unique reagents. 1242 Data and code availability 1243 • This paper does not report original code. 1244 • Any additional information required to reanalyze the data reported in this paper is available from the lead 1245 contact upon request. 1246 Key resources table 1247 REAGENT or RESOURCE SOURCE IDENTIFIER Antibodies p-RSK-1 (pS380.20A) Santa Cruz Biotechnology sc-136476 p-RSK-1 (Thr 348) Santa Cruz Biotechnology sc-101770 Phospho-p90RSK (Thr359/Ser363) Antibody Cell signaling technology #9344 RSK1 Antibody Cell signaling technology #9333 RSK1/RSK2/RSK3 (D7A2H) Rabbit mAb Cell signaling technology #14813 Phospho-GSK-3β (Ser9) (D85E12) XP® Rabbit mAb Cell signaling technology #5558 GSK-3β (D5C5Z) XP® Rabbit mAb Cell signaling technology #12456 Phospho-Akt (Thr308) (244F9) Rabbit mAb Cell signaling technology #4056 Phospho-Akt (Ser473) (D9E) XP® Rabbit mAb Cell signaling technology #4060 Akt1 (C73H10) Rabbit mAb Cell signaling technology #2938 Phospho-Tau (Ser396) (PHF13) Mouse mAb Cell signaling technology #9632 Tau (D1M9X) XP® Rabbit mAb Cell signaling technology #46687 Anti-HIV1 p55 + p24 + p17 antibody Abcam ab63917 MAP2 Polyclonal antibody proteintech 17490-1-AP Alexa Fluor 647 Anti-RBFOX3 (NeuN) BioLegend 608453 Alexa Fluor 647 anti-Tau phosphor (Ser396) BioLegend 829005 Anti-Tau BioLegend 806701 Actin antibody Santa Cruz Biotechnology sc-47778 GAPDH antibody Santa Cruz Biotechnology sc-25778 Alexa Fluor 488 goat anti-rabbit Invitrogen A11008 Alexa Fluor 568 goat anti-mouse Invitrogen A11004 IRDye 680RD Li-cor (Lincoln, NE, USA) Cat# 926-68071; RRID: AB_10956166 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint IRDye 680LT Li-cor (Lincoln, NE, USA) Cat# 926-68022; RRID: AB_10715072 IRDye 800CW Li-cor (Lincoln, NE, USA) Cat# 926-32211; RRID: AB_621843 APC anti-human CD4 BioLegend Cat# 317416 PE anti-human CD184 (CXCR4) BioLegend Cat# 306505 PE anti-human CD4 BioLegend Cat# 357403 APC/Cyanine& anti0human CD195 (CCR5) BioLegend Cat# 359110 Chemicals Cocaine NIH Lipofectamine 2000 Reagent Invitrogen 11668027 Trizma Base Sigma-Aldrich T1503 Glycine Fisher Chemical G46-1 Sodium chloride Fisher Chemical S271-1 Sodium dodecyl sulfate Bio-Rad 161-0302 Acrylamide Fisher Chemical O1065-500 Bis-acrylamide Hoefer GR142-100 EDTA Sigma-Aldrich E6758-500G Potassium chloride Sigma-Aldrich P9541-1KG BSA (Fraction V) RPI Research Products A30075-100.0 Ammonium persulfate Fisher Chemical S25178 TEMED Fisher Chemical BP150-20 Pierce BCA Reagent A Thermo Scientific 23228 Pierce BCA Reagent B Thermo Scientific 23224 2-Mercaptoethanol Sigma-Aldrich M3148-250ML 1-Butanol Fisher Chemical A399-500 RPMI-1640 Medium (1X) Cytiva HyClone Laboratories SH30027.02 Penicillin-Streptomycin Gibco 15140-122 Fetal Bovine Serum Gibco 10082147 Nonidet P-40 Substitute Sigma-Aldrich 74385-1L Triton X-100 Sigma-Aldrich T8787 DL-Dithiothreitol Sigma-Aldrich D0632-1G HEPES Buffer Corning 25-060-Cl PMSF Thermo Fisher Scientific 36978 PageRuler Prestained Ladder Thermo Fisher Scientific 26617 Nitrocellulose blotting membrane Amersham 10600006 RNeasy Plus Mini Kit Qiagen 74134 High-Capacity cDNA Reverse Transcription Kit Thermo Fisher Scientific 4374967 Hoechst 33342, Invitrogen H1399 ProLong glass Antifade Mountant Invitrogen P36980 Anti-Adherence Rinsing solution Stemcell Technologies 07010 Inhibitors BI-D1870 Selleckhem S2843 CHIR 99021 Tocris 4423 Cell lines H80 cell line A gift Jurkat cell line ATCC TIB-152 MT-4 cell line NIH AIDS reagent ARP-120 U937 cell line ATCC CRL-1593.2 HEK293T cell line ATCC CRL-3216 SH-SY5Y cell line ATCC CRL-2266 Microglial cell line (C20) A gift Virus HIV replication-competent virus (HIV-1 strain 93/TH/051; R5- and X4-tropic virus isolated from a seropositive individual in Thailand) NIH AIDS reagent ARP-2165 Software Prism 9 GraphPad Version 9.1.2 Odyssey Infrared Imaging LI-COR Version 3.0.30 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint ImageJ NIH Version 1.53e Biorender Others 1.5 mL microcentrifuge tubes Caplugs Evergreen 214-3721-010 Biofloat 3D cell culture plates Sarstedt 83.3925.400 Eppendorf Centrifuge 5810R Eppendorf 5811000015 Sorvall™ Legend™ Micro 21R Microcentrifuge Thermo Scientific™ 75002447 VWR Analog Heat Block VWR 12621-104 CHROMATE-4300-N Awareness Technology 4300 Mini Trans-Blot Transfer Cell Bio-Rad 1703930 EVOS M7000 Imaging System Thermo Fisher Scientific AMF7000 1248

1249 1. De Cock KM, Jaffe HW, Curran JW: Reflections on 40 Years of AIDS . Emerg Infect Dis 2021, 1250 27(6):1553-1560. 1251 2. Swinton MK, Sundermann EE, Pedersen L, Nguyen JD, Grelotti DJ, Taffe MA, Iudicello JE, Fields JA: 1252 Alterations in Brain Cannabinoid Receptor Levels Are Associated with HIV -Associated 1253 Neurocognitive Disorders in the ART Era: Implications for Therapeutic Strategies Targeting the 1254 Endocannabinoid System. Viruses 2021, 13(9). 1255 3. Irollo E, Luchetta J, Ho C, Nash B, Meucci O: Mechanisms of neuronal dysfunction in HIV -1256 associated neurocognitive disorders. Cell Mol Life Sci 2021, 78(9):4283-4303. 1257 4. Heaton RK, Clifford DB, Franklin DR, Jr., Woods SP, Ake C, Vaida F, Ellis RJ, Letendre SL, Marcotte 1258 TD, Atkinson JH et al : HIV-associated neurocognitive disorders persist in the era of potent 1259 antiretroviral therapy: CHARTER Study. Neurology 2010, 75(23):2087-2096. 1260 5. Marino J, Maubert ME, Mele AR, Spector C, Wigdahl B, Nonnemacher MR: Functional impact of 1261 HIV-1 Tat on cells of the CNS and its role in HAND. Cell Mol Life Sci 2020, 77(24):5079-5099. 1262 6. Sil S, Hu G, Liao K, Niu F, Callen S, Periyasamy P, Fox HS, Buch S: HIV-1 Tat-mediated astrocytic 1263 amyloidosis involves the HIF-1alpha/lncRNA BACE1-AS axis. PLoS Biol 2020, 18(5):e3000660. 1264 7. Avila J, Lucas JJ, Perez M, Hernandez F: Role of tau protein in both physiological and 1265 pathological conditions. Physiol Rev 2004, 84(2):361-384. 1266 8. Morris M, Maeda S, Vossel K, Mucke L: The many faces of tau. Neuron 2011, 70(3):410-426. 1267 9. Iqbal K, Liu F, Gong CX, Grundke-Iqbal I: Tau in Alzheimer disease and related tauopathies. Curr 1268 Alzheimer Res 2010, 7(8):656-664. 1269 10. Wang Y, Mandelkow E: Tau in physiology and pathology. Nat Rev Neurosci 2016, 17(1):5-21. 1270 11. Harris RB, Martin RJ: Metabolic response to a specific lipid -depleting factor in parabiotic rats. 1271 Am J Physiol 1986, 250(2 Pt 2):R276-286. 1272 12. Buee L, Bussiere T, Buee-Scherrer V, Delacourte A, Hof PR: Tau protein isoforms, phosphorylation 1273 and role in neurodegenerative disorders. Brain Res Brain Res Rev 2000, 33(1):95-130. 1274 13. Arriagada PV, Growdon JH, Hedley -Whyte ET, Hyman BT: Neurofibrillary tangles but not senile 1275 plaques parallel duration and severity of Alzheimer's disease . Neurology 1992, 42(3 Pt 1):631-1276 639. 1277 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint 14. Braak H, Braak E: Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol 1278 1991, 82(4):239-259. 1279 15. Chen Y , Yu Y: Tau and neuroinflammation in Alzheimer's disease: interplay mechanisms and 1280 clinical translation. J Neuroinflammation 2023, 20(1):165. 1281 16. Zhang J, Zhang Y , Wang J, Xia Y, Zhang J, Chen L: Recent advances in Alzheimer's disease: 1282 Mechanisms, clinical trials and new drug development strategies. Signal Transduct Target Ther 1283 2024, 9(1):211. 1284 17. Kaul M, Garden GA, Lipton SA: Pathways to neuronal injury and apoptosis in HIV -associated 1285 dementia. Nature 2001, 410(6831):988-994. 1286 18. Gonzalez-Scarano F, Martin-Garcia J: The neuropathogenesis of AIDS . Nat Rev Immunol 2005, 1287 5(1):69-81. 1288 19. Maggirwar SB, Tong N, Ramirez S, Gelbard HA, Dewhurst S: HIV-1 Tat-mediated activation of 1289 glycogen synthase kinase-3beta contributes to Tat-mediated neurotoxicity. J Neurochem 1999, 1290 73(2):578-586. 1291 20. Kaul M, Lipton SA: Mechanisms of neuronal injury and death in HIV-1 associated dementia. Curr 1292 HIV Res 2006, 4(3):307-318. 1293 21. Buch S, Yao H, Guo M, Mori T, Su TP , Wang J: Cocaine and HIV -1 interplay: molecular 1294 mechanisms of action and addiction. J Neuroimmune Pharmacol 2011, 6(4):503-515. 1295 22. Sonti S, Tyagi K, Pande A, Daniel R, Sharma AL, Tyagi M: Crossroads of Drug Abuse and HIV 1296 Infection: Neurotoxicity and CNS Reservoir. Vaccines (Basel) 2022, 10(2). 1297 23. Clare K, Park K, Pan Y , Lejuez CW, Volkow ND, Du C: Neurovascular effects of cocaine: relevance 1298 to addiction. Front Pharmacol 2024, 15:1357422. 1299 24. Tyagi M, Weber J, Bukrinsky M, Simon GL: The effects of cocaine on HIV transcription. J Neurovirol 1300 2016, 22(3):261-274. 1301 25. Sahu G, Farley K, El-Hage N, Aiamkitsumrit B, Fassnacht R, Kashanchi F, Ochem A, Simon GL, Karn 1302 J, Hauser KF et al: Cocaine promotes both initiation and elongation phase of HIV-1 transcription 1303 by activating NF-kappaB and MSK1 and inducing selective epigenetic modifications at HIV -1 1304 LTR. Virology 2015, 483:185-202. 1305 26. Sharma AL, Shafer D, Netting D, Tyagi M: Cocaine sensitizes the CD4(+) T cells for HIV infection 1306 by co-stimulating NFAT and AP-1. iScience 2022, 25(12):105651. 1307 27. Tyagi M, Bukrinsky M, Simon GL: Mechanisms of HIV Transcriptional Regulation by Drugs of 1308 Abuse. Curr HIV Res 2016, 14(5):442-454. 1309 28. Sharma AL, Tyagi P , Khumallambam M, Tyagi M: Cocaine-Induced DNA-Dependent Protein Kinase 1310 Relieves RNAP II Pausing by Promoting TRIM28 Phosphorylation and RNAP II 1311 Hyperphosphorylation to Enhance HIV Transcription. Cells 2024, 13(23). 1312 29. Zicari S, Sharma AL, Sahu G, Dubrovsky L, Sun L, Yue H, Jada T, Ochem A, Simon G, Bukrinsky M 1313 et al: DNA dependent protein kinase (DNA-PK) enhances HIV transcription by promoting RNA 1314 polymerase II activity and recruitment of transcription machinery at HIV LTR. Oncotarget 2020, 1315 11(7):699-726. 1316 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint 30. Naim A, Farooqui AM, Badruddeen, Khan MI, Akhtar J, Ahmad A, Islam A: The Role of Kinases in 1317 Neurodegenerative Diseases: From Pathogenesis to Treatment . Eur J Neurosci 2025, 1318 61(11):e70156. 1319 31. Jiang G, Xie G, Li X, Xiong J: Cytoskeletal Proteins and Alzheimer's Disease Pathogenesis: 1320 Focusing on the Interplay with Tau Pathology. Biomolecules 2025, 15(6). 1321 32. Rankin CA, Sun Q, Gamblin TC: Pre-assembled tau filaments phosphorylated by GSK -3b form 1322 large tangle-like structures. Neurobiol Dis 2008, 31(3):368-377. 1323 33. Rankin CA, Sun Q, Gamblin TC: Tau phosphorylation by GSK -3beta promotes tangle -like 1324 filament morphology. Mol Neurodegener 2007, 2:12. 1325 34. Cross DA, Alessi DR, Cohen P, Andjelkovich M, Hemmings BA: Inhibition of glycogen synthase 1326 kinase-3 by insulin mediated by protein kinase B. Nature 1995, 378(6559):785-789. 1327 35. Sen T, Saha P, Jiang T, Sen N: Sulfhydration of AKT triggers Tau-phosphorylation by activating 1328 glycogen synthase kinase 3beta in Alzheimer's disease . Proc Natl Acad Sci U S A 2020, 1329 117(8):4418-4427. 1330 36. Sayas CL, Avila J: GSK-3 and Tau: A Key Duet in Alzheimer's Disease. Cells 2021, 10(4). 1331 37. Domise M, Didier S, Marinangeli C, Zhao H, Chandakkar P , Buee L, Viollet B, Davies P , Marambaud 1332 P, Vingtdeux V: AMP-activated protein kinase modulates tau phosphorylation and tau pathology 1333 in vivo. Sci Rep 2016, 6:26758. 1334 38. Yoshida H, Goedert M: Phosphorylation of microtubule-associated protein tau by AMPK-related 1335 kinases. J Neurochem 2012, 120(1):165-176. 1336 39. Yoshimura Y, Ichinose T, Yamauchi T: Phosphorylation of tau protein to sites found in Alzheimer's 1337 disease brain is catalyzed by Ca2+/calmodulin -dependent protein kinase II as demonstrated 1338 tandem mass spectrometry. Neurosci Lett 2003, 353(3):185-188. 1339 40. Kirouac L, Rajic AJ, Cribbs DH, Padmanabhan J: Activation of Ras-ERK Signaling and GSK-3 by 1340 Amyloid Precursor Protein and Amyloid Beta Facilitates Neurodegeneration in Alzheimer's 1341 Disease. eNeuro 2017, 4(2). 1342 41. Rawat P , Sehar U, Bisht J, Selman A, Culberson J, Reddy PH: Phosphorylated Tau in Alzheimer's 1343 Disease and Other Tauopathies. Int J Mol Sci 2022, 23(21). 1344 42. Shimamura A, Ballif BA, Richards SA, Blenis J: Rsk1 mediates a MEK -MAP kinase cell survival 1345 signal. Curr Biol 2000, 10(3):127-135. 1346 43. Doehn U, Hauge C, Frank SR, Jensen CJ, Duda K, Nielsen JV, Cohen MS, Johansen JV, Winther BR, 1347 Lund LR et al: RSK is a principal effector of the RAS -ERK pathway for eliciting a coordinate 1348 promotile/invasive gene program and phenotype in epithelial cells. Mol Cell 2009, 35(4):511-522. 1349 44. Vassal M, Cruz AC, Rebelo S, Martins F: Neurons in a Dish: A Review of In Vitro Cell Models for 1350 Studying Neurogenesis. J Neurochem 2026, 170(1):e70344. 1351 45. Jarabo P, de Pablo C, Herranz H, Martin FA, Casas -Tinto S: Insulin signaling mediates 1352 neurodegeneration in glioma. Life Sci Alliance 2021, 4(3). 1353 46. Portela M, Venkataramani V, Fahey-Lozano N, Seco E, Losada -Perez M, Winkler F, Casas-Tinto S: 1354 Glioblastoma cells vampirize WNT from neurons and trigger a JNK/MMP signaling loop that 1355 enhances glioblastoma progression and neurodegeneration. PLoS Biol 2019, 17(12):e3000545. 1356 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint 47. Slanzi A, Iannoto G, Rossi B, Zenaro E, Constantin G: In vitro Models of Neurodegenerative 1357 Diseases. Front Cell Dev Biol 2020, 8:328. 1358 48. Rai SN, Dilnashin H, Birla H, Singh SS, Zahra W, Rathore AS, Singh BK, Singh SP: The Role of 1359 PI3K/Akt and ERK in Neurodegenerative Disorders. Neurotox Res 2019, 35(3):775-795. 1360 49. Pan J, Yao Q, Wang Y, Chang S, Li C, Wu Y, Shen J, Yang R: The role of PI3K signaling pathway 1361 in Alzheimer's disease. Front Aging Neurosci 2024, 16:1459025. 1362 50. Chu E, Mychasiuk R, Hibbs ML, Semple BD: Dysregulated phosphoinositide 3-kinase signaling 1363 in microglia: shaping chronic neuroinflammation. J Neuroinflammation 2021, 18(1):276. 1364 51. Tyagi M, Pearson RJ, Karn J: Establishment of HIV latency in primary CD4+ cells is due to 1365 epigenetic transcriptional silencing and P-TEFb restriction. J Virol 2010, 84(13):6425-6437. 1366 52. Sanjana NE, Shalem O, Zhang F: Improved vectors and genome -wide libraries for CRISPR 1367 screening. Nat Methods 2014, 11(8):783-784. 1368 53. Doench JG, Fusi N, Sullender M, Hegde M, Vaimberg EW, Donovan KF, Smith I, Tothova Z, Wilen C, 1369 Orchard R et al: Optimized sgRNA design to maximize activity and minimize off -target effects 1370 of CRISPR-Cas9. Nat Biotechnol 2016, 34(2):184-191. 1371 54. Richards SA, Dreisbach VC, Murphy LO, Blenis J: Characterization of regulatory events 1372 associated with membrane targeting of p90 ribosomal S6 kinase 1 . Mol Cell Biol 2001, 1373 21(21):7470-7480. 1374 55. Donadoni M, Cakir S, Bellizzi A, Swingler M, Sariyer IK: Modeling HIV-1 infection and NeuroHIV in 1375 hiPSCs-derived cerebral organoid cultures. J Neurovirol 2024, 30(4):362-379. 1376 56. Prisacar M, Esser S, Hausherr M, Karacora B, Vyushkova Y , Eisenacher M, Grugel R, Marcus K, 1377 Eggers B: Systematic Analysis of SH -SY5Y Differentiation Protocols and Neuronal Subtype 1378 Abundance. Cell Mol Neurobiol 2025, 45(1):104. 1379 57. Kaul M, Ma Q, Medders KE, Desai MK, Lipton SA: HIV-1 coreceptors CCR5 and CXCR4 both 1380 mediate neuronal cell death but CCR5 paradoxically can also contribute to protection . Cell 1381 Death Differ 2007, 14(2):296-305. 1382 58. Buckley S, Byrnes S, Cochrane C, Roche M, Estes JD, Selemidis S, Angelovich TA, Churchill MJ: The 1383 role of oxidative stress in HIV-associated neurocognitive disorders. Brain Behav Immun Health 1384 2021, 13:100235. 1385 59. Hanger DP, Hughes K, Woodgett JR, Brion JP, Anderton BH: Glycogen synthase kinase-3 induces 1386 Alzheimer's disease-like phosphorylation of tau: generation of paired helical filament epitopes 1387 and neuronal localisation of the kinase. Neurosci Lett 1992, 147(1):58-62. 1388 60. Lovestone S, Reynolds CH, Latimer D, Davis DR, Anderton BH, Gallo JM, Hanger D, Mulot S, 1389 Marquardt B, Stabel S et al : Alzheimer's disease -like phosphorylation of the microtubule -1390 associated protein tau by glycogen synthase kinase-3 in transfected mammalian cells. Curr Biol 1391 1994, 4(12):1077-1086. 1392 61. Cho JH, Johnson GV: Glycogen synthase kinase 3beta phosphorylates tau at both primed and 1393 unprimed sites. Differential impact on microtubule binding. J Biol Chem 2003, 278(1):187-193. 1394 62. Hooper C, Killick R, Lovestone S: The GSK3 hypothesis of Alzheimer's disease . J Neurochem 1395 2008, 104(6):1433-1439. 1396 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint 63. Hernandez F, Lucas JJ, Avila J: GSK3 and tau: two convergence points in Alzheimer's disease. J 1397 Alzheimers Dis 2013, 33 Suppl 1:S141-144. 1398 64. Forster JI, Koglsberger S, Trefois C, Boyd O, Baumuratov AS, Buck L, Balling R, Antony PM: 1399 Characterization of Differentiated SH -SY5Y as Neuronal Screening Model Reveals Increased 1400 Oxidative Vulnerability. J Biomol Screen 2016, 21(5):496-509. 1401 65. Mamik MK, Hui E, Branton WG, McKenzie BA, Chisholm J, Cohen EA, Power C: HIV-1 Viral Protein 1402 R Activates NLRP3 Inflammasome in Microglia: implications for HIV -1 Associated 1403 Neuroinflammation. J Neuroimmune Pharmacol 2017, 12(2):233-248. 1404 66. Zink WE, Zheng J, Persidsky Y , Poluektova L, Gendelman HE: The neuropathogenesis of HIV -1 1405 infection. FEMS Immunol Med Microbiol 1999, 26(3-4):233-241. 1406 67. Zipp F, Bittner S, Schafer DP: Cytokines as emerging regulators of central nervous system 1407 synapses. Immunity 2023, 56(5):914-925. 1408 68. Hashimoto O, Hepler TD, Tynan A, Torres A, Li JH, Brines M, Tracey KJ, Chavan SS: Central neurons 1409 encode interleukin-1beta signals and mediate stress -induced inflammation. J Exp Med 2026, 1410 223(4). 1411 69. Sui Z, Sniderhan LF, Fan S, Kazmierczak K, Reisinger E, Kovacs AD, Potash MJ, Dewhurst S, Gelbard 1412 HA, Maggirwar SB: Human immunodeficiency virus-encoded Tat activates glycogen synthase 1413 kinase-3beta to antagonize nuclear factor-kappaB survival pathway in neurons. Eur J Neurosci 1414 2006, 23(10):2623-2634. 1415 70. Kehn-Hall K, Guendel I, Carpio L, Skaltsounis L, Meijer L, Al -Harthi L, Steiner JP , Nath A, Kutsch O, 1416 Kashanchi F: Inhibition of Tat-mediated HIV-1 replication and neurotoxicity by novel GSK3-beta 1417 inhibitors. Virology 2011, 415(1):56-68. 1418 1419 Figure Legends 1420 Figure 1: Neuronal characteristics and HIV co -receptor profile of H80 cells. (A) Immunofluorescence 1421 staining of H80 cells for NeuN using specific primary antibodies, with Hoechst (blue) counterstaining and 1422 unstained as controls. NeuN was detected, confirming the neuronal -like phenotype of H80 cells. (B) 1423 Comparative immunofluorescence staining of H80 cells, microglial cells, SH-SY5Y cells (positive control), and 1424 HEK293T cells (negative control) demonstrated that MAP2 expression was restricted to H80 and SH -SY5Y 1425 cells. The inclusion of HEK293T as a negative control and SH -SY5Y as a positive control confirmed assay 1426 specificity. Detection of MAP2 in H80 cells indicates that these cells exhibit features of differentiated neurons. 1427 (C) Immunoblotting verified MAP2 expression in H80 cells but not in HEK293T or microglial cells, further 1428 supporting the neuronal identity of H80 cells. Collectively, these findings demonstrate that H80 cells express 1429 canonical neuronal markers, including NeuN and MAP2, positioning them as a relevant model for investigating 1430 neuron-related molecular mechanisms, implicated in neurodegenerative disease and HIV/drug -induced 1431 neurotoxicity. (D) Flow cytometry analysis of HIV receptor expression on HEK293T, U937, and H80 cells. 1432 Cells were blocked with 2% BSA plus Fc block and co-stained for either CD4 and CXCR4 (APC anti-human 1433 CD4, BioLegend cat. no. 317416; PE anti-human CD184 \[CXCR4], BioLegend cat. no. 306505) or CD4 and 1434 CCR5 (PE anti -human CD4, BioLegend cat. no. 357403; APC/Cyanine7 anti -human CD195 \[CCR5], 1435 BioLegend cat. no. 359110). HEK293T cells (negative control) lacked CD4, CXCR4, and CCR5 expression, 1436 whereas U937 cells (positive control) expressed all three markers. H80 cells did not express CD4 or CCR5 1437 but showed detectable CXCR4 expression. These findings suggest that H80 cells lack main HIV receptor 1438 CD4 and chemokine receptor CCR5 for HIV entry, despite expressing CXCR4. 1439 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Figure 2: HIV exposure induces RSK1 upregulation and promotes Tau phosphorylation in H80 cells. 1440 (A) Immunoblot analysis of total cell lysates from Jurkat T cells infected with replication -competent HIV or 1441 uninfected (Ctrl), confirming HIV infection by detection of the HIV p24 capsid protein. Culture supernatants 1442 from uninfected Jurkat cells (control) and from HIV -infected cells (containing replication competent virion 1443 particles) were collected and used to expose H80 cells via spinoculation (2 h at 1,000 rpm). Cells were 1444 subsequently reseeded, and on the following day subjected to a second round of spinoculation with HIV 1445 virions under the same conditions, followed by seeding onto 100 -mm plates. (B) Twenty-four hours post -1446 exposure, total RNA was extracted from H80 cells and analyzed by reverse transcription –quantitative PCR 1447 (RT–qPCR) for the expression of IL-1β, TNF-α, and RSK1, relative normalized to GAPDH. HIV-exposed cells 1448 showed transcriptional upregulation of IL -1β, TNF-α, and RSK1 relative to No HIV controls. (C) After 48 h 1449 post-exposure, cells were subjected to immunofluorescence staining using antibodies against phosphorylated 1450 Tau at Ser396 and Tau. HIV exposure enhanced increased Tau phosphorylation compared to No HIV exposed 1451 controls. Yellow arrows indicate representative phosphorylation -positive sites in the immunofluorescence 1452 images. The scale bar represents 10 µm. (D–E) To validate these findings, H80 cells were cultured in four 1453 independent dishes (two biological replicates per condition), and whole cell lysates were collected 48 h after 1454 HIV exposure. Protein lysates were prepared separately from each dish and quantified. Equal amounts of 1455 protein were loaded and analyzed by immunoblotting for RSK1, RSK1/2/3, phosphorylated Tau (p-Tau-S396), 1456 and Tau, with actin or total protein as loading controls. Immunoblotting confirmed upregulation of RSK1, 1457 increased phosphorylation of Tau at Ser396, and a moderate elevation in Tau protein levels in HIV-exposed 1458 H80 cells. Densitometric analysis of immunoblots was performed by normalizing band intensities to β actin or 1459 total protein, with values expressed relative to the control (Ctrl/No HIV). (F) To assess whether H80 cells were 1460 susceptible to HIV infection, lysates from HIV‑exposed and No HIV exposed H80 cells were examined for HIV 1461 p24 by immunoblotting, alongside Jurkat T cells included as positive (infected) and negative (uninfected) 1462 controls. Immunoblots are representative of at least three independent biological replicates. Data are 1463 presented as mean ± S.D. Statistical significance was assessed using an unpaired, two tailed Student’s t test. 1464 Significance is indicated as P < 0.05 (*) and P < 0.01 (**). 1465 Figure 3: Cocaine activates and upregulates RSK1 and promotes Tau phosphorylation in H80 cells. 1466 (A) Schematic representation of the protocol for the IF and Immunoblot assay detailing treatment with the 1467 chronic cocaine and HIV exposure (B) Immunofluorescence analysis of H80 cells chronically exposed to 1468 cocaine (twice daily for 2 days) revealed a marked increase in phosphorylated Tau at Ser396 (p-Tau-Ser396) 1469 compared with untreated controls, while Tau levels remained unchanged, indicating that cocaine enhances 1470 Tau phosphorylation without altering overall Tau protein. Yellow arrows indicate representative 1471 phosphorylation-positive sites in the immunofluorescence images. The scale bar represents 10 µm. ( C–D) 1472 Immunoblot analysis of whole -cell lysates from H80 cells treated with cocaine, HIV, or both revealed that 1473 cocaine modestly increased total RSK1 expression and its phosphorylation at Ser380, a marker of catalytic 1474 activation, while minimally affecting Thr348 phosphorylation. HIV exposure produced a robust increase in 1475 total RSK1 and phosphorylation at multiple sites (Ser380, Thr348, Thr359, and Ser363), far exceeding the 1476 effects of cocaine alone. Both cocaine and HIV significantly elevated p -Tau-Ser396 relative to untreated 1477 controls, with HIV exerting a stronger effect. Co-exposure to cocaine and HIV resulted in a higher, though not 1478 strictly additive, increase in Tau phosphorylation, suggesting convergence on overlapping signaling pathways. 1479 Densitometric analysis of immunoblots was performed by normalizing band intensities to β actin or total 1480 protein, with values expressed relative to the control (ctrl/No HIV). (E) Schematic representation of the 1481 protocol for immunoblots with acute cocaine and HIV exposure. (F-G) H80 cells of different passages were 1482 cultured in eight independent dishes (two biological replicates per condition), and whole ‑cell lysates were 1483 collected 15 min after exposure to cocaine, HIV, or the combined treatment. Protein lysates were prepared 1484 separately from each dish and quantified. Equal amounts of protein were loaded and analyzed by 1485 immunoblotting for p -RSK1 S380, p -RSK1 Thr359 and S363, and RSK1, with actin as loading controls. 1486 Immunoblotting confirmed activation of RSK1, increased phosphorylation of RSK1 at S380, Thr359 and S363, 1487 in HIV -exposed H80 cells. Densitometric analysis of immunoblots was performed by normalizing band 1488 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint intensities to RSK1, with values expressed relative to the control (Ctrl/No HIV). Immunoblots are 1489 representative of at least three independent biological replicates. Data are presented as mean ± S.D. 1490 Statistical significance was assessed using one way ANOVA with Dunnett’s multiple comparisons test. 1491 Significance is indicated as P < 0.05 (*) and P < 0.01 (**). Together, these findings implicate RSK1 activation 1492 as a key mediator of cocaine - and HIV-driven Tau phosphorylation, highlighting a shared molecular axis 1493 underlying neurodegenerative processes in the context of substance use and HIV exposure. 1494 Figure 4: HIV and cocaine converge on GSK3β inactivation via Ser9 phosphorylation. (A-B) Immunoblot 1495 analysis of Jurkat T cells infected with HIV or uninfected (control) revealed a marked increase in GSK3 β 1496 phosphorylation at the inhibitory Ser9 site (p-GSK3β-Ser9) upon HIV infection, while GSK3β levels remained 1497 unchanged, indicating functional inactivation of GSK3 β during viral infection in immune cells. (C-D) Acute 1498 exposure of H80 cells (15 min) to supernatants from HIV-infected Jurkat cells induced a pronounced increase 1499 in Ser9 phosphorylation compared to control (uninfected) supernatant, demonstrating that HIV exposure 1500 rapidly modulates host kinase signaling even in non -permissive cells lacking CD4. (E-F) H80 cells were 1501 cultured in eight independent dishes (two biological replicates) under each identical condition and whole-cell 1502 lysates were collected 15 min after cocaine, HIV exposure and cocaine plus HIV exposure. Protein lysates 1503 were prepared separately from each dish and quantified. Equal amounts of protein were loaded and analyzed 1504 by immunoblotting for p-GSK3β S9, and GSK3β. Immunoblotting confirmed inactivation of GSK3β, increased 1505 phosphorylation of GSK3β at S9, in cocaine, HIV exposed and cocaine along with HIV -exposed H80 cells 1506 compared to Ctrl/No HIV. Densitometric analysis of immunoblots was performed by normalizing band 1507 intensities to GSK3 β, with values expressed relative to the control (Ctrl/No HIV). Immunoblots are 1508 representative of at least three independent biological replicates. (G-H) Immunoblot analysis of H80 cells 1509 exposed for 48 h to cocaine (chronic exposure), HIV virions, or both (Schematic representation in Figure 3A) 1510 revealed robust Ser9 phosphorylation under all conditions, while GSK3β levels remained constant, confirming 1511 post-translational regulation rather than changes in protein abundance. Combined exposure to HIV and 1512 cocaine produced an inhibitory effect of GSK3 β similar to individual treatments. Densitometric analysis of 1513 immunoblots was performed by normalizing band intensities to GSK3β, with values expressed relative to the 1514 control (Ctrl/No HIV). Immunoblots are representative of at least three independent biological replicates. Data 1515 are presented as mean ± S.D. Statistical significance was assessed using an unpaired, two tailed Student’s t 1516 test (for B and D) or one way ANOVA with Dunnett’s multiple comparisons test (for F and H). Significance is 1517 indicated as P < 0.05 (*) and P < 0.01 (**). 1518 Figure 5: Cocaine activates AKT signaling in H80 cells, whereas HIV exposure does not activate AKT. 1519 (A) Schematic representation of the protocol for the IF and Immunoblot assay detailing treatment with the 1520 chronic cocaine and HIV exposure (B) Immunofluorescence analysis of H80 cells chronically exposed to 1521 cocaine (twice daily for 2 days) revealed a robust increase in phosphorylated AKT at Ser473 (p-AKT-Ser473) 1522 compared with untreated controls, indicating strong activation of the AKT signaling pathway. Hoechst staining 1523 was used for nuclear visualization. AKT levels remained unchanged, confirming that the observed effect 1524 reflects post -translational regulation rather than changes in protein abundance. Yellow arrows indicate 1525 representative phosphorylation-positive sites in the immunofluorescence images. The scale bar represents 1526 10 µm. HIV exposure alone did not alter p -AKT-Ser473 levels under the same conditions (data in 1527 supplementary). (C) Immunoblot analysis of whole -cell lysates from H80 cells exposed to cocaine, HIV, or 1528 both for 48 h demonstrated that cocaine significantly increased phosphorylation of AKT at both Thr308 and 1529 Ser473, modifications essential for full kinase activation. HIV exposure alone did not affect AKT 1530 phosphorylation, while combined treatment mirrored the effect of cocaine alone, indicating that cocaine exerts 1531 a dominant influence on AKT activation. (D) Densitometric quantification confirmed a significant increase in 1532 AKT phosphorylation at Thr308 and Ser473 in cocaine-treated and HIV+cocaine-treated cells, whereas HIV 1533 alone had no measurable impact. AKT protein levels remained constant across all conditions. Densitometric 1534 analysis of immunoblots was performed by normalizing band intensities to AKT, with values expressed relative 1535 to the control (Ctrl/No HIV). Immunoblots are representative of at least three independent biological replicates. 1536 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Data are presented as mean ± S.D. Statistical significance was assessed using one way ANOVA with 1537 Dunnett’s multiple comparisons test. Significance is indicated as P < 0.01 (**) and ns denotes not significant. 1538 Figure 6: RSK1 functions upstream of GSK3 β to mediate Tau phosphorylation induced by HIV 1539 exposure and cocaine. (A & B ) H80 cells were cultured in twelve independent dishes (three biological 1540 replicates per condition) under each identical condition and whole -cell lysates were collected 48 h after 1541 cocaine, HIV exposure and cocaine plus HIV exposure. Protein lysates were prepared separately from each 1542 dish and quantified. Equal amounts of protein were loaded and analyzed by immunoblotting. Immunoblot 1543 analysis revealed a significant increase in Tau phosphorylation at Ser396 (p -Tau-S396), a marker of 1544 pathological Tau, and RSK-1 expression under all conditions (cocaine, HIV exposure and both) compared 1545 with untreated controls. Concurrent analysis demonstrated enhanced phosphorylation of GSK3β at Ser9 (p-1546 GSK3β-Ser9), indicating functional inactivation of GSK3 β under all conditions, while GSK3 β and Tau levels 1547 remained unchanged. Densitometric analysis of immunoblots was performed by normalizing band intensities 1548 to β actin or total protein, with values expressed relative to the control (Ctrl/No HIV). (C & D) To delineate the 1549 relative contributions of RSK1 and GSK3 β, H80 cells were pretreated for 24 h with selective inhibitors BI -1550 D1870 (RSK1 inhibitor) or CHIR -99021 (GSK3 β inhibitor) prior to HIV and/or cocaine exposure. 1551 Immunoblotting and densitometry revealed that BI -D1870 effectively suppressed RSK1 activation and 1552 significantly reduced Tau phosphorylation, whereas CHIR-99021 failed to alter Tau phosphorylation induced 1553 by HIV or cocaine, confirming that Tau modification is primarily mediated through RSK1-dependent signaling. 1554 Inhibition of RSK1 also reversed GSK3 β inactivation, as evidenced by reduced Ser9 phosphorylation, 1555 suggesting a hierarchical relationship in which RSK1 lies upstream of GSK3 β. Densitometric analysis of 1556 immunoblots was performed by normalizing band intensities to GAPDH, with values expressed relative to the 1557 control (Ctrl/No HIV). (E) An acute time point study was conducted for 1 h, 3 h and 6 h with cocaine and 1558 cocaine along with HIV, Immunoblot results show enhanced phosphorylation of GSK3 β at Ser9 (p-GSK3β-1559 Ser9), indicating functional inactivation of GSK3 β at 6h while simultaneously seen the enhanced tau 1560 phosphorylation at S396. Immunoblots are representative of at least three independent biological replicates. 1561 Data are presented as mean ± S.D. Statistical significance was assessed using one way ANOVA with 1562 Dunnett’s multiple comparisons test. Significance is indicated as P < 0.05 (*) and P < 0.01 (**). 1563 Figure 7: CRISPR -Cas9-mediated RSK1 knockout reveals its essential role in Tau phosphorylation 1564 and AKT-GSK3β signaling. / RSK1 as a critical upstream regulator of Tau phosphorylation and AKT 1565 signaling, functioning hierarchically above GSK3 β and contributing to multiple signaling pathways 1566 relevant to neurodegeneration. (A–B) Immunoblot analysis confirmed successful RSK1 knockout (RSK1 1567 KO) in H80 cells and demonstrated reduced inhibitory phosphorylation of GSK3 β at Ser9 (p-GSK3β-Ser9), 1568 indicating reactivation of GSK3 β kinase activity upon loss of RSK1. RSK1 KO also led to a significant 1569 reduction in Tau phosphorylation at Ser396 (p-Tau-S396) compared with control cells, while total Tau levels 1570 remained unchanged. Despite GSK3 β reactivation, Tau phosphorylation did not recover, suggesting that 1571 RSK1 mediates site -specific Tau phosphorylation independently of GSK3 β. Quantitative analysis from 1572 multiple independent experiments confirmed a significant decrease in p -Tau-S396 in RSK1 KO cells, 1573 establishing RSK1 as essential for efficient Tau phosphorylation. (C-D) Analysis of AKT signaling revealed 1574 that RSK1 knockout severely impaired AKT activation, as evidenced by a marked reduction in phosphorylation 1575 at Thr308 and Ser473, and also decreased total AKT protein levels. These findings indicate that RSK1 1576 positively regulates both AKT activation and AKT protein stability. Densitometric analysis of immunoblots was 1577 performed by normalizing band intensities to β actin or total protein, with values expressed relative to the 1578 control. Immunoblots are representative of at least three independent biological replicates. Data are 1579 presented as mean ± S.D. Statistical significance was assessed using one way ANOVA with Dunnett’s multiple 1580 comparisons test. Significance is indicated as P < 0.05 (*), P < 0.01 (**) and P < 0.001 (***). 1581 Figure 8: RSK1 overexpression modulates AKT and GSK3β signaling in H80 cells. / RSK1 as a central 1582 upstream modulator of AKT and GSK3 β signaling (A–B) Immunoblot analysis confirmed robust 1583 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint overexpression of RSK1 following transient transfection with a CMV-driven RSK1 construct, as evidenced by 1584 a marked increase in total RSK1 protein and elevated levels of RSK isoforms (RSK1/2/3). Overexpression 1585 also resulted in a significant increase in phosphorylation of RSK1 at Ser380, an autophosphorylation site 1586 associated with partial catalytic activation, while phosphorylation at Thr348 decreased relative to empty vector 1587 transfected control. (C–D) RSK1 overexpression induced a robust increase in AKT phosphorylation at both 1588 Thr308 and Ser473, two canonical regulatory sites required for full AKT activation, without altering total AKT 1589 protein levels. These findings indicate that RSK1 positively regulates AKT activity predominantly through post-1590 translational mechanisms. (E–F) Overexpression of RSK1 also enhanced phosphorylation of GSK3β at Ser9, 1591 a canonical inhibitory site that suppresses GSK3 β kinase activity, while total GSK3 β levels remained 1592 unchanged. Densitometric analysis of immunoblots was performed by normalizing band intensities to β actin 1593 or total protein, with values expressed relative to the empty vector transfected control. Immunoblots are 1594 representative of at least three independent biological replicates. Data are presented as mean ± S.D. 1595 Statistical significance was assessed using an unpaired, two tailed Student’s t test. Significance is indicated 1596 as P < 0.01 (**) and ns denotes not significant. 1597 Figure 9: HIV exposure and cocaine induce RSK1 ‑dependent Tau phosphorylation in neuronal 1598 monolayers, 3D spheroid, and brain organoid models. (A) Schematic representation of the immunoblot 1599 experimental workflow illustrating HIV and cocaine exposure in a three‑dimensional spheroid culture system. 1600 (B) SH‑SY5Y neuronal cells were exposed to HIV for 48 h, followed by cell lysis and immunoblot analysis. 1601 HIV exposure resulted in upregulation of RSK1, increased inhibitory phosphorylation of GSK3β at Ser9, and 1602 enhanced phosphorylation of Tau at Ser396. (C) These immunoblot findings were recapitulated in a 3D 1603 multicellular spheroid model composed of H80 neurons, microglia, and SH ‑SY5Y cells, confirming the 1604 reproducibility of HIV ‑induced signaling responses in a heterogeneous cellular context. (D) Immunoblot 1605 analysis in a three ‑dimensional organoid model further validated HIV ‑ and cocaine‑induced activation and 1606 regulation of RSK1, phosphorylation GSK3 β S9, and Tau phosphorylation S396, demonstrating the 1607 robustness of this signaling axis across increasingly complex neuronal systems. 1608 Figure 10: Model summarizing HIV ‑ and cocaine ‑induced Tau phosphorylation in neuronal cells. 1609 Proposed schematic illustrating distinct yet convergent signaling mechanisms by which HIV exposure and 1610 cocaine promote Tau phosphorylation in H80 neuronal cells. HIV exposure induces a robust and sustained 1611 activation and upregulation of RSK1, which drives Tau phosphorylation through a pathway that is independent 1612 of AKT signaling while concurrently promoting inhibitory phosphorylation of GSK3 β at Ser9. In contrast, 1613 cocaine exposure engages a partially overlapping but mechanistically distinct pathway, characterized by 1614 modest RSK1 induction and strong activation of AKT, as evidenced by phosphorylation at Thr308 and Ser473. 1615 Activated AKT subsequently catalyzes inhibitory phosphorylation of GSK3β at Ser9, leading to its functional 1616 inactivation. Despite GSK3β inactivation under both conditions, Tau phosphorylation persists, indicating the 1617 existence of a GSK3β‑independent mechanism regulated by RSK1. Collectively, these findings identify RSK1 1618 as a central signaling hub that integrates viral and substance ‑induced signaling to drive Tau dysregulation, 1619 highlighting its critical role in neurodegenerative processes relevant to HIV ‑associated neurocognitive 1620 disorders. 1621 1622 1623 1624 1625 1626 1627 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Figure1 1628 1629 1630 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Figure 2 1631 1632 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Figure 3 1633 1634 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Figure 4 1635 1636 1637 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Figure 5 1638 1639 1640 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Figure 6 1641 1642 1643 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Figure 7 1644 1645 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Figure 8 1646 1647 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Figure 9 1648 1649 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint Figure 10 1650 1651 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted April 16, 2026. ; https://doi.org/10.64898/2026.04.14.718541doi: bioRxiv preprint