HIV-1 gp120-induced lysosomal stress responses are controlled by TRPML1 redox sensors Authors: Nirmal Kumar, Braelyn Liang, Jonathan D Geiger Published: 2026 ABSTRACT AbstractBackgroundIncreased lysosomal stress responses (LSR) are commonly implicated in the pathogenesis of neurodegenerative disorders including HIV-1-associated neurocognitive disorders (HAND). The HIV-1 envelope glycoprotein gp120 causes LSR, increases levels of ferrous iron (Fe2+) in the cytosol and in mitochondria, disrupts the reactive species interactome (RSI), and increases neural cell death.MethodsUsing SH-SY5Y human neuroblastoma and U87MG human astrocytoma cells treated with gp120 and pharmacological modulators, we evaluated redox signaling and LSR by redox-sensitive fluorescent probes, spinning-disk confocal microscopy, flow cytometry, Western blotting, and immunostaining.ResultsHere, we report that TRPML1, an endolysosome redox-sensitive cation channel, is mechanistically involved in gp120-induced neurotoxicity. TRPML1 was activated by gp120-induced increases in cytosolic reactive oxygen species (ROS) and resulted in release of Fe2+ from endolysosomes in levels sufficient to increase cytosolic levels of Fe2+ and ROS as well as decrease levels of hydrogen sulfide (H2S). Reduced glutathione normally buffers intracellular Fe2+, but gp120 decreased endolysosome glutathione levels and disrupted this regulatory control mechanism thereby promoting TRPML1-mediated Fe2+ efflux from endolysosomes. TRPML1 redox activation led to changes to the RSI in endolysosomes including increased ROS, lipid peroxidation, nitric oxide, and sulfane sulfur as well as decreased H2S. These changes were accompanied by increased cysteine oxidation of luminal proteins and endolysosome deacidification. Pharmacological inhibition of TRPML1 or knocking down expression levels of TRPML prevented these effects.ConclusionOur findings suggest that TRPML1 redox activation controls gp120-induced endolysosome dysfunction and iron/redox imbalance, and further implicates TRPML1 in the pathogenesis of HAND.Keywords: HIV-1 gp120, lysosomal stress responses, reactive species interactome, reactive sulfur species, reactive nitrogen species, reactive oxygen species, reactive carbonyl species, TRPML1, redox sensor Introduction Human immunodeficiency virus-1 (HIV-1) negatively affects brain function prior to HIV-1 seroconversion [1] and causes a broad spectrum of cognitive, motor, and behavioral deficits collectively termed HIV-associated neurocognitive disorders (HAND) [2]. Antiretroviral therapy has now improved the quality of life and longevity of people living with HIV-1 (PLWH) such that PLWH are now living almost full life spans, but the prevalence of HAND has increased and it is now estimated to affect about 50% of PLWH [2,3]. The neuropathogenesis of HAND is complex and associated with HAND are chronic long-lasting neuroinflammation, increased oxidative stress, and neurodegeneration [2]. The HIV-1 envelope glycoprotein gp120 [4] continues to be implicated in the pathogenesis of HAND [5,6]; it is neurotoxic, it causes neuroinflammation, it leads to endolysosome and mitochondrial dysfunction, it increases levels of reactive oxygen species (ROS) and reactive nitrogen species (RNS), it decreases antioxidant defense activity [5–10], and it increases the oxidation of DNA, RNA, proteins, and lipids [11,12]. In addition to increases in ROS in the CNS of HAND patients [12,13] and in neurons treated with gp120 [14], other short-lived reactive molecules may also be involved in HAND [13,15]. The reactive species interactome (RSI) is a relatively new concept of cellular stress signaling that involves reactive sulfur species (RSS), RNS, ROS, and reactive carbonyl species (RCS); the RSI also includes interactions between the various reactive species and their biological targets [16,17]. Elevated levels of RCS were reported in brain tissue and cerebrospinal fluid of HAND patients [15,18–21]. Moreover, increased levels of metal ions that cause redox catastrophe in cells, including iron, have been described in HAND patients and in gp120-treated neurons [22–24]. Endolysosomes are acidic organelles well known for their physiologically and pathologically important roles they play in degradation of metabolites and cellular debris, nutrient sensing, iron storage, redox signaling [23,24], neurodegeneration and HAND [25,26]. Endolysosome alkalization (de-acidification) leads to alternations in iron levels and because endolysosomes have readily releasable stores of Fe2+, endolysosome dysfunction continues to be implicated in the onset of numerous pathological conditions especially those where increased levels of Fe2+ have been implicated 24,27–31. Others and we have shown that gp120 disrupts endolysosome pH, Fe2+, and RSI homeostasis and that the release of Fe2+ from endolysosomes into the cytosol was sufficient to increase levels of cytosolic and mitochondrial Fe2+ and ROS as well as increase neural cell death [7,28,32–34]. Endolysosome-resident transient receptor potential mucolipin 1 (TRPML1) channels transport Ca2+, Fe2+, and Zn2+ divalent cations, and help control endolysosome homeostatic processes, autophagy, redox signaling, and oxidative stress-induced cell death [35,36]. Loss-of-function mutations in the TRPML1 gene leads to mucolipidosis type 4, and reduced TRPML1 protein expression levels or activity can cause neurodegeneration [37–39]. TRPML1 activation can also result in neurotoxic increases in cytosolic Ca2+, Zn2+, and ROS, oxidative stress-induced neuronal injury, and cell death [39–43]. Recent studies have demonstrated that TRPML1 can be directly activated by ROS and act as a redox sensor in endolysosomes [36,41]. Thus, activation of TRPML1 activity or increased protein expression levels of TRPML1, possibly through redox activation, may contribute to gp120-induced neurotoxicity and the pathogenesis of neurodegenerative disorders. Here, we determined the extent to which redox sensing of TRPML1 channels controls gp120-induced endolysosome dysfunction and disruption of iron and redox homeostasis. We demonstrate that TRPML1 redox activation links gp120 to endolysosome Fe2+ release, RSI disruption, protein cysteine oxidation, and impaired endolysosome acidification. Inhibition of TRPML1 prevented these effects thus highlighting TRPML1 as a promising therapeutic target to restore iron and redox homeostasis and protect against gp120-induced endolysosome dysfunction and neurotoxicity in HAND. Results Endolysosome TRPML1 channels controlled gp120-induced decreases in endolysosome Fe2+ and increases in cytosolic Fe2+ Fe2+ release from endolysosomes is sufficient to account for gp120-induced increases in intracellular Fe2+ levels [28]. To determine the involvement of endolysosome TRPML1 channels, we measured endolysosome Fe2+ using LysoRhonox-1 and showed that gp120 (500 pM, 4 h) significantly decreased endolysosome Fe2+ levels and that this effect was blocked by pretreatment with the TRPML1 inhibitors ML-SI1 (10 μM, 1 h) and Ned-19 (1 μM, 1 h) (Figure 1A and B). gp120 significantly increased cytosolic Fe2+ levels measured with PhenGreen FL DA, and pretreatment with the TRPML1 inhibitors ML-SI1, Ned-19 and YM201 (5 μM, 1 h) blocked these increases (Figure 1C). YM201 alone decreased basal cytosolic Fe2+ levels (Figure 1C). Given that cytosolic Fe2+ accumulation increases ferritin protein levels to sequester excess iron [44], we measured protein expression levels of ferritin H (FTH1) and found that pretreatment with Ned-19 (1 μM, 1 h) significantly (p < 0.0125) blocked gp120-induced (500 pM, 6 h) increases in FTH1 protein expression levels (Figure 1D and E). For stable TRPML1 knockdown cells, TRPML1 protein expression levels were significantly decreased by about 41% (Figure 1F), basal cytosolic Fe2+ levels and gp120-induced Fe2+ increases were significantly decreased (Figure 1G), and FAC-induced increases in cytosolic Fe2+ were significantly blocked (Figure 1H). Conversely, pretreatment with the TRPML1 activator NAADP-AM (1 μM, 1 h) potentiated gp120-induced increases in cytosolic Fe2+ levels (Figure 1I). gp120-induced decreases in endolysosome Fe2+ and increases in cytosolic Fe2+ were mediated by endolysosome TRPML1 channels. (A and B) Endolysosome Fe2+ levels in SH-SY5Y cells were measured with LysoRhoNox-1 by flow cytometry and data were presented as fold changes of mean fluorescence intensity (MFI). gp120-induced decreases in endolysosome Fe2+ levels were significantly decreased by pre-treatment (1 h) with the TRPML1 inhibitors Ned-19 (A, 1 μM) and ML-SI1 (B, 10 μM). (C) Cytosolic Fe2+ levels in SH-SY5Y cells were measured with Phen Green FL DA by flow cytometry and data were presented as reciprocals of MFI (1/MFI). gp120-induced increases in cytosolic Fe2+ levels were significantly decreased by the TRPML1 inhibitors ML-SI1 (10 µM), YM201 (5 µM), and Ned-19 (1 µM). YM201, but not ML-SI1 or Ned-19 significantly decreased basal cytosolic Fe2+ levels. (D, E) Representative Western blot image (D) and quantification of ferritin H (FTH1) levels in SH-SY5Y cells showed that gp120-induced increases in protein expression levels of FTH1 were blocked by 1 h treatments with Ned-19 (1 µM, 1 h). (F) Representative immunofluorescence images of TRPML1 (green) and nuclei stained with Hoechst 33342 (blue) in U87MG cells transfected with control shRNA or TRPML1 knockdown (TRPML1 KD); TRPML1 KD significantly decreased protein expression levels of TRPML1 by ~41%. (G) TRPML1 KD significantly decreased basal cytosolic Fe2+ levels and significantly decreased gp120-induced increases in cytosolic Fe2+ levels. (H) TRPML1 KD significantly reduced ferric ammonium citrate (FAC)-induced increases in cytosolic Fe2+ levels. (I) Pre-treatment (1 h) with the TRPML1 agonist NAADP-AM (1 µM) potentiated gp120-induced increases in cytosolic Fe2+ levels in SH-SY5Y cells. Data were shown as means and SEM with individual data points (n = 4–11) included on each bar. Two-way ANOVA with Tukey's multiple comparison tests were used for statistical analyses. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Reduced glutathione (GSH) decreased levels of endolysosome Fe2+ and blocked gp120-induced increases in levels of intracellular Fe2+ TRPML1 is a redox sensing channel and GSH is a major anti-oxidant with metal -buffering actions [45]. Using LysoRhonox-1, we found that exogenous GSH (1 mM) but not oxidized GSH (GSSG) significantly decreased basal endolysosome Fe2+ levels (Figure 2A). GSH in combination with gp120 (500 pM) further decreased endolysosome Fe2+ levels (Figure 2A). As a control, we found that neither GSH (1 mM) nor GSSG (1 mM) altered Fe2+ binding to LysoRhonox-1 (Supplementary Figure 5A). Using PhenGreen FL DA, GSH (1 mM) significantly decreased basal levels of Fe2+ in the cytosol and significantly blocked gp120-induced increases Fe2+ in the cytosol; GSSG (1 mM) had no significant effect (Figure 2B). GSH and GSSG interfered with Fe2+ binding to PhenGreen FL (Supplementary Figure 5B) but not FerroOrange (Supplementary Figure 5C). Accordingly, using FerroOrange, we found that GSH significantly decreased intracellular Fe2+ levels, and GSH and to a lesser extent GSSG blocked gp120-induced increases in intracellular Fe2+ (Figure 2C). Inhibition of endogenous GSH biosynthesis with N-methylmaleimide (NMM, 200 μM) decreased endolysosome Fe2+ levels (Figure 2D) and significantly increased cytosolic Fe2+ levels (Figure 2E). Reduced glutathione decreases levels of endolysosome and cytosolic Fe2+, and blocks gp120-induced decreases in endolysosome and increases in intracellular levels of Fe2+. All measurements were performed in SH-SY5Y cells. (A) Using LysoRhoNox-1, gp120 (500 pM), GSH (1 mM), and gp120 plus GSH significantly decreased endolysosome Fe2+ levels. (B) Using Phen Green FL DA, GSH significantly decreased basal levels of cytosolic Fe2+ and significantly blocked gp120-induced increases in cytosolic Fe2+ levels. (C) Using FerroOrange, GSH significantly decreased basal levels of intracellular Fe2+, and GSH and GSSG significantly blocked gp120-induced increases in intracellular Fe2+ levels. (D and E) Inhibition of GSH biosynthesis with N-methylmaleimide (NMM, 200 µM) significantly decreased levels of endolysosome Fe2+ and significantly increased cytosolic Fe2+ levels. Data were illustrated as means and SEM with individual data points included (n = 5 to 12). Two-way ANOVA with Tukey's multiple comparison tests were used for statistical analyses. ns = non-significant, ***p < 0.001, ****p < 0.0001. gp120-induced increases in cytosolic ROS, Fe2+, and decreases in H2S were blocked by ROS and TRPML1 inhibitors​​​​​​ Using CM-H2DCFDA, gp120 (500 pM, 4 h) significantly increased cytosolic ROS levels, and those increases were blocked by the antioxidant NAC (5 mM) and by Ned-19 (Figure 3A and B). As a measure of ROS, hydrogen peroxide (H2O2) concentration-dependently increased levels of cytosolic Fe2+ (Figure 3C), H2O2 (100 μM)-induced increases in cytosolic Fe2+ were blocked by Ned-19 and ML-SI1 (Figure 3D), and gp120-induced increases in cytosolic Fe2+ levels were blocked by NAC (5 mM) and Trolox (200 μM) (Figure 3E). Using SF7-AM, gp120-induced (500 pM, 4 h) decreases in cytosolic H2S levels were blocked by Ned-19 (Figure 3F). Consistent with iron involvement in H2S regulation [46], H2O2 (250 μM) significantly increased cytosolic ROS, serum deprivation (0% FBS) significantly increased cytosolic H2S levels, and iron loading with FAC (100 μM) significantly decreased H2S levels (Supplementary Figure 1A and B). gp120-induced increases in cytosolic ROS, Fe²⁺, and H2S were blocked by ROS and TRPML1 inhibitors. All experiments were conducted in SH-SY5Y cells. (A and B) Using CM-H2DCFDA, gp120 (500 pM) significantly increased cytosolic ROS levels and those increases were significantly blocked by the antioxidant NAC (5 mM, 1 h) and the TRPML1 inhibitor Ned-19 (1 µM, 1 h). (C and D) Using Phen Green FL DA, H2O2 concentration-dependently increased cytosolic Fe2+ levels and both Ned-19 (1 µM, 1 h) and ML-SI1 (10 µM, 1 h) significantly blocked H2O2-induced increases in cytosolic Fe2+ levels. (E) The antioxidants N-acetylcysteine (NAC, 5 mM, 1 h) and Trolox (200 µM, 1 h) both significantly blocked gp120-induced increases in cytosolic Fe2+ levels. (F) Using SF7-AM, Ned-19 (1 µM, 1 h) significantly blocked gp120-induced decreases in cytosolic H2S levels. Data were shown as means and SEM with individual data points (n = 5 to 12). Two-way ANOVA with Tukey's multiple comparison tests were used for statistical analyses. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Ned-19 blocked gp120-induced increases in endolysosome ROS, lipid peroxidation, and nitric oxide gp120 disrupts the RSI in endolysosomes [28] and here we tested the extent to which TRPML1 redox activation mediated these effects. In LysoTracker-positive endolysosomes, gp120 (500 pM, 4 h) significantly increased CellROX (ROS), BODIPY C11 (lipid peroxidation), and siRNO (nitric oxide) fluorescence (Figure 4A–F); effects were significantly blocked by pretreatment with Ned-19 (1 μM, 1 h) (Figure 4). gp120 activates NOX2 [47] and the NOX2 inhibitor apocynin (500 nM) significantly prevented gp120-induced increases in endolysosome ROS levels (Supplementary Figure 2A and B). As positive controls, H2O2 (200 μM) significantly increased endolysosome ROS, RLS3 (10 μM) significantly increased endolysosome lipid peroxidation, and sodium nitroprusside (SNP, 100 μM) significantly increased endolysosome nitic oxide (Supplementary Figure 2C–H). Ned-19 blocked gp120-induced increases in endolysosome ROS, lipid peroxidation, and nitric oxide. (A, C, E) Representative images of SH-SY5Y cells treated with vehicle (control), Ned-19 (1 μM, 1 h), gp120 (500 pM, 4 h) and Ned-19 plus gp120, and stained for ROS (CellROX, red), lipid peroxidation (Bodipy C11, green), nitric oxide (SiRNO, red), endolysosomes (LysoTracker, green or red), and nuclei (Hoechst 33342, blue). Scale bars were 15 µm (A) and 10 µm (C and E). (A, C, E) Representative Imaris-generated surface rendering of Lysotracker-positive endolysosomes containing ROS, lipid peroxidation, and nitric oxide fluorescence. Scale bar = 1 µm. (B, D, F) Images were analyzed with Imaris software, and data were presented as fold-changes of mean fluorescence intensity (MFI) for CellROX, Bodipy C11 or SiRNO fluorescence within LysoTracker-positive endolysosomes. Each data point represents the fluorescence MFI within Lysotracker-positive endolysosomes from one field of view with a minimum of 100 cells analyzed per condition across at least two biological replicates. Data were shown as means and SEM with individual data points (n = 6 to 7). Two-way ANOVA with Tukey's multiple comparison tests were used for statistical analyses. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Ned-19 blocked gp120-induced decreases in endolysosome GSH and H2S, but not increases in H2Sn gp120 (500 pM, 4 h) significantly decreased endolysosome GSH and H2S levels, and significantly increased sulfane sulfur (H2Sn) levels (Figure 5A–D). Ned-19 (1 μM, 1 h) significantly blocked gp120-induced decreases in GSH and H2S levels (Figure 5A–D). In contrast, Ned-19 did not significantly affect gp120-induced increases in H2Sn levels (Figure 5E and F). Ned-19 alone significantly increased endolysosome GSH, H2S and H2Sn levels (Figure 5A–F). As additional validation controls, serum deprivation (0% FBS), which increases production of endogenous reactive sulfur species, significantly increased endolysosome H2S and H2Sn levels, and H2O2 exposure significantly decreased endolysosome GSH levels (Supplementary Figure 3A–F). Ned-19 blocked gp120-induced decreases in endolysosome GSH and H2S, but not increases in H2Sn. (A, C, E) Representative images of SH-SY5Y cells treated (1 h) with vehicle (control), Ned-19 (1 μM), gp120 (500 pM), and Ned-19 plus gp120. Cells were stained for GSH (RT-AM-GSH, blue/green), H2S (P3, green), sulfane sulfur (SSP4, green), endolysosomes (LysoTracker, red), and nuclei (Hoechst 33342, blue). Scale bars = 10 µm. (A, C, E) Representative Imaris-generated surface rendering of LysoTracker-positive endolysosomes containing GSH, H2S, or H2Sn. Scale bar = 1 µm. (B, D, F) Images were analyzed with Imaris software and data were presented as fold-changes of mean fluorescence intensity (MFI) of RT-AM GSH, P3 and SSP4 fluorescence within LysoTracker-positive endolysosomes. Each data point represents MFI within Lysotracker-positive endolysosomes from one field of view with a minimum of 100 cells analyzed per condition across at least two biological replicates. Data were shown as means and SEM with individual data points (n = 6 to 7). Two-way ANOVA with Tukey's multiple comparison tests were used for statistical analyses. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Ned-19 blocked gp120-induced increases in protein cysteine oxidation in endolysosomes Using DCP-Rho1, a dye that labels oxidized proteins at sulfenic acid sites [48], gp120 (500 pM, 1 h) significantly increased DCP-Rho1 fluorescence, Ned-19 (1 μM, 1 h) significantly decreased DCP-Rho1 fluorescence, and Ned-19 significantly blocked gp120-induced increases in DCP-Rho1 fluorescence in LysoTracker-positive endolysosomes (Figure 6A and B). Probe specificity was confirmed by blocking DCP-Rho1 staining with dimedone and increasing DCP-Rho1 staining with H2O2 (Supplementary Figure 4A and B). Ned-19 blocked gp120-induced increases in endolysosome protein cysteine oxidation. (A) Representative images of SH-SY5Y cells pre-treated for 1 h with Ned-19 (1 μM) or vehicle before treatment for 1 h with gp120 (500 pM). Cells were stained for protein cysteine oxidation (PCO; DCP-Rho-1, red), endolysosomes (LysoTracker, green) and nuclei (Hoechst 33342, blue). Scale bars = 10 µm. The bottom panels showed representative Imaris-generated surface rendering of Lysotracker-positive endolysosomes containing PCO fluorescence. Scale bar = 1 µm. (B) Quantification of the fold changes of mean fluorescence intensity (MFI) of DCP-Rho-1 staining within LysoTracker-positive endolysosomes showed that Ned-19 significantly decreased PCO, gp120 significantly increased PCO, and Ned-19 blocked gp120-induced increases in endolysosome PCO. Each data point represents the MFI of DCP-Rho1 staining within Lysotracker-positive endolysosomes from one field of view with a minimum of 133 cells analyzed per condition across at least two biological replicates. Data were means and SEM with individual data points (n = 7). Two-way ANOVA with Tukey's multiple comparison tests were used for statistical analyses. ****p < 0.0001. gp120-induced endolysosomal de-acidification was blocked by Ned-19 and antioxidants Using the ratiometric probe LysoSensor Yellow/Blue DND-160, we found that gp120 (500 pM, 30 min) significantly increased endolysosome pH (Figure 7) and pretreatment with Ned-19 (1 μM, 30 min) blocked this effect (Figure 7A). Because TRPML1 is a redox-sensitive channel, we next tested whether gp120-induced increases in ROS contributed to endolysosome de-acidification. Pretreatment with the NOX2 inhibitor apocynin (500 nM, 30 min) or the antioxidant NAC (5 mM, 30 min) prevented gp120-induced de-acidification (Figure 7B and C). Similarly, supplementation with reduced GSH (1 mM, 30 min) also blocked gp120-induced pH changes (Figure 7D). Ned-19 and anti-oxidants blocked gp120-induced endolysosomal de-acidification. (A–D) Endolysosome pH was measured in SH-SY5Y cells using LysoSensor Yellow/Blue DND-160. Cells were pre-treated for 30 min with Ned-19, apocynin, NAC, GSH or vehicle and then for 30 min with gp120 (500 pM). Each data point (n) represents the mean pH value from cells in a single well of a 96-well plate. Data were shown as means and SEM with individual data points (n = 7 to 10). Two-way ANOVA with Tukey's multiple comparison tests were used for statistical analyses. **p < 0.01, ***p < 0.001, ****p < 0.0001. Discussion The neurocognitive deficits that range from mild to severe with HAND have been linked to many factors including neuroinflammation and elevated levels of neurotoxic HIV-1 proteins, iron, reactive species, and organellar dysfunction [5,7,13,15,23,29]. Endolysosomes are acidic organelles that, like other organelles, exhibit stress responses to various insults; for endolysosomes this has been referred to as lysosomal stress response [49]. Others and we have shown that endolysosomes are ‘master regulators of cellular iron’ and iron continues to be linked directly to the generation of reactive species [25,28,29,50]. Others and we have also shown that HIV-1 proteins have adverse effects on the morphology and functions of endolysosomes including increasing levels of intracellular iron, redox catastrophe, bioenergetic crisis, mitochondrial dysfunction, and increased neural cell death [24,28]. Here, we extend such work by demonstrating that reactive species once generated can feedback onto TRPML1 endolysosome cation channels that function as redox sensors and thereby induce a vicious cycle of gp120-induced iron-redox disruption, endolysosome dysfunction, inter-organellar neurotoxic signaling, and cell death. gp120-induced increases in ROS activated TRPML1 channels to increase Fe2+ release from endolysosomes, increase cytosolic levels of Fe2+ and ROS, decrease levels of cytosolic H2S, disrupt the endolysosome RSI, and increase endolysosome luminal protein cysteine oxidation and pH (Figure 8). A model of gp120-induced TRPML1 redox activation and endolysosome dysfunction. gp120 is internalized and traffics to endolysosomes where it activates NOX2 and increases levels of ROS. Elevated ROS activates TRPML1, depletes endolysosome glutathione, and increases the release of Fe2+ into the cytosol. Increases in cytosolic Fe2+ and ROS and decreases in H2S initiate a vicious cycle of endolysosome iron release and redox disruption. TRPML1 redox activation further drives endolysosome iron and redox imbalance, impairs acidification, reduces antioxidant defenses, and induces protein cysteine oxidation and other oxidative damage. The end result is endolysosome dysfunction and increased cellular stress responses. Blocking TRPML1 channels may represent a potential therapeutic target against gp120-induced endolysosome dysfunction. Illustration was created using BioRender. Endolysosomes contain high levels of readily releasable Fe2+ that are sufficient to account for insult-induced increases of Fe2+ and ROS in the cytosol and in mitochondria [24,28,32,50]. Insult-induced release of endolysosome Fe2+ occurs via multiple channels including two-pore and TRPML channels [43,51,52]. TRPML1 channels are especially important because they are activated by NAADP, the essential phosphoinositide PI(3,5)P2, and ROS [36,53,54]. Others and we have shown that gp120, iron overload, and other insults increase endolysosome Fe2+ release, increase cytosolic and mitochondrial levels of Fe2+ and ROS, cause mitochondrial membrane depolarization, and increase neural cell death [24,28,43,55]. Here, we showed that gp120-induced effects were TRPML1-dependent; pharmacological inhibition of TRPML channels with ML-SI1, Ned-19, and YM201636 all blocked while TRPML1 activation with NAADP-AM potentiated gp120-induced Fe2+ release from endolysosomes. These findings are consistent with findings that NAADP-AM enhanced whereas Ned-19 prevented TRPML1-mediated iron overload-induced cytotoxicity [43]. We also observed that FAC induced a small but statistically significant increase in cytosolic Fe2+ in TRPML1 knockdown cells; a finding likely mediated by TRPML2 channels as previously reported [43]. Cytosolic Fe2+ accumulation was accompanied by elevated ferritin expression; a cellular homeostatic mechanism that helps buffer against excess labile Fe2+[56–58]. Reduced GSH, a thiol-containing redox reducing molecule, is an important endogenous anti-oxidant [45,59]. GSH and its oxidized partner GSSG protect against oxidative damage and iron dyshomeostasis intracellularly and in endolysosomes [60,61]. Because exogenous GSH and GSSG do not freely diffuse across plasma membranes, their effects on intracellular redox and iron homeostasis are mediated through extracellular metabolism and precursor uptake. Specifically, extracellular GSH cleavage is catalyzed by the cell-surface enzyme γ-glutamyl transpeptidase to generate cysteinylglycine, which is further metabolized into constituent amino acids including cysteine. Cysteine is then transported into cells and used for de novo GSH synthesis. Through this mechanism, exogenous GSH modulates intracellular GSH levels and redox balance [62–65]. Given the close relationship between GSH and the labile Fe2+ pool as well as its role in regulating intracellular iron trafficking, changes in intracellular GSH are known to affect Fe2+ buffering and redox state [59,63,66–68]. Consistent with this, neurodegenerative disorders commonly show iron accumulation alongside GSH depletion [69], reflecting a pathogenic interplay in which iron-driven oxidative stress promotes GSH consumption and lipid peroxidation [69]. Here, we found that exogenous GSH, but not GSSG, decreased basal levels of endolysosome Fe2+ as well as enhanced gp120-induced decreases in endolysosome Fe2+ levels. This is consistent with the ability of GSH to act as a luminal reductant thereby increasing the availability of readily releasable Fe2+ for TRPML1-mediated efflux [70]. Furthermore, GSH blocked gp120-induced increases in cytosolic Fe2+; findings consistent with reports of others that cytosolic GSH can buffer iron following its export from endolysosomes [59,66] and our findings that inhibition of GSH biosynthesis with NMM decreased basal endolysosome Fe2+ levels and increased cytosolic Fe2+ levels. GSH depletion elevated ferritin expression [71] and GSH regulated Zn2+ release from TRPM7-positive GSH-enriched organelles [72]; findings that strengthen the physiological importance of thiol-based gating of metal fluxes from acidic organelles. TRPML1 plays dual roles in cellular signaling and homeostasis. Under physiological conditions, TRPML1 activation supports autophagy-mediated removal of damaged organelles, stimulates ROS-dependent antioxidant pathways, and promotes cell survival [36,73]. Under pathological conditions, TRPML1 activation can increase the release of endolysosome Fe2+, Ca2+, and Zn2+, increase levels of oxidants, disrupt autophagy and mitochondrial function, and cause neuronal death [30,34,51]. Here, we showed that gp120-induced increases in intracellular ROS activated TRPML1 to increase the release of Fe2+ from endolysosomes into the cytosol; effects blocked by the antioxidants NAC and Trolox. These findings are consistent with previous reports that NAC and Trolox inhibited TRPML1-mediated cell death and insult-induced increases in cytosolic levels of Zn2+, Ca2+, and ROS [41,43,73]. Further, and consistent with findings of others [30,34,51], we found that H2O2 dose-dependently increased cytosolic Fe2+ levels, and pharmacological inhibition of as well as knockdown of TRPML1 channels decreased H2O2-induced increases in cytosolic Fe2+ levels. Moreover, TRPML1 redox activation decreased levels of cytosolic H2S; Fe2+-induced increases in ROS can oxidize H2S to other reactive sulfur species thereby depleting H2S levels [74] and oxidative stress can upregulate the anti-oxidant enzyme SOD1, a potent oxidizer of H2S [74]. We also found that gp120-induced TRPML1 activation increased endolysosome ROS, LPO and NO levels, and decreased H2S, GSH and Fe2+ levels. These redox disruptions appeared to impair organelle function as evidenced by increased luminal protein cysteine oxidation and endolysosome de-acidification. Importantly, these effects were prevented by pretreatment with Ned-19. Thus, TRPML1 redox activation appears to link endolysosome Fe2+ release to a vicious self-amplifying cycle that promotes neuronal oxidative stress, mitochondrial dysfunction, and cell death. NOX2 is a major source of ROS within endolysosomes and gp120 can activate NOX2 and increase levels of ROS and neuronal cell death [47,75,76]. Here, we found that pharmacological inhibition of NOX2 with apocynin blocked gp120-induced increases in endolysosome ROS. These results appear to be consistent with reports that HIV infection activates endolysosome NOX2 to generate ROS [77] and that TRPML1/TFEB signaling can further potential NOX2-dependent ROS production [36,78]. Hydrogen peroxide can increase oxidation of membrane phospholipids involving Fe2+ and Fenton-like chemical reactions [25,50,79,80]. Previously it was reported that gp120 increased LPO [28,33] and induced cortical neuronal death via increases in NO and superoxide ions [33]. Here, we found that gp120-induced increases in endolysosome LPO and NO levels. Furthermore, increased NO production within endolysosomes led to peroxynitrite-mediated LPO increases, inhibition of vacuolar ATPase activity, and endolysosome de-acidification [81–83]. Ned-19 alone increased levels of endolysosome H2S, H2Sn, and GSH, blocked gp120-induced decreases in H2S, GSH, and Fe2+, and restored endolysosome acidification. gp120 and HIV infection have been shown to decrease intracellular GSH and H2S levels [84,85]. The increased H2Sn observed here could be due to gp120-induced increases in LPO because H2Sn accumulates during LPO and contributes to LPO mitigation [86]. Importantly, H2S, H2Sn, and GSH are synthesized by cysteine desulfuration, and together they provide protection against oxidative stress, ferroptosis, and neurodegeneration [87]. Endolysosomes are enriched in cystine, the oxidized dimeric form of cysteine, and endolysosome de-acidification has been linked to cystine depletion, lipid peroxidation, and ferroptosis [88,89]. Thus, TRPML1 redox activation-induced endolysosome de-acidification may decrease cysteine availability, decrease endolysosome H2S and GSH levels, and thereby exacerbate oxidative stress and cellular vulnerability to ferroptosis. Elevated levels of reactive species can cause thiol imbalance, protein oxidation, morphological alterations, functional impairments, and neuronal cell death [45]. Endolysosomes are particularly vulnerable to oxidative damage; oxidation of v-ATPase and cathepsin B/L impairs endolysosome functions [90,91] and gp120-induced endolysosome de-acidification can decrease cathepsin B activity and cathepsin B cysteine oxidation [91]. Here, we found that Ned-19 alone decreased endolysosome protein cysteine oxidation below baseline levels, that gp120 increased endolysosome protein cysteine oxidation, and that Ned-19 blocked gp120-induced increases in endolysosome protein cysteine oxidation. Thus, TRPML1 activity can help regulate endolysosome luminal redox balance, can regulate levels of GSH, and GSH can modify protein cysteines though reversible glutathionylation, a mechanism that protects proteins from irreversible over-oxidation to sulfinic or sulfonic acids [92]. TRPML1 activation can de-acidify endolysosomes [39,40,93], Ned-19 can normalize endolysosome pH [39], and these actions are likely mediated through ROS and other reactive species production [81,90,94]. NAC restored endolysosome pH [95,96], GSH enhanced luminal enzymatic activity and acidification [97], and NOX-2-derived superoxide impaired endolysosome acidification [90,98]. Here, gp120-induced de-acidification was rescued by NAC, GSH, Ned-19, and the NOX inhibitor apocynin. Together with our previous work showing that the endolysosome-specific iron chelator deferoxamine (DFO) prevented gp120-induced endolysosome de-acidification, redox imbalance, and cell death [28,30,32,99,100], these observations indicate that lysosomal stress responses are characterized by endolysosomal dysfunction and iron dyshomeostasis, can be modulated either by limiting TRPML1 redox activation or by targeting endolysosomal iron. These findings, obtained in SH-SY5Y and U87MG cells, are consistent with similar lysosomal stress responses we have observed in established cell lines, mouse primary cultures, and human cortical neurons. In these systems, diverse insults-including silica nanoparticles, LDL cholesterol, HIV-1 gp120 and Tat proteins, SARS-CoV-2 S1 protein, human endogenous retrovirus K envelope protein, drugs of abuse, and weak base therapeutic drugs-induce endolysosomal de-acidification, and iron and redox imbalance, and cell death [28,32,99–103]. These changes are prevented by endolysosome-specific iron chelator DFO [28,30,32,99]. Together, these results indicate that lysosomal stress disrupts iron and redox homeostasis through TRPML1 redox dependent mechanisms. Targeting endolysosomal iron and TRPML1 redox signaling may be promising approaches to protect against gp120-induced endolysosome dysfunction, iron/redox imbalance, neurotoxicity, and HAND progression. Materials and methods Cell cultures SH-SY5Y human neuroblastoma and U87MG human astrocytoma cells were maintained in Dulbecco's Modified Eagle Medium (DMEM; Invitrogen, Carlsbad, U.S.A., Cat. #11995) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (Invitrogen, Cat. #15140122). Cells were grown in T75 flasks and after reaching 80%–90% confluency were sub-cultured using 0.025% trypsin (Invitrogen, Carlsbad U.S.A., Cat. #25200056). All cultures were maintained in a humidified incubator at 37 °C with 5% CO2 and cells were not used beyond their tenth passage. Reagents and treatments Recombinant HIV-1 IIIB gp120 was purchased from ImmunoDX (Woburn, MA, U.S.A., Cat. #1001). Ferric ammonium citrate (FAC; Cat. #I72-500) was purchased from Fisher Scientific U.S.A.. Trans-Ned 19 (Cat. #3954) and NAADP tetrasodium salt (Cat. #3905) were purchased from Tocris (Minneapolis, U.S.A.). Sodium nitroprusside (SNP; Cat. #PHR1423), and sodium hydrosulfide (NaHS; Cat. #161527) were purchased from Sigma-Aldrich (St. Louis, U.S.A.). gp120 was prepared as a 0.1 µM stock solution and diluted in ultrapure distilled water to the desired concentration immediately before use. The working concentration of gp120 (500 pM; ~60 ng/mL) is less than the reported serum levels reported in HIV-1 positive patients (120 to 960 ng/mL) [104]. Accordingly, a concentration of 500 pM gp120 was used to model physiological relevance exposure, consistent with our previous studies [28]. Unless otherwise indicated, cells were incubated with gp120 or drugs for 1 to 4 h to investigate early effects on endolysosome function. Measurement of endolysosome, cytosolic and intracellular Fe2+, cytosolic ROS, and cytosolic H2S Endolysosome Fe2+ levels were measured using LysoRhonox-1. Cytosolic Fe2+, ROS, and H2S were measured with Phen Green™ FL diacetate (PGFL-DA), CM-H2DCFDA, and SF7-AM, respectively. Intracellular Fe2+ levels were measured with FerroOrange. The probes were purchased from the following suppliers; LysoRhonox-1 (Lumiprobe Life Science Solutions, MD, U.S.A.; Cat. #3317), PGFL-DA (Invitrogen, MA, U.S.A.; Cat. #P14313), CM-H2DCFDA (Invitrogen, Cat. #C6827), SF7-AM (Cayman Chemicals, Cat. #1416872-50-8), and FerroOrange (Dojindo, Cat. #F374). SH-SY5Y cells were seeded at 6 × 105 cells/well in 24-well culture plates and incubated at 37 °C with 5% CO2 overnight. Cells were pre-treated with inhibitors or vehicle for 1 h followed by treatment with gp120 (500 pM) or vehicle for an additional 4 h. After treatment, cells were incubated at 37 °C with the following dyes in DMEM; 5 µM LysoRhonox-1 for 20 min, 10 µM PGFL-DA for 20 min, 10 µM CM-H2DCFDA for 20 min, 0.3 µM SF7-AM for 25 min, and 1 µM FerroOrange for 20 min. After staining, cells were washed three-times with PBS and resuspended in 600 μL PBS. Stained untreated cells served as baseline controls, and cells treated with ferric chloride (FeCl3, 100 μM) or ferric ammonium citrate (FAC, 50 μM) were included as positive controls for iron assays. Flow cytometry was performed on an Attune NxT flow cytometer (ThermoFisher, Waltham, U.S.A.). A minimum of 10,000 events were collected per condition. Fluorescence was recorded at the following settings; LysoRhonox-1 (Ex/Em = 561/590 nm), PLGF-DA (Ex/Em = 488/530 nm), CM-H2DCFDA (Ex/Em = 488/530 nm), SF7-AM (Ex/Em = 488/530 nm), and FerroOrange (Ex/Em = 561/585 nm). Each sample was measured at least three-times, and the mean MFI value was used for analyses. Western blot analysis of ferritin H (FTH1) SH-SY5Y cells were seeded in 100 mm2 dishes and were incubated at 37 °C with 5% CO2 overnight. Cells were pretreated for 1 h with Ned-19 (1 μM) or vehicle, followed by treatment for 6 h with gp120 (500 pM). After treatment, cells were washed twice with ice-cold PBS and lysed in RIPA lysis buffer (ThermoScientific, Cat. #89900) supplemented with a protease inhibitor cocktail (ThermoScientific, Cat. #78429). Lysates were clarified by centrifugation (~16000 × g for 15 min in 4 °C), and protein concentrations were determined using a modified Bradford assay (ThermoScientific, Cat. #A55866). Equal amounts of protein (20 μg) were separated by SDS-PAGE on 4%–12% bis-Tris gel and were transferred onto PVDF membranes using the iBlot 3 system (ThermoFisher). Membranes were blocked in 5% (w/v) non-fat milk in TBST for 1 h at room temperature and then incubated overnight at 4 °C with anti-FTH1 antibody (1:1000, Cell Signaling Technology, Cat. #4393). After washing, membranes were incubated with HRP-conjugated secondary antibody (1:3000) for 1 h at room temperature. Signals were detected using SuperSignal West Femto chemiluminescent substrate (ThermoFisher, Cat. #34095) and imaged with a ChemiDoc Imaging System (BioRad). GAPDH (1:2000) was used as a loading control. Band intensities were quantified using imageJ (Fiji) software. Western blots were performed as three independent experiments (n = 3) and quantification was based on three biological replicates. Measurement of endolysosome ROS, NO, lipid peroxidation, GSH, H2S, and sulfane sulfur SH-SY5Y cells were seeded in 35 mm2 glass-bottom culture dishes and were incubated at 37 °C with 5% CO2 overnight. Cells were preincubated with Ned-19 (1 μM) or vehicle for 1 h followed by gp120 (500 pM) treatment for 4 h. After treatment, cells were stained in phenol red-free DMEM (Invitrogen, Cat. #21063029) with one of the following probes; CellROX™ Deep Red (5 μM, Invitrogen, Cat. #C10422) for ROS (30 min), BODIPY™ 581/591 C11 (10 μM, Invitrogen, Cat. #D3861) for lipid peroxidation (25 min), siRNO (5 μM, Sigma-Aldrich, Cat. #SCT053) for NO (30 min), RT-AM GSH (1 μM, Kerafast, Inc. MA U.S.A., Cat. #EBY001) for reduced glutathione (10 min), P3 (10 μM, Sigma-Aldrich, Cat. #5.34329) for H2S (20 min), and SSP4 (5 μM, Dojindo, Cat. #SB10-10) for sulfane sulfur (20 min). LysoTracker™ (50 nM) to label endolysosomes and Hoechst 33342 (1 μg/mL, ThermoFisher, Cat. #62249) to label nuclei were added for the final 7 min of incubations. Cells were washed three-times with PBS (Invitrogen, Cat. #14190144) and resuspended in live cell imaging solution (Invitrogen, Cat. #A59688DJ). Imaging was performed using an Andor Dragonfly 200 spinning-disk confocal microscope (Oxford Instruments, Concord, U.S.A.) equipped with a 63× oil immersion objective and multichannel acquisition. The following microscopy settings used were as follows; CellROX (Ex/Em = 637/698 nm), siRNO (Ex/Em = 637/690 nm), P3 and SSP4 (Ex/Em = 488/530 nm), RT-AM-GSH (Ex/Em = 405 or 530 nm), BODIPY C11 (Ex/Em = 488/530 nm for oxidized green and 581/600 nm for reduced red), LysoTracker deep red (647/668 nm, Invitrogen, Cat. #L12492) and LysoTracker Green (Ex/Em = 488/521 nm) for endolysosomes, and Hoechst 33342 (Ex/E = 405/445 nm) for nuclei. Mean fluorescence intensity (MFI) was analyzed using Imaris software (v9.9). Endolysosome protein cysteine oxidation measurements SH-SY5Y cells were seeded in 35 mm2 culture dishes and incubated overnight at 37 °C and 5% CO2. Cells were pre-incubated with Ned-19 (1 μM) or vehicle for 1 h before addition of gp120 (500 pm) or vehicle for an additional 1 h. DCP-Rho1 (10 μM, Kerafast, Inc. MA U.S.A., Cat. #EE0031) was added for the final 15 min, and LysoTracker™ Green DND-26 (50 nM) together with Hoechst 33342 (1 μg/mL) were added for the final 7 min. Cells were washed two-times with DMEM, two-times with PBS, and then resuspended in live-cell imaging solution. Imaging was performed on an Andor Dragonfly 200 spinning disk confocal microscope equipped with 63× oil immersion objective. Acquisition settings were as follows; Lysotracker Green (Ex/Em = 488/521 nm), DCP-Rho1 (Ex/Em = 561/590 nm), and Hoechst 33342 (Ex/Em = 405/445 nm). Images were analyzed using Imaris software (v9.9), and MFI was quantified within LysoTracker-positive endolysosomes. DCP-Rho1 selectively reacts with sulfenic acid modifications of cysteine residues and has been previously validated as a selective probe for protein cysteine oxidation [48,105]. To further confirm specificity, we pretreated with dimedone, a selective sulfenic acid scavenger to decrease DCP-Rho-1 MFI and the positive control H2O2 (100 μM) to increase DCP-Rho-1 MFI. Endolysosome pH measurements Endolysosome pH was quantified using the ratiometric probe LysoSensor Yellow/Blue DND-160 (5 μM, Invitrogen, Waltham, MA, U.S. Cat. #L7545) [28]. SH-SY5Y cells were seeded in black, clear-bottom 96-well culture plates (Invitrogen, Waltham, MA, U.S. Cat. #165305) and incubated overnight at 37 °C and 5% CO2. Cells were pretreated with inhibitors or vehicle for 30 min followed by treatment with gp120 (500 pM) for an additional 30 min. Cell culture media was replaced with imaging buffer (140 mM NaCl, 2.5 mM KCl, 1.8 mM CaCl2, 1 mM MgCl2, and 20 mM HEPES, pH 7.40) containing 5 μM LysoSensor Yellow/Blue DND‐160. Cells were incubated for 4 min at room temperature in the dark, rinsed three-times with imaging buffer to remove excess dye, and immediately analyzed. Fluorescence intensity was measured with a BioTek Synergy H1 microplate reader with dual excitation wavelengths 340 nm (F340) and 380 nm (F380); emission was recorded at 527 nm. Relative endolysosomal pH was measured from the F340/F380 ratio and pH was quantified from calibration curves using GraphPad Prism as described previously [28]. Effects of GSH and GSSG on Fe2+detection by fluorescence probes LysoRhoNox (5 μM, Ex/Em = 569/581), PhenGreen FL DA (5 μM, 488/530 nm), and FerroOrange (1 μM, 561/570 nm) were diluted in PBS in black clear-bottom 96-well culture plates with or without 1 mM GSH or GSSG. FeCl2 was then added, and fluorescence changes were measured with a BioTek Synergy H1 microplate reader. Generation of stable TRPML1 knockdown U87MG cells U87MG cells were transfected with either TRPML1 shRNA plasmid (Santa Cruz, Sc-44519) or control shRNA plasmid (Sc-108060) using Jet prime reagent as per manufacturer's instructions. After 36 h, cells were selected in puromycin (20 μg/mL; Invitrogen) for 1 week to generate stable knockdown lines. Knockdown efficiency was confirmed by immunofluorescence staining. For immunofluorescence, cells were fixed with 4% paraformaldehyde (15 min, room temperature) and permeabilized with 0.1% Triton-X 100 in PBS (10 min, room temperature). Nonspecific binding was blocked with 10% goat serum in PBS (1 h, room temperature). Cells were incubated overnight at 4 °C with anti-MCOLN1 antibody (1:500, NOVUS biologicals, CO, U.S.A., Cat. #NBP1-92152) followed by washing and incubation with Alexa Fluor 488-conjugated secondary antibodies (1:1000) for 1 h at room temperature. Nuclei were counterstained with Hoechst 33342. Imaging was performed using our spinning disk confocal microscope with a 63× oil immersion objective. Acquisition settings included the GFP channel (Ex/Em = 488/521 nm) for TRPML1 and the DAPI channel (Ex/Em = 405/445 nm) for Hoechst. Images were analyzed using Imaris software (v9.9). Statistics and reproducibility All experiments were independently repeated to ensure reproducibility. Comparisons between two groups were performed using Student's t-test. For multiple group comparison, one-way or two-way ANOVA with Tukey's post hoc test was used. Normality was assed using Shapiro–Wilk's test and Kolmogorov–Smirnov's test, and outliers were identified using the ROUT method (ROUT = 1%). A p < 0.05 was considered statistically significant. All analyses were performed using GraphPad Prism (v9.4–9.5). Supplementary Material Supplementary Material Supplementary data_clean copy.docx Funding Statement This work was supported by the National Institute of General Medical Sciences [P20GM139759], the National Institute of Mental Health [R01MH119000], the National Institute of Neurological Disorders and Stroke [2R01NS065957], and the National Institute on Drug Abuse [2R01DA032444]. N.K. conceived the study, designed and performed experiments, collected and analyzed data, interpreted results, and wrote the original draft. B.L. assisted with experiments and validation. J.D.G. contributed to conceptualization of the study, provided supervision, and edited the manuscript. Acknowledgements This work was supported by the National Institute of General Medical Sciences [P20GM139759], the National Institute of Mental Health [R01MH119000], the National Institute of Neurological Disorders and Stroke [2R01NS065957], and the National Institute on Drug Abuse [2R01DA032444]. N.K. conceived the study, designed and performed experiments, collected and analyzed data, interpreted results, and wrote the original draft. B.L. assisted with experiments and validation. J.D.G. contributed to conceptualization of the study, provided supervision, and edited the manuscript. Disclosure statement No potential conflict of interest was reported by the author(s). Data availability statement All data supporting findings of this study are included in the manuscript and supplementary materials. Additional information is available from the corresponding author upon reasonable request. Preprint A preprint of this article is available on bioRxiv at https://doi.org/10.64898/2026.03.02.709165. Supplemental material Supplemental data for this article can be accessed at https://doi.org/10.1080/13510002.2026.2679333. References Associated Data Supplementary Materials Supplementary Material Supplementary data_clean copy.docx Data Availability Statement All data supporting findings of this study are included in the manuscript and supplementary materials. Additional information is available from the corresponding author upon reasonable request.