A Non-Canonical Role for IRE1α Links ER and Mitochondria as Key Regulators of Astrocyte Dysfunction: Implications in Methamphetamine use and HIV-Associated Neurocognitive Disorders

Abstract

Astrocytes are one of the most numerous glial cells in the central nervous system (CNS) and provide essential support to neurons to ensure CNS health and function. During a neuropathological challenge, such as during human immunodeficiency virus (HIV)-1 infection or (METH)amphetamine exposure, astrocytes shift their neuroprotective functions and can become neurotoxic. Identifying cellular and molecular mechanisms underlying astrocyte dysfunction are of heightened importance to optimize the coupling between astrocytes and neurons and ensure neuronal fitness against CNS pathology, including HIV-1-associated neurocognitive disorders (HAND) and METH use disorder. Mitochondria are essential organelles for regulating metabolic, antioxidant, and inflammatory profiles. Moreover, endoplasmic reticulum (ER)-associated signaling pathways, such as calcium and the unfolded protein response (UPR), are important messengers for cellular fate and function, including inflammation and mitochondrial homeostasis. Increasing evidence supports that the three arms of the UPR are involved in the direct contact and communication between ER and mitochondria through mitochondria-associated ER membranes (MAMs). The current study investigated the effects of HIV-1 infection and chronic METH exposure on astrocyte ER and mitochondrial homeostasis and then examined the three UPR messengers as potential regulators of astrocyte mitochondrial dysfunction. Using primary human astrocytes infected with pseudotyped HIV-1 or exposed to low doses of METH for 7 days, astrocytes had increased mitochondrial oxygen consumption rate (OCR), cytosolic calcium flux and protein expression of UPR mediators. Notably, inositol-requiring protein 1α (IRE1α) was most prominently upregulated following both HIV-1 infection and chronic METH exposure. Moreover, pharmacological inhibition of the three UPR arms highlighted IRE1α as a key regulator of astrocyte metabolic function. To further explore the regulatory role of astrocyte IRE1α, astrocytes were transfected with an IRE1α overexpression vector followed by activation with the proinflammatory cytokine interleukin 1β. Overall, our findings confirm IRE1α modulates astrocyte mitochondrial respiration, glycolytic function, morphological activation, inflammation, and glutamate uptake, highlighting a novel potential target for regulating astrocyte dysfunction. Finally, these findings suggest both canonical and non-canonical UPR mechanisms of astrocyte IRE1α. Thus, additional studies are needed to determine how to best balance astrocyte IRE1α functions to both promote astrocyte neuroprotective properties while preventing neurotoxic properties during CNS pathologies.

Keywords: astrogliosis; metabolic function; mitochondria-associated ER membranes; neurodegeneration; neuroinflammation; unfolded protein response.

FIGURE 1

Pseudotyped HIV-1 infection in astrocytes is characterized by HIV-1 DNA integration and productive protein expression. (A) Schematic of the proposed experimental design to assess the effects of METH exposure and HIV-1 infection on astrocyte ER-mitochondrial homeostasis. (B) Pseudotyped HIV-1 was constructed by co-transfecting an HIV-1 plasmid (pHIV-1) with vesicular stomatitis virus glycoprotein plasmid (pVSVg) in HEK 293 T cells. (C–E) Primary human astrocyte (PHA) cultures were infected with (C,D) 500 RT pseudotyped HIV-1 for 5 days or (E) 1000 RT pseudotyped HIV-1 for 7 days followed by (C) HIV-1 DNA integration assay or (D) Simple Wes to detect expression of HIV-1 proteins p24 or Nef. Vinculin was used as an internal control. (E) Immunocytochemistry staining of HIV-1 protein p24 (red), astrocyte marker glial fibrillary acidic protein (GFAP, green) with nuclear DNA labeled with DAPI (blue). Experimental illustrations were made with BioRender.com.

FIGURE 2

Chronic METH exposure and HIV-1 infection increase astrocyte metabolic activity. Astrocytes were treated with (A–G) METH (50 or 250 nM) or (H–N) infected with pseudotyped HIV-1 (100–1000 RT) for 7 days prior to Seahorse XF Cell Mito Stress Test. (A,H) Representative metabolic profile tracings from a single astrocyte donor are graphed over time. Fold changes in oxygen consumption rates (OCR) quantifying (B,I) basal respiration, (C,J) ATP production, (D,K) maximal respiration, (E,L) spare respiratory capacity, (F,M) non-mitochondrial OCR, and (G,N) proton leak were graphed for statistical comparisons. Individual dots on graphs represent the averaged data from a minimum of six replicates per biological donor. Statistical significance was determined by one-way ANOVA followed by Tukey’s post hoc for multiple comparisons.

FIGURE 3

METH and HIV-1 upregulate ER/UPR signaling mediators. Astrocytes were treated with (C,E) METH (50 nM) for 7 days (blue bars) or (D,F) pseudotyped HIV-1 (500 RT) for 7 days (red bars) before (A–D) calcium imaging or (E,F) protein analysis via Simple Wes. (A–D) Astrocytes were transfected with a GFP-calmodulin calcium sensor (GCaMP6s) for 48 h prior to calcium flux analysis. (A,B) Time series confocal imaging was used to measure changes in fluorescence every 500 ms for a total of 5 min (600 cycles). (B) Representative calcium flux line tracings illustrate the change in astrocyte calcium flux (ΔF) at any given time point, with control media or METH (250 μM) at 20 cycles (10 s) and ionomycin (10 μM) at 450 cycles (225 s). Calcium flux was calculated by: ΔF = (F-F₀)/(F_max-F₀), where F is the fluorescence intensity at any given time; F₀ is the baseline (1 – 20 cycles) fluorescence intensity, and F_max is the maximum fluorescence intensity when exposed to ionomycin (450 – 600 cycles). (C,D) Area under the curve (AUC) was calculated by the sum of ΔF following treatment with control media or METH (250 μM) at 20 cycles (10 sec) and before ionomycin (10 μM) at 450 cycles (225 s). Individual dots represent the average AUC from a minimum of 20 cells per biological donor and are graphed as fold changes. One-way ANOVA was performed for statistical analysis followed by Fisher’s LSD test for stand-alone comparisons to account for sensitivity of calcium flux variation across different biological donors. (E–H) Protein expression of BiP, ATF6, PERK, and IRE1α was measured via Simple Wes post-treatment of (E,G) chronic (7 days) METH (50 nM) or (F,H) HIV-1 infection (7 days; 500 RT). (E,F) Representative blot images are illustrated from two separate biological donors per post-treatment paradigm. (G,H) Data from a minimum of four donors are compiled for graphical representation. Individual dots on graphs represent fold changes to vinculin for separate biological donors. Statistics were performed using ratio-paired t-tests for individual targets per condition.

FIGURE 4

Inhibition of IRE1α decreases astrocyte metabolic function. (A–K) Astrocytes were treated with pharmacological inhibitors for the three UPR arms [ATF6 (AEBSF; 100 μM), PERK (GSK2606414; 1 μM), IRE1α (STF-083010; 60 μM)] for 3 h prior to (A–J) Seahorse Mito Stress Test for metabolic assessment or (K) extracellular lactate dehydrogenase (LDH) assay for cytotoxicity. Representative (A) OCR and (B) ECAR profile tracings from a single astrocyte donor were graphed over time. Compiled data from five separate biological donors quantifying fold changes in (C) basal respiration, (D) ATP production, (E) maximal respiration, (F) spare respiratory capacity, (G) non-mitochondrial OCR, (H) proton leak, (I) basal ECAR, and (J) maximal ECAR were graphed for statistical comparisons. Statistical significance was determined via one-way ANOVA followed by Tukey’s post hoc for multiple comparisons. Each dot on graphs represents the averaged data from a minimum of six replicates per biological donor.

FIGURE 5

IRE1α inhibition can partially restore astrocyte metabolic function following HIV-1 infection and chronic METH exposure. (A–E) Astrocytes were treated with METH (50 nM; blue bars) or infected with pseudotyped HIV-1 (500 RT; red bars) for 7 days followed by IRE1α pharmacological inhibition (STF-083010, 60 μM) for 3 h prior to Seahorse Mito Stress Test. (A) Representative metabolic OCR profile tracing is illustrated from a single astrocyte donor. Compiled data quantifying fold changes in (B) basal respiration, (C) ATP production, (D) maximal respiration, and (E) spare respiratory capacity were graphed for statistical comparisons. Rescue experiments were performed twice in two separate biological donors to acquire four separate data sets. A minimum of six replicates were performed per experiment. Statistical significance was determined by one-way ANOVA followed by Tukey’s post hoc for multiple comparisons.

FIGURE 6

Astrocyte IRE1α regulates both mitochondrial respiration and glycolytic activity through distinct mechanisms. (A–I) Astrocytes were transfected with an IRE1α overexpression vector (gray bars) or backbone (white bars) and then treated with IL-1β for 24 h (checkered bars) prior to functional assessments. (B,C) Cellular lysates were collected and assayed by Simple Wes to quantify (B) IRE1α and (C) BiP expression levels. Vinculin was used as an internal control. (D,F,G) Mitochondrial respiration and (E,H,I) glycolytic activity were assessed by Seahorse metabolic assay. (D) OCR and (E) ECAR profile tracings from a representative astrocyte donor were graphed over time. Fold changes in (F) basal respiration, (G) maximal respiration, (H) basal ECAR, and (I) maximal ECAR were graphed for statistical comparisons. Individual dots on graphs represent the averaged data from a minimum of 6 replicates per biological donor. Significance was determined by one-way ANOVA and Tukey’s post hoc for multiple comparisons. Experimental Illustration was made with BioRender.com.

FIGURE 7

IRE1α overexpression augments cytokine expression and increases glutamate clearance in human astrocytes. Backbone and IRE1α transfected astrocytes were treated with IL-1β for 24 h. (A–D) Cells were immunolabeled with antibodies specific for IRE1α (red) and the astrocyte marker glial fibrillary acidic protein (GFAP, green). Nuclear DNA was labeled with DAPI (blue). (E–G) Experiments in four separate biological donors were analyzed to quantify morphological activation. Individual dots on graphs represent compiled fold-changes calculated from duplicate wells and/or triplicate images per condition for each biological astrocyte donor. (E) GFAP intensity was measured across full-well scans using SoftMax Pro and normalized to DAPI. (F) Process length of individual astrocytes was manually traced and measured using ImageJ Software. (G) Percent morphological activation was calculated based on the number of astrocytes presenting with ‘reactive’ morphology divided by the total number of astrocytes imaged. (H) CCL2 and (I) CXCL8 levels were assessed by an ELISA, and expression was normalized to metabolic activity prior to calculating fold changes to backbone. (J) Astrocytes were treated with 400 nM glutamate for 24 h. Remaining glutamate levels were quantified by fluorescent assay to calculate% glutamate clearance followed by fold change for each individual donor. Individual dots on graphs represent compiled data from triplicate experiments per biological astrocyte donor. Significance was determined by one-way ANOVA and Tukey’s post hoc for multiple comparisons.

SCHEME 1

METH and HIV-1 alter astrocyte function to induce neurotoxicity, which determines the pathology of HAND and METH use disorders. Neuroinflammation, oxidative stress and glutamate excitotoxicity are hallmarks of neurodegenerative pathology and are all propagated by astrocyte dysfunction. During a pathological challenge, such as METH exposure and HIV-1 infection, astrocytes became reactive leading to a shift in their neurotrophic functions to become neurotoxic. This reactive state, often termed as astrogliosis, is characterized by an increased inflammatory phenotype which promotes neuroinflammation. Moreover, decreased provision of essential metabolic and antioxidant support to neurons and increased release of toxic radicals such as reactive oxygen and nitrogen species (ROS/RNS) are triggers for oxidative stress. Finally, excitotoxicity arises from an impaired uptake of excess glutamate between synaptic junctions. Our studies emphasize the role of astrocyte ER-associated mechanisms and mitochondrial dysfunction following METH exposure and HIV-1 infection as potential underlying mechanics controlling astrocyte dysfunction and astrocyte-associated neurodegeneration. Experimental Illustration was made with BioRender.com.