Acidic Nanoparticles Prevent HIV Pre‐Exposure Prophylaxis (PrEP)‐Induced Oligodendrocyte Impairments by Restoring Lysosomal pH in Adolescent Models

ABSTRACT

A disproportionate percentage of adolescents are diagnosed with human immunodeficiency virus (HIV) in the United States each year. Preexposure prophylaxis (PrEP), an antiretroviral regimen, is effective at preventing the transmission of HIV to adolescents at substantial risk for acquiring HIV. However, other select antiretrovirals have been shown to cause white matter deficits in experimental models. Adolescents taking PrEP are uniquely vulnerable to myelin impairments as the adolescent brain undergoes high rates of myelination. Here, we report that PrEP significantly reduced oligodendrocyte maturation in adolescent rats. Furthermore, cultures of primary rat oligodendrocyte progenitors treated with PrEP showed inhibited oligodendrocyte differentiation through deacidification of lysosomes resulting in lysosomal accumulation of myelin proteins. Acidic nanoparticle co‐administration with PrEP prevented PrEP‐induced oligodendrocyte maturation impairments both in vivo and in vitro. These studies suggest uninfected adolescents are vulnerable to PrEP‐induced oligodendrocyte impairments and identify maintenance of lysosome pH as a critical factor in antiretroviral design.

Main Points

• Pre‐exposure prophylaxis (PrEP) inhibits oligodendrocyte differentiation

• PrEP inhibits oligodendrocyte differentiation through lysosome deacidification

• Acidic nanoparticles prevent PrEP‐induced inhibition of oligodendrocyte differentiation

1. Introduction

Myelin, the lipid‐rich plasma membrane of oligodendrocytes in the central nervous system (CNS), provides insulation and metabolic support (Nave 2010) to axons, allowing for fast (Huxley and Stämpeli 1949), efficient, and coordinated (Chorghay et al. 2018; Pease‐Raissi and Chan 2021) neurotransmission. Myelination is a prolonged and complex process that is both temporally and spatially constrained, proceeding from caudal to rostral (Dean et al. 2015) and proceeding in three main temporal waves: in childhood, adolescence, and adulthood (de Faria et al. 2021). During childhood, white matter tracts such as the corpus callosum (Reynolds et al. 2019) are predominately myelinated, whereas during adolescence, rostral gray matter areas such as the prefrontal cortex (Grydeland et al. 2019; Yakovlev, and Lecours (1967)) continue to be myelinated. Thus, because myelination is largely a postnatal event continuing into the third decade of life, oligodendrocytes are well‐suited to play an integral role in the developing adolescent central nervous system.

Each year, 13‐ to 24‐year‐olds disproportionately account for the number of individuals newly diagnosed with human immunodeficiency virus (HIV) in the United States (https://www.cdc.gov/hiv/library/reports/hiv‐surveillance/vol‐28‐no‐3/index.html n.d.). Preexposure prophylaxis (PrEP), a once daily antiretroviral (ARV) regimen, is an effective (Mayer et al. 2020; Grant et al. 2014; Castillo‐Mancilla et al. 2013) method to prevent the transmission of HIV in adolescents at substantial risk for acquiring HIV; however, it is not without risk as studies have shown that select antiretroviral drugs can lead to neurocognitive impairments in people with HIV (PWH) even if their viral load is undetectable (Jernigan et al. 2011; Heaton et al. 2010; (Borjabad et al. 2011)). White matter deficits are commonly seen in PWH, including decreases in myelin (Alakkas et al. 2019; Gongvatana et al. 2011; Corrêa et al. 2015; Kelly et al. 2014; Leite et al. 2013; Tate et al. 2011; Solomon et al. 2020). Our lab has shown that primary oligodendrocyte precursor cell cultures treated with therapeutic concentrations of select ARV drugs displayed dose‐dependent decreases in oligodendrocyte maturation (Festa et al. 2019, 2023; Roth et al. 2020; Jensen et al. 2015). Several mechanisms have been identified, including lysosomal dysfunction due to the weakly basic nature of some of the antiretroviral drugs (Festa et al. 2019, 2023). The drugs that compose a common formulation of PrEP, emtricitabine (FTC) and tenofovir disoproxil fumarate (TDF), are also weak bases that are known to accumulate and deacidify lysosomes (Poole and Ohkuma 1981).

Lysosomes play an integral role in maintaining cellular homeostasis by degrading intracellular macromolecules, sensing nutrient levels, transporting proteins, processing lipids, and participating in calcium signaling (Settembre et al. 2013). Functional lysosomes are especially important in oligodendrocytes as they must process and transport large quantities of lipids and proteins necessary for the generation of the myelin membrane (Guo et al. 2018). The protein proteolipid protein (PLP), critical for myelin formation, is trafficked as a synthesized protein to the plasma membrane in lysosomes (Trajkovic et al. 2006) and proteolytically functional lysosomes are required for the release of PLP from storage to the nascent myelin membrane (Guo et al. 2018). The function of lysosomal proteases is dependent on the maintenance of a low intralysosomal pH (4.5–5) and slight deacidification disrupts the function of essential enzymes by as much as 80% (Barrett 1970).

Here, we evaluated the effects of PrEP on oligodendrocyte development in vitro and in adolescent rats. We demonstrated that PrEP inhibits oligodendrocyte differentiation through the deacidification of lysosomes, resulting in the accumulation of proteins critical for myelin formation within lysosomes, especially PLP. Furthermore, we were able to prevent PrEP‐induced lysosome dysfunction and inhibition of oligodendrocyte maturation by co‐administering poly lactic‐co‐glycolic acid (PLGA) acidic nanoparticles to restore an acidic pH in the lysosomal lumen at the time of PrEP treatment both in vitro and in vivo. This work identifies that uninfected adolescents on PrEP may be vulnerable to myelin alterations and provides a metric for the development of new prophylactic therapies to protect this uniquely vulnerable population.

2. Results

2.1. PrEP Treatment Impairs Oligodendrocyte Maturation in Cortex of Adolescent Rats

Previously, ART drugs across multiple classes (integrase strand transfer inhibitors (INSTI) and protease inhibitors (PI)) have been shown to inhibit remyelination following demyelination in adult mice (Festa et al. 2023; Roth et al. 2020; Jensen et al. 2015). However, the effect of ART drugs on developmental myelination during adolescence was not investigated. Here, the effects of two ART drugs that commonly compose PrEP, emtricitabine (FTC) and tenofovir disoproxil fumarate (TDF), were examined on oligodendrocyte differentiation in male and female adolescent rats. First, we administered a range of concentrations of the two drugs to rats, took plasma samples by submandibular bleed at 3 h, and used mass spectrometry to determine drug concentrations comparable to the maximum plasma (C _max) values observed in patients (Table S1) (Mathias 2012). PrEP (64.35 mg/kg TDF and 17.14 mg/kg FTC) was administered daily for 3 weeks starting at 3 weeks of age via oral gavage as this period approximates adolescence in humans (Semple et al. 2013). Previous human post‐mortem studies have found concentrations of FTC and TDF in brain tissue from individuals with HIV who continued to take ART until the time of death which were similar to the plasma C _max, suggesting that FTC and TDF do enter the CNS (Ferrara et al. 2020). Interestingly, the concentrations of select ART drugs, including tenofovir, the active drug form of both TDF and TAF, were significantly greater in white matter areas compared to gray matter areas (Ferrara et al. 2020). Furthermore, the concentration of tenofovir in the postmortem brain was 25 times greater than previously published CSF tenofovir drug concentrations (Ferrara et al. 2020). While not all antiretrovirals were examined, Ferrara et al. (2020) demonstrated that ART drugs can enter the CNS and suggest that CSF values may not accurately reflect the concentration of ART in the brain.

First, oligodendrocyte maturation was examined in the frontal cortex as this area is still undergoing high rates of myelination in adolescents (Grydeland et al. 2019) (Figure 1a). The myelin protein proteolipid protein (PLP) was used as a marker for mature oligodendrocytes. Male adolescent rats treated with PrEP had significantly less PLP expression compared to vehicle‐treated animals in the frontal cortex (Figure 1b). Furthermore, to determine the effects of PrEP on the number of mature oligodendrocytes, brain sections were labeled with an antibody to aspartoacylase (ASPA, labels oligodendrocyte cytoplasm). PrEP treatment significantly decreased the number of ASPA+ cells compared to vehicle‐treated animals in the frontal cortex (Figure 1c). Additionally, sections were labeled with the oligodendrocyte precursor cell (OPC) marker of neuro/glial antigen 2 (NG2). PrEP treatment significantly increased NG2 fluorescent intensity compared to vehicle‐treated animals in the frontal cortex (Figure 1d). Hence, PrEP treatment significantly decreased oligodendrocyte maturation in the rostral cortex and increased the fluorescent intensity of OPC staining. The latter may be due to OPCs migrating into the area as is seen in MS and its models (Festa et al. 2023; Roth et al. 2020; Jensen et al. 2015; Nait‐Oumesmar et al. 2007).

FIGURE 1.

PrEP treatment impairs oligodendrocyte maturation in the cortex and corpus callosum of male adolescent rats. (a) Schematic of daily oral gavage treatment of rats from 3‐ to 6‐weeks‐old. Generic coronal section of a rat brain for rostral cortex and corpus callosum. The boxes represent the areas of analysis. All images were taken, and analysis was done at 6‐weeks‐old after 3 weeks of treatment. (b) Representative images of PLP staining within the frontal cortex. (c) Quantification of PLP intensity normalized to untreated rats in the frontal cortex (p = 0.0001, t = 9.747, df = 8). (d) Representative images of ASPA staining within the frontal cortex. (e) Quantification of ASPA cell number normalized to untreated rats in the frontal cortex (p = 0.0001, t = 4.856, df = 8). (f) Representative images of NG2 staining within the frontal cortex. (g) Quantification of NG2 intensity normalized to untreated rats in the frontal cortex (p = 0.0010, t = 5.059, df = 8). (h) Representative images of PLP staining within the corpus callosum of 6‐week‐old rats following 3 weeks of treatment. (i) Quantification of PLP intensity normalized to untreated rats in the corpus callosum (p = 0.1886, t = 1.437, df = 8). (j) Representative images of ASPA staining within the corpus callosum. (k) Quantification of ASPA cell number normalized to untreated rats in the corpus callosum (p = 0.0013, t = 7.806, df = 8). (l) Representative images of NG2 staining within the corpus callosum. (m) Quantification of NG2 intensity normalized to untreated rats in the corpus callosum (p = 0.0006, t = 5.413, df = 8). Bolded points represent the average of individual animals and smaller, lighter points represent values from individual sections. All analysis was done using each animal as an n, and statistical significance was determined by unpaired, two‐tailed Student's t test. n = 5 animals per group and mean ± SEM. All scale bars = 50 μm. All schematics created with BioRender.com.

Next, oligodendrocyte maturation in the rostral corpus callosum was examined by PLP expression (Figure 1e). In contrast to the frontal cortex, PLP expression was not significantly different in the rostral corpus callosum of PrEP‐treated rats compared to control‐treated animals (Figure 1f). However, there were significantly fewer ASPA+ cells in the corpus callosum of PrEP‐treated rats (Figure 1g). Lastly, PrEP treatment significantly increased NG2+ intensity in the corpus callosum (Figure 1h). Taken together, PrEP significantly decreases the number of mature oligodendrocytes in the rostral cortex and corpus callosum of adolescent rats on PrEP. For all analysis, we have confirmed that vehicle exposure does not significantly impact the readout and is indistinguishable from untreated conditions (data not shown). All experiments were repeated in female rats, and PrEP had the same effect in both sexes (Figure S1).

2.2. PrEP Reversibly Inhibits Oligodendrocyte Maturation Independent of Cell Death In Vitro

Because PrEP inhibited oligodendrocyte maturation in vivo, the potential mechanisms underlying the impairment of oligodendrocyte differentiation were investigated in vitro. OPC cultures were purified and stimulated to differentiate into oligodendrocytes over a 72‐h period using a well‐established rat primary cell culture system (Feigenson et al. 2009). Oligodendrocyte maturation was examined in vitro using stage‐specific lineage markers for OPCs (A2B5), immature oligodendrocytes (galactocerebroside; GalC), and mature oligodendrocytes (proteolipid protein; PLP) following 72 h of differentiation (Figure 2a). Cultures were treated at the time of differentiation with the drugs composing PrEP both individually and together at concentrations corresponding to the reported human plasma C _max concentrations as well as 1/10th C _max and 3× C _max (0.7, 7, and 20 μM for FTC and 0.06, 0.6, and 1.7 μM for TDF) (Mathias 2012).

FIGURE 2.

PrEP reversibly inhibits oligodendrocyte maturation independent of cell death in vitro. (a) Schematic of in vitro differentiation and cell stage markers. Created with Biorender.com (b) Representative images of primary oligodendrocyte cultures treated for 72 h with DMSO or PrEP at various physiologically relevant concentrations (GalC, green, immature oligodendrocyte; PLP, magenta, mature oligodendrocyte marker; DAPI, blue, nuclear marker). (c) Representative images of primary oligodendrocyte cultures treated as in (b), OPC (A2B5, green), cell death (Draq7, magenta) and nucleus (DAPI, blue). (d) Quantification of GalC+ cells as a percentage of total cells (DAPI+) and normalized to untreated (one‐way ANOVA, F (8) = 89.47, p < 0.0001). (e) Quantification of PLP+ cells as a percentage of total cells (DAPI+) and normalized to untreated (one‐way ANOVA, F (8) = 56.85, p < 0.0001). (f) Quantification of A2B5+ cells as a percentage of total cells (DAPI+) and normalized to untreated (one‐way ANOVA, F (8) = 2.884, p = 0.1027). (g) Quantification of Draq7‐ cells as a percentage of total cells (DAPI+) and normalized to untreated (one‐way ANOVA, F (8) = 1.681, p = 0.2475). For (d–g) statistical significance determined by one‐way ANOVA followed by Bonferroni's multiple comparisons test. (h) Schematic of timing of treatment with PrEP C _max treatment and recovery period when cells were changed to differentiation media without PrEP C _max or DMSO. Representative images of GalC (green; immature oligodendrocyte), PLP (magenta; mature oligodendrocyte), and DAPI (blue; nucleus). (i) Quantification of GalC+ cells (top) or PLP+ cells (bottom) as a percentage of total cells (DAPI+) and normalized to untreated across four timepoints: 72 h (red) with PrEP or DMSO followed by 24, 48, or 72 h of differentiation medium without PrEP or DMSO. Statistical significance determined by two‐way ANOVA followed by Bonferroni's multiple comparisons test. GalC: Interaction (p = 0.0082, F (3, 16) = 5.580), main effect of PrEP (p < 0.0001, F (1, 16) = 37.11), and main effect of time (p = 0.0016, F (3, 16) = 8.127. PLP: Interaction (p < 0.0001, F (3, 16) = 17.90), main effect of PrEP (p < 0.0001, F (1, 16) = 131.4), and main effect of time (p < 0.0001, F (3, 16) = 17.97). For all, n = 3 biological replicates and error bars = mean ± SEM. All scale bars = 50 μm.

Treatment with either FTC or TDF significantly decreased the number of mature oligodendrocytes and did not change the number of OPCs or cell viability (Figure S2). As both FTC and TDF had the same effects on oligodendrocyte differentiation, we next examined the combined effect, PrEP. PrEP treatment across all three doses significantly decreased the number of GalC+ cells after 72 h compared to vehicle (Figure 2b,d). Similarly, the number of PLP+ cells was significantly decreased in PrEP treated cultures compared with vehicle (Figure 2b,e). Furthermore, the number of A2B5+ OPCs was not significantly different between PrEP treated and vehicle controls (Figure 2c,f). To determine if the decrease in immature and mature oligodendrocytes was due to cell death, we utilized Draq7, a cell impermeable dye, to determine cell viability. There were no significant differences in Draq7+ cells across PrEP treatments (Figure 2c,g). Finally, to determine if the effects of PrEP on oligodendrocyte maturation were reversible, OPCs were differentiated in the presence of PrEP C _max for 72 h as above, then changed to media without PrEP to allow for recovery. The number of GalC+ and PLP+ cells was examined by ICC after 24, 48, or 72 h of recovery (Figure 2h). Differentiation was comparable to vehicle control after 48 h of recovery for GalC+ and at 72 h for PLP+ (Figure 2i). Together, these data demonstrate that PrEP reversibly inhibits oligodendrocyte maturation rather than inducing cell death.

2.3. PrEP Treatment Impairs Lysosome Function Through Deacidification

As PrEP inhibited oligodendrocyte maturation both in vivo and in vitro, the mechanism mediating the inhibition of oligodendrocyte maturation was investigated. FTC and TDF treatment has been shown to deacidify lysosomes in neurons (Hui et al. 2019) and microglia (Tripathi et al. 2019) in vitro. Due to the acidic nature of the lysosomal lumen (pH 4–5), even drugs that are weakly basic such as TDF and FTC can result in a slight increase in the pH of lysosomes and inactivate pH sensitive enzymes. Lysosomes within PrEP treated oligodendrocytes were assessed using multiple complementary assays for lysosomal deacidification, functionality, and number. These included LysoSensor to assess lysosome pH, LysoTracker to assess the number of acidic organelles, DQ‐BSA to assess lysosome proteolytic function, and LAMP1 to assess endolysosomal number. Lysosomal pH was quantitatively determined using the ratiometric probe LysoSensor Yellow/Blue DND‐160. PrEP treatment resulted in a significant increase in the average pH of lysosomes compared to untreated oligodendrocyte cultures (Figure 3a). As LysoSensor measured the pH across all cell types in the oligodendrocyte cultures, we next examined the number of acidic organelles in GalC+ oligodendrocytes using LysoTracker. PrEP treatment significantly decreased the number of Lysotracker+ vesicles compared to vehicle across all three doses (Figure 3b,c). To determine the function of the lysosomes within PrEP treated oligodendrocytes, DQ‐BSA, a self‐quenching fluorescent BSA which only fluoresces upon proteolytic cleavage within functional lysosomes, was used in GalC+ oligodendrocytes. Oligodendrocytes treated with PrEP had significantly decreased DQ‐BSA intensity than vehicle across all PrEP treatments (Figure 3d,e). Finally, to determine if lysosome number changed, LAMP1 was used as an endolysosomal marker within GalC+ oligodendrocytes to assess lysosome (including endosomes) number independent of pH. PrEP treatment significantly decreased LAMP1 intensity in GalC+ oligodendrocytes at the C _max and 3× C _max doses (Figure 3f,g). Thus, PrEP treatment significantly decreased the number of functional and acidic lysosomes in oligodendrocytes. The ability of FTC or TDF to individually deacidify lysosomes and impair lysosome function was also examined. Both FTC and TDF decreased the number, function, and acidity of lysosomes within oligodendrocytes (Figure S3). As both FTC and TDF contributed to the effects of PrEP, the rest of the experiments were carried out using the combination of the two drugs, referred to as PrEP.

FIGURE 3.

PrEP treatment impairs lysosome function through deacidification. (a) Quantification of lysosome pH by LysoSensor in PrEP or DMSO treated cultures (one‐way ANOVA, F (4, 10) = 10.58, p = 0.0013). (b) Representative images of Lysotracker (green) in GalC+ (magenta) oligodendrocytes in DMSO or PrEP C _max treated cultures after 72 h of treatment. (c) Quantification of the intensity of Lysotracker in GalC+ oligodendrocytes normalized to untreated (one‐way ANOVA, F(3, 8) = 285.5, p < 0.0001). (d) Representative images of DQ‐BSA (green) in GalC+ (magenta) oligodendrocytes in DMSO or PrEP C _max treated cultures after 72 h of treatment. (e) Quantification of the intensity of DQ‐BSA in GalC+ oligodendrocytes normalized to untreated (one‐way ANOVA, F (3, 8) = 39.85, p < 0.0001). (f) Representative images of GalC+ oligodendrocytes (magenta) and LAMP1 puncta (green) in DMSO or PrEP C _max treated cultures after 72 h of treatment. (g) Quantification of the intensity of LAMP1 staining in GalC+ oligodendrocytes normalized to untreated (one‐way ANOVA, F (3, 8) = 20.95, p < 0.0004). For all analysis, statistical significance was determined by one‐way ANOVA followed by Bonferroni's multiple comparisons test. n = 3 biological replicates and error bars = mean ± SEM. All scale bars = 10 μm.

2.4. PrEP Treatment Prevents Cell Surface Expression of PLP In Vitro

To determine how PrEP‐induced lysosome deacidification inhibited oligodendrocyte differentiation, the transportation of myelin proteins to the membrane of differentiating oligodendrocytes was examined. First, the myelin protein PLP was examined as during differentiation, it is transported to the plasma membrane as a synthesized protein through the endolysosomal pathway (Trajkovic et al. 2006; Shen et al. 2016; Feldmann et al. 2011). Furthermore, PLP insertion into the plasma membrane from the lysosome requires the pH‐sensitive protease cathepsin D, which has an 80% decrease in proteolytic activity when lysosomal pH is above 5 (Barrett 1970). PLP+ oligodendrocytes treated with PrEP were significantly smaller than vehicle‐treated oligodendrocytes as determined by Sholl analysis (Figure 4a,b); however, PLP intensity per cellular area was significantly greater in PrEP‐treated oligodendrocytes compared to vehicle (Figure 4c). As we hypothesized that PLP transport is affected by PrEP treatment, we verified that the cellular area of PrEP‐treated cells was, in general, smaller than untreated by also performing Sholl analysis on cells labeled with GalC, which stains immature and mature oligodendrocytes (Figure 4d). There was also a significant decrease in cell size, although the area labeled by GalC+ was slightly larger than that labeled by PLP+(Figure 4e). Total GalC intensity was not significantly different in PrEP‐treated cultures compared to vehicle‐treated (Figure 4f). Whole cell western blot analysis of PLP protein expression demonstrated that PrEP‐treated cultures had similar total cellular PLP expression to vehicle (Figure 4g). PrEP treatment resulted in significantly fewer PLP+ oligodendrocytes (Figure 2e), suggesting that there is more PLP per cell in PrEP‐treated cultures compared with vehicle treatment. MBP total cellular protein expression was also examined. Unlike PLP, MBP is transported to the plasma membrane as mRNA and locally translated (Trajkovic et al. 2006). Total cellular MBP was significantly less in PrEP C _max and 3× C _max treated oligodendrocytes compared with vehicle (Figure 4h). To determine if our observation that PrEP‐treated oligodendrocytes exhibited an increase in PLP per oligodendrocyte was due to an intracellular accumulation of PLP, surface PLP protein expression was examined by cell surface biotinylation assay. PrEP‐treated cultures had significantly less cell surface PLP protein expression compared to vehicle‐treated cultures across all three PrEP doses (Figure 4i). Thus, PrEP treatment inhibits cell surface expression of PLP during differentiation.

FIGURE 4.

PrEP treatment inhibits cell surface expression of PLP in vitro. (a) Representative images of PLP+ oligodendrocytes in cultures treated with DMSO or PrEP for 72 h. (b) Quantification of the size of PLP+ oligodendrocytes by Sholl analysis normalized to untreated (two‐way ANOVA, interaction (p = 0.0660, F (21, 64) = 1.646), main effect of PrEP (p < 0.0001, F (3, 64) = 10.55), and main effect of distance (p < 0.0001, F (7, 64) = 14.00)followed by Bonferroni's multiple comparisons test. (c) Quantification of PLP intensity per cellular area normalized to untreated (one‐way ANOVA, F (3, 8) = 9.397, p = 0.0053). (d) Representative images of GalC+ oligodendrocytes in cultures treated with DMSO or PrEP for 72 h. (e) Quantification of the size of GalC+ oligodendrocytes by Sholl analysis normalized to untreated (two‐way ANOVA, interaction (p = 0.0329, F (21, 64) = 1.773), main effect of PrEP (p < 0.0001, F (3, 64) = 18.22), and main effect of distance (p < 0.0001, F (7, 64) = 11.84) followed by Bonferroni's multiple comparisons test. (f) Quantification of GalC intensity per cellular area normalized to untreated (one‐way ANOVA, F (3, 8) = 0.5386, p = 0.6690). (g) Immunoblot of PLP in DMSO and PrEP treated cultures and quantification of PLP protein per alpha‐tubulin protein normalized to untreated (one‐way ANOVA, F (3, 8) = 1.197, p = 0.3711). (h) Immunoblot of MBP in control, DMSO and PrEP treated cultures and quantification of MBP protein per alpha‐tubulin protein normalized to untreated (one‐way ANOVA, F (3, 8) = 14.80 p < 0.0013). (i) Immunoblot of PLP from whole cell protein lysates or cell surface protein lysates of DMSO and PrEP treated cultures. Quantification of the ratio of PLP in cell surface samples compared to PLP protein in whole cell (total PLP) samples normalized to untreated (one‐way ANOVA, F (3, 8) = 27.37, p < 0.0001). For all analysis, Bonferroni's multiple comparisons test. n = 3 biological replicates and error bars = mean ± SEM. All scale bars = 10 μm.

2.5. Acidic Nanoparticle Treatment Prevents Lysosome Impairments Within PrEP‐Treated Oligodendrocytes In Vitro

As PrEP deacidifies lysosomes, we next tested if acidification of lysosomes at the time of PrEP treatment could prevent PrEP‐induced inhibition of oligodendrocyte maturation. Poly (lactic‐co‐glycolic acid) (PLGA) acidic nanoparticles (aNPs) were co‐administered at the time of PrEP treatment. PLGA nanoparticles are FDA approved and are characterized by low cytotoxicity and high bioavailability (Blasi 2019). PLGA nanoparticles are localized to the lysosomes through endocytosis and hydrolyzed into lactic acid and glycolic acid (Zeng et al. 2019; Baltazar et al. 2012). As in Figure 2, 1/10th C _max, C _max, and 3× C _max PrEP treatment resulted in significantly fewer GalC+ and PLP+ oligodendrocytes compared with control (Figure 5c,d). Cultures treated with PrEP and aNPs (Figure 5a) had significantly more PLP+ (Figure 5c) and GalC+ (Figure 5d) oligodendrocytes compared to PrEP only. Lastly, aNPs colocalized with lysotracker within GalC positive oligodendrocytes (Figure 5e). Thus, acidic nanoparticle cotreatment prevents PrEP‐induced inhibition of oligodendrocyte maturation.

FIGURE 5.

Acidic nanoparticles prevent PrEP‐induced inhibition of oligodendrocyte maturation in vitro. (a) Representative images of primary rat oligodendrocyte cultures treated with DMSO or PrEP for 72 h at various physiologically relevant concentrations (GalC, green; PLP, magenta; DAPI, blue). (b) Representative images of primary rat oligodendrocytes cultures (GalC, green, PLP, magenta, and DAPI, blue) treated with DMSO and aNPs or PrEP and aNPs for 72 h at various physiologically relevant concentrations. (c) Quantification of GalC+ cells as a percentage of total cells (DAPI+) and normalized to untreated in cultures treated with DMSO or PrEP with and without aNPs (two‐way ANOVA, interaction (p = 0.0007, F (3, 16) = 9.602), PrEP (p = 0.0009, F (3, 16) = 9.301), and aNPs (p = 0.0016, F (1, 16) = 59.31). (d) Quantification of PLP+ cells as a percentage of total cells (DAPI+) and normalized to untreated in cultures treated with DMSO or PrEP with and without aNPs (two‐way ANOVA, interaction (p < 0.0001, F (3, 16) = 27.75), PrEP (p < 0.0001, F (3, 16) = 50.64), and acidic nanoparticles (p < 0.0001, F (1, 16) = 157.7). (c, d) Clear bars represent vehicle treatment and shaded bars represent acidic nanoparticle treatment. (e) Representative image of CY5 tagged aNPs (magenta) and lysosomes (Lysotracker; green) in GalC+ (turquoise) untreated oligodendrocytes. For all analyses, statistical significance determined by two‐way ANOVA followed by Bonferroni's multiple comparisons test. n = 3 biological replicates and error bars = mean ± SEM. Scale bars = 50 μm in (a) and (b) and 10 μm in (e).

As previous work has demonstrated that aNPs are able to reacidify lysosomes (Zeng et al. 2019; Baltazar et al. 2012), the ability of aNPs to prevent PrEP‐induced lysosome impairments was investigated. Acidic nanoparticles administered at the time of PrEP treatment significantly increased LAMP1 intensity compared to C _max and 1/10th C _max PrEP alone (Figure 6a,b). Next, the number of acidic organelles was determined using LysoTracker. aNP treatment at the time of PrEP treatment significantly increased LysoTracker intensity compared to PrEP alone across all three doses (Figure 6c,d). Thus, aNPs prevent PrEP‐induced decreases in the total number of lysosomes and the number of acidic lysosomes.

FIGURE 6.

Acidic nanoparticle treatment partially prevents lysosome impairments within PrEP‐treated oligodendrocytes. (a) Representative images of PLP+ oligodendrocytes (magenta) and LAMP1 puncta (green) in rat primary oligodendrocyte cultures 72 h after treatment with DMSO or PrEP C _max with and without aNPs. (b) Quantification of the intensity of LAMP1 staining in PLP+ oligodendrocytes normalized to untreated (two‐way ANOVA, interaction (p = 0.2339, F (3, 16) = 1.577), PrEP (p < 0.0001, F (3, 16) = 16.43), and acidic nanoparticles (p = 0.0049, F (1, 16) = 10.65). (c) Representative images of Lysotracker (green) in GalC+ (magenta) oligodendrocytes in rat primary oligodendrocyte cultures 72 h after treatment with DMSO or PrEP C _max with and without aNPs. (d) Quantification of the intensity of Lysotracker in GalC+ oligodendrocytes normalized to untreated (two‐way ANOVA, interaction (p = 0.0416, F (3, 16) = 3.454), PrEP (p < 0.0001, F (3, 16) = 38.61), and acidic nanoparticles (p = 0.0002, F (1, 16) = 24.20). For all analysis, statistical significance determined by two‐way ANOVA followed by Bonferroni's multiple comparisons test. n = 3 biological replicates and error bars = mean ± SEM. All scale bars = 10 μm.

Furthermore, we wanted to determine if acidic nanoparticle treatment at the time of PrEP treatment was able to prevent PrEP‐induced lysosome dysfunction and deacidification. As above, we used DQ‐BSA to measure the proteolytic function of lysosomes. Oligodendrocytes treated with PrEP and aNPs had significantly greater DQ‐BSA intensity than PrEP alone across all three treatments (Figure 7a,b). Finally, as the data in Figure 6 suggest, there are fewer acidic organelles and functional lysosomes in the context of PrEP, we quantitatively determined the pH of these acidic organelles, including lysosomes, using the ratiometric probe LysoSensor Yellow/Blue DND‐160. aNPs co‐treatment with PrEP resulted in a significant decrease in the average pH of lysosomes (optimal pH of 4.5–5.0) compared to PrEP alone across all three treatment doses within oligodendrocytes (Figure 7g). Thus, aNPs prevent PrEP‐induced lysosome dysfunction and deacidification in oligodendrocytes.

FIGURE 7.

Acidic nanoparticle treatment partially prevents PrEP‐induced lysosome deacidification and dysfunction in oligodendrocytes. (a) Representative images of DQ‐BSA (green) in GalC+ (magenta) in rat primary oligodendrocyte cultures 72 h after treatment with DMSO or PrEP C _max with and without aNPs. (b) Quantification of the intensity of DQ‐BSA in GalC+ oligodendrocytes normalized to untreated (two‐way ANOVA, interaction (p = 0.8892, F (3, 16) = 0.2082), PrEP (p = 0.0009, F (3, 16) = 9.219, and acidic nanoparticles (p < 0.0001, F (1, 16) = 95.98). (c) Quantification of lysosome pH by Lysosensor normalized to untreated in rat primary oligodendrocyte cultures 72 h after treatment with DMSO or PrEP with and without aNPs (two‐way ANOVA, interaction (p = 0.0054, F (3, 16) = 5.097), PrEP (p < 0.0001, F (3, 16) = 18.96), and acidic nanoparticles (p < 0.0001, F (1, 16) = 50.26). For all analysis, statistical significance determined by two‐way ANOVA followed by Bonferroni's multiple comparisons test. n = 3 biological replicates and error bars = mean ± SEM. All scale bars = 10 μm.

2.6. Acidic Nanoparticle Treatment Prevents PLP Accumulation Within Lysosomes

As acidic nanoparticle cotreatment prevented PrEP induced lysosome deacidification, the ability of acidic nanoparticles to restore cell surface expression of PLP in the presence of PrEP was examined. As in Figure 4a,b, PLP per cell area was significantly greater in PrEP treated cells compared to vehicle (Figure 8a,b). When acidic nanoparticles were co‐administered with PrEP, PLP intensity per cell area was not significantly different from vehicle (Figure 8a,b). Recapitulating the results in Figure 4c, PLP+ oligodendrocytes were significantly smaller upon PrEP treatment compared to vehicle treated oligodendrocytes. This effect was ablated when aNPs were administered at the time of PrEP treatment as PLP+ oligodendrocytes were significantly larger compared to PrEP alone treated oligodendrocytes as assessed by Sholl analysis (Figure 8a,c). Finally, using the same cell surface biotinylation assay as in Figure 4, cell surface PLP was analyzed by western blot. Once again, PrEP treatment alone recapitulated Figure 4i as PrEP C _max treated oligodendrocytes had significantly less cell surface PLP compared to vehicle treated oligodendrocytes. PrEP and aNPs co‐treated oligodendrocytes had significantly more cell surface PLP compared to PrEP treatment alone across all three doses of PrEP treatment (Figure 8d,e). Therefore, PrEP treatment reduces PLP surface expression and aNP treatment at the time of PrEP is sufficient to restore cell surface PLP protein expression through prevention of lysosome deacidification and dysfunction.

FIGURE 8.

Acidic nanoparticle treatment prevents PrEP‐induced intracellular PLP accumulation. (a) Representative images of PLP+ oligodendrocytes in rat oligodendrocyte cultures treated for 72 h with DMSO or PrEP with and without aNPs. (b) Quantification of PLP intensity per cellular area normalized to untreated (two‐way ANOVA, interaction (p = 0.0024, F (3, 16) = 7.499), PrEP (p = 0.0009, F (3, 16) = 9.280), and acidic nanoparticles (p < 0.0001, F (1, 16) = 48.64). (c) Quantification of the size of PLP+ oligodendrocytes by Sholl analysis normalized to untreated (two‐way ANOVA, interaction (p = 0.0348, F (21, 56) = 1.851), distance (p = 0.1221, F (2.512, 20.09) = 2.248), and treatment (p = 0.2476, F (1, 16) = 1.680). DMSO vs. C _max at 35.0645 uM p = 0.0473. (d) Immunoblot of PLP from whole cell lysates or cell surface lysates from rat oligodendrocyte cultures treated for 72 h with DMSO or PrEP with and without aNPs. (e) Quantification of the ratio of PLP protein in cell surface samples to whole cell (total) PLP protein and normalized to untreated (two‐way ANOVA, interaction (p = 0.0057, F (1, 8) = 13.99), PrEP (p < 0.0046, F (1, 8) = 15.17), and acidic nanoparticles (p < 0.0025, F (1, 8) = 18.86). For all analysis, statistical significance determined by two‐way ANOVA followed by Bonferroni's multiple comparisons test. n = 3 biological replicates and error bars = mean ± SEM. All scale bars = 10 μm.

2.7. Acidic Nanoparticles Prevent PrEP‐Induced Inhibition of Oligodendrocyte Maturation in the Frontal Cortex of Adolescent Rats

As aNPs can prevent PrEP‐induced inhibition of oligodendrocytes in vitro, we determined if co‐administration of aNPs at the time of PrEP treatment could prevent the PrEP‐induced inhibition of oligodendrocyte maturation in our adolescent rat model (Figure 1). As aNPs cannot cross the blood‐brain barrier, aNPs were administered intranasally. Previous work has demonstrated that intranasal administration of aNPs is a successful way to deliver aNPs to the brain (Mamik et al. 2016; Sharma et al. 2015; Xiao et al. 2013).

First, intranasal delivery of aNPs was confirmed in sagittal sections of 6‐week‐old rats 1 h following intranasal administration of fluorescent PLGA nanoparticles (Figure 9a). The effects of PrEP treatment on oligodendrocyte maturation in the frontal cortex recapitulated observations in Figure 1 such that PrEP decreased PLP intensity, decreased the number of oligodendrocytes labeled with ASPA, and increased the intensity of NG2, a marker of oligodendrocyte progenitor populations (Figure 9b–g). Furthermore, aNP treatment at the time of PrEP treatment (PrEP+aNPs) prevented PrEP‐mediated inhibition of oligodendrocyte maturation as measured by an increase in PLP intensity and ASPA+ cell number compared with PrEP treatment alone (Figure 9b,c,e,f). This finding was complemented by a significant decrease in NG2 intensity in the frontal cortex of PrEP+aNPs treated animals as compared to PrEP only (Figure 9d,g) suggestive of a decrease in oligodendrocyte progenitor cells. Thus, aNP treatment prevents PrEP‐induced inhibition of oligodendrocyte maturation and PLP expression in the frontal cortex of adolescent rats.

FIGURE 9.

Acidic nanoparticle co‐treatment mitigates PrEP‐induced reduction in PLP and oligodendrocyte number and induction of NG2+ oligodendrocyte precursor cells in the frontal cortex of adolescent rats. (a) Schematic of intranasal administration of aNPs in rats. Inserted image is of acidic nanoparticles (magenta) in OPCs (NG2+, green) following intranasal administration in the olfactory bulb of 6‐week‐old rats. Created with Biorender.com. (b–d) All images were taken, and analysis performed at 6‐weeks of age following 3 weeks of treatment with DMSO, DMSO + aNPs, PrEP, PrEP+aNPs, or untreated. (b) Representative images of PLP staining within the frontal cortex. (c) Representative images of ASPA staining within the frontal cortex. (d) Representative images of NG2 staining within the frontal cortex. (e) Quantification of PLP intensity normalized to untreated rats in the frontal cortex (two‐way ANOVA, interaction (p = 0.6352, F (1, 16) = 0.2339), PrEP (p = 0.7489, F (1, 16) = 0.1060), and acidic nanoparticles (p = 0.1363, F (1, 16) = 2.460). (f) Quantification of ASPA cell number normalized to untreated rats in the frontal cortex (two‐way ANOVA, interaction (p = 0.0095, F (1, 16) = 8.688), PrEP (p = 0.0147, F (1, 16) = 7.418), and acidic nanoparticles (p = 0.0027, F (1, 16) = 12.61). (g) Quantification of NG2 intensity normalized to untreated rats in the frontal cortex (two‐way ANOVA, interaction (p = 0.0071, F (1, 16) = 9.517), PrEP (p < 0.0001, F (1, 16) = 34.78), and acidic nanoparticles (p = 0.0562, F (1, 16) = 4.237). Bolded points represent average of individual animals and smaller, lighter points represent values from individual sections. All analysis was done using each animal as an n and statistical significance was determined by two‐way ANOVA followed by Bonferroni's multiple comparisons test. n = 5 animals per group and mean ± SEM. Scale bars = 10 μm in (a) and 50 μm in (b–d).

2.8. Acidic Nanoparticles Prevent PrEP‐Induced Inhibition of Oligodendrocyte Maturation in the Corpus Callosum of Adolescent Rats

Next, the ability of aNPs to prevent PrEP inhibition of oligodendrocyte maturation in the rostral corpus callosum was examined. Once again, PrEP treatment compared to vehicle recapitulated the results in Figure 1; PrEP did not alter PLP intensity but decreased the number of ASPA positive cells and increased the intensity of NG2, a marker of oligodendrocyte progenitor populations (Figure 10b–g). As there was no change in PLP intensity between PrEP and vehicle, there was no difference between rats that received aNPs at the time of PrEP treatment and PrEP only‐treated animals (Figure 10b,e). While aNPs+PrEP treatment did not result in a significant increase in ASPA+ cell number compared to PrEP alone, ASPA+ cell number in PrEP+aNPs was increased but was not significantly different than vehicle treated animals (Figure 10c,f). Furthermore, NG2 intensity in PrEP+aNPs‐treated animals was not significantly different from vehicle and PrEP+aNPs‐treated animals had significantly less NG2 intensity in the rostral corpus callosum compared to PrEP alone (Figure 10d,g). Thus, aNP treatment prevented significant PrEP‐induced changes in oligodendrocyte number in the corpus callosum.

FIGURE 10.

Number of oligodendrocytes in the corpus callosum are indistinguishable between adolescent rats co‐treated with acidic nanoparticles and PrEP and rats treated with vehicle as a control despite no observed differences in PLP. (a) Representative images of PLP staining within the corpus callosum. (b) Representative images of ASPA staining within the corpus callosum. c. Representative images of NG2 staining within the corpus callosum. (a–c) All analysis and images were done at 6‐weeks‐old following 3 weeks of treatment with DMSO, DMSO+aNPs, PrEP, PrEP+aNPs, or untreated. (d) Quantification of PLP intensity normalized to untreated in the corpus callosum (two‐way ANOVA, interaction (p = 0.0045, F (1, 16) = 10.90), PrEP (p < 0.0001, F (1, 16) = 53.78), and acidic nanoparticles (p = 0.0032, F (1, 16) = 11.95). (e) Quantification of ASPA cell number normalized to untreated rats in the corpus callosum (two‐way ANOVA, interaction (p = 0.2903, F (1, 16) = 1.196), PrEP (p < 0.0059, F (1, 16) = 10.08), and acidic nanoparticles (p = 0.0247, F (1, 16) = 6.144). (f) Quantification of NG2 intensity normalized to untreated rats in the corpus callosum (two‐way ANOVA, interaction (p = 0.0098, F (1, 16) = 8.591), PrEP (p < 0.0001, F (1, 16) = 33.61), and acidic nanoparticles (p = 0.0068, F (1, 16) = 9.635). Bolded points represent average of individual animals and smaller, lighter points represent values from individual sections. All analysis was done using each animal as an n and statistical significance was determined by two‐way ANOVA followed by Bonferroni's multiple comparisons test. n = 5 animals per group and mean ± SEM. All scale bars = 50 μm.

3. Discussion

HIV PrEP remains a key tool for ending the HIV epidemic in the United States. Prescriptions for FTC/TDF PrEP increased from 8800 in 2012 to nearly 220,000 in 2018 (Harris et al. 2019; Sullivan et al. 2018). As more and more people take PrEP, minimizing any damaging side effects of the drug becomes critical. The goal of this study is to determine the best form of PrEP; one that is highly effective against HIV acquisition and minimizes negative effects on the CNS. Our prior studies suggest that select antiretroviral drugs from several different classes can inhibit oligodendrocytes from differentiating in vitro (Festa et al. 2019, 2023; Roth et al. 2020; Jensen et al. 2015) and oligodendrocyte differentiation and remyelination after a demyelinating event in vivo (Festa et al. 2023; Roth et al. 2020; Jensen et al. 2015). The effects of the drugs composing PrEP on oligodendrocytes have not yet been examined. Here we examined the effects of FTC/TDF PrEP on oligodendrocyte differentiation during adolescence, identified a mechanism for this effect, and examined the potential for acidic nanoparticles to mitigate that effect.

We show that adolescent rats orally gavaged with PrEP daily from 3 to 6 weeks old had significantly fewer mature oligodendrocytes (ASPA+) and significantly more oligodendrocyte precursor cells (OPCs; NG2) in the cortex and corpus callosum compared to vehicle (Figure 1). Despite the decrease in the number of oligodendrocytes upon PrEP treatment, the intensity of the myelin protein PLP was only significantly decreased in the cortex of PrEP‐treated adolescent rats compared to vehicle controls (Figure 1). As myelination proceeds from caudal to rostral, it is possible that the corpus callosum was already myelinated at the time of treatment, whereas the cortex, which is more rostral, was not. Additionally, white matter tracts tend to be myelinated during early childhood (Reynolds et al. 2019) and gray matter areas continue to be myelinated through adolescence (Grydeland et al. 2019; Yakovlev, and Lecours (1967)), which would also support the idea that the corpus callosum was already myelinated at the time of PrEP treatment, whereas the cortex was still being myelinated. These data suggest the importance of the timing of PrEP exposure on brain regions that are still undergoing myelination and may determine which regions are impacted. Furthermore, they suggest that PrEP may not affect existing myelin but rather inhibits oligodendrocyte differentiation, which is supported by our in vitro data, although the impact on myelination specifically warrants further investigation. There are two alternative hypotheses. First, it is possible that already differentiated oligodendrocytes can compensate for the reduced number of oligodendrocytes by producing additional PLP‐containing myelin sheaths in this region (Bacmeister et al. 2020). This is something that remains to be examined in the context of PrEP. Second, it is possible that there is a significant loss of PLP within the corpus callosum, but it is not able to be observed due to the density of oligodendrocytes within the corpus callosum compared to the sparsity within the cortex. These findings are the first, to our knowledge, to examine the effects of ART on the oligodendrocyte population during the adolescence phase of development in vivo and raise intriguing implications for the treatment of humans with these compounds during phases of development in which oligodendrocytes are differentiating and white matter is developing.

Mechanistically, we determined that PrEP inhibits oligodendrocyte maturation by deacidifying lysosomes and decreasing the number of functional and acidic lysosomes in vitro (Figure 3). This work supports reports by others demonstrating that treatment with FTC or TDF alone deacidifies lysosomes in neurons (Hui et al. 2019) and microglia (Tripathi et al. 2019) in vitro. The two antiretrovirals composing PrEP, FTC and TDF, are nucleoside mimics and weak bases, which are known to accumulate within lysosomes and deacidify the lysosomal lumen (Poole and Ohkuma 1981). This mechanism accounts for the ability of ART drugs from distinct classes, such as bictegravir, saquinavir, and darunavir, to inhibit oligodendrocyte differentiation by deacidifying lysosomes (Festa et al. 2023). Additionally, previous work from our group has demonstrated that ART drugs can impair the integrated stress response (Roth et al. 2020) and induce oxidative stress in oligodendrocytes (Jensen et al. 2015). How oxidative stress and the ISR affect lysosomal dysfunction remains to be examined. The lysosome participates in cell signaling with both the ER and mitochondria; it is possible that lysosome dysfunction could activate the ISR in the ER, oxidative stress in the mitochondria, or both. This remains to be investigated in oligodendrocytes in the context of ART, as well as in other conditions. Furthermore, it is possible that PrEP‐induced deacidification of lysosomes has secondary effects due to lysosome signaling. For instance, mechanistic target of rapamycin (mTOR) is located on the lysosome and is responsible for coordinating cell growth with cellular nutrient, energy, and growth factor levels. mTOR can form a complex known as mTORC1, which initiates lipid synthesis, a requirement for proper myelination (Kelly et al. 2014; Leite et al. 2013). In oligodendrocytes, balanced mTORC1 activity is required for proper CNS myelination, as both too little and too much mTORC1 activity results in hypomyelination (Tate et al. 2011). Lysosomes regulate mTORC1 activity depending on the pH of the lysosome. Thus, it is possible that PrEP treatment could also affect the mTOR signaling pathway. Together, these results, with previous work from our group (Festa et al. 2019, 2023) and others (Folts et al. 2016), suggest that maintaining acidic lysosomal pH is critical for oligodendrocyte differentiation. Furthermore, these results extend beyond PrEP and have implications for other weak base drugs that have been shown to deacidify endolysosomes (Halcrow et al. 2024), as they may also impair oligodendrocyte differentiation during adolescence via lysosome deacidification and dysfunction, which warrants further investigation.

Because FTC or TDF have been shown to deacidify lysosomes in neurons (Hui et al. 2019) and microglia (Tripathi et al. 2019), it is possible that there is an indirect effect on oligodendrocyte differentiation, mediated by those cell types or from astrocytes, although the effect of PrEP on astrocytes has not been examined to our knowledge. However, our in vitro data is consistent with our in vivo findings and the starting material in our cultures is approximately 95% oligodendrocyte progenitor cells with less than 5% astrocytes, few microglia, and no surviving neurons. As our data cannot rule out an effect of other neural cell types, future studies are needed to address the impact of PrEP drugs on astrocytes and microglia.

Our observation that lysosome number is decreased following deacidification is unexpected. Lysosome deacidification has been shown to result in reduced mTORC1 (Hu et al. 2016) signaling and nuclear translocation of transcription factor EB (TFEB) (Fedele and Proud 2020). TFEB is the master transcription regulator of lysosome biogenesis (Sardiello et al. 2009), and its function depends on its cellular localization (Napolitano et al. 2022). Under non‐stress conditions, mTORC1 on the lysosome phosphorylates TFEB, and it remains in the cytosol (Puertollano et al. 2018). In contrast, when the lysosome is stressed, such as deacidification, mTORC1 detaches from the lysosome, and as a result, TFEB is not phosphorylated and can translocate to the nucleus, and genes crucial to lysosome biogenesis are transcribed (Napolitano et al. 2022). Based on this alone, following lysosome deacidification, lysosome number might be expected to be greater than or equal to control. However, the aforementioned experiments were not done in oligodendrocytes. Among their many attributes that are distinct from other cell types, oligodendrocytes require high levels of cholesterol (Saher et al. 2005) in order to produce the lipid‐rich oligodendrocyte membrane that forms myelin. Evidence from HEK293t cells indicates that mTORC1 remains attached to the lysosome under conditions of high lysosomal cholesterol, retaining TFEB in a phosphorylated state, resulting in cytosolic localization and reducing lysosome biogenesis (Castellano et al. 2017). Thus, it is possible that any expected increase in lysosome number through TFEB following deacidification is countered due to activation of mTORC1 by cholesterol and ultimately inactivation of TFEB. An alternative explanation for the decrease in lysosomes is simply that we examined lysosome number following three days of drug treatment. As a result, it is possible that there is an initial increase in lysosome biogenesis by TFEB following deacidification, but by the three‐day timepoint, this increase has subsided due to prolonged lysosome stress.

To our knowledge, our results are also the first to demonstrate the impact of ART drugs on PLP trafficking via the lysosomes. We found PLP to be retained within oligodendrocytes and not trafficked to the cell membrane during differentiation in the presence of PrEP. We hypothesize that this is due to endolysosomal dysfunction as it is restored by co‐treatment with acidic nanoparticles. Oligodendrocyte differentiation is uniquely vulnerable to impairment via lysosome dysfunction as production of the myelin sheath involves a several thousand‐fold (Pfeiffer et al. 1993) expansion of the area of the oligodendrocyte plasma membrane. This process requires trafficking a large volume of lipids and myelin proteins to the plasma membrane through the endolysosomal pathway for placement in the nascent myelin membrane. PLP is the most abundant of these lysosomally transported proteins (Gongvatana 2011; Corrêa et al. 2015). The intracellular signals triggering the release of PLP from endosome/lysosome stores remain to be determined. However, it has been shown that cathepsin D, a pH‐sensitive lysosomal protease, is necessary for the release of PLP from the endosome/lysosome stores to the cell membrane (Guo et al. 2018). When cathepsin D is knocked down in oligodendrocytes, PLP accumulates within late endosomes/lysosomes and inhibits oligodendrocyte maturation (Guo et al. 2018). Our data demonstrate that treatment with PrEP increases the lysosomal pH to a range that would reduce cathepsin D activity by 80% (Barrett 1970), consistent with this as a mechanism for our observed reduction in surface PLP.

In vitro experiments demonstrated that PrEP reversibly inhibits oligodendrocyte maturation independent of cell death. This leads to the question, what happens when an individual stops taking PrEP? We anticipate that based on our in vitro ‘washout’ experiments that oligodendrocyte maturation would also recover in vivo following the termination of PrEP administration. Future experiments should examine if oligodendrocyte differentiation recovers in adult rats that received PrEP during adolescents.

The ability of aNPs to rescue this phenotype both in vivo and in vitro represents a novel therapy to be combined with weak base drugs. In our previous works (Festa et al. 2019, 2023), we were able to prevent inhibition of oligodendrocyte maturation by several ART drugs in vitro using an agonist of the lysosomal transmembrane non‐selective ion channel TRPML1 (mucolipin transient receptor potential channel 1), which also results in reacidification of the lysosome. Additionally, we previously reported that TRPML1 agonism prevented pH changes beyond appropriate ranges in the context of these ART drugs (Festa et al. 2023). However, as TRPML1 has multiple roles within cells, including release of cholesterol from intralysosomal stores and participation in calcium signaling (Zeevi et al. 2007), we sought to determine if direct manipulation of lysosome pH using PLGA aNPs was sufficient to prevent PrEP‐induced inhibitions of oligodendrocyte maturation. Neutral PLGA nanoparticles are only hydrolyzed into the acidic components lactic acid and glycolic acid within acidic cellular compartments such as the lysosome (Zeng et al. 2019; Baltazar et al. 2012). As a result, aNPs represent a potential therapeutic approach to be combined with PrEP to prevent white matter impairments in adolescents.

While the objective is different in the context of HIV infection, intranasal administration of nanoparticles loaded with ART has been shown to increase the efficacy of ART delivery to infected macrophages within the brain (Zhou et al. 2024). FTC and TDF are two of the three drugs that compose the frontline combination of ART for PWH. In this case, it is imperative that ART is able to cross the blood‐brain barrier to prevent HIV reservoirs in the brain, as HIV alone is known to have CNS impairments (Alakkas et al. 2019; Gongvatana et al. 2011; Corrêa et al. 2015; Leite et al. 2013). Thus, intranasal delivery of acidic nanoparticles loaded with ART serves as a potential therapy to both increase delivery of ART to infected cells within the brain and to minimize effects of the drugs on non‐infected cells such as the oligodendrocyte.

Even though HIV PrEP is a highly effective strategy to prevent the acquisition of HIV (Mayer et al. 2020; Grant et al. 2014; Castillo‐Mancilla et al. 2013), it is still possible for an individual to acquire HIV while on PrEP, as PrEP is not 100% effective at preventing the acquisition of HIV and strict adherence is required for PrEP to be effective (Grant et al. 2014; Castillo‐Mancilla et al. 2013; Cottrell et al. 2016), raising the question as to what effects PrEP could have on individuals who acquire HIV after having been on PrEP. Studies have shown that as much as 50% of people with HIV (PWH) whose viral load is adequately suppressed by antiretroviral drugs still show behavioral and cognitive deficits, a condition known as HIV‐associated neurocognitive deficits or HAND (Saylor et al. 2016). Antiretroviral agents themselves have been implicated in contributing to HAND, which includes thinning of the corpus callosum, loss of structural integrity, and dysregulation of transcripts for myelin proteins Borjabad et al. (2011); Corrêa et al. 2015; Kelly et al. 2014; Leite et al. 2013; Tate et al. 2011; Solomon et al. (2020). Therefore, it is possible that adolescent PrEP use could negatively affect an individual that acquires HIV later in life, which would be important to interrogate as PrEP use expands.

Finally, it is important to examine the impact of other antiretroviral compounds on white matter development, function, and integrity that are used as part of PrEP. Our findings suggest the importance of considering the dose strength and duration. Further, our observation that changes in oligodendrocyte differentiation are reversible also encourages consideration of dosing strategies of PrEP that permit recovery of white matter during days off the compounds. Recent changes to the suggested drug regimen for partners of people with HIV now include other antiretroviral drugs and multiple dosing strategies such as long‐acting injectables and on‐demand or 2‐1‐1, two pills 24 h before and 1 pill for the next 2 days after engaging in behaviors that increase the risk of acquiring HIV. One such long‐lasting drug, cabotegravir, is delivered by an intramuscular injection once a month for the first 2 months and then once every 2 months. The use of long‐acting injectable antiretrovirals, like cabotegravir, emphasizes the importance of understanding the potential CNS effects of these drugs as they can remain in the body for up to 12 months after a single dose in long‐lasting injectable form (Spreen et al. 2014). Additionally, the effects of on‐demand versus daily administration (modeled here) must be examined in the future to determine if sporadic use alleviates the effects on white matter observed here or if, albeit short‐term, a higher dosage of PrEP is equal to or worse for oligodendrocyte differentiation in adolescents.

Taken together, these results highlight the importance of continuing to develop PrEP antiretroviral drugs that not only prevent the acquisition of HIV but minimize CNS damage. In an HIV‐negative individual, it is not necessary for PrEP antiretroviral drugs to cross the blood‐brain barrier as there is no HIV in the CNS in these individuals. One possibility moving forward would be to develop a PrEP cocktail that is composed of antiretrovirals that are not blood‐brain barrier permeable. Finally, concurrent therapies can be administered, such as intranasal acidic nanoparticles, to prevent PrEP‐induced CNS impairments in adolescents without HIV.

4. Materials and Methods

4.1. Reagents

Reagents and chemicals used were LysoTracker Red DND‐99 (ThermoFisher Scientific; L7528), LysoSensor Yellow/Blue DND‐160 (ThermoFisher Scientific; L7545), DAPI (ThermoFisher Scientific; D1306), DQ‐BSA Red (ThermoFisher Scientific; D12051), Pierce Cell Surface Biotinylation and Isolation Kit (ThermoFisher Scientific, A44390), PLGA nanoparticles (Phosphorex; LGFG100, LG100), emtricitabine (MedChemExpress; HY‐17427), and tenofovir disoproxil fumarate (MedChemExpress; HY‐13782).

4.2. Antibodies

The following antibodies were used for immunohistochemical analysis of brain tissue sections: aspartoacylase (ASPA; mature oligodendrocytes; GeneTex; GTX110699; 1:500), neural/glial antigen 2 (NG2; OPCs; Millipore; AB5320; 1:200) and proteolipid protein (PLP; mature oligodendrocytes; AA3 rat hybridoma; 1:1). The following antibodies were used for in vitro immunocytochemistry: A2B5 (OPCs; mouse IgM hybridoma; 1:1) (Eisenbarth et al. 1979), galactocerebroside (GalC; immature oligodendrocytes; mouse IgG3 hybridoma; 1:5) (Raff et al. 1978), proteolipid protein (PLP; mature oligodendrocytes; AA3 rat hybridoma; 1:1), and LAMP1 (Enzo Life Sciences; ADI‐VAM‐EN001‐F; 1:100). The following antibodies were utilized for immunoblotting: myelin basic protein (MBP; Biolegend; 808,401; 1:1000), PLP (AA3 rat hybridoma, 1:1000), and alpha tubulin (α‐tubulin; Sigma‐Aldrich; T5168; 1:10,000).

4.3. Animal Use

All experiments were conducted in accordance with the guidelines set forth by the Children's Hospital of Philadelphia Institutional Animal Care and Use Committee.

4.4. Mass Spectrometry Study

Rats were used to determine the maximum plasma concentration of TDF and FTC via mass spectrometry. Ten rats underwent oral gavage and were treated with one of five doses of FTC (8.57, 17.14, 34.28, 51.42, or 68.56 mg/kg) and TDF (12.87, 25.47, 51.48, 77.22, or 102.96 mg/kg). Blood was collected via the submandibular vein 3 h following oral gavage, and plasma was isolated and processed for mass spectrometry analysis. Based on the results, animals were treated with 64.35 mg/kg TDF and 17.14 mg/kg FTC from 3 to 6 weeks old.

4.5. Oral Gavage of PrEP

Fifteen male and 15 female rats were daily orally gavaged from 3 to 6 weeks old. Five untreated animals were allowed to remain in the cage and were not handled. Five vehicle treated animals were orally gavaged 5% DMSO at the same volume as PrEP treated rats. Five PrEP treated animals were orally gavaged PrEP (64.35 mg/kg TDF and 17.14 mg/kg FTC) at the same time each day. Animals were weighed every other day and the doses of TDF and FTC were adjusted based on the weight of the animals. Animals never received a volume greater than 10 mL/kg by oral gavage.

4.6. Intranasal Administration of PLGA Nanoparticles

PLGA nanoparticles (Phosphorx; cat # LGFG100) were administered intranasally daily in male rats from 3 to 6 weeks old. 20 μL of 37.5 mg/mL PLGA nanoparticle solution was dissolved in saline and administered drop by drop in both nostrils of the rat using a gel loading pipette tip and pipettor (Musumeci et al. 2018). 20 μL dose was administered in (4) 5 μL doses, alternating between nostrils, and each dose was given drop by drop to allow the animal to inhale. Rats were scuffed in a supine position, with care taken to ensure that the neck and head of the animal were held parallel to the floor. Animals were monitored for any signs of respiratory distress, and researchers observed the rats in their home cage following administration.

4.7. Immunohistochemistry

Rats were deeply anesthetized with isoflurane and intracardially perfused with ice‐cold PBS followed by 4% paraformaldehyde (pH 7.4). Brains were removed and the olfactory bulb and cerebellum were removed using a brain matrix. Brains were postfixed in 4% paraformaldehyde for 24 h followed by immersion in a 30% sucrose solution for 72 h. Brains were cryopreserved and embedded in optimal cutting temperature compound (OCT) followed by coronal or sagittal sectioning (12 μm) on a cryostat (Leica Microsystems, Exton, PA). Sections were blocked in 10% normal goat serum for 1 h at RT and then incubated overnight with primary antibodies diluted in 2% normal goat serum in PBS. Sections were rinsed three times in 1X PBS and incubated with fluorescent secondary antibodies in 1X PBS (1:200) for 1 h at RT. Sections were rinsed three times in 1X PBS prior to being stained with DAPI (1:10,000) for 5 min and mounted with ProLong Gold anti‐fade reagent (ThermoFisher Scientific; P36930). Images were acquired on a DMi8 Leica inverted confocal microscope (Leica Microsystems) using a 40× objective/1.3 NA.

4.8. Primary Rat Oligodendrocyte Precursor Cell Cultures

Primary rat cells were isolated from P1–4 Sprague Dawley rats (Charles River Laboratories, Wilmington, MA) and plated on T75 flasks (Jensen et al. 2015). Upon reaching confluency, OPCs were purified using the “shake‐off” method (McCarthy and De Vellis 1980). Briefly, T75 flasks were rotated on an orbital shaker set to 250 rpm and incubated overnight at 37°C. The following day, cells were filtered using a 20 μm nylon mesh (Merck Millipore, Darmstadt, Germany) and centrifuged at 1500 rpm for 5 min at 4°C. The supernatant was discarded, and the pellet was resuspended in 5 mL Neurobasal media supplemented with B27 and incubated on a bacteriological petri dish for 15 min at 37°C. The supernatant was collected and centrifuged at 1500 rpm for 5 min at 4°C. The pellet was resuspended in Neurobasal media supplemented with B27 and growth factors: PDGF (2 ng/mL), NT3 (1 ng/mL), and FGF (10 ng/mL) and plated on 24‐well plates with coverslips (immunocytochemistry), 10 cm dishes (immunoblotting), or a 96‐well plate (lysosomal pH measurements).

4.9. Drug Treatment

Primary rat OPCs were grown in 24‐well plates with coverslips or 10 cm dishes for immunocytochemistry or immunoblotting respectively until they reached approximately 70% confluency. To differentiate OPCs into mature oligodendrocytes, growth medium was replaced with differentiation medium containing 50% DMEM, 50% Ham's F12, pen/strep, 2 mM glutamine, 50 μg/mL transferrin, 5 μg/mL putrescine, 3 ng/mL progesterone, 2.6 ng/mL selenium, 12.5 μg/mL insulin, 0.5 μg/mL T4, 0.3% glucose, and 10 ng/mL biotin. At the time of differentiation, cells were treated with vehicle (DMSO), emtricitabine (700 nM, 0.7 μM, and 7 μM), tenofovir disoproxil fumarate (0.6 μM, 6 μM, and 20 μM) prior to staining or protein collection. On average, typically 30%–50% of OPC differentiation into mature oligodendrocytes is observed in untreated conditions.

4.10. Immunoblotting

Whole cell extracts of primary rat oligodendrocytes were prepared in ice cold lysis buffer (25 mM Tris (pH 7.4), 10 mM EDTA, 10% SDS, 1% Triton‐X 100, 150 mM NaCl, protease and phosphatase inhibitor cocktail) followed by sonication and centrifugation at 10,000 rpm at 4°C for 30 min. Protein concentrations were determined by Pierce BCA Protein Assay (ThermoFisher Scientific, 23,225). Protein (7.5 μg) was loaded onto a 4%–12% Bis‐Tris gradient gel and after separation, proteins were transferred to Immobilin‐FL membranes. Membranes were blocked in NAP (Non Animal Protein)‐BLOCKER(G‐BIOSCIENCES, 786‐190)1:1 with TBS‐T for 1 h at RT and then incubated overnight at 4°C with primary antibodies in 1:2 NAP in TBS‐T. Following three washes in TBS‐T, membranes were incubated with fluorescent probe‐conjugated secondary antibodies in 1:2 NAP in TBS‐T. Membranes were visualized using an Odyssey Infrared Imaging System (LiCOR). Densitometric analysis of band intensities was conducted using ImageJ. All bands were normalized to the loading control.

4.11. Immunocytochemistry

Following drug treatment, coverslips were washed three times in DMEM/F12 and incubated in primary antibodies to surface antigens (A2B5 and/or GalC) for 30 min at RT. Coverslips were then rinsed three times with DMEM/F12 and incubated in appropriate fluorescent conjugated secondary antibodies for 30 min at RT. Cells were then rinsed three times in DMEM/F12, fixed in 4% paraformaldehyde for 8 min, and permeabilized with 0.1% Triton‐X 100 for 5 min. Cells were then incubated in primary antibodies to internal antigens (PLP and LAMP1) for 30 min at RT. Coverslips were rinsed three times in 1X PBS and incubated in appropriate secondary antibodies for 30 min at RT. Lastly, cells were rinsed, counterstained with DAPI (1:10,000) for 5 min, and then mounted on slides with ProLong Gold anti‐fade reagent. For Draq7 staining, the same steps were followed with the omission of permeabilization. Cells were imaged either using a Keyence BZ‐X‐700 digital fluorescent microscope (Keyence Corporation) for OPC and oligodendrocyte cell counts or a DMi8 Leica inverted confocal microscope equipped with a 40× objective/1.3 NA. Quantification of oligodendrocyte lineage counts was conducted by hand counting the number of OPCs, immature oligodendrocytes, mature oligodendrocytes, and DAPI+ cells across 10 fields/coverslips and 3 coverslips per condition for each biological replicate. The percentage of each cell type was calculated and normalized to the untreated condition.

4.12. LysoSensor Yellow/Blue Experiments

After 72 h of differentiation and drug treatments, differentiation medium was removed and fresh media with 5 μM LysoSensor Yellow/Blue was added for 5 min. Cells were then rinsed three times in isotonic solution (105 mM sodium chloride, 5 mM potassium chloride, 10 mM HEPES buffer, 5 mM sodium bicarbonate, 60 mM mannitol, 5 mM glucose, 0.5 mM magnesium chloride, and 1.3 mM calcium chloride) and then left in isotonic solution for imaging. For the generation of the standard curve, pH standards (4.0–6.5) were generated by preparing a 10 μM H⁺/Na⁺ ionophore monensin and 20 μM H⁺/K⁺ ionophore nigericin dissolved in 20 mM 2‐(N‐morpholino) ethane sulfonic acid (MES), 110 mM potassium chloride, and 20 mM sodium chloride followed by the adjustment of pH with hydrogen chloride and sodium hydroxide. Data was generated by exciting each well at 340 nm and 380 nm with emission at 520, and the ratio between each excitation was calculated followed by extrapolation based on the standard curve.

4.13. LysoTracker Experiments

Following 72 h of drug treatment, cells were incubated with 50 nM LysoTracker Red DND‐99 in fresh differentiation medium for 2 h at 37°C. After this incubation, cells were rinsed three times with DMEM/F12, and immunocytochemistry for oligodendrocyte lineage markers was performed as described above. Images were acquired on a DMi8 Leica inverted confocal microscope equipped with a 40×/1.3 NA objective and 3× optical multiplier (120× total magnification). Imaging fields were chosen at random, and Z‐stacks from 10 fields across 3 coverslips/condition for each biological replicate were acquired. The channel of interest was separated and thresholded to isolate the puncta (0.5–1.0 μm²). The number of puncta per cell was quantified using the particle analysis plugin in ImageJ.

4.14. DQ‐BSA Red Experiments

Differentiated oligodendrocytes were treated with 10 μg/mL DQ‐BSA in fresh differentiation medium for 6 h at 37°C. Cells were then rinsed three times with DMEM/F12 and then stained for oligodendrocyte lineage markers as outlined above. Images were acquired on a DMi8 Leica inverted confocal microscope equipped with a 40×/1.3 NA objective and 3× optical multiplier (120× total magnification). Imaging fields were chosen at random, and Z‐stacks from 6 fields/coverslips and 3 coverslips/condition for each biological replicate were acquired. The channel of interest was separated, and a threshold was set to isolate the puncta (0.5–1.0 μm²). The number of puncta per cell was quantified using the particle analysis plugin in ImageJ.

4.15. Cell Surface Biotinylation Assay and Immunoblot

After 72 h of differentiation, cell surface proteins of oligodendrocyte cultures were biotinylated using the Pierce Cell Surface Biotinylation and Isolation Kit (ThermoFisher Scientific, A44390). Cells were washed with ice‐cold PBS, incubated with biotin substrate (EZ‐Link Sulfo‐NHS‐SS‐Biotin) for 10 min at room temperature, and whole cell extracts of primary rat oligodendrocytes were prepared in ice‐cold lysis buffer (25 mM Tris (pH 7.4), 10 mM EDTA, 10% SDS, 1% Triton‐X 100, 150 mM NaCl, protease and phosphatase inhibitor cocktail), and the cell pellet was transferred to ice for a 30 min incubation. At this step, a fraction of the sample was saved as the whole cell protein sample, and total protein concentrations were determined by the Pierce BCA Protein Assay (ThermoFisher Scientific, 23,225). Cell surface biotinylated proteins were extracted using a NeutrAvidin Agarose column and then eluted to form the cell surface protein samples. The same total protein concentrations were added to the column for capture. Both total protein and biotinylated samples were run on the same gel for each treatment group. Protein (7.5 μg) was loaded onto a 4%–12% Bis‐Tris gradient gel, and after separation, proteins were transferred to Immobilin‐FL membranes. Membranes were blocked in NAP (Non Animal Protein)‐BLOCKER(G‐BIOSCIENCES, 786‐190)1:1 with TBS‐T for 1 h at RT and then incubated overnight at 4°C with primary antibodies in 1:2 NAP in TBS‐T. Following three washes in TBS‐T, membranes were incubated with fluorescent probe‐conjugated secondary antibodies in 1:2 NAP in TBS‐T. Membranes were visualized using an Odyssey Infrared Imaging System (LiCOR). Densitometric analysis of band intensities was conducted using ImageJ. All bands were normalized to the loading control.

4.16. Experimental Design and Statistical Analysis

The number of animals used for in vivo studies was calculated by power analysis using G* Power version 3.1.9.4 (Faul et al. 2009). Sample sizes were calculated using an independent measures study design with α = 0.05 and power = 0.8. Male rats were randomly assigned to groups. All in vitro experiments were performed with at least n = 3 biological replicates, and each biological replicate used OPCs derived from a different litter of P0–P4 rat pups. Outliers were neither encountered nor removed for any analysis. Data are represented as mean ± SEM for in vitro and in vivo experiments. A p < 0.05 was considered statistically significant. P and n are reported in the figure legends. All statistical analysis was performed with GraphPad Prism, version 9.0 (GraphPad Software).

Author Contributions

C.C.L., J.B.G., and K.L.J.‐S. developed the project. C.C.L. designed, performed, and analyzed all experiments. L.K.F. aided and provided guidance across all experiments. M.C.‐B. assisted C.C.L. with daily oral gavage and intranasal treatments in rats from 3 to 6 weeks old (Figures 9 and 10). T.D.J. conducted DQ‐BSA experiments in the context of PrEP and PLGA nanoparticles with guidance from C.C.L. C.H.M. provided the PLGA nanoparticles for initial characterization experiments and provided expertise on their uses. C.C.L., J.B.G., and K.L.J.‐S. wrote the manuscript. All authors provided feedback on the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1. Supporting Information.

Long, C. C. , Festa L. K., Cruz‐Berrios M., et al. 2025. “Acidic Nanoparticles Prevent HIV Pre‐Exposure Prophylaxis (PrEP)‐Induced Oligodendrocyte Impairments by Restoring Lysosomal pH in Adolescent Models.” Glia 73, no. 10: 1967–1988. 10.1002/glia.70050.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.