Atorvastatin differentially regulates the interactions of cocaine and amphetamine with dopamine transporters

Abstract

The function of dopamine transporters (DAT) is regulated by membrane cholesterol content. A direct, acute removal of membrane cholesterol by methyl-β-cyclodextrin (MβCD) has been shown to reduce dopamine (DA) uptake and release mediated by DAT. This is of particular interest because a few widely prescribed statins that lower peripheral cholesterol levels are blood-brain barrier (BBB) penetrants, and therefore could alter DAT function through brain cholesterol modulation. The goal of this study was to investigate the effects of prolonged atorvastatin treatment (24 hr) on DAT function in neuroblastoma 2A (N2A) cells stably expressing DAT. We found that atorvastatin treatment effectively lowered membrane cholesterol content in a concentration-dependent manner. Moreover, atorvastatin treatment markedly reduced DA uptake and abolished cocaine inhibition of DA uptake, independent of surface DAT levels. These deficits induced by atorvastatin treatment were reversed by cholesterol replenishment. However, atorvastatin treatment did not change amphetamine (AMPH)-induced DA efflux. This is in contrast to a small but significant reduction in DA efflux induced by acute depletion of membrane cholesterol using MβCD. This discrepancy may involve differential changes in membrane lipid composition resulting from chronic and acute cholesterol depletion. Our data suggest that the outward-facing conformation of DAT, which favors the binding of DAT blockers such as cocaine, is more sensitive to atorvastatin-induced cholesterol depletion than the inward-facing conformation, which favors the binding of DAT substrates such as AMPH. Our study on statin-DAT interactions may have clinical implications in our understanding of neurological side effects associated with chronic use of BBB penetrant statins.

1. Introduction

The dopamine transporter (DAT) is an integral plasma membrane protein containing 12 transmembrane helices, an extracellular N-terminus, and an intracellular C-terminus. The DAT localizes to plasma membranes of dopaminergic dendrites and axon terminals and functions to terminate dopamine (DA) transmission by taking up extracellular DA into presynaptic dendrites and axon terminals. Dysregulation of DAT function is associated with abnormal DA transmission observed in various neurological diseases, including substance use disorder, Parkinson’s disease, and attention deficit/hyperactive disorder [see review in (Vaughan & Foster 2013)].

The DAT is the site of action for psychostimulants, including cocaine and amphetamine (AMPH). As a DAT blocker, cocaine binds to the DAT to inhibit DA reuptake, resulting in an increase in extracellular DA levels in the brain. As a DAT substrate, AMPH is taken up into the cells by the DAT and binds directly to the vesicular monoamine transporter 2 on DA-containing vesicles, inducing DA release from vesicles into the cytoplasm through a carrier-mediated exchange (Sulzer et al. 2005; Freyberg et al. 2016). The DAT then reversely transports cytoplasmic DA into the synaptic cleft - a process called DA efflux - resulting in increased extracellular DA levels. Based on the crystal structure and computational modeling of two bacterial DAT homologues, the leucine transporters and multi-hydrophobic amino acid transporters, the DAT is presumed to constantly adopt an outward- or an inward-facing conformation to accommodate the high-affinity binding of DAT blockers (e.g., cocaine) or substrates (e.g., DA and AMPH), respectively (Kazmier et al. 2014; Krishnamurthy & Gouaux 2012; Malinauskaite et al. 2014; Shi et al. 2008; Shaikh & Tajkhorshid 2010; Wang et al. 2015). Notably, the crystal structure of an inhibitor-bound Drosophila DAT reveals the presence of a cholesterol molecule, which is hypothesized to stabilize the inhibitor-bound, outward-facing conformation of the DAT (Penmatsa et al. 2013). Further, using atomistic molecular dynamics simulations, the DAT was shown to undergo structural changes from an outward-facing to an inward-facing conformation in the absence of cholesterol; accordingly, these conformational changes are prevented in the presence of cholesterol (Zeppelin et al. 2018). Therefore, cholesterol homeostasis appears to critical for maintaining an equilibrium between the outward- and inward-facing states. Accumulating evidence further indicates the role of cholesterol in the regulation of DAT function. For example, acute cholesterol depletion with methyl-β-cyclodextrin (MβCD), a chelator of membrane cholesterol, markedly reduces DA uptake in various cultured cell lines and in striatal synaptosomes prepared from animal brains (Jones et al. 2012; Adkins et al. 2007; Morissette et al. 2018). Moreover, cholesterol replenishment to cholesterol-depleted HEK293 cells restores DAT uptake function (Jones et al. 2012). Importantly, cholesterol addition to rat striatal synaptosomes increases the binding of the cocaine analog [¹²⁵I]RTI-55 (Hong & Amara 2010), suggesting that the DAT favors the outward-facing conformation in an enriched membrane cholesterol environment. Lastly, cholesterol depletion by MβCD reduces AMPH-induced DA efflux in HEK293 cells (Jones et al. 2012). These data suggest that cholesterol influences the kinetics of DAT conformational states, and therefore the efficacy and potency of DAT ligands.

The DAT-cholesterol interaction has significant clinical implications, as cholesterol-lowering statins - in particular, blood-brain barrier (BBB) penetrant statins such as atorvastatin and simvastatin - may affect brain cholesterol metabolism. Though DAT uptake function is disrupted upon membrane cholesterol removal by MβCD, a major limitation of such an approach is that cholesterol depletion by MβCD is fast-acting (~30 min after treatment) and thus is considered an acute process. This process does not mimic statin-induced peripheral cholesterol reduction, which takes several months to occur. Moreover, statins and MβCD remove cholesterol through different mechanisms. Thus, the goal of the current study was to investigate whether atorvastatin, a lipophilic statin, would effectively reduce membrane cholesterol content and disrupt DAT function. Cholesterol is synthesized in a multistep enzymatic process starting with acetyl-CoA (Russell 1992). Acetyl-CoA and acetoacetyl-CoA are converted to 3-hydroxy-3-methylglutaryl (HMG)-CoA, which is further reduced to mevalonate by HMG-CoA reductase (HMGCR). Through a series of enzymatic reactions, mevalonate is converted to squalene, lanosterol, and finally to cholesterol. The HMGCR, a rate-limiting enzyme of cholesterol synthesis, is the primary pharmacological target of statin drugs. Statins competitively bind to the catalytic domain of HMGCR, blocking the conversion of HMG-CoA to mevalonate, and therefore reducing cholesterol production (McFarland et al. 2014; Istvan & Deisenhofer 2001). Because circulating cholesterol does not cross the BBB, brain cholesterol is produced locally. Cholesterol is primarily synthesized in oligodendrocytes during brain development and thereafter is mostly synthesized in astrocytes (Dietschy & Turley 2004; Pfrieger 2003). It has been reported that chronic use of statins, especially BBB penetrant statins such as atorvastatin and simvastatin, is associated with cognitive impairment and depression in humans (Kirsch et al. 2003; While & Keen 2012; Schilling et al. 2017). However, the relationship between statin treatment, DAT function, and DA transmission has yet to be established. Thus, the main goal of this study was to investigate the effects of prolonged atorvastatin exposure on DAT function in neuroblastoma cells stably expressing human DAT (hDAT). Cocaine and AMPH were used to examine the outward- and inward-facing conformation of DAT, respectively. The effects of acute membrane cholesterol depletion by MβCD on DAT uptake and release function were assessed as a comparison to the prolonged effects of membrane cholesterol depletion by atorvastatin.

2. Materials and methods

2.1. Chemicals

Atorvastatin (ATE959438188), MβCD (C4555), water-soluble cholesterol (cholesterol-loaded MβCD, ~40–45 mg of cholesterol/g of MβCD, C4951), and dopamine (H8502) were purchased from Millipore Sigma (St. Louis, MO). Lipoprotein depleted fetal bovine serum (LPDS, #880100) was purchased from KALEN Biomedical LLC (Germantown, MD). [³H]DA (NET116000, specific activity, 54.1 Ci/mmol; PerkinElmer) was obtained from PerkinElmer Life and Analytical Sciences (Waltham, MA). D-amphetamine (A3278) was purchased from Millipore Sigma. Cocaine was obtained from National Institute on Drug Abuse (Bethesda, MD). Other chemicals were purchased from Thermo Fisher Scientific unless otherwise specified.

2.2. Stable cell line and cell culture

The neuroblastoma 2a (N2A) cell line stably expressing hDAT (N2A-hDAT), a generous gift from Dr. Margaret Gnegy (University of Michigan, Ann Arbor, MI), was characterized previously (Chen et al., 2013; Furman et al., 2009). Cells were cultured in Dulbecco’s modified Eagle’s (DMEM) medium (Corning Life Science, Corning, NY) supplemented with 10% fetal bovine serum (FBS), 100 units/ml penicillin, and 100 μg/ml streptomycin. The stable expression of hDAT was selected using 400 μg/ml geneticin G418 (Invitrogen).

2.3. Atorvastatin and MβCD treatment

To remove exogenous cholesterol from the culture medium and increase the effectiveness of atorvastatin treatment on cholesterol reduction, cells were treated with vehicle or atorvastatin (0.1, 0.3, 1 or 10 μM) for 24 hr in DMEM supplemented with 10% lipoprotein-deficient serum. These doses were chosen based on previous publications (Kolyada et al. 2001; Dong et al. 2011; Warita et al. 2014). Atorvastatin was dissolved in dimethyl sulfoxide (DMSO). The final concentration of DMSO across all experiments was ≤0.1%. The same concentration of DMSO was applied to vehicle-treated cells. For acute depletion of membrane cholesterol, cells were washed with serum free DMEM three times, followed by incubation with MβCD (2.5, 5 or 7.5 mM) or vehicle for 30 min at 37°C in DMEM. These doses were chosen based on a previous publication showing their effectiveness in removal of membrane cholesterol (Jones et al. 2012).

2.4. Cell viability assay

Following treatment with atorvastatin (0.1–10 μM) or vehicle for 24 hr in lipid-deficient serum, cells were washed with PBS twice, and cell viability was determined using a trypan blue exclusion assay. Cells were stained with trypan blue solution (Thermo Fisher Scientific, #15250061) for 3 min at room temperature, and the cells were rinsed once with PBS and counted using a hematocytometer under a light microscope. Cells that were trypan blue-positive were considered nonviable, whereas cells that were trypan blue-negative were considered viable. The percentage of viable cells was calculated using the following formula: viable cells (%) = (total viable cells/total cells) x100. Experiments were performed four independent times.

2.5. Amplex Red cholesterol assay

Unesterified cholesterol content in N2A-hDAT cells was measured using an Amplex Red cholesterol assay kit (Invitrogen, A12216). Briefly, after treatment with atorvastatin (0.1, 0.3, 1 or 10 μM for 24 hr), MβCD (2.5, 5 or 7.5 mM for 30 min) or an appropriate vehicle, cells were harvested and lysed in cholesterol reaction buffer (0.1 M potassium phosphate, pH 7.4, 50 mM NaCl, 25 mM cholic acid, and 0.1% Triton X-100) included in the Amplex Red cholesterol assay kit. Cells were then centrifuged at 12,000 × g for 20 min at 4°C. The supernatant was collected and incubated with Amplex Red working solution (150 μm Amplex Red, 1 U/ml cholesterol oxidase, and 1 U/ml horseradish peroxidase dissolved in cholesterol reaction buffer) at 37°C for 30 min. As a result of the reaction, the red-fluorescent oxidation product, resorufin, was produced and detected at 560 nm using a Victor3 plate reader (Perkin Elmer, Waltham, MA). Some experiments were performed in the presence of cholesterol esterase to determine the total cholesterol content. Cholesterol content was calculated based on the linear regression of the cholesterol standard curve and normalized to the amount of total protein. Protein concentration was measured using a Pierce BCA Protein Assay Kit (ThermoFisher Scientific, #23225).

2.6. [³H]DA uptake

To examine the effect of atorvastatin or MβCD treatment on DA uptake function of DAT, [³H]DA uptake was performed as described previously (Furman et al. 2009). Briefly, cells were washed with Krebs-Ringer-HEPES buffer (KRH) (25 mM HEPES, pH 7.4, 125 mM NaCl, 4.8 mM KCl, 1.2 mM KH₂PO₄, 1.3 mM CaCl₂, 1.2 mM MgSO₄, and 5.6 mM glucose) following atorvastatin (0.3, 1 or 10 μM, 24 hr), MβCD (5 mM, 30 min) or a corresponding vehicle treatment. Then, cells were incubated with 15 nM [³H]DA in the presence of increasing concentrations of non-labeled DA (10 nM - 10 μM) supplemented with 250 μM ascorbic acid to prevent DA oxidation. The DA uptake assay was performed at room temperature and was terminated 10 min after incubation by rapidly washing the cells three times using cold KRH buffer. The nonspecific uptake of [³H]DA was determined in the presence of 300 μM DA. Next, cells were lysed in a solubilization buffer (50 mM Tris, 150 mM NaCl and 1% Triton X-100). The radioactivity retained in these cells was counted in CytoSint Liquid Scintillation Cocktail (MP Biomedicals, SKU: 01882453-CF) using Beckman LS 5801 liquid scintillation counter (Beckman Coulter, Fullerton, CA). The DA binding affinity (K_m) for the DAT and the maximal DA uptake velocity (V_max) were determined by the nonlinear regression analysis using the Michaelis-Menten equation. The V_max values were normalized to protein content and are presented as pmole/mg protein/min.

To determine whether atorvastatin-induced deficit in DA uptake was due to reduced membrane cholesterol content, cells were replenished with water-soluble cholesterol (0.1 – 1 mM) for 2 hr at 37°C following atorvastatin (1 μM) or vehicle treatment. These concentrations and the duration of cholesterol treatment were chosen based on previous publications (Morissette et al. 2018; Zidovetzki & Levitan 2007; Sponne et al. 2004; Jones et al. 2012). Data are presented as relative to the control condition.

2.7. Cocaine inhibition of [³H]DA uptake

To determine the effects of atorvastatin or MβCD treatment on the ability of cocaine to inhibit DA uptake, cells were treated with atorvastatin (1 μM, 24 hr), MβCD (5 mM, 30 min) or an appropriate vehicle followed by pre-incubation with increasing concentrations of cocaine (100 nM − 10 μM) for 10 min at room temperature before [³H]DA uptake assay as described previously (Guptaroy et al. 2011). The nonspecific uptake of [³H]DA was determined in the presence of 100 μM cocaine. To examine whether treatment with water-soluble cholesterol (250 μM, 2 hr at 37°C) could reverse the effect of atorvastatin, cells were incubated with vehicle or water-soluble cholesterol (250 μM) for 2 hr at 37°C before the cocaine inhibition of [³H]DA uptake assay. The IC₅₀ values were determined by the nonlinear regression analysis using the Log[inhibitor] vs. normalized response model. Data are presented as the percentage of specific radioactivity relative to the vehicle treatment.

2.8. AMPH-stimulated [³H]DA efflux

The effect of atorvastatin or MβCD treatment on AMPH-stimulated DA efflux was examined using a procedure described previously (Jones et al. 2012; Furman et al. 2009). Briefly, cells plated on 6-well plates were treated with atorvastatin (1 μM, 24 hr), MβCD (5 mM, 30 min) or an appropriate vehicle followed by preloading with 30 nM [³H]DA and 1 μM non-labeled DA in KRH buffer containing ascorbic acid (250 μM) at room temperature for 30 min. Cells were then quickly rinsed with KRH. Next, 700 μL of KRH were incubated with cells for 5 min followed by removal and replenishment with the same volume of fresh KRH. We repeated the wash step three times to establish a stable DA baseline. Then, cells were incubated with AMPH (1 μM or 10 μM) or vehicle (KRH) for 5 min followed by solution removal, addition of fresh KRH and incubation for another 5 min. This procedure was repeated four times for a total of 20 min following AMPH stimulation. The [³H]DA radioactivity in each collection was measured. Data are presented as a percent of basal DA radioactivity, which is defined as the last basal count of DA radioactivity right before AMPH treatment.

2.9. Biotinylation of surface DAT

Basal surface levels of DAT in atorvastatin or vehicle treated cells were measured by surface biotinylation as described previously (Chen et al. 2013; Chen et al. 2009). After 24 hr of atorvastatin (1 μM) or vehicle treatment, cells were washed with cold PBS/Ca/Mg buffer and incubated with membrane non-permeable sulfo-NHS-SS-biotin (1.5 mg/mL in PBS/Ca/Mg, Thermo Fisher Scientific, #21331) for 1.5 hr at 4°C. Excessive biotin was quenched by 0.1 M glycine in PBS for 15 min at 4°C. Next, the cells were lysed in a solubilization buffer containing protease inhibitor cocktails (Millipore Sigma, #P8465). The biotinylated proteins were pulled down using strepavidin beads (Santa-Cruz, sc-2003) at 4°C overnight with gentle agitation. The biotinylated proteins were eluded from the beads with sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) loading buffer containing dithiothreitol (DTT). Biotinylated and total lysate proteins were separated on a 10% SDS-PAGE gel with electrophoresis and transferred to nitrocellulose membranes. The membrane was blocked with 5% non-fat milk in Tris-buffered saline containing 1% Tween-20 (TBST) and then probed with rat anti-DAT (Millipore Sigma, MAB369) overnight at 4°C. Next, the membrane was incubated with goat anti-rat IgG-HRP (Cell Signaling Technology, 7077S) followed by application of Pierce Chemiluminescence Substrate. The band intensity was captured using Bio-Rad Chemi Doc-Touch system and quantified using FIJI software (NIH). The intensity of the biotinylated DAT band was normalized to that of the total DAT, whereas the intensity of the DAT band from lysates was normalized to that of GAPDH. Data are presented relative to the vehicle group.

2.10. Western blotting of HMGCR proteins

Cells treated with atorvastatin or vehicle were lysed in the solubilization buffer containing the protease cocktail as described above. Proteins were separated on an 8% SDS-PAGE gel with electrophoresis and transferred to nitrocellulose membranes. The membrane was blocked with 5% non-fat milk in Tris-buffered saline containing 1% Tween-20 (TBST) and then incubated with mouse anti-HMGCR (Santa Cruz, sc-271595) overnight at 4°C. Next, mouse IgGkappa BP-HRP (Santa Cruz, sc-516102) was added to the membrane followed by an application of Pierce Chemiluminescence Substrate. GAPDH was used as an internal control and probed with mouse anti-GAPDH (Santa Cruz, sc-365062). The intensity of the HMGCR band was normalized to that of GAPDH. Data are presented relative to the vehicle condition.

2.11. Data analysis

Graph Pad Prism 9 (La Jolla, CA, USA) was used for statistical analysis. All data are presented with scatterplots when possible. The outlier test (Rout, Q=1%) was performed for each data set and significant outliers were removed from further analyses if applicable. The nonlinear regression analysis was performed to calculate the values of IC₅₀, K_m and V_max. Paired Student t-tests were used to examine the total and free cholesterol content in N2A cells. A one-way or two-way ANOVA was used to examine the main effects of atorvastatin, cholesterol replenishment, MβCD, cocaine, and AMPH on DA uptake dynamics or DA efflux. Post hoc Bonferroni or Sidak’s multiple comparisons tests were performed when appropriate. A value of p≤ 0.05 was considered statistically significant.

3. Results

3.1. Atorvastatin treatment reduced membrane cholesterol content in a concentration-dependent manner

To determine whether atorvastatin treatment effectively reduced cholesterol content, cells were treated with vehicle or atorvastatin (0.1, 0.3, 1, or 10 μM) for 24 hr in lipoprotein-deficient medium. In the periphery, cholesterol is present in unesterified (free, membrane) and esterified (storage, cytosolic) forms. Unesterified cholesterol is localized within plasma membranes and is biologically active, whereas esterified cholesterol (cholesteryl ester) is present as lipid droplets in the plasma for storage, and therefore is not directly involved in the membrane function [see review in (Axmann et al. 2019)]. Using an Amplex Red cholesterol assay kit, we measured both unesterified (membrane) and total cholesterol content in the absence and presence of cholesterol esterase, respectively, in drug naïve N2A cells. There was no significant difference in the total and membrane cholesterol content (Fig. 1A, N=7, paired t-test), suggesting that cholesterol is not stored and is instead utilized in N2A cell membranes. Therefore, only membrane cholesterol was measured in atorvastatin-treated cells (Fig. 1B). A one-way ANOVA revealed a significant main effect of atorvastatin concentration on membrane cholesterol content, F (4, 30) = 14.69, p<0.0001. When compared to vehicle treatment (8.40 ± 0.56 ng/mg protein, N=7), a post hoc Bonferroni test indicated a significant reduction of membrane cholesterol content by atorvastatin at concentrations of 0.3 μM (6.15 ± 0.50 ng/mg protein, N=7, p<0.01), 1 μM (4.97 ± 0.37 ng/mg protein, N=7, p<0.001), and 10 μM (4.38 ± 0.30 ng/mg protein, N=7, p<0.0001). Cholesterol content in cells treated with atorvastatin at 0.1 μM (7.32 ± 0.37 ng/mg protein, N=7) did not differ from vehicle treatment.

Fig. 1.

Atorvastatin treatment reduced cholesterol content in N2A cells. Cells were treated with atorvastatin (0.1 –10 μM) or vehicle for 24 hr in DMEM supplemented with 10% lipoprotein-deficient serum. Total and unesterified cholesterol content were determined in cell lysates in the presence and absence of cholesterol esterase, respectively, using an Amplex Red cholesterol assay. (A) Cholesterol in N2A cells is unesterified (membrane) cholesterol. There was no difference in the membrane and total cholesterol content. (B) Atorvastatin (Ator) treatment reduced cholesterol content in a concentration-dependent manner. (C) Representative western blot images of 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR, ∼95 kDa) following Ator treatment. (D) Ator treatment led to an increase in HMGCR experssion in a concentration-dependent manner. Data are presented with scatter plots as well as mean ± SEM. **p<0.01, ****p<0.0001 vs. vehicle

The effect of atorvastatin treatment on the expression of HMGCR, the rate-limiting cholesterol synthesis enzyme, was determined using western blotting. As shown in Figs. 1C & 1D, atorvastatin treatment induced a concentration-dependent increase in HMGCR expression (~95 kDa). A one-way ANOVA revealed a significant main effect of atorvastatin concentration, F (4,18) = 33.17, p<0.0001. Posthoc Bonferroni test indicated a significant increase in the HMGCR expression when cells were treated with atorvastatin at 0.3 μM, t(18) = 4.088, p=0.0028; 1 μM, t(18) = 8.111, p<0.0001; and 10 μM t(18) = 9.439, p<0.0001, when compared to vehicle treatment. Notably, western blotting of HMGCR proteins produced multiple bands, which is indicative of the cleavage of HMGCR proteins. This cleavage has been demonstrated in several model systems and is mediated through various mechanisms (Menzies et al. 2018; Moriyama et al. 1998; Moriyama et al. 2001).

The effect of atorvastatin treatment on cell viability was determined using trypan blue staining. There was no significant cell death when the concentration of atorvastatin treatment was lower than 10 μM (Fig. S1). We did not include any concentration of atorvastatin exceeding 10 μM because cells were notably detached from cell culture plates following 30 μM atorvastatin treatment (24 hr). This is consistent with reported apoptosis in various cell lines in the presence of high concentrations of atorvastatin in serum free medium (Deng et al. 2019; Crescencio et al. 2009; Zhao et al. 2019). In addition, a longer treatment (e.g. 48 hr) with atorvastatin (>= 1μM) resulted in at least 50% of cells detached from the plate, indicating cell death.

3.2. Atorvastatin treatment reduced DA uptake and cholesterol replenishment reversed the effect

The function of DA uptake was assessed in the presence of increasing concentrations of DA following atorvastatin or vehicle treatment and plotted using the nonlinear regression analysis (Fig. 2A). The values of V_max and K_m were extrapolated from the concentration curve (Figs. 2B & 2C). A one-way ANOVA revealed a significant main effect of atorvastatin concentration on V_max, F (3, 25) = 11.81, p<0.0001. When compared to vehicle treatment (28.44 ± 1.88 pmole/mg protein/min, N=11), a post hoc Bonferroni test indicated a significant reduction in V_max values in cells treated with 1 μM (16.94 ± 1.18 pmole/mg protein/min, N=6, p=0.0004) and 10 μM atorvastatin (12.07 ± 0.99 pmole/mg protein/min, N=6, p<0.0001). The values of K_m were reduced in cells treated with 1μM (1.47 ± 0.24 μM, N=6, p=0.039) and 10 μM (1.07 ± 0.19 μM, N=6, p=0.0015) when compared to vehicle treatment (2.31 ± 0.20 μM, N=11). Treatment with atorvastatin at 0.3 μM did not significantly alter either V_max (23.97 ± 2.33 pmole/mg protein/min, N=6) or K_m values (2.99 ± 0.21 μM, N=6).

Fig. 2.

The effect of atorvastatin treatment and cholesterol replenishment on DA uptake in N2A cells stably expressing hDAT. (A) Cells were treated with increasing concentrations of atorvastatin (Ator) for 24 hr in 10% lipoprotein-deficient serum followed by [³H]DA (15 nM) uptake in the presence of increasing concentrations of unlabeled DA (100 nM − 30 μM) for 10 min at room temperature. Treatment with atorvastatin at 0.3 μM (N=6), 1 μM (N=6) or 10 μM (N=6) reduced DA uptake in a concentration-dependent manner when compared to vehicle treatment (N=11). (B & C) The values of the maximal DA uptake rate (V_max) and K_m were significantly lower in cells treated with 1 and 10 μM atorvastatin when compared to vehicle treatment. (D) Replenishment of water-soluble cholesterol (250 and 500 μM, 2 hr at 37°C) to cells following treatment with atorvastatin (1 μM, 24 hr, N=8/group) restored DA uptake. For non-atorvastatin treated cells, addition of 500 μM and 1000 μM cholesterol significantly reduced DA uptake (N=5/group). (E) Following atorvastatin treatment for 24 hr, surface DAT was biotinylated using membrane non-permeable sulfo-NHS-SS-biotin (1.5 mg/ml) for 1.5 hr at 4°C. The biotinylated proteins were pulled down using streptavidin beads. The biotinylated proteins were eluded from the beads with SDS-PAGE loading buffer containing DTT. Representative western blot images for biotinylated DAT, total DAT and GAPDH are presented. (F & G) Quantitation of surface DAT and total DAT expression levels. For surface DAT levels, data were calculated as the surface DAT levels vs. the total DAT levels, and presented as relative to the vehicle treatment. Total DAT levels were normalized to GAPDH levels and presented as relative to vehicle treatment. Data are presented with mean ± SEM *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 vs. vehicle

To examine if the atorvastatin-induced reduction in DA uptake resulted from a loss of membrane cholesterol content, various concentrations of water-soluble cholesterol were added to atorvastatin (1 μM)-treated cells to replenish cholesterol content. The [³H]DA uptake assay was performed immediately after cholesterol replenishment (Fig. 2D). A two-way ANOVA revealed significant main effects of atorvastatin treatment, F (1, 55) = 11.03, p = 0.0016; cholesterol concentration, F (4, 55) = 3.565, p = 0.0118; and atorvastatin-cholesterol interactions, F (4, 55) = 9.501, p<0.0001. In vehicle-treated cells, cholesterol addition produced a significant, descending trend in reducing DA uptake across the concentrations analyzed by a simple linear regression (Y=−0.039*X + 96.19, F (1,23) = 32.35, p<0.0001). Post hoc Bonferroni tests revealed that addition of 500 μM (N=5, t(55) = 3.452, p = 0.0485) and 1mM cholesterol (N=5, t(55) = 3.762, p = 0.0185) significantly reduced DA uptake when compared to the vehicle group. For atorvastatin-treated cells, post hoc Bonferroni test indicated that replenishment with 250 μM (N=8, t(55) = 5.002, p=0.0003) and 500 μM (N=8, t(55) = 4.205, p=0.0043) cholesterol significantly improved DA uptake compared to vehicle treatment.

3.3. Atorvastatin treatment did not alter basal DAT surface expression

To determine whether reduced DAT uptake function resulted from a change in surface DAT level, basal DAT surface expression was determined using surface biotinylation assay (Fig. 2E). Surface DAT levels were normalized to total DAT levels and are presented as relative to controls, while total DAT levels were normalized to GAPDH levels (Figs. 2F & 2G). A one-way ANOVA did not indicate a significant main effect of atorvastatin concentration on the expression of surface DAT, F(4,33)= 0.405, p= 0.8035, and total DAT, F(4,33) = 0.880, p=0.487. These data suggest that reduced DAT function is independent of surface DAT levels.

3.4. Atorvastatin treatment reduced the ability of cocaine to inhibit DA uptake, and cholesterol replenishment reversed this effect

To investigate whether atorvastatin treatment alters the interaction of cocaine with the DAT, we examined the ability of cocaine to inhibit DA uptake. Cells treated with atorvastatin (1 μM) or vehicle were pre-incubated with various concentrations of cocaine (1 nM – 10 μM) in the presence of 15 nM [³H]DA followed by DA uptake assay. Atorvastatin treatment shifted the cocaine concentration-response curve rightward (Fig. 3A). There was a significant increase of cocaine IC₅₀ values in atorvastatin-treated cells when compared to the vehicle treatment (Veh: 1.081 ± 0.143 μM, N=6; Atorvastatin: 4.109 ± 0.493 μM, N=7; p=0.0002), indicating a reduced potency of cocaine to inhibit DA uptake (Fig. 3B). Replenishment of water-soluble cholesterol (250 μM) to cells treated with atorvastatin (1 μM) shifted the cocaine concentration-response curve to the left, bringing it close to the control condition (Fig. 3A). Post hoc Bonferroni analysis showed a significant decrease in cocaine IC₅₀ values from cells replenished with cholesterol (N=6, 2.070 ± 0.510 μM, p<0.01) when compared to cells treated with atorvastatin but without replenishment (4.11 ± 0.493 μM), while there was no significant difference between cholesterol replenishment and vehicle treatment (1.08 ± 0.143 μM, N=7).

Fig. 3.

Atorvastatin treatment reduced the ability of cocaine to inhibit DA uptake and cholesterol replenishment ameliorated the deficit. Cells were treated with atorvastatin (Ator, 1 μM, 24 hr), pre-incubated with increasing concentrations of cocaine (100 nM − 10 μM) for 10 min, and then 15 nM [³H]DA uptake was conducted. Some cells treated with atorvastatin were incubated with vehicle or water-soluble cholesterol (250 μM, 2 hr, 37°C) followed by cocaine inhibition of [³H]DA uptake assay. (A) Atorvastatin treatment shifted the cocaine concentration-response curve to the right and water-soluble cholesterol replenishment prevented the shift. (B) The IC₅₀ values for cocaine inhibition in cells treated with vehicle, atorvastatin (Ator), and Ator+Chol. **p<0.01, ***p<0.001 vs. vehicle

3.5. Atorvastatin treatment did not change AMPH-induced DA Efflux

To examine whether atorvastatin altered the ability of the DAT to reversely transport DA from the intracellular to extracellular space (DA efflux) in the presence of AMPH, cells treated with atorvastatin (1 μM) or vehicle were preloaded with 30 nM [³H]DA for 30 min followed by incubation with AMPH (1 μM or 10 μM) for 5 min. The incubation medium was collected and then replaced at 5-min intervals. The amount of [³H]DA activity in each collection was assessed by liquid scintillation counting. Because of the group difference in the total amount of DA taken up into the cells, data are presented as a percent of the baseline instead of absolute DA radioactivity counts so that a direct comparison of DA efflux over the basal could be made between groups. There was no group difference in DA efflux induced by AMPH at both 1μM (Fig. 4A) and 10 μM (Fig. 4B), suggesting that atorvastatin treatment does not change the ability of AMPH to induce DA efflux mediated by the DAT.

Fig. 4.

Atorvastatin treatment did not alter amphetamine (AMPH)-induced, DAT-mediated DA efflux. Cells treated with atorvastatin (Ator, 1 μM, 24 hr) were preloaded with 30 nM [³H]DA and 1 μM unlabeled DA for 30 min. Cells were quickly washed with KRH buffer and then incubated with KRH for 5 min followed by removal and replenishment with the same volume of fresh KRH. This step was repeated three times to establish a stable DA baseline. Next, cells were incubated with AMPH (1 or 10 μM) for 5 min followed by KRH removal, addition of fresh KRH and incubation for another 5 min. This procedure was repeated 4 times for a total of 20 min after AMPH incubation. The radioactivity of [³H]DA in each collection was counted. Data were calculated as a percent of baseline [³H]DA radioactivity and presented as Mean ± SEM. Atorvastatin treatment did not alter AMPH-induced DA efflux at both 1 μM (Fig. 4A) and 10 μM (Fig. 4B) concentrations (N=4/group).

3.6. Acute cholesterol depletion with MβCD reduced DA uptake, cocaine inhibition of DA uptake, and AMPH-induced DA efflux

To determine whether the effect of prolonged atorvastatin treatment on DAT function could be replicated by acute cholesterol depletion, we treated cells with MβCD, which acutely removes cholesterol from membranes to the intracellular space and causes cholesterol esterification (Wiegand et al. 2003). Cells were incubated with various concentrations of MβCD (2.5, 5 and 7.5 mM, 30 min at 37°C) or vehicle in serum free medium. Unesterified (membrane) cholesterol was measured in the absence of cholesterol esterase in cell lysates (Fig. 5A). A one-way ANOVA revealed a significant main effect of MβCD concentration on cholesterol content, F (3,17) =29.7, p<0.0001. Posthoc Bonferroni test showed that MβCD treatment significantly reduced membrane cholesterol content at 2.5 mM, t(17) = 4.49, p=0.0010; 5.0 mM, t(17) = 7.72, p<0.0001; and 7.5 mM, t(17) = 8.40, p<0.0001.

Fig. 5.

Acute methyl-β cyclodextrin (MβCD) treatment reduced DA uptake, cocaine inhibition of DA uptake and amphetamine (AMPH)-induced DA efflux. (A) Cells were treated with 2.5, 5 and 7.5 mM MβCD in serum free medium for 30 min at 37°C. Unesterified (membrane) cholesterol was measured in the absence of cholesterol esterase using an Amplex Red cholesterol assay kit. There was a concentration-dependent reduction in membrane cholesterol content. (B) 15 nM [³H]DA uptake was performed in the presence of increasing concentrations of unlabeled DA following treatment with MβCD (5 mM, 30 min). The saturation curves for DA uptake are presented for both vehicle and MβCD treatment. (C) The V_max value for DA uptake from MβCD-treated cells was significantly lower than that from vehicle-treated cells. (D) Cells treated with MβCD (5 mM, 30 min) or vehicle were pre-incubated with increasing concentrations of cocaine for 10 min followed by addition of 15 nM [³H]DA for 10 min. Cocaine concentration-response curves are presented. (E) The IC₅₀ value for cocaine inhibition in MβCD-treated cells was significantly higher than in vehicle-treated cells. (F) AMPH-induced DA efflux was conducted following MβCD (2.5 mM or 5 mM, 30 min) or vehicle treatment and 30 min incubation with 30 nM [³H]DA and 1 μM unlabeled DA. Cells were quickly washed with KRH buffer and then incubated with KRH for 5 min followed by removal and replenishment with the same volume of fresh KRH. This step was repeated three times to establish a stable DA baseline. Next, cells were incubated with AMPH (10 μM) for 5 min followed by KRH removal, addition of fresh KRH and incubation for another 5 min for a total of four times. The radioactivity of [³H]DA in each collection was counted. Data were calculated as a percent of baseline [³H]DA radioactivity and presented as Mean ± SEM. Both concentrations of MβCD resulted in a significant reduction in AMPH-induced DA efflux. **p<0.01 vs. vehicle treatment.

It was reported previously that MβCD treatment (5 mM, 30 min at 37°C) reduces cholesterol content by ~63% in HEK293 cells; this reduction is associated with lower DA uptake and AMPH-induced DA efflux (Jones et al. 2012). Thus, we used this concentration of MβCD in N2A cells. To examine the effect of MβCD on DA uptake, cells were treated with vehicle or MβCD (5 mM, 30 min), followed by [³H]DA uptake in the presence of increasing concentrations of unlabeled DA as described for atorvastatin treatment. There was a notable difference in DA uptake saturation curves between vehicle and MβCD treatment (Fig. 5B). The V_max values (pmole/mg protein/min) for vehicle and MβCD treatment were 33.99 ± 6.56 and 14.34 ±3.24, respectively (Fig. 5C). A student t-test indicated that MβCD treatment significantly reduced V_max, t(8)=2.687, p=0.0276. The K_m values (μM) were 3.006 ± 0.497 and 2.069 ± 0.295 for the vehicle and MβCD group, respectively, which did not differ statistically. The reduced DA uptake by MβCD treatment is in agreement with the effect of atorvastatin treatment.

To assess the ability of cocaine to inhibit DA uptake, cells were treated with vehicle or MβCD (5 mM, 30 min). Then cells were pre-incubated with increasing concentrations of cocaine for 10 min followed by the 15 nM [³H]DA uptake assay. Treatment with MβCD resulted in a rightward shift in the cocaine concentration-response curve (Fig. 5D). The IC₅₀ values (μM) for vehicle and MβCD treatment are 3.35 ± 0.86 and 8.51 ±1.10, respectively (Fig. 5E). A student t-test showed that MβCD treatment significantly increased the IC₅₀ values for cocaine inhibition of DA uptake, t(8)=3.704, p=0.006, suggesting that the ability of cocaine to inhibit DA uptake is reduced by MβCD.

To examine the effect of MβCD on AMPH-induced, DAT-mediated DA efflux, cells were treated with vehicle or MβCD (2.5 or 5 mM, 30 min). We used 2.5 mM MβCD concentration because this concentration reduced cholesterol content by 40%, which was similar to the reduction induced by 1 μM atorvastatin. Cells were incubated with [³H]DA (30 nM) along with of 1μM unlabeled DA for 30 min followed by AMPH (10 μM) treatment for 5 min as described for atorvastatin treatment. Data were calculated as a percent of its own baseline DA radioactivity for each group. For cells treated with 2.5 mM MβCD, a two-way ANOVA indicated a significant main effect of cholesterol treatment, F (1,55) = 11.13, p=0.0014. Sidak’s multiple comparisons test showed that there was a significant reduction in AMPH-induced DA efflux in MβCD-treated cells (N=3) when compared to vehicle-treated cells (N=7) shown in Fig. 5E. For cells treated with 5.0 mM MβCD, a two-way ANOVA indicated a significant interaction effect, F (7,69)=2.82, p=0.012. Sidak’s multiple comparisons test showed a significant decrease in AMPH-induced DA efflux in MβCD-treated cells (N=3) when compared to vehicle-treated cells (N=7) shown in Fig. 5G.

4. Discussion

In the present study, we show that both acute and prolonged cholesterol reduction by MβCD and atorvastatin treatment, respectively, lowered membrane cholesterol content in N2A cells, markedly reduced DAT uptake function and attenuated the ability of cocaine to inhibit DA uptake. The reduction in DA uptake and cocaine inhibition of DA uptake following atorvastatin treatment was mitigated by cholesterol replenishment. Although atorvastatin and MβCD reduced the amount of membrane cholesterol equivalently, atorvastatin had no effect on AMPH-induced DA efflux, whereas MβCD treatment induced a small but significant reduction in DA efflux. Our data suggest that atorvastatin can modulate DAT function through inhibition of cholesterol synthesis.

4.1. Atorvastatin reduces membrane cholesterol content in a concentration-dependent manner

In the periphery, cholesterol exists in unesterified and esterified forms. Only unesterified cholesterol, not cholesteryl ester, was detected in N2A cells, suggesting that cholesterol in N2A cells is utilized on cell membranes. The effect of prolonged atorvastatin treatment on cholesterol content was determined in N2A cells that were cultured in the lipoprotein-deficient serum instead of FBS during atorvastatin treatment. It has been reported that the presence of an external cholesterol source (e.g. FBS) in HEK293 cells blocks statin’s effect on cholesterol reduction because cells can replenish cholesterol from FBS (Cole et al. 2005). We found that both atorvastatin and MβCD produced a concentration-dependent reduction in cholesterol content in N2A cells. A marked reduction in cholesterol content was previously reported following atorvastatin treatment (1 μM, 24 hr) to human hepatoma HepG2 cells (Scharnagl et al. 2001) and to rat RN46A-B14 cells (Deveau et al. 2021). However, statins have also been reported ineffective in reducing cholesterol content. For example, simvastatin treatment (12 μM, 24 hr) in HEK293T cells and miaPaCa-2 pancreatic cancer cells in FBS free medium has no effect on cholesterol levels (Gbelcova et al. 2013). Because cell lines differ in their cholesterol content (Hovde et al. 2019), the efficacy of statin treatment in reducing cholesterol may vary. Moreover, we found that HMGCR expression was increased following atorvastatin treatment (Fig. 1D), which was presumably a compensatory change resulting from lowered cholesterol content induced by prolonged atorvastatin treatment. Despite the marked induction of HMGCR expression, cholesterol content remained reduced. This reduction may be related to the long elimination half-life of atorvastatin, ranging from 14.7 to 57.6 hours in human plasma (Posvar et al. 1996).

4.2. Atorvastatin reduces DAT uptake function without altering surface DAT levels

Previous studies have shown that the DA uptake function is sensitive to a direct removal or addition of membrane cholesterol. For example, treatment with MβCD (2.5 – 10 mM) reduces DA uptake V_max values in N2A cells, HEK293 cells, striatal synaptosomes prepared from rats and monkey brains (Adkins et al. 2007; Jones et al. 2012; Morissette et al. 2018). Herein, we replicated the finding that MβCD (5 mM) reduced DA uptake in N2A cells. As previous studies have focused exclusively on direct manipulations of membrane cholesterol content using MβCD and water-soluble cholesterol, no study has investigated whether statins, which inhibit de novo cholesterol synthesis, influence DAT function. The present study found that atorvastatin treatment reduced DA uptake; importantly, replenishment of water-soluble cholesterol to atorvastatin-treated cells attenuated the deficit in DA uptake. Our observation is consistent with a previous report showing that cholesterol replenishment to MβCD-treated cells or synaptosomes restored DA uptake function (Jones et al. 2012). Interestingly, simvastatin treatment (30 mg/kg/day, PO, 2 or 4 weeks) to Long-Evans rats reduced cholesterol content in various brain regions as well as serotonin (5-HT) transporter (SERT) reuptake in platelets (Vevera et al. 2016), indicating that SERT function is also sensitive to cholesterol modulation. However, simvastatin treatment (1 & 10 μM, 24 hr) to serotonergic RN46A-B14 cells significantly increased serotonin (5-HT) uptake, and cholesterol addition did not restore SERT function (Deveau et al. 2021), suggesting a cholesterol-independent effect of simvastatin. The opposing effects of simvastatin on SERT uptake function between animals and cultured cells may be attributed to the effectiveness of reducing cholesterol in these two models. Simvastatin, a prodrug, needs to be hydrolyzed to its active β-hydroxyacid form to be effective (Vickers et al. 1990). Therefore, without activation, simvastatin in cultured RN46A-B14 cells may produce a marginal effect on cholesterol reduction.

Atorvastatin treatment did not alter basal DAT surface levels, suggesting that reduced DA uptake function following atorvastatin treatment cannot be attributed to DAT availability. Our observation is similar to previous reports that acute membrane cholesterol depletion with MβCD treatment reduces DA uptake but has no effect on basal DAT surface expression (Adkins et al. 2007; Jones et al. 2012). Because the DAT is distributed in both lipid raft and non-raft membrane microdomains (Foster et al. 2008), constitutive internalization of the DAT may be differentially regulated depending on their compartmental localization on plasma membranes in different cell lines. For example, the DAT was reported to undergo internalization through a clathrin-dependent pathway, as treatment with filipin, a drug that inhibits caveolae- and lipid raft-mediated endocytosis, does not influence constitutive DAT trafficking in porcine aortic endothelia cells (Sorkina et al. 2005). However, the DAT in human neuroblastoma SK-N-MC cells does not co-localize with clathrin-coated pits at the cell surface and acute clathrin inhibition with Pitstop-2 does not inhibit basal DAT internalization (Wu et al. 2015), suggesting that constitutive DAT internalization is clathrin-independent. Moreover, the DAT is constitutively internalized and recycled, and the half-life of surface DAT is approximately 13 min when measured in heterologous systems (Loder & Melikian 2003; Boudanova et al. 2008). Perhaps a reduction in cholesterol content induced by atorvastatin disrupts endocytic and exocytic membrane trafficking events, resulting in no change in DAT surface expression, as observed in the present study. Future studies are required to investigate whether constitutive and agonist-induced DAT internalization and recycling are abolished when cholesterol content is reduced.

4.3. The ability of cocaine to inhibit DA uptake is more sensitive to atorvastatin treatment than the ability of AMPH to induce DA efflux

Computational modeling illustrates that membrane cholesterol stabilizes the outward-facing DAT conformation (Zeppelin et al. 2018). In agreement, membrane cholesterol addition increases the binding of the cocaine analog [³H]WIN35428 to the DAT (Hong & Amara 2010). The present study added another line of evidence showing that an overall reduction in membrane cholesterol via inhibition of cholesterol synthesis as well as acute depletion by MβCD reduced the ability of cocaine to inhibit DA uptake. Our data support the notion that atorvastatin disrupts the outward-facing DAT conformation, preventing the high-affinity binding of cocaine to the DAT. It is worth noting that addition of high concentrations (0.5 and 1 μM) of water-soluble cholesterol to control cells produced a significant reduction in DA uptake, suggesting that cholesterol homeostasis on plasma membranes is important for DA uptake function.

In contrast to atorvastatin-induced reduction in cocaine inhibition of DA uptake, AMPH-induced DA efflux was not altered by atorvastatin treatment (1 μM, 24 hr), indicating that DAT uptake function is more sensitive to cholesterol depletion than DAT efflux function. It is reasonable to presume that a greater depletion of cholesterol content may eventually lead to reduced DAT efflux function. Because a longer treatment of atorvastatin (1 μM) or higher concentrations of atorvastatin treatment led to notable cell detachment from the plate in the present of the lipoprotein-deficient medium, we tested this hypothesis using acute cholesterol depletion with different concentrations of MβCD. It was reported previously that MβCD at 2.5 mM and 5 mM (30 min at 37°C) reduces AMPH-induced DA efflux in HEK293 and LLC-PK cells (Jones et al. 2012). These two concentrations of MβCD (30 min at 37°C) reduced ~40% and ~60% cholesterol content in N2A cells, and also produced a small but significant reduction in AMPH-induced DA efflux in the current study. It is interesting to note that cholesterol reduction by 2.5 mM MβCD is similar to that by 1 μM atorvastatin; however, DA efflux was not altered by atorvastatin treatment. Because the V_max of DA uptake was reduced by both drugs and therefore, uptake of AMPH, a substrate of DAT, is presumably reduced under both conditions. It is intriguing that reduced AMPH uptake would influence DA efflux in MβCD- but not atorvastatin-treated cells. A potential explanation for the discrepancy may be related to the distinct mechanistic actions of these two drugs. It has been shown that even when MβCD and statins induce the same amount of cholesterol depletion, they differ in their effects on membrane dipolar potential, which is an important electrostatic property of organized molecular assemblies on membranes (Sarkar et al. 2017). Therefore, cholesterol depletion by MβCD and statins might lead to differential recruitment of signaling molecules and subsequent differential regulations of DAT interactions with other membrane proteins. Another potential explanation is that atorvastatin treatment (24 hr), but not acute MβCD treatment (30 min), may produce changes in the lipid composition of plasma membranes to compensate for the lack of cholesterol to restore the inward-facing conformation but not the outward-facing conformation. Using lipidomic analysis, it was reported that acute treatment with MβCD (1.25, 2.5 and 5 mM, 30 min at 37°C) primarily depletes membrane cholesterol, does not alter other lipid composition and content of plasma membrane vesicles, and has no impact on the unsaturation/length of plasma membrane lipids in HEK293 cells (McGraw et al. 2019). However, atorvastatin depletes cholesterol by competitively inhibiting cholesterol synthesis, and therefore presumably changes the content of other lipids that use cholesterol as a precursor. It was reported that prolonged simvastatin treatment (5 mM, 48 hrs) to human heptatoma Hep3B cells in the presence of FBS in the culture medium alters the content of many lipid species, including phosphatidylinositol and arachidonic acid (Schooneveldt et al. 2021). Therefore, it is of future interest to investigate the effects of other altered lipid species by statin treatment on DAT function. Lastly, it is reasonable to speculate that atorvastatin and MβCD may differentially modulate DAT phosphorylation, which is known to influence AMPH-induced DA efflux. DAT phosphorylation at and near the N-terminus is necessary for AMPH-induced DA efflux (Khoshbouei et al. 2004; Challasivakanaka et al. 2017; Foster et al. 2012; Guptaroy et al. 2009; Fog et al. 2006). Perhaps, atorvastatin treatment, but not MβCD treatment, may maintain AMPH-induced DAT phosphorylation and therefore DA efflux, which warrants further investigation.

It is important to note that DAT requires sodium for cocaine binding and substrate (DA and AMPH) translocation. The electrochemical gradient of Na+ provides the energy to move substrate through the transporter against its concentration gradient. AMPH-induced DA efflux is dependent on extracellular sodium concentrations as substitution of sodium with chloride abolishes the efflux process (Fraser et al. 2014). There are a few inconsistent reports of the potential effects of cholesterol depletion by MβCD on Na+/K+-ATPase activity in purified enzyme systems, which may lead to disruption of transmembrane sodium gradient and DAT function. A recent paper, using extracted membranes from pig kidney, reported that high concentrations of MβCD (30, 45, or 60 mM, 2 hrs at 37°C) reduced Na+/K+ activity, whereas concentrations of MβCD lower than 30 mM (e.g. 10 and 20 mM) did not significantly alter Na+/K+ ATPase activity (Garcia et al. 2019). Given the amount of cholesterol depletion and drug treatment conditions used in the current study, it is unlikely that Na+/K+-ATPase activity is disrupted. Further, because the transmembrane sodium gradient is required for both DA uptake and efflux function of the DAT, the observation that DA efflux is not impacted by atorvastatin treatment suggests that sodium gradient is unlikely disrupted. Regardless, future studies should confirm that the integrity of sodium gradient is not impaired when membrane cholesterol is removed. To summarize, our data suggest that the outward-facing DAT conformation, which favors the binding of DAT blockers such as cocaine, is more sensitive to cholesterol disruption by atorvastatin than the inward-facing DAT conformation, which favors the binding of DAT substrates such as AMPH, in a neuronal-like environment.

5. Conclusions and Perspectives

Our finding on statin-DAT interactions may have significant implications for our understanding of the side effects associated with chronic use of BBB penetrant statins. Nearly 30% of adults 40 years and older in the United States are on a statin prescription. Several statins (e.g., simvastatin, atorvastatin and fluvastatin) are BBB penetrants and can achieve high levels in the brain (Hamelin & Turgeon 1998; Schachter 2005). Cholesterol depletion in the brain from chronic statin treatment could alter the structural conformation and function of membrane proteins including the DAT, thus producing unwanted side effects. It has been reported that cognitive impairment occurs a few months after the start of statin therapy or after an increase in dosage (Wagstaff et al. 2003; Ott et al. 2015). Among the cases of reported cognitive impairment, the vast majority of patients are on simvastatin or atorvastatin medication (Sahebzamani et al. 2014). Future investigations should focus on understanding the potential association of reduced brain cholesterol content with dysregulation of DA transmission and DA-related symptoms including cognition, anxiety, depression and motor activity. Lastly, this study also suggests a need to develop tools for imaging brain cholesterol metabolism, which would guide a personalized and safer treatment duration for patients who are on a statin prescription.

Highlights.

• Both atorvastatin and MβCD reduce membrane cholesterol content

• Both atorvastatin and MβCD reduce dopamine uptake and cocaine inhibition of dopamine uptake

• Cholesterol replenishment prevents atorvastatin-induced reduction in dopamine uptake and cocaine inhibition

• MβCD, but not atorvastatin, reduces amphetamine-induced, dopamine transporter-mediated dopamine efflux

• The outward-facing conformation of DAT is more sensitive to atorvastatin treatment than the inward-facing conformation

Abbreviations

AMPH

amphetamine

BBB

blood-brain barrier

DA

dopamine

DAT

dopamine transporter

DMSO

dimethyl sulfoxide

DTT

dithiothreitol

FBS

fetal bovine serum

HMG

3-hydroxy-3-methylglutaryl

HMGCR

HMG-CoA reductase

LPDS

lipoprotein-deficient serum

MβCD

methyl-β-cyclodextrin