Apolipoprotein E isoform dependently affects Tat-mediated HIV-1 LTR transactivation

U87MG glioblastoma cells were cultured in 1× DMEM (Invitrogen) supplemented with 10% fetal calf serum and penicillin/streptomycin (Invitrogen). U87MG cells were stably transfected with luciferase reporter gene under the control of HIV-1 LTR promoter following neomycin (Sigma-Aldrich) selection pressure. These cells were provided generously by Dr. Lena Al-Harthi (Rush University, Chicago). ApoE2 and ApoE3 were purchased from Sigma-Aldrich, and ApoE4 from Abcam. Human plasma HDL were obtained from Kalen Biomedical and ApoE-deprived HDL3 (d = 1.125–1.21 g/ml) were prepared [29].

U87MG cells were seeded at a confluency level of about 50% (10,000 cells) on 96 well plates 1 day prior to being taken for experimentation. Cells were incubated with ApoE isoforms with or without HDL in the absence and presence of 2 μg/ml HIV-1 Tat (ABL Inc. and NIH AIDS program) protein and chloroquine (100 μM). Forty-eight hours post-incubation, luciferase activity assays (Promega) were performed and relative luminescence was quantified using a fluorometer/luminometer plate reader (Spectra MAX GEMINI EM). Similar protocols were followed using cells treated with HDL, LRP1 receptor-associated protein (RAP from American Research products), ApoE mimetic peptide (H-Gly-Arg-Leu-Val-Gln-Tyr-Arg-Gly-Glu-Val-Gln-Ala-Met-Leu-Gly-Gln-Ser-Thr-Glu-Glu-Leu-Arg-Val-Arg-Leu-Ala-Ser-His-Leu-Arg-Lys-Leu-Arg-Lys-Arg-Leu-Leu-Arg-Asp-OH from Anaspac Inc.), or ApoE4 structure corrector (PH002, Calbiochem).

U87MG cells were transfected with negative control siRNA (Ambion) and specific targeted LRP1 siRNA (sense: CAGACAAGAUUGAACGGAUtt and anti-sense: AUCCGUUCAAUCUUGUCUGTc from Ambion) by Jetprime reagent (Polyplus). Twenty-four hours post-transfection, cells were treated with HIV-1 Tat protein in the presence of chloroquine (100 μM) for 48 h and then taken for luciferase assay. Transfection efficiency was determined by protein levels of LRP1.

Cells were harvested and lysed in 1× RIPA lysis buffer (Thermofisher) plus 10 mM NaF, 1 mM Na₃VO₄, and Protease Inhibitor Cocktail (Sigma). After centrifugation (14,000×g for 10 min at 4 °C), supernatants were collected, and protein concentrations were determined with a DC protein assay (Bio-Rad). Proteins (10 μg) were separated by SDS-PAGE (12% gel) and transferred to polyvinylidene difluoride membranes (Millipore). The membranes were incubated overnight at 4 °C with antibodies against GAPDH (Abcam) and LRP1 (Abcam). The blots were developed with enhanced chemiluminescence, and bands were visualized and analyzed by our LI-COR Odyssey Fc Imaging System.

Quantitative analysis of HIV-1 Tat internalization was performed using a method as described previously, but with minor modifications [30]. Cells plated on glass-bottom 35-mm² tissue culture dishes were co-incubated with ApoE2, ApoE3, or ApoE4 (5,10 μg/ml) and FITC-labeled HIV-1 Tat (5 μg/ml; purchased from Rockland) for 90 min at 37 °C in the presence of chloroquine (100 μM). Cells were washed with an acid wash solution (0.2 M acetic acid, 0.5 M NaCl, pH 2.8) at 4 °C for 10 min and then washed with ice-cold PBS for 5 min to remove surface-bound HIV-1 Tat-FITC. Cells were fixed in 4% paraformaldehyde, and images were taken with a confocal laser-scanning microscope (Zeiss LSM800). All experiments were performed in triplicate. The average integrated intensity of HIV-1 Tat-FITC green signal per cell was calculated for each dish using ImageJ software.

All data were expressed as means ± SD. Statistical significance between two groups was analyzed with a Student’s t test, and statistical significance among multiple groups was analyzed with one-way ANOVA plus a Tukey post hoc test. p < 0.05 was considered to be statistically significant.

For our studies, we measured Tat-mediated HIV-1 LTR transactivation using U87MG glioblastoma cells stably transfected with HIV-1 LTR-driven luciferase. With this system, concentration-dependent increases in Tat-mediated HIV-1 LTR transactivation were observed when exogenous HIV-1 Tat protein was co-applied at an optimal concentration of 100 μM chloroquine (Additional file 1: Figure S1), a weak base that neutralizes the acidic environment of lysosomes, prevents HIV-1 Tat degradation, and enhances nuclear delivery of HIV-1 Tat [26, 31‐33]. Thus, many factors related to HIV-1 infection could affect Tat-mediated LTR transactivation. Given that gp120 and TNFα have both been shown to neutralize endolysosome pH [34‐36], we determined the extent to which gp120 and TNFα affected Tat-mediated LTR transactivation in the absence of chloroquine, and neither affected Tat-mediated LTR transactivation (Additional file 1: Figure S2). However, we demonstrated that other endolysosome de-acidifying reagents acted in a similar way as chloroquine to enhance Tat-mediated LTR transactivation (Additional file 1: Figure S2), and these reagents include LYS01 (a free base) [37], bafilomycin (a specific vacuolar ATPase inhibitor), and KH7 (a selective soluble adenylyl cyclase inhibitor) [38]. We showed that 100 μM of chloroquine enhanced Tat-mediated LTR transactivation at Tat concentration as low as 0.5 μg/ml (35 nM) (Additional file 1: Figure S1). In the present study, we chose higher concentrations of Tat at 2 μg/ml (140 nM) to induce robust LTR transactivation so that we can demonstrate confidently whether ApoE or blocking LRP1 could affect Tat-mediated LTR transactivation. Using the above conditions (Tat at 2 μg/ml in the presence of 100 μM chloroquine), we next determined the mechanisms by which HIV-1 Tat enters the cells and the involvement of LRP1 in the induction of HIV-1 LTR transactivation in part because LRP1 is involved with HIV-1 Tat internalization [27]. First, using a specific LRP1 antagonist, receptor-associated protein (RAP), we demonstrated that RAP concentration dependently and significantly restricted Tat-mediated HIV-1 LTR transactivation (Fig. 1a). Next, we determined the extent to which siRNA knockdown of LRP1 affected Tat-mediated HIV-1 LTR transactivation. LRP1 protein expression levels were significantly decreased in cells transfected with LRP1 specific siRNA as compared to control (scrambled) siRNA (Fig. 1b). Further, the ability of HIV-1 Tat in conjunction with chloroquine to increase HIV-1 LTR transactivation was significantly attenuated in cells transfected with LRP1 specific siRNA as compared to control (scrambled) siRNA (Fig. 1c). Thus, siRNA knockdown of LRP1 blocked significantly [39] Tat-mediated LTR transactivation. Together, these data indicate that LRP1 is involved in Tat-mediated LTR transactivation in this system.

ApoE-rich lipoprotein is the major carrier for cholesterol transport in the brain and is a well-characterized ligand of LRP1. To determine the extent to which ApoE-rich lipoprotein affects Tat-mediated HIV-1 LTR transactivation, we needed to mimic the in situ structure and composition of ApoE-rich lipoproteins in the brain. ApoE-rich lipoproteins synthesized in situ in the brain is thought to be HDL-like particles composed of phospholipids and un-esterified cholesterol [39‐41]. We first determined the extent to which plasma-derived HDL (which contains low levels of ApoE) or ApoE-deprived HDL3 affects Tat-induced LTR transactivation, and we found neither plasma-derived HDL nor ApoE-deprived HDL3 at the concentrations of 50 μg/ml affected Tat-induced LTR transactivation (data not shown). To model the brain in situ ApoE-rich lipoproteins, we pre-incubated ApoE2, ApoE3, or ApoE4 with ApoE-deprived HDL3 with a fixed ratio of 0.4 for ApoE protein/HDL protein to form ApoE2-HDL, ApoE3-HDL, and ApoE4-HDL. This ratio was based on the findings that the phospholipid content of human plasma HDL is about 29% and the protein content of human plasma HDL is about 40%; thus, the ratio of phospholipids/protein content of HDL is about 0.7 [42]. Furthermore, the ApoE/phospholipid ratio in CSF is about 0.6 [43]. Therefore, if one assumes that the phospholipid (HDL) content is similar in plasma and CSF, the calculated ratio of ApoE protein/HDL protein content in CSF is 0.42. Using this 0.4 ratio, we demonstrated that ApoE2-HDL (Fig. 2a) and ApoE3-HDL (Fig. 2b), significantly and concentration dependently restricted Tat-mediated HIV-1 LTR transactivation. In contrast, ApoE4-HDL only restricted Tat-mediated HIV-1 LTR transactivation at the two highest concentrations of tested (Fig. 2c). ApoE4-HDL was less potent and less effective than were ApoE2-HDL and ApoE3-HDL at restricting HIV-1 Tat-mediated HIV-1 LTR transactivation (Fig. 2d); calculated IC₅₀ values were 0.08 μg/ml for ApoE2-HDL, 0.03 μg/ml for ApoE3-HDL, and 10.0 μg/ml for ApoE4-HDL. Because ApoE2, ApoE3, and ApoE4 were all purified recombinant proteins and may contain endotoxin, the presence of endotoxin in these recombinant proteins may affect Tat-induced HIV-1 LTR transactivation. Thus, we further determined the extent to which endotoxin affects HIV-1 LTR transactivation. We found that endotoxin did not have any effect on Tat-mediated HIV-1 LTR transactivation in U87MG cells (Additional file 1: Figure S3).

We have shown that ApoE isoform dependently affected Tat-mediated HIV-1 LTR transactivation in the presence of HDL and that HDL alone does not affect Tat-mediated HIV-1 LTR transactivation. Although we did not determine the extent to which ApoE-HDL affects HIV-1 infection, our finding seems contradict with a previous report that non-lipidated ApoE4 at 10 μg/ml enhances HIV-1 infection in vitro [12]. Thus, we determined the extent to which ApoE isoforms in the absence of HDL-affected Tat-mediated HIV-1 LTR transactivation. We demonstrated that, in the absence of HDL, ApoE2 (Fig. 3a) and ApoE3 (Fig. 3b), but not ApoE4 (Fig. 3c) concentration dependently restricted Tat-mediated HIV-1 LTR transactivation. By comparing results of the effects of ApoEs on Tat-mediated HIV-1 LTR transactivation in the absence or presence of HDL, it was clear that ApoE2 (Fig. 3d), ApoE3 (Fig. 3e), and ApoE4 (Fig. 3f) were all more potent and effective at restricting Tat-mediated HIV-1 LTR transactivation when tested in the presence of HDL (a fixed ratio of 0.4 for ApoE protein/HDL protein); IC₅₀ values for ApoE2 were 0.08 μg/ml in the presence of HDL and 2.2 μg/ml in the absence of HDL, and for ApoE3 were 0.03 μg/ml in the presence of HDL and 1.4 μg/ml in the absence of HDL. In the presence of HDL, ApoE4 at the two highest concentrations slightly restricted Tat-mediated HIV-1 LTR transactivation; however, ApoE4 concentration dependently enhanced Tat-mediated HIV-1 LTR transactivation in the absence of HDL (Fig. 3f). Thus, our finding is consistent with the previous report that non-lipidated ApoE4 enhances HIV-1 infection [12].

ApoE and HIV-1 Tat are competitive ligands for LRP1, and ApoE can restrict HIV-1 Tat internalization (26). Accordingly, we next determined the extent to which ApoE2, ApoE3, and ApoE4 affected internalization of HIV-1 Tat. Control cells (Fig. 4a) and cells treated with ApoE4 at 5 and 10 μg/ml (Fig. 4a) exhibited a typical punctate staining pattern for internalized FITC-labeled HIV-1 Tat. ApoE2 at 5 and 10 μg/ml (Fig. 4a), and ApoE3 at 5 and 10 μg/ml (Fig. 4a) displayed less punctate staining of internalized FITC-labeled HIV-1 Tat as compared to control cells. When fluorescence intensity was quantified, FITC-labeled HIV-1 Tat internalization was decreased significantly and concentration dependently for ApoE2 and ApoE3 (Fig. 4b). In contrast, ApoE4 was ineffective in blocking Tat internalization (Fig. 4b).

ApoE, a protein with 191 amino acids, has a receptor-binding region at residues 134–150 [44], and biological functions of ApoE are maintained with small fragments consisting of residues 141–150 [45]. We next determined the extent to which an ApoE mimetic peptide (residues 131–169) could affect Tat-mediated HIV-1 LTR transactivation. The ApoE mimetic peptide concentration dependently restricted Tat-mediated HIV-1 LTR transactivation (Fig. 5a). Small-molecule structure correctors (ApoE4 structure correctors) have been shown to modify the structure of ApoE4 to be more of a ApoE3-like phenotype [46]. Using the ApoE4 structure corrector at concentrations ranging from 2.5 to 10 μg/ml, we found that the ApoE4 structure corrector significantly and concentration dependently enhanced the efficiency of ApoE4 in restricting Tat-mediated LTR transactivation (Fig. 5b).