Conceived and designed the experiments: ADB SAR. Performed the experiments: CKB KS GDB. Analyzed the data: GDB ADB SAR. Wrote the paper: ADB SAR.
The authors have declared that no competing interests exist.
HIV infected patients have an increased susceptibility to liver disease due to Hepatitis B Virus (HBV), Hepatitis C Virus (HCV), alcoholic, and non-alcoholic steatohepatitis. Clinically, this results in limited options for antiretroviral therapy and accelerated rates of liver disease, causing liver disease to be the second leading cause of death for HIV infected patients. The mechanisms causing this propensity for liver dysfunction during HIV remains unknown.
We demonstrate that HIV and/or the HIV glycoprotein gp120 ligation of CXCR4 on hepatocytes selectively up-regulates TRAIL R2 expression and confers an acquired sensitivity to TRAIL mediated apoptosis which is mediated by JNK II, but not p38 nor G-proteins.
These findings suggest that HIV infection renders hepatocytes more susceptible to liver injury during disease states associated with enhanced TRAIL production such as HBV, HCV, or steatohepatitis.
Liver disease has become the second most common cause of death of HIV infected patients
In HIV infected patients, liver disease occurs most commonly from Hepatitis C Virus (HCV) co-infection, active Hepatitis B Virus (HBV) infection, non-alcoholic steatohepatitis (NASH)
Both HIV associated T cell depletion as well as liver injury are regulated, at least in part, by disordered apoptosis. In particular, in both circumstances, the apoptosis inducing ligand TRAIL (TNF related apoptosis inducing ligand) as well as its cognate receptors have been implicated. TRAIL exerts apoptotic effects by binding to and signaling either TRAIL Receptor 1 and/or TRAIL Receptor 2 expressed on the surface of target cells. Each of HIV, HBV, and HCV alter the sensitivity of cells they infect by modulating expression of these receptors
It remains unknown how the HIV infection accelerates the disease course of HCV infection compared to mono-HCV infection. While it has been suggested that the immunodeficiency associated with HIV infection may be the reason HCV progresses more rapidly, two lines of evidence suggest this may not be the case: first, HCV progression is faster in HIV infected patients with normal immune function
Based upon the combined observations that 1) HCV progression is faster in HIV co-infection; 2) this effect is not necessarily solely due to immunodeficiency; and 3) TRAIL dysregulation characterizes both HIV disease and HCV liver injury, we hypothesized that the ability of HIV to modulate TRAIL signaling in T cells might also occur in hepatocytes. We therefore tested whether HIV particles or HIV proteins altered either TRAIL receptor expression or TRAIL sensitivity of the human hepatocyte cells, Huh7.
Dulbecco's modification of Eagle's medium (DMEM) (Mediatech Inc., Manassas, VA). Fetal bovine serum (FBS) (Atlanta Biologicals Inc., Lawrenceville, GA). JNK II inhibitor, Pertusis toxin (PT), SB 203580 (Calbiochem and, San Diego, CA). Recombinant superkiller TRAIL(skTRAIL) (Alexis Biochemicals, San Diego, CA). Recombinant HIV-1 IIIB envelope glycoprotein 120 (gp120) and recombinant human T- cell receptor sCD4 (Immuno Diagnostics Inc., Woburn, MA). Bovine serum albumin (BSA) (Sigma Chemical Co, St. Louis, MO), anti-actin and anti-tubulin antibodies (Molecular Probes, Carlsbad, CA). AMD3100 (NIH AIDS Research and Reference Reagent Program, Germantown, MD). Antibodies to TRAIL R1, R2, R3, R4 for flow cytometry (BD Bioscience, San Diego, CA) and TRAIL R2 for Western blot (Capralogics Inc., Hardwick, MA), active caspase-3 (BD Biosciences). The MTS, CytoTox96 non-radioactive cytotoxicity assay kit (Promega, Madison, WI).
The human hepatocyte cell line Huh7 (a kind gift from Dr. G. Gores, Mayo Clinic, Rochester, MN) were cultured at 37°C DMEM with 10% FBS and 10,000 µg/ml penicillin/streptomycin, and 200 mM glutamine. The Huh7 cells were plated at a density of 0.1×105 cells/well in 96-well plates in a final media volume of 200 µL. Cells were pre-incubated with either AMD3100 (10 µM), Pertussis toxin (1 µg/µl), JNK II inhibitor (400 nM), or SB203580 (8 µM), a p38 inhibitor for one hour at 37 °C. Then, 5 µg/mL HIV glycoprotein 120 (HIV gp120) IIIB or purified X4 HIV IIIb was added for six hours followed by 100 ng/mL of skTRAIL for twelve hours. Cell viability was assessed by using 20 µL of the methanethiosulfonate reagent (MTS) by measuring absorption at 490 nm using spectrophotometric plate reader (EL800, Bio-tek instruments Inc., Winooski, VT). The mean viability was calculated and the results were expressed as percentage (%) of MTS reduction.
Huh7 cells (4×105) were treated with 10 µg gp120 for 0–3 hours and lysed in lysis buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 2 mM EDTA, 0.1% Triton-X100, 2 µg/ml aprotinin, 5 µg/ml leupeptin, 1 µM PMSF and 1 µM Saponin) on ice. 50 µg of protein was resolved on a 15% sodium dodecyl sulfate-polyacrylamide (SDS-PAGE) gel and electrotransferred to polyvinylidene fluoride (PVDF) membranes (Millipore Corporation, Bedford, MA). Membranes were blocked with 5% nonfat milk in TBST [Tris base (10 mM), NaCl (150 mM), Tween (0.1%), pH 7.5] for two hours at room temperature and probed with goat anti TRAIL R2 (1∶1000) antibodies overnight at 4°C. Actin (Sigma, St. Louis, MO) was used as a loading control and intensity was measured using EL Logic 2200 imaging software system.
Huh7 cells (4×105) were harvested, washed in cold PBS, and fixed in 2% paraformaldehyde overnight at 4°C. Cells were permeabilized with 0.1% NP-40 for 15 minutes (active caspase-3) on ice, blocked with 2% BSA/PBS and stained with phycoerythrin (PE) conjugated anti human anti TRAIL R1, R2, R3, R4 mAb (1∶40) or rabbit polyclonal anti-active caspase-3 antibodies (1∶40) or isotype control (Abcam Inc., Cambridge, MA) for one hour on ice, and analyzed by flow cytometry (FACScan flow cytometer, BD Biosciences, San Diego, CA).
All results were expressed as the mean±standard deviation (SD) and repeated at least three times as indicated in the figures. Statistical comparisons were made by student's
In order to determine whether HIV alters the regulation of TRAIL sensitivity in hepatocytes, concentrated purified whole live HIV IIIb was used as a clinically relevant source of HIV gp120. Since Huh7 cells express the HIV receptor CXCR4, cells were pre-incubated or not with the CXCR4 receptor decoy AMD3100 and then treated with HIV IIIb. After six hours, skTRAIL was added, and the following day the cells were fixed, permeabilized, stained with antibodies that recognize the active form of caspase-3, and analyzed by flow cytometry. Untreated hepatocytes have spontaneous levels of apoptosis of <5%, while the addition of TRAIL increased apoptosis only minimally, confirming the relative resistance of these cells to TRAIL. In addition, and consistent with prior observations
Huh7 cells were pre-incubated or not with AMD3100 for one hour followed by treatment with purified HIV IIIb or mock virus for six hours. Selective points were then treated with skTRAIL for twelve hours. Cells were then fixed, permeabilized, and stained with fluorescently labeled anti-active caspase-3 antibodies and analyzed by flow cytometry. Camptothesin (10 µM) was used as positive control for caspase-3 activation. (A) A single representative dot plot for each group is given. In each plot, the percentage of cells positive for active caspase-3 is shown in the upper right quadrant. (B) Values are expressed as the mean±SD of three independent experiments.
Since human hepatocytes express the HIV chemokine co-receptor CXCR4 on the surface,
(A) Huh7 cells were treated with HIV gp120 or control protein for six hours, stained with fluorescently conjugated antibodies specific for TRAIL receptors and analyzed by flow cytometry. BSA control protein is shown as a dotted line, and HIV gp120 is shown as a solid line. A single representative histogram from three independent experiments is given for each group. (B) Huh7 cells were treated with HIV gp120 for one, two, or three hours, lysed, and resolved on 15% SDS-PAGE, electrotransferred to PVDF membrane, and probed for TRAIL R2. Densitometer readings are presented as the relative intensity of TRAIL R2 normalized to actin. The data represents the mean±SD of three independent experiments. (C) Huh7 cells were pre-incubated or not with AMD3100 for one hour followed by treatment with HIV gp120 or BSA and analyzed for TRAIL R2. Data of mean channel fluorescence represents the mean±SD of three independent experiments.
Because individuals with HIV infection are more prone to liver disease than non-infected individuals, and given that HIV gp120 makes immune cells sensitive to bystander TRAIL mediated death, we sought to determine whether HIV gp120 also causes the hepatocytes to become sensitive to TRAIL mediated apoptosis. Human hepatocytes were incubated with HIV gp120 or control protein for six hours and then stimulated with TRAIL. HIV gp120 alone induces hepatocyte death (p<.002), as previously described
(A) Huh7 cells were pre-incubated or not with AMD3100 for one hour and then treated with HIV gp120 or BSA for six hours followed by treatment of skTRAIL. Cell viability was determined by reduction of MTS by live cells. (B) Huh7 cells were pre-incubated with JNK II inhibitor, p38 inhibitor or the G-protein inhibitor, Pertussis toxin, for one hour followed by treatment with HIV gp120 or control protein for six hours. Cell viability was determined by reduction of MTS in live cells. Values are expressed as the mean±SD of three independent experiments from duplicate wells.
HIV gp120 triggers a number of intracellular signals after binding CXCR4. Gp120/CXCR4 T cell death is associated with p38 phosphorylation and activation,
Given that HIV gp120 induced TRAIL sensitivity can be abrogated by pre-incubating hepatocytes with JNK II kinase inhibitors, we questioned whether the concomitant increase in TRAIL R2 expression on hepatocytes is also mediated by the JNK II kinase. Huh7 cells were transfected with siRNA for JNK I, JNK II, p38 kinase, or scramble control, followed by treatment with HIV gp120. After 18 hours of HIV gp120 treatment, the cells were lysed and blotted for TRAIL R2.
Huh7 cells were treated with either BSA or gp120 in the presence or absence siRNA constructs for JNK I, JNK II, and p38; and analyzed for TRAIL R2 content (A). The results presented below are pooled TRAIL R2 densitometry normalized to Actin. Parallel experiments demonstrate that siRNA for JNK I, JNK II, and p38 did not alter TRAIL R2 expression (B), and these siRNA constructs were specific for the proteins against which they were directed (C).
Liver disease is a frequent cause of morbidity and mortality in HIV infected individuals. Recently, the leading cause of death for hospitalized HIV patients in the United States became liver disease (MMWR)
During HIV infection, CD4+ T cells are depleted by multiple mechanisms; including HIV gp120 binding to the chemokine co-receptor CXCR4 and resulting in resting CD4+ T cell death
In the current report, we have extended the previous understanding of how HIV gp120 may contribute to hepatotoxicity. First, as previously reported, gp120 ligation of CXCR4 expressed on hepatocytes can cause hepatocyte death directly in a minority of cells
Since human hepatocyte cell lines as well as freshly isolated primary human hepatocytes express CXCR4 on the cell surface, we tested the hypothesis that HIV gp120 could bind to CXCR4 and up-regulate TRAIL R2 and consequently induce TRAIL sensitivity in human hepatocytes, thereby making the cells vulnerable to apoptosis in settings of high TRAIL expression (
Taken together with reports that HBV infection
Schematic model of synergy between HIV gp120/CXCR4-induced TRAIL sensitivity and other proapoptotic stimuli (e.g., HBV, HCV, Bile salts), causing hepatocyte death and consequent cirrhosis.
HIV gp120 death and signaling can be mediated through several possible pathways. In immune cells, HIV gp120 binds CXCR4 and causes a G-protein independent death that signals via p38