Hepatitis C Virus and HIV Envelope Proteins Collaboratively Mediate Interleukin-8 Secretion through Activation of p38 MAP Kinase and SHP2 in Hepatocytes*

Hepatitis C virus (HCV) infects (cid:1) 40% of human immunodeficiency virus (HIV) patients, and the resulting hepatic dysfunction that occurs is the primary cause of death in patients with co-infection. We hypothesized that hepatocytes exposed to HCV and HIV proteins might be susceptible to injury via an “innocent bystander” mechanism. To assess this, we studied the effects of envelope proteins, E2 of HCV and gp120 of HIV, in model HepG2 cells. Upon co-stimulation with HCV-E2 and HIV-gp120, we observed a potent proinflammatory response with the induction of IL-8. Furthermore, our studies revealed that HCV-E2 and HIV-gp120 act collaboratively to trigger a specific set of downstream signaling pathways that include activation of p38 mitogen-activated protein (MAP) kinase and the tyrosine phosphatase, SHP2. Both specific inhibitors of p38 MAP kinase and sodium vanadate, a potent protein-tyrosine phosphataseinhibitor,blockedIL-8productioninadose-dependentmanner.Theroleofp38MAPkinaseandSHP2wasfurtherdefinedbytransientlyoverexpressingdominantnegativemutantsoftheseproteinsintoHepG2cells.Thesestudiesrevealedthatoverexpressionofaninactivep38MAPkinaseorSHP2mutantpartiallyabrogatedHCV-E2-andHIV-gp120-inducedIL-8produc-tion.FurtherstudiesrevealedthatIL-8inductionwasnotmediatedthroughactivationoftheNF-(cid:1)

Hepatitis C virus (HCV) 1 infection is frequently found in human immunodeficiency virus (HIV)-infected persons, due to common modes of transmission. It has been shown that overall prevalence of HCV infection among HIV-infected individuals is ϳ30 -50%, with rates of co-infection as high as 90% in intravenous drug users and almost 100% in hemophiliacs (1,2). Since the advent of antiretroviral therapy, HCV/HIV-co-infected individuals have experienced increasing HCV-related mortality and morbidity due to the increased severity of HCV-related liver disease (3)(4)(5). The more rapid progression of hepatic disease with co-infection is currently unexplained. It has also been shown that co-infected patients have a higher incidence of fibrosis (6,7). Furthermore, a greater degree of portal, periportal, and lobular inflammation (centrilobular fibrosis, cholestasis, and granulocytic cholangiolitis) was reported in patients with HCV/HIV co-infection than with HCV alone (8). Inflammation is characterized by the invasion of activated leukocytes into the injured tissue. This process is critically dependent upon the rapid expression of chemokines (9). Direct HCV infection of hepatocytes may not be immunologically contained in the face of HIV-related immune dysfunction. However, other mechanisms may be operative, including the toxic effects of HCV and HIV proteins on uninfected cells.
We hypothesize that hepatocytes exposed to HCV and HIV might be subjected to signaling effects produced by the binding of viral proteins to the cell surface, independent of cell infection and viral replication. There is a precedent for such "innocent bystander" effects, whereby viral envelope proteins can induce cell dysfunction and death by triggering specific signal transduction pathways (10 -12). In the present study, we have shown that the HCV envelope protein E2 and HIV envelope glycoprotein gp120 act collaboratively to trigger a specific set of downstream signaling events that lead to induction of the proinflammatory chemokine interleukin-8 (IL-8). Higher expression of IL-8 could contribute to the inflammation reported in HCV/HIV co-infection.
IL-8 was the first identified member of the still growing chemokine family and represents the prototype human chemokine (13). IL-8 synthesis, at low or undetectable levels in normal noninflamed tissue, can be induced in vivo as well as in a wide variety of cells in vitro by proinflammatory cytokines such as IL-1 or tumor necrosis factor (14,15). Secretion of IL-8 can also be a direct consequence of contact with pathogens like bacteria (16,17), viruses (18,19), and cell-stressing agents (20 -23). IL-8, a potent chemoattractant (24), causes optimal immune responses such as inflammation by the activation of neutrophils (25,26). IL-8 has been shown to play a role in the pathogenesis of various diseases, including rheumatoid arthritis, psoriasis, asthma, pancreatitis, acute respiratory distress syndrome, and sepsis (27). Its level in plasma or inflammatory biological fluids is often correlated with the severity of the pathology and/or the outcome of the patients (28,29).
To our knowledge, the regulation of IL-8 gene expression in HCV/HIV co-infection has not yet been studied. In this report, we have characterized a novel signaling pathway cooperatively activated by HCV-E2 and HIV-gp120 and mediated by p38 MAP kinase and SHP2 that may contribute to proinflammatory signals leading to IL-8 production.

EXPERIMENTAL PROCEDURES
Materials-HCV-E2 subtype 1a and M-tropic HIV-gp120 were obtained from Immuno Diagnostics, Inc. (Woburn, MA). T-tropic HIV-gp120 derived from the HIV IIIB strain was purchased from Protein Sciences (Meriden, CT). All of the proteins were expressed in insect cells using the baculovirus expression system and found to be highly pure and endotoxin-free. Antibodies to SHP2, p38, Grb2, Shc, and PY99 were procured from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Phospho-p38 (Thr 180 /Tyr 182 ) was from BIOSOURCE International (Camarillo, CA), and SB202190, SB203580, SB356095, and okadaic acid were from Calbiochem. Lipopolysaccharide (LPS; Escherichia coli 0111:B4), phosphonitrophenyl phosphate (pNPP) and sodium orthovanadate were from Sigma. Minimum essential Eagle's medium was purchased from Invitrogen, Dulbecco's modified Eagle's medium was from Cellgro (Herndon, VA), and the fetal calf serum was from Sigma.
Cell Culture and Stimulation of Cells-Human hepatocellular carcinoma cells (HepG2) were obtained from ATCC (HB-8065). Cells were cultured in ATCC medium (minimum essential Eagle's medium with 2 mM L-glutamine and Earle's buffered salt solution adjusted to contain 1.5 g/liter sodium bicarbonate, 0.1 nM nonessential amino acids, 1.0 mM sodium pyruvate, and 10% fetal bovine serum) supplemented with 1% penicillin (10,000 IU/ml) and streptomycin (10,000 g/ml) at 37°C in 5% CO 2 in a water-saturated atmosphere.
Cells cultured to about 65% confluence were starved for 3 h in the serum-free medium and then stimulated with 1.5 nM HCV-E2 and 0.8 nM HIV-gp120 (i.e. 100 ng/ml of each dissolved in cell culture medium). Controls received the appropriate volume of cell culture medium. For the studies with inhibitors, SB202190, okadaic acid, N ␣ -p-tosyl-L-lysine chloromethyl ketone (TLCK), and lactacystin were dissolved in Me 2 SO and added to the culture medium 45 min prior to stimulation at various concentrations. After stimulation for the designated time, cells were lysed with radioimmune precipitation assay buffer (50 mM Tris-HCl, pH 7.4, 1% Nonidet P-40, 150 mM NaCl) containing 1 mM phenylmethylsulfonyl fluoride; 10 g/ml aprotinin, leupeptin, and pepstatin; 10 mM sodium vanadate; 10 mM sodium fluoride; and 10 mM sodium pyrophosphate. Total cell lysates were clarified by centrifugation at 10,000 ϫ g for 20 min. Protein concentrations were determined with a Bio-Rad protein assay.
The human microvascular endothelial cell line (HMEC-1) was obtained from the Centers for Disease Control and Prevention (Atlanta, GA). Cells were cultured in Dulbecco's modified Eagle's medium (with 2 mM L-glutamine, 4.5 g/liter glucose, and 10% fetal bovine serum) supplemented with 1% penicillin (10,000 IU/ml) and streptomycin (10,000 g/ml) at 37°C in 5% CO 2 in a water-saturated atmosphere.
Cells cultured to about 65% confluence were starved for 3 h in the serum-free medium and then stimulated with LPS at a concentration of 1 g/ml in cell culture medium. Controls received the appropriate volume of cell culture medium. After stimulation for 24 h, the supernatants were collected and used for the IL-8 assay.
Immunoprecipitation and Western Blot Analysis-For the immunoprecipitation studies, equal amounts of protein from each sample were clarified by 50% of protein A-Sepharose beads for 1 h at 4°C. The cell lysates were incubated with specific antibodies at 4°C for a minimum of 2 h. The immune complexes were bound to protein A-Sepharose beads overnight at 4°C. Nonspecific bound proteins were removed by washing the Sepharose beads three times with radioimmune precipitation assay buffer (50 mM Tris-HCl, pH 7.4, 1% Nonidet P-40, 150 mM NaCl) and once with phosphate-buffered saline. Bound proteins were boiled with 2ϫ Laemmli sample buffer, and the precipitated proteins were separated on 10% SDS-polyacrylamide gel.
Using the semidry Western blotting method, the electrophoretically separated proteins were transferred onto nitrocellulose membrane. Nonspecific binding was blocked with 5% nonfat milk protein in TBST (20 mM Tris, pH 7.4, 137 mM NaCl, and 10% Tween 20), and the blot was probed with primary antibody at a 1:1000 dilution in milk solution at 4°C overnight. After extensive washing with TBST, the blot was incubated with secondary antibody (donkey anti-rabbit IgG or sheep antimouse IgG conjugated to horseradish peroxidase) for 2 h at room temperature. After further washing with TBST, the immunoblot was developed with the ECL system (PerkinElmer Life Sciences) following the manufacturer's instructions.
DNA Constructs and Transfection Procedures-The cDNAs for SHP2 wild type and SHP2 inactive mutant in a Cldn 6368 expression vector were generously provided by Dr. Benjamin G. Neel (Beth Israel Deaconess Medical Center, Boston, MA). The plasmid for the inactive protein-tyrosine phosphatase was made by mutating Cys 459 of SHP2 to Ser. The p38 MAP kinase wild type and p38 MAP kinase mutant were a kind gift from Dr. Roger J. Davis (University of Massachusetts Medical School, Worcester, MA). The plasmids, pCMV5-FLAG-p38 MAP kinase and p38 MAP kinase AGF mutant (Thr and Tyr at positions 180 and 182 of p38 MAP kinase mutated to Ala and Phe, respectively), were cloned as previously described (30). Transfection of HepG2 cells was performed with LipofectAMINE (Invitrogen) for 48 h. Briefly, cells grown to 70% confluence in 35-mm plates were incubated with DNA (2 g of total)-LipofectAMINE (10 l) complexes in serum-free medium overnight. Then the cells were grown in the presence of media with 10% serum. The cells were washed with phosphatebuffered saline and starved with serum-free medium for 3 h and used for stimulation as described above. To assess transfection efficiency, cells co-transfected with pSV-␤-galactosidase control vector (Promega, Madison, WI) were processed to determine the galactosidase activity using a Promega kit according to the manufacturer's recommendations.
IL-8 Measurement-IL-8 secretion was measured by ELISA (Endogen, Woburn, MA). The cells grown to 65% confluence were stimulated as described above for 24 h. The supernatants collected were assayed for IL-8 using an ELISA assay kit according to the manufacturer's instructions.
p38 in Vitro Kinase Assay-Following stimulation, cell lysates were immunoprecipitated with p38 rabbit polyclonal antibody as described above. The beads were washed twice in radioimmune precipitation assay buffer and once in kinase buffer (20 mM HEPES, pH 7.5, 10 mM MgCl 2 , 50 mM NaCl, 1.5 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, 0.05 mM dithiothreitol, 0.1 mM Na 3 VO 4 ). Kinase assays were then performed using 5 Ci of [␥ 32 P]ATP/reaction and 2 g of glutathione S-transferase-ATF-2 (Cell Signaling, Beverly, MA). After 30 min of incubation at 30°C, the reaction was terminated by boiling with 2ϫ SDS sample buffer. The samples were separated by SDS-PAGE on a 12% polyacrylamide gel, dried, and subjected to autoradiography.
In Vitro SHP2 Phosphatase Assay-After stimulation, the cell lysates were immunoprecipitated with SHP2 rabbit polyclonal antibody as described above. The beads were washed twice with radioimmune precipitation assay buffer and once with phosphatase assay buffer (100 mM sodium acetate at pH 5.0 and 1.6 mM dithiothreitol). The immunoprecipitates were incubated in phosphatase assay buffer containing 10 mM pNPP as the substrate at 30°C for 1 h. The reaction was terminated by the addition of 0.2 N sodium hydroxide. Absorbance of the reaction mixture was measured immediately at 410 nm.
Electrophoretic Mobility Shift Assays (EMSAs)-Preparations of nuclear and cytoplasmic extracts were performed with an NE-PER TM nuclear and cytoplasmic extraction reagent kit (Pierce). Nuclear extracts were prepared after stimulation for 20 min as described above. The double-stranded probe containing the NF-B consensus site 5Ј-AGTTGAGGGGACTTTCCCAGGC-3Ј and the AP-1 consensus oligonucleotide 5Ј-CGCTTGATGAGTCAGCCGGAA-3Ј from Promega were biotin-labeled using a Biotin 3Ј-end DNA labeling kit (Pierce). EMSA was performed using a LightShift TM chemiluminescent EMSA kit (Pierce) according to the manufacturer's instructions.
Transient Transfection and Luciferase Reporter Assay-HepG2 and HMEC-1 cells were cultured in 24-well plates and grown to 70% confluence. The cells were transfected with 0.5 g/well of pNFB-Luc vector or pTal-Luc vector (negative control) (BD Biosciences Clontech, Palo Alto, CA) using LipofectAMINE Plus reagent according to the manufacturer's protocol (Invitrogen) for 48 h. To assess transfection efficiency, cells were co-transfected with pSV-␤-galactosidase control vector (Promega). The cells were washed with phosphate-buffered saline and starved with serum-free medium for 3 h. Then the transfectants were stimulated with 1.5 nM HCV-E2 and 0.8 nM HIV-gp120 (i.e. 100 ng/ml each) in HepG2 cells for 20 min or with LPS (1 g/ml) in HMEC-1 cells for 45 min. Using the luciferase reporter assay kit (BD Biosciences Clontech, Palo Alto, CA), the cells were lysed, and the luciferase reporter assay was performed as per the manufacturer's instructions. Chemiluminescent detection of the luciferase activity was performed with a luminometer. ␤-Galactosidase activity was determined using a Promega kit to verify the reproducibility of the quadruplicate transfections in all experiments.
Statistical Analysis-All of the experiments were reproducible and were carried out in duplicates or quadruplicates. Each set of experiments was repeated at least three times with similar results, and a representative one is shown. The results are presented as the means Ϯ S.D. Student's t test for paired samples was used to determine statistical significance. Differences were considered statistically significant at a value of p Յ 0.05.

RESULTS
HCV-E2 in Conjunction with HIV-gp120 Induces IL-8 Production-Hepatic inflammation has been shown to be an important marker in HCV-infected HIV patients (31,32). In HCV patients with liver inflammation and fibrosis, elevated serum levels of adhesion molecules, E-selectin, IL-8, and tumor necrosis factor-␣ were reported (33,34). Our initial studies revealed that one of the components of the HCV structural proteins, HCV-E2, slightly induced IL-8 production in a dose-dependent manner (data not shown).
Following these initial studies, we sought to model events in hepatocytes in hosts infected with both HCV and HIV. HIV-gp120 is known to be an important signaling molecule through its binding to the surface chemokine receptors CXCR4, CCR5, and CCR3 (35). We observed robust expression of CXCR4 and CCR3 and lower amounts of CCR5 on both HepG2 and primary hepatocytes by using semiquantitative PCR (data not shown). We also observed low expression of CD4 in HepG2 cells by flow cytometry as well as expression of CD81 on both HepG2 cells FIG. 1. Induction of IL-8 by HCV-E2 and HIV-gp120. HepG2 cells were plated in 24-well plates and grown in complete minimum essential Eagle's medium as described under "Experimental Procedures." At 65% confluence, the cells were washed with serum-free medium and treated with 0 -15 nM (0 -1000 ng/ml) concentrations of HCV-E2 and 0 -8 nM (0 -1000 ng/ml) concentrations of HIV-gp120 (A) or HCV-E2 (1.5 nM), gp120 (0.8 nM), M-gp120 (0.8 nM), HCV-E2 plus gp120, HCV-E2 plus M-gp120, or various combinations of activated and/or heat-inactivated (⌬) E2 (1.5 nM), gp120 (0.8 nM), and M-gp120 (0.8 nM), as indicated (B). C, primary hepatocytes, plated in 24-well plates and growing in hepatocyte basal medium in 0.5% fetal calf serum, were untreated (UN) or treated with HCV-E2 (1.5 nM) and HIV-gp120 (0.8 nM) or heat-inactivated E2 and gp120 (⌬E2 and ⌬gp120) as indicated. D, HepG2 cells were pretreated with SB356095 (a CXCR4 inhibitor) at different concentrations of 0.1, 1, and 10 nM for 45 min and then stimulated with HCV-E2 (1.5 nM) and HIV-gp120 (0.8 nM). The supernatants were collected after 24 h and assayed by ELISA (Endogen) for the production of IL-8. Data represent one of three independent experiments performed in quadruplicates. The results are shown as the means Ϯ S.D. *, p Յ 0.05. gp120 represents T-gp120.
p38 MAP Kinase and SHP2 Regulate IL-8 Induction and primary hepatocytes by PCR (data not shown). We then examined the effects of T-cell tropic HIV-gp120 derived from HIV IIIB, an X4 strain that specifically binds to CXCR4 (35), on HCV-E2-induced IL-8 production. HIV-gp120 enhanced the production of IL-8 from HepG2 cells in the presence of HCV-E2 (Fig. 1A). It is evident from the data that HCV-E2 and HIV-gp120 exhibit an optimal cooperativity at concentrations of 1.5 nM HCV-E2 (100 ng/ml) and 0.8 nM HIV-gp120 (100 ng/ml). Subsequently, we performed all of our experiments at these concentrations. The induction of IL-8 appeared to be specific, since we did not detect induction of MCP-1, MIP-1␣, MIP-1␤, or tumor necrosis factor in these treated cells, as determined by ELISA (R&D Systems, Minneapolis, MN) (data not shown).
To assess the specificity of HIV-gp120 and the chemokine receptor CXCR4, we further analyzed the effect of macrophage cell tropic HIV-gp120 (M-gp120) derived from YU2, an R5 strain that binds to the CCR5 receptor. HepG2 cells were stimulated with HCV-E2 (1.5 nM) and T-tropic gp120 (gp120, 0.8 nM) or M-tropic gp120 (0.8 nM) either separately or in combination at these concentrations. Heat-inactivated (⌬) proteins were used in a similar fashion and served as controls for the observed effects. Fig. 1B shows the cooperative effect of HCV-E2 and HIV-gp120 on IL-8 production in comparison with untreated or HCV-E2-or HIV-gp120-treated samples. Moreover, we did not find any significant difference in the degree of IL-8 induction between the T-tropic and M-tropic HIV-gp120. Furthermore, various combinations of the heat-inactivated proteins were unable to induce IL-8 as compared with the active proteins. A similar result was obtained with primary hepatocytes (Fig. 1C). These initial data suggest that there might be an enhanced inflammatory response in hepatocytes upon simultaneous exposure to HCV-E2 and HIV-gp120. This could contribute to the clinical finding of frequent and often rapidly progressive inflammation and fibrosis in patients co-infected with these two viruses.
The binding of HCV-E2 to CD81 has been well established (36 -38). However, several studies have shown that HCV associates with low density lipoprotein (LDL), suggesting that the virus could use the LDL receptor as a receptor (39,40). Moreover, HCV-E2 has been proposed to interact with negatively charged molecules such as glycosaminoglycans (41). Therefore, the downstream signaling induced by HCV-E2 may be due to its binding with one of these receptors. However, the binding of T-tropic HIV-gp120 to CXCR4 has been well established (42). To address the role of the CXCR4 receptor in HCV-E2-and HIV-gp120-induced IL-8 production, a dose-dependent study was done with the CXCR4 inhibitor SB356095. We found a gradual block in the induction of IL-8 (13, 29, and 43%) upon increasing concentrations of SB356095 (Fig. 1D). The cells were found to be viable at these concentrations of the inhibitor. However, the inhibition was not complete, emphasizing the partial involvement of the CXCR4 receptor in the downstream signaling induced by HCV-E2 and HIV-gp120.
p38 MAP Kinase Activation Contributes to the Secretion of IL-8 Induced by HCV-E2 and HIV-gp120 -It is known that members of the MAP kinase family regulate several transcription factors that are important in the proinflammatory response (43,44). IL-8 gene expression has been shown to be dependent on different MAP kinase activators in various cell types (45,46). We found that HCV-E2 or HIV-gp120 alone modestly induced p38 MAP kinase activity. However, this induction was augmented (over 3-fold) upon co-treatment with HCV-E2 and HIV-gp120 ( Fig. 2A). Phosphorylation studies of p38 MAP kinase correlated well with its activity (Fig. 2, B and  C, top panels). Equal amounts of p38 protein were present in each lane (Fig. 2, B and C, bottom panels). Recent studies have also shown that inhibition of p38 MAP kinase activity suppresses the IL-8 gene expression, thereby regulating inflammatory response (47,48). SB202190 and SB203580, specific inhibitors of p38 MAP kinase, were used to assess the contribution of p38 MAP kinase to IL-8 production. Fig. 3 (A and B) clearly indicates that the p38 MAP kinase inhibitors significantly decreased IL-8 production in a dose-dependent manner in cells stimulated with HCV-E2 plus HIV-gp120. To further confirm the role of p38 MAP kinase in IL-8 induction, we transiently transfected HepG2 cells with either wild type or a kinase-inactive mutant of p38 MAP kinase. These studies revealed that mutant p38 MAP kinase reduced IL-8 induction by HCV-E2 and HIV-gp120 in HepG2 cells (Fig. 3C). A concentration dependent decrease in IL-8 production induced by HCV-E2 and HIV-gp120 was observed when HepG2 cells were transiently transfected with various DNA concentrations of the p38 MAP kinase-inactive mutant (p38 MT) (Fig. 3D). Appropriate concentrations of vector DNA were also transfected and used as controls. However, it appears that the mutant is partially active, since, even at higher DNA concentrations (10 g/ml), we could not achieve complete inhibition of IL-8. Transfection efficiency was determined by the ␤-galactosidase assay (data not shown). This indicates that p38 MAP kinase partially mediates HCV-E2-and HIV-gp120-induced IL-8 production. p38 MAP Kinase and SHP2 Regulate IL-8 Induction HCV-E2 plus HIV-gp120 Induces Interaction of p38 MAP Kinase with SHP2 and Adaptor Molecules-To identify the signaling molecules that regulate p38 MAP kinase activity, we analyzed the proteins that interact with p38 MAP kinase upon HCV-E2 and HIV-gp120 treatment. We found that HCV-E2 and HIV-gp120 treatment induced the association of the tyrosine phosphatase SHP2 with p38 MAP kinase (Fig. 4A). We also found a significant increase in the association between p38 MAP kinase and the adaptor molecules Grb2 and Shc (Fig. 4C). Reciprocal immunoprecipitations with SHP2, Grb2, or Shc antibodies and Western blot analysis with p38 MAP kinase antibody were performed to confirm their respective interactions with p38 MAP kinase (Fig. 4, B, D, and E). It has been previously shown that SHP2 interacts with Grb2 and Shc in hepatocyte growth factor-induced signaling (49). We found a similar interaction of Grb2 and Shc with SHP2 during HCV-E2 and HIV-gp120 co-stimulation (data not shown). The association of SHP2, Shc, and Grb2 has also been observed to activate MAP kinase in activated endothelial cells at sites of inflammation (50). To determine the specificity of interaction, immunoprecipitated p38 MAP kinase was probed with various antibodies of signaling molecules such as LDL receptor, phosphatidylinositol 3-kinase, Bad, FasL, c-Jun, and c-Fos. None of these molecules showed any physical interaction with p38 MAP kinase (data not shown).
Role of SHP2 in the IL-8 Secretion Induced by HCV-E2 and HIV-gp120 -The association of p38 MAP kinase with SHP2 and adaptor molecules led us to analyze the role of SHP2 in HCV-E2-and HIV-gp120-induced IL-8 secretion. SHP2 showed a marked time-dependent increase in tyrosine phosphorylation upon HCV-E2 and HIV-gp120 co-stimulation (Fig. 5A). However, no time-dependent change in tyrosine phosphorylation of SHP1 was observed under similar conditions (data not shown).
A previous study in a human promyelocyte cell line (an HL-60 subline) reported that upon okadaic acid and vanadate stimulation, there was a marked increase in IL-8 mRNA gene expression (23). Thus, we next studied the involvement of SHP2 in IL-8 production by using sodium orthovanadate, a p38 MAP Kinase and SHP2 Regulate IL-8 Induction tyrosine phosphatase inhibitor. We used okadaic acid, a serine/ threonine phosphatase inhibitor, as a control. Interestingly, we found significant inhibition of IL-8 production in a dose-dependent manner in the presence of vanadate (Fig. 5B). However, the presence of okadaic acid did not inhibit IL-8 secretion when HepG2 cells were co-stimulated with HCV-E2 and HIV-gp120 (Fig. 5C). To further confirm the role of SHP2 in HCV-E2-plus HIV-gp120-induced IL-8 production, HepG2 cells were transiently transfected with wild type SHP2 and the catalytically inactive SHP2 mutant (C459S). Transfection efficiency was determined by ␤-galactosidase assay (data not shown). Functional analysis of these transfectants revealed that wild type SHP2 (SHP2 WT) exerted no significant effect on IL-8 induction, whereas overexpression of the catalytically inactive mutant of SHP2 (SHP2 MT) resulted in a reduction in IL-8 (Fig. 5D). These results suggest that SHP2 could be involved in mediating IL-8 expression induced by HCV-E2 and HIV-gp120.
p38 MAP Kinase Regulates the Phosphorylation of SHP2-To explore the order in which these signals are transmitted, we determined the activity of p38 MAP kinase in the presence of vanadate, a tyrosine phosphatase inhibitor, or okadaic acid as a control inhibitor. We did not find any change in the enhanced activity of p38 MAP kinase induced by HCV-E2 and HIV-gp120 in the presence of either vanadate or okadaic acid (Fig. 6, A and  B). These data indicate that SHP2 might act downstream of p38 MAP kinase. In parallel, we found a decrease in the phosphorylation of SHP2 in the presence of SB202190 and SB203580, specific inhibitors of p38 MAP kinase (Fig. 6C). We also observed a significant decrease in SHP2 phosphatase activity in the presence of specific p38 MAP kinase inhibitors (Fig. 6D). Similarly, overexpression of the p38 MAP kinaseinactive mutant also showed reduced SHP2 phosphorylation (Fig. 6E) and phosphatase activity (Fig. 6F), thus confirming the study with the p38 MAP kinase inhibitors. This result suggests that p38 MAP kinase may regulate the phosphorylation and activity of SHP2.

Role of NF-B and AP-1 in IL-8 Expression Induced by HCV-E2
and HIV-gp120 -In general, NF-B is considered a key factor in the regulation of inflammation because of its ability to control the expression of numerous inflammatory mediators (51)(52)(53). Recently, it has been found that p38 MAP kinase regulates NF-B-driven gene expression (54). Therefore, NF-B could be a prospective candidate involved in the expression of IL-8 induced by HCV-E2 and HIV-gp120. NF-B (p65 subunit) was found to associate with p38 MAP kinase in a time-dependent manner (Fig.  7A). However, in the presence of TLCK, an inhibitor of NF-B, we found no significant blocking of the IL-8 production induced by HCV-E2 and HIV-gp120 (Fig. 7B). We also found no significant inhibition of IL-8 induction in the presence of lactacystin, a proteasome inhibitor (Fig. 7C). However, in HMEC-1 cells, both inhibitors decreased significantly the LPS-induced IL-8 production when compared with the stimulated control, which confirms the effectiveness of these inhibitors. Proteosomes have been shown to activate NF-B by degrading IB (55-58). Further- more, we observed that HCV-E2 and HIV-gp120 treatment did not induce the DNA binding activity of NF-B (Fig. 8A). Densitometric quantification of the blot demonstrated no obvious increase in the DNA binding activity of NF-B when compared with the unstimulated sample (data not shown). To further confirm that NF-B is not activated in this signaling pathway, a luciferase reporter assay was done. For this assay, we transiently transfected HepG2 cells with the pNF-B-Luc vector which contains an NF-B consensus sequence fused to the TATA-like promoter (P TAL ) and the firefly luciferase gene, which acts as the reporter. The control vector pTal-Luc, which contains only P TAL and the reporter gene, was also transfected to serve as a negative control. The transfectants stimulated with HCV-E2 and HIV-gp120 showed no change in luciferase activity (Fig. 8B, right  panel). As a positive control, the HMEC-1 cells were transfected with these vectors, followed by stimulation with LPS, which showed a significant increase in luciferase activity (Fig. 8B, left  panel). Transfection efficiency was determined by measuring the ␤-galactosidase activity (data not shown). The fold increase was calculated based on the unstimulated group, which was transfected with the pTal-Luc vector in each cell line.
p38 MAP kinase has also been shown to regulate AP-1-dependent processes (59). In our study, we found an increase in the DNA binding activity of AP-1 (Fig. 8C). Furthermore, blocking of p38 MAP kinase with SB202190 reduced the DNA binding activity of AP-1 (Fig. 8C, lanes 4 and 5). Densitometric quantification of the blot demonstrated that the DNA binding activity of AP-1 was increased 1.5-fold upon HCV-E2 and HIV-gp120 stimulation when compared with the unstimulated control. However, in the presence of SB202190, there was no significant difference in the DNA binding activity among the unstimulated and stimulated samples. These results suggest that IL-8 induced by HCV-E2 and HIV-gp120 follows a novel signaling pathway, which is mediated by p38 MAP kinase and SHP2 via an NF-B-independent mechanism and possibly through AP-1-driven transcription.

DISCUSSION
The mechanism of the enhanced progression of hepatic disease in 35-50% of HIV patients co-infected with HCV is not known (1, 60 -62). The pathogenesis of HCV/HIV co-infection is also poorly understood. We have shown in an in vitro model system that HCV and HIV structural proteins at concentrations within the range of K d induce the secretion of the proinflammatory molecule IL-8. The release of a potent inflammatory chemokine like IL-8 would function to foster local inflammatory changes that induce fibrosis and ultimately cirrhosis. This suggests that HCV and HIV structural proteins might injure uninfected hepatocytes through an "innocent bystander" mechanism.
In HIV patients infected with HCV, hepatic inflammation has been shown to be an important marker of the disease (31,32). Furthermore, in HCV patients with liver inflammation and fibrosis, elevated serum levels of IL-8, tumor necrosis factor-␣, adhesion molecules, and E-selectin have been reported (33,34). Moreover, prior data on liver specimens from patients with HCV infection indicate robust expression of IL-8, thus confirming the pathophysiological relevance of our findings (45,63). In addition to the chemotactic properties of IL-8, this chemokine is a natural antagonist of interferon (64).
Petracca et al. (38) reported that HCV-E2 binds to CD81 with a K d value of 1.8 nM in vitro. However, we do not know the physiological concentration of HCV-E2 in vivo in HCV-infected FIG. 6. p38 MAP kinase regulates the phosphorylation of SHP2. HepG2 cells were pretreated with orthovanadate (vanadate; a tyrosine phosphatase inhibitor) (A) or with okadaic acid (serine/threonine phosphatase inhibitor) (B) for 45 min and then unstimulated (Ϫ) or stimulated (ϩ) with HCV-E2 and HIV-gp120 for 20 min. Cell lysates (1 mg of protein) of each sample were immunoprecipitated (IP) with anti-p38 antibody, and the immunocomplexes were assayed for p38 MAP kinase activity using both glutathione S-transferase-ATF2 as substrate and radiolabeled [␥-32 P]ATP. C, HepG2 cells pretreated with SB202190 or SB203580 (specific p38 MAP kinase inhibitors) for 45 min were stimulated with HCV-E2 and HIV-gp120 for 20 min. The cell lysates were immunoprecipitated with anti-SHP2 antibody and probed with anti-phosphotyrosine antibody (pTyr) to detect the phosphorylation of SHP2 (top panel). Equal amounts of protein were seen when the blot was stripped and reprobed with anti-SHP2 antibody (bottom panel). D, SHP2 phosphatase activity was determined in the immunocomplexes of SHP2 using pNPP as the substrate at 410 nm. E, HepG2 cells were untransfected or transiently transfected with pCMV (control) vector, pCMV5-FLAG-p38 (AGF) mutant (p38 MT), or pCMV5-FLAG-p38 wild type (p38 WT). After 48 h, the efficiency of transfection was determined by ␤-galactosidase assay. The transfectants were then unstimulated (UN) or stimulated with HCV-E2 (1.5 nM) and HIV-gp120 (0.8 nM) for 20 min. The cell lysates were immunoprecipitated with anti-SHP2 antibody and probed with anti-phosphotyrosine antibody (pTyr) to detect the phosphorylation of SHP2 (E, top panel). Equal amounts of protein were seen when the blot was stripped and reprobed with anti-SHP2 antibody (E, bottom panel). F, SHP2 phosphatase activity in the transfectants was determined in the immunocomplexes of SHP2 using pNPP as the substrate at 410 nm. The data represent one of three independent experiments done. *, p Յ 0.05. TCL, total cell lysates; Ab, antibody control; WB, Western blot; DMSO, Me 2 SO. patients. We have titrated the optimal concentration for HCV-E2 to induce IL-8 production to be 1.5 nM. In vivo measurements of HIV-gp120 show that its concentration ranges from 0.1 to 1 nM (65,66). Moreover, it has been reported that the K d value for the HIV-gp120 and CD4 interaction in vitro is within the range of 1-4 nM (67). Therefore, we have determined the optimal concentration for HIV-gp120 to induce IL-8 production to be 0.8 nM. Our results show that downstream signaling and maximum induction of IL-8 occurred when cells were exposed simultaneously to both HIV T-tropic or M-tropic gp120 and HCV-E2 proteins at concentrations within the range of K d and thus probably reflecting pathophysiological conditions. In this setting of co-infection, collaborative signaling induced by two distinct viruses is novel and extends the past studies of viral pathogenesis that have been performed with each virus alone.
We observed activation of p38 MAP kinase following collaborative stimulation with HCV-E2 and HIV-gp120. Furthermore, p38 MAP kinase was shown to mediate IL-8 induction. p38 MAP kinase has been reported to participate in the inflammatory responses triggered by LPS, IL-1, or tumor necrosis factor-␣ (30, 68, 69). Moreover, it has also been shown to regulate IL-8 secretion in intestinal inflammation and vascular diseases (45,46). Although p38 MAP kinase has been shown to regulate the expression of proinflammatory genes, its upstream and downstream regulators are not well known.
In the present study, we have further defined the p38 MAP kinase-mediated signaling pathway that leads to HCV-E2-and HIV-gp120-induced IL-8 production. We observed an enhanced association of p38 MAP kinase with the adaptor proteins Grb2 and Shc upon HCV-E2 and HIV-gp120 stimulation. Reciprocal immunoprecipitation studies conducted also confirmed these interactions. Adaptor proteins have recently been shown to regulate MAP kinase activities (70). Furthermore, TAB-1, an adaptor protein with no catalytic activity, has been shown to activate p38 MAP kinase (71).
We also found that HCV-E2 and HIV-gp120 induced the tyrosine phosphorylation of SHP2. Moreover, overexpression of a dominant negative mutant of SHP2 significantly reduced IL-8 induction. Although protein tyrosine phosphatases have been shown to influence IL-8 expression (23), this, to our knowledge, is the first report that shows directly that IL-8 expression could be positively regulated by SHP2.
We have observed that p38 MAP kinase associates with SHP2 and regulates its phosphorylation and activity upon stimulation with HCV-E2 and HIV-gp120. We also have shown that tyrosine phosphatase inhibitors did not block p38 MAP kinase activity. This is in contrast to other reports that suggest that protein phosphatases act as negative regulators of MAP kinases (72,73). Taken together, these findings suggest that p38 MAP kinase and SHP2 mediate HCV-E2-and HIV-gp120induced IL-8 production. and HIV-gp120 (0.8 nM) for 20 min. The cells were harvested, and nuclear extracts were prepared. EMSA was performed using 4 g of nuclear extract with biotin-labeled NF-B oligonucleotide (5Ј-AGTTGAGGGGACTTTCCCAGGC-3Ј) in the presence of 1-2 g of poly(dI⅐dC). B, HepG2 cells and HMEC-1 cells were transiently transfected with the pTal-Luc (control) vector or pNF-B-Luc vector. After 48 h, the transfection efficiency was determined by measuring the ␤-galactosidase activity. The transfectants of HepG2 cells were stimulated with HCV-E2 (1.5 nM) and HIV-gp120 (0.8 nM) for 20 min, whereas the transfectants of HMEC-1 cells were stimulated with LPS (1 g/ml) for 45 min. The cells were lysed, and the luciferase activity was detected by chemiluminescence (BD Biosciences Clontech) using a luminometer. The fold increase was calculated based on the unstimulated sample, which was transfected with the pTal-Luc vector in each cell line. C, HepG2 cells were pretreated with SB202190 (a specific p38 MAP kinase inhibitor) for 45 min and stimulated with HCV-E2 (1.5 nM) and HIV-gp120 (0.8 nM) for 20 min. The cells were harvested, and nuclear extracts were prepared. EMSA was performed using 4 g of nuclear extract with biotin-labeled AP-1 oligonucleotide (5Ј-CGCTTGAT-GAGTCAGCCGGAA-3Ј) in the presence of 1-2 g of poly(dI⅐dC). The reaction mixtures of both the EMSAs were electrophoresed on a 6% native polyacrylamide gel. Electrophoresed samples were transferred to nylon membrane, and the cross-linked labeled DNA was detected by chemiluminescence (Pierce). The data represent one of three independent experiments performed in quadruplicates. The results are shown as the means Ϯ S.D. *, p Յ 0.05. p38 MAP Kinase and SHP2 Regulate IL-8 Induction p38 MAP kinase has been shown to regulate NF-B and AP-1 transcriptional factors (54, 74 -76). NF-B has been shown to activate various inflammatory genes and has also been reported to play a central role in inflammatory diseases (51)(52)(53). Although we found an increased association of p38 MAP kinase with NF-B, its DNA binding activity was not enhanced upon HCV-E2 and HIV-gp120 treatment. In addition, the luciferase reporter assay was done to confirm that NF-B was not activated upon HCV-E2 and HIV-gp120 stimulation. Furthermore, inhibitors of the NF-B pathway did not block HCV-E2-and HIV-gp120-induced IL-8 production. Apart from NF-B, IL-8 gene expression has been shown to be regulated by AP-1-driven processes (77)(78)(79). In this investigation, we found an increase in the DNA binding activity of the AP-1 complex, which was blocked by the p38 MAP kinase inhibitor SB202190. p38 MAP kinase is known to up-regulate IL-8 expression by activating NF-B and AP-1. However, in certain cases such as the presence of proteasome inhibitors (80) or during sepsis (81) and cell stress (82), AP-1-driven transcription of IL-8 was activated even when the NF-B pathway is inhibited.
Based on our study, the cell dysfunction and inflammatory response induced by HCV-E2 and HIV-gp120 could be due in part to an "innocent bystander" mechanism. We have delineated a novel signaling pathway for this inflammatory response induced by these HCV and HIV proteins, which is mediated by p38 MAP kinase and SHP2 but is independent of NF-B activation. These signaling studies could help in the development of therapeutic strategies for controlling the severe liver damage observed in HCV/HIV co-infected patients.