The HIV Matrix Protein p17 Promotes the Activation of Human Hepatic Stellate Cells through Interactions with CXCR2 and Syndecan-2

Background The human immunodeficiency virus type 1 (HIV-1) p17 is a matrix protein involved in virus life's cycle. CXCR2 and Syndecan-2, the two major coreceptors for the p17 protein, are expressed in hepatic stellate cells (HSCs), a key cell type involved in matrix deposition in liver fibrotic disorders. Aim In this report we have investigated the in vitro impact of p17 on HSCs transdifferentiation and function and underlying signaling pathways involved in these processes. Methods LX-2 cells, a human HSC line, and primary HSC were challenged with p17 and expressions of fibrogenic markers and of p17 receptors were assessed by qRT-PCR and Western blot. Downstream intracellular signaling pathways were evaluated with qRT-PCR and Western blot as well as after pre-treatment with specific pathway inhibitors. Results Exposure of LX2 cells to p17 increases their contractile force, reshapes the cytoskeleton fibers and upregulates the expression of transdifferentiation markers including αSMA, COL1α1 and endothelin-1 through the activation of Jak/STAT and Rho signaling pathways. These effects are lost in HSCs pre-incubated with a serum from HIV positive person who underwent a vaccination with a p17 peptide. Confocal laser microscopy studies demonstrates that CXCR2 and syndecan-2 co-associate at the plasma membrane after exposure to p17. Immunostaining of HIV/HCV liver biopsies from co-infected patients reveals that the progression of liver fibrosis correlates with a reduced expression of CXCR2. Conclusions The HIV matrix protein p17 is pro-fibrogenic through its interactions both with CXCR2 and syndecan-2 on activated HSCs.


Introduction
In the era of effective antiretroviral therapy (ART), liver disease is the second most common cause of death among persons with human immunodeficiency virus (HIV) infection and a strong association between immunodeficiency and risk of liver-related death, exists [1,2]. In addition, hepatitis B and C virus (HBV and HCV) and non-alcoholic steatohepatitis (NASH) are found frequently in HIV infected persons greatly increasing the burden for liver diseases [3,4]. The mechanisms that drives liver injury in HIV infected persons are several and include both direct and indirect pathways. To date, amongst the HIV proteins, only the HIV envelope protein gp120 has been demonstrated to exert a direct profibrogenic action on humans Hepatic Stellate cells (HSCs), thus identifying a direct mechanism possibly linking HIV infection with liver fibrogenesis via envelope proteins [5][6][7]. Indeed, the HIV envelope protein gp120 directly acts on HSCs by binding and activating both CCR5 or CXCR4 receptors (the two major HIV co-receptors) and syndecans [5][6][7][8]. Syndecans are type I transmembrane cell surface heparan sulfate proteoglycans (HSPGs) that function as co-factors in cell-cell adhesion, in linking cells to ligands in the extracellular matrix, and in the binding of cellular growth factors [9,10]. They also function as co-receptors for HIV-1 entry into primary target cells [11,12]. Indeed, the enzymatic removal of cell surface heparan sulfates or the addition of soluble heparan sulfates, while leaving CD4 and chemokine receptors (CCR5 and CXCR4) unchanged, drastically reduces HIV-1 adhesion to and entry into specific target cells, including CD4+ T cells, Hela cells and macrophages [11].
The HIV matrix protein p17 is a structural protein that plays important functions in the viral replication cycle such as the recruitment of the viral surface/transmembrane gp120/gp41 envelope protein complex into virions [13] as well as the targeting of Pr55Gag proteins to their assembly sites at the plasma membrane of infected cells [14,15]. The p17 exerts its biological activity on immune cells (T lymphocytes, monocytes and dendritic cells) upon interaction with the IL-8 receptors, CXCR1 and CXCR2 [16]. Furthermore, an interaction between p17 and syndecans (specifically syndecan-2 and -4) has been documented in HeLa cells and in human activated CD4+ T cells. 16,17 The intracellular signal activated by p17 upon interaction with the IL8 receptors or syndecans involves the activation of both Rho/ ROCK1 and JaK/STAT1 pathways [17][18][19].
Hepatic fibrosis and cirrhosis are chronic scarring processes of the liver which associate with increased and altered deposition of extracellular matrix (ECM). In the setting of chronic liver injury, HSCs undergo a process of trans-differentiation from a resting, fatstoring phenotype to a myofibroblast-like phenotype characterized by expression of fibroblastic cell markers such as a-smooth muscle actin (a-SMA). Activated HSCs release increased amounts of ECM components, such as a1-collagen type I, that contribute significantly to the fibrotic changes in cirrhosis [20,21]. In addition, activated HSCs gain contractile phenotype and is well established that the contractile force generated by HSCs contributes to the regulation of sinusoidal blood flow and the development of portal hypertension [22]. A number of vasoactive molecules can trigger contractile response in HSCs with endothelin-1 being the most potent constrictor [23,24].
The aims of the present study were to investigate the effect of p17 on HSC activation, collagen-I and endothelin-1 expression through interactions with both CXCR2 and syndecan-2 and to explore underlying intracellular pathways involved in these interactions. Present results demonstrate that p17 pirates the IL-8 receptors and syndecans to modulates HSCs function. Of clinical relevance we demonstrate that anti-p17 antibodies obtained from a patient included in an anti-p17 vaccination trial [25] protect HSCs from activation caused by the viral protein.

Results
LX2 and primary HSCs express CXCR2 and syndecan-2 and their stimulation with p17 promotes expression of fibrogenic markers Since the HIV matrix protein p17 exerts its pro-inflammatory effects through the binding with CXCR2 and syndecan-2 [16,17,19], we have first examined whether LX2 cells, a human immortalized HSC line, express these two proteins. As shown in Figure 1 A and B, LX2 cells express CXCR2 and syndecan-2 as demonstrated by Western blotting analysis and RT-PCR. However, stimulation with escalating doses of p17 (from 0.1 to 10 mg/ ml) resulted in a down-regulation of the expression of CXCR2 in these cells. Treating LX2 cells with p17 also resulted in a significant increase in the expression of collagen-I and endothelin-1 as confirmed by both Western blotting and RT-PCR while the induction of a-SMA was weak and did not changes significantly in cells exposed to 1 or 10 mg/ml of p17 ( Fig.1 A, B and Fig. S1).
To extend the relevance of these observations to the pathophysiology of human stellate cells, primary stellate cells were stimulated with escalating doses of p17 (from 0.1 to 10 mg/ml) and the relative mRNA expression of CXCR2, syndecan-2, collagen-I, a-SMA and endothelin-1 examined by Real-Time PCR. Results from these experiments confirmed that primary human HSCs express CXCR2 and syndecan-2 and that the stimulation of these cells with p17 results in an induction of collagen-I, a-SMA and endothelin-1 comparable to tha observed in LX2 cells. Similarly to LX2 cells, exposure of human primary HSCs to 10 mg/ml p17 resulted in a significant downregulation of CXCR2 mRNA (Fig. 1C).

Stimulation of LX2 cells with p17 reshapes cytoskeleton fibers
During the process of transdifferentiation HSCs undergo changes in the composition and amounts of cytoskeleton fibers. In particular, protein expression of focal adhesion protein (such as a-SMA and vinculin) as well as that of intermediate filaments (such as desmin and vimentin) changes during HSC activation [21,[26][27]. Thus, we have investigated the expression of vimentin and a-SMA in LX2 cells stimulated 24 h with p17 by confocal immunofluorescence. As shown in Figure 2, exposure of LX2 cells to p17 resulted in an increased organization of intermediate filaments as detected by confocal analysis of vimentin ( Fig. 2A), while minor changes in a-SMA staining was seen in p17 treated cells in comparison with not stimulated cells (Fig. 2B).

Pro-fibrogenic effects of p17 were neutralized by therapeutical vaccination
Having shown that the HIV matrix protein p17 regulates the expression of proteins involved in HSC functions, such as collagen deposition and contraction, we have next investigated whether these effects could be reversed using the serum of an HIV positive patient who underwent a vaccination with a p17 peptide [18,25]. As illustrated in Figure 3A-C, exposure of LX2 cells to p17 neutralizing antibodies retained in the sera of a vaccinated patient completely abrogates the effects of p17 in terms of induction of collagen-I, a-SMA and endothelin-1. Of interest, p17 mediated induction of a-SMA and endothelin-1, but not that of collagen-I was almost completely reversed even by using the serum taken before vaccination indicating that p17 effects could be neutralized with minimal amount of antibodies. As a control for this experiment we used a serum obtained from an HIV-negative subject. As shown in Figure 3A-C, the serum from this subject failed to block the fibrogenic effects exerted by p17 on LX2 cells ( Fig. 3A-C).
Since endothelin-1 is a well characterized molecule involved in HSC contraction [23] and p17 treatment associates with upregulation of endothelin-1 mRNA and protein, we then investigated whether the contractile response of LX2 cells to p17 was enedotelin-1 dependent. To measure the effect of p17 on the contraction of LX2 cells, we used the established assay of hydrated collagen gels [28]. As shown in Figure 2 D, exposure of gel for 6 hour to 10% FBS (positive control), resulted in a significant decrease of the collagen surface area. Compared with positive control, the treatment with p17 caused a similar reduction of the collagen surface area after 18-hour incubation. Interestingly, pretreatment of LX2 cells with serum taken before vaccination failed to reverse the p17 mediated LX2 contraction while the postvaccination serum almost completely abrogated these effects (Fig. 2E). All together these data indicated that anti-p17 immune-neutralization leads to a specific reversion of p17 mediated activities on LX2 cells.
p17 induction of collagen-I and a-SMA is CXCR2 dependent and occurs via Rho/ROCK-1 pathway CXCR2 and/or Syndecan-2 dependent. First, LX2 cells were pretreated with a CXCR2 inhibitor (SB-265610) before exposure to p17. Results obtained by Real-Time PCR analysis demonstrated that the CXCR2 inhibitor effectively reversed the p17-mediated induction of collagen-I and a-SMA, while it failed in reversing the induction of endothelin-1 (Figure 4 A). In a second set of experiments, LX2 cells were pre-treated with a syndecan-2 antibody before exposure to p17. Results from these experiments demonstrated that the blockade of syndecan-2 reversed the p17mediated induction of collagen-I, a-SMA and endothelin-1 (Figure 4 B). Since the intracellular signaling activated by CXCR2 involves the activation of Rho/ROCK-1 [17] and one of the major consequences of activation of Rho kinase is the phosphorylation of myosin light chains (MLC) [29], we have examined whether p17 induces the phosphorylation of MLC, which is critically involved in the contraction of HSC [30]. Results from these investigations demonstrated that exposure of LX2 cells to p17 increases the amount of phosphorylated MLC. In particular, the p17 mediated phosphorylation of MLC occurs after 159 stimulation of LX2 cells and still remains constant up to 60 minutes ( Fig. 4 C and Fig. S2-A). To further confirm that the pathway triggered by p17 involves the activation of Rho kinase we have pre-treated LX2 cells with a specific Rho kinase inhibitor (Y-27632) before exposure to p17. As illustrated in Figure 4 D, confirming the results obtained with the CXCR2 inhibitor SB-265610, the treatment of LX2 cells with the Rho inhibitor Y-27632 completely reversed the p17-mediated induction of collagen-I and a-SMA but not that of endothelin-1. All together these data demonstrated that p17 induction of collagen-I and a-SMA occurs via CXCR2 activation of Rho kinase and suggested that the p17-mediated up-regulation of endothelin-1 could take place with another additional mechanism which involves syndecan-2 activation. Since it has been well recognized that signal transducers and activators of transcription (STAT) proteins regulate the expression of HSC markers (i.e. endothelin-1 and collagen-I) [30][31][32][33] and the HIV matrix protein p17 signals even by activating the JaK/STAT pathway [18] we next investigated whether p17 induces phosphorylation of STAT proteins (i.e. STAT1 and STAT3) in LX2 cells. As illustrated in Figure 5 A and B, we found that exposure to p17 increases the level of phosphorylated STAT1 protein after 30 minutes as well as increases the level of phosphorylated STAT3 protein after 5 minutes of stimulation (

CXCR2 and syndecan-2 co-localize after stimulation of LX2 cells with p17
We next investigated the hypothesis that CXCR2 and syndecan-2 could interact as a result of their binding with p17. Results from immunoprecipitation experiments demonstrated the existence of a multiprotein complex between the proteins CXCR2, syndecan-2 and RACK1 which is clearly assembled already in basal conditions. Of interest, the interaction between these proteins increased after 5 min stimulation of LX2 cells with p17 ( Fig. 6 A and Fig. S3 A and B). These results were further confirmed by immunoprecipitation experiments with RACK-1 antibodies. Indeed, results from these experiments not only confirmed that both CXCR2 and syndecan-2 interact with RACK1 after exposure of LX2 cells to p17 but also highlighted that RACK-1 interacts with JAK-1 ( Fig. 6 B and Fig. S3 C-E).
The mutual relations between CXCR2 and syndecan-2 were further confirmed by analyzing LX2 cells at the confocal microscope. As illustrated in Figure 6 C, a different subcellular localization of CXCR2 between control and p17-treated cells has been found as resting LX2 cells exhibited both cell surface and cytoplasmic diffuse staining while p17-treated cells shown a ''punctated vesicular-like'' staining. Furthermore, results obtained making a double staining CXCR2/syndecan-2 demonstrated that these two proteins coexist in areas where the signals (red for syndecan-2 and green for CXCR2) are superimposed (Figure 6 D). All together these data demonstrated that the HIV matrix protein p17 promotes the formation of a multiprotein complex characterized by the proteins CXCR2, syndecan-2, RACK-1 and JAK-1.

Progression of liver fibrosis correlates with a reduction of CXCR2 protein expression in vivo
To investigate if the above described in vitro observation has a clinical readout we have examined the expression of syndecan-2 and CXCR2 in liver biopsies obtained from HCV monoinfected and HCV/HIV co-infected patients with various degree of liver fibrosis (see Table 1 for staging of fibrosis). As shown in Figure 7 A and B, the investigation of the p17 co-receptors reveals a trend indicating that while protein expression of syndecan-2 did not change significantly during the progression of liver fibrosis (from mild to severe irrespective of viral status), that of CXCR2, which occurs primarily in neutrophils and along the hepatic sinusoids (inset of Figure 7 C), was greatly reduced in liver biopsies obtained from patients with severe fibrosis (Figure 7 C), thus confirming the in vitro observations that during the activation of HSCs the expression of CXCR2 is down-regulated.

Discussion
In the present study we have provided evidence that the HIV matrix protein p17 exerts profibrogenic effects on HSCs. This conclusion emerges from the following results: (i) LX2 cells and human primary HSCs express CXCR2 and syndecan-2, the two main plasma membrane receptors for p17; (ii) exposure of LX2 and HSCs to p17 drives the production of ECM components, such as a1-collagen type 1; (iii) p17 induces the reorganization of intermediate filaments such as vimentin and increases the contractile force of LX2 by inducing endothelin-1; (iv) p17 immune-neutralization completely abrogates the effects of p17 on HSCs contraction and induction of collagen-I, a-SMA and endothelin-1.
Results from the present study are consistent with hypothesis that p17 hijacks two alternative signaling pathways in LX2 cells i.e. the CXCR2/Rho pathway and the Jak/STAT pathway. Rho is a small, monomeric guanosine triphosphate-binding protein from the Ras superfamily [34]. The Rho kinase regulates downstream mediators, such as myosin light chain (MLC), through phosphorylation. Phosphorylated MLC (pMLC2) binds to actin and stabilizes the stress fibers necessary for migration and contraction [35]. Several studies have demonstrated that the Rho pathway is activated by several regulatory cytokines during fibrogenesis (i.e. TGFb1 and PDGF) and is responsible for assembly of actin stress fibers, contractility and chemotaxis of HSCs [21]. Furthermore, inhibition of Rho kinase reduces MLC phosphorylation leading to a decreased activity of the promoter for myofibroblast marker a-SMA and to a reduced deposition of type I collagen [36][37][38]. Results presented in this study demonstrated that the treatment of LX2 cells with p17 results in a significant induction of MLC phosphorylation and that Y-27632, a selective inhibitor of Rho kinase, completely abrogates the profibrogenic effects exerted by p17 on these cells. Similarly, we found that while the pathway activated by CXCR2 was responsible for the p17 mediated induction of type I collagen and a-SMA, the inhibition of receptor and post-receptor checkpoints in the CXCR2 pathway, failed to modulate endothelin-1 gene expression. All together these results indicated that the CXCR2/Rho pathway regulates the The contractile force of LX2 was only in part due to Rho activation of MLC. Indeed, our results support the notion that p17 might activate the JaK/STAT pathway, which in turn induces the transcription/expression of endothelin-1. Consistent with this view we have detected a marked activation of STAT1 and STAT3 after exposure of LX2 cells to p17. Of note, when LX2 cells were preincubated with STAT-1 and STAT3 inhibitors the effects exerted by p17 on endothelin-1, collagen-I and a-SMA were reduced. These results indicate that regulation of these effector functions by p17 take place through the activation of various, but partially convergent, pathways. Moreover, we found that the pre-incuba-tion of LX2 cells with an anti-syndecan-2 antibody exerts the same regulatory effect of the STAT inhibitors. These results confirm previous observations that syndecans regulates the JaK/STAT signal transduction [39].
An important observation we have made in this study was that the exposure of both LX2 cells and primary human HSCs to p17 drives a robust reduction of CXCR2 expression while the expression of syndecan-2 remains unchanged. The negative regulation of CXCR2 expression that occurs in activated HSCs during their trans-differentiation is probably p17 independent. Indeed, it has been previously demonstrated that CXCR2 expression declines in a non-specific manner during LX2 transdifferentiation [40]. In particular, when LX2 cells were cultured in presence of 1% FBS (i.e. quiescent condition) they express much higher levels of CXCR2 compared to cells cultures with 10% FBS (i.e. activated condition) [40]. To provide translational information to this observation we have examined how the expression of p17 putative receptors changes with progression of liver fibrosis in HIV/HCV coinfected and HCV monoinfected patients. These immunohistochemical studies revealed that CXCR2 expression was non detectable in liver biopsies from patients with severe fibrosis while the expression of syndecan-2 and a-SMA remained stable, thus confirming our in vitro results that activation of HSCs, per se, causes the downregulation of the CXCR2 receptor.
Confirming a previous study [18], we have shown that exposure of LX2 to the HIV matrix protein p17 drives the formation of a multiprotein complex between the IL-8 receptor CXCR2, syndecan-2, RACK1 and JAK1. These results came from immunoprecipitation experiments and were further confirmed by confocal studies which provide evidence that CXCR2 colocalizes with syndecan-2 after 5 minutes exposure of LX2 cells to p17.
An important finding of the present study is the demonstration that the serum of an HIV positive patient who underwent a vaccination program [25] with a p17 protein contains anti-p17 antibodies which effectively neutralize the biological effects of the viral protein. The fact that a serum taken from the same HIV infected person before vaccination, but not a serum from an HIV negative healthy subject, was partially effective in reversing some effects of p17, is consistent with the presence of low title p17 neutralizing antibodies [18]. Since liver fibrosis is an important complication of long-term infection in HIV infected persons, present results suggest a possible clinical application of anti-p17 vaccination in the prevention of liver complication during HIV infection.
In conclusion, we have demonstrated that the HIV matrix protein p17 hijacks the CXCR2 and syndecan-2 signaling pathways and exerts profibrogenic effects in HSCs in vitro. Our results therefore, provide an additional explanation for liver fibrosis in HIV infected persons. Indeed, the presence of circulating p17 particles in HIV/HCV coinfected or in HIV monoinfected patients may not only be responsible for local activation and recruitment of inflammatory cells, but also for the direct activation of HSCs in the liver.  For studies focused on determining the impact of HIV p17 protein on the expression of collagen-I, a-SMA, CXCR2, syndecan-2 and endothelin-1, cells were serum starved for 48 hours prior to challenge with 0.1, 1 and 10 mg/ml p17 for 18 hours. At the end of the incubation cells were lysed for isolation of total RNA and proteins.

Materials and Methods
To investigate the effect of the serum of an HIV patient who is subjected to vaccination with a recombinant p17 peptide, serum starved LX2 cells were pre-incubated 2 h with the serum of the patient taken before and after the vaccination, diluted 1:100 in culture medium, followed by additional treatment with 2 mg/ml p17 for 18 hours. As a control, LX2 cells were also stimulated with a serum taken from an healthy donor. After stimulation cells were lysed for isolation of total RNA and proteins.
For studies focused on the impact of HIV p17 protein on the expression of collagen-I, a-SMA, CXCR2, syndecan-2 and endothelin-1, cells were serum starved for 24 hours prior to challenge with 0.1, 1 and 10 mg/ml p17 for 18 hours. At the end of the incubation cells were lysed for isolation of total RNA.

Vaccination protocol
The HIV matrix protein p17 was immune-neutralized by incubation with sera from an HIV infected patient enrolled in a Phase 1 study designed to investigate the safety and immunogenicity of recombinant p17 peptide in HIV. The therapeutic vaccination was performed using a 20 amino acids peptide, named AT20-KLH (SGGELDRWEKIRLRPGGKKK). The vaccination protocol n. MED-AT20-001 Eudract Number 2008-001465-29 had been approved by the Ethical committee of Regione Umbria (Italy) on June 25, 2010 authorization n. 1558/10. Authorization for collecting and using blood samples from HIV infected persons for ex vivo testing was also granted by the ethical committee of Regione Umbria (Italy) on July 22, 2010 (authorization number CEAS 1654/20). An informed written consent was obtained from each participant to the study. The presence of anti-p17 antibodies in the blood of these subjects had been verified by micro-Elisa as previously described [18].

Quantitative Real-Time PCR
Total RNA was extracted with 1 ml Trizol Reagent (Invitrogen) according to the manufacturer's instructions. One mg total RNA was reverse-transcribed using the enzyme Super Script II (Invitrogen) and quantitative Real time Polymerase Chain Reaction (qRT-PCR) was performed using primers indicated in Table 2. The PCR mixture was prepared by adding the following reagents: 50 ng cDNA, 0.2 mM of each primer and 12.5 ml of 2X SYBR Green qPCR master mix (Invitrogen) in a final volume of 25 ml. All reactions were performed in triplicate and the thermal cycling conditions were: 2 min at 95uC, followed by 40 cycles of 95uC for 20 s, 55uC for 20 s and 72uC for 30 s in iCycler iQ instrument (Biorad). The results of Real-Time PCR were normalized with the housekeeping gene HPRT and expressed as 2 2(DDCt) .

Gel contraction assay
Contractility of LX2 was evaluated using collagen gel lattices on 12-well culture plates. Briefly, wells were filled with 3 ml PBS containing 1% bovine serum albumin for 1 hour at 37uC, washed twice with PBS, and air-dried. Type 1 bovine collagen (3 mg/mL; Sigma) was adjusted according to the manufacturer's instructions. Wells were filled with 1 ml collagen solution and incubated for 1 hour at 37uC to allow gelation. LX2 cells were trypsinized, suspended in DMEM (1610 5 cells/mL) supplemented with 1% fetal bovine serum and antibiotics, and plated on the collagen gels (2 mL cell suspension/well). After incubation overnight to allow cell attachment, LX2 were serum starved over-night. Gels were then detached from the plates. LX2 were pre-incubated for 2 hours with the serum of an HIV patient taken before and after vaccination with a p17 peptide (diluted 1:100 in culture medium). After pre-incubation LX2 cells were stimulated for 18 hours with p17 (2 mg/ml). Culture medium containing 10% FBS was used as a positive control. The gels were photographed after 18 hours. Surface area of the collagen gels was measured using digital image analysis software.

Immunoprecipitation protocol
To study the p17 intracellular signaling LX2 cells were serum starved over-night and then stimulated with p17 (2 mg/ml) for 5, 15, 30 and 60 minutes. After the stimulation, cells were washed 3 times with ice-cold PBS and lysed with an insulin syringe in 500 ml E1A lysis buffer (250 mM NaCl, 50 mM Hepes pH 7.0, 0.1% NP40, 5 mM EDTA). Lysates were incubated 20 minutes in ice, clarified by centrifugation at 12000 rpm for 20 minutes at 4uC and quantified with Bradford reagent (PIERCE). 200 mg total proteins were pre-cleared on a rotating wheel for 1 h at 4uC using protein A Sepharose beads (Amersham Biosciences). Immunoprecipitation was performed overnight at 4uC with the followings antibodies: 1 mg anti-RACK-1 antibody (Santa Cruz -sc-17754), 1 mg anti-CXCR2 antibody (Santa Cruz -sc-7304) or 1 mg anti-IgG as a negative control antibody in the presence of 40 ml of protein A Sepharose (Amersham Biosciences). The resultant immunoprecipitates were washed five times with 1 ml of lysis buffer and resuspended in 40 ml of 2X SDS Sample buffer (100 mM Tris HCl pH 6.8, 5% b-mercaptoethanol, 4% SDS, 20% glycerol). Anti CXCR2 immunoprecipitates were used for western blotting using the antibodies syndecan-2, RACK-1 and CXCR2. Anti RACK-1 immunoprecipitates were used for western blotting using the antibodies syndecan-2, RACK-1, JAK-1 (Santa Cruz -sc-7228) and CXCR2.

Confocal Immunofluorescence
LX2 cells were plated in 2 chamber polystyrene vessels (1610 5 cell per chamber) previously coated with poly-l-lysyne (Sigma). Cells were serum starved overnight and then treated with p17 (2 mg/ml) for 5 minutes or 24 hours. After stimulation cells were washed in PBS and fixed with PFA 4% in PBS for 15 minutes. For CXCR2 and Syndecan double staining cells were incubated in a blocking solution containing BSA 6% in PBS. For vimentin and a-SMA single staining cells were incubated in PBS containing BSA 6% and triton 0,3%. Primary antibodies were incubated for 1 hour in PBS with the following dilutions: rat anti-human syndecan-2 (R&D systems) 1:40; mouse anti CXCR2 (Santa Cruz) 1:100; mouse anti vimentin (Dako) 1:50; mouse anti a-SMA (Novus Biologicals) 1: 100. Cells were washed in PBS and then incubated with secondary antibodies Alexa Fluor 488 goat anti mouse IgG and Alexa Fluor 546 goat anti rat IgG (Life technologies) (1:100 in PBS) for 30 minutes. After washing in PBS, slow fade gold antifade reagent was applied to slides and covered with cover slip.

Immunohistochemistry of human liver biopsies
Human liver tissues used in the present study were obtained from needle biopsy previously performed in HCV-HIV coinfected patients and HCV-positive-HIV-negative patients in order to asses severity and evolution of their hepatic disease. Informed written consent was obtained from 8 patients (see Table 1). Sections of 4 mm thick were performed by original formalin-fixed, paraffinembedded tissue blocks. Sections were immunostained for the following antibodies: aSMA (clone asm-1, Menarini, ready to use), CXCR2 (clone E-2, Santa Cruz, dilution 1:250) and syndecan-2 (clone 305507, R&D systems, dilution 1:50). The primary antibody was detected using a biotin-free polymeric-horseradish peroxidase (HRP)-linker antibody conjugate system (Bond Polymer Refine Detection, Leica) conducted with the Bond III automated immunostainer (Leica). The cases illustrated in Figure 6 came from patients 3 and 6 respectively (see Table 1).

Statistical analysis
All results are expressed as mean 6 standard error (SE). Comparisons of more than two groups were made with a one-way ANOVA with post-hoc Tukey's test. Differences were considered statistically significant when P was ,0.05.