The HIV matrix protein p17 promotes the activation of human hepatic stellate cells through interactions with CXCR2 and Syndecan-2

Abstract

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.

Figure 1. Effect of HIV p17 on stellate cell activation.

(A–B) Serum starved LX2 cells were stimulated with 0.1, 1 and 10 µg/ml p17 for 18 hours. (A) Immunoblot of syndecan-2 (Syn-2), CXCR2, αSMA, pro-col1α1, endothelin-1 (ET-1) and Tubulin. (B) qRT-PCR of syn-2, CXCR2, αSMA, collagen-I and ET-1. Data represent the mean of 4 experiments. Values are normalized relative to HPRT mRNA and are expressed relative to those of untreated cells, which were arbitrarily set to 1. *P<0.05 versus not treated cells. (C) Serum starved human primary hepatic stellate cells were stimulated with 0.1, 1 and 10 µg/ml p17 for 18 hours. At the end of stimulations relative mRNA expression of syn-2, CXCR2, αSMA, collagen-I and ET-1 was evaluated by Real-Time PCR. Data represent the mean of 4 experiments. Values are normalized relative to HPRT mRNA and are expressed relative to those of untreated cells, which were arbitrarily set to 1. *P<0.05 versus not treated cells.

Figure 2. Effect of p17 on expression and distribution of cytoskeleton fibers in LX2 cells.

Confocal immunofluorescence of vimentin (A) and α-SMA (B) in LX2 cells left not treated (a) or stimulated with 2 µg/ml p17 for 24 hours (b–c).

Figure 3. Blocking antibodies against p17 inhibit LX2 activation.

(A–C) Serum starved LX2 cells were pre-incubated 2 h with the serum taken from and healthy donor (HD) or with the serum of an HIV patient taken before (preV) and after (postV) the vaccination with a recombinant p17 peptide, diluted 1∶100 in culture medium, followed by additional treatment with 2 µg/ml p17 for 18 hours. qRT-PCR of Col1α1 (A), αSMA (B) and ET-1. Values are normalized relative to HPRT mRNA and are expressed relative to those of untreated cells, which were arbitrarily set to 1. *P<0.05 versus not treated cells. #P<0.05 versus p17 stimulated cells. (D) Hydrated collagen lattices, photographed 18 hours after stimulation of LX2 with 10 µg/ml p17 alone (a) or in the presence of the serum of an HIV patient taken before (b) and after (c) the vaccination with a recombinant p17 peptide, diluted 1∶100 in culture medium. Contraction of LX2 cells in response to 10% FBS was also evaluated after 6 h incubation. (E) Mean percent gel surface area of collagen gels compared with initial surface area. Data are mean ± SE of 3 experiments. *P<0.05 versus control cells. #P<0.05 versus p17 stimulated cells. (nt: not treated cells).

Figure 4. p17 activates CXCR2/Rho pathway on LX2 cells.

(A) Serum starved LX2 cells were pre-incubated for 2 h with 100 nM SB265610 followed by additional treatment with 2 µg/ml p17 for 18 hours. qRT-PCR of collagen-I, αSMA and ET-1. Values are normalized relative to HPRT mRNA and are expressed relative to those of untreated cells, which were arbitrarily set to 1. *P<0.05 versus not treated cells. #P<0.05 versus p17 stimulated cells. (B) Serum starved LX2 cells were pre-incubated for 2 h with 4 µg anti-syndecan-2 antibody followed by additional treatment with 2 µg/ml p17 for 18 hours. qRT-PCR of collagen-I, αSMA and ET-1. Values are normalized relative to HPRT mRNA and are expressed relative to those of untreated cells, which were arbitrarily set to 1. *P<0.05 versus not treated cells. #P<0.05 versus p17 stimulated cells. (C) Time course of phosphoMLC and total MLC expression in LX-2 cells treated with 2 µg/ml of p17. (C) LX2 cells were pre-incubated for 2 h with indicated concentrations of Rho kinase inhibitor Y-27632 prior to stimulation with 2 µg/ml p17 for 18 hour. qRT-PCR of collagen-I, αSMA and ET-1. Values are normalized relative to HPRT mRNA and are expressed relative to those of untreated cells, which were arbitrarily set to 1. *P<0.05 versus not treated cells. #P<0.05 versus p17 stimulated cells.

Figure 5. HIV matrix protein p17 activates JAK/STAT pathway on LX2 cells.

(A) Time course of phosphoSTAT1 and total STAT1 expression in LX-2 cells treated with 2 µg/ml of p17. (B) Time course of phosphoSTAT3 and total STAT3 expression in LX-2 cells treated with 2 µg/ml of p17. (C–E) LX2 cells were pre-incubated for 2 h with the STAT1 inhibitor Fludarabine (5 µM) or with the STAT3 inhibitor 5,15 DPP (5 µM) prior to stimulation with 2 µg/ml p17 for 18 hour. qRT-PCR of collagen-I (C), αSMA (D) and ET-1 (E). Values are normalized relative to HPRT mRNA and are expressed relative to those of untreated cells, which were arbitrarily set to 1. *P<0.05 versus not treated cells. #P<0.05 versus p17 stimulated cells.

Figure 6. p17 drives the formation of a multiprotein complex containing CXCR-2, syn-2, RACK-1 and JAK-1 on LX2 cells.

LX2 cells were serum starved and stimulated for 5, 15, 30 and 60 minutes with 2 µg/ml of p17. (A) Co-immuneprecipitation of CXCR2 with syndecan-2 and RACK-1. (B) Co-immuneprecipitation of RACK1 with syndecan-2, CXCR2 and JAK1. (C) Confocal immunofluorescence of LX2 cells left untreated (a) or exposed to p17 for 5 minutes and stained with anti-CXCR2 antibody. (D) Immunofluorescence analysis of syndecan-2 and CXCR2 localization on LX2 cells left untreated (NT) or stimulated 5 minutes with p17. Cells were stained with the indicated antibodies. Red: syn-2. Green: CXCR2. The red and green signals were electronically merged for co-localization analysis.

Figure 7. Progression of liver fibrosis is associated with a down-regulation of CXCR2 receptor.

Liver biopsies from HIV/HCV coinfected and HCV monoinfected patients were serially sectioned and stained with the following antibodies: αSMA, syndecan-2 and CXCR2. (A) Patient n. 3 HCV positive. Diffusely and intensely sinusoidal walls staining with αSMA (Magnification 200x). Patient n.6 HIV/HCV positive. Focally and weakly sinusoidal walls staining with αSMA (Magnification 200x). (B) Patient n. 3 HCV positive and patient n. 6 HIV/HCV positive. In both cases strongly cytoplasmatic staining of stellate cells with syndecan-2 (Magnification 200x) (C) Patient n. 3 HCV positive. Focally and weakly sinusoidal walls staining with CXCR2 (Magnification 200x; inset 400x). Patient n.5 HIV/HCV positive. Absence of staining with CXCR2 (Magnification 200x)