Figure 1.
HIV-1 gp120 induces hyperpermeability of a lymphatic cell monolayer.
(A) Permeability, as determined by the translocation of FITC-conjugated dextran particles through an L-LEC monolayer in transwell chambers. Cells were incubated with M-gp120, T-gp120 or controls at various concentrations for 18 hours. Dextran particles were added, and fluorescence assessed after 5 minutes. Permeability was calculated based on the relative fluorescence of media in lower chambers of HIV-1 gp120-treated cells vs. the controls (“0”). Data indicate the mean ± SD of 3 independent experiments. (*** p<0.001). (B) Immunofluorescent analysis of CXCR4 expression in L-LECs. Scale bars = 20 µm. Representative images are shown.
Figure 2.
HIV-1 gp120 modulates the expression of fibronectin and Slit2 in L-LECs.
(A) Representative Western blot analysis of fibronectin (FN) expression in L-LEC cytoplasm and supernatant. Cells were serum-starved for 1 hour and incubated with indicated concentrations of HIV-1 gp120 for 18 hours before harvesting protein. GAPDH used as loading control. (B) Robo4 and FN expression in L-LECs by confocal microscopy. L-LECs were cultured in chamber slides and incubated with either HIV-1 gp120 (500 ng/ml), Slit2 (500 ng/ml) or a control for 15 minutes before fixing and staining cells. Red = Robo4; Green = FN; Blue = DAPI. Scale bars = 10 µm. (C) Robo4 immunoprecipitation of total and phosphorylated FN (p-Ser/Ther) by Western blot analysis in L-LECs. Cells were incubated with either HIV-1 gp120 (500 ng/ml) or a control for 15 minutes before protein from total cell lysates was collected for Robo4 immunoprecipitation. Membrane was stripped and reprobed for Robo4 expression as a loading control.
Figure 3.
The differential effects of HIV-1 gp120 concentrations on Slit2 expression in L-LECs.
Representative RT-PCR analysis (DNA gel) of Slit2 expression in L-LECs after incubation with designated concentrations of HIV-1 gp120 for 18 hours prior to performing RT-PCR. β-actin was amplified as an internal control.
Figure 4.
HIV-1 gp120 induces hyperpermeability in L-LECs through activation of α5β1 integrin.
(A) Representative Western blot analysis of phosphorylated α5β1 integrin in L-LECs after 2 hours of serum starvation and subsequent incubation for designated times with HIV-1 gp120 (500 ng/ml) or Slit2N (500 ng/ml). GAPDH used as loading control. (B) Phosphorylated α5β1 integrin and HIV-1 gp120 expression and co-localization in L-LECs by confocal microscopy. L-LECs were cultured in chamber slides and incubated with HIV-1 gp120 (500 ng/ml) for 15 minutes before fixing and staining cells. Scale bars = 10 µm. (C) Permeability through an L-LEC monolayer as previously described. L-LEC monolayers were pretreated with a neutralizing anti-integrin β1 antibody or a normal IgG control for 2 hours before incubating with M-gp120 or T-gp120 (both 500 ng/ml) for 18 hours. Data indicate the mean ± SD of 3 independent experiments. (**p<0.01; *** p<0.001).
Figure 5.
Slit2N inhibits HIV-1 gp120-induced lymphatic hyperpermeability by blocking the interaction between α5β1 integrin and Robo4.
(A) Robo4 immunoprecipitation of α5β1 integrin by Western blot analysis in L-LECs. Cells were incubated with either Slit2N (500 ng/ml), HIV-1 gp120 (500 ng/ml) or their respective controls (“−”) for 15 minutes before protein from total cell lysates was collected for Robo4 immunoprecipitation. Membranes were stripped and reprobed for Robo4 expression as a loading control. (B) Robo4 immunoprecipitation of α5β1 integrin by Western blot analysis in L-LECs. Cells were incubated with either Slit2N (500 ng/ml) or a control for 2 hours and then treated with HIV-1 gp120 (500 ng/ml) for times indicated before the protein from total cell lysates was collected for Robo4 immunoprecipitation. Membrane was stripped and reprobed for Robo4 expression as a loading control. (C) Permeability through an L-LEC monolayer as previously described. L-LEC monolayers were pretreated with Slit2N (500 ng/ml) or a control for 2 hours before incubating with M-gp120 or T-gp120 (both 500 ng/ml) for 18 hours. Data indicate the mean ± SD of 3 independent experiments. (**p<0.01; *** p<0.001 for treatment with Slit2N versus vehicle control).
Figure 6.
Slit2N attenuates HIV-1 virus-induced lymphatic hyperpermeability.
(A) Permeability through a dermal lymphatic endothelial cell (D-LEC) monolayer as previously described. D-LEC monolayers were pretreated with Slit2N (500 ng/ml) or a control for 2 hours before incubating with M-gp120 (500 ng/ml) for 18 hours. (B) Permeability through an L-LEC monolayer as previously described. L-LEC monolayers were pretreated with Slit2N (500 ng/ml) or a control for 1 hour, followed by incubation with HIV-1 virions (4.0×106 TCID 50/ml) or gp120 (500 ng/ml) for 5 hours. The viral load corresponds to a gp120 concentration of 425 ng/ml. For (A) and (B), data indicate the mean ± SD of 3 independent experiments. (*p<0.05; *** p<0.001 for treatment with Slit2N versus vehicle control).
Figure 7.
Slit2N and Robo4 influence gp120-induced hyperpermeability in L-LECs by modulating c-Src kinase activation and signaling.
(A) Representative Western blot analysis of phosphorylated c-Src in L-LECs after preincubation with either Slit2N (500 ng/ml) or a control for 2 hours before treatment with HIV-1 gp120 (500 ng/ml) for times indicated. GAPDH used as loading control. (B) Robo4 immunoprecipitation of Myc-tagged Slit2 by Western blot analysis in 293 cells. 293 s were co-transfected with a Robo4 expression plasmid and either a Myc-tagged Slit2 expression plasmid or a vector control. Cells were incubated for 48 hours before the protein from total cell lysates was collected for Robo4 immunoprecipitation. (C) Representative Western blot analysis of phosphorylated c-Src in L-LECs after transfection with either Robo4-specific siRNAs or a control siRNA for 48 hours before treatment with HIV-1 gp120 (500 ng/ml) for times indicated. GAPDH used as loading control. (D) Permeability through an L-LEC monolayer as previously described. An L-LEC monolayer was pretreated with a Src kinase inhibitor (2 µM) or DMSO for 2 hours before incubating with HIV-1 gp120 (500 ng/ml) or a control for 18 hours. Data indicate the mean ± SD of 3 independent experiments. (**p<0.01 for treatment with the Src kinase inhibitor versus DMSO control).
Figure 8.
Robo4 expression levels influence HIV-1 gp120-induced lymphatic hyperpermeability.
(A) Representative Western blot analysis of Robo4 expression in L-LECs, 24 hours after transfection with a mixture of Robo4-specific siRNAs or a negative control siRNA. GAPDH used as loading control. (B) Permeability through an L-LEC monolayer as previously described. L-LECs transfected with control siRNAs or Robo4-specific siRNAs were seeded into the upper chamber of transwell plates and incubated with Slit2N (500 ng/ml), HIV-1 gp120 (500 ng/ml), or a control, for 18 hours. Data indicate the mean ± SD of 3 independent experiments. (**p<0.01 for treatment with Slit2N or HIV-1 gp120 versus negative control of the L-LECs transfected with control siRNAs).
Figure 9.
The fibronectin domains of Robo4 are critical for gp120-induced hyperpermeability of L-LEC monolayers.
(A) Representative Western blot analysis of phosphorylated c-Src in L-LECs pretreated with Slit2N (500 ng/ml) or a control for 1 hour, then stimulated with fibronectin (FN) [1 µg/ml (1×); 10 µg/ml (10×)] for 20 minutes as indicated. GAPDH used as loading control. (B) Permeability through an L-LEC monolayer as previously described. L-LECs were transiently transfected with expression plasmids encoding wild-type Robo4 (WT), mutant Robo4 (MT), or a vector control (V). After 48 hours, cells were plated for the permeability assay per manufacturer's instructions. L-LEC monolayers were incubated overnight with 500 ng/ml HIV-1 gp120 or a control. Data are represented as the percentage increase in permeability of each cell type monolayer incubated with gp120 vs. control. Data indicate the mean ± SD of 3 independent experiments. (*p<0.05, ***p<0.001). (C) Representative Western blot analysis of phosphorylated c-Src and ERK1/2 in L-LECs transiently transfected with expression plasmids encoding wild-type Robo4 (WT), mutant Robo4 (MT), or a vector control (V). After 48 hours, the cells were serum-starved for 2 hours and stimulated with HIV-1gp120 (500 ng/ml) or a control for 15 minutes as indicated. GAPDH used as loading control.
Figure 10.
Hypothetical model for the interactions of HIV-1 gp120, FN, α5β1 integrin, Robo4 and Slit2, and their effect on the permeability of lymphatic endothelium.
Robo4 and α5β1 integrin are expressed on the surface of lymphatic endothelial cells. Under physiologic conditions, endogenous Slit2 interacts with Robo4 via its Ig domains, and inhibits c-Src signaling. FN also interacts with Robo4 (via its FN type III domains), and binds to α5β1 integrin on the endothelial cell surface. This contributes further to the dynamic regulation of Robo4 signaling by Slit2 and FN, to maintain the integrity of the lymphatic endothelial barrier. Upon HIV infection, gp120 significantly elevates FN levels and complexes with FN. FN then activates α5β1 integrin, which results in enhanced intracellular signaling through α5β1 integrin, a stronger interaction between FN and Robo4, and disruption of Robo4 signaling. These changes activate the c-Src signaling pathway and induce hyperpermeability of the lymphatic endothelial barrier. Exogenous Slit2 may protect the lymphatic channels from HIV-induced vasculopathy by interacting with Robo4 to restore Robo4 signaling, block the c-Src pathway, and inhibit HIV-induced lymphatic hyperpermeability.