Alterations in lymph node architecture occur with HIV infection and contribute to immunological derangements. We previously showed that matrix fibronectin stabilized HIV and increased HIV infection of PBL. We showed increased fibronectin deposition in lymph nodes of HIV-infected patients. However, we did not detect a difference in fibronectin synthesis between uninfected and infected PBL. Therefore, we hypothesized that interactions of HIV-infected cells with fibroblasts resulted in increased fibronectin deposition. We detected increased fibronectin deposition by immunofluorescence on fibroblasts cocultured with HIV-infected PBL. We also found a 6-fold increase in fibronectin mRNA levels in fibroblasts cocultured with HIV-infected PBL by real-time PCR. Furthermore, when HIV-infected PBL were added to reporter fibroblasts stably transfected with a fibronectin promoter, we found a 1.5- to 2-fold increase in promoter activity. Since conditioned medium from HIV-infected PBL also increased fibronectin promoter activity, we hypothesized that a soluble factor such as TGFβ was responsible for increased fibronectin secretion. Pretreatment of supernatant from HIV-infected PBL with a neutralizing Ab to TGFβ1 abrogated the increased fibronectin promoter activity. We confirmed that HIV-infected PBL produced increased TGFβ1 by ELISA. Using Mv1Lu reporter cells, we found a 2- to 3-fold increase in biologically active TGFβ in supernatants of HIV-infected PBL. Finally, we determined that HIV infection did not change the percentage of active TGFβ. Our data suggest that HIV-infected lymphocytes indirectly contribute to lymph node remodeling by secretion of TGFβ1, which increases fibronectin synthesis by fibroblasts.

Interactions of HIV with extracellular matrix influence its pathogenicity. HIV may encounter matrix proteins during initial transmission, as in the setting of genital ulceration, in which a provisional wound-healing matrix is deposited. In addition, during HIV infection extensive remodeling of lymph nodes occurs, with fibrotic changes and loss of the normal lymph node architecture. The destruction of normal lymph node architecture greatly decreases the ability to mount an appropriate immune response to other pathogens. This results in increased viral replication, increased susceptibility to opportunistic infections, and an accelerated clinical course. Changes in the extracellular matrix composition of lymph nodes, where the majority of virions are produced, will alter the context in which HIV interacts with target cells. Furthermore, the degree of fibrosis in the lymph node at the initiation of highly active antiretroviral therapy (HAART)3 may limit the effective response (1). Thus, information regarding the mechanisms of lymph node remodeling may increase the efficacy of HAART.

Our previous work showed that extracellular fibronectin significantly enhanced the interaction of HIV with lymphocytes (2, 3). We previously reported that HIV virions adhered efficiently to superfibronectin, an in vitro model for matrix fibronectin (3). Superfibronectin-bound virions had enhanced infectivity and maintained their infectivity significantly longer than unbound virions (2). Additionally, proteolytic fragments of fibronectin containing the III1-C region also enhanced HIV infection. We also showed increased fibronectin deposition within lymph nodes of HIV-infected patients. However, we did not detect an increase in fibronectin synthesis by HIV-infected lymphocytes. Thus, the cause of increased fibronectin synthesis in HIV infection is unknown. We hypothesized that HIV-infected lymphocytes promote fibronectin deposition by interstitial fibroblasts through secretion of profibrotic cytokines.

PBMC were isolated from the blood of healthy HIV-seronegative volunteers by Ficoll-Hypaque density gradient centrifugation and depleted of monocytes by several rounds of adherence to tissue culture plastic to obtain PBL. CD4+ or CD8+ T lymphocytes were isolated using Ab-coated magnetic beads (Miltenyi Biotec, Auburn, CA) and grown in RPMI 1640 medium (Cellgro, Herndon, VA) supplemented with 10% FCS (Sigma-Aldrich, St. Louis, MO), penicillin-streptomycin (Life Technologies, Gaithersburg, MD), and 100 U/ml IL-2 (Proleukin; Chiron, Emeryville, CA). Cells were verified as >95% CD4+ or CD8+ by flow cytometric analysis.

HIVIIIB virus was purified by ultracentrifugation from the virus-producing cell line H9/HIVIIIB (AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Rockville, MD, and contributed by Dr. R. Gallo) as previously described (4) or purchased from Advance Biotechnology (Rockville, MD). For HIV infection, cells (106) were stimulated with PHA (5 μg/ml) for 2 days and then incubated with virus (equivalent to 4 ng of p24 Ag in a 100-μl volume) for 2 h at 37°C. Cells were subsequently washed and resuspended in 2 ml of medium. Infection was monitored by measurement of p24 Ag in cell-free supernatant (Coulter HIV-1 P24 Ag assay kit; Coulter, Seattle, WA). As controls, cells were stimulated with PHA alone. Cells were used 7 days after infection, unless otherwise indicated. Viability was similar in uninfected and HIV-infected cells at days 7 and 14 (day 7: 91 ± 0.8% vs 88 ± 2%; day 14: 67 ± 6% vs 63 ±12%).

The fibroblast cell line IMR90 was obtained from American Type Culture Collection (Manassas, VA). Fibroblasts 3T3 cells stably transfected with a fibronectin promoter upstream of a luciferase reporter gene were a gift from Dr. J. Roman (Emory University, Atlanta, GA) (5). Mink lung epithelial cells (Mv1Lu) stably transfected with a portion of the plasminogen activator inhibitor 1 (PAI-1) promoter upstream of a luciferase reporter gene were a gift from J. Munger (New York University School of Medicine, New York, NY) (6). Cells were grown in DMEM supplemented with 10% FCS and penicillin/streptomycin. Monoclonal anti-TGFβ1 Ab (clone1D11) was used for neutralization of TGFβ1 bioactivity (R&D Systems, Minneapolis, MN). Mouse anti-human fibronectin Ab (HFN7.1), developed by R. J. Klebe, was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the National Institute of Child Health and Human Development and maintained by the University of Iowa, Department of Biological Sciences (Iowa City, IA). FITC-conjugated anti-mouse IgG was obtained from Vector Laboratories (Burlingame, CA). Full-length recombinant HIVIIIB gp120 and CXCR4 mAb (12G5) were obtained from the AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health. Polyclonal Ab to gp120 was obtained from Immunodiagnostics, (Bedford, MA). Polyclonal Abs to Smad2/3 and phospho-Smad2/3 (Ser433/435) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

HIV-infected or uninfected PBL (106 cells) were added to a confluent monolayer of IMR90 fibroblasts in Nunc chamber well slides and cocultured for 24 h in DMEM with 1% FCS. After washing with PBS, fibroblasts were fixed with 2.5% paraformaldehyde and incubated with anti-fibronectin Ab HFN7.1 (1:100) for 2 h at room temperature. Unbound Ab was removed by washing with PBS followed by incubation with FITC-conjugated anti-mouse IgG (Vector Laboratories) for 1 h. To obtain baseline, IMR90 cells alone were used. Three independent experiments were performed. Quantification was performed with a MetaMorph imaging program as follows. Five independent images per condition were captured with Adobe Photoshop Version 6 and then imported into the MetaMorph imaging program. Lower threshold level was set based on background. The number of pixels above the threshold level was determined for each field. Fibronectin deposition index was calculated as mean number of pixels per experimental condition over mean number of pixels of IMR alone. Statistical significance was determined with paired Student t tests.

HIV-infected PBL or CD4 lymphocytes were added to a confluent monolayer of IMR90 fibroblasts (3:1 ratio) and cocultured for 24 h. As a control, uninfected but stimulated lymphocytes were cocultured with IMR90 cells. In some experiments, cell-free supernatant from HIV-infected and uninfected cells was added to IMR90 cells. After the 24-h incubation, cells were washed with PBS and total cellular RNA was extracted by the SV Total RNA Isolation System (Promega, Madison, WI). RNA concentrations were determined by spectrophotometric analysis of the samples at 260 nm. As additional controls, RNA was isolated from lymphocytes or IMR90 cells alone.

Total RNA was reverse transcribed using the reverse transcription system Improm II (Promega). In brief, RNA (1 μg) and random primers (0.5 μg) were combined in 5 μl of nuclease-free water and denatured at 70°C. Reverse transcription was conducted at 40°C for 60 min by adding 200 U of reverse transcriptase and 1 U of recombinant RNasin RNase inhibitor (Promega). The samples were then heated for 15 min at 70°C to terminate the reverse transcriptase.

PCR primers used to amplify fibronectin, collagen I and collagen IV, and TGFβ (Qiagen, Valencia, CA) are shown in Table I. To control for sample loading, all PCR samples were amplified using the Quantum RNA 18S Internal Standards kit (Ambion, Austin TX). For light cycler reactions, a master mix of the following reaction components was prepared to the indicated end concentration: 6.4 μl of water, 1.2 μl of MgCl2 (4 mM), 0.2 μl of fibronectin sense primer (0.4 μM), 0.2 μl of fibronectin antisense primer (0.4 μM), and 1.0 μl of LightCycler Fast Start DNA Master SYBR Green I (Roche, Indianapolis, IN). An aliquot of 9 μl of LightCycler master mix was added to the LightCycler glass capillaries and 1 μl of reverse transcriptase product was added as PCR template. The capillaries were centrifuged and placed into the LightCycler rotor. The following LightCycler protocol was used: denaturation program (95°C for 10 min), a three segment amplification program repeated 40 times, melting curve program of 60–90°C, and finally a cooling down program to 4°C. The standard curve for quantifying the mRNA copy number was established by using the known copy number of fibronectin plasmid ranging from 1012 to 105 copies. Serially diluted standard DNA template was PCR amplified and a standard curve was generated by using the threshold cycle of templates in the known number of copies. Product purity was verified by melting curve analysis after amplification. mRNA copy numbers of the samples were calculated using LightCycler Data Analysis Software Version 3.5.3 (Roche).

Table I.

Primer sequences

PrimersSequenceProduct size (bp)
Fibronectin Sense: CAAGTATGAGAAGCCTGGGTCT 213 
 Antisense: TGAAGATTGGGGTGTGGAAG  
Collagen I Sense: GCTTCACCTACAGCGTCAC 201 
 Antisense: TGGGATGGAGGGAGTTTACA  
Collagen IV Sense: AGGCCTAGTGGTCCGAATCT 157 
 Antisense: CCTGGCTTGAAAAACAGCTC  
TGFβ1 Sense: GGGACTATCCACCTGCAAGA 239 
 Antisense: CCTCCTTGGCGTAGTAGTCG  
PrimersSequenceProduct size (bp)
Fibronectin Sense: CAAGTATGAGAAGCCTGGGTCT 213 
 Antisense: TGAAGATTGGGGTGTGGAAG  
Collagen I Sense: GCTTCACCTACAGCGTCAC 201 
 Antisense: TGGGATGGAGGGAGTTTACA  
Collagen IV Sense: AGGCCTAGTGGTCCGAATCT 157 
 Antisense: CCTGGCTTGAAAAACAGCTC  
TGFβ1 Sense: GGGACTATCCACCTGCAAGA 239 
 Antisense: CCTCCTTGGCGTAGTAGTCG  

PCR for collagen I and collagen IV were conducted in 50-μl reaction mixtures containing PCR buffer, 2 mM MgCl2, 0.2 mM dNTPs, 1 μM of each primer, and 1.25 U of Taq bead hot start DNA polymerase (Promega). The reaction mixture was incubated at 94°C for 2 min and subjected to 25 cycles consisting of 1 min at 94°C, 1 min at 60°C, 2 min at 72°C, followed by a single cycle of 5 min at 72°C. PCR for TGFβ was performed as above except for an annealing temperature of 55°C. The PCR-amplified products were analyzed by 1% agarose/Tris-acetate gel electrophoresis. Densitometry was performed with ImageJ Version 1.3 (http://rsb.info.nih.gov/ij).

IMR90 cells were cocultured with HIV-infected or uninfected PBL as above and lysed in buffer containing 50 mM Tri-HCl (pH 7.4), 150 mM NaCl, 0.25% sodium deoxycholate, 1% IGEPAL CA-630, 1 mM EGTA, 1 mM PMSF, 1 mM sodium vanadate, 1 mM sodium fluoride, 1 mg/ml each of aprotonin, leupeptin, and pepstatin. Equal amounts of protein were separated by SDS-PAGE under reducing conditions, transferred to Immobilon, and blocked for 1 h. Blots were incubated with anti-Smad Ab for 1 h, followed by peroxidase-conjugated secondary Ab for 1 h and then developed with ECL (Amersham, Arlington Heights, IL). As controls, blots were stripped, then reprobed with Ab to total Smad2/3.

Cell-free supernatants (100 μl) from HIV-infected or uninfected cells (day 7 or 14) were added to reporter 3T3 cells (105/ml) for 16 h. At the end of the incubation, the cells were washed and lysed using the Luciferase Assay System. In some experiments, supernatants were pretreated with anti-TGFβ1-neutralizing Ab or, as controls, anti-IL-1 Ab (R&D Systems) or control IgG Ab (Vector Laboratories) for 15 min before incubation with 3T3 cells. As an additional control, varying concentrations of rTGFβ1 (R&D Systems) was added to 3T3 cells and luciferase activity was measured. Data are reported as the mean luciferase activity of triplicate wells ± SD.

ELISA.

At day 2, 7, or 14 after HIV infection, PBL, CD4, or CD8 cells were resuspended in serum-free medium for 24 h. Cell-free supernatant was collected and total TGFβ1 was quantified by a Quantikine human TGFβ1 ELISA kit (R&D Systems) per the manufacturer’s direction. The concentration of TGFβ1 in the samples was determined by interpolation from the standard curve generated with rTGFβ1. Data are reported as the mean level of triplicate samples ± SD.

Luciferase assay for TGFβ1 bioactivity.

Mv1Lu reporter cells (105/ml) were grown overnight in a 96-well black plate (Costar, Corning, NY) at 37°C and incubated with cells or cell-free supernatant (100 μl) from HIV-infected or uninfected PBL, CD4, or CD8 cells at 37°C overnight. Cells were washed, lysed, and luciferase activity was measured using the Luciferase Assay System (Promega). In some experiments, the supernatant was heat activated at 90°C for 3 min before addition to Mv1Lu to measure total TGFβ1. To verify specificity, supernatants were preincubated with anti-TGFβ1 Ab (0.5 μg/ml) at room temperature for 1 h. Data are reported as the mean luciferase activity of triplicate wells ± SD.

At day 14 after HIV infection, cell-free supernatant from PBL, CD4, or CD8 cells was collected and analyzed using the Beadlyte Human 22-Plex Cytokine Detection System (Upstate Biotechnology, Lake Placid, NY). The lyophilized mixed standard was resuspended in blank cell culture medium and serially diluted. Samples or standards were incubated with the 22-Plex cytokine capture bead suspension array in a 96-well filter plate for 2 h at room temperature. The beads were washed and biotinylated reporter 22-plexantibodies were added for 1.5 h. Streptavidin-PE was then added to each well. After a 30-min incubation, stop solution was added and the beads were washed and resuspended in assay buffer. The median fluorescence intensity of 50 beads per cytokine was read using a Luminex 100 Instrument (Luminex, Austin, TX). Concentrations were interpolated from standard curves. Samples were run in duplicate.

We previously showed that fibronectin deposition was increased in lymph nodes during HIV infection but did not find a difference in fibronectin synthesis by HIV-infected lymphocytes (2). Therefore, we hypothesized that interactions of HIV-infected lymphocytes with fibroblasts would result in increased fibronectin synthesis by fibroblasts. To test this, we cocultured HIV-infected lymphocytes with fibroblasts and examined fibronectin deposition by immunofluorescence. We found increased fibronectin deposition in IMR90 fibroblasts cocultured with HIV-infected PBL for 24 h (Fig. 1,A). We quantitated the differences in fibronectin deposition using MetaMorph Image analysis software. We found a 20% increase in fibronectin deposition in IMR90 cells cocultured with HIV-infected PBL compared with uninfected PBL (p = 0.008; Fig. 1 B).

FIGURE 1.

Increased fibronectin deposition by fibroblasts cocultured with HIV-infected PBL. A, IMR90 fibroblasts were cocultured with HIV-infected (left panel) or uninfected (right panel) PBL for 24 h and fixed with 2.5% paraformaldehyde followed by incubation with anti-fibronectin Ab and FITC-conjugated secondary Ab. B, Quantitative analysis of fibronectin deposition. Fibronectin deposition index was defined as the mean number of pixels above threshold per experimental condition over the mean number of pixels of IMR alone.

FIGURE 1.

Increased fibronectin deposition by fibroblasts cocultured with HIV-infected PBL. A, IMR90 fibroblasts were cocultured with HIV-infected (left panel) or uninfected (right panel) PBL for 24 h and fixed with 2.5% paraformaldehyde followed by incubation with anti-fibronectin Ab and FITC-conjugated secondary Ab. B, Quantitative analysis of fibronectin deposition. Fibronectin deposition index was defined as the mean number of pixels above threshold per experimental condition over the mean number of pixels of IMR alone.

Close modal

To determine whether the increased fibronectin deposition was due to increased fibronectin RNA, we extracted RNA from cocultured fibroblasts and quantitated the fibronectin mRNA expression by real-time PCR. We observed a 6-fold increase in fibronectin mRNA copy number in fibroblasts cocultured with HIV-infected PBL compared with uninfected PBL (Fig. 2, A and B). There was also a 5-fold increase in fibronectin mRNA copy number when IMR90 cells were incubated with infected supernatant compared with uninfected supernatant. There was no significant difference in fibronectin mRNA expression between uninfected and infected PBL alone.

FIGURE 2.

Matrix RNA expression by fibroblasts cocultured with HIV-infected PBL. A, Fibronectin mRNA copy number from fibroblasts cocultured with HIV-infected PBL by real-time PCR. B and C, Amplified products of fibronectin (213 bp; B, top left panel), collagen I (201 bp), and collagen IV (157 bp) (C, top right panel) were separated on a 1% agarose gel and detected by ethidium bromide staining. As a control for loading, samples were amplified with primers for 18S ribosomal RNA (bottom panels).

FIGURE 2.

Matrix RNA expression by fibroblasts cocultured with HIV-infected PBL. A, Fibronectin mRNA copy number from fibroblasts cocultured with HIV-infected PBL by real-time PCR. B and C, Amplified products of fibronectin (213 bp; B, top left panel), collagen I (201 bp), and collagen IV (157 bp) (C, top right panel) were separated on a 1% agarose gel and detected by ethidium bromide staining. As a control for loading, samples were amplified with primers for 18S ribosomal RNA (bottom panels).

Close modal

To determine whether coculture of HIV-infected lymphocytes with fibroblasts induced increased RNA expression of other matrix proteins, we examined collagen I and collagen IV RNA expression by RT-PCR. We observed a 2.5-fold increase in collagen I mRNA expression by IMR90 when cocultured with HIV-infected PBL compared with uninfected PBL (Fig. 2 C). However, there was no difference in collagen IV expression when IMR90 was cocultured with either uninfected or infected PBL.

To extend our findings, we examined the effect of HIV-infected lymphocytes on fibronectin promoter activity. We cocultured HIV-infected PBL, CD4, or CD8 with 3T3 fibroblasts containing a luciferase reporter gene driven by a fibronectin promoter. We found up to a 1.6-fold increase in luciferase activity when 3T3 cells were cocultured with HIV-infected lymphocytes (Fig. 3,A). To determine whether cell contact was necessary for the observed increase or whether a soluble factor was involved, we incubated 3T3 cells with cell-free supernatants from infected PBL (days 7 and 14). We again observed up to a 2.1-fold increase in promoter activity, which correlated with progressive infection (Fig. 3,B). Interestingly, we also found increased promoter activity when cells were incubated with HIV-treated CD8 cells or their supernatants. Since our data suggested that a soluble factor was involved, we asked whether TGFβ1, a profibrotic cytokine, was responsible for the increased fibronectin promoter activity. First, addition of rTGFβ1 increased fibronectin promoter activity (Fig. 3,C). Second, when the supernatant was treated with anti-TGFβ1 Ab, the fibronectin promoter activity was reduced to basal level (Fig. 3,B), suggesting that TGFβ1 secretion was the primary mediator of the increased fibronectin promoter activity. In contrast, pretreatment with IgG or neutralizing Ab to IL-1 did not decrease promoter activity (Fig. 3 B).

FIGURE 3.

Increased fibronectin promoter activity in fibroblasts cocultured with HIV-infected lymphocytes. A, Uninfected or HIV-infected PBL, CD4, and CD8 cells were cocultured with 3T3 reporter cells overnight. Cells were then lysed and luciferase activity was measured. Data are presented as average luciferase activity of triplicate samples ± SD. B, Cell-free supernatant from uninfected and HIV-infected PBL, CD4, and CD8 was incubated overnight with 3T3 cells reporter cells. In some cases, supernatant was preincubated with anti-TGFβ1 Ab (0.5 μg/ml), anti-IL-1 Ab, or control IgG. Cells were washed and lysed and luciferase activity was measured. Data are presented as average luciferase activity of triplicate samples ± SD. C, rTGFβ was added to 3T3 reporter cells overnight. Cells were washed and lysed and luciferase activity was measured. Data are presented as average luciferase activity of triplicate samples ± SD.

FIGURE 3.

Increased fibronectin promoter activity in fibroblasts cocultured with HIV-infected lymphocytes. A, Uninfected or HIV-infected PBL, CD4, and CD8 cells were cocultured with 3T3 reporter cells overnight. Cells were then lysed and luciferase activity was measured. Data are presented as average luciferase activity of triplicate samples ± SD. B, Cell-free supernatant from uninfected and HIV-infected PBL, CD4, and CD8 was incubated overnight with 3T3 cells reporter cells. In some cases, supernatant was preincubated with anti-TGFβ1 Ab (0.5 μg/ml), anti-IL-1 Ab, or control IgG. Cells were washed and lysed and luciferase activity was measured. Data are presented as average luciferase activity of triplicate samples ± SD. C, rTGFβ was added to 3T3 reporter cells overnight. Cells were washed and lysed and luciferase activity was measured. Data are presented as average luciferase activity of triplicate samples ± SD.

Close modal

Since our results suggested that TGFβ1 secretion was responsible for increased fibronectin synthesis, we asked whether there was increased TGFβ1 production by HIV-infected PBL compared with uninfected PBL. PBL and CD4 cells were stimulated with PHA for 48 h and infected with HIV. At indicated time points, cell-free supernatant was collected and total TGFβ1 was measured by ELISA. HIV infection progressively increased total TGFβ1 secretion from PBL compared with the uninfected PBL (Fig. 4,A). Total TGFβ1 secretion for infected PBL and CD4 cells increased >3-fold by day 14. Interestingly, TGFβ1 produced by infected PBL (∼3000 ng/ml) was nearly 10 times more compared with that of CD4 cells (∼400 ng/ml) on day 14. We also measured total TGFβ secretion by adding heat-activated supernatants to Mv1Lu reporter cells that are stably transfected with a TGFβ-responsive promoter element upstream of a luciferase reporter gene. This method also confirmed increased TGFβ secretion in HIV-infected cells (Fig. 4,B). Simultaneous measurement of TGFβ by ELISA and reporter cell assay showed similar fold increases (Fig. 4,C). Furthermore, we observed a 2- to 3-fold increase in TGFβ mRNA expression from HIV-infected PBL, CD4, and CD8 cells (day 14) compared with uninfected cells (Fig. 4,D). We also analyzed supernatants for changes in other cytokines by multiplex flow cytometry. We found several other cytokines elevated by >2-fold, including IL-1, IL-13, and IFN-inducible protein 10. However, neutralizing Abs to these cytokines did not affect fibronectin synthesis (Fig. 3 B and data not shown).

FIGURE 4.

HIV infection increases total TGFβ1 secretion by lymphocytes. A, Supernatant from uninfected and HIV-infected PBL and CD4 cells was collected on days 2, 7, and 14 after infection. Total TGFβ1 (active and latent) in the conditioned supernatant were measured by ELISA. Data are reported as average of triplicate samples ± SD. B, Cell-free supernatant was collected and heat activated from uninfected and HIV-infected PBL, CD4, and CD8 cells (day 14). Supernatants were incubated with Mv1Lu reporter cells overnight at 37°C, then cells were washed and lysed and luciferase activity was measured. As a control, supernatants were preincubated with anti-TGFβ1 Ab before addition to Mv1Lu reporter cells. Data are presented as average luciferase activity of triplicate samples ± SD. C, Cell-free supernatant was collected from PBL (days 7 and 14) and heat activated. Total TGFβ was simultaneously measured by ELISA (right) and Mv1Lu reporter assay (left). D, Increased TGFβ RNA expression by HIV-infected cells. RNA from uninfected or HIV-infected PBL, CD4, or CD8 cells was reverse transcribed and PCR amplified with TGFβ-specific primers. Amplicons were separated on a 1.5% agarose gel and detected by ethidium bromide staining (top panel). As a control for loading, samples were amplified with primers for 18S ribosomal RNA (bottom panel).

FIGURE 4.

HIV infection increases total TGFβ1 secretion by lymphocytes. A, Supernatant from uninfected and HIV-infected PBL and CD4 cells was collected on days 2, 7, and 14 after infection. Total TGFβ1 (active and latent) in the conditioned supernatant were measured by ELISA. Data are reported as average of triplicate samples ± SD. B, Cell-free supernatant was collected and heat activated from uninfected and HIV-infected PBL, CD4, and CD8 cells (day 14). Supernatants were incubated with Mv1Lu reporter cells overnight at 37°C, then cells were washed and lysed and luciferase activity was measured. As a control, supernatants were preincubated with anti-TGFβ1 Ab before addition to Mv1Lu reporter cells. Data are presented as average luciferase activity of triplicate samples ± SD. C, Cell-free supernatant was collected from PBL (days 7 and 14) and heat activated. Total TGFβ was simultaneously measured by ELISA (right) and Mv1Lu reporter assay (left). D, Increased TGFβ RNA expression by HIV-infected cells. RNA from uninfected or HIV-infected PBL, CD4, or CD8 cells was reverse transcribed and PCR amplified with TGFβ-specific primers. Amplicons were separated on a 1.5% agarose gel and detected by ethidium bromide staining (top panel). As a control for loading, samples were amplified with primers for 18S ribosomal RNA (bottom panel).

Close modal

TGFβ1 is secreted as a latent form and requires activation to exert its effect. As an indicator of TGFβ1 activation, we used Mv1Lu reporter cells transfected with a portion of the PAI-1 promoter upstream of a luciferase reporter gene. The PAI-1 promoter contains a well-characterized TGFβ-responsive element. Therefore, if active TGFβ1 is present, an increase in luciferase activity will be detected. To determine whether HIV-infected lymphocytes secrete biologically active TGFβ, we incubated supernatants from uninfected and infected cells with Mv1Lu reporter cells and measured luciferase activity after 16 h. We found an increase in biologically active TGFβ secreted by HIV-infected cells (Fig. 5,A). To verify specificity, supernatants were pretreated with an anti-TGFβ Ab, which eliminated the increased luciferase activity. Treatment of Mv1Lu with HIV alone did not affect luciferase activity. We asked whether soluble gp120 was sufficient to increase TGFβ activity. Incubation of full-length recombinant (monomeric) gp120 with CD8, PBL, or CD4 did not cause increased TGFβ activity (Fig. 5,B and data not shown). However, when anti-gp120 Ab was then added, TGFβ secretion increased to levels comparable to addition of HIV to CD8 cells (Fig. 5 B). Treatment of CD8 cells with anti-gp120 Ab alone (without soluble gp120 or HIV) did not alter TGFβ secretion. In addition, pretreatment of CD8 cells with anti-CXCR4 Ab blocked the increased TGFβ secretion. Thus, our data show that interactions of HIV envelope protein gp120 with the chemokine receptor CXCR4 is sufficient to induce TGFβ secretion by CD8 cells.

FIGURE 5.

TGFβ1 secreted by HIV-infected lymphocytes is biologically active. A, Mv1Lu reporter cells were incubated with supernatants from uninfected and HIV-infected PBL, CD4, or CD8 lymphocytes (day 14). After 16 h, luciferase activity was measured. As a control, indicated supernatants were pretreated with an anti-TGFβ1 Ab. As an additional control, HIV was directly added to Mv1Lu cells. Data are presented as average luciferase activity of triplicate samples ± SD. B, CD8 cells were incubated with HIV, HIV plus anti-CXCR4 Ab, recombinant gp120, gp120 Ab alone, or gp120 plus anti-gp120 Ab. Supernatant was collected on day 14 and incubated as above with MvLu1. Data are presented as average luciferase activity of triplicate samples ± SD. C, Smad activation in fibroblasts. IMR-90 fibroblasts were cocultured with HIV-infected PBL overnight and then lysed in buffer containing phosphatase inhibitors. Equal amounts of protein were separated by reducing SDS-PAGE and blotted with the Ab to phospho-Smad2/3.

FIGURE 5.

TGFβ1 secreted by HIV-infected lymphocytes is biologically active. A, Mv1Lu reporter cells were incubated with supernatants from uninfected and HIV-infected PBL, CD4, or CD8 lymphocytes (day 14). After 16 h, luciferase activity was measured. As a control, indicated supernatants were pretreated with an anti-TGFβ1 Ab. As an additional control, HIV was directly added to Mv1Lu cells. Data are presented as average luciferase activity of triplicate samples ± SD. B, CD8 cells were incubated with HIV, HIV plus anti-CXCR4 Ab, recombinant gp120, gp120 Ab alone, or gp120 plus anti-gp120 Ab. Supernatant was collected on day 14 and incubated as above with MvLu1. Data are presented as average luciferase activity of triplicate samples ± SD. C, Smad activation in fibroblasts. IMR-90 fibroblasts were cocultured with HIV-infected PBL overnight and then lysed in buffer containing phosphatase inhibitors. Equal amounts of protein were separated by reducing SDS-PAGE and blotted with the Ab to phospho-Smad2/3.

Close modal

To determine whether the increased TGFβ bioactivity was due to increased secretion or increased activation, we determined the proportion of active TGFβ present in each sample. To do this, supernatants were divided and one aliquot was heat activated (total TGFβ) and the other aliquot was left untreated (active TGFβ) before adding to Mv1Lu cells. The percentage of active TGFβ was similar in HIV-infected and uninfected lymphocytes (Table II). This suggests that HIV infection causes a proportionate increase in total and active TGFβ.

Table II.

Increased TGFβ bioactivity is due to increased secretion, not increased activationa

Cells% Active TGFβ
Uninfected PBL 13.43 
HIV-infected PBL 14.3 
Uninfected CD4 13.46 
HIV-infected CD4 13.71 
Uninfected CD8 8.31 
HIV-infected CD8 12.87 
Cells% Active TGFβ
Uninfected PBL 13.43 
HIV-infected PBL 14.3 
Uninfected CD4 13.46 
HIV-infected CD4 13.71 
Uninfected CD8 8.31 
HIV-infected CD8 12.87 
a

Supernatants from uninfected and infected cells were divided and either heat treated to activate TGFβ (total TGFβ) or untreated (active TGFβ) prior to incubation with Mv1Lu cells. After 16 h, luciferase activity was measured and presented as the percentage of luciferase activity from untreated samples (active TGFβ) to heat-treated samples.

TGFβ signaling can occur through Smad-dependent and Smad-independent pathways (7). TGFβ interaction with its receptor leads to activation and phosphorylation of Smad2/3, which then translocates from the cytoplasm to the nucleus and induces gene expression. Therefore, we examined the level of phosphorylated Smad2/3 in fibroblasts cocultured with HIV-infected or uninfected PBL. We found coculture with HIV-infected PBL caused a significant increase in phosphorylated Smad (Fig. 5 C), showing Smad activation in this cell system.

To gain insight into the mechanism of matrix remodeling observed with HIV infection, we examined the effect of HIV-infected lymphocytes on fibronectin production by fibroblasts. Our study shows that coculture of HIV-infected lymphocytes with fibroblasts leads to increased fibronectin deposition, increased fibronectin RNA synthesis, and increased fibronectin promoter activity. We demonstrated that this effect did not require direct contact between lymphocytes and fibroblasts and instead was mediated by a soluble factor. We showed that HIV infection increases TGFβ mRNA and biologically active TGFβ1 protein by lymphocytes and that the increased fibronectin synthesis was due to TGFβ1. Furthermore, coculture of fibroblasts with HIV-infected cells activated the Smad pathway, the major TGFβ signaling pathway. In addition to fibronectin, HIV-infected lymphocytes increased fibroblast collagen I mRNA, another matrix protein typically up-regulated during fibrosis, but not collagen IV mRNA, a basement membrane component. Thus, HIV-infection specifically promotes a profibrotic environment that is fibronectin-enriched. In conjunction with our previous work that showed fibronectin-bound HIV was more stable and infectious than unbound virus, we propose a model in which HIV alters the microenvironment to one in which infectivity and viral stability are enhanced (Fig. 6).

FIGURE 6.

Model of HIV interactions with extracellular fibronectin.

FIGURE 6.

Model of HIV interactions with extracellular fibronectin.

Close modal

HIV infection results in alterations of the normal cytokine expression profile, that, in turn, contribute to HIV pathogenesis. TGFβ1, a member of an important family of immunomodulating cytokines, is overexpressed during HIV infection (8). TGFβ1 overexpression in HIV may contribute to the immune dysfunction by inhibition of T cell activation, suppression of B cell function, and inhibition of IFN-γ production (9). However, in addition to its immunomodulatory effects, TGFβ is an important mediator of the fibrotic response (10). In contrast to its effect on most cell types, TGFβ stimulates the growth of fibroblasts and smooth muscle cells and promotes stromal connective tissue formation. Coculture of fibroblasts with HIV-infected cells activated the Smad pathway. Recent work showed that TGFβ-dependent fibronectin synthesis required Smad activation (11, 12). Our data also suggest a role of Smad signaling in fibronectin synthesis, although we cannot rule out the contribution of Smad-independent pathways, such as c-Jun N-terminal kinase (13) or Connective tissue growth factor pathways (14). Although we found elevation of additional cytokines besides TGFβ after HIV infection, neutralizing Abs to these cytokines failed to block fibronectin synthesis. Thus, our data support the contribution of TGFβ overexpression to the fibrotic process in HIV infection, in addition to its immunosuppressive effects.

TGFβ1 is secreted as a latent protein that is noncovalently associated with latency-associated peptide and requires activation to exert its effects. Several mechanisms induce TGFβ activation including cleavage by plasmin or metalloproteinases and interaction with integrins (15). One study showed that apoptotic lymphocytes released TGFβ (16). However, since we showed similar viability between HIV-infected and uninfected cells, this is unlikely to be the explanation in our system. A role for reactive oxygen species in TGFβ1 activation in lymphocytes was also reported (16). However, our data do not support increased activation of TGFβ1 as a mechanism for increased activity, but rather increased total secretion.

Interestingly, we found that incubation of CD8 cells with HIV resulted in increased TGFβ1 secretion and increased fibronectin promoter activity. Although HIV does not typically infect CD8 cells, interactions of HIV with chemokine receptors or other cell surface proteins may transduce signaling pathways that increase TGFβ1 expression. For example, exposure of astrocytes in vitro to HIV resulted in enhanced TGFβ1 expression (17). Tat protein was reported to induce TGFβ release from macrophages (18). In another study, HIV Ags induced TGFβ1 secretion from PBMC derived from HIV-infected patients, independent of Tat (19). CD8 cells were the major cell type responsible for increased TGFβ secretion (19). Our results confirm that HIV can alter TGFβ expression without viral infection. Soluble gp120 alone did not induce TGFβ1 secretion in CD8, CD4, or PBL. However, subsequent addition of anti-gp120 Ab induced TGFβ secretion, suggesting that cross-linking of the chemokine receptor is necessary for TGFβ induction. Pretreatment of CD8 cells with anti-CXCR4 Ab also blocked TGFβ increase. Therefore, our data support a model wherein gp120 on HIV can interact with CXCR4 chemokine receptor and cause increased TGFβ mRNA and secretion without the productive infection of cells.

HIV infection results in a state of chronic activation that produces fibrosis in the lymph node tissue. Determination of the factors that contribute to the disruption of the lymph node microenvironment is critical since the degree of fibrosis may be a predictor of response to HAART (1). Advanced fibrosis is generally considered irreversible, even after the initiating agent is removed. However, there is increasing evidence that regression of established fibrosis is possible. With initiation of HAART, lymphoid tissue architecture can improve with reformation of organized lymphoid follicles and follicular dendritic cell networks (20). In patients with liver fibrosis and cirrhosis, treatment for hepatitis C with antiviral drugs resulted in histological regression of fibrosis in almost half of the patients (21). Animal models of liver fibrosis also show that improvement of established fibrosis is possible (22, 23). HAART can indirectly contribute to improved lymph node architecture by a decrease in HIV-infected cells, which in turn, decreases secreted TGFβ and eliminates the profibrotic stimulus. Because lymph node fibrosis may impair immune reconstitution (1), addition of antifibrotic therapies may increase the effectiveness of HAART.

In summary, we show that HIV-infected lymphocytes contribute to immune dysfunction by creation of a fibronectin-rich, fibrotic environment that also facilitates HIV infection of naive lymphocytes. This effect is mediated by increased TGFβ1 secretion by HIV-infected lymphocytes, which then acts on local fibroblasts and activates the Smad pathway. The extracellular matrix is a dynamic environment that impacts on the ability of HIV to infect cells and contributes to immune dysfunction. Insight into factors that contribute to lymph node fibrosis is important to develop therapies aimed at preventing the initial matrix remodeling. Additionally, therapies aimed at facilitating the resolution of lymph node fibrosis in select patients may be appropriate adjuvant therapy to HAART.

We thank Dr. Robert Coombs, Joan Dragavon, and G. Aggarwal for advice and assistance.

1

This work was supported by National Institutes of Health Grant RO1 HL57890 and University of Washington Center for AIDS Research National Institutes of Health Grant AI 30731.

3

Abbreviations used in this paper: HAART, highly active antiretroviral therapy; PAI-1, plasminogen activator inhibitor 1.

1
Schacker, T. W., P. L. Nguyen, G. J. Beilman, S. Wolinsky, M. Larson, C. Reilly, A. T. Haase.
2002
. Collagen deposition in HIV-1 infected lymphatic tissues and T cell homeostasis.
J. Clin. Invest.
110
:
1133
.
2
Greco, G., S. Pal, R. Pasqualini, L. M. Schnapp.
2002
. Matrix fibronectin increases HIV stability and infectivity.
J. Immunol.
168
:
5722
.
3
Tellier, M. C., G. Greco, M. Klotman, A. Mosoian, A. Cara, W. Arap, E. Ruoslahti, R. Pasqualini, L. M. Schnapp.
2000
. Superfibronectin, a multimeric form of fibronectin, increases HIV infection of primary CD4+ T lymphocytes.
J. Immunol.
164
:
3236
.
4
Leblond, V., C. Legendre, G. Gras, N. Dereuddre-Bosquet, C. Lafuma, D. Dormont.
2000
. Quantitative study of β1-integrin expression and fibronectin interaction profile of T lymphocytes in vitro infected with HIV.
AIDS Res. Hum. Retroviruses
16
:
423
.
5
Michaelson, J. E., J. D. Ritzenthaler, J. Roman.
2002
. Regulation of serum-induced fibronectin expression by protein kinases, cytoskeletal integrity, and CREB.
Am. J. Physiol.
282
:
L291
.
6
Abe, M., J. G. Harpel, C. N. Metz, I. Nunes, D. J. Loskutoff, D. B. Rifkin.
1994
. An assay for transforming growth factor-β using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct.
Anal. Biochem.
216
:
276
.
7
Derynck, R., Y. E. Zhang.
2003
. Smad-dependent and Smad-independent pathways in TGF-β family signalling.
Nature
425
:
577
.
8
Kekow, J., W. Wachsman, J. A. McCutchan, M. Cronin, D. A. Carson, M. Lotz.
1990
. Transforming growth factor β and noncytopathic mechanisms of immunodeficiency in human immunodeficiency virus infection.
Proc. Natl. Acad. Sci. USA
87
:
8321
.
9
Lotz, M., P. Seth.
1993
. TGF beta and HIV infection.
Ann. NY Acad. Sci.
685
:
501
.
10
Taipale, J., J. Saharinen, J. Keski-Oja.
1998
. Extracellular matrix-associated transforming growth factor-β: role in cancer cell growth and invasion.
Adv. Cancer Res.
75
:
87
.
11
Itoh, S., M. Thorikay, M. Kowanetz, A. Moustakas, F. Itoh, C. H. Heldin, P. ten Dijke.
2003
. Elucidation of Smad requirement in transforming growth factor-β type I receptor-induced responses.
J. Biol. Chem.
278
:
3751
.
12
Isono, M., S. Chen, S. W. Hong, M. C. Iglesias-de la Cruz, F. N. Ziyadeh.
2002
. Smad pathway is activated in the diabetic mouse kidney and Smad3 mediates TGF-β-induced fibronectin in mesangial cells.
Biochem. Biophys. Res. Commun.
296
:
1356
.
13
Hocevar, B. A., T. L. Brown, P. H. Howe.
1999
. TGF-β induces fibronectin synthesis through a c-Jun N-terminal kinase-dependent, Smad4-independent pathway.
EMBO J.
18
:
1345
.
14
Chen, M. M., A. Lam, J. A. Abraham, G. F. Schreiner, A. H. Joly.
2000
. CTGF expression is induced by TGF-β in cardiac fibroblasts and cardiac myocytes: a potential role in heart fibrosis.
J. Mol. Cell. Cardiol.
32
:
1805
.
15
Munger, J. S., X. Huang, H. Kawakatsu, M. J. Griffiths, S. L. Dalton, J. Wu, J. F. Pittet, N. Kaminski, C. Garat, M. A. Matthay, et al
1999
. The integrin αvβ6 binds and activates latent TGF β1: a mechanism for regulating pulmonary inflammation and fibrosis.
Cell
96
:
319
.
16
Chen, W., M. E. Frank, W. Jin, S. M. Wahl.
2001
. TGF-β released by apoptotic T cells contributes to an immunosuppressive milieu.
Immunity
14
:
715
.
17
da Cunha, A., R. W. Jackson, L. Vitkovic.
1995
. HIV-1 non-specifically stimulates production of transforming growth factor-β1 transfer in primary astrocytes.
J. Neuroimmunol.
60
:
125
.
18
Zauli, G., B. R. Davis, M. C. Re, G. Visani, G. Furlini, M. La Placa.
1992
. tat protein stimulates production of transforming growth factor-β1 by marrow macrophages: a potential mechanism for human immunodeficiency virus-1-induced hematopoietic suppression.
Blood
80
:
3036
.
19
Garba, M. L., C. D. Pilcher, A. L. Bingham, J. Eron, J. A. Frelinger.
2002
. HIV antigens can induce TGF-β1-producing immunoregulatory CD8+ T cells.
J. Immunol.
168
:
2247
.
20
Zhang, Z.-Q., T. Schuler, W. Cavert, D. W. Notermans, K. Gebhard, K. Henry, D. V. Havlir, H. F. Gunthard, J. K. Wong, S. Little, et al
1999
. Reversibility of the pathological changes in the follicular dendritic cell network with treatment of HIV-1 infection.
Proc. Natl. Acad. Sci. USA
96
:
5169
.
21
Poynard, T., J. McHutchison, M. Manns, C. Trepo, K. Lindsay, Z. Goodman, M. H. Ling, J. Albrecht.
2002
. Impact of pegylated interferon α-2b and ribavirin on liver fibrosis in patients with chronic hepatitis C.
Gastroenterology
122
:
1303
.
22
Iredale, J. P., R. C. Benyon, J. Pickering, M. McCullen, M. Northrop, S. Pawley, C. Hovell, M. J. Arthur.
1998
. Mechanisms of spontaneous resolution of rat liver fibrosis: hepatic stellate cell apoptosis and reduced hepatic expression of metalloproteinase inhibitors.
J. Clin. Invest.
102
:
538
.
23
Murphy, F. R., R. Issa, X. Zhou, S. Ratnarajah, H. Nagase, M. J. Arthur, C. Benyon, J. P. Iredale.
2002
. Inhibition of apoptosis of activated hepatic stellate cells by tissue inhibitor of metalloproteinase-1 is mediated via effects on matrix metalloproteinase inhibition: implications for reversibility of liver fibrosis.
J. Biol. Chem.
277
:
11069
.