Skip Navigation

Preservation of Intrahepatic Hepatitis C Virus (HCV)-Specific CD4+ T Cell Responses despite Global Loss of CD4+ T Cells in HCV/HIV Coinfection

  1. Dr. Chihiro Morishima1,
  2. Margaret C. Shuhart2,
  3. Christina S. Yoshihara1,a,
  4. Denise M. Paschal1,
  5. Melissa A. Silva1,
  6. Lisa V. Thomassen3,
  7. Scott S. Emerson4 and
  8. David R. Gretch1,2
  1. 1 Department of Laboratory Medicine, University of Washington School of Medicine, Seattle
  2. 2 Department of Medicine, University of Washington School of Medicine, Seattle
  3. 3 Department of Pathology, University of Washington School of Medicine, Seattle
  4. 4 Department of Biostatistics, University of Washington School of Medicine, Seattle
  1. Reprints or correspondence: Dr. Chihiro Morishima 325 9th Ave. Box 359690 Harborview Medical Center Dept. of Laboratory Medicine University of Washington School of Medicine Seattle WA 98104–2499 (chihiro{at}u.washington.edu).

  1. Presented in part: 12th International Symposium on Hepatitis C Virus and Related Viruses, Montreal, Canada, 2–6 October 2005 (poster 47).

  • ↵Present affiliation: Washington State University Pullman.

 
Next Section

Abstract

Background. Human immunodeficiency virus (HIV) coinfection and low peripheral blood CD4+ T cell counts are associated with increased hepatitis C liver disease.

Methods. Hepatitis C virus (HCV)-specific CD4+ T cell responses were assessed using interferon (IFN)-γ enzyme-linked immunospot assays on peripheral blood mononuclear cells and expanded liver lymphocytes from HCV-monoinfected and HCV/HIV-coinfected subjects. Cell frequencies were determined using flow cytometry.

Results. HIV coinfection was associated with decreased CD4+ T cell percentages in both peripheral blood (21% vs. 48%; P < .0001) and liver (15% vs. 36%; P < .0001) and with reduced responsiveness of peripheral CD4+ T cells to HCV antigens compared with HCV monoinfection (22% vs. 45%; P = .021). However, intrahepatic HCV-specific responses were maintained in HCV/HIV coinfection, compared with HCV monoinfection (38% vs. 32%; P = .7). Notably, the presence of HCV-specific responses was not related to the frequency of liver CD4+ T cells (P = .4). Circulating and liver CD4+ T cell percentages were correlated (r = 0.58; P < .0001). Circulating percentages were also inversely associated with liver fibrosis stage among HCV/HIV-coinfected subjects (P = .029). Neither hepatic CD4+ T cell percentages nor HCV-specific IFN-γ responses in the liver or periphery predicted stage.

Conclusions. Despite decreases in peripheral blood HCV-specific CD4+ T cell responses and intrahepatic CD4+ T cell percentages, intrahepatic HCV-specific CD4+ IFN-γ responses were preserved in HCV/HIV coinfection.

Current therapies for HIV infection have led to a significant increase in life expectancy for infected individuals. Within this context, coinfection with hepatitis C virus (HCV) has emerged as a major health problem, affecting as many as 30% of HIV-infected individuals in the United States [1]. Of great concern is the observation that accelerated liver fibrosis progression and increased rates of liver failure are found in HCV/HIV coinfection, compared with HCV monoinfection [2].

The role of immune effectors in HCV liver fibrosis progression is poorly understood. However, several groups have observed that lower peripheral blood CD4+ T cell counts are associated with more advanced liver disease stage in HCV/HIV-coinfected patients [3–6], raising the possibility that CD4+ T cells could be protective against liver fibrosis progression.

HCV-specific CD4+ T cell responses have been detected at low levels in HCV/HIV-coinfected subjects. Lauer et al. first reported that peripheral CD4+ lymphoproliferative responses to HCV antigens were virtually nonexistent in HCV/HIV-coinfected subjects [7]. Another group described that CD4+ liver lymphocyte interferon (IFN)-γ enzyme-linked immunospot (ELI-Spot) responses to core, NS3, and NS5 occurred at similar frequencies in both HCV-infected and HCV/ HIV-coinfected patients; however, the median response for each antigen individually among 30 patients was zero [8]. In the present study, we sought to advance our understanding of the relationships among CD4+ T cell frequencies and HCV-specific IFN-γ responses in peripheral blood mononuclear cells (PBMCs) and the liver, measured simultaneously in a large cohort of HCV-infected patients with and without HIV.

Previous SectionNext Section

Subjects, Materials, and Methods

Subjects. One hundred and nine subjects were recruited sequentially from Harborview Medical Center clinics after written informed consent through a University of Washington (UW) Institutional Review Board-approved protocol. All subjects undergoing clinical evaluation met the same criteria for liver biopsy regardless of HIV status. Twenty-one HIV-positive subjects with low CD4+ T cell counts and no history of HIV treatment consented to liver biopsies as part of a research protocol. Subjects were excluded if they had any acute infection, positive hepatitis B surface antigen, or any history of HCV treatment. Ninety-nine (90.8%) subjects had a parenteral risk factor for HCV infection. In subjects with a history of injection drug use (IDU) (n = 94), the duration of HCV infection was defined as the difference between the year of enrollment and the year of IDU onset, minus 2 years, based on seroepidemiologic data from injection drug users in the Seattle-King County region (median time from IDU onset to HCV seroconversion, 2 years) [9].

Clinical laboratory data, PBMC and liver biopsy samples were obtained within 30 days of enrollment. Peripheral blood ELISpot experiments were initiated before and concluded during the liver T cell expansion experiments, leaving only 28 patients with both peripheral and liver data available. All HIV-negative subjects had confirmatory HIV antibody testing performed. Serum HCV RNA levels (lower limit of detection [LLOD], 50 IU/mL) [10] and HCV genotype [11] were determined at the UW Clinical Virology Laboratory. Through July 2003, HIV-1 RNA quantitation was determined using a bDNA assay (LLOD, 50 copies/mL; Bayer Diagnostics). Later determinations were performed using real-time reverse-transcriptase polymerase chain reaction (LLOD, 30 copies/mL) [12]. Liver biopsy grade (0–4) and stage (0–4) were assigned by a single, blinded hepatopathologist using the Batts-Ludwig system [13].

Peripheral blood IFN-γ ELISpot assay. Freshly isolated PBMCs (200,000/well) were cultured with HCV protein antigens SOD (superoxide dismutase)-c22 (core), SOD-c100 (NS4), SOD-c200 (NS3 and NS4), SOD-c25 (core, NS3, and NS4), and SOD-NS5, as well as control SOD alone (all 10 μg/mL; gifts of Dr. M. Houghton, Chiron Corp.) and Candida and tetanus proteins as described elsewhere [14] at 37°C (5% CO2). Recombinant human interleukin (rhIL-2; 20 U/mL) was added on day 6 of culture. Cells were transferred to ELISpot plates (Millipore) coated with a primary IFN-γ antibody (Mabtech) at ∼day 10 and restimulated with the same antigens. Two days later, IFN-γ ELISpots were detected using a second biotinylated IFN-γ antibody (Mabtech), streptavidin-horseradish peroxidase colorimetric detection (BD Pharmingen), and 3-amino9-ethylcarbazole substrate kits (Vector Laboratories). Four replicates were used for media control and antigen wells, whereas 8 were used for SOD. Assays were considered valid if the phytohemagglutinin (PHA) response was positive. Spots were counted using an automated ELISpot reader (Immunospot; Cellular Technology). To be considered positive, the results were required to exceed the following thresholds, which were set a priori using data from 18 HCV-uninfected control subjects: 3 of 4 replicate wells each was required to have twice the number of spots per well as the mean SOD spots and greater than the mean SOD plus 6 spots for all antigens, except for c100, which required that the number of test spots be greater than the mean SOD plus 11 spots. Any subject with a positive IFN-γ response to 1 or more HCV antigens by these criteria was considered “positive” for the assay.

Intrahepatic T cell expansion. Liver biopsy tissue was washed, minced, and cultured in a 6-well plate in RPMI 1640 with HEPES containing 10% heat-inactivated fetal bovine serum (FBS; Gemini BioProducts), 1% Pen-Strep, 2 mmol/L l-glutamine, and 1 μmol/L indinavir (to limit HIV replication; indinavir was obtained from the AIDS Research and Reference Repository, Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health) at 37°C(5% CO2). Anti-CD3 antibody (OKT3 [30 ng/mL]; Ortho Biotech), after the method described by Riddell and Greenberg [15], or PHA (1.6 μg/mL; Remel) was used to nonspecifically stimulate lymphocytes in equal numbers of HCV-monoinfected and HCV/HIV-coinfected subjects. Eighty to 100 U/mL rhIL-2 was added on day 1 and every 2–3 days thereafter. Restimulation with irradiated allogeneic feeder cells, an irradiated allogeneic B-lymphoblastoid cell line, and OKT3 or PHA was performed every 12–14 days, for a total of 2–3 stimulations. After expansion, all liver lymphocytes were cryopreserved in 10% DMSO (Sigma Aldrich) in FBS before use in ELISpot assays. HIV p24 antigen testing of culture supernatants (sensitivity, 30 pg/mL) showed no evidence of active HIV replication. The stimulus used for expansion (anti-CD3 vs. PHA) and the number of times a culture was restimulated did not significantly affect the results (all; data not shown).

Preparation of liver lymphocytes for direct ex vivo analysis. Fresh liver biopsy tissue was minced and treated with 0.025% type IV collagenase (Sigma) and 0.002% DNAse (Benzonase Nuclease; EMD Biosciences) in RPMI 1640 with HEPES containing 10% FBS at 37°C for 30 min. Cells were washed twice before antibody labeling.

Enumeration of liver-infiltrating CD4+ T lymphocytes. Isotype-matched controls and the following fluorescently conjugated antibodies were used to label lymphocytes: α-CD4-fluorescein isothiocyanate (MY31), α-CD56-phycoerythrin (SK1), α-CD8-peridinin-chlorophyll-protein (PerCP) complex (SK3) (all from BD Biosciences), and α-CD3-allophycocyanin (UCHT1; Beckman Coulter). Cells were fixed with 1% paraformaldehyde before data collection using a Becton Dickinson FACSCalibur and analyzed using FlowJo software for Macintosh (version 6.3.3; Tree Star).

HCV-specific INF-γ ELISpot assay for cultured liver lymphocytes. One day before assay, expanded and cryopreserved liver lymphocytes (“effectors”) were thawed and allowed to “rest” overnight at 37°C (5% CO2). The next day, uncultured PBMCs were thawed and depleted using α-CD3 and α-CD56 Microbeads (Miltenyi Biotec), LD columns, and the MidiMACS Separator (Miltenyi Biotec) to prepare antigen presenting cells (APCs). Effectors and APCs (10:1) were combined on coated IFN-γ ELISpot plates for 48 h before developing. Protein antigens and data analyses were as described above for the cultured ELISpot assay.

Statistical methods. Groups were compared using the χ2 test, the 2-sided Student's t test for unequal variances, or the Wilcoxon rank-sum test, where appropriate. Correlations were assessed using the Spearman correlation coefficient. No adjustment was made for multiple comparisons. Multivariate analysis was conducted using linear or logistic regression with robust variance estimates, where appropriate. P < .05 was considered significant. Statistical analyses were conducted using Stata (version 8.0; StataCorp).

Previous SectionNext Section

Results

HIV-positive subjects were younger, included more males, and had shorter HCV disease duration and lower CD4+ T cell counts (table 1). Differences in age, sex, and disease duration between the 2 groups were considered in multivariate analyses for all results and did not influence the reported associations between the predictor of interest and the outcome, unless stated otherwise.

Less common detection of peripheral blood IFN-γ ELI- Spot responses in HCV/HIV coinfection than in HCV monoinfection. Using freshly isolated PBMCs and a cultured IFN-γ ELISpot assay to improve the sensitivity of detection, fewer HCV/HIV-coinfected subjects had HCV-specific CD4+ T cell responses detected compared with those infected with HCV alone (11/49 [22%] vs. 19/42 [45%], respectively; P = .021) (figure 1B). The summed number of spot-forming cells from HCV-antigen wells was also different between the 2 groups (mean, 1164/106 vs. 414/106 PBMCs for HCV and HCV/HIV groups, respectively; P = .04). Among the 19 HCV-monoinfected and 11 HCV/HIV-coinfected subjects with positive results, the number of HCV antigens recognized was similar (P = .32) (figure 1C), as was the distribution of responses across HCV antigens (P = .99) (figure 1D).

Figure 1.
View larger version:
Figure 1.

Less frequent detection of peripheral blood-cultured interferon (IFN)-γ enzyme-linked immunospot (ELISpot) responses in hepatitis C virus (HCV)/HIV coinfection than in HCV monoinfection. Representative examples of ELISpot results from 4 HCV-infected and 4 HCV/HIV-coinfected subjectswith positive responses to HCV antigens are shown (A). Each dot represents the no. of spot-forming units per 106 peripheral blood mononuclear cells (PBMCs) detected in a single well. Lines indicate the mean spot-forming units per 106 PBMCs for the antigen tested, and plus signs (+) indicate those antigens for which results were considered positive using criteria described in Subjects, Materials, and Methods. “M” refers to the media control for tetanus (Tet) and Candida (Cand) responses, whereas superoxide dismutase (SOD) serves as the negative control for all recombinant HCV antigens tested. The no. of subjects with HCV-specific IFN-γ responses to HCV protein antigens were compared between those infected with HCV (white bars) or HCV/HIV (black bars) using the χ2 test (B). The total no. of consecutively enrolled subjects tested is shown below the X-axis in panel B. Results from those same subjects with positive results are shown in panels C and D. The no. of HCV antigens recognized in those with positive responses was not different between the groups (P = .32, Wilcoxon rank-sum test); the horizontal lines indicate the mean values (C). The overall distribution of HCV antigens recognized was not different between the groups (P = .99, Fisher's exact test; D).

The frequency of recall responses to Candida and tetanus was also evaluated. Fewer HCV/HIV-coinfected subjects exhibited positive recall responses than HCV-monoinfected subjects (tetanus, 34.7% vs. 85.4%, respectively; Candida, 46.9% vs. 92.7%; both tetanus and Candida, 24.5% vs. 82.9%; P < .001 , for all comparisons). Furthermore, recall responses correlated with HCV antigen responses in all subjects and in the HCV/ HIV-coinfected group alone (P < .01 , for both, χ2 test) (data not shown).

Detection of similar numbers of HCV-monoinfected and HCV/HIV-coinfected subjects with HCV-specific IFN-γ ELISpot responses from liver lymphocytes. IFN-γ responses from intrahepatic CD4+ T cells were assessed in 43 consecutively

enrolled patients with expanded liver lymphocytes available. The clinical and demographic characteristics of these patients were not different from the overall cohort (data not shown). No difference was found in the frequency of HCV-specific responses when comparing 21 HCV-infected and 22 HCV/HIVcoinfected patients (38.1% vs. 31.8%, respectively; P = .7, χ2 test) (figure 2B). The summed spot-forming cells/106 liver lymphocytes for all HCV antigens was also similar when comparing the 2 disease groups (396 vs. 404 for HCV vs. HCV/HIV, respectively; P = .96). Furthermore, the mean number of antigens recognized and the distribution of responses across HCV antigens were not different between the 2 groups (P = .9 [figure 2C] and P = .40 [figure 2D], respectively). CD4+ T cell depletion experiments confirmed that IFN-γ responses were arising specifically from the helper T cell subset (data not shown). Finally, no differences were found in the frequency of subjects with positive IFN-γ responses to Candida or tetanus antigens (P = 1.0, P = .66, and P = 1.0, for Candida, tetanus, or either response, respectively).

Figure 2.
View larger version:
Figure 2.

Detection of similar nos. of subjects with hepatitis C virus (HCV) monoinfection and HCV/HIV coinfection with intrahepatic interferon (IFN)-γ enzyme-linked immunospot (ELISpot) responses. All ELISpot results from the 15 subjects with positive responses to HCV antigens are shown (A). Each dot represents the no. of spot-forming units per 106 liver lymphocytes detected in a single well. Lines indicate the mean spot-forming units per 106 liver lymphocytes for the antigen tested, and plus signs (+) indicate those antigens for which results were considered positive using criteria described in Subjects, Materials, and Methods. “M” refers to the media control for tetanus (Tet) and Candida (Cand) responses, whereas superoxide dismutase (SOD) serves as the negative control for all recombinant HCV antigens tested. Liver lymphocytes were nonspecifically expanded and cryopreserved before testing as described in Subjects, Materials, and Methods. The no. of sequentially enrolled subjects with intrahepatic HCV-specific IFN-γ responses to HCV protein antigens were compared between those infected with HCV (white bars) or HCV/HIV (black bars) using the χ2 test (B). The total no. of subjects tested is shown below the X-axis in panel B. Results from those same subjects with positive results are shown in panels C and D. The no. of HCV antigens recognized in those with positive responses was not different between the groups (P =.9, Wilcoxon rank-sum test); the horizontal lines indicate the mean values (C). The overall distribution of HCV antigens recognized was not different between the groups (P = .40, Fisher's exact test; D).

HCV-specific IFN-γ responses in one compartment not predicted by responses in another, but overlapping specificities detected when responses present in both. To determine the relationship between peripheral blood and liver responses, 28 patients with IFN-γ ELISpot results available from both compartments were analyzed (table 2). Similar to the overall cohort, peripheral blood HCV-specific IFN-γ responses were reduced in the HCV/HIV-coinfected group (n = 15), compared with the HCV-monoinfected group (n = 13; 13.3% vs. 30.8%), although the findings were not significant likely because of the smaller sample size (P = .26, Fisher's exact test). Only 1 coinfected and 2 monoinfected patients had HCV-specific responses detected in both compartments; in those cases the HCV antigens recognized overlapped but were not identical (table 3). No association was found between the occurrence of IFN-γ- positive HCV-specific responses in the peripheral blood with that in the liver (P = 1.0, Fisher's exact test). However, IFN-γ ELISpot responses were found more commonly in the liver than in the peripheral blood for the HCV/HIV-coinfected group (7/15 [46.7%] vs. 2/15 [13.3%], respectively), as well as in the HCV-monoinfected group (7/13 [53.8%] vs. 4/13 [30.8%], respectively).

Figure 3.
View larger version:
Figure 3.

Decreased CD4+ T cell percentages in liver lymphocyte cultures from hepatitis C virus (HCV)/HIV coinfection, compared with HCV monoinfection. Liver lymphocytes were expanded nonspecifically and analyzed for the percentage of CD4+ T cells present at the end of the culture period (before cryopreservation) using antibodies specific for CD3, CD56, CD4, and CD8 as described in Subjects, Materials, and Methods (A). The percentage of CD4+ T cells was derived from gating on the subpopulation of live lymphocytes (determined using forward and side scatter characteristics) that expressed CD3. The data shown are the percentage of CD4+ CD8-CD56- T cells among CD3+ T cells. Twenty thousand to 30,000 gated events were collected per sample. The horizontal lines indicate the mean percentage of CD4+ T cells for each group. The percentage of CD4+ T cells among cultured liver lymphocytes at serial time points during the culture process was also determined (B). The white circles and dashed lines denote results from the HCV-monoinfected samples, whereas the black circles and solid lines denote results from the HCV/HIV-coinfected samples. The absolute mean change over time in all the cultures tested was 0.6%. When data from the first, second, and third time points were compared in pairwise fashion using a 2-sided t test, no significant differences were identified. Four clinically indicated ultrasound-guided biopsies yielded larger amounts of liver biopsy material (1 HCV infected and 3 HCV/ HIV coinfected) and were analyzed directly ex vivo by flow cytometric analysis using antibodies for CD3, CD56, CD4, and CD8 as described in Subjects, Materials, and Methods (C and D). The plots in panel C show the percentage of cells expressing CD3, CD56, both, or neither cell surface molecules as a percentage of total lymphocytes. The lymphocyte population was derived using forward and side scatter characteristics. The percentage of CD4+ and CD8+ T cells was derived from gating on the subpopulation of live lymphocytes that expressed CD3 but not CD56 (D). All possible gated events were collected for the fresh ex vivo liver lymphocyte samples (mean, 5000 events; range, 2000–10,000 events). Note that the fresh ex vivo data from panels C and D are also included in panel B at time 0; a small sample from the same biopsy was concurrently cultured as described and the percentage of CD4+ T cells determined at sequential time points.

Figure 4.
View larger version:
Figure 4.

Relationship between liver CD4+ T cell percentages and peripheral CD4+ T cell percentages in hepatitis C virus (HCV)-infected subjects. The percentage of expanded liver lymphocytes expressing CD3 and CD4 (as determined for figure 3, Y-axis) was plotted against the percentage of peripheral blood CD3+ CD4+ T cells (X-axis) obtained within 1 month of liver biopsy. HCV-monoinfected subjects are shown by white circles, and HCV/ HIV-coinfected subjects are shown by black circles. A strong association was found for the overall cohort (n = 60; r = 0.58; P < .0001, Spearman correlation coefficient) and was somewhat weaker for the HCV/HIV-infected cohort considered alone (n = 33; r = 0.43; P = .015). No association was found for the HCV-monoinfected group alone (P = .7).

Table 1.
View larger version:
Table 1.

Subject characteristics according to HIV status.

Table 2.
View larger version:
Table 2.

No. of subjects with hepatitis C virus (HCV)-specific interferon-g enzyme-linked immunospot responses in peripheral blood (PB) and liver.

Table 3.
View larger version:
Table 3.

Hepatitis C virus (HCV)-specific interferon (IFN)-γ enzyme-linked immunospot (ELISpot) responses in subjects with responses in both peripheral blood (PB) and liver.

Detection of intrahepatic HCV-specific CD4+ T cell responses not precluded by decreased liver CD4+ T cell percentages in HCV/HIV coinfection. We sought to determine whether overall liver CD4+ T cell percentages could account for the similar frequency of HCV-specific CD4+ T cell responses in HCV-and HCV/HIV-infected subjects. However, the percentage of CD4+ T cells in liver lymphocyte cultures from 33 HCV/HIV-coinfected subjects (mean ± SD, 14.8 ± 14.5) was significantly diminished compared with those from 27 HCV-infected subjects (35.6 ± 17.7) (P < .0001) (figure 3A).

We investigated the stability of the CD4+ T cell percentages from expanded populations over time by serially assaying CD4+ T cell percentages in the cultures just before the second and third of 3 rounds of stimulation and before cryopreservation (figure 3B). In addition, 4 patients had larger biopsy samples available, allowing for an assessment of CD4+ T cell percentages present in the liver infiltrate directly ex vivo as well as at sequential cultured time points (figure 3C and 3D). Similar percentages of CD4+ T cells were found at the first and last expanded time points in both HCV-monoinfected and HCV/ HIV-coinfected liver lymphocyte cultures (overall mean increase in absolute CD4+ T cell frequency, 0.6%). In addition, CD4+ T cell percentages from the fresh ex vivo experiments were similar to those from expanded lymphocyte populations (figure 3D).

No relationship was detected between cultured CD4+ liver lymphocyte percentages and the presence of HCV-specific responses (P = .4, t test). It is notable that HCV-specific IFN-γ responses could be detected in HCV/HIV-coinfected patients with extremely low percentages of CD4+ T cells in their liver cultures (e.g., 2.8% and 3.5%); these responses were eliminated with CD4+ T cell depletion.

CD4+ T cell percentages in the liver reflect CD4+ T cell percentages in the circulation. Others have reported that HCV liver disease progression is related to peripheral blood CD4+ T cell counts [3–6]. Thus, it was of interest to determine whether liver CD4+ T cell percentages were related to peripheral blood CD4+ T cell counts. Because peripheral CD4+ T cell counts are subject to daily variations in total lymphocyte counts, we also assessed for a relationship between intrahepatic CD4+ T cell percentages and peripheral blood CD4+ T cell percentages. Expanded liver CD4+ T cell percentages were correlated with both peripheral blood CD4+ T cell percentages (r = 0.58; P < .0001 , Spearman's correlation) (figure 4) and the peripheral CD4+ T cell count (r = 0.48 ; P = .0001) in the overall cohort.

Association of peripheral blood CD4+ T cell percentages and T cell counts with liver fibrosis stage. Associations between the immunological findings and the demographic or clinical variables were investigated. No significant relationship was found between the frequency of peripheral blood or liver CD4+ IFN-γ responses and the peripheral blood CD4+ T cell count or percentages, regardless of how the analyses were performed. Subject demographics, serum alanine aminotransferase level, HCV genotype, HCV level, HIV level, HIV treatment with highly active antiretroviral therapy, and liver disease grade or stage also were not associated with HCV-specific CD4+ T cell responses in peripheral blood or liver or with liver CD4+ T cell percentages using the appropriate univariate and multivariate analyses (data not shown). However, similar to findings from other cohorts [3–6], both peripheral blood CD4+ T cell counts and CD4+ T cell percentages were inversely associated with liver fibrosis stage in our HCV/HIV-coinfected subjects (n = 49; multivariate logistic regression adjusting for age, sex, and hepatitis C duration, P = .017 and P = .029, respectively).

Previous SectionNext Section

Discussion

In this study, we comprehensively examined IFN-γ ELISpot responses and CD4+ T cell frequencies in the peripheral blood and liver compartments of treatment-naive HCV-infected patients with and without HIV coinfection. A major finding is that HCV-specific (and recall antigen-specific) IFN-γ responses are preserved in the liver during HCV/HIV coinfection, despite considerable reductions in such responses in the peripheral blood and a very significant reduction in overall CD4+ T cell percentages in the liver. Thus, the major detriment of HIV coinfection on HCV-specific CD4+ IFN-γ responses was observed in the periphery and not liver. This result is remarkable considering the evidence that HIV coinfection dramatically reduced total CD4+ T cell populations in both the peripheral blood and liver, raising the important question of how HCV-specific CD4+ T cell responses are maintained in the liver. One possible answer is that nonrandom depletion or exclusion of CD4+ T cell subpopulations could be occurring in the livers of HCV/HIV-coinfected individuals. An interesting candidate for specific depletion/exclusion would be CD4+ regulatory T cells, which have been detected in the livers of patients with HCV monoinfection [16] and appear to be responsible for blunting HCV-specific immune responses during chronic infection [17, 18]. Thus, it is tempting to speculate that the relatively high frequency of intrahepatic HCV-specific responses in HCV/HIVcoinfected patients could be due to the specific absence of regulatory T cells, which have the ability to limit IFN-γ responses. Experiments are currently under way to examine this possibility.

Our systematic evaluation of peripheral blood and liver compartments in these patients has revealed other important observations. For example, we describe for the first time that peripheral blood CD4+ T cell percentages predict liver CD4+ T cell percentages in HCV/HIV coinfection. By contrast, peripheral CD4+ T cell immunity did not necessarily predict intrahepatic immunity, or vice versa, in a subset of patients who had both peripheral blood and liver responses examined. Finally, our results independently confirm and expand on previous reports that examined HCV-specific CD4+ T cell responses in liver or PBMC compartments separately [8, 19]. Graham et al. reported similar frequencies of HCV-specific IFN-γ responses in the liver between HCV-monoinfected and HCV/HIV-coinfected subjects [8]. However, the increased level of the IFN-γ responses in the present study permitted a characterization of the number of antigens recognized, as well as the breadth of the responses, both of which were similar between the groups (figures 1 and 2). Likewise, we now extend to adult populations the previously reported finding that peripheral blood HCV-specific IFN-γ responses by CD4+ T cells are diminished in HCV/HIV-coinfected, compared with HCVmonoinfected, children [19].

The amount of excess liver biopsy tissue available usually yields insufficient liver lymphocyte numbers for functional testing, necessitating lymphocyte expansions using standard techniques [8, 15, 20]. We therefore considered whether the decreased CD4+ T cell percentages found in HCV/HIV-coinfected subjects could have been due to a proliferative defect of hepatic CD4+ T cells. However, the lack of significant change over time in the liver lymphocyte culture composition, as well as the direct ex vivo liver CD4+ lymphocyte percentages, confirmed the data from expanded populations. Furthermore, immunohistochemical analyses by Canchis et al. demonstrated similar results in 38 HCV/HIV-coinfected and 41 HCV-monoinfected subjects [21]. Thus, the present study confirms that these differences are found using a completely different approach. As for the detection of HCV-specific peripheral blood CD4+ T cell responses, our short-term antigen-specific expansion step to increase the sensitivity of detection of HCV-specific responses produced results comparable to those reported using direct ex vivo IFN-γ ELISpot assays in children [19]. These findings reassure us that the differences found in liver CD4+ T cell percentages and peripheral blood responses between the 2 groups were not artifacts of the particular assays we chose to employ.

It seems intuitive that peripheral blood CD4+ T cell percentages are related to liver CD4+ T cell percentages as found here in HCV/HIV coinfection, because the peripheral blood likely serves as the pool from which liver lymphocytes are drawn. However, it was surprising that no relationship was found between liver CD4+ T cell percentages and liver fibrosis stage, because peripheral blood CD4+ T cell percentages were related to fibrosis stage in this cohort. Possible obstacles to identifying a relationship between these 2 parameters include inadequate sample size, the need to expand liver lymphocytes, and sampling error inherent in liver biopsies. It should be noted, however, that Canchis et al. also found no association between liver CD4+ T cell frequencies enumerated in situ and liver disease stage in 38 HCV/HIV-coinfected subjects [21].

Finally, we investigated the relationship between intrahepatic IFN-γ production and liver fibrosis severity, because IFN-γ has known antifibrotic properties [22, 23]. However, no such association was found. These results were not altogether surprising given that CD8+ T cells and NK cells can also elaborate IFN-γ and were not included in these analyses. Furthermore, Graham et al. recently studied as many as 107 HCV/HIV-coinfected patients and found only a weak inverse correlation between peripheral blood CD4+ IFN-γ ELISpot frequencies and liver fibrosis stage (r = -0.22; P = .02) [24]. It is notable that our and other studies [8, 14] investigating liver lymphocyte function utilized expanded cells, leaving open the possibility that dysfunctional intrahepatic lymphocytes that play a role in liver fibrosis were not expanded. Thus, T cell functions not measured here or the function of HCV-nonspecific T cells could be related to liver fibrosis stage and should be evaluated in future studies.

In conclusion, we show for the first time in a single HCV/ HIV-coinfected cohort that HCV-specific IFN-γ responses by liver CD4+ T cells were preserved in the face of significantly decreased liver CD4+ T cell percentages, compared with subjects with HCV monoinfection. These diminished CD4+ T cell percentages were predicted by decreased peripheral blood CD4+ T cell counts and percentages. However, no relationships were found between intrahepatic CD4+ T cell percentages or HCV-specific IFN-γ responses and liver fibrosis stage. Future experiments are needed to define the mechanisms underlying the differing frequencies of HCV-specific CD4+ T cell responses observed in the periphery and liver of HCV/HIV-coinfected individuals, as well as the role of CD4+ T cells, if any, in liver fibrosis progression.

Previous SectionNext Section

Acknowledgments

We thank T. Mathisen and C. Glenn for help with recruitment of subjects, M. Gano and N. Lejarcegui for technical assistance, E. Glynn and J. Do for help with preparation of the manuscript, and R. D. Harrington, M. J. McElrath, B. Wood, and A. E. T. Yeo for helpful discussions and guidance. We also thank K. Crawford and M. Houghton for the generous gift of hepatitis C virus protein antigens.

Previous SectionNext Section

Footnotes

  • Potential conflicts of interest: none reported.

  • Financial support: University of Washington General Clinical Research Center (grants M01RR-00037 and R01 AI49168–01 to D.R.G. and and R21 AI53817–01 to C.M.); University of Washington Center for AIDS Research, a National Institutes of Health-funded program (2004 developmental grant P30-AI27757 to C.M.).

  • Received November 14, 2006.
  • Accepted February 18, 2007.
Previous Section
 

References

  1. 1.↵
    1. Sherman KE,
    2. Rouster SD,
    3. Chung RT,
    4. Rajicic N
    . Hepatitis C virus prevalence among patients infected with human immunodeficiency virus: a cross-sectional analysis of the US adult AIDS Clinical Trials Group. Clin Infect Dis 2002;34:831-7.
    Abstract/FREE Full Text
  2. 2.↵
    1. Graham CS
    . Influence of human immunodeficiency virus infection on the course of hepatitis C virus infection: a meta-analysis. Clin Infect Dis 2001;33:562-9.
    Abstract/FREE Full Text
  3. 3.↵
    1. Benhamou Y,
    2. Bochet M,
    3. Di M-V,
    4. et al
    . Liver fibrosis progression in human immunodeficiency virus and hepatitis C virus coinfected patients. Hepatology 1999;30:1054-8.
    CrossRefMedlineWeb of Science
  4. 4.
    1. Eyster ME,
    2. Diamondstone LS,
    3. Lien JM,
    4. Ehmann WC,
    5. Quan S,
    6. Goedert JJ
    . Natural history of hepatitis C virus infection in multitransfused hemophiliacs: effect of coinfection with human immunodeficiency virus. The Multicenter Hemophilia Cohort Study. J Acquir Immune Defic Syndr 1993;6:602-10.
    MedlineWeb of Science
  5. 5.
    1. Telfer P,
    2. Sabin C,
    3. Devereux H,
    4. Scott F,
    5. Dusheiko G,
    6. Lee C
    . The progression of HCV-associated liver disease in a cohort of haemophilic patients. Br J Haematol 1994;87:555-61.
    MedlineWeb of Science
  6. 6.↵
    1. Telfer PT,
    2. Brown D,
    3. Devereux H,
    4. Lee CA,
    5. DuSheiko GM
    . HCV RNA levels and HIV infection: evidence for a viral interaction in haemophilic patients. Br J Haematol 1994;88:397-9.
    MedlineWeb of Science
  7. 7.↵
    1. Lauer GM,
    2. Nguyen TN,
    3. Day CL,
    4. et al
    . Human immunodeficiency virus type 1-hepatitis C virus coinfection: intraindividual comparison of cellular immune responses against two persistent viruses. J Virol 2002;76:2817-26.
    Abstract/FREE Full Text
  8. 8.↵
    1. Graham CS,
    2. Curry M,
    3. He Q,
    4. et al
    . Comparison of HCV-specific intrahepatic CD4+ T cells in HIV/HCV versus HCV. Hepatology 2004;40:125-32.
    CrossRefMedlineWeb of Science
  9. 9.↵
    1. Hagan H,
    2. McGough JP,
    3. Thiede H,
    4. Weiss NS,
    5. Hopkins S,
    6. Alexander ER
    . Syringe exchange and risk of infection with hepatitis B and C viruses. Am J Epidemiol 1999;149:203-13.
    Abstract/FREE Full Text
  10. 10.↵
    1. Cook L,
    2. Ng KW,
    3. Bagabag A,
    4. Corey L,
    5. Jerome KR
    . Use of the MagNA pure LC automated nucleic acid extraction system followed by real-time reverse transcription-PCR for ultrasensitive quantitation of hepatitis C virus RNA. J Clin Microbiol 2004;42:4130-6.
    Abstract/FREE Full Text
  11. 11.↵
    1. Davidson F,
    2. Simmonds P,
    3. Ferguson JC,
    4. et al
    . Survey of major genotypes and subtypes of hepatitis C virus using RFLP of sequences amplified from the 5' non-coding region. J Gen Virol 1995;76:1197-204.
    Abstract/FREE Full Text
  12. 12.↵
    1. Li CC,
    2. Seidel KD,
    3. Coombs RW,
    4. Frenkel LM
    . Detection and quantification of human immunodeficiency virus type 1 p24 antigen in dried whole blood and plasma on filter paper stored under various conditions. J Clin Microbiol 2005;43:3901-5.
    Abstract/FREE Full Text
  13. 13.↵
    1. Batts KP,
    2. Ludwig J
    . Chronic hepatitis: an update on terminology and reporting. Am J Surg Pathol 1995;19:1409-17.
    MedlineWeb of Science
  14. 14.↵
    1. Rothman AL,
    2. Morishima C,
    3. Bonkovsky HL,
    4. et al
    . Associations among clinical, immunological, and viral quasispecies measurements in advanced chronic hepatitis C. Hepatology 2005;41:617-25.
    CrossRefMedlineWeb of Science
  15. 15.↵
    1. Riddell SR,
    2. Greenberg PD
    . The use of anti-CD3 and anti-CD28 monoclonal antibodies to clone and expand human antigen-specific T cells. J Immunol Methods 1990;128:189-201.
    CrossRefMedlineWeb of Science
  16. 16.↵
    1. Rushbrook SM,
    2. Ward SM,
    3. Unitt E,
    4. et al
    . Regulatory T cells suppress in vitro proliferation of virus-specific CD8+ T cells during persistent hepatitis C virus infection. J Virol 2005;79:7852-9.
    Abstract/FREE Full Text
  17. 17.↵
    1. Sugimoto K,
    2. Ikeda F,
    3. Stadanlick J,
    4. Nunes FA,
    5. Alter HJ,
    6. Chang KM
    . Suppression of HCV-specific T cells without differential hierarchy demonstrated ex vivo in persistent HCV infection. Hepatology 2003;38:1437-48.
    MedlineWeb of Science
  18. 18.↵
    1. Cabrera R,
    2. Tu Z,
    3. Xu Y,
    4. et al
    . An immunomodulatory role for CD4+ CD25+ regulatory T lymphocytes in hepatitis C virus infection. Hepatology 2004;40:1062-71.
    CrossRefMedlineWeb of Science
  19. 19.↵
    1. Harcourt G,
    2. Gomperts E,
    3. Donfield S,
    4. Klenerman P
    . Diminished frequency of HCV-specific interferon γ secreting CD4+ T cells in human immunodeficiency virus/hepatitis C virus coinfected patients. Gut 2006;55:1484-7.
    Abstract/FREE Full Text
  20. 20.↵
    1. Shoukry NH,
    2. Grakoui A,
    3. Houghton M,
    4. et al
    . Memory CD8+ T cells are required for protection from persistent hepatitis C virus infection. J Exp Med 2003;197:1645-55.
    Abstract/FREE Full Text
  21. 21.↵
    1. Canchis PW,
    2. Yee HT,
    3. Fiel MI,
    4. et al
    . Intrahepatic CD4+ cell depletion in hepatitis C Virus/HIV-coinfected patients. J Acquir Immune Defic Syndr 2004;37:1125-31.
    CrossRefMedlineWeb of Science
  22. 22.↵
    1. Shi Z,
    2. Wakil AE,
    3. Rockey DC
    . Strain-specific differences in mouse hepatic wound healing are mediated by divergent T helper cytokine responses. Proc Natl Acad Sci USA 1997;94:10663-8.
    Abstract/FREE Full Text
  23. 23.↵
    1. Baroni GS,
    2. D'Ambrosio L,
    3. Curto P,
    4. et al
    . Interferon gamma decreases hepatic stellate cell activation and extracellular matrix deposition in rat liver fibrosis. Hepatology 1996;23:1189-99.
    CrossRefMedlineWeb of Science
  24. 24.↵
    1. Graham CS,
    2. Wells A,
    3. Liu T,
    4. et al
    . Antigen-specific immune responses and liver histology in HIV and hepatitis C coinfection. AIDS 2005;19:767-73.
    MedlineWeb of Science
| Table of Contents

This Article

  1. J Infect Dis. (2007) 196 (4): 577-586. doi: 10.1086/519386
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most