Expression of programmed cell death protein 1 and T‐cell immunoglobulin‐ and mucin‐domain‐containing molecule‐3 on peripheral blood CD4+CD8+ double positive T cells in patients with chronic hepatitis C virus infection and in subjects who spontaneously cleared the virus

Abstract Chronic hepatitis C virus (HCV) infection is characterized by increased proportion of CD4+CD8+ double positive (DP) T cells, but their role in this infection is unclear. In chronic hepatitis C, immune responses to HCV become functionally exhausted, which manifests itself by increased expression of programmed cell death protein 1 (PD‐1) and T‐cell immunoglobulin‐ and mucin‐domain‐containing molecule‐3 (Tim‐3) on T cells. The aim of our study was to determine PD‐1 and Tim‐3 phenotype of DP T cells in subjects with naturally resolved and chronic HCV infection. Peripheral blood mononuclear cells from 16 patients with chronic infection and 14 subjects who cleared HCV in the past were stained with anti‐CD3, anti‐CD4, anti‐CD8, anti‐PD‐1 and anti‐Tim‐3 antibodies and, in 12 HLA‐A*02‐positive subjects, MHC class I pentamer with HCV NS31406 epitope. In chronic and past HCV infection, proportions of total DP T cells and PD‐1+ DP T cells were similar but significantly higher than in healthy controls. DP T cells were more likely to be PD‐1+ than either CD4+ or CD8+ single positive (SP) T cells. HCV‐specific cells were present in higher proportions among DP T cells than among CD8+ SP T cells in both patient groups. Furthermore, while the majority of HCV‐specific DP T cells were PD‐1+, the proportion of HCV‐specific CD8+ T cells which were PD‐1+ was 4.9 and 1.9 times lower (chronic and past infection, respectively). PD‐1 and Tim‐3 were predominantly expressed on CD4highCD8low and CD4lowCD8high cells, respectively, and co‐expression of both markers was uncommon.


| INTRODUC TI ON
Adaptive immune responses play a critical role in the outcome of hepatitis C virus (HCV) infection. 1 Strong and HCV-specific CD8+ and CD4+ cellular immunity is indispensable for the spontaneous clearance of HCV infection, which is observed in 20%-50% of patients with newly acquired infection but is rare in patients with an already established chronic infection. [2][3][4][5] Suppression of various mechanisms resulting in suboptimal virusspecific T-cell responses has been described for a number of chronic viral infections such as hepatitis B virus (HBV), 6-8 HCV, 9 lymphocytic choriomeningitis virus (LCMV) 10 and human immunodeficiency virus (HIV). [11][12][13] One of the major mechanisms driving the persistence of HCV infection is T-cell exhaustion which results in weak antigen-specific T-cell responses. 14,15 These defects, which are due to continuous antigen stimulation, progress with the duration of infection and are accompanied by increased expression of inhibitory molecules such as programmed cell death protein 1 (PD-1) and T-cell immunoglobulinand mucin-domain-containing molecule-3 (Tim-3). [15][16][17] The existence of peripheral blood CD4+ CD8+ double positive (DP) T cells was described both in humans and animals such as rats, mice, chickens, monkeys and swine. 18,19 Two subpopulations of these cells are distinguished: CD4 high CD8 low and CD4 low CD8 high , reflecting the predominance of either CD4 or CD8 expression on their surface. 20 Although the origin of these cells is still debatable, dominant expression (ie 99%) of CD8αβ heterodimer rather than CD8αα homodimer on their surface implies they are derived from thymic and not gut environment. 20 DP T cells are mostly functional/effector memory cells specific for antigens of pathogens encountered throughout life. 21 In healthy blood donors, these cells constitute around 1% of CD3+ cells]. 21 However, increased frequencies of up to 20% were reported in chronic viral infections and these cells were found to represent highly proliferative and active population expressing FasL and IFN-γ. [22][23][24][25] Due to the activated phenotype exhibited by DP T cells, they could be hypothetically prone to apoptosis. However, levels of TUNEL expression in DP T cells and CD8 + T cells were reported to be similar. 20 DP T cells (predominantly CD4 high CD8 low ) are commonly found in peripheral blood and liver in patients with chronic HCV infection. 21,26 When compared to their single positive (SP) CD4+ or CD8+ counterparts, DP T cells were found to display shorter telomeres indicating they experienced more cell divisions. 21 In the available animal model of HCV infection (chimpanzee), the proportion of DP T cells negatively correlated with serum viral load, suggesting their involvement in the control of HCV infection. 21 Information on DP T cells in HCV infection is still limited, however. In particular, the exhaustion phenotype of these cells, their specificity towards HCV antigens and their prevalence in subjects who spontaneously recovered from HCV infection have not been reported.
In our study, we analysed exhaustion markers expression on total and HCV-specific DP T cells in patients with chronic HCV infection, in subjects who spontaneously cleared HCV infection, and in healthy controls.

| Patients
Sixteen patients with recently diagnosed chronic HCV infection and 14 subjects who spontaneously cleared HCV infection in the past were recruited among outpatients of the Warsaw Hospital for Infectious Diseases. The latter patients were referred because of anti-HCV-positive status but were eventually found to be HCV-RNA-negative and without any evidence of liver disease. Eleven healthy anti-HCV-negative volunteers with no evidence of liver disease (normal ALT activity levels) served as controls. Peripheral blood mononuclear cells (PBMCs) were isolated from 10 mL of EDTA-anticoagulated blood by density gradient centrifugation (Lymphoprep, Stemcell Technologies Inc

| HLA-A*02 typing
The presence of HLA-A*02 allele was verified by flow cytometry using anti-HLA-A*02-FITC antibody (BD Pharmingen) and by quantitative PCR as described elsewhere. 28

| Statistical analysis
Differences in the proportions of cell subpopulations were analysed by Mann-Whitney test. All P-values were two-tailed and considered significant when <0.05.

| Ethical statement
The study was approved by the Bioethical Committee of the Medical University of Warsaw (KB/247/2013), and all subjects provided written informed consent.

| CD4+ CD8+ DP T cells
The proportion of DP T cells in peripheral blood were similar in patients with chronic infection and in subjects who spontaneously cleared HCV (median 1.9% of CD3+ cells vs 1.8%; NS) but significantly higher than in healthy controls (median 0.7%, P < 0.01); ( Table 2). Within DP T cells, the population of CD4 high CD8 low cells was 2.7-fold higher (mean) than the population of CD4 low CD8 high cells in both groups of patients. However, in healthy controls, it was the population of CD4 low CD8 high cells that predominated ( Table 2).
Examples of dot plots of DP T-cell populations in two patients with chronic HCV infection are shown in Figure 2.

| PD-1 and Tim-3 phenotype of CD4+ CD8+ DP T cells and CD4+/CD8+ SP T cells
The representing expression of PD-1 and Tim-3 on gated DP T cells significantly higher than among healthy controls (median 24.6%, P < 0.01; Figure 3A). Furthermore, expression of PD-1 both in chronic and past HCV infection was more common on DP T cells than on CD4+ or CD8+ SP T cells (P < 0.01). However, this was not the case in healthy controls.
The proportion of DP T cells expressing Tim-3 was similar in all groups including healthy controls and higher than the proportion of SP CD4+ cells expressing Tim-3 (P < 0.01; Figure 3B). However, the proportion of both SP CD4+ and SP CD8+ cells expressing Tim-3 was significantly higher in chronic than in past infection (P < 0.01; Figure 3B).
Co-expression of both PD-1 and Tim-3 exhaustion markers on CD4+ and CD8+ SP T cells was significantly more common in chronic infection than either in past infection or in healthy controls (P < 0.05; Figure 3C). Furthermore, DP T cells were more likely to be PD-1+ Tim-3+ than either SP CD4+ or CD8+ cells, although not all differences were statistically significant ( Figure 3C).

| PD-1 and Tim-3 phenotype of CD4 high CD8 low and CD4 low CD8 high DP T-cell subpopulations
Programmed cell death protein 1 was present on higher proportion of CD4 high CD8 low DP T cells than CD4 low CD8 high DP T cells both in chronic (median 46.7% vs 19.3%; P < 0.0001) and past (median 44.2% vs 28.6%; P = 0.003) infection ( Figure 4A). Conversely, Tim-3 was present on higher proportion of CD4 low CD8 high DP T cells compared to CD4 high CD8 low DP T cells, both in chronic (median 3.1% vs 0.7%; P = 0.009) and past (median 1.4% vs 0.6%; P = 0.0035) infection ( Figure 4B). These differences were not present among healthy controls.
There were no significant differences between chronic and past HCV infection in the proportions of PD-1+ or Tim-3+ CD4 high CD8 low and CD4 low CD8 high populations. However, PD-1+ CD4 high CD8 low cells were significantly more prevalent in chronic infection than in healthy controls. Furthermore, there were no significant differences between chronic and past HCV infection in the proportions of PD-1+ Tim-3+ CD4 high CD8 low and CD4 low CD8 high populations.

| PD-1 exhaustion phenotype of HCV-specific DP T cells and CD8+ SP T cells
Hepatitis C virus-specific cells were analysed in 12 patients with HLA-A*02 allele: three were chronically infected, while nine spontaneously recovered from HCV infection (Table 3). In chronic infection, 0.2%-2.5% of DP T cells and 0.006%-0.098% of CD8+ SP T cells were specific for HCV NS3 1406 epitope KLSGLGLNAV. The HCV-specific DP T cells were found to be mainly PD-1+ (mean 51.3%), while only 10.5% of HCV-specific CD8+ SP T cells were PD-1+.
In subjects who recovered from HCV infection, 0.17%-1.8% of DP T cells and 0.003%-0.163% of CD8+ SP T cells were reactive to HCV NS3 1406 epitope. Again, the HCV-specific DP T cells were found to be mostly PD-1+ cells (mean 54.9%) which was in contrast to HCV-specific CD8+ SP cells, of which only 28.2% were PD-1+.
An analysis of DP T-cell population specific for HLA-A*02-restricted HCV NS3 1406 KLSGLGLNAV epitope together with PD-1 expression in patient #29 is shown in Figure 5.

| D ISCUSS I ON
In the current study, we characterized the population of CD4+ and reverse proportion of CD4 high CD8 low subsets found in healthy controls imply that the observed effects were indeed due to HCV exposure. The proportions of DP T cells expressing PD-1 and/or Tim-3 were found to be similar in chronic and past HCV infection which suggests that expression of these markers on these cells is long-lasting and perhaps irreversible. Since healthy controls had significantly lower PD-1 expression, PD-1 expression on DP T cells seems to be highly affected by exposure to HCV.
There are no published data regarding the PD-1 and Tim-3 phenotype of DP T cells but a number of studies analysed SP T cells.
Thus, studies on chronic HIV infection suggest that PD-1 is a marker of early T-cell exhaustion, representing a stage of already impaired proliferation, but still relatively well-preserved T-cell function, including cytokine production. 29 In contrast, Tim-3 seems to be a marker of more advanced T-cell exhaustion associated with HIV disease progression 30 and was also shown to be up-regulated in HCV infection. 31 In our study, the proportions of CD8+ SP T cells expressing Tim-3 were higher in chronic than in past infection, which suggests that the expression of Tim-3 on these cells is maintained for the du- Interestingly, we found that CD4 high CD8 low and CD4 low CD8 high subsets differed with respect to exhaustion markers expres-

| CON CLUS IONS
In summary, we found total DP T-cell proportions and PD-1+ DP Tcell subpopulations to be similar in chronic and past HCV infection but higher than in healthy controls. Furthermore, the expression of PD-1 on total DP T cells was higher than on either CD4+ or CD8+ SP T cells. HCV-specific cells were of higher frequency among DP T cells and were more likely to express PD-1 when compared to HCV-specific CD8+ SP T cells. Our findings suggest that DP T cells play an important, perhaps critical role in the immune response to HCV, but at the same time seem to be prone to incur an inhibitory phenotype.

CO N FLI C T O F I NTE R E S T S
None declared.