The cross-sectional case-control study was carried out in accordance with the guidelines of the International Conference on Harmonization Guidelines and the Declaration of Helsinki. The study protocols were reviewed by the Institutional Ethical Committee (IEC) of the Government Medical College, Theni, for necessary approval for the conduct of the research (Ref. No. 2544/ME1/18 and Ref. No. 1515/MEIII/21). Prior approval was also obtained from the Tamil Nadu State AIDS Control Society, Chennai, India (TANSACS Approval- 00529/TANSACS/M&E/2019 NACO-T-11020/08/2020_NACO) for the conduct of work. Institutional Biosafety Committee (IBSC) approval was secured (Ref. No.: CUTN/SLS/1st IBSC/2020/04). All the human subjects were adults and written consents were duly obtained from all the participants.

HBV-infected individuals with plasma HBsAg and anti-HBc positivity (n = 13), HCV-infected individuals as determined by anti-HCV (n = 8), HIV-infected individuals (as per the criteria of the National AIDS Control Organization (NACO), India) (n = 7), and healthy controls (HCs) (n = 10) were recruited into the cross-sectional study. The HIV-infected patients were on ART as per NACO recommendations (at least for over five years). Peripheral blood was obtained from all the participants by a trained phlebotomist. HCs were identified as individuals free from HBV, HCV, HIV, Mycobacterium tuberculosis infections as well as HBV/HIV co-infections.

Plasma aspartate aminotransferase (AST), alanine aminotransferase (ALT), $\gamma$-glutamyl transferase (GGT), and alkaline phosphatase (ALP) levels were measured on a Semi-automatic Analyzer (Rapid Diagnostics Star 20, Hyderabad, India) using commercial kits procured from TRUEchemie, Brussels, Belgium; AST assay kit (Lot. No. A1921102; TRUEchemie, Brussels, Belgium) with the cut-off values set at 33 U/L, ALT assay kit (Lot. No. A1521101; TRUEchemie, Brussels, Belgium) with the cut-off values set at 35 U/L, GGT assay kit (Lot. No. G1221122; TRUEchemie, Brussels, Belgium), with the cut-off values set at 49 U/L (male) and 32 U/L (female), ALP assay kit (Lot. No. A1321111; TRUEchemie, Brussels, Belgium) with the cut-off values set at 141 U/L, respectively.

HIV infection was diagnosed using the conventional three-kit method advocated by the NACO. According to the manufacturer’s instructions, the rapid immunochromatographic tests Comb Aids-RS (Arkray Healthcare, Mumbai, India), VoXpress HIV-1/2 (Voxtur Bio, Mumbai, India), and Meriscreen HIV 1-2 WB (Mumbai, Merillife, India) were employed for diagnosing HIV infection. At inclusion, all HIV-positive individuals were receiving active antiretroviral therapy (ART). According to the manufacturer’s instructions, anti-CD45-PE-Cy5 (Cat. No. 05-8405-02) and anti-CD4-PE (Cat. No. 05-8405-01) (Sysmex Partec GmbH, Gorlitz, Germany) fluorochrome-tagged antibodies were used for the immunophenotyping. For absolute CD4${}^{+}$ T-cell counts, 2 mL of whole blood in EDTA tubes was used.

The Pathodetect™ (Mylab Discovery, Pune, India) quantitative Real-Time PCR was used to quantify the viral loads of HBV and HCV using an in vitro nucleic acid amplification assay on a QuantStudio 5 real-time PCR (Applied Biosystems, ThermoFisher Scientific, MA, USA). We determined HIV-1 viral load using the Abbott Real-time HIV-1 assay (Abbott, Abbot Park, IL, USA), via an in vitro reverse transcription-polymerase chain reaction assay with the sensitivity 40 copies/mL volume of the sample.

Cytokines were measured using the Bio-plex Pro Human Cytokine Grp I Panel 17-plex kit (BioRad, Hercules, CA, USA) that quantifies the levels of MCP-1, G-CSF, GM-CSF, IL-7, IL-12 (p70), IL-1$\beta$, MIP-1$\beta$, TNF-$\alpha$, CXCL8 (IL-8), IFN-$\gamma$, IL-6, IL-2, IL-4, IL-5, IL-13, IL-17 and IL-10 following the manufacturer’s instructions.

Ten milliliters of peripheral blood were collected by venipuncture, and stored in lithium heparin BD Vacutainer (BD Biosciences, Franklin Lakes, NJ, USA) tubes at room temperature. PBMCs were extracted using a commercial Sepmate™ (Stemcell Technologies, Vancouver, Canada) by density gradient centrifugation. Cell viability was determined by 0.4% Trypan blue vital staining. Purified PBMCs were suspended in a Bambanker™ serum-free cell freezing medium (Nippon Genetics Europe GmbH, Duren, Germany) for storage at –80 °C, not more than 3 months before use in the experiments. PBMCs were thawed in a water bath at 37 °C before use in the experiments.

Mononuclear cells were incubated with PMA (50 ng/mL) and ionomycin (500 ng/mL) or cultured in RPMI containing 10% FBS (R10) alone. Samples were incubated at a concentration of 10 µg/mL, and Golgi Plug (brefeldin A) and Golgi Stop (monensin) were included at 10 µg/mL. Samples were incubated for overnight at 37 °C in 5% CO${}_2$ and then permeabilized using Fix & Perm reagents (BD Bioscience) and stained intracellularly with BV421–conjugated anti–IFN-$\gamma$ and APC–conjugated anti–TNF-$\alpha$. At the end of stimulation, cells were washed once with FACS wash (PBS containing 2% [vol/vol] FBS and 0.25% of sodium azide) and surface stained with anti-CD3, anti-TCR7.2 (3C10), anti-Ki67 (B56) anti-CD8 (SK1), anti-CD4 (OKT4). Cell stain at room temperature for 30 minutes. Cells were then fixed with Cytofix/Cytoperm (BD Pharmingen) for 20 minutes at 4 °C and washed with Perm wash (BD Pharmingen). Cells were then incubated for 30 minutes at 4 °C with antibodies specific to IFN-$\gamma$ and TNF-$\alpha$, washed once with Perm wash and once with FACS wash, and resuspended in PBS containing 1% formalin. Cells were acquired on a BD FACS Canto II Immunocytometric system. FlowJo for Windows, Ver.10.0.8 (FlowJo LLC, Ashland, OR, USA) was used to perform the analysis. At least 100,000 events were acquired for each sample.

We examined the percentages and expression of biomarkers on distinct subsets of T cells, MAIT and Tfh cells between the four study groups. For multiple group comparisons, categorical variables were examined using the Chi-square test of Fisher’s Exact Test, while continuous variables were tested using non–parametric Kruskal–Wallis Test. If the p values were 0.05, four–way comparisons were made between the four patient groups using the Mann–Whitney Test. The Spearman’s Rank correlation was used to compare the correlation between two continuous variables. The association between the surface markers, and functional markers and plasma viral load (PVL) were assessed using the linear regression model. ${}^{*}$p 0.05, ${}^{**}$0.01, ${}^{***}$0.001, and ${}^{****}$0.0001 were used to determine significance. GraphPad Prism Ver.6.0 software (GraphPad, La Jolla, CA, USA) was used to perform all the analyses.

The four groups, non-randomized study design consisted of 38 individuals. Thirteen subjects with chronic HBV infection who tested positive for HBsAg, anti-HBc as well as HBV DNA: Group 1 (G1), eight subjects with HCV RNA positive and anti-HCV positive; G2, seven subjects with HIV RNA positive, and 10 healthy controls (HC) (G4). The samples were collected between September and October of 2021. As per the analytical parameters, 54% HBV-infected individuals, 87% HCV-infected individuals, and 47.5% HIV-infected individuals were diagnosed with signs of liver injury, while 46% of the HBV-infected, 13% of the HCV-infected, and 52.5% of the HIV-infected participants were chronically infected without any underlying clinical or biochemical signs of liver injury (Table 1).

The flow cytometry gating strategy to delineate the CD4${}^{\text{+hi}}$, CD4${}^{\text{+lo}}$ and CD8${}^{+}$ T cells is depicted in Fig. 1A. The CD8${}^{+}$ T cell levels were significantly higher for HIV (Fig. 1B) as expected. There was negligible alteration in CD69, ICOS, or PD-1 (CD279)-expressing T cells between the different groups (Fig. 1C,D). CD69 was significantly increased on CD4${}^{\text{+hi}}$ T cells in chronic HBV, HCV and HIV in comparison with HCs (p 0.05, p 0.01, and p 0.05, respectively). The expression of co-stimulatory (ICOS) and co-inhibitory (PD-1) markers were significantly higher in CD4${}^{\text{+hi}}$ T cells in HBV (p 0.05), HCV-infected (p 0.01) and HIV-infected (p 0.01) individuals. Likewise, CD69${}^{+}$CD4${}^{\text{+lo}}$ T cell levels (p 0.01) PD-1${}^{+}$CD4${}^{\text{+lo}}$ T cell levels and (p 0.001) were significantly higher in all infected groups whereas the levels of ICOS${}^{+}$CD4${}^{\text{+lo}}$ T cells were only significantly higher in chronic HBV (p 0.01) infection. CD69${}^{+}$CD8${}^{+}$ T cell levels were significantly increased in HBV (p 0.05) and HCV-infected individuals (p 0.01). Furthermore, all groups had significantly higher levels of PD-1${}^{+}$CD8${}^{+}$ T cells (Fig. 1C,D). Together, these data highlight that markers of T-cell activation coupled with exhaustion were increasingly expressed on T cells in chronic HBV, HCV and HIV infections.

Fig. 1.

Immune activation and exhaustion markers associated with CD4${}^{\mathrm{+}}$ T cells and CD8${}^{\mathrm{+}}$ T cells. (A) Gating strategy for CD69, ICOS and PD-1 expression on peripheral CD4${}^{\text{+hi}}$, CD4${}^{\text{lo}}$, and CD8${}^{+}$ T cell populations. Lymphocytes were gated from whole human PBMCs using height and area of forward scatter, then singlet gates were utilized to remove doublet populations. This was followed by lymphocyte gating using forward and side scatters areas. This was followed by a total CD3${}^{+}$ cell gate against TCR iV$\alpha$7.2. From CD3${}^{+}$ cells, total CD8${}^{+}$ cells CD4${}^{\text{+hi}}$, and CD4${}^{\text{+lo}}$ were gated out. From this CD8${}^{+}$, CD4${}^{\text{+hi}}$, and CD4${}^{\text{+lo}}$ we determined CD69${}^{+}$, PD-1 and ICOS. (B) The results of these gates are three T cell populations: CD8${}^{+}$, CD4${}^{\text{+hi}}$, and, CD4${}^{\text{+lo}}$. Comparison of the levels of, CD4${}^{\text{+lo}}$, CD4${}^{\text{+hi}}$ and CD8${}^{+}$ among patients chronically infected with HBV, HCV, HIV, and HCs. (C) CD69, PD-1 and ICOS expression was determined by using a CD69, PD-1, and ICOS mean fluorescence intensity (MFI), respectively, which was used as a negative control for CD69, PD-1, and ICOS staining and allowed for accurate gating on the positive populations only. (D) Expression (MFI) of CD69, ICOS, and PD-1 in CD4${}^{\text{+hi}}$, CD4${}^{\text{lo}}$ and CD8${}^{+}$ T cells among patients chronically-infected with HBV, HCV, HIV and HCs. The level of expression of each marker was reflected by the color scale of the heatmap. The cells were compared across the four groups by the Kruskal–Wallis test. Post-hoc Mann–Whitney U tests were subsequently performed only for those biomarkers with a Kruskal–Wallis test p value of 0.05. p 0.05 are considered significant; *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001. n.s. represents ‘not significant’.

Next, we set out to study the MAIT and Tfh cells across the different study groups with a pre-determined gating strategy (Fig. 2A). Total TCR iV$\alpha$7.2${}^{+}$ MAIT cells were significantly lower in all the infected groups (p 0.01) compared with HCs. The CD4${}^{+}$MAIT cell levels were significantly increased in HCV and HBV groups (p 0.01) but not in the HIV-infected group compared to HCs. The level of CD8${}^{+}$MAIT cells (p 0.01) as compared to HCs were higher in all the groups (Fig. 2B). The circulating Tfh cell levels in HIV-infected individuals were significantly higher (p 0.001) in HIV infected group, but comparable in the HBV and HCV-infected groups (Fig. 2C). The CD69${}^{+}$CD4${}^{+}$MAIT cell levels were significantly increased in the HIV and HBV groups (0.05) but not in the HCV-infected group. PD-1${}^{+}$CD4${}^{+}$MAIT cell levels were significantly elevated in all the infected groups (Fig. 2D). The ICOS${}^{+}$CD8${}^{+}$ MAIT cells were significantly decreased for HCV and HBV groups, and the PD-1${}^{+}$CD8${}^{+}$ T cell levels were highly significant in HIV-infected individuals with p 0.001 (Fig. 2D). Spearman correlation of CD4${}^{+}$ TCR iV$\alpha$7.2, CD8${}^{+}$ TCR iV$\alpha$7.2 with CD69, ICOS, and PD-1 is presented in Fig. 2E. The functional markers for Tfh viz., CD27, ICOS and PD-1 were comparable between HBV, HCV and HCs. Nonetheless, PD-1 was highly elevated in the HIV-infected group suggesting that Tfh cells become activated (see Fig. 2F).

Fig. 2.

Expression of immune activation and exhaustion markers on mucosal-associated invariant T cells and follicular T helper cells. (A) Gating strategies for mucosal-associated invariant T cells and follicular T helper cells. Total CD3${}^{+}$ cells were gated against TCR iV$\alpha$7.2 (MAIT cells). From MAIT cells CD8${}^{+}$ and CD4${}^{+}$ T cell populations were gated. From CD3${}^{+}$ cells, CD4${}^{+}$ cells were gated against CXCR5. The different gates were determined: TCR iV$\alpha$7.2, CD4${}^{+}$ TCR iV$\alpha$7.2, CD8${}^{+}$ TCR iV$\alpha$7.2, CD4${}^{+}$ CXCR5. (B) Comparison of the levels of TCR iV$\alpha$7.2, CD4${}^{+}$ TCR iV$\alpha$7.2, CD8${}^{+}$ TCR iV$\alpha$7.2 among individuals chronically infected with HBV, HCV, HIV, and HCs. (C) Comparison of the levels of TFH among patients chronically infected with HBV, HCV, HIV, and HCs. (D) Percentage level of CD4${}^{+}$ TCR iV$\alpha$7.2, CD8${}^{+}$ TCR iV$\alpha$7.2 were compared with CD69, ICOS expression and PD-1. (E) Expression (mean fluorescence intensity, MFI) of CD69, ICOS, and PD-1 in CD4${}^{+}$ TCR iV$\alpha$7.2, CD8${}^{+}$ TCR iV$\alpha$7.2 with among individual chronically infected with HBV, HCV, HIV, and HCs. (F) Percentage levels of TFH were compared with CD27, PD-1, and ICOS expression. The level of expression of each marker was reflected by the color scale of the heatmap. The cells were compared across the four groups by the Kruskal–Wallis test. Post-hoc Mann–Whitney U tests were subsequently performed only for those biomarkers with a Kruskal–Wallis test p value 0.05. p 0.05 are considered significant; *p 0.05, **p 0.01, ***p 0.001. n.s. represents ‘not significant’.

Next, we compared the cytokine-producing ability of T cells with respect to CD4${}^{\text{+hi}}$, CD4${}^{\text{+lo}}$ and CD8${}^{+}$ T cells. Our results indicates that there was a significant increase in the cytokine${}^{+}$ CD4${}^{+}$ T cells and CD8${}^{+}$ T cells during chronic viral infection when compared to HCs. The comparison analysis of CD4${}^{\text{+hi}}$ IFN-$\gamma$ was significant in HBV and HIV-infected individuals (p 0.0001). Ki67${}^{+}$CD4${}^{\text{+hi}}$ was significantly higher in HIV-infected individuals (p 0.0001). CD4${}^{\text{+lo}}$TNF-$\alpha$ was significant in HCV-infected individuals (p 0.001). CD4${}^{\text{+lo}}$IFN-$\gamma$ and Ki67 was highly significant in HIV (p 0.0001) infection. CD8${}^{+}$ IFN-$\gamma$ and Ki67 had significance (p 0.0001) in HBV and HCV-infected subjects, respectively (Fig. 3A,B).

Fig. 3.

Functional markers on conventional T cells and mucosal-associated invariant T cells (A) Gating strategies of intercellular cytokines in conventional and unconventional T cells. Total CD3${}^{+}$ cells were gated against TCR iV$\alpha$7.2 (MAIT cells). From MAIT cells CD8${}^{+}$ and CD4${}^{+}$ T cell populations were gated whereas CD3${}^{+}$ was gated as CD4${}^{\text{+hi}}$, CD4${}^{\text{+lo}}$, and CD8${}^{+}$ T cells. The functional markers of different gates were determined: CD4${}^{+}$ TCR iV$\alpha$7.2, CD8${}^{+}$ TCR iV$\alpha$7.2, CD4${}^{+}$, and CD8${}^{+}$ cells. (B) Comparison of the levels of TNF-$\alpha$, IFN-$\gamma$, and Ki67 in CD4${}^{\text{+hi}}$, CD4${}^{\text{+lo}}$ and CD8${}^{+}$ T cells among individuals chronically-infected with HBV, HCV, HIV, and HCs. (C) Comparison of the levels of CD4${}^{+}$ MAIT cells, and CD8${}^{+}$ MAIT cells among patients chronically-infected with HBV, HCV, HIV, and HCs. (D) The level of expression of TNF-$\alpha$, IFN-$\gamma$, Ki67 with CD4${}^{\text{+hi}}$, CD4${}^{\text{+lo}}$, CD8${}^{+}$, CD4${}^{+}$ TCR iV$\alpha$7.2, CD8${}^{+}$ TCR iV$\alpha$7.2. The level of expression for each marker was reflected by the color scale of the heatmap. The cells were compared across the four groups by the Kruskal–Wallis test. Post-hoc Mann–Whitney U tests were subsequently performed only for those with a Kruskal–Wallis test p value of 0.05. p 0.05 are considered significant; *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001. †, TNF-$\alpha$ did not express in the CD4${}^{\text{hi}}$ population. n.s. represent not significant.

Next, we also looked at the proliferating potential of T cells, which indicated that Ki67${}^{+}$ T cells were higher in chronic viral infections as compared to HCs. Then, we examined the production of cytokines by the CD4${}^{+}$ and CD8${}^{+}$ MAIT cells. Both CD4${}^{+}$ and CD8${}^{+}$ MAIT cells expressed higher levels of TNF-$\alpha$, and IFN-$\gamma$, together with higher proliferating ability as compared to HCs. The proliferating MAIT cells, Ki67${}^{+}$CD4${}^{+}$MAIT and Ki67${}^{+}$CD8${}^{+}$ MAIT cells, had higher significance in HIV and HBV-infected individuals (Fig. 3C) compared to the HCV-infected group. Spearman correlation analysis and the significance of various cytokines in chronic HBV, HCV and HIV-infected individuals are shown in Fig. 3D.

In this study, we observed that the infected groups shared a similar activation and cytokine profiles with total CD4${}^{+}$, CD8${}^{+}$ T cells along with Tfh, and MAIT cells displaying elevated activation and cytokine levels (Fig. 4A,B). The cells that showed $>$2-fold for each chronic infection were identified and displayed in Venn diagrams after the cytokine fold change was graded by descending order (Fig. 4C,D). The analysis revealed that among all the immune cells, seven altered phenotypic profiles viz., CD4${}^{\text{+lo}}$TNF-$\alpha$, MAIT CD4${}^{+}$TNF-$\alpha$, MAIT CD4${}^{+}$IFN-$\gamma$, MAIT CD8${}^{+}$IFN-$\gamma$, CD4${}^{\text{+lo}}$IFN-$\gamma$, CD4${}^{\text{+hi}}$ IFN-$\gamma$, CD8${}^{+}$ IFN-$\gamma$ was common among patients chronically-infected with HBV, HCV, and HIV. Ki67${}^{+}$MAIT CD4${}^{+}$ and CXCR5 CD4${}^{\text{+lo}}$ was common among those chronically infected with HCV and HIV. CD4${}^{+}$MAIT was common among chronic HBV and HCV-infected individuals. (Fig. 4C,D and Supplementary Fig. 1). The activation, exhaustion and functional markers in association with viral load of T-cell subsets are presented in Table 2.

Fig. 4.

Immune cell profiling in chronic HBV-, HCV-, and HIV-infected individuals. (A,B) The fold change in immune cells with identified intracellular and extracellular markers in individuals chronically infected with HBV, HCV and HIV normalized against HCs. (C) Bar plot depicting mean 2-fold change among identified markers in specific immune cell type. (D) Venn diagram showing immune cells that are upregulated $>$2-fold. The Venn diagram identified a profile of seven immune cell populations that are commonly expressed among chronically HBV-, HCV-, and HIV-infected individuals. (v) %CD4${}^{\text{hi}}$; (w) %MAIT CD4${}^{+}$; (x) %CD4${}^{\text{lo}}$ TNF-$\alpha$${}^{+}$, %MAIT CD4${}^{+}$ TNF-$\alpha$${}^{+}$, %MAIT CD4${}^{+}$ IFN-$\gamma$${}^{+}$, %MAIT CD8${}^{+}$ IFN-$\gamma$${}^{+}$, %CD4${}^{\text{hi}}$ IFN-$\gamma$${}^{+}$, %CD4${}^{\text{lo}}$ IFN-$\gamma$${}^{+}$, CD8${}^{+}$ IFN-$\gamma$${}^{+}$; (y) %MAIT CD4${}^{+}$ Ki67${}^{+}$, %CD4${}^{\text{lo}}$ CXCR5${}^{+}$; (z) %CD4${}^{\text{hi}}$ CXCR5${}^{+}$, %MAIT CD8${}^{+}$ TNF-$\alpha$${}^{+}$.

We performed a network analysis for the six predictors of PVL, viz., CD4${}^{\text{+lo}}$TNF-$\alpha$, CD69${}^{+}$CD8${}^{+}$, CD69${}^{+}$CD4${}^{+}$MAIT, PD-1${}^{+}$CD4${}^{\text{+hi}}$, PD-1${}^{+}$CD8${}^{+}$, Ki67${}^{+}$CD4${}^{+}$MAIT cells (Fig. 5A). CD69${}^{+}$CD8${}^{+}$ showed a positive correlation with CD69${}^{+}$ CD4${}^{+}$MAIT, PD-1${}^{+}$CD8${}^{+}$, and a negative correlation with CD4${}^{\text{+lo}}$TNF-$\alpha$. Further, CD69${}^{+}$CD4${}^{+}$MAIT had a positive correlation with PD-1${}^{+}$CD8${}^{+}$ and a negative correlation with CD4${}^{\text{+lo}}$TNF-$\alpha$. Moreover, PD-1${}^{+}$CD4${}^{\text{+hi}}$ correlated positively with Ki67${}^{+}$CD4${}^{+}$ and CD4${}^{\text{+lo}}$TNF-$\alpha$. We also found that PD-1${}^{+}$CD8${}^{+}$ were positively correlated with Ki67${}^{+}$CD4${}^{+}$. Ki67${}^{+}$CD4${}^{+}$MAIT were negatively correlated with CD4${}^{\text{+lo}}$TNF-$\alpha$. In short, CD4${}^{\text{+lo}}$TNF-$\alpha$ had a positive correlation with PD-1${}^{+}$CD4${}^{\text{+hi}}$ and a negative correlation with CD69${}^{+}$CD8${}^{+}$, CD69${}^{+}$ CD4${}^{+}$MAIT, and Ki67${}^{+}$CD4${}^{+}$MAIT (Fig. 5B). Together, we found that the markers associated with immune activation, proliferation and exhaustion were differentially expressed and correlated with the different T cell phenotypes.

Fig. 5.

Predictors of plasma viral load. (A,B) Network analysis of the six predictors of plasma viral load. (A) depicts the complexity of the interactions between the six predictors. (B) Spearman correlation between the six predictors of plasma viral load. (C) The six markers were subjected to multivariate linear regression analysis to determine the markers that independently predict the plasma viral load. (D) Expression of TNF-$\alpha$ and their association with plasma cytokines. (E) Plasma level of cytokines associated with plasma viral load. Variables with p values 0.05 were considered independent predictors in their respective models. *, **, **** represent p 0.05, 0.01, and 0.0001, respectively.

Next, we performed multivariate linear regression analysis among the six subsets of T cells to determine independent prediction over the PVLs of HBV, HCV and HIV-infected individuals. The multivariate linear regression analysis of cellular markers were as follows: CD69${}^{+}$CD8${}^{+}$, –7.24 (p = 0.014), CD69${}^{+}$CD4${}^{+}$MAIT, –13.98 (p = 0.0001), PD-1${}^{+}$CD4${}^{\text{+hi}}$, 0.04 (p = 0.011), PD-1${}^{+}$CD8${}^{+}$, 9.99 (p = 0.013), Ki67${}^{+}$CD4${}^{+}$MAIT, 3.42 (p = 0.009) and CD4${}^{\text{+lo}}$TNF-$\alpha$, –3.75 (p = 0.023) (Fig. 5C). We also determined the expression of CD4${}^{\text{+lo}}$TNF-$\alpha$ and its association with plasma cytokines. We observed that CD4${}^{\text{+lo}}$TNF-$\alpha$ was significant among almost all the cytokines (Fig. 5D). The PVLs across all the three viral infections were associated with various cytokines of which 13 revealed positive correlations (Fig. 5E). These 13 cytokines, in turn, had an inverse correlation with the PVL. In HBV PVL IL-5, IL-6, and IL-17 were significant (p 0.01), and IL-2 (p 0.05) with a negative correlation. In comparison with HCV PVL, IFN-$\gamma$ was significant (p 0.01); IL-6 and IL-13 also revealed significance (p 0.05) with a negative correlation. In the case of HIV PVL, G-CSF, IL-1$\beta$, IL-2, IFN-$\gamma$, IL-6, and TNF-$\alpha$ were negatively significant (p 0.01). Among all the infected individuals IL-17, and MIP-1$\beta$ were inversely significant (p 0.01): IL-8, IL-10, TNF-$\alpha$ (p 0.05). Together, our data suggest that the inverse correlation of cellular markers may independently play a vital role in determining the plasma viral load of chronic HBV, HCV and HIV-infected patients.