SHMT2 regulates CD8+ T cell senescence via the reactive oxygen species axis in HIV-1 infected patients on antiretroviral therapy

Summary

Background

Although antiretroviral therapy (ART) effectively inhibits viral replication, it does not fully mitigate the immunosenescence instigated by HIV infection. Cellular metabolism regulates cellular differentiation, survival, and senescence. Serine hydroxymethyltransferase 2 (SHMT2) is the first key enzyme for the entry of serine into the mitochondria from the de novo synthesis pathway that orchestrates its conversion glutathione (GSH), a key molecule in neutralising ROS and ensuring the stability of the immune system. It remains incompletely understood whether SHMT2 is involved in the senescence of CD8+ T cells, crucial for immune vigilance against HIV.

Methods

HIV-infected individuals receiving antiretroviral therapy were enrolled in our study. SHMT2-siRNA was electroporated into T cells to disrupt the gene expression of SHMT2, followed by the quantification of mRNA levels of crucial serine metabolism enzymes using real-time PCR. Immunophenotyping, proliferation, cellular and mitochondrial function, and senescence-associated signalling pathways were examined using flow cytometry in CD8+ T cell subsets.

Findings

Our findings revealed that CD8+ T cells in HIV-infected subjects are inclined towards senescence, and we identified that SHMT2, a key enzyme in serine metabolism, plays a role in CD8+ T cell senescence. SHMT2 can regulate glutathione (GSH) synthesis and protect mitochondrial function, thus effectively controlling intracellular reactive oxygen species (ROS) levels. Moreover, SHMT2 significantly contributes to averting immunosenescence and sustaining CD8+ T cell competence by modulating downstream DNA damage and phosphorylation cascades in pathways intricately linked to cellular senescence. Additionally, our study identified glycine can ameliorate CD8+ T cell senescence in HIV-infected individuals.

Interpretation

Decreased SHMT2 levels in HIV-infected CD8+ T cells affect ROS levels by altering mitochondrial function and GSH content. Increased ROS levels activate senescence-related signalling pathways in the nucleus. However, glycine supplementation counteracts these effects and moderates senescence.

Funding

This study was supported by grants from the National Key R&D Program of China (2021YFC2301900-2021YFC2301901), National Natural Science Foundation of China (82372240), and Department of Science and Technology of Liaoning Province Project for the High-Quality Scientific and Technological Development of China Medical University (2022JH2/20200074).

Research in context.

Evidence before this study

HIV infection leads to the early CD8+T cells cellular senescence despite antiretroviral therapy (ART), and the mechanism remains unclear. Cellular metabolism regulates cellular differentiation, survival and senescence. Amino acid metabolism has been demonstrated to be important in several senescence-related illnesses. Serine hydroxymethyltransferase 2 (SHMT2) is the first key enzyme for the entry of serine into the mitochondria from the de novo synthesis pathway that orchestrates its conversion glutathione (GSH), a key molecule in neutralising reactive oxygen species (ROS) and ensuring the stability of the immune system.

Added value of this study

Our study investigated the expression of SHMT2 in CD8+ T cells of HIV-infected individuals and its effect on cellular senescence and, related mechanisms. We found that the decreased expression of SHMT2 in CD8+ T cells of HIV-infected individuals affects the antioxidant capacity of glutathione (GSH) and leads to an increase in the levels of intracellular ROS. Ultimately, this leads to DNA damage and the phosphorylation cascade associated with γH2AX and p53 in the nucleus. Glycine, a direct metabolic offshoot of SHMT2, can alleviate CD8+ T cell senescence in HIV-infected individuals.

Implications of all the available evidence

These data highlight SHMT2 as a novel metabolic enzyme that affects cellular senescence, with anti-senescence effects on CD8+ T cells in HIV-infected individuals. Glycine, a metabolite of the SHMT2 enzyme, holds considerable potential as an agent to arrest CD8+ T cell senescence in HIV infection.

Introduction

Following HIV infection, a state of consistent chronic immune activation and inflammation often culminate in immune senescence.1, 2, 3 While antiretroviral therapy (ART) is capable of inhibiting viral replication, it falls short of fully rectifying the immunosenescence triggered by HIV infection.^4,5 CD8+ T cells are pivotal in the immune surveillance of HIV-infected patients.⁶ Their susceptibility to senescence is greater than that of CD4+ T cells,⁷ and leads to diminished function and compromised immune surveillance which has been identified as a significant precursor to serious non-AIDS events (SNAEs).⁸ Despite considerable research efforts, the underlying mechanism by which CD8+ T cells reach senescence has not been fully illustrated and efficacious intervention strategies need to be investigated.

Studies have revealed that the senescence of CD8+ T cells is intimately entwined with alterations in metabolic substances,⁹ including upregulated glucose metabolism,¹⁰ lipid metabolism,¹¹ and amino acid metabolism dysregulation.¹² The serine metabolic pathway is pivotal in amino acid metabolism, encompassing several subprocesses including de novo synthesis, the folate cycle (one-carbon metabolism), the methionine cycle, and the transsulphuration pathway. This metabolic cascade is indispensable for DNA synthesis and repair, oxidative stress regulation, DNA methylation, and other vital processes intricately linked with senescence.¹³ Amino acid metabolism has been shown to be important in several senescence-related illnesses: branched-chain amino acids have been implicated in accelerating cellular senescence through mTOR signalling and enhancing tumour suppression,¹⁴ while the inhibition of glutamine catabolism has been observed to selectively clear senescent cells, thereby improving age-related organ dysfunction.¹⁵ Additionally, serine metabolism has been associated with attenuated senescence in dental pulp stem cells,¹⁶ and cysteine supplementation has been observed to stimulate the activation of the p53-p21 pathway leading to melanoma senescence.¹⁷ However, the correlation between amino acid metabolism and CD8+ T cell senescence remains ambiguous. Consequently, the metabolic routes linked to senescence in HIV-infected CD8+ T cells, including the pivotal metabolic enzymes driving senescence, must be further investigated.

Serine hydroxymethyltransferase 2 (SHMT2) is the first key enzyme for the entry of serine into the mitochondria from the de novo synthesis pathway.¹⁸ SHMT2 orchestrates the conversion of serine to glycine, subsequently producing glutathione (GSH), a key molecule in neutralising ROS and ensuring the stability of the immune system.¹⁹ Additionally, in the context of Jurkat cell lines, SHMT2 influences mitochondrial functionality, playing a role in tRNA synthesis.²⁰ SHMT2's activity is also critical to T cell proliferation.²¹ Interestingly, diminished activation of senescent CD8+ T cells correlates with a reduction in genes linked to SHMT2, and the observed deficiencies in senescent T cells can be offset with glycine supplementation.²² Notably, while these observations are compelling, concrete evidence delineating the relationship between SHMT2 and cellular senescence remains elusive.

In this study, we demonstrated that CD8+ T cells from HIV-infected individuals exhibit an increased propensity towards senescence, accompanied by a notable downregulation of SHMT2 expression. A decrease in SHMT2 expression in CD8+ T cells leads to a decrease in GSH and abnormal mitochondrial function. This causes an increase in intracellular ROS levels, which in turn leads to the activation of ROS-DDR-p53 and ROS-p16, leading to senescence. To counteract cellular senescence and functional decline, we administered glycine, a downstream product of SHMT2. Collectively, our findings underscore the important role of SHMT2 in driving the senescence of CD8+ T cells.

Methods

Study subjects

In this study, HIV-infected individuals who had been receiving suppressive ART and HIV-negative individuals without immune system-related diseases were included. The subjects were recruited from the First Hospital of China Medical University. 148 HIV-infected individuals and 93 HIV-negative controls were enrolled for the in vivo study as shown by the data in Figs. 1 and 2 and Supplementary Fig. S1. Among these HIV-infected individuals there were 145 males and 3 females (Mean ± SD: age, 40 ± 12 years, CD4+ T cell, 593 ± 245 cells/μL; duration on ART, 70 ± 37 months. Viral load<50 copies/mL).

Fig. 1.

Phenotypic detection of senescencein T cell subsets. (a and b) The levels of KLRG1 expression in CD4+ and CD8+ T cells from HIV negative controls (n = 11) and HIV-infected individuals (n = 32). (c and d) Correlation analysis of the percentages of CD4+KLRG1+ and CD8+KLRG1+ T cells with CD4+ T cells count from HIV-infected individuals (n = 22). (e–g) Representative data and summary graphs for KLRG1 expression in CD4+ and CD8+ T cell subsets (N, CM, EM, and EMRA) from HIV-infected individuals (n = 31). Data are analysed by Mann–Whitney U test in a-b; Spearman linear regression test in c-d; Wilcoxon signed rank test in g. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001. HC, HIV negative controls; HIV, HIV-infected individuals; N, naive; CM, central memory; EM, effector memory; EMRA, terminally differentiated effector memory cells.

Fig. 2.

SHMT2 gene expression is negatively correlated with senescence in HIV-infected individuals' CD8+ T cells. (a–c) Correlation analysis of the percentages of CD98 (n = 25), GLUT1 (n = 20) and CD36 (n = 20) with KLRG1 in CD8+ T cells from HIV-infected individuals. (d) The expression of SHMT2 mRNA in PBMC from HIV negative controls (n = 20) and HIV-infected individuals (n = 20). (e) The expression of SHMT1 mRNA in PBMC from HIV negative controls (n = 20) and HIV-infected individuals (n = 20). (f) The expression of PHGDH mRNA in PBMC from HIV negative controls (n = 20) and HIV-infected individuals (n = 20). (g) The expression of SHMT2 mRNA (left) in CD8+ T cells from HIV negative controls (n = 22) and HIV-infected individuals (n = 22). The expression of SHMT2 MFI (right) in CD8+ T cells from HIV negative controls (n = 21) and HIV-infected individuals (n = 28). (h) The expression of SHMT2 mRNA (left) in CD4+ T cells from HIV negative controls (n = 30) and HIV-infected individuals (n = 26). The expression of SHMT2 MFI (right) in CD4+ T cells from HIV negative controls (n = 21) and HIV-infected individuals (n = 28). (i) Correlation analysis of the percentages of KLRG1 with SHMT2 mRNA expression in CD8+ T cells from HIV negative controls (n = 10) and HIV-infected individuals (n = 10). Data are analysed by Spearman linear regression test in a-c and i; Mann–Whitney U test in d-h. ∗p < 0.05. HC, HIV negative controls; HIV, HIV-infected individuals.

Immune cell purification and culture

Peripheral blood was collected and processed using density gradient centrifugation to isolate peripheral blood mononuclear cells (PBMCs). CD3+, CD4+, and CD8+ T lymphocytes were enriched by negative selection using according Human T cell, CD4+ or CD8+ T cell isolation kit (STEMCELL, Canada, cat#17951, cat#17952, cat#17953) according to the manufacturer’s protocol depending on different assay requirements. For experiments involving exogenous glycine or ROS intervention, glycine (Beyotime, China, cat#ST085) or N-AC (MCE, USA, cat#HY-B0215) were introduced. The sorted cells were stimulated with Dynabeads Human T-Activator CD3/CD28 (Thermo Fisher, USA, cat#11131D) in the presence or absence of glycine or N-AC for 24 h. All primary PBMCs or lymphocytes were incubated at 37 °C, 5% CO2 in RPMI 1640 medium (Hyclone, USA, cat#11875119) containing 10% fetal bovine serum (FBS) (LONSERA, China, cat#A511-001) and 1% penicillin and streptomycin mixture (TBD Science, China, cat#PS2004HY).

Electrotransfection

T cells were transfected using a Human T Cell Nucleofector™ Kit (Lonza, Switzerland, cat#VPA-1002) and an electroporation apparatus (Lonza, Switzerland, Nucleofector 2b device), following the manufacturer’s protocol. For knockdown assays, 100 pmol of either siRNA targeting SHMT2 (Thermo Fisher, USA, cat#4392420-s12823) or a negative control siRNA (Thermo Fisher, USA, cat#4390843) were electroporated into human primary T cells. Transfections were performed using the U-014 electroporation program on the Nucleofector 2b device. 24 h following the completion of the electrotransfection of the cells, T cells were stimulated with anti-CD3/CD28 Dynabeads (6 μL/10⁶ cells) for 72 or 24 h at 37 °C in 5% CO₂ in 96-well plates for proliferation detection or other subsequent treatments.

For overexpression assays, 3 μg of either an empty vector or SHMT2 overexpression plasmid (Hanbio Biotechnology) were electroporated into human primary T cells. Forty-eight hours after transfection, T cells were stimulated with anti-CD3/CD28 Dynabeads (6 μL/10⁶ cells) for 24 h at 37 °C in 5% CO₂ and cultured with subsequent treatments as appropriate.

Immune profiling by flow cytometry

Phenotype analysis was performed by flow cytometry on freshly isolated lymphocytes. T cell phenotype was assessed using the following fluorochrome-conjugated monoclonal antibodies: PerCp-cy5.5-CD3 (RRID:AB_893299), APC-cy7-CD4 (RRID:AB_314085), FITC-CD8 (RRID:AB_1877178), PE-cy7-CD45RA (RRID:AB_10708879), APC-CCR7 (RRID:AB_10915474), PE-KLRG1 (RRID:AB_2572156), APC-CD36 (RRID:AB_398480), PE-GLUT1 (R&D, USA, cat#FAB1418P), BV-510-CD98 (RRID: AB_27422788). Monoclonal antibodies were added to cell, placed in the dark at 4 °C for 20 min, then washed with Phosphate Buffered Saline (PBS).

For examining intracellular SHMT2, cells were first labelled with LIVE/DEAD stain and cell surface protein antibodies as described above, then fixed with 4% paraformaldehyde for 10 min at room temperature, followed by permeabilisation with 90% pre-cooled methanol on ice. Cells were stained with SHMT2 rabbit monoclonal primary antibody (Abcam, UK, cat#ab316328), isotype control (Abcam, UK, cat#ab172730), or an unlabelled control for 1 h at room temperature, followed by staining with goat anti-rabbit IgG AF647 secondary antibody (Thermo Fisher, USA, cat#A32733) for 30 min at room temperature.

Detection of the T cell phenotype after SHMT2 intervention or glycine intervention was performed by flow cytometry on treated and cultured cells. The following monoclonal antibodies were used: PE-CD28 (RRID:AB_400368), APC-KLRG1 (RRID:AB_10641560), BV-421-CD57 (RRID:AB_2562459), APC-PD-1 (RRID:AB_940473), BV421-TIM3 (RRID:AB_2632853), FITC-TIGIT (RRID:AB_2572530), PE-CTLA4 (RRID:AB_2566796), PerCP-7-AAD (BD Biosciences, USA, cat#559925). Cell surface markers were stained as above, followed by the addition of 7-AAD for 2 min at room temperature, then assayed on the machine. To detect nuclear proteins, including PE-cy7-ki-67 (RRID:AB_2562871), PE-p53 (RRID:AB_2896863), FITC-p16 (RRID:AB_396464), and AF647-γH2AX (RRID:AB_2114994), cells were stained with LIVE/DEAD™ Fixable Aqua Dead Cell Stain Kit (Invitrogen, USA, Cat# L34966), antibodies for cell surface markers, fixed and permeabilised using the Transcription Factor Buffer Set (RRID:AB_2869424), and then stained intra-nucleus with the above mentioned intracellular antibodies. To detect cell surface proteins, PE-CD107a (RRID:AB_396135), and intracellular proteins APC-IFN-γ (RRID:AB_315443), PE-cy7-TNF-α (RRID:AB_2204079) and BV510-Granzyme-B (RRID:AB_2738174), antibody to CD107a was added to the medium 12 h before cell harvest, and the additional antibodies were added following the same protocol as described for nuclear staining, except for using the Fixation/Permeabilization Solution Kit (RRID:AB_2869010).

For apoptosis analysis, cells were first treated and collected, followed by preliminary surface staining with specific antibodies. The cells were then resuspended in Binding Buffer containing Annexin V (RRID:AB_2561298) and stained for 20 min. After washing off the antibodies, the cells were resuspended in Binding Buffer containing 7-AAD and stained for 2 min, then immediately analysed on a flow cytometer. For all in vitro assays, samples were run in LSR II cytometer or FACSCanto™ II cytometer (BD Biosciences, USA), and data were analysed with FacsDiva™ and FlowJo™.

Cell proliferation assay

Following the manufacturer’s instructions, cells were labelled using the CellTrace™ Violet Cell Proliferation Kit (Thermo Fisher, USA, Cat#C34557) 24 h after electroporation. The cells were then cultured under the Dynabeads human T-activator CD3/CD28 stimulation for 72 h. For the glycine supplementation experiment, PBMCs from HIV-infected individuals were labelled with the CellTrace™ Violet Cell Proliferation Kit (Thermo Fisher). The cells were then cultured for 72 h under the stimulation of soluble 5 μg/mL anti-CD3 antibody (RRID:AB_395736) and 1 μg/mL anti-CD28 antibody (RRID:AB_396068), with or without the addition of glycine. After collecting the cells, surface staining and 7-AAD detection were performed, followed by flow cytometry analysis.

Mitochondria assays

T cells were resuspended with 25 nM MitoTracker™ Orange CMTMRos (MTO, Thermo Fisher, USA, cat#M7510) and 50 nM MitoTracker™ Green FM (MTG, Thermo Fisher, USA, cat#M7514). For the detection of superoxide levels in mitochondria, T cells were resuspended with 5 mM MitoSOX™ (Invitrogen, USA, cat#M36008). The cells were then incubated at 37 °C for 30 min in the dark. Following this, dead cells were stained using the LIVE/DEAD™ Fixable Aqua Dead Cell Stain Kit, and cell surface labelling was performed with APC-CD8 (RRID:AB_2869906).

Measurement of GSH and GSSG

The GSH/GSSG ratio was evaluated using the GSH and GSSG Assay Kit (Beyotime, China, cat#S0053) in accordance with the manufacturer's instructions. Briefly, CD8+ T cells isolated from HIV-infected individuals were transfected with SHMT2-siRNA or negative control for 24 h. Following treatment, the GSH and GSSG of 0.5 million CD8+ T cells were measured according to the instructions of the kit.

RNA extraction and quantitative real-time PCR

To detect gene expression of SHMT2, SHMT1, PHGDH, CD57, MKI67 and GAPDH, total RNA was isolated by RNeasy Micro kit (Qiagen, Germany, cat#74034). Then the RNA was reverse transcribed using PrimScript™ RT reagent kit (TAKARA, Japan, cat#RR037A) according to the instructions provided by the manufacturer. The real-time PCR reactions for the detection of mRNA were performed using TB Green Premix Ex Taq II (Takara, Japan, cat#RR820B), with according primer sets (Beijing Genomics Institute, BGI). Changes in mRNA expression were calculated using the 2^−ΔΔCt method.²³

Gene Expression Omnibus database analysis

Gene expression profiles from 3 pairs of mice SHMT2 overexpression models were downloaded from GEO (GSE139523). The dataset was derived from sorted B220+ cells lymphomas between SHMT2 overexpression and control mice. Firstly, the quality of the raw data was assessed using FastQC v0.11.8 and MultiQC v1.7. Low-quality sequences and junctions were trimmed using trim-galore. Salmon software was used to map sequencing files to Mouse Genome Assembly GRCm39 and quantify gene expression. Differential gene expression was performed using the DEseq2 program. Differentially expressed genes were visualized using the ggplot2 program with an FDR adjusted p-value of less than 0.05 and an absolute log2 fold change greater than 1 as the standard filter.²⁴ Differentially expressed genes were analysed by the KEGG signalling pathway through the ClusterProfiler package, while the normalised gene expression matrix was enriched for gene set enrichment analysis (GSEA). All the above processes were run and visualised through R 4.0 and R studio.

Statistical analysis

Data analysis was performed using GraphPad Prism 9.0. The Mann–Whitney U test was used for comparisons between unpaired groups. The Wilcoxon signed-rank test was applied to paired groups. One-way ANOVA was used for comparisons among three groups, followed by Dunn’s multiple comparison test for post-hoc analysis. Spearman’s rank correlation was used to assess relationships between variables. Data were recorded as mean and standard deviation. p values < 0.05 were considered statistically significant.

Ethics

The study was approved by the Ethics Committee at the First Hospital of China Medical University (approval number: 2019-150-4). All participants provided written informed consent prior to their enrolment.

Role of funders

The study funders had no role in the study design, data collection, data analysis, data interpretation, or writing of the report.

Results

KLRG1 expression is increased on T cells of HIV-infected patients

To assess T cell senescence in HIV-infected individuals, we evaluated the expression of KLRG1 on T cells. KLRG1 was a senescence associated molecule,²⁵ which was characterised in RBL-2H3 mast cells from rats²⁶ and initially considered as a unique marker for replicative senescence of murine CD8+ T cells.²⁷ We found that the expression of KLRG1 was markedly elevated in the CD4+ T cell subset (Fig. 1a) as well as the CD8+ T cell subset (Fig. 1b) among HIV-infected patients, which is consistent with previous studies.^28,29 Furthermore, we explored if this senescent phenotype was tied to disease progression by scrutinising the CD4+ T cell counts. Notably, we found a significant inverse relationship between the percentage of KLRG1 expression and CD4+ T cells within the HIV-infected group. This correlation manifested in both CD4+ and CD8+ T cells (Fig. 1c and d), insinuating that T cell senescence may be intertwined with the trajectory of HIV disease.

To delve deeper into cellular senescence across various T cell subsets, we assessed KLRG1 expression on CD4+ and CD8+ T cells with different differentiation status (Fig. 1e–g).³⁰ We found an increasing expression of KLRG1 on CD4+ T cells with advanced differentiation from naive (N, CCR7+CD45RA+), central memory (CM, CCR7+CD45RA-), effector memory (EM, CCR7-CD45RA-), to terminally differentiated effector memory cells (EMRA, CCR7-CD45RA+) (Fig. 1g). For the CD8+ T cell subsets, a differential expression of KLRG1 was observed in all groups except the EM and EMRA cells (Fig. 1g). Further analysis revealed that CD8+ T cells consistently displayed elevated KLRG1 expression across all subsets compared to CD4+ T cells (Fig. 1g). This underscores the heightened vulnerability of CD8+ T cells to senescence compared to CD4+ T cells in the HIV-positive cohort, as previously reported.

SHMT2 gene expression is negatively correlated with senescence in HIV-infected individuals' CD8+ T cells

Emerging research suggests that metabolic alterations can drive cellular senescence.^31,32 To determine the relevance of the metabolic profile of HIV-infected CD8+ T cells to senescence, we examined the surface expression of amino acid (CD98), lipid (CD36), and glucose (GLUT1) transporters.^10,33,34 Interestingly, we found an inverse relationship between CD98 and KLRG1 in CD8+ T cells (Fig. 2a), whereas GLUT1 (Fig. 2b) and CD36 (Fig. 2c) showed no correlation in CD8+ T cells. In addition, we evaluated the expression levels of CD98 (Supplementary Fig. S1a), GLUT1 (Supplementary Fig. S1b), and CD36 (Supplementary Fig. S1c) in CD8+ T cell subsets with different differentiation status. We found the decreased expression of CD98 but not GLUT1 and CD36 on CD8+ T cells with advanced differentiation from naive (N, CCR7+CD45RA+), central memory (CM, CCR7+CD45RA-), effector memory (EM, CCR7-CD45RA-) to terminally differentiated effector memory cells (EMRA, CCR7-CD45RA+) in HIV-infected individuals (Supplementary Fig. S1a–c). These results suggest that CD8+ T cell senescence may be potentially linked to amino acid metabolism rather than glucose metabolism or lipid metabolism.

As a type I amino acid transporter, CD98 plays a crucial role in the serine metabolism pathway.^35,36 In addition, serine metabolism is important for T cell activation during senescence.¹⁹ The elusive connection between alterations in serine metabolism and the process of cellular senescence remains an enigma yet to be unravelled. Therefore, we turned our attention to key serine metabolic enzymes, namely PHGDH, SHMT1, and SHMT2.¹³ Initially, our analysis of gene expression in PBMCs from HIV negative control and HIV-infected cohorts revealed a marked decrease in SHMT2 expression in the latter (Fig. 2d), while the expression of SHMT1 and PHGDH showed no significant differences (Fig. 2e and f). Delving deeper, we discerned a pronounced downregulation of SHMT2 in CD8+ T cells from HIV-infected individuals, a trend absent in CD4+ T cells as compared with controls (Fig. 2g and h). Correlation analyses further illuminated an inverse relationship between SHMT2 transcription and KLRG1 surface expression in CD8+ T cells (Fig. 2i), as well as a negative correlation with CD57 (Supplementary Fig. S1d) and a positive correlation with Ki67 expression (Supplementary Fig. S1e). These findings suggest that SHMT2 might be an underappreciated player in CD8+ T cell senescence. Investigating the potential role of SHMT2 in modulating senescence is a pivotal next step in our research trajectory.

SHMT2 prevents cellular senescence and maintains CD8+ T cell function

Given the important role of SHMT2 in one-carbon folate metabolism,³⁷ which is closely related to proliferation, we detected the role of SHMT2 in T cell proliferation by utilising Ki-67 as an indicator and employing CellTrace Violet labelling.³⁸ Silencing SHMT2 via siRNA, we found that the SHMT2 knockdown led to a significant reduction in the proliferation and the MFI of Ki-67 in CD8+ T cells (Fig. 3a), underscoring SHMT2's potential to promote T cell proliferation. Next, we assessed TNF-α and IFN-γ expression in CD8+ T cells post-SHMT2 interference. Notably, following the downregulation of SHMT2, there was a significant reduction in IFN-γ (Fig. 3b and Supplementary Fig. S2a) and TNF-α (Fig. 3c and Supplementary Fig. S2b). Recently, the coexpression of IFN-γ and TNF-α has been used to assess the functionality of tumour-reactive CD8+ T cells.³⁹ We also analysed IFN-γ+TNF-α+ coexpressing T cells and found that these cells were also significantly reduced following SHMT2 inhibition (Fig. 3d). Additionally, we measured the levels of the degranulation marker CD107a and the cytotoxic molecule Granzyme-B. The results showed that both CD107a (Fig. 3e) and Granzyme-B (Fig. 3f) levels were significantly reduced after SHMT2 interference. This reinforced that the function of CD8+ T cells is hampered when SHMT2 is suppressed.

Fig. 3.

SHMT2 maintains CD8+ T cells function and prevents cellular senescence. (a) Representative data and summary graphs showing cell proliferation levels as assessed by CellTrace™ Violet in CD8+ T cells from HIV negative controls (n = 6) and Ki-67 expression in CD8+ T cells from HIV-infected individuals (n = 11). (b and c) Representative data (left) and summary graphs (right) for IFN-γ, and TNF-α expression in CD8+ T cells from HIV-infected individuals (n = 13). (d) The expression of CD8+ T cells co-expressing IFN-γ and TNF-α from HIV-infected individuals (n = 13). (e and f) Representative data (left) and summary graphs (right) for CD107a expression in CD8+ T cells from HIV-infected individuals (n = 13), and Granzyme-B expression in CD8+ T cells from HIV negative controls (n = 13). (g) Representative data (left) and summary graphs (right) for KLRG1 expression in CD8+ T cells from HIV-infected individuals (n = 23). (h) Representative data (left) and summary graphs (right) for CD57 and CD28 in CD8+ T cells from HIV-infected individuals (n = 23). (i) The levels of KLRG1 expression in CD8+ T cells from HIV-infected individuals (n = 6). (j) The levels of CD57 and CD28 expression in CD8+ T cells from HIV-infected individuals (n = 6). The p values are determined by Wilcoxon signed rank test. ∗p < 0.05, ∗∗p < 0.01. NC, Negative control; SHMT2 i, SHMT2 RNA interference; SHMT2 OE, SHMT2 overexpress; GZMB, Granzyme-B.

While previous studies have touched on SHMT2's multifaceted regulatory roles in cellular physiology,⁴⁰ its direct influence on cellular senescence remains uncharted. In our exploration, we observed elevated expression of established senescence markers like KLRG1, CD57, and CD28 (Fig. 3g and h) post-SHMT2 interference. When SHMT2 was overexpressed in CD8+ T cells, the expression of KLRG1, CD57, and CD28 was reduced (Fig. 3i, j and Supplementary Fig. S2e). Additionally, we also investigated the effect of SHMT2 knockdown on T cell exhaustion and apoptosis. The results showed that SHMT2 knockdown did not affect the expression of PD-1, TIM3, TIGIT, and CTLA4 on CD8+ T cells (Supplementary Fig. S2c), nor did it impact the percentage of early apoptosis, late apoptosis, or total apoptotic cells (Supplementary Fig. S2d). Decreased SHMT2 expression triggers senescence and dysfunction of CD8+ T cells according to our data.

Reductions in SHMT2 leads to altered ROS levels in CD8+ T cells via mitochondrial dysfunction and abnormal GSH metabolism

To elucidate the mechanistic pathways underpinning the influence of SHMT2 expression on cellular senescence, we procured an RNA-seq dataset (GSE139523) from the GEO and carried out correlation assessments. After differential gene analysis (Fig. 4a), GSEA indicated heightened expression of oxidative phosphorylation pathways, an indicator of mitochondrial function, following SHMT2 overexpression (Fig. 4b). KEGG enrichment analysis of the differential genes indicated that GSH metabolism exhibited an elevation upon SHMT2 overexpression (Fig. 4c). A previous study has shown that glutathione contributes to the maintenance of mitochondrial function.⁴¹ To study whether SHMT2 modulated mitochondrial functionality, we gauged mitochondrial depolarization by quantifying both mitochondrial mass (MM) and mitochondrial membrane potential using the specific mitochondrial probes, MTG and MTO. Post-SHMT2 silencing, there was a notable surge in CD8+ T cell mitochondrial depolarization (Fig. 4d). Additionally, our scrutiny of glutathione (GSH) and its oxidised form (GSSG) within CD8+ T cells revealed a marked dip in GSH levels upon SHMT2 silencing compared to controls (Fig. 4e). This observation suggested that SHMT2 knockdown decreased GSH concentrations and led to the mitochondrial dysfunction of CD8+ T cells.

Fig. 4.

Reduction in SHMT2 leads to altered CD8+ T cell ROS levels via mitochondrial dysfunction and abnormal GSH metabolism. (a and b) Differential gene expression analysis and GSEA analysis were performed after SHMT2 overexpression in mouse lymphomas using the mouse RNA-seq dataset (GSE139523) from the Gene Expression Omnibus database. (c) KEGG enrichment analysis was performed on sorted B220+ cells from lymphomas of SHMT2 overexpression and control mice, using the mouse RNA-seq dataset (GSE139523) from the Gene Expression Omnibus database. (d) Representative data (left) and summary graphs (right) for mitochondrial depolarization in CD8+ T cells from HIV negative controls (n = 4) and HIV-infected individuals (n = 11). (e) The GSH/GSSG ratio was measured to indicate GSH levels in CD8+ T cells from HIV-infected individuals (n = 8). (f) Representative data (left) and summary graphs (right) for mitochondrial superoxide in CD8+ T cells from HIV negative controls (n = 8) and HIV-infected individuals (n = 11). The p values are determined by Wilcoxon signed rank test. ∗∗p < 0.01, ∗∗∗∗p < 0.0001. NC, Negative control; SHMT2 i, SHMT2 RNA interference.

Mitochondria dysfunction and decreased GSH typically leads to the release of ROS, which further aggravates mitochondrial dysfunction. Based on the above-mentioned abnormalities in both mitochondrial and GSH systems after knockdown of SHMT2, we assayed mitochondrial ROS (MITOSOX). Consistent with our speculation, knockdown of SHMT2 resulted in a significant increase in MITOSOX (Fig. 4f and Supplementary Fig. S3). This suggests that SHMT2 maintains cellular redox homeostasis, and reductions in SHMT2 lead to altered CD8+ T cell ROS levels.

SHMT2 decreases the activation of senescence signalling in the nucleus through regulation of ROS

The attenuation of SHMT2 results in heightened ROS concentrations, which act as an integral conduit for signalling between the mitochondria and the nucleus.⁴² Given this context, we probed deeper into SHMT2's potential influence on nuclear senescence signalling pathways. The DNA damage response (DDR) is an inherent reaction to DNA insults instigated by elevated ROS levels. Positioned downstream of DDR, the protein p53 stands as a hallmark of cellular senescence and is concurrently subject to negative regulation in the cell cycle.⁴³ To discern the ramifications of diminishing SHMT2 on the DDR-p53 axis, we probed the levels of γH2AX (a DDR benchmark)⁴⁴ and phosphorylated p53. Remarkably, both γH2AX and phosphorylated p53 concentrations increased post-SHMT2 suppression. However, the levels of γH2AX (Fig. 5a) and phosphorylated p53 (Fig. 5b) decreased after inhibition of ROS by adding N-AC. This indicated that SHMT2 depletion promotes a surge in ROS, instigating DNA damage and amplifying phosphorylated p53 levels. Additionally, ROS has the ability to potentiate the downstream element p16, another senescence indicator under negative cell cycle regulation.⁴⁵ Assessing p16 levels unveiled an increase post-SHMT2 curtailment, which was countered upon ROS inhibition by adding N-AC, underscoring the interplay between SHMT2 reduction, ROS augmentation, and an increase in p16 (Fig. 5c). Our study shows that SHMT2 expression is reduced in HIV-affected CD8+ T cells. This reduction leads to mitochondrial dysfunction and reduced GSH levels, which in turn leads to elevated ROS and activation of the nuclear senescence signalling cascade.

Fig. 5.

Phosphorylation profiling reveals senescence signalling cascade in CD8+ T cells following SHMT2 RNA interference. (a) Representative data (upper) and summary graphs (lower) for γH2AX phosphorylation in CD8+ T cells from HIV-infected individuals (n = 12). (b) Representative data (upper) and summary graphs (lower) for p53 phosphorylation in CD8+ T cells from HIV-infected individuals (n = 12). (c) Representative data (upper) and summary graphs (lower) for p16 in CD8+ T cells from HIV-infected individuals (n = 8). The p values are determined by one-way ANOVA with Dunn’s multiple comparison test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. NC, Negative control; SHMT2 i, SHMT2 RNA interference; SHMT2 i + N-AC, Addition of N-AC following SHMT2 RNA interference.

Glycine ameliorates CD8+ T cell senescence in HIV-infected individuals

While we have established the influence of SHMT2 on CD8+ T cell senescence and its associated mechanisms, strategies to counteract the deleterious effects stemming from decreased SHMT2 levels remained elusive. Earlier investigations hinted that augmenting SHMT2 metabolic enzyme functionality could lead to B-cell lymphoma emergence,⁴⁶ signifying a potential oncogenic threat from SHMT2. Therefore, SHMT2 itself might not be a prudent target for anti-senescence interventions. Guided by this knowledge, we investigated whether glycine, a direct metabolic offshoot of SHMT2,⁴⁷ could provide anti-senescence benefits upon the CD8+ T cells in HIV-infected subjects.

We introduced exogenous glycine to in vitro cultures, to assess if glycine could mitigate the adverse impacts arising from diminished SHMT2 expression. Subsequent detection of senescence markers showed that glycine resulted in a significant decrease in both the proportion of CD57 + CD28-cells and KLRG1 expression (Fig. 6a, b, and Supplementary Fig. S4a). Next, we assessed whether glycine might counteract the compromised CD8+ T cell functionality associated with SHMT2 suppression. We found that glycine bolstered cell functional markers including IFN-γ, TNF-α, CD107a (Fig. 6c–e, and Supplementary Fig. S4b–d). This underscored glycine's potential to counterbalance the adverse outcomes on CD8+ T cells associated with SHMT2 reduction.

Fig. 6.

Immunophenotypic and signalling pathways of CD8+ T cells after SHMT2 modulation with glycine supplementation. (a and b) The expression of CD57, CD28 and KLRG1 in CD8+ T cells from HIV-infected individuals (n = 17). (c–e) The expression of IFN-γ, CD107a and TNF-α in CD8+ T cells from HIV-infected individuals (n = 13). (f) Representative data (left) and summary graphs (right) for mitochondrial depolarization in CD8+ T cells from HIV-infected individuals (n = 11). (g) The GSH/GSSG ratio was measured to indicate GSH levels in CD8+ T cells from HIV-infected individuals (n = 8). (h) Representative data (left) and summary graphs (right) for mitochondrial superoxide in CD8+ T cells from HIV-infected individuals (n = 9). (i–k) Representative data (left) and summary graphs (right) for γH2AX, p53 and p16 phosphorylation in CD8+ T cells from HIV-infected individuals (n = 8). The p values are determined by Wilcoxon signed rank test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. SHMT2 i, SHMT2 RNA interference; SHMT2 i + Gly, Addition of exogenous glycine following SHMT2 RNA interference.

Our study found that SHMT2 expression is reduced in CD8+ T cells from HIV-infected individuals. Therefore, we also conducted in vitro experiments with glycine supplementation in CD8+ T cells from HIV-infected individuals. The results showed that after exogenous glycine supplementation, the proportion of CD57 + CD28-cells (Supplementary Fig. S5a) and the expression of KLRG1 (Supplementary Fig. S5b) were significantly reduced. Additionally, cell functional markers such as IFN-γ, TNF-α, CD107a, Granzyme-B (Supplementary Fig. S5c–f), and cell proliferation capacity (Supplementary Fig. S5g) were restored. Consistent with our previous findings, glycine supplementation did not affect levels of exhaustion or apoptosis in CD8+ T cells (Supplementary Fig. S5h and i).

Glycine decreases phosphorylation in nuclear senescence signalling pathways by reducing mitochondrial ROS

Subsequently, we investigated whether the anti-senescence effects of glycine in HIV-infected CD8+ T cells were associated with an increase in ROS, which could be induced by SHMT2 inhibition. Our data showed that glycine significantly attenuated mitochondrial depolarization in CD8+ T cells (Fig. 6f), while causing a significant increase in intracellular GSH levels (Fig. 6g), which collectively led to a decrease in mitochondrial ROS (Fig. 6h and Supplementary Fig. S4e). Our in vitro evaluation showed that glycine supplementation effectively restored mitochondrial function and increased GSH concentration, reducing ROS levels.

Although our study suggests that glycine can inhibit the increase in ROS, the level to which this inhibition can alleviate ROS-induced signalling in nuclear senescence remains undetermined. Therefore, using supplemental glycine, we further analysed the ROS-linked nuclear senescence signalling cascades. Our findings pointed to glycine's potential to diminish the phosphorylation metrics of γH2AX (Fig. 6i), p53 (Fig. 6j), and p16 (Fig. 6k). Thus, glycine appears adept at tempering the ROS-DDR-p53 and ROS-p16 activations. We found that in SHMT2 knockdown T cells, glycine ameliorates CD8+ T cell senescence and decreases phosphorylation in nuclear senescence signalling pathways by reducing mitochondrial ROS levels.

Discussion

After HIV infection, persistent immune activation and inflammation can lead to immune senescence, in which the immune system loses its ability to respond effectively to pathogens or cancer cells, resulting in a poor prognosis for HIV-infected patients.48, 49, 50 In comparison to other immune cells, CD8+ T cells exhibit increased susceptibility to senescence, thus we propose they represent a novel avenue for controlling immune senescence, albeit the precise mechanism has remained obscure until now.⁷ In this study, we identified the role of the serine metabolism key enzyme, SHMT2, in CD8+ T cell senescence among HIV-infected individuals, which protects mitochondrial function, thereby effectively controlling intracellular reactive oxygen species (ROS) levels. Moreover, SHMT2 significantly contributes to averting immune senescence and sustaining CD8+ T cell competence by modulating downstream DNA damage and phosphorylation cascades in pathways intricately linked to cellular senescence (Fig. 7).

Fig. 7.

The schematic diagram shows that reduction of SHMT2 in HIV-infected CD8+ T cells causes immune senescence.

Our first finding was that the serine catabolic enzyme SHMT2 affects CD8+ T cell senescence and function in HIV infection. Recent studies have found that changes in serine levels are associated with senescence.51, 52, 53 Serine metabolism in mice ameliorates oxidative stress and inflammation in the hypothalamus during senescence.⁵¹ There is also a possible involvement of the glycine-serine-threonine metabolic axis in longevity and its associated molecular mechanisms.⁵² In HIV infection, SHMT2 has been reported to be a novel and important regulator of HIV-1 Tat protein levels in infected T cells, affecting viral transcription.⁵⁴ However, there have been no reports on the relationship between SHMT2 and serine metabolism with ageing in HIV infection to the best of our knowledge. We observed that SHMT2, a key enzyme in serine metabolism, is decreased in CD8+ T cells and negatively correlates with indicators of senescence in HIV-infected individuals. In an in vitro knockdown assay of CD8+ T cells from HIV-infected patients, we demonstrated that a decrease in SHMT2 leads to an increase in senescence markers and a decrease in CD8+ T cell function.

It is widely recognised that exhausted and senescent T cells share several overlapping phenotypic and functional characteristics, such as defective proliferative activity, impaired cytotoxic activity, and increased cell cycle arrest. However, each state has unique molecular and developmental signatures, such as surface molecules, etc. The principal feature of exhausted T cells is the elevated expression of a panel of inhibitory receptors, including PD-1, CTLA-4, Tim-3, LAG-3, BTLA, TIGIT, CD244, CD160, and others. Previous studies have revealed the increased expression of these inhibitory receptors in HIV infection, as well as senescence-associated markers, including CD28, CD57, and KLRG-1, etc. We studied the effects of SHMT2 on exhaustion markers, including PD-1, CTLA-4, Tim-3, and TIGIT. We found that SHMT2 knockdown did not affect the expression of these molecules on CD8+ T cells. We also studied the effect of glycine on exhaustion markers, and found glycine supplementation did not affect PD-1, CTLA-4, Tim-3, and TIGIT expression on CD8+ T cells from HIV infected patients. Our results help to justify that in HIV infected patients with both immune exhaustion and senescence, SHMT2 mainly regulates T cell senescence and maintenance of their function.

We secondly found that SHMT2 downregulation leads to ROS-triggered DNA damage, which activates cellular senescence-related pathways, ultimately leading to cellular senescence in HIV-infected individuals. Mitochondria are key sources of cellular ROS and energy, and are essential for maintaining the homeostasis and function of CD8+ T cells.^55,56 Moreover, the weakening of mitochondrial function is closely related to the process of cellular senescence.^57,58 In our study, we observed that interfering with SHMT2 expression resulted in increased ROS levels. The elevated ROS content can likely be attributed to two factors. First, the lowered SHMT2 might induce structural and functional impairment in mitochondria, thus contributing to the increased production of ROS. This is consistent with studies in the Jurkat cell lines.⁹ Second, decreased GSH synthesis may compromise the cellular antioxidant defence system, thereby impeding effective ROS scavenging and triggering aberrant ROS accumulation.⁴² Following the exogenous supplementation of glycine, which serves as both a product of SHMT2 metabolism and a substrate for GSH, there was a significant decrease in ROS levels. This led us to conclude that insufficient GSH synthesis might be a primary factor contributing to the elevated ROS levels.

DNA damage caused by ROS is thought to be a major cause of cellular senescence, therefore, we analysed the factors associated with DNA damage and the maintenance of the DDR.⁵⁹ Our results showed that a decrease in SHMT2 led to an increase in the phosphorylation levels of γH2AX, p53, and p16. Previous studies have indicated that when elevated intracellular ROS levels lead to DNA damage, a DNA repair mechanism is initiated, involving the phosphorylation of the histone H2A family member, H2AX, by ATM, ATR, and PRKDC, thereby forming γH2AX.⁵¹ The effects triggered by the downregulation of SHMT2 are similar to the phenomenon induced by serine deficiency in tumour cells.⁶⁰ In this case, the p53-p21 signalling axis is activated, leading to p21-mediated G1 phase arrest as well as a decrease in the number of S-cycle cells. This process leads to cell cycle arrest, curtailing cellular proliferation and diminishing functionality.⁶¹ Through exogenous glycine supplementation, we succeeded in decreasing the ROS levels and thus the phosphorylation levels of ROS-γH2AX-p53. Furthermore, glycine supplementation restored mitochondrial function and promoted intracellular GSH synthesis, thereby slowing down the cellular senescence process. Our data suggest that SHMT2 affects senescence by modulating the level of ROS-γH2AX-p53 and ROS-p16, and that glycine may be an effective therapeutic target to ameliorate CD8+ T cell senescence and restore function in HIV-infected individuals.

Our study has limitations. We did not perform multi-centre studies and all subjects were recruited from one hospital. It’s because: 1) the difficulties of timely transferring of the freshly collected sample to avoid alteration of expression following freezing processes from different centres to our hospital; and 2) the consistency of the data acquired from different labs by flowcytometer cannot be assured because the detection results are still dependent on the measuring instruments and other factors including reagents and sample processing procedures. Multi-centre study which includes patients from hospitals of different districts would improve the generalizations of the findings.

Overall, our study reveals that reduced SHMT2 in CD8+ T cells from HIV-infected individuals promotes cellular senescence through reduced GSH and abnormal mitochondrial function, causing elevated intracellular ROS levels. This leads to increased levels of phosphorylation of downstream DNA damage and cellular senescence-related pathways. In addition, we identified an anti-senescence effect exerted by glycine in CD8+ T cells. Our study provides new insights into the mechanisms of immune senescence in HIV infection.

Contributors

Zi-Ning Zhang and Hong Shang conceived and designed the experiments. Qi-Sheng Zhang and Jia-Ning Wang performed the experiments and analysed the data. Tian-Ling Yang, Si-Yao Li, Jia-Qi Li, and Ding-Ning Liu contributed reagents, materials, and analysis tools. Zi-Ning Zhang, Hong Shang, Qi-Sheng Zhang, and Jia-Ning Wang wrote the article. All authors revised the manuscript and approved it for publication. Note that more than one author has accessed and verified the data, in this case: Zi-Ning Zhang, Hong Shang, Qi-Sheng Zhang, and Jia-Ning Wang.

Data sharing statement

The raw data supporting the conclusions of this article will be made available by the authors without undue reservation.

Declaration of interests

All authors declare no conflicts of interest.

Appendix ASupplementary data

Supplementary Fig. S1

Fig. S1: Metabolic markers in CD8+ T cell subsets and the relationship between SHMT2 and senescence in HIV-infected individuals. (a-c) The levels of CD98, GLUT1 and CD36 expression in CD8+ T cell subsets (N, CM, EM and EMRA) from HIV negative controls (n=20) and HIV-infected individuals (n=20). (d) Correlation analysis of CD57 mRNA and SHMT2 mRNA expression in CD8+ T cells from HIV negative controls (n=12) and HIV-infected individuals (n=12). (e) Correlation analysis of Ki-67 mRNA and SHMT2 mRNA expression in CD8+ T cells from HIV negative controls (n=12) and HIV-infected individuals (n=12).Data are analysed by Mann-Whitney U test with comparisons between HIV-infected individuals and HIV negative controls in a-c; Wilcoxon signed rank test with comparisons between CD8+ T cell subsets in a-c; Spearman linear regression test in d-e. ∗p < 0.05, ∗∗p < 0.01,∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. HC, HIV negative controls; HIV, HIV-infected individuals; N, naive; CM, central memory; EM, effector memory; EMRA, terminally differentiated effector memory cells.

Supplementary Fig. S2

Fig. S2: SHMT2 preserves CD8+ T cell function and prevents cellular senescence. (a-b) The MFI of IFN-γ and TNF-α in CD8+ T cells from HIV-infected individuals (n=13). (c) The expression of PD-1, TIM-3, TIGIT and CTLA4 in CD8+ T cells from HIV negative controls (n=16). (d) The levels of early apoptosis, late apoptosis and total apoptosis in CD8+ T cells from HIV negative controls (n=11). (e) The MFI of SHMT2 in CD8+ T cells HIV negative controls (n=4) and HIV-infected individuals (n=10). The p values are determined by Wilcoxon signed rank test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. NC, Negative control; SHMT2 i, SHMT2 RNA interference; SHMT2 OE, SHMT2 overexpress; MFI, Mean fluorescence intensity.

Supplementary Fig. S3

Fig. S3: Reductions in SHMT2 lead to mitochondrial dysfunction in CD8+ T cells. The expression of mitochondrial superoxide in CD8+ T cells from HIV negative controls (n=8) and HIV-infected individuals (n=11). The p values are determined by Wilcoxon signed rank test. ∗∗∗∗p < 0.0001. NC, Negative control; SHMT2 i, SHMT2 RNA interference.

Supplementary Fig. S4

Fig. S4: Immunophenotypic and signalling pathways of CD8+ T cells regulated by supplementing Glycine to modulate SHMT2. (a) The MFI of KLRG1 in CD8+ T cells from HIV-infected individuals (n=17). (b–d) The MFI of IFN-γ, CD107a and TNF-α in CD8+ T cells from HIV-infected individuals (n=13). (e) Representative data (left) and summary graphs (right) for mitochondrial superoxide in CD8+ T cells from HIV-infected individuals (n=9). The p values are determined by Wilcoxon signed rank test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. SHMT2 i, SHMT2 RNA interference; SHMT2 i + Gly, Addition of exogenous glycine following SHMT2 RNA interference.

Supplementary Fig. S5

Fig. S5: Supplementation of glycine in CD8+ T cells from HIV-infected individuals inhibits senescence and restores cell function. (a) The levels of CD57 and CD28 expression in CD8+ T cells from HIV-infected individuals and HIV-infected individuals treated with Glycine (n=15). (b) The levels of KLRG1 expression in CD8+ T cells from HIV-infected individuals and HIV-infected individuals treated with Glycine (n=14). (c-e) The expression of CD107a, IFN-γ and TNF-α in CD8+ T cells from HIV-infected individuals and HIV-infected individuals treated with Glycine (n=15). (f) The expression of Granzyme-B in CD8+ T cells from HIV-infected individuals and HIV-infected individuals treated with Glycine (n=15). (g) Cell proliferation levels as assessed by CellTrace™ Violet in CD8+ T cells from HIV-infected individuals and HIV-infected individuals treated with Glycine (n=7). (h) The expression of PD-1, TIM-3, TIGIT and CTLA4 in CD8+ T cells from HIV-infected individuals and HIV-infected individuals treated with Glycine (n=15). (i) The levels of early apoptosis, late apoptosis and total apoptosis in CD8+ T cells from HIV-infected individuals and HIV-infected individuals treated with Glycine (n=17). The p values are determined by Wilcoxon signed rank test. ∗p < 0.05, ∗∗p < 0.01.HIV, HIV-infected individuals; HIV + Gly, Addition of exogenous glycine to HIV-infected individuals; GZMB, Granzyme-B.