Serine Is an Essential Metabolite for Effector T Cell Expansion

•Enzymes of SGOC metabolism are upregulated upon T cell activation•Serine is required for optimal T cell proliferation in vitro and in vivo•Serine supports de novo purine biosynthesis in proliferating T cells•One-carbon units from formate can bypass serine starvation During immune challenge, T lymphocytes engage pathways of anabolic metabolism to support clonal expansion and the development of effector functions. Here we report a critical role for the non-essential amino acid serine in effector T cell responses. Upon activation, T cells upregulate enzymes of the serine, glycine, one-carbon (SGOC) metabolic network, and rapidly increase processing of serine into one-carbon metabolism. We show that extracellular serine is required for optimal T cell expansion even in glucose concentrations sufficient to support T cell activation, bioenergetics, and effector function. Restricting dietary serine impairs pathogen-driven expansion of T cells in vivo, without affecting overall immune cell homeostasis. Mechanistically, serine supplies glycine and one-carbon units for de novo nucleotide biosynthesis in proliferating T cells, and one-carbon units from formate can rescue T cells from serine deprivation. Our data implicate serine as a key immunometabolite that directly modulates adaptive immunity by controlling T cell proliferative capacity. During immune challenge, T lymphocytes engage pathways of anabolic metabolism to support clonal expansion and the development of effector functions. Here we report a critical role for the non-essential amino acid serine in effector T cell responses. Upon activation, T cells upregulate enzymes of the serine, glycine, one-carbon (SGOC) metabolic network, and rapidly increase processing of serine into one-carbon metabolism. We show that extracellular serine is required for optimal T cell expansion even in glucose concentrations sufficient to support T cell activation, bioenergetics, and effector function. Restricting dietary serine impairs pathogen-driven expansion of T cells in vivo, without affecting overall immune cell homeostasis. Mechanistically, serine supplies glycine and one-carbon units for de novo nucleotide biosynthesis in proliferating T cells, and one-carbon units from formate can rescue T cells from serine deprivation. Our data implicate serine as a key immunometabolite that directly modulates adaptive immunity by controlling T cell proliferative capacity. T lymphocytes are sentinels of the adaptive immune system tailored to identify and eliminate threats to the host. In response to antigen-specific and inflammatory signals, naive T cells transition to an activated state marked by rapid growth and proliferation, resulting in the generation of a T effector (Teff) cell pool sufficient to mediate protective immunity. One of the fundamental programs triggered by T cell receptor (TCR) activation is the reprogramming of cellular metabolism to support the demands of Teff cell expansion (Buck et al., 2015Buck M.D. O’Sullivan D. Pearce E.L. T cell metabolism drives immunity.J. Exp. Med. 2015; 212: 1345-1360Crossref PubMed Scopus (708) Google Scholar, Wang and Green, 2012Wang R. Green D.R. Metabolic checkpoints in activated T cells.Nat. Immunol. 2012; 13: 907-915Crossref PubMed Scopus (330) Google Scholar). This metabolic shift is characterized by increased expression of nutrient transporters and glycolytic enzymes early after activation (Macintyre et al., 2014Macintyre A.N. Gerriets V.A. Nichols A.G. Michalek R.D. Rudolph M.C. Deoliveira D. Anderson S.M. Abel E.D. Chen B.J. Hale L.P. Rathmell J.C. The glucose transporter Glut1 is selectively essential for CD4 T cell activation and effector function.Cell Metab. 2014; 20: 61-72Abstract Full Text Full Text PDF PubMed Scopus (668) Google Scholar, Wang et al., 2011Wang R. Dillon C.P. Shi L.Z. Milasta S. Carter R. Finkelstein D. McCormick L.L. Fitzgerald P. Chi H. Munger J. Green D.R. The transcription factor Myc controls metabolic reprogramming upon T lymphocyte activation.Immunity. 2011; 35: 871-882Abstract Full Text Full Text PDF PubMed Scopus (1316) Google Scholar). Glycolytic flux provides energy and building blocks for T cell expansion, but also influences effector function by regulating cytokine production by Teff and T memory (Tmem) cells (Cham and Gajewski, 2005Cham C.M. Gajewski T.F. Glucose availability regulates IFN-gamma production and p70S6 kinase activation in CD8+ effector T cells.J. Immunol. 2005; 174: 4670-4677Crossref PubMed Scopus (235) Google Scholar, Chang et al., 2013Chang C.H. Curtis J.D. Maggi Jr., L.B. Faubert B. Villarino A.V. O’Sullivan D. Huang S.C. van der Windt G.J. Blagih J. Qiu J. et al.Posttranscriptional control of T cell effector function by aerobic glycolysis.Cell. 2013; 153: 1239-1251Abstract Full Text Full Text PDF PubMed Scopus (1323) Google Scholar, Gubser et al., 2013Gubser P.M. Bantug G.R. Razik L. Fischer M. Dimeloe S. Hoenger G. Durovic B. Jauch A. Hess C. Rapid effector function of memory CD8+ T cells requires an immediate-early glycolytic switch.Nat. Immunol. 2013; 14: 1064-1072Crossref PubMed Scopus (325) Google Scholar). The requirement of specific nutrients to support T cell function raises the possibility that the metabolic microenvironment and nutrient availability can impact immunity by influencing T cell function. One pathway that has emerged as a key metabolic node in proliferating cells is one-carbon metabolism, which encompasses a network of interconnected biochemical pathways that facilitate the transfer of one-carbon units for biosynthesis (Locasale, 2013Locasale J.W. Serine, glycine and one-carbon units: cancer metabolism in full circle.Nat. Rev. Cancer. 2013; 13: 572-583Crossref PubMed Scopus (971) Google Scholar) (Figure 1A). Folate intermediates such as tetrahydrofolate (THF) are active carriers of one-carbon units used for de novo nucleotide synthesis. Connected to the folate cycle is the methionine cycle, which is involved in the synthesis of amino acids, lipids, and S-adenosyl methionine (SAM), the latter serving as the primary methyl donor for cellular methylation reactions. The methionine cycle is linked to the trans-sulphuration pathway, which produces glutathione, an important cellular antioxidant. Therapeutics that disrupt folate metabolism such as methotrexate are used to treat certain autoimmune conditions, including rheumatoid arthritis and Crohn’s disease (Cesarini et al., 2016Cesarini M. Festa S. Papi C. Methotrexate in Crohn’s disease: a new face for an old drug?.Expert Rev. Gastroenterol. Hepatol. 2016; 10: 1135-1144Crossref Scopus (5) Google Scholar, Cronstein, 1996Cronstein B.N. Molecular therapeutics. Methotrexate and its mechanism of action.Arthritis Rheum. 1996; 39: 1951-1960Crossref PubMed Scopus (224) Google Scholar), highlighting the clinical importance of one-carbon metabolism to immune cell function. While THF and its derivatives function as carriers of methyl groups during one-carbon metabolism, one of the major carbon donors to the pathway is the non-essential amino acid serine. Serine is converted to glycine by the enzyme serine hydroxymethyl transferase (Shmt), which transfers a methyl group from serine to THF, thus helping to drive the folate cycle (Figure 1A). Serine can be taken up by cells from exogenous sources or synthesized de novo from the glycolytic intermediate 3-phosphoglycerate (3-PG) (Pearce et al., 2013Pearce E.L. Poffenberger M.C. Chang C.H. Jones R.G. Fueling immunity: insights into metabolism and lymphocyte function.Science. 2013; 342: 1242454Crossref PubMed Scopus (823) Google Scholar). How serine is sourced for cellular biosynthesis varies widely among proliferating cells. Many cancer cells are highly dependent on exogenous serine for biosynthesis (Labuschagne et al., 2014Labuschagne C.F. van den Broek N.J. Mackay G.M. Vousden K.H. Maddocks O.D. Serine, but not glycine, supports one-carbon metabolism and proliferation of cancer cells.Cell Rep. 2014; 7: 1248-1258Abstract Full Text Full Text PDF PubMed Scopus (358) Google Scholar, Maddocks et al., 2013Maddocks O.D. Berkers C.R. Mason S.M. Zheng L. Blyth K. Gottlieb E. Vousden K.H. Serine starvation induces stress and p53-dependent metabolic remodelling in cancer cells.Nature. 2013; 493: 542-546Crossref PubMed Scopus (619) Google Scholar), while others increase de novo serine biosynthesis from glucose to drive growth (Locasale et al., 2011Locasale J.W. Grassian A.R. Melman T. Lyssiotis C.A. Mattaini K.R. Bass A.J. Heffron G. Metallo C.M. Muranen T. Sharfi H. et al.Phosphoglycerate dehydrogenase diverts glycolytic flux and contributes to oncogenesis.Nat. Genet. 2011; 43: 869-874Crossref PubMed Scopus (784) Google Scholar, Possemato et al., 2011Possemato R. Marks K.M. Shaul Y.D. Pacold M.E. Kim D. Birsoy K. Sethumadhavan S. Woo H.K. Jang H.G. Jha A.K. et al.Functional genomics reveal that the serine synthesis pathway is essential in breast cancer.Nature. 2011; 476: 346-350Crossref PubMed Scopus (1117) Google Scholar). Collectively, the serine, glycine, one-carbon (SGOC) metabolic network helps coordinate nucleotide, NADPH, and glutathione biosynthesis to support macromolecular biosynthesis and redox balance in cancer cells (Mehrmohamadi et al., 2014Mehrmohamadi M. Liu X. Shestov A.A. Locasale J.W. Characterization of the usage of the serine metabolic network in human cancer.Cell Rep. 2014; 9: 1507-1519Abstract Full Text Full Text PDF PubMed Scopus (94) Google Scholar). In this study, we investigated the dynamics of SGOC metabolism and serine availability on T cell function. We report that serine is an essential metabolite for optimal T cell expansion following antigenic stimulation. T cells cultured without exogenous serine or with Shmt inhibition show no impairment in activation, bioenergetics, or cytokine production, but fail to proliferate efficiently. Strikingly, these results translate in vivo, as antigen-specific T cell expansion is impaired in animals maintained on a serine-restricted diet. Our data implicate serine metabolism as an intrinsic regulator of T cell proliferative capacity—independent of glycolysis—that is required for optimal T cell responses in vivo. Recent work has identified the SGOC metabolic network as a metabolic feature of many cancer cells (Mehrmohamadi et al., 2014Mehrmohamadi M. Liu X. Shestov A.A. Locasale J.W. Characterization of the usage of the serine metabolic network in human cancer.Cell Rep. 2014; 9: 1507-1519Abstract Full Text Full Text PDF PubMed Scopus (94) Google Scholar) (Figure 1A). To better understand the dynamics of SGOC metabolism during T cell responses in vivo, we analyzed the expression of a subset of SGOC network genes in CD8+ TCR-transgenic OT-I T cells (specific for OVA257–264) responding to OVA-expressing Listeria monocytogenes (LmOVA) infection in vivo, as previously generated by the Immunological Genome Project (Best et al., 2013Best J.A. Blair D.A. Knell J. Yang E. Mayya V. Doedens A. Dustin M.L. Goldrath A.W. Immunological Genome Project ConsortiumTranscriptional insights into the CD8(+) T cell response to infection and memory T cell formation.Nat. Immunol. 2013; 14: 404-412Crossref PubMed Scopus (219) Google Scholar) (Figure 1B). We observed a clustering of SGOC network gene products increased in early CD8+ Teff cells (0.5–2 days post-infection [dpi], correlating with the stage of rapid Teff cell expansion following infection). These included genes involved in de novo serine biosynthesis (Psat1, Phgdh, and Psph), the THF cycle (both mitochondrial [Mthfd2 and Mthfd1l] and cytosolic [Mthfd1]), and the methionine cycle (Mat2a, Mtrr, and Mtr). The largest increase was observed in Shmt1 and Shmt2 expression (Figure 1B), which regulates the entry of serine-dependent carbon into the cytosolic and mitochondrial THF cycles, respectively. Heightened expression of these genes was largely restricted to early Teff cells, with expression returning to baseline levels (as observed in naive CD8+ T cells) by 6 dpi (Figure 1B). The exception to this trend was the expression of Tyms and Dhfr, which together mediate a shunt of the THF cycle involved in thymidine biosynthesis (Figure 1A) that remained increased in Teff cells at 6 dpi. We next verified the expression of SGOC network components in mouse and human T cells by qPCR. Expression of mRNA transcripts for the serine biosynthesis pathway (Phgdh, Psat, and Psph), serine entry into one-carbon metabolism (Shmt1 and Shmt2), and thymidine biosynthesis (Dhfr and Tyms) was induced in both CD8+ (Figure 1C) and CD4+ T cells (Figure S1A, available online) upon stimulation with anti-CD3 and anti-CD28 antibodies. We observed a corresponding increase in protein expression of several of these enzymes upon T cell activation (Figure 1D). Activated human T cells also displayed a broad upregulation of genes involved in serine biosynthesis and both the cytosolic and mitochondrial THF pathways (Figures 1E and S1B). Of note, expression of Mthfd2, an mTORC1-responsive enzyme of the mitochondrial THF cycle that mediates production of 10-formyl-THF for de novo purine biosynthesis (Ben-Sahra et al., 2016Ben-Sahra I. Hoxhaj G. Ricoult S.J. Asara J.M. Manning B.D. mTORC1 induces purine synthesis through control of the mitochondrial tetrahydrofolate cycle.Science. 2016; 351: 728-733Crossref PubMed Scopus (444) Google Scholar), was induced by TCR stimulation in both mouse and human CD8+ T cells (Figures 1C and 1E). These data confirm upregulation of the SGOC metabolic network as a hallmark of activated Teff cells in vitro and in vivo, both in mouse and human. We next assessed the dynamics of serine metabolism in Teff cells using stable isotope tracer analysis (SITA). We first cultured proliferating mouse and human Teff cells with U-[13C]-glucose and examined the steady-state incorporation of 13C-glucose-derived carbon into serine and glycine using gas chromatography-mass spectrometry (GC-MS). Approximately 30% of the intracellular serine pool was labeled from glucose in mouse Teff cells, with ∼20% of serine carbon being fully labeled (m+3) (Figure 2A). Glucose-dependent labeling of serine was lower in activated human CD8+ T cells, constituting ∼20% of the intracellular serine pool (Figure 2B). Partially labeled serine isotopomers (m+1 and m+2) were also detected, likely due to the inter-conversion of 13C-labeled and unlabeled serine and glycine, indicating the bi-directional nature of this metabolic pathway. The majority of intracellular serine did not contain carbon from 13C-glucose (m+0), suggesting additional routes for serine entry into one-carbon metabolism in Teff cells. We next assessed the contribution of extracellular serine to SGOC metabolism in T cells by culturing activated Teff cells with U-[13C]-serine. Rapid extracellular serine uptake was observed in Teff cells, with ∼75% of the intracellular serine pool labeled as 13C-serine (m+3) and reaching steady state within 15 min of culture (Figure 2C). These data suggest that most of the intracellular serine pool in Teff cells comes from the extracellular environment. However, carbon from both 13C-glucose-derived serine (Figures 2A and 2B) and U-[13C]-serine (Figure 2C) was readily metabolized into 13C-glycine (m+2), indicating that serine from both de novo biosynthesis and extracellular sources contributed to the one-carbon pool in Teff cells. THF moves one-carbon units through various oxidation states using a series of connected enzymatic redox reactions. N5-N10-methyl-THF conversion to N10-formyl-THF by Mthfd1 or Mthfd2 is used for purine nucleotide synthesis. One-carbon units in the THF pool can also be used to generate deoxythymidine monophosphate (dTMP) or to synthesize methionine (Figure 1A). SITA using U-[13C]-glucose or U-[13C]-serine revealed no labeling of methionine from these carbon sources (Figure 2D). These data indicated that one-carbon units from either de novo synthesized serine or extracellular serine contributed to the THF cycle, but did not continue into the methionine cycle. Given that most of the intracellular serine in activated T cells was taken up from the extracellular environment (Figure 2), we questioned whether extracellular serine availability could impact T cell responses in vivo. To test this, we maintained mice for 2 weeks on a diet lacking serine and glycine to decrease circulating levels of these amino acids. Since serine and glycine are non-essential amino acids, removal of these nutrients from the rodent diet is well tolerated in vivo (Maddocks et al., 2013Maddocks O.D. Berkers C.R. Mason S.M. Zheng L. Blyth K. Gottlieb E. Vousden K.H. Serine starvation induces stress and p53-dependent metabolic remodelling in cancer cells.Nature. 2013; 493: 542-546Crossref PubMed Scopus (619) Google Scholar). Compared to control animals, mice maintained on the serine/glycine-free diet for 2–4 weeks displayed reduced levels of serine and glycine in their blood, while the abundance of other amino acids was not affected (Table S1). Compared to animals fed a control diet containing serine and glycine, analysis of immune cell numbers from the spleens of animals fed serine/glycine-free chow for 2 or 4 weeks revealed no significant change in numbers of lymphocytes (CD4+ T cells, CD8+ T cells, B cells, NK [natural killer] cells, and NKT cells), as well as eosinophils, monocytes, neutrophils, and dendritic cells (Table S2). We next examined T cell responses to LmOVA infection in a low serine/glycine environment in vivo. Congenic CD45.1+ mice were maintained on a control or serine/glycine-free diet for 2 weeks prior to adoptive transfer of naive CD45.2+ OT-I CD8+ T cells, and then infected 1 day later with a modified attenuated strain of LmOVA (Figure 3A). At 7 dpi, splenocytes from infected mice were analyzed for the expansion of OVA-specific CD45.2+ T cells by flow cytometry. We observed a significant reduction in both the percentage and number of OVA-specific CD8+ CD45.2+ T cells in mice maintained on the serine-restricted feed (Figure 3B), corresponding to an ∼70% reduction in responding antigen-specific CD8+ T cells in these mice. This translated into an overall reduction in the number of IFN-γ- and TNF-α-producing effector CD8+ T cells following re-stimulation with OVA257 peptide (SIINFEKL) ex vivo (Figures S2A and S2B). Serine restriction did not affect overall numbers of bystander CD8+ (Figure S2C) and CD4+ (Figure S2D) T cells following LmOVA infection, consistent with our previous observation that overall immune cell numbers were not affected by the serine/glycine-free feed (Table S2). Next, we examined the effect of serine restriction on endogenous T cell responses to Listeria infection. C57BL/6 mice maintained on control or serine/glycine-free feed for 2 weeks were infected with LmOVA, and endogenous CD8+ and CD4+ T cell responses were assessed 7 days later (Figure 3C). Similar to Figure 3B, the percentage and number of endogenous OVA-specific CD8+ T cells were significantly reduced in animals maintained on the serine-restricted diet 7 dpi, compared to animals on the control diet (Figure 3D). Analysis of splenocytes ex vivo following re-stimulation with OVA257 peptide revealed over a 50% reduction in both the number and percentage of IFN-γ-producing CD8+ T cells for animals maintained on the serine-restricted diet (Figure 3E). The number of IFN-γ-producing CD4+ Th1 cells responding to LLO190 peptide stimulation was similarly reduced in mice fed the serine/glycine-free chow (Figure S2E), indicating that antigen-specific CD4+ T cell responses were similarly affected by serine restriction in vivo. Interestingly, the quantity of cytokine produced on a per-cell basis by CD8+ and CD4+ Teff cells responding to LmOVA, as determined by the mean fluorescence intensity (MFI) of the IFN-γ+ cell population, was not affected by diet, suggesting that dietary serine restriction did not affect the functionality of the T cells that could respond to infection (Figures 3F and S2F). Analysis of overall CD8+ and CD4+ T cell numbers in infected mice revealed no difference in total T cell numbers (Figure S2G). Collectively, these data indicated that dietary serine restriction could impact antigen-specific T cell expansion in vivo. One consequence of impaired expansion of Teff cells is a reduced memory T cell pool and reduced T cell responses and pathogen clearance upon re-challenge (Harty and Badovinac, 2008Harty J.T. Badovinac V.P. Shaping and reshaping CD8+ T-cell memory.Nat. Rev. Immunol. 2008; 8: 107-119Crossref PubMed Scopus (427) Google Scholar). To assess whether serine availability during T cell priming affects protective immunity, we maintained C57BL/6 mice on control or serine/glycine-free feed for the duration of the primary LmOVA response (7 dpi), after which all mice were switched to the control feed containing serine and glycine for an additional 25 days (Figure 3G). Tracking OVA-specific CD8+ T cells in the peripheral blood of mice over the course of the immune response to LmOVA infection revealed similar contraction of OVA-specific CD8+ T cells between immunized mice maintained on the control or serine/glycine-free diet for the first 7 days of the infection (Figure S2H). Mice fed the control diet for the entire duration of the experiment displayed a slightly higher percentage of circulating memory OVA-specific CD8+ T cells at 21 dpi compared to mice fed a serine-free diet for the first 7 days of infection (Figure S2H). Mice were then re-challenged with a lethal dose of virulent LmOVA at 32 dpi under serine-replete conditions (Figure 3G). Analysis of serine and glycine abundance in the blood prior to secondary challenge revealed no difference in serine and glycine levels in animals initially maintained on the serine/glycine-free feed relative to controls (Figure S2I). Despite having normal blood serine/glycine levels, mice primed under serine/glycine-free conditions displayed reduced secondary T cell responses to virulent LmOVA, exhibiting reduced numbers of OVA-specific (Figure 3H) and IFN-γ-producing (Figure 3I) CD8+ T cells 5 days post-re-challenge. This impaired secondary T cell response corresponded to a decrease in bacterial clearance in mice originally immunized under serine/glycine-free conditions (Figures 3J and S2J). These data confirmed that serine availability during immunization could impact T cell-mediated immunity in vivo. We next investigated why serine was essential for antigen-specific T cell expansion in vivo. We first assessed the impact of serine availability on the proliferation of T cells from C57BL/6 mice in vitro following polyclonal activation with anti-CD3 and anti-CD28 antibodies. Total T cell proliferation measured by thymidine incorporation (Figure 4A) or the proliferation of CD4+ or CD8+ T cells by dye dilution assay (Figure 4B) revealed clear reductions in T cell proliferation when stimulated in medium containing glucose and glutamine but lacking serine and glycine. Human CD8+ T cells displayed similar reductions in TCR-stimulated proliferation when deprived of serine and glycine (Figure S3A). To assess whether serine was required for the continued proliferation of activated T cells, we expanded T cells in complete medium for 2 days, then continued culture in complete medium or switched the cells to medium lacking serine and glycine. Teff cells cultured without serine and glycine displayed reduced proliferation (Figure 4C) but remained viable (Figure S3B). Despite displaying reduced proliferation, T cells cultured under serine-free conditions displayed comparable expression of activation markers (CD69, CD25, and CD44) (Figure 4D) and continued to produce IFN-γ at similar levels to T cells grown under full serine conditions (Figure 4E). Thus, serine availability affected T cell proliferative capacity without impacting their ability to become activated or acquire effector functions. One possibility for the reduced expansion of antigen-specific T cells under serine-free conditions in vivo is that serine availability affects the stimulatory capacity of antigen-presenting cells (APCs) such as dendritic cells (DCs). DCs activated with LPS under serine/glycine-free conditions showed comparable surface expression of activation markers (CD86 and CD40) (Figure S3C) and secreted cytokines (TNF-α and IL-12p40) at similar levels to DCs activated under full medium conditions (Figure S3D), indicating that serine limitation did not affect the stimulatory capacity of APCs in vitro. T cell bioenergetics can also be a regulator of T cell proliferative capacity. Inhibiting mitochondrial OXPHOS or limiting the glycolytic capacity of T cells has been shown to suppress T cell proliferation (Chang et al., 2013Chang C.H. Curtis J.D. Maggi Jr., L.B. Faubert B. Villarino A.V. O’Sullivan D. Huang S.C. van der Windt G.J. Blagih J. Qiu J. et al.Posttranscriptional control of T cell effector function by aerobic glycolysis.Cell. 2013; 153: 1239-1251Abstract Full Text Full Text PDF PubMed Scopus (1323) Google Scholar, Macintyre et al., 2014Macintyre A.N. Gerriets V.A. Nichols A.G. Michalek R.D. Rudolph M.C. Deoliveira D. Anderson S.M. Abel E.D. Chen B.J. Hale L.P. Rathmell J.C. The glucose transporter Glut1 is selectively essential for CD4 T cell activation and effector function.Cell Metab. 2014; 20: 61-72Abstract Full Text Full Text PDF PubMed Scopus (668) Google Scholar). We measured the extracellular acidification rate (ECAR, a readout of glycolysis) and oxygen consumption rate (OCR, a measure of OXPHOS) (Blagih et al., 2015Blagih J. Coulombe F. Vincent E.E. Dupuy F. Galicia-Vázquez G. Yurchenko E. Raissi T.C. van der Windt G.J. Viollet B. Pearce E.L. et al.The energy sensor AMPK regulates T cell metabolic adaptation and effector responses in vivo.Immunity. 2015; 42: 41-54Abstract Full Text Full Text PDF PubMed Scopus (392) Google Scholar) of Teff cells cultured with or without serine and glycine, and found no impact of serine availability on T cell bioenergetics (Figure 4F). Analysis of culture medium revealed similar glucose and glutamine consumption rates between serine-starved and control Teff cells (Figure S3E). Teff cells readily metabolized glucose to lactate (13C-Glc→lactate (m+3)) and the TCA cycle intermediate citrate (13C-Glc→citrate (m+2)) regardless of serine and glycine availability (Figure 4G). These data indicated that SGOC metabolism was functioning independently of glucose utilization and cellular bioenergetics in Teff cells yet was required for T cell proliferation even when glucose was abundant. We next assessed whether the decrease in T cell proliferation triggered by serine limitation was due to the reduced flux of serine into the one-carbon metabolism pathway. We used a dual Shmt inhibitor (RZ-2994) with high specificity to both Shmt1 and Shmt2 to block serine metabolism to glycine in T cells (Figure 5A). RZ-2994 was identified from a series of pyrazolopyrans active against plant and Plasmodium SHMTs and optimized for activity against human SMHT1 and SHMT2 (Witschel et al., 2015Witschel M.C. Rottmann M. Schwab A. Leartsakulpanich U. Chitnumsub P. Seet M. Tonazzi S. Schwertz G. Stelzer F. Mietzner T. et al.Inhibitors of plasmodial serine hydroxymethyltransferase (SHMT): cocrystal structures of pyrazolopyrans with potent blood- and liver-stage activities.J. Med. Chem. 2015; 58: 3117-3130Crossref PubMed Scopus (42) Google Scholar). We verified that RZ-2994 could inhibit both Shmt enzymes in vitro, with an IC50 of 5 and 13 nM for Shmt1 and Shmt2, respectively (Figure 5B). Teff cells cultured with U-[13C]-glucose or U-[13C]-serine displayed a significant decrease in abundance of labeled glycine (m+2) when cultured with RZ-2994, verifying that this compound could inhibit the entry of carbon from both extracellular and de novo serine into the one-carbon pathway in Teff cells (Figures 5C and 5D). We next activated naive T cells in vitro with anti-CD3 and anti-CD28 antibodies along with RZ-2994 to assess the impact of blocking serine entry into one-carbon metabolism on T cell proliferation. RZ-2994 reduced the proliferation of both CD4+ and CD8+ T cells regardless of whether it was added at the initial point of activation (Figure 5E) or to actively proliferating T cell cultures (Figure 5F). RZ-2994 lowered T cell proliferation in a dose-dependent manner regardless of the presence of serine or glycine in the culture medium (Figure S4A). Interestingly, RZ-2994 had minimal effects on T cell viability, even at higher drug concentrations that induced maximal inhibition of T cell proliferation (>2.5 μM) (Figure S4B). We chose 1.25 μM as the concentration of RZ-2994 used for all subsequent experiments, as this was the EC50 dose for inhibiting cell proliferation that did not significantly affect T cell viability (Figure 5G). Similar to Teff cells cultured in the absence of serine and glycine, T cells cultured with RZ-2994 displayed no changes in the expression of activation markers (CD25 and CD44) (Figure S4C). Furthermore, RZ-2994 did not alter the metabolism of glucose by T cells, as the conversion of U-[13C]-glucose to lactate or citrate was unaffected by RZ-2994 treatment (Figure 5H). Given that T cell bioenergetics were unaffected by serine availability, we hypothesized that SGOC-mediated biosynthesis may run in parallel to glycolysis to control Teff cell proliferation. Input of serine-derived carbon into the THF cycle yields one-carbon units as well as glycine, both critical for de novo purine nucleotide synthesis (Figure 6A) (Labuschagne et al., 2014Labuschagne C.F. van den Broek N.J. Mackay G.M. Vousden K.H. Maddocks O.D. Serine, but not glycine, supports one-carbon metabolism and proliferation of cancer cells.Cell Rep. 2014; 7: 1248-1258Abstract Full Text Full Text PDF PubMed Scopus (358) Google Scholar). Following culture of proliferating Teff cells with U-[13C]-serine, we found that purine nucleotide