The different OxPhos activity promoted by EVR and RAPA in Treg cells may associate with changes in mitochondrial membrane potential (ΔΨm), a critical parameter of mitochondrial functional integrity. To address this possibility, we measured ΔΨm with the TMRE fluorescent dye in the 5-day expanded untreated (control) and drug-treated cells. The reliability of the TMRE uptake to measure ΔΨm was confirmed for each condition by the collapse of the ΔΨm upon preincubation with the decoupler proton ionophore FCCP. To rule out any biased results as a consequence of changes in mitochondrial mass, the TMRE loading results were normalized for mitochondrial content and were expressed as the ratios between TMRE and MitoTracker Green FM fluorescences as reported elsewhere.35 The MitoTracker Green FM results revealed the increase in mitochondrial mass in RAPA-treated cells (447 ± 45 mean fluorescence intensity [MFI]) compared with EVR (385 ± 29 MFI,) and untreated cells (367 ± 42 MFI). Once normalized, the ΔΨm results were not significantly different among different conditions (Figure 4A). Because the activation of the PI3K/mTOR pathway is known to negatively influence autophagy and RAPA is reportedly promoting autophagy in different cell settings,36 , 37 we next asked whether EVR induced similar effects in Treg cells. The analyses with the Cyto-ID Green detection reagent demonstrated that EVR also increased the formation of autophagosomal vacuoles (6288 ± 427 MFI) compared with untreated control cells (5518 ± 345). However, the vacuole formation in EVR samples was significantly lower than in RAPA samples (7245 ± 222). Addition of CLQ to measure autophagic flux produced similar accumulation of autophagosomes in response to both drugs (2143 ± 134 MFI for EVR and 2214 ± 68 for RAPA), both significantly higher than in the untreated group (1442 ± 48) (Figure 4B).
The presence of RAPA or EVR did not induce any significant phenotypic difference in the 5-day expanded cells. The mean fluorescent intensity (MFI) of CD25 and FoxP3 expression was similar under the 2 drug conditions, as well as the expression of other Treg cell-related cell markers including CTLA-4, CD49d (depicted in Figure S6, SDC, http://links.lww.com/TP/B645), PD1, PDL-1, OX40, GITR, and CD86 (data not shown). The functional properties of expanded Treg cells in the presence of EVR or RAPA were assessed by their ability to suppress the proliferation of Tconv cell. The data demonstrated the equivalent suppressive function elicited by EVR- and RAPA-Treg cells (Figure S7, SDC, http://links.lww.com/TP/B645). The functional properties of expanded Treg cells in the presence of EVR or RAPA were assessed by their ability to suppress the proliferation of Tconv cells. The data demonstrated the equivalent suppressive function elicited by EVR- and RAPA-Treg cells (Figure S7, SDC, http://links.lww.com/TP/B645).
Manufacturing clinical grade cellular products for application in adoptive immunotherapy requires the ex vivo expansion of the original pool of cells. To validate the use of EVR to generate sufficient cell numbers of high-quality, clinical-grade Treg cells, freshly isolated CD4+CD25+ cells were expanded in the absence or in the- presence of EVR (100 nM) or RAPA (100 nM). Addition of EVR or RAPA promoted similar cell growth rates (Figure 5A and B), although the cells in EVR-medium experienced a temporary delay in early stages of culture (Figure S8, SDC, http://links.lww.com/TP/B645). In the absence of rapalogs, the long-term expansion produced significantly larger cell yields (Figure 5B). However, these cells displayed a reduced suppressor activity when compared with drug-treated cells, whereas no significant differences in suppressive capacity were noted between both rapalog treatments (Figure 5C). The demethylation of the CpG dinucleotides at the highly conserved TSDR region of the FOXP3 locus is necessary for Treg cell lineage stability.38 Both RAPA- and EVR-treated cells show low expression of methylation levels across the 9 CpG sites of the TSDR (ranges between 5.9% and 12.1% and 6.2% to 13.1%, respectively, in 6 samples), whereas the untreated cells displayed a broader range of methylation (between 12.3% and 43.6%, n = 6) (Figure 5D). The same treatments in conventional T cells produce methylation levels at the TSDR CpG sites ranging from 81% to 97% (n = 8). The high purity of the initial CD4+CD25+ Treg cells was sustained along the 21-day ex vivo expansion period in both RAPA- and EVR-treated cells. The absence of rapalogs in the expansion cell culture produced a final population with inconsistent contamination of CD8+ T cells (ranging from less than 3% to 27% of total T cells in 6 experiments) and with a population of CD4+CD25+ displaying lower expression of FoxP3 and CD25 markers compared with EVR- or RAPA-treated cells (Figure 6A). Further phenotype analysis of these cells showed also a different profile of several phenotype markers expressed distinctly in Treg cells and Tconv cells (Figure S9, SDC, http://links.lww.com/TP/B645), including higher intracellular IL-10, Helios, CCR4, CTLA4, and CD36, and lower expression of TIGIT and PD1 in rapalog-treated cells (Figure 6B), whereas no substantial differences were noted between RAPA and EVR cell culture phenotypes. The functional and phenotypic similarities between long-term expanded drug-treated cells also concur with the convergence of the oxidative metabolism (Figure 7) and glycolytic rates (Figure S5B, SDC, http://links.lww.com/TP/B645) in EVR- and RAPA-treated Treg cells, which, coincidently, displayed equivalent measurements in mitochondrial mass, mΔψ, and autophagy (not shown).
The allosteric mTOR inhibitors RAPA and EVR are increasingly used in transplantation to minimize the dosage of CNIs in an attempt to reduce the risk of nephrotoxicity and incidence of malignancy.39-41 Comparative pharmacokinetics suggest that EVR exhibits greater intestinal absorption compared with RAPA.42 The relative hydrophobicity of RAPA makes it readily absorbed through the skin and is used in custom topical preparations.43 In contrast, RAPA systemic clearance is half that of EVR,44 , 45 which allows for EVR to reach faster steady-state levels after the initiation of treatment and faster elimination after withdrawal. To date, there are no clinical trials directly comparing EVR and RAPA in cancer therapy or transplant, and there is limited literature on the characterization of the effects induced by EVR on T cells. A comparative study by Roat et al46 among liver transplant patients under CsA or EVR revealed that patients taking EVR had a higher percentage of total and naïve CD4+ T cells than those treated with CsA, a lower percentage and functional response of CD8+ T cells, and a higher percentage of Treg cells. Levistky et al47 reported a significant amplification of newly generated and natural Treg cells in a mixed lymphocyte culture with EVR compared with mycophelonate, RAPA, and tacrolimus. Huijts et al26 showed that mTOR inhibition by RAPA or EVR increased the immunosuppressive capacity of the total Treg cell enriched population caused by the increased frequency of Treg cells but not by the alteration of the suppressive activity per cell.
Here we confirmed the advantage of adding rapalogs for the ex vivo expansion of functional, clinical grade Treg cells and performed a comparative assessment of mechanisms of action and efficacy of EVR and RAPA. Our results demonstrated a similar efficacy of EVR and RAPA to expand functional Treg cells, although EVR treatment showed an early delay in cell growth (Figure S8, SDC, http://links.lww.com/TP/B645). During this early phase, both drugs reduced the glycolytic rates in Treg cells, but only RAPA enhanced the mitochondrial OxPhos activity compared with untreated cells (Figures 2 and 3). The oligomycin-induced inhibition of OxPhos activity resulted in the rapid metabolic shift from OxPhos to aerobic glycolysis (Figure S5, SDC, http://links.lww.com/TP/B645), which was further corroborated in separate ECAR experiments (Figure 3E). This oligomycin-dependent increase in glycolytic rates illustrates the metabolic plasticity of Treg cells previously suggested by Procaccini et al35 and is consistent with the ability of expanding Treg cells to use glucose as a substrate for both mitochondrial respiration and aerobic glycolysis33 , 34 even in long-term expanded Treg cells (Figure S4B, SDC, http://links.lww.com/TP/B645). In contrast, the glycolytic rates remained steady in Tconv cell upon mitochondrial OxPhos inhibition. The robust spare respiratory capacity and high glycolytic reserve levels in proliferating Treg cells add further evidence of the cells’ metabolic adaptability to sustain their intracellular ATP demand. This bioenergetic plasticity may allow Treg cells to experience temporary metabolic stress without triggering cell death in a similar way as reported in some cancer cells.48 Our results also revealed qualitative differences between Treg cells and Tconv cells in OxPhos substrate utilization, as reflected by the major contribution of FA in Treg cells and non-FA in Tconv cells. The inability of Tconv cells to oxidize glucose suggests the use of alternative non-FA, likely amino acids,49-50 as preferential mitochondrial substrates. In the context of the ex vivo expansion of Treg cells, with the cells growing under conditions of unlimited nutrient availability, RAPA-treated cells appear to exhibit more active metabolism than EVR-treated cells during the early culture phase, which may account for their faster cell expansion growth seen in our study. As suggested for CD8+ memory T cells,49 the increase in mitochondrial mass (Figure 4A) may contribute to the higher oxidative and glycolytic capacity of RAPA-treated Treg cells.
The differences between EVR and RAPA in the regulation of the Treg cell metabolic responses were associated with a different pattern of mTOR signaling activation. Both drugs produced similar attenuation on the phosphorylation of 4EBP1 and p70S6k, 2 main downstream effectors of mTORC1, as well as similar compensatory overactivation of ERK. However, the balance between mTORC1 and mTORC2 activities was differently perturbed; although both treatments increased the expression levels of mTORC2-dependent phosphorylation of AKT in Ser-473 when compared with untreated control cells, the increment was significantly lower in EVR-treated cells. The reduced total AKT expression in EVR-treated cells is likely contributing to the partial overactivation of AKT. However, we cannot discard the participation of different feedback loops and/or compensatory mechanisms within the complex PI3K/AKT/mTOR signaling cluster.27 Because the activation of mTORC2-dependent AKT is a critical marker for increased glycolysis in T cells51 and in agreement with the mTORC2 necessary role in cell growth, proliferation, and survival,28-30 we can rationally speculate a functional link between the weaker activation of AKT, the reduced glycolytic phenotype, and the slower proliferative rates in the early stages of EVR-treated Treg cell culture. In contrast, the suppressive function or the phenotypic profile of expanding Treg cells were independent of this metabolic and signaling fluctuations, which is consistent with the different molecular circuitries that regulate expansion and suppressor activity in de novo differentiated FoxP3+ Treg cells described in a previous study.25 This possibility is also in line with the direct regulation of the suppressive function of Treg cells by mTORC1.31 , 32
The increased activity of mTORC1 is generally perceived as a potent inhibitor of autophagy52 and, consequently, the mTORC1-inhibitor RAPA is widely used to induce autophagy.37 Consistent with the same mTORC1 inhibitory capacity of RAPA and EVR in Treg cells, both drugs induced similar increase of autophagic flux. A potential cause for the increased autophagosome formation in RAPA-treated Treg cells may rely on their high mitochondrial mass.53 , 54 The combination of high autophagosome formation and mitochondrial mass raised the possibility of autophagic stress in RAPA-treated cells. However, long-term cell viability, expansion, phenotype, and function did not differ between EVR and RAPA treatments, suggesting that the differences in autophagy processes did not exceed a threshold value to elicit any measurable damaging effect in the cell. In this context, the metabolic plasticity of Treg cells and their ability to redirect the energy metabolism toward glycolysis may also contribute to minimizing the potential damage associated with high mitochondrial OxPhos activity.55 Additional evidence against the likelihood of autophagy stress induction in RAPA-treated Treg cells was generated. First, we previously reported that autophagy-deficient Treg cells exhibit a significant decrease in ΔΨm as well as metabolic and functional deficits25; in contrast, none of these parameters were similarly altered in the current study after the exposure of Treg cells to RAPA. Importantly, the suppressor activity was consistently equivalent between both treatments throughout the ex vivo expansion process. Second, our findings suggest that the differences induced by EVR and RAPA in day 5 of the ex vivo culture are temporary, as evidenced by the subsequent expansion rates as well as the phenotypic, functional, and metabolic profile progressions of both cell treatments.
In the absence of EVR or RAPA, the expansion of Treg cells, even in our conditions of low activation, may produce a significant degree of contaminant CD8+ non-Treg cells. In addition, the fact that the expression levels of standard Treg cell markers such as CD25 and FoxP3 are low in untreated CD4+ T cells brings into question the degree of purity of these Treg cells, further unsettled by the higher methylation status of the TSDR-FOXP3 region. In the absence of Treg cell-specific membrane markers, the discrimination between effector and Treg cells for clinical use may be challenging. From the results generated in this study, we are currently analyzing the link between the different expression of TIGIT and CD36 in untreated and treated Treg cells and their functional capacities. On the other hand, the genomic location of FOXP3 on X chromosome should caution from a sex-biased expression. X chromosome inactivation in female mammals generates a transcriptionally silent inactive X chromosome (Xi) that, in case of FOXP3, remains highly methylated.56-58 However, sex differences remain on the methylation status of the FOXP3 gene even after corrected with a factor of 2. Although there is no evidence to date that FOXP3 is among the immune-related genes that escape X chromosome inactivation in humans,56 , 59 other sex-specific differences are arising with respect to the increased expression of FOXP3 in females, including androgen-dependent sensitivity of FOXP3 expression60 and the potential role of some functional FOXP3 variants.61-66 These sex-specific epigenetic states and regulatory cues are likely to have important implications for understanding sex dimorphic variability of Treg cells in health and disease and strongly support the stratification of the Treg cell studies based on sex.
Similar to our pilot study (NCT03284242), the polyclonal Treg cell yield required in phase I/II clinical trials is in the range of 1 to 10 × 108 cells. We choose to expand the cells in a rather low activation regimen (relative low dose of IL2 and bead concentration) to reduce the response of potential contaminating effector cells. Addition of RAPA or EVR will further support this purpose while allowing an expansion rate of 80 times the original cell yield. From the initial leukapheresis product, we obtain a number of CD25+ Treg cells ranging from 50 to 80 × 106, which allow us to reach (and exceed) the intended Treg cell number. In agreement with others,20-22 our results suggest that the absence of rapalogs in the expansion cell culture may represent a significant risk of contamination with unwanted non-Treg cells.
Our results also revealed the efficacy of ex vivo EVR Treg cell expansion and support EVR as a potential alternative to RAPA in the generation of clinical grade Treg cells. We reported several novel key findings regarding the distinct mechanisms of action of EVR in short-term cultured Treg cells, including the lower mTORC2 activity associated with an overall reduced metabolism and slow early expansion rates compared with RAPA. This initial EVR-Treg cell expansion delay, however, was overcome at later stages, and both RAPA and EVR treatments produced a similar number of competent Treg cells with equivalent phenotype and functional suppressive activity. Overall, our findings support the implementation of a common immunosuppressive EVR-based regimen in the transplant patient that includes the adoptive infusion of ex vivo EVR-expanded autologous Treg cells.
1. Elens L, Bouamar R, Shuker N, et al. Clinical implementation of pharmacogenetics in kidney transplantation: calcineurin inhibitors in the starting blocks. Br J Clin Pharmacol. 2014;77:715–728.
2. Bamoulid J, Staeck O, Halleck F, et al. The need for minimization strategies: current problems of immunosuppression. Transpl Int. 2015;28:891–900.
3. Ballou LM, Lin RZ. Rapamycin and mTOR kinase inhibitors. J Chem Biol. 2008;1:27–36.
4. Budde K, Becker T, Arns W, et al. Everolimus-based, calcineurin-inhibitor-free regimen in recipients of de-novo kidney transplants: an openlabel, randomised, controlled trial. Lancet. 2011;377:837–847.
5. Zuckermann A, Wang SS, Epailly E, et al. Everolimus immunosuppression in de novo heart transplant recipients: what does the evidence tell us now? Transplant Rev (Orlando). 2013;27:76–84.
6. Yee ML, Tan HH. Use of everolimus in liver transplantation. World J Hepatol. 2017;9:990–1000.
7. Sterneck M, Kaiser GM, Heyne N, et al. Long-term follow-up of five yr shows superior renal function with everolimus plus early calcineurin inhibitor withdrawal in the PROTECT randomized liver transplantation study. Clin Transplant. 2016;30:741–748.
8. Lin M, Mittal S, Sahebjam F, et al. Everolimus with early withdrawal or reduced-dose calcineurin inhibitors improves renal function in liver transplant recipients: a systematic review and meta-analysis. Clin Transplant. 2017;31:e12872.
9. Saliba F, De Simone P, Nevens F, et al. Renal function at two years in liver transplant patients receiving everolimus: results of a randomized, multicenter study. Am J Transplant. 2013;13:1734–1745.
10. De Simone P, Nevens F, De Carlis L, et al. Everolimus with reduced tacrolimus improves renal function in de novo liver transplant recipients: a randomized controlled trial. Am J Transplant. 2012;12:3008–3020.
11. Fischer L, Klempnauer J, Beckebaum S, et al. A randomized, controlled study to assess the conversion from calcineurin-inhibitors to everolimus after liver transplantation—PROTECT. Am J Transplant. 2012;12:1855–1865.
12. Fischer L, Saliba F, Kaiser GM, et al. Three-year outcomes in de novo liver transplant patients receiving Everolimus with reduced tacrolimus: follow-up results from a randomized, multicenter study. Transplantation. 2015;99:1455–1462.
13. Sakaguchi S, Yamaguchi T, Nomura T, et al. Regulatory T cells and immune tolerance. Cell. 2008;133:775–787.
14. Li Y, Koshiba T, Yoshizawa A, et al. Analyses of peripheral blood mononuclear cells in operational tolerance after pediatric living donor liver transplantation. Am J Transplant. 20044:2118–2125.
15. Demirkiran A, Kok A, Kwekkeboom J, et al. Low circulating regulatory T-cell levels after acute rejection in liver transplantation. Liver Transpl. 2006;12:277–284.
16. Hanidziar D, Koulmanda M. Inflammation and the balance of Treg and Th17 cells in transplant rejection and tolerance. Curr Opin Organ Transplant. 2010;15:411–415.
17. Wang Y, Zhang M, Liu ZW, et al. The ratio of circulating regulatory T cells (Tregs)/Th17 cells is associated with acute allograft rejection in liver transplantation. PLoS One. 2014;9:e112135.
18. Teshima T, Maeda Y, Ozaki K. Regulatory T cells and IL-17-producing cells in graft-versus-host disease. Immunotherapy. 2011;3:833–852.
19. Lam AJ, Hoeppli RE, Levings MK. Harnessing advances in T regulatory cell biology for cellular therapy in transplantation. Transplantation. 2017;101:2277–2287.
20. Battaglia M, Stabilini A, Roncarolo MG. Rapamycin selectively expands CD4+ CD25+ FoxP3+ regulatory T cells. Blood. 2005;105:4743–4748.
21. Strauss L, Whiteside TL, Knights A, et al. Selective survival of naturally occurring human CD4 + CD25 + Foxp3 + regulatory T cells cultured with rapamycin. J Immunol. 2007;178:320–329.
22. Fraser H, Safinia N, Grageda N, et al. A rapamycin-based GMP-compatible process for the isolation and expansion of regulatory T cells for clinical trials. Mol Ther Methods Clin Dev. 2018;8:198–209.
23. Vilchez V, Turcios L, Butterfield DA, et al. Evidence of the immunomodulatory role of dual PI3K/mTOR inhibitors in transplantation: an experimental study in mice. Transpl Int. 2017;30:1061–1074.
24. Tresoldi E, Dell’Albani I, Stabilini A, et al. Stability of human rapamycin-expanded CD4 + CD25 + T regulatory cells. Haematologica. 2011;96:1357–1365.
25. Ellis GI, Zhi L, Akundi R, et al. Mitochondrial and cytosolic roles of PINK1 shape induced regulatory T-cell development and function. Eur J Immunol. 2013;43:3355–3360.
26. Huijts CM, Santegoets SJ, Quiles Del Rey M, et al. Differential effects of inhibitors of the PI3K/mTOR pathway on the expansion and functionality of regulatory T cells. Clin Immunol. 2016;168:47–54.
27. Coquillard C, Vilchez V, Marti F, et al. mTOR signaling in regulatory T cell differentiation and expansion. SOJ Immunol. 2015;3:1–10.
28. Zou Z, Chen J, Liu A, et al. mTORC2 promotes cell survival through c-Myc–dependent up-regulation of E2F1. J Cell Biol. 201511:105–122.
29. Eltschinger S, Loewith R. TOR complexes and the maintenance of cellular homeostasis. Trends Cell Biol. 2016;26:148–159.
30. Gaubitz C, Prouteau M, Kusmider B, et al. TORC2 structure and function. Trends Biochem Sci. 2016;41:532–545.
31. Park Y, Jin HS, Lopez J, et al. TSC1 regulates the balance between effector and regulatory T cells. J Clin Invest. 2013;123:5165–5178.
32. Zeng H, Yang K, Cloer C, et al. mTORC1 couples immune signals and metabolic programming to establish T(reg)-cell function. Nature. 2013;499:485–490.
33. Gerriets VA, Kishton RJ, Nichols AG, et al. Metabolic programming and PDHK1 control CD4+ T cell subsets and inflammation. J Clin Invest. 2015;125:194–207.
34. Michalek RD, Gerriets VA, Jacobs SR, et al. Cutting edge: distinct glycolytic and lipid oxidative metabolic programs are essential for effector and regulatory CD4+ T cell subsets. J Immunol. 2011;186:3299–3303.
35. Procaccini C, Carbone F, Di Silvestre D, et al. The proteomic landscape of human ex vivo regulatory and conventional T cells reveals specific metabolic requirements. Immunity. 2016;44:406–442. [Published erratum appears in Immunity
36. Claus C, Chey S, Heinrich S, et al. Involvement of p32 and microtubules in alteration of mitochondrial functions by rubella virus. J Virol. 2011;85:3881–3892.
37. Harder LM, Bunkenborg J, Andersen JS. Inducing autophagy: a comparative phosphoproteomic study of the cellular response to ammonia and rapamycin. Autophagy. 2014;10:339–355.
38. Polansky JK, Kretschmer K, Freyer J, et al. DNA methylation controls Foxp3 gene expression. Eur J Immunol. 2008;38:1654–1663.
39. Sumethkul V, Tankee P, Worawichawong S, et al. Ten-year follow-up of parmacokinetics-guided very early cyclosporine minimization synchronized with everolimus initiation in de novo kidney transplantation. Transplant Proc. 2017;49:1743–1746.
40. Yanik EL, Siddiqui K, Engels EA. Sirolimus effects on cancer incidence after kidney transplantation: a meta-analysis. Cancer Med. 2015;4:1448–1459.
41. Knoll GA, Kokolo MB, Mallick R, et al. Effect of sirolimus on malignancy and survival after kidney transplantation: systematic review and meta-analysis of individual patient data. BMJ. 2014;349:g667.
42. Lamoureux F, Picard N, Boussera B, et al. Sirolimus and everolimus intestinal absorption and interaction with calcineurin inhibitors: a differential effect between cyclosporine and tacrolimus. Fundam Clin Pharmacol. 2012;26:463–472.
43. Rauktys A, Lee N, Lee L, et al. Topical rapamycin inhibits tuberous sclerosis tumor growth in a nude mouse model. BMC Dermatol. 2008;8:1.
44. Kirchner GI, Meier-Wiedenbach I, Manns MP. Clinical pharmacokinetics of everolimus. Clin Pharmacokinet. 2004;43:83–95.
45. Moes DJ, Press RR, den Hartigh J, et al. Population pharmacokinetics and pharmacogenetics of everolimus in renal transplant patients. Clin Pharmacokinet. 2012;51:467–480.
46. Roat E, De Biasi S, Bertoncelli L, et al. Immunological advantages of everolimus versus cyclosporin a in liver-transplanted recipients, as revealed by polychromatic flow cytometry. Cytometry A. 2012;81:303–311.
47. Levitsky J, Miller J, Huang X, et al. Immunoregulatory effects of everolimus on in vitro alloimmune responses. PLoS One. 2016;11:e0156535.
48. Hao W, Chang C-PB, Tsao C-C, et al. Oligomycin-induced bioenergetic adaptation in cancer cells with heterogeneous bioenergetic organization. J Biol Chem. 2010;285:12647–12654.
49. van der Windt GJ, Pearce EL. Metabolic switching and fuel choice during T-cell differentiation and memory development. Immunol Rev. 2012;249:27–42.
50. Klysz D, Tai X, Robert PA, Craveiro M, et al. Glutamine-dependent α-ketoglutarate production regulates the balance between T helper 1 cell and regulatory T cell generation. Sci Signal. 2015;8:ra97.
51. Wahl DR, Byersdorfer CA, Ferrara JL, et al. Distinct metabolic programs in activated T cells: opportunities for selective immunomodulation. Immunol Rev. 2012;249:104–115.
52. Noda T, Ohsumi Y. Tor, a phosphatidylinositol kinase homologue, controls autophagy in yeast. J Biol Chem. 1998;273:3963–3966.
53. Hailey DW, Rambold AS, Satpute-Krishnan P, et al. Mitochondria supply membranes for autophagosome biogenesis during starvation. Cell. 2010;141:656–667.
54. Cook KL, Soto-Pantoja DR, Abu-Asab M, et al. Mitochondria directly donate their membrane to form autophagosomes during a novel mechanism of parkin-associated mitophagy. Cell Biosci. 2014;4:16.
55. Jeong DW, Kim TS, Cho IT, et al. Modification of glycolysis affects cell sensitivity to apoptosis induced by oxidative stress and mediated by mitochondria. Biochem Biophys Res Commun. 2004;313:984–991.
56. Carrel L, Willard HF. X-inactivation profile reveals extensive variability in X-linked gene expression in females. Nature. 2005;434:400–404.
57. Fontenot JD, Rasmussen JP, Williams LM, et al. Regulatory T cell lineage specification by the forkhead transcription factor foxp3. Immunity. 2005;2:329–341.
58. Cottrell S, Jung K, Kristiansen G, et al. Discovery and validation of 3 novel DNA methylation markers of prostate cancer prognosis. J Urol. 2007;177:1753–1758.
59. Wang J, Syrett CM, Kramer MC, et al. Unusual maintenance of X chromosome inactivation predisposes female lymphocytes for increased expression from the inactive X. Proc Natl Acad Sci U S A. 2016;113:E2029–38.
60. Walecki M, Eisel F, Klug J, et al. Androgen receptor modulates Foxp3 expression in CD4+CD25+Foxp3+ regulatory T-cells. Mol Biol Cell. 2015;26:2845–2857.
61. Lan Y, Tang XS, Qin J, et al. Association of transcription factor FOXP3 gene polymorphism with genetic susceptibility to systematic lupus erythematosus in Guangxi Zhuang population. Zhonghua Yi Xue Yi Chuan Xue Za Zhi. 2010;27:433–436.
62. Wu Z, You Z, Zhang C, et al. Association between functional polymorphisms of Foxp3 gene and the occurrence of unexplained recurrent spontaneous abortion in a Chinese Han population. Clin Dev Immunol. 2012896458.
63. Yang Q, Chen Y, Yong W. FOXP3 genetic variant and risk of acute coronary syndrome in Chinese Han population. Cell Biochem Funct. 2013;31:599–602.
64. Jahan P, Cheruvu R, Tippisetty S, et al. Association of FOXP3 (rs3761548) promoter polymorphism with nondermatomal vitiligo: a study from India. J Am Acad Dermatol. 2013;69:262–266.
65. Jafarzadeh A, Jamali M, Mahdavi R, et al. Circulating levels of interleukin-35 in patients with multiple sclerosis: evaluation of the influences of FOXP3 gene polymorphism and treatment program. J Mol Neurosci. 2015;55:891–897.
66. Gholami M, Esfandiary A, Vatanparast M, et al. Genetic variants and expression study of FOXP3 gene in acute coronary syndrome in Iranian patients. Cell Biochem Funct. 2016;34:158–162.