Article | Published:

Single-cell gene expression reveals a landscape of regulatory T cell phenotypes shaped by the TCR

Nature Immunologyvolume 19pages291301 (2018) | Download Citation



CD4+ T regulatory cells (Treg) are central to immune homeostasis, their phenotypic heterogeneity reflecting the diverse environments and target cells that they regulate. To understand this heterogeneity, we combined single-cell RNA-seq, activation reporter and T cell receptor (TCR) analysis to profile thousands of Treg or conventional CD4+FoxP3 T cells (Tconv) from mouse lymphoid organs and human blood. Treg and Tconv pools showed areas of overlap, as resting ‘furtive’ Tregs with overall similarity to Tconvs or as a convergence of activated states. All Tregs expressed a small core of FoxP3-dependent transcripts, onto which additional programs were added less uniformly. Among suppressive functions, Il2ra and Ctla4 were quasiconstant, inhibitory cytokines being more sparsely distributed. TCR signal intensity did not affect resting/activated Treg proportions but molded activated Treg programs. The main lines of Treg heterogeneity in mice were strikingly conserved in human blood. These results reveal unexpected TCR-shaped states of activation, providing a framework to synthesize previous observations of Treg heterogeneity.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Change history

  • 03 May 2018

    In the version of this article initially published, the Supplementary Note was missing. The Supplementary Note has now been provided online and is cited in the Methods section of the article. The error has been corrected in the HTML and PDF version of the article.


  1. 1.

    Josefowicz, S. Z., Lu, L. F. & Rudensky, A. Y. Regulatory T cells: mechanisms of differentiation and function. Annu. Rev. Immunol. 30, 531–564 (2012).

  2. 2.

    Panduro, M., Benoist, C. & Mathis, D. Tissue Tregs. Annu. Rev. Immunol. 34, 609–633 (2016).

  3. 3.

    Campbell, D. J. & Koch, M. A. Phenotypical and functional specialization of FOXP3+ regulatory T cells. Nat. Rev. Immunol. 11, 119–130 (2011).

  4. 4.

    Luo, C. T. & Li, M. O. Transcriptional control of regulatory T cell development and function. Trends. Immunol. 34, 531–539 (2013).

  5. 5.

    Vignali, D. A., Collison, L. W. & Workman, C. J. How regulatory T cells work. Nat. Rev. Immunol. 8, 523–532 (2008).

  6. 6.

    Koch, M. A. et al. The transcription factor T-bet controls regulatory T cell homeostasis and function during type 1 inflammation. Nat. Immunol. 10, 595–602 (2009).

  7. 7.

    Linterman, M. A. et al. Foxp3+ follicular regulatory T cells control the germinal center response. Nat. Med. 17, 975–982 (2011).

  8. 8.

    Miyara, M. et al. Functional delineation and differentiation dynamics of human CD4+ T cells expressing the FoxP3 transcription factor. Immunity 30, 899–911 (2009).

  9. 9.

    Kleinewietfeld, M. et al. CCR6 expression defines regulatory effector/memory-like cells within the CD25+CD4+ T-cell subset. Blood 105, 2877–2886 (2005).

  10. 10.

    Smigiel, K. S. et al. CCR7 provides localized access to IL-2 and defines homeostatically distinct regulatory T cell subsets. J. Exp. Med. 211, 121–136 (2014).

  11. 11.

    Cretney, E., Kallies, A. & Nutt, S. L. Differentiation and function of Foxp3+ effector regulatory T cells. Trends. Immunol. 34, 74–80 (2013).

  12. 12.

    Hill, J. A. et al. Foxp3 transcription-factor-dependent and -independent regulation of the regulatory T cell transcriptional signature. Immunity 27, 786–800 (2007).

  13. 13.

    Gavin, M. A. et al Single-cell analysis of normal and FOXP3-mutant human T cells: FOXP3 expression without regulatory T cell development. Proc. Natl Acad. Sci. USA 103, 6659–6664 (2006).

  14. 14.

    Miyao, T. et al. Plasticity of Foxp3+ T cells reflects promiscuous Foxp3 expression in conventional T cells but not reprogramming of regulatory T cells. Immunity 36, 262–275 (2012).

  15. 15.

    Rubtsov, Y. P. et al. Stability of the regulatory T cell lineage in vivo. Science 329, 1667–1671 (2010).

  16. 16.

    Zemmour, D., Pratama, A., Loughhead, S. M., Mathis, D. & Benoist, C. Flicr, a long noncoding RNA, modulates Foxp3 expression and autoimmunity. Proc. Natl Acad. Sci. USA 114, E3472–E3480 (2017).

  17. 17.

    Lathrop, S. K. et al. Peripheral education of the immune system by colonic commensal microbiota. Nature 478, 250–254 (2011).

  18. 18.

    Li, M. O. & Rudensky, A. Y. T cell receptor signalling in the control of regulatory T cell differentiation and function. Nat. Rev. Immunol. 16, 220–233 (2016).

  19. 19.

    Levine, A. G., Arvey, A., Jin, W. & Rudensky, A. Y. Continuous requirement for the TCR in regulatory T cell function. Nat. Immunol. 15, 1070–1078 (2014).

  20. 20.

    Hashimshony, T., Wagner, F., Sher, N. & Yanai, I. CEL-Seq: single-cell RNA-Seq by multiplexed linear amplification. Cell Rep. 2, 666–673 (2012).

  21. 21.

    Klein, A. M. et al. Droplet barcoding for single-cell transcriptomics applied to embryonic stem cells. Cell 161, 1187–1201 (2015).

  22. 22.

    Wagner, A., Regev, A. & Yosef, N. Revealing the vectors of cellular identity with single-cell genomics. Nat. Biotechnol. 34, 1145–1160 (2016).

  23. 23.

    Zilionis, R. et al. Single-cell barcoding and sequencing using droplet microfluidics. Nat. Protoc. 12, 44–73 (2017).

  24. 24.

    Sawant, D. V. & Vignali, D. A. Once a Treg, always a Treg? Immunol. Rev. 259, 173–191 (2014).

  25. 25.

    van der Maaten, L. & Hinton, G. Visualizing high-dimensional data using t-SNE. J. Mach. Learn. Res. 9, 2579–2605 (2008).

  26. 26.

    Kim, H. J. et al. Stable inhibitory activity of regulatory T cells requires the transcription factor Helios. Science 350, 334–339 (2015).

  27. 27.

    Sage, P. T. & Sharpe, A. H. T follicular regulatory cells. Immunol. Rev. 271, 246–259 (2016).

  28. 28.

    Wakamatsu, E., Mathis, D. & Benoist, C. Convergent and divergent effects of costimulatory molecules in conventional and regulatory CD4+ T cells. Proc. Natl. Acad. Sci. USA 110, 1023–1028 (2013).

  29. 29.

    Samstein, R. M. et al. Foxp3 exploits a pre-existent enhancer landscape for regulatory T cell lineage specification. Cell 151, 153–166 (2012).

  30. 30.

    Kitagawa, Y. et al. Guidance of regulatory T cell development by Satb1-dependent super-enhancer establishment. Nat. Immunol. 18, 173–183 (2017).

  31. 31.

    Gavin, M. A., Clarke, S. R., Negrou, E., Gallegos, A. & Rudensky, A. Homeostasis and anergy of CD4+CD25+ suppressor T cells in vivo. Nat. Immunol. 3, 33–41 (2002).

  32. 32.

    Meredith, M., Zemmour, D., Mathis, D. & Benoist, C. Aire controls gene expression in the thymic epithelium with ordered stochasticity. Nat. Immunol. 16, 942–949 (2015).

  33. 33.

    Hennig, C. Cluster-wise assessment of cluster stability. Comput. Stat. Data. Anal. 52, 258–271 (2007).

  34. 34.

    Cretney, E. et al. The transcription factors Blimp-1 and IRF4 jointly control the differentiation and function of effector regulatory T cells. Nat. Immunol. 12, 304–311 (2011).

  35. 35.

    Feng, Y. et al. A mechanism for expansion of regulatory T-cell repertoire and its role in self-tolerance. Nature 528, 132–136 (2015).

  36. 36.

    Liu, Y. et al. Inhibition of p300 impairs Foxp3+ T regulatory cell function and promotes antitumor immunity. Nat. Med. 19, 1173–1177 (2013).

  37. 37.

    Roychoudhuri, R. et al. BACH2 represses effector programs to stabilize T(reg)-mediated immune homeostasis. Nature 498, 506–510 (2013).

  38. 38.

    Ouyang, W. et al. Novel Foxo1-dependent transcriptional programs control T(reg) cell function. Nature 491, 554–559 (2012).

  39. 39.

    Jaitin, D. A. et al. Massively parallel single-cell RNA-seq for marker-free decomposition of tissues into cell types. Science 343, 776–779 (2014).

  40. 40.

    Shalek, A. K. et al. Single-cell RNA-seq reveals dynamic paracrine control of cellular variation. Nature 510, 363–369 (2014).

  41. 41.

    Moran, A. E. et al. T cell receptor signal strength in Treg and iNKT cell development demonstrated by a novel fluorescent reporter mouse. J. Exp. Med. 208, 1279–1289 (2011).

  42. 42.

    Pesenacker, A. M. et al. A regulatory T-cell gene signature is a specific and sensitive biomarker to identify children with new-onset type 1 diabetes. Diabetes 65, 1031–1039 (2016).

  43. 43.

    Fontenot, J. D., Rasmussen, J. P., Gavin, M. A. & Rudensky, A. Y. A function for interleukin 2 in Foxp3-expressing regulatory T cells. Nat. Immunol. 6, 1142–1151 (2005).

  44. 44.

    Mahmud, S. A. et al. Costimulation via the tumor-necrosis factor receptor superfamily couples TCR signal strength to the thymic differentiation of regulatory T cells. Nat. Immunol. 15, 473–481 (2014).

  45. 45.

    Liu, Z. et al. Immune homeostasis enforced by co-localized effector and regulatory T cells. Nature 528, 225–230 (2015).

  46. 46.

    Beyer, M. et al. Repression of the genome organizer SATB1 in regulatory T cells is required for suppressive function and inhibition of effector differentiation. Nat. Immunol. 12, 898–907 (2011).

  47. 47.

    Zhou, X. et al. Instability of the transcription factor Foxp3 leads to the generation of pathogenic memory T cells in vivo. Nat. Immunol. 10, 1000–1007 (2009).

  48. 48.

    Sefik, E. et al. Individual intestinal symbionts induce a distinct population of RORγ+ regulatory T cells. Science 349, 993–997 (2015).

  49. 49.

    Malissen, B. & Bongrand, P. Early T cell activation: integrating biochemical, structural, and biophysical cues. Annu. Rev. Immunol. 33, 539–561 (2015).

  50. 50.

    Ashouri, J. F. & Weiss, A. Endogenous Nur77 is a specific indicator of antigen receptor signaling in human T and B cells. J. Immunol. 198, 657–668 (2017).

  51. 51.

    Bettelli, E. et al. Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells. Nature 441, 235–238 (2006).

  52. 52.

    Liston, A. et al Differentiation of regulatory Foxp3+ T cells in the thymic cortex. Proc. Natl. Acad. Sci. USA 105, 11903–11908 (2008).

  53. 53.

    Ferraro, A. et al Interindividual variation in human T regulatory cells. Proc. Natl Acad. Sci. USA 111, E1111–E1120 (2014).

  54. 54.

    Kim, D. et al. TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome. Biol. 14, R36 (2013).

  55. 55.

    Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome. Biol. 15, 550 (2014).

  56. 56.

    Grün, D. et al. Single-cell messenger RNA sequencing reveals rare intestinal cell types. Nature 525, 251–255 (2015).

  57. 57.

    Krijthe, J. H. Rtsne: T-Distributed Stochastic Neighbor Embedding using a Barnes–Hut Implementation. GitHub (2015).

  58. 58.

    Hennig, C. fpc: Flexible Procedures for Clustering. R package version 2.1-10. Comprehensive R Archive Network (2015).

  59. 59.

    Godec, J. et al. Compendium of immune signatures identifies conserved and species-specific biology in response to inflammation. Immunity 44, 194–206 (2016).

  60. 60.

    Baker, F. J., Lee, M., Chien, Y. H. & Davis, M. M. Restricted islet-cell reactive T cell repertoire of early pancreatic islet infiltrates in NOD mice. Proc. Natl. Acad. Sci. USA 99, 9374–9379 (2002).

  61. 61.

    Burzyn, D. et al. A special population of regulatory T cells potentiates muscle repair. Cell 155, 1282–1295 (2013).

  62. 62.

    Bolotin, D. A. et al. MiXCR: software for comprehensive adaptive immunity profiling. Nat. Methods 12, 380–381 (2015).

Download references


We thank E. Sefik for help with colon profiling and K. Hattori, C. Araneo K. Waraska, M. Thorsen and R. Steen for help with mice, profiling, sorting and sequencing. This work was supported by NIH grants AI051530, AI116834 and AI125603 to C.B./D.M. D.Z. and E.K. were supported by a PhD fellowship from the Boehringer Ingelheim Fonds.

Author information


  1. Division of Immunology, Department of Microbiology and Immunobiology, Harvard Medical School, and Evergrande Center for Immunologic Diseases, Harvard Medical School and Brigham and Women’s Hospital, Boston, MA, USA

    • David Zemmour
    • , Evgeny Kiner
    • , Diane Mathis
    •  & Christophe Benoist
  2. Department of Systems Biology, Harvard Medical School, Boston, MA, USA

    • Rapolas Zilionis
    •  & Allon M. Klein
  3. Institute of Biotechnology, Vilnius University, Vilnius, Lithuania

    • Rapolas Zilionis


  1. Search for David Zemmour in:

  2. Search for Rapolas Zilionis in:

  3. Search for Evgeny Kiner in:

  4. Search for Allon M. Klein in:

  5. Search for Diane Mathis in:

  6. Search for Christophe Benoist in:


D.Z., E.K. and R.Z. performed the experiments; D.Z., A.M.K., C.B. and D.M. designed the study, and analyzed and interpreted the data; D.Z., D.M. and C.B. wrote the manuscript.

Competing interests

The authors declare no competing interests.

Corresponding authors

Correspondence to Diane Mathis or Christophe Benoist.

Integrated supplementary information

  1. Supplementary Figure 1 Quality-control plots and algorithm to identify contaminant cells in scRNA-seq.

    a. Scatterplot matrix comparing the collapsed gene expression profiles of the Treg scRNAseq replicates, with Pearson correlation and P value (t test). b. Scatterplot matrix comparing the variability in gene expression (Fano Factor = variance/mean) within each of the different Treg scRNAseq replicates, with Pearson correlation and P value (t test). c. Illustration of the naïve Bayes algorithm used to find contaminant cell in the datasets. We calculated the likelihood for each single cell transcriptome to be drawn from any of the 249 immune cell types profiled by microarrays by the Immgen consortium (Heng, T. S. & Painter, M. W. The Immunological Genome Project: networks of gene expression in immune cells. Nat. Immunol. 9, 1091–1094 (2008)). Venn diagrams: Distribution of the most likely Immgen immune cell type assigned in each independent Tconv and Treg scRNAseq dataset. d. Treg vs Tconv changes in gene expression in pooled scRNAseq (x-axis) compared with the changes in a classic population RNAseq experiment (y-axis). Highlighted genes are genes are the canonical Treg signature genes12.

  2. Supplementary Figure 2 Furtive Tregs can be found in multiple independent datasets and express FoxP3 by flow cytometry.

    a. Distribution of the number of detected genes in Tconvs (in E), Tregs (outside E) and furtive Tregs (in E). E corresponds to the cell cluster delineated in Fig. 2a. Error bars correspond to the mean +- 1 standard deviation. b. Distribution of the number of uniquely identified mRNA reads per cell in Tconvs (in E), Tregs (outside E) and furtive Tregs (in E). E corresponds to the cell cluster delineated in Fig. 2a. Error bars correspond to the mean +- 1 standard deviation. c. Two-dimensional tSNE plot of TCRβ+ CD4+ T cells sequenced using the 10X Genomics platform. Tregs (red) and Tconvs (black) were identified using the naive Bayes algorithm that compares each single-cell transcriptome with each of the 249 immune-cell states profiled by the Immgen Consortium (see Methods). The Tconv area used to identify furtive Tregs is shown along with the proportion of Tregs and Tconvs relative to the total number of Tregs and Tconvs. d. Same tSNE plots as in c, but split to show Treg and Tconv cells separately (top row) and cells in which Foxp3 transcript was detected (bottom row). e. Two-dimensional tSNE plot of index-sorted TCRβ+ CD4+ FoxP3-GFP+ Tregs and FoxP3-GFP- Tconvs sequenced by CEL-Seq. The tSNE is split to show Tregs and Tconvs separately. The Tconv area used to identify furtive Tregs is shown along with the proportion of Tregs and Tconvs relative to the total number of Tregs and Tconvs. f. FoxP3-GFP protein expression (flow cytometry intensity) distribution of indexed-sorted Tconvs, Tregs and furtive Tregs shown in e.

  3. Supplementary Figure 3 FoxP3 binds to the core Treg-signature genes.

    a. Core Treg signature genes at the single cell level are the genes that were found to be consistently differentially expressed (>80%) when we compare a sample of closely related Treg and Tconv in the transcriptional space. For each single Treg cell (n = 708), we compared the expression profiles of its 50 closest Tregs and 50 closest Tconvs (by Pearson correlation distance) and deemed significant genes with a P value <0.05. We report in this plot the average of the gene expression fold change across the 708 comparisons (x axis) and the fraction of tests where this difference was significant (y axis). Canonical Treg signature genes are highlighted in red and blue12. b. Chromatin profiles for core Treg signature genes . Top to bottom: ATAC-seq in Tconvs, ATAC-seq in Tregs, FoxP3 Chip-Seq in Tregs (data from 29).

  4. Supplementary Figure 4 Robustness of the splenic Treg clusters.

    a. Scatterplot of the average gene expression in single Treg cells and variability of expression (Fano factor = variance/mean). Top 100 most variable genes expressed in >1% cells are in red and were used in Fig. 4 for clustering. b. Jaccard index of the Treg clusters after kmeans bootstrapping (measures how many times cells consistently clustered together). c. Silhouette plot of the Treg clusters after kmeans bootstrapping (measure of how close is each single cell to the cells in the same clusters vs to the cells in the neighbor cluster). d. Biclustering expression heatmap, as Fig. 4a, for 2 independent splenic Treg scRNAseq datasets (n = 2035 and n = 1,294 cells). Columns: Single Treg cells assigned to the same 6 clusters as Fig. 4 using a likelihood model (see Methods). Rows: most characteristic genes found in the 6 Treg clusters (per Fig. 4a). The proportion of Tregs in each cluster in indicated. e. Circular projection plot of the likelihood for each single Treg cell in the 2 independent datasets to belong to any of the 6 mouse clusters described in Fig. 4.

  5. Supplementary Figure 5 Signature enrichment in single Treg cell clusters.

    Single-cell Treg expression of various genesets computed or obtained from previous Treg studies (reference included). Differently expressed signatures between clusters are represented here, ordered by hierarchal clustering. Single Treg cells were ordered by clusters (columns). P values by ANOVA.

  6. Supplementary Figure 6 Nr4a1-GFP expression correlates with an enrichment in cluster 3 and an impoverishment in cluster 1 and 2 cells.

    Circular projection plot of the likelihood for each single Treg cells in the Nr4a1-GFP high, medium and low independent datasets to belong to any of the 6 mouse clusters described in Fig. 4 (n = 1000 cells in each plot). The proportion of cells in each cluster is indicated.

  7. Supplementary Figure 7 Human Treg and Tconv single-cell datasets: Quality-control plots and definition of the core human Treg transcripts.

    a. Scatterplot matrix comparing the collapsed gene expression profiles of the Treg and Tconv scRNAseq replicates, with Pearson correlation and P value (t test). b. Scatterplot matrix comparing the variability in gene expression (Fano Factor = variance/mean) within each of the different Treg and Tconv scRNAseq replicates, with Pearson correlation and P value (t test). c. Treg vs Tconv changes in the gene expression in pooled scRNAseq (x-axis) compared with the changes in an independent population microarrray experiment (y-axis)53. Highlighted genes are genes are the canonical Treg signature genes. d. Same tSNE plots of human Treg and Tconv as in Fig. 7b, but split to show the contribution of each donor. e. Core human Treg signature. The same analysis as described in Fig. S2 was performed. This scatterplot compares the average fold change in gene expression (x axis) with the fraction of tests (y axis) that were significantly differently expressed between human Tregs and Tconvs (similar to Supplementary Fig. 3a). f. Nr4a1-GFP expression distribution in Tregs isolated from the blood and spleen of three mice.

Supplementary information

  1. Supplementary Text and Figures

    Supplementary Figures 1–7

  2. Life Sciences Reporting Summary

  3. Supplementary Table 1

    Single-cell datasets created in this study

  4. Supplementary Table 2

    Top 100 most variable genes amongst splenic Tregs

  5. Supplementary Table 3

    Average gene expression in each splenic Treg cluster

  6. Supplementary Table 4

    Differentially expressed genes between splenic Treg clusters

  7. Supplementary Table 5

    DNA primer for TCR sequencing

  8. Supplementary Data

About this article

Publication history





Further reading

  • Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations

    • Sonya A. MacParland
    • , Jeff C. Liu
    • , Xue-Zhong Ma
    • , Brendan T. Innes
    • , Agata M. Bartczak
    • , Blair K. Gage
    • , Justin Manuel
    • , Nicholas Khuu
    • , Juan Echeverri
    • , Ivan Linares
    • , Rahul Gupta
    • , Michael L. Cheng
    • , Lewis Y. Liu
    • , Damra Camat
    • , Sai W. Chung
    • , Rebecca K. Seliga
    • , Zigong Shao
    • , Elizabeth Lee
    • , Shinichiro Ogawa
    • , Mina Ogawa
    • , Michael D. Wilson
    • , Jason E. Fish
    • , Markus Selzner
    • , Anand Ghanekar
    • , David Grant
    • , Paul Greig
    • , Gonzalo Sapisochin
    • , Nazia Selzner
    • , Neil Winegarden
    • , Oyedele Adeyi
    • , Gordon Keller
    • , Gary D. Bader
    •  & Ian D. McGilvray

    Nature Communications (2018)