Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Early specification of CD8+ T lymphocyte fates during adaptive immunity revealed by single-cell gene-expression analyses


T lymphocytes responding to microbial infection give rise to effector cells that mediate acute host defense and memory cells that provide long-lived immunity, but the fundamental question of when and how these cells arise remains unresolved. Here we combined single-cell gene-expression analyses with 'machine-learning' approaches to trace the transcriptional 'roadmap' of individual CD8+ T lymphocytes throughout the course of an immune response in vivo. Gene-expression signatures predictive of eventual fates could be discerned as early as the first T lymphocyte division and may have been influenced by asymmetric partitioning of the receptor for interleukin 2 (IL-2Rα) during mitosis. Our findings emphasize the importance of single-cell analyses in understanding fate determination and provide new insights into the specification of divergent lymphocyte fates early during an immune response to microbial infection.

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Figure 1: Gating strategy and experimental approach.
Figure 2: Effector and memory CD8+ T lymphocyte subsets are molecularly distinct on a single-cell level.
Figure 3: Early heterogeneity of gene expression in individual CD8+ T lymphocytes during an immune response.
Figure 4: Classifier analysis allows prediction of eventual fates of individual CD8+ T lymphocytes.
Figure 5: Temporal model for predicting the differentiation paths of individual CD8+ T lymphocytes.
Figure 6: Asymmetric segregation of IL-2Rα during the division of T lymphocytes influences the eventual fate of daughter cells.

Accession codes

Primary accessions

Gene Expression Omnibus


  1. Ahmed, R. & Gray, D. Immunological memory and protective immunity: understanding their relation. Science 272, 54–60 (1996).

    CAS  PubMed  Google Scholar 

  2. Joshi, N.S. et al. Inflammation directs memory precursor and short-lived effector CD8+ T cell fates via the graded expression of T-bet transcription factor. Immunity 27, 281–295 (2007).

    CAS  PubMed  PubMed Central  Google Scholar 

  3. Masopust, D., Kaech, S.M., Wherry, E.J. & Ahmed, R. The role of programming in memory T-cell development. Curr. Opin. Immunol. 16, 217–225 (2004).

    CAS  PubMed  Google Scholar 

  4. Sallusto, F., Lenig, D., Forster, R., Lipp, M. & Lanzavecchia, A. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 401, 708–712 (1999).

    CAS  PubMed  Google Scholar 

  5. Stemberger, C. et al. A single naive CD8+ T cell precursor can develop into diverse effector and memory subsets. Immunity 27, 985–997 (2007).

    CAS  PubMed  Google Scholar 

  6. Gerlach, C. et al. One naive T cell, multiple fates in CD8+ T cell differentiation. J. Exp. Med. 207, 1235–1246 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  7. Buchholz, V.R. et al. Disparate individual fates compose robust CD8+ T cell immunity. Science 340, 630–635 (2013).

    CAS  PubMed  Google Scholar 

  8. Gerlach, C. et al. Heterogeneous differentiation patterns of individual CD8+ T cells. Science 340, 635–639 (2013).

    CAS  PubMed  Google Scholar 

  9. Chang, J.T. et al. Asymmetric T lymphocyte division in the initiation of adaptive immune responses. Science 315, 1687–1691 (2007).

    CAS  PubMed  Google Scholar 

  10. Chang, J.T. et al. Asymmetric proteasome segregation as a mechanism for unequal partitioning of the transcription factor T-bet during T lymphocyte division. Immunity 34, 492–504 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  11. Kaech, S.M., Hemby, S., Kersh, E. & Ahmed, R. Molecular and functional profiling of memory CD8 T cell differentiation. Cell 111, 837–851 (2002).

    CAS  PubMed  Google Scholar 

  12. Best, J.A. et al. Transcriptional insights into the CD8+ T cell response to infection and memory T cell formation. Nat. Immunol. 14, 404–412 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  13. Sarkar, S. et al. Functional and genomic profiling of effector CD8 T cell subsets with distinct memory fates. J. Exp. Med. 205, 625–640 (2008).

    CAS  PubMed  PubMed Central  Google Scholar 

  14. Guo, G. et al. Resolution of cell fate decisions revealed by single-cell gene expression analysis from zygote to blastocyst. Dev. Cell 18, 675–685 (2010).

    CAS  PubMed  Google Scholar 

  15. Buganim, Y. et al. Single-cell expression analyses during cellular reprogramming reveal an early stochastic and a late hierarchic phase. Cell 150, 1209–1222 (2012).

    CAS  PubMed  PubMed Central  Google Scholar 

  16. Dalerba, P. et al. Single-cell dissection of transcriptional heterogeneity in human colon tumors. Nat. Biotechnol. 29, 1120–1127 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  17. Lu, R., Neff, N.F., Quake, S.R. & Weissman, I.L. Tracking single hematopoietic stem cells in vivo using high-throughput sequencing in conjunction with viral genetic barcoding. Nat. Biotechnol. 29, 928–933 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  18. Ichii, H. et al. Role for Bcl-6 in the generation and maintenance of memory CD8+ T cells. Nat. Immunol. 3, 558–563 (2002).

    CAS  PubMed  Google Scholar 

  19. Kaech, S.M. & Cui, W. Transcriptional control of effector and memory CD8+ T cell differentiation. Nat. Rev. Immunol. 12, 749–761 (2012).

    CAS  PubMed  PubMed Central  Google Scholar 

  20. Kallies, A., Xin, A., Belz, G.T. & Nutt, S.L. Blimp-1 transcription factor is required for the differentiation of effector CD8+ T cells and memory responses. Immunity 31, 283–295 (2009).

    CAS  PubMed  Google Scholar 

  21. Pearce, E.L. et al. Control of effector CD8+ T cell function by the transcription factor Eomesodermin. Science 302, 1041–1043 (2003).

    CAS  PubMed  Google Scholar 

  22. Rutishauser, R.L. et al. Transcriptional repressor Blimp-1 promotes CD8+ T cell terminal differentiation and represses the acquisition of central memory T cell properties. Immunity 31, 296–308 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  23. Szabo, S.J. et al. Distinct effects of T-bet in TH1 lineage commitment and IFN-γ production in CD4 and CD8 T cells. Science 295, 338–342 (2002).

    CAS  PubMed  Google Scholar 

  24. Yang, C.Y. et al. The transcriptional regulators Id2 and Id3 control the formation of distinct memory CD8+ T cell subsets. Nat. Immunol. 12, 1221–1229 (2011).

    CAS  Article  PubMed  Google Scholar 

  25. Zhou, X. et al. Differentiation and persistence of memory CD8+ T cells depend on T cell factor 1. Immunity 33, 229–240 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  26. Warren, L., Bryder, D., Weissman, I.L. & Quake, S.R. Transcription factor profiling in individual hematopoietic progenitors by digital RT-PCR. Proc. Natl. Acad. Sci. USA 103, 17807–17812 (2006).

    CAS  PubMed  PubMed Central  Google Scholar 

  27. Masopust, D., Vezys, V., Marzo, A.L. & Lefrancois, L. Preferential localization of effector memory cells in nonlymphoid tissue. Science 291, 2413–2417 (2001).

    CAS  PubMed  Google Scholar 

  28. Schluns, K.S., Kieper, W.C., Jameson, S.C. & Lefrancois, L. Interleukin-7 mediates the homeostasis of naive and memory CD8 T cells in vivo. Nat. Immunol. 1, 426–432 (2000).

    CAS  PubMed  Google Scholar 

  29. Wherry, E.J. et al. Lineage relationship and protective immunity of memory CD8 T cell subsets. Nat. Immunol. 4, 225–234 (2003).

    CAS  PubMed  Google Scholar 

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

    Google Scholar 

  31. Freund, Y. & Mason, L. The Alternating Decision Tree Learning Algorithm. In Proceedings of the Sixteenth International Conference on Machine Learning (eds. Bratko, I. & Dzeroski, S.) 124–133 (Morgan Kaufmann Publishers, 1999).

  32. Freund, Y. Invited talk: Drifting games, boosting and online learning. In Proceedings of the 26th Annual International Conference on Machine Learning 162, (ACM, 2009).

  33. Beerenwinkel, N. & Drton, M. A mutagenetic tree hidden Markov model for longitudinal clonal HIV sequence data. Biostatistics 8, 53–71 (2007).

    PubMed  Google Scholar 

  34. Bulla, J. & Bulla, I. Stylized facts of financial time series and hidden semi-Markov models. Comput. Stat. Data Anal. 51, 2192–2209 (2006).

    Google Scholar 

  35. Feau, S., Arens, R., Togher, S. & Schoenberger, S.P. Autocrine IL-2 is required for secondary population expansion of CD8+ memory T cells. Nat. Immunol. 12, 908–913 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  36. Williams, M.A., Tyznik, A.J. & Bevan, M.J. Interleukin-2 signals during priming are required for secondary expansion of CD8+ memory T cells. Nature 441, 890–893 (2006).

    CAS  PubMed  PubMed Central  Google Scholar 

  37. Kalia, V. et al. Prolonged interleukin-2Ralpha expression on virus-specific CD8+ T cells favors terminal-effector differentiation in vivo. Immunity 32, 91–103 (2010).

    CAS  PubMed  Google Scholar 

  38. Pipkin, M.E. et al. Interleukin-2 and inflammation induce distinct transcriptional programs that promote the differentiation of effector cytolytic T cells. Immunity 32, 79–90 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  39. Obar, J.J. & Lefrancois, L. Early signals during CD8 T cell priming regulate the generation of central memory cells. J. Immunol. 185, 263–272 (2010).

    CAS  PubMed  Google Scholar 

  40. Afkarian, M. et al. T-bet is a STAT1-induced regulator of IL-12R expression in naive CD4+ T cells. Nat. Immunol. 3, 549–557 (2002).

    CAS  PubMed  Google Scholar 

  41. Lighvani, A.A. et al. T-bet is rapidly induced by interferon-γ in lymphoid and myeloid cells. Proc. Natl. Acad. Sci. USA 98, 15137–15142 (2001).

    CAS  PubMed  PubMed Central  Google Scholar 

  42. Gebhardt, T. et al. Memory T cells in nonlymphoid tissue that provide enhanced local immunity during infection with herpes simplex virus. Nat. Immunol. 10, 524–530 (2009).

    CAS  PubMed  Google Scholar 

  43. Masopust, D. et al. Dynamic T cell migration program provides resident memory within intestinal epithelium. J. Exp. Med. 207, 553–564 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  44. Olson, J.A., McDonald-Hyman, C., Jameson, S.C. & Hamilton, S.E. Effector-like CD8+ T cells in the memory population mediate potent protective immunity. Immunity 38, 1250–1260 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  45. Tang, F. et al. mRNA-Seq whole-transcriptome analysis of a single cell. Nat. Methods 6, 377–382 (2009).

    CAS  PubMed  Google Scholar 

  46. Shalek, A.K. et al. Single-cell transcriptomics reveals bimodality in expression and splicing in immune cells. Nature 498, 236–240 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  47. Bendall, S.C. et al. Single-cell mass cytometry of differential immune and drug responses across a human hematopoietic continuum. Science 332, 687–696 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  48. Moon, J.J. et al. Tracking epitope-specific T cells. Nat. Protoc. 4, 565–581 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

Download references


We thank S. Hedrick, J. Bui, A. Goldrath, S. Schoenberger and members of the Chang and Yeo laboratories for discussions and critical reading of the manuscript. Supported by the US National Institutes of Health (DK080949, OD008469 and AI095277 to J.T.C., and HG004659 and NS075449 to G.W.Y.), the UCSD Digestive Diseases Research Development Center (DK80506), the California Institute for Regenerative Medicine (RB1-01413 and RB3-05009 to G.W.Y.), the National Science Foundation (B.K.), the Alfred P. Sloan Foundation (G.W.Y.) and the Howard Hughes Medical Institute (J.T.C.).

Author information

Authors and Affiliations



J.A. and J.T.C. designed experiments; J.A., P.J.M. and S.H.K. did experiments. B.K. and G.W.Y. analyzed data; and J.A., B.K., G.W.Y. and J.T.C. wrote the manuscript.

Corresponding authors

Correspondence to Gene W Yeo or John T Chang.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Single-cell gene-expression data acquisition.

(a) Fluidigm 96.96 IFC array heatmap representing individual qPCR reactions of 94 Taqman gene expression assays in 94 single CD8+ T cells. Heatmap is representative of at least three independent experiments per cell population. Negative (no template) and positive controls (cDNA from an in vitro stimulated CD8+ T cell bulk population sample) were included in every array. (b) Data matrix containing raw Ct values of all single-cell qRT-PCR reactions used in the analyses. Individual lines along the vertical axis represent single cells. Expression data from 1,300 single cells representing naïve (149 cells), division 1 (144), distal daughter (68), proximal daughter (83), day 3 (143), day 5 (154), day 7 (134), Tmp (62), Tsle (89), Tcm (138), and Tem (136) cells were used for analyses.

Supplementary Figure 2 t-distributed stochastic neighbor embedding (tSNE) analysis.

tSNE reduces the dimensions of a multivariate dataset (94 dimensions for each of the 94 genes in our analysis). Each data point (a single cell) is assigned a location in a two- or three-dimensional map to illustrate potential clusters (populations) of neighboring cells, which contain similar gene expression patterns.

Supplementary Figure 3 Single CD8+ T lymphocytes responding to microbe exhibit the greatest divergence in gene expression early after infection.

(a) The top five genes that drive variance in mean gene expression within and between the indicated CD8+ T lymphocyte populations were identified by JSD analysis. Red bars represent the divergence of each gene. The top five genes that drive intra-population divergence for a single population are listed in the box located at the intersection of that population with itself. For example, the top five genes driving intra-population divergence for naïve cells are Ifngr1, Ptprc, Ccr5, Psmb7, and Sell. The top five genes that drive inter-population divergence between any two cell populations are listed in the box located at the intersection of those two populations. For example, the top five genes driving inter-population divergence for naïve and division 1 cells are Klf2, Lgals1, Irf4, Il2ra, and Myc. (b) The JSD analysis shown in Figure 3b was repeated by sub-sampling the populations so that each pair was compared with equal sizes. This analysis confirmed that the inter-population divergence measurement was not affected by unequal group sizes.

Supplementary Figure 4 Supervised analysis approaches.

(a) Decision tree built from the data consisting of several predictive rules comparing expression of Ptprc, Ccl5, and Sell to decide whether a cell is more Tcm- or Tsle-like; two terminal nodes labeled "…" depict a continuation of the decision tree. The full decision tree is available at: (b) We evaluated the Tcm vs. Tsle misclassification error as a function of the classifier complexity (number of trees in the ensemble). The training error (blue curve) for a single instance of the classifier was calculated on the entire gene expression dataset of 138 Tcm and 89 Tsle cells. The generalization error (red dots) was estimated by the leave-one-out cross-validation procedure. Briefly, each cell in the dataset was set aside and a separate instance of the classifier was trained on the remaining cells, which was used to predict the class of the set-aside cell. The number of held-out cells that were misclassified is reported as the cross-validation error as an approximation to the generalization error of the classifier on future gene expression data from similar cell populations. It is clear that an ensemble of 10 – 20 trees is sufficient to discriminate between Tcm and Tsle cells without overfitting. (c) The significance and robustness of each proposed differentiation path for the HMM model was measured by performing 10 random initializations of the HMM parameters and 100 random shuffles of the data. Each panel compares the cumulative distribution functions (CDFs) of log-likelihoods for the proposed model on the real data (green for linear, blue for bifurcating, and pink for the best performing model) to the CDF of log-likelihoods for the same model on randomly reshuffled data. These CDFs were compared by the 2-sample Kolmogorov-Smirnov test, whose p-value is shown above each panel. The last panel (bottom right) shows the reproducibility of the best model by comparing the log-likelihood CDFs of the model on the original data (pink) versus a bootstrap resampled version of the same data (red). Insets include each proposed differentiation path. Transition states include pre-memory (p-Mem), pre-short-lived-effector (p-Tsle), and common progenitor (Pre-X). (d) Cells in early states of differentiation (division 1, day 3, day 5) were ranked by their Tsle- or memory-like expression profiles. Cells were then linked to sorted naïve and sorted Tsle, Tem and Tcm cells by bootstrap resampling, forming hypothetical differentiation paths that were analyzed with a Hidden Markov Model. Shown is the matrix of probabilities that a CD8+ T cell will transition from one state (vertical axis) to another (horizontal axis).

Supplementary Figure 5 Change in log gene expression associated with each transition phase during specification of CD8+ T lymphocyte fates.

The absolute change in expression of each of the 94 genes during each unique transition is shown: naïve to pre-memory, naïve to pre-Tsle, pre-Tsle to Tsle, pre-memory to Tcm, and pre-memory to Tem.

Supplementary Figure 6 Frequencies of the progeny of adoptively transferred IL-2RαhiCD62Llo and IL-2RαloCD62Lhi cells.

IL-2RαhiCD62Llo or IL-2RαloCD62Lhi cells that had undergone their first division were sorted and adoptively transferred into infection-matched recipients (n=13). The frequencies of transferred CD45.1+ cells were measured in the blood following adoptive transfer and their numbers are shown as a percentage of total CD8+ T cells (a) at day 7 and (b) at multiple timepoints following the primary infection. Data are representative of two independent experiments. Error bars indicate s.e.m. (Kolmogorov-Smirnov test)

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–6 and Supplementary Table 1 (PDF 1257 kb)

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Arsenio, J., Kakaradov, B., Metz, P. et al. Early specification of CD8+ T lymphocyte fates during adaptive immunity revealed by single-cell gene-expression analyses. Nat Immunol 15, 365–372 (2014).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing