Information-dense analysis for information-dense understanding
Editorial

Information-dense analysis for information-dense understanding

Nikola A. Ivica*, Mandeep Kaur*, Guangan Hu, Jianzhu Chen

Koch Institute for Integrative Cancer Research and Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02139, USA

*These authors contributed equally to this work.

Correspondence to: Jianzhu Chen. Koch Institute for Integrative Cancer Research and Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02139, USA. Email: jchen@mit.edu.

Submitted Oct 17, 2016. Accepted for publication Oct 28, 2016.

doi: 10.21037/tcr.2016.11.21


Using mass cytometry coupled with cellular barcoding, Wong et al. (1) report a high-dimensional map of human T-cell trafficking receptors and cytokine profiles in several different tissues. Specifically, the authors analyzed T cells from eight human organs, including healthy donors and patients with different underlying conditions, with a range of three to five individual samples per organ. Perhaps unsurprisingly, their findings highlight the dynamic and complex nature of T cells in different organs and stress the importance of looking beyond peripheral blood when analyzing human T cell phenotype and function. The large amount of data generated by mass cytometry will undoubtedly serve as a reference point for future information-dense studies of human lymphocytes, but the main finding in this paper, the astonishing heterogeneity of T-cell trafficking markers and cytokine expression, suggests that regulation of T cell distribution and function in humans is more complex than previously recognized. Finally, this study complements and corroborates several previous studies, most notably those by Thome et al. (2,3) and Sathaliyawala et al. (4), which demonstrated differential distribution of human memory T cells in different organs and across decades of life. This resource should be especially useful for researchers who aim to engineer T cells for disease interventions, such as cancer immunotherapy.

In accordance with the bulk of literature, Wong et al. show that leukocytes vary in composition in different human organs (5). Cord blood and peripheral blood mononuclear cells (PBMCs) mainly consist of T cells; with most of cord blood T cells being naïve in phenotype, while T cells in PBMCs exhibiting either naïve or memory phenotype. The secondary lymphoid organs tonsil and spleen share a similar make-up consisting predominantly of B cells. In the four non-lymphoid tissues examined (liver, colon, lung, and skin), T cells predominate with most of the T cells displaying an effector memory phenotype (CD45RACCR7) as would be expected for adult tissues. While B cells from all anatomical sites consistently expressed CCR6, CCR7 and CXCR5, T cells from different organs had distinct expression of trafficking receptors. Interestingly, no consensus set of T cell trafficking markers were discernible in any of the organs examined.

The heterogeneity of T cell surface receptors, especially trafficking markers, is striking. Unlike B cells that have the ability to secrete antibodies and act at a distance, T cells need to make a direct contact with target cells to exert their effector functions. One would expect a common set of trafficking receptors expressed on different T cell subtypes from the same tissue. Wong et al. convincingly show that this is not the case. There are several mutually non-exclusive interpretations of this result. One possibility is that T cell subsets differ intrinsically in tissue trafficking. Other possibilities relate to data analysis and the small sample size (3 to 5 donors) per organ from patients with different underlying conditions. In the current study, most of the data (except for trafficking receptors in Supplemental Figure 2) is shown as grouped analysis of all individual samples. The comparison of T cells from various tissues from healthy and diseased donors with different underlying conditions inevitably increased the heterogeneity of T cell markers. In support of this possibility, studies have shown gender-specific differences in the T cell repertoire (6) and the strength of the immune response (7). Similarly, multiple antigen stimulations are known to up-regulate distinct programs in the same T cells, which would contribute to donor-to-donor differences (8). It would be informative to visualize individual donor data for all parameters and analyze how this relates to the averages presented here. In addition, although Wong et al. analyzed 41 trafficking markers, the number may still be insufficient to capture the “area-codes” for different T cell subsets in different tissues (9).

Consistent with previous reports in mice (10), Wong et al. found that most of the tissue-resident memory T cells (TRM) express CD103 and CD69 as well as integrin β7, a subunit of the CD103 complex. Similarly, both human and mouse skin-resident TRM cells express CLA and CCR4 but not CCR10. Interestingly, Wong et al. also found CD69+CD103 TRM cells in different organs, which may be migratory T cells from the circulation that are not completely restricted by the local microenvironment or they may be in the process of conversion to TRM cells outside of their resident tissues. Furthermore, Wong et al. were able to divide both CD4+ and CD8+ TRM cells into four subtypes based on the expression of CLA, HLA-DR and CD25 for CD4+ T cells and CCR5, CD49a, CD45RA, CD45RO, and CD161 for CD8+ T cells. The proportion of these subtypes varied in different human organs and exhibited different cytokine expression, suggesting heterogeneity of TRM in humans.

Mass cytometry can analyze close to 100 different markers of individual cells with essentially no spectral overlap. The closest information-dense experimental approach is single-cell RNA-seq. How do the results from the two methods compare? Unfortunately, single T cell RNA-seq has only been reported for melanoma-infiltrating T cells in human (11). A comparison of the frequency of T cells positive for the 31 trafficking markers and cytokines shows no correlation between CD8+ T cells from skin analyzed by mass cytometry and 889 single CD8+ T cells from melanoma analyzed by RNA-seq. This comparison is probably inadequate as T cells from the two studies are likely different. While a more appropriate comparison has to await future studies, it is worth considering the advantages and disadvantages of each approach. Compared to mass cytometry, single cell RNA-seq provides transcriptome information at the whole genome level. However, an important advantage of mass cytometry analysis is the avoidance of false-positive results that stem from post-transcriptional and post-translational modifications that cannot be captured by transcriptome analysis.

A recent study reported human T cell receptor (TCR) repertoire across decades of life by comparing data from cord blood to PBMCs from donors of increasing age (6). The results show the highest diversity of TCR in cord blood and reduced TCR diversity in PBMCs with age. A separate study reported a difference in TCR repertoire of CD4+ and CD8+ T effector memory (TEM) cells in secondary lymphoid organs with increased clonal expansion of CD8+ but not CD4+ T cells (2). It would be interesting to investigate the relationship between diverse T cell phenotypes and function from different tissues and TCR repertoire diversity. Do the same T cell clones in different locations display heterogeneity in trafficking receptors and cytokine profiles? Are antigen-specific TCR clones specifically retained in tissues by up-regulation of trafficking receptors such as CD103? Does expression of CD103 dictate the retention and subsequent expansion of specific clones?

An improved understanding of tissue-specific T cell phenotypes, trafficking receptors and cytokine profiles is extremely valuable for the development of T cell based immunotherapies. For instance, this knowledge can aid in the development of engineered T cells that can be targeted to specific organs based on T cell trafficking receptor combinations identified by Wong et al. Most chimeric antigen receptor T cells (CAR-T) to-date are not designed with tissue targeting moieties based on T cell trafficking and significant on-target off-tumor and neurological toxicities (12) have been observed with these T cells, some of which can be attributed to the whole-body distribution of systemically delivered CAR-T cells. Based on data provided by Wong et al., although complex, expression of combinations of trafficking receptors may aid in the targeting of engineered T cells to specific tissues. For instance, in applications where engineered T cells are to be targeted to the colon, expression of CCR5 and CD103 should aid in directing these cells to the colon. Such tissue targeting could potentially minimize unwanted side effects due to systemic distribution. A similar approach can be utilized in antibody therapies with bi-, tri- or multivalent antibodies, with one arm of the antibody recognizing specific trafficking receptors, such as CLA for skin homing, for tissue-specific targeting.

Additionally, recognizing the diverse nature of T cells from different organs will be important in the selection of the source of T cells for specific applications. Cord blood T cells which are predominantly naïve (Supplementary Figure S3A and B, Wong et al.) may be an appropriate source for T cell engineering approaches in which antigen specific memory T cells are not desired. CAR-T therapies may benefit from using PBMCs as the source of T cells based on recent observations that administration of a defined ratio of CD4:CD8 CAR-T cells results in superior therapy efficacy (13). On the flip side, the use of TRM cells from specific organs in applications such as adoptive T cell immunotherapy (ACT) could assist in limiting the distribution of these cells to other organs when local administration is not an option.

It is reported here that a subset of helper CD4+ cells that produce interleukin-9 (IL-9) is mainly found in human skin. These IL-9+ Th9 cells were not readily detectable in other tissues analyzed. IL-9 is a pleiotropic cytokine with a multitude of effects on numerous immune cells (14). T cells are the main source for IL-9 and accordingly, IL-9 was originally identified as a T cell growth factor (15). More recently, Th9 cells were implicated in anti-tumor immunity against melanoma in mouse models (16,17). Purwar et al. showed that loss of IL-9 promotes melanoma development, and that higher numbers of Th9 cells are detected in normal human skin in comparison to patient melanoma metastases supporting an anti-tumor role for these T cells in patients. These findings along with the observations presented by Wong et al. suggest that using Th9 cells in adoptive T-cell therapy may provide robust anti-tumor immunity in melanoma. Moreover, since IL-9 producing T helper cells are concentrated in the skin as suggested here, and the effect of IL-9 in preventing tumor growth was found to be independent of B and T cells by Purwar and colleagues (17), delivery of IL-9 via nanoparticles or liposomes decorated with skin homing receptors such as CLA, could be a novel approach for local delivery of IL-9 in melanoma. Similarly, different cytokine combinations may boost the immune system in different tissues and help against cancer as an addition to standard immune modulatory therapies.


Acknowledgements

Funding: This work was supported in part by the National Research Foundation Singapore through the Singapore-MIT Alliance for Research and Technology’s Infectious Disease IRG research programme, Cottrell Research Fund, and the Koch Institute Support (core) Grant P30-CA14051 from the National Cancer Institute. M Kaur is supported by the KI Quinquennial Cancer Research Fellowship.


Footnote

Provenance: This is a Guest Editorial commissioned by the Section Editor Hao Feng (Experimental Surgical Research, Department of General, Visceral, Transplant, Vascular and Thoracic Surgery, Hospital of the LMU Munich, Munich, Germany).

Conflicts of Interest: The authors have no conflicts of interest to declare.

Comment on:Wong MT, Ong DE, Lim FS, et al. A High-Dimensional Atlas of Human T Cell Diversity Reveals Tissue-Specific Trafficking and Cytokine Signatures. Immunity 2016;45:442-56.


References

  1. Wong MT, Ong DE, Lim FS, et al. A High-Dimensional Atlas of Human T Cell Diversity Reveals Tissue-Specific Trafficking and Cytokine Signatures. Immunity 2016;45:442-56. [Crossref] [PubMed]
  2. Thome JJ, Bickham KL, Ohmura Y, et al. Early-life compartmentalization of human T cell differentiation and regulatory function in mucosal and lymphoid tissues. Nat Med 2016;22:72-7. [Crossref] [PubMed]
  3. Thome JJ, Yudanin N, Ohmura Y, et al. Spatial map of human T cell compartmentalization and maintenance over decades of life. Cell 2014;159:814-28. [Crossref] [PubMed]
  4. Sathaliyawala T, Kubota M, Yudanin N, et al. Distribution and compartmentalization of human circulating and tissue-resident memory T cell subsets. Immunity 2013;38:187-97. [Crossref] [PubMed]
  5. Thome JJ, Farber DL. Emerging concepts in tissue-resident T cells: lessons from humans. Trends Immunol 2015;36:428-35. [Crossref] [PubMed]
  6. Britanova OV, Shugay M, Merzlyak EM, et al. Dynamics of Individual T Cell Repertoires: From Cord Blood to Centenarians. J Immunol 2016;196:5005-13. [Crossref] [PubMed]
  7. Fish EN. The X-files in immunity: sex-based differences predispose immune responses. Nat Rev Immunol 2008;8:737-44. [Crossref] [PubMed]
  8. Wirth TC, Xue HH, Rai D, et al. Repetitive antigen stimulation induces stepwise transcriptome diversification but preserves a core signature of memory CD8(+) T cell differentiation. Immunity 2010;33:128-40. [Crossref] [PubMed]
  9. Masopust D, Schenkel JM. The integration of T cell migration, differentiation and function. Nat Rev Immunol 2013;13:309-20. [Crossref] [PubMed]
  10. Mueller SN, Mackay LK. Tissue-resident memory T cells: local specialists in immune defence. Nat Rev Immunol 2016;16:79-89. [Crossref] [PubMed]
  11. Tirosh I, Izar B, Prakadan SM, et al. Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq. Science 2016;352:189-96. [Crossref] [PubMed]
  12. Bonifant CL, Jackson HJ, Brentjens RJ, et al. Toxicity and management in CAR T-cell therapy. Mol Ther Oncolytics 2016;3:16011. [Crossref] [PubMed]
  13. Turtle CJ, Hanafi LA, Berger C, et al. Immunotherapy of non-Hodgkin's lymphoma with a defined ratio of CD8+ and CD4+ CD19-specific chimeric antigen receptor-modified T cells. Sci Transl Med 2016;8:355ra116. [Crossref] [PubMed]
  14. Kaplan MH, Hufford MM, Olson MR. The development and in vivo function of T helper 9 cells. Nat Rev Immunol 2015;15:295-307. [Crossref] [PubMed]
  15. Goswami R, Kaplan MH. A brief history of IL-9. J Immunol 2011;186:3283-8. [Crossref] [PubMed]
  16. Lu Y, Hong S, Li H, et al. Th9 cells promote antitumor immune responses in vivo. J Clin Invest 2012;122:4160-71. [Crossref] [PubMed]
  17. Purwar R, Schlapbach C, Xiao S, et al. Robust tumor immunity to melanoma mediated by interleukin-9-producing T cells. Nat Med 2012;18:1248-53. [Crossref] [PubMed]
Cite this article as: Ivica NA, Kaur M, Hu G, Chen J. Information-dense analysis for information-dense understanding. Transl Cancer Res 2016;5(Suppl 6):S1078-S1081. doi: 10.21037/tcr.2016.11.21