Skip to main content

Thank you for visiting nature.com. 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.

  • Technical Report
  • Published:

Quantifying in situ adaptive immune cell cognate interactions in humans

Abstract

Two-photon excitation microscopy (TPEM) has revolutionized the understanding of adaptive immunity. However, TPEM usually requires animal models and is not amenable to the study of human disease. The recognition of antigen by T cells requires cell contact and is associated with changes in T cell shape. We postulated that by capturing these features in fixed tissue samples, we could quantify in situ adaptive immunity. Therefore, we used a deep convolutional neural network to identify fundamental distance and cell-shape features associated with cognate help (cell-distance mapping (CDM)). In mice, CDM was comparable to TPEM in discriminating cognate T cell–dendritic cell (DC) interactions from non-cognate T cell–DC interactions. In human lupus nephritis, CDM confirmed that myeloid DCs present antigen to CD4+ T cells and identified plasmacytoid DCs as an important antigen-presenting cell. These data reveal a new approach with which to study human in situ adaptive immunity broadly applicable to autoimmunity, infection, and cancer.

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

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: TPEM to measure antigen-specific CD4+ T cells interacting with dendritic cells.
Fig. 2: Development of CDM3.
Fig. 3: The sensitivity and specificity of TPEM and CDM3 are comparable.
Fig. 4: Segmentation and shape of T cell nuclei.
Fig. 5: Identification of pDCs as an antigen-presenting cell in lupus nephritis.
Fig. 6: Confirmation of pDCs as antigen-presenting cell in lupus nephritis.

Similar content being viewed by others

Data availability

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

References

  1. Miller, M. J., Wei, S. H., Parker, I. & Cahalan, M. D. Two-photon imaging of lymphocyte motility and antigen response in intact lymph node. Science 296, 1869–1873 (2002).

    Article  CAS  Google Scholar 

  2. Miller, M. J., Safrina, O., Parker, I. & Cahalan, M. D. Imaging the single cell dynamics of CD4+T cell activation by dendritic cells in lymph nodes. J. Exp. Med. 200, 847–856 (2004).

    Article  CAS  Google Scholar 

  3. Mempel, T. R., Henrickson, S. E. & Von Andrian, U. H. T-cell priming by dendritic cells in lymph nodes occurs in three distinct phases. Nature 427, 154–159 (2004).

    Article  CAS  Google Scholar 

  4. Stoll, S., Delon, J., Brotz, T. M. & Germain, R. N. Dynamic imaging of T cell-dendritic cell interactions in lymph nodes. Science 296, 1873–1876 (2002).

    Article  Google Scholar 

  5. Germain, R. N., Robey, E. A. & Cahalan, M. D. A decade of imaging cellular motility and interaction dynamics in the immune system. Science 336, 1676–1681 (2012).

    Article  CAS  Google Scholar 

  6. Masedunskas, A. et al. Intravital microscopy: a practical guide on imaging intracellular structures in live animals. Bioarchitecture 2, 143–157 (2012).

    Article  Google Scholar 

  7. Secklehner, J., Lo Celso, C. & Carlin, L. M. Intravital microscopy in historic and contemporary immunology. Immunol. Cell Biol. 95, 506–513 (2017).

    Article  Google Scholar 

  8. You, S. et al. Intravital imaging by simultaneous label-free autofluorescence-multiharmonic microscopy. Nat. Commun. 9, 2125 (2018).

    Article  Google Scholar 

  9. Kobat, D., Horton, N. G. & Xu, C. In vivo two-photon microscopy to 1.6-mm depth in mouse cortex. J. Biomed. Opt. 16, 106014 (2011).

    Article  Google Scholar 

  10. Yew, E., Rowlands, C. & So, P. T. Application of multiphoton microscopy in dermatological dtudies: a minireview. J. Innov. Opt. Health Sci. 7, 1330010 (2014).

    Article  Google Scholar 

  11. Fisher, D. T. et al. Intraoperative intravital microscopy permits the study of human tumour vessels. Nat. Commun. 7, 10684 (2016).

    Article  CAS  Google Scholar 

  12. Gerner, M. Y., Kastenmuller, W., Ifrim, I., Kabat, J. & Germain, R. N. Histo-cytometry: a method for highly multiplex quantitative tissue imaging analysis applied to dendritic cell subset microanatomy in lymph nodes. Immunity 37, 364–376 (2012).

    Article  CAS  Google Scholar 

  13. Goltsev, Y. et al. Deep profiling of mouse splenic architecture with CODEX multiplexed imaging. Cell 174, 968–981.e915 (2018).

    Article  CAS  Google Scholar 

  14. Keren, L. et al. A structured tumor-immune microenvironment in triple negative breast cancer revealed by multiplexed ion beam imaging. Cell 174, 1373–1387.e1319 (2018).

    Article  CAS  Google Scholar 

  15. Giesen, C. et al. Highly multiplexed imaging of tumor tissues with subcellular resolution by mass cytometry. Nat. Methods 11, 417–422 (2014).

    Article  CAS  Google Scholar 

  16. Azizi, E. et al. Single-cell map of diverse immune phenotypes in the breast tumor microenvironment. Cell 174, 1293–1308.e1236 (2018).

    Article  CAS  Google Scholar 

  17. Arazi, A. et al. The immune cell landscape in kidneys of lupus nephritis patients. bioRxiv https://doi.org/10.1101/363051 (2018).

  18. Chevrier, S. et al. An immune atlas of clear cell renal cell carcinoma. Cell 169, 736–749.e718 (2017).

    Article  CAS  Google Scholar 

  19. Liarski, V. et al. Quantitative cell distance mapping in human nephritis reveals organization of in situ adaptive immune responses. Sci.Trans. Med. 6, 230ra46 (2014).

    Article  Google Scholar 

  20. Zhang, Q. et al. CD8+ effector T cell migration to pancreatic islet grafts is dependent on cognate antigen presentation by donor graft cells. J. Immunol. 197, 1471–1476 (2016).

    Article  CAS  Google Scholar 

  21. Simoni, Y. et al. Bystander CD8+ T cells are abundant and phenotypically distinct in human tumour infiltrates. Nature 557, 575–579 (2018).

    Article  CAS  Google Scholar 

  22. Martin-Cofreces, N. B., Baixauli, F. & Sanchez-Madrid, F. Immune synapse: conductor of orchestrated organelle movement. Trends. Cell Biol. 24, 61–72 (2014).

    Article  CAS  Google Scholar 

  23. Dustin, M. L. & Groves, J. T. Receptor signaling clusters in the immune synapse. Annu Rev Biophys 41, 543–556 (2012).

    Article  CAS  Google Scholar 

  24. Lesserre, R. & Alcover, A. Microtubule dynamics and signal transduction at the immunological synapse: new partners and new connections. EMBO J. 31, 4100–4102 (2012).

    Article  Google Scholar 

  25. Monks, C. R., Freiberg, B. A., Kupfer, H., Sciaky, N. & Kupfer, A. Three-dimensional segregation of supramolecular activation clusters in T cells. Nature 395, 82–86 (1998).

    Article  CAS  Google Scholar 

  26. Dustin, M. L. et al. A novel adaptor protein orchestrates receptor patterning and cytoskeletal polarity in T-cell contacts. Cell 94, 667–677 (1998).

    Article  CAS  Google Scholar 

  27. Tourret, M. et al. T Cell polarity at the immunological synapse Is required for CD154-dependent IL-12 secretion by dendritic cells. J. Immunol. 185, 6809–6818 (2010).

    Article  CAS  Google Scholar 

  28. Blanchard, N. et al. Strong and durable TCR clustering at the T/dendritic cell immune synapse is not required for NFAT activation and IFN-γ production in human CD4 T Cells. J. Immunol. 173, 3062–3072 (2004).

    Article  CAS  Google Scholar 

  29. Dustin, M. L. The cellular context of T cell signaling. Immunity 30, 482–492 (2009).

    Article  CAS  Google Scholar 

  30. van Panhuys, N., Klauschen, F. & Germain, R. N. T-cell-receptor-dependent signal intensity dominantly controls CD4+ T cell polarization in vivo. Immunity 41, 63–74 (2014).

    Article  Google Scholar 

  31. Malherbe, L., Mark, L., Fazilleau, N., McHeyzer-Williams, L. & McHeyzer-Williams, M. G. Vaccine adjuvants alter TCR-based selection thresholds. Immunity 28, 698–709 (2009).

    Article  Google Scholar 

  32. Baumbartner, C. K., Ferrante, A., Nagaoka, M., Gorski, J. & Malherbe, L. P. Peptide-MHC class II complex stability governs CD4 T cell clonal selection. J. Immunol. 184, 573–581 (2010).

    Article  Google Scholar 

  33. Olson, E. Particle shape factors and their use in image analysis-part 1: theory. J. GXP Compl. 15, 85–90 (2011).

    Google Scholar 

  34. R Core Team. R: A language and environment for statistical computing. v3.4.1 (R Foundation for Statistical Computing, Vienna, Austria; 2017).

  35. Polliack, A. et al. Identification of human B and T lymphocytes by scanning electron microscopy. J. Exp. Med. 138, 607–624 (1973).

    Article  CAS  Google Scholar 

  36. Sallusto, F. & Lanzavecchia, A. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor alpha. J. Exp. Med. 179, 1109–1118 (1994).

    Article  CAS  Google Scholar 

  37. Guermonprez, P., Valladeau, J., Zitvogel, L., Thery, C. & Amigorena, S. Antigen presentation and T cell stimulation by dendritic cells. Annu. Rev. Immunol. 20, 621–667 (2002).

    Article  CAS  Google Scholar 

  38. Swiecki, M. & Colonna, M. The multifaceted biology of plasmacytoid dendritic cells. Nat. Rev. Immunol. 15, 471–485 (2015).

    Article  CAS  Google Scholar 

  39. Villadangos, J. A. & Young, L. Antigen-presentation properties of plasmacytoid dendritic cells. Immunity 29, 352–361 (2008).

    Article  CAS  Google Scholar 

  40. Villani, A. C. et al. Single-cell RNA-seq reveals new types of human blood dendritic cells, monocytes, and progenitors. Science 356, 273–284 (2017).

    Article  Google Scholar 

  41. Chang, A. et al. In situ B cell-mediated immune responses and tubulointerstitial inflammation in human lupus nephritis. J. Immunol. 186, 1849–1860 (2011).

    Article  CAS  Google Scholar 

  42. Delon, J., Kaibuchi, K. & Germain, R. N. Exclusion of CD43 from the immunological synapse is mediated by phosphorylation-regulated relocation of the cytoskeletal adaptor moesin. Immunity 15, 691–701 (2001).

    Article  CAS  Google Scholar 

  43. Allenspach, E. J. et al. ERM-dependent movement of CD43 defines a novel protein complex distal to the immunological synapse. Immunity 15, 739–750 (2001).

    Article  CAS  Google Scholar 

  44. Hutton, L. Using statistics to assess the performance of neural network classifiers. Johns Hopkins APL Tech. Dig. 13, 291–299 (1992).

    Google Scholar 

  45. Razi, M. & Athappilly, K. A comparative predictive analysis of neural networks (NNs), nonlinear regression and classification and regression tree (CART) models. Expert Syst. Appl. 29, 65–74 (2005).

    Article  Google Scholar 

  46. MATLAB 9.1 R2016b and Signal Processing Toolbox (The MathWorks Inc., 2016).

  47. Midway2. https://rcc.uchicago.edu/support-and-services/midway2 (University of Chicago Research Computing Center Cluster).

  48. Abadi, M. et al. TensorFlow: Large-scale machine learning on heterogeneous systems. arXiv https://arxiv.org/abs/1603.04467 (2015).

  49. Glorot, X & Yoshua B. Understanding the difficulty of training deep feedforward neural networks. In Proc. 13th International Conference on Artificial Intelligence and Statistics. 249–256 (PMLR, 2010).

  50. van der Walt, S. et al. scikit-image: image processing in Python. PeerJ 2, e453 (2014).

    Article  Google Scholar 

Download references

Acknowledgements

The research reported in this publication was supported by the National Institute of Allergy and Infectious Diseases (NIH) under Award Number U19 AI082724 (M.R.C.) and the National Institute of Arthritis and Musculoskeletal and Skin Diseases under Award Number K08 AR068421 (V.M.L.). Funding was also provided by R01 AR055646 (M.R.C.), T32 EB002103 (A.S.), and U01 CA195564 (M.L.G.). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Computational support was provided by the SIRAF cluster, with special thanks to C.-W. Chan (Cancer Center Support Grant P30 CA014599). GPU computation was supported by the University of Chicago Research Computing Center. Computational support on the Beagle supercomputer was provided by the NIH through resources provided by the Computation Institute and the Biological Sciences Division of the University of Chicago and Argonne National Laboratory, under grant 1S10OD018495-01, with thanks to L. Pesce and J. Urbanski. All imaging was performed at the University of Chicago Integrated Light Microscopy Facility, with thanks to V. Bindokas and C. Labno. We thank R. Abraham for careful reading of this manuscript.

Author information

Authors and Affiliations

Authors

Contributions

V.M.L. prepared the manuscript and was involved in CDM3 development (automated and manual approaches) and development of the tuned neural network analytic pipeline, assisted with convolutional network development, supervised all manual classification of cell types, and performed all data analyses; A.S. was responsible for the development, training, and testing of the image segmentation and the Tensorflow-based deep convolutional neural network, and conducted data analysis; N.v.P. performed all two-photon animal experiments and related data analyses; J.A. performed all immunofluorescence staining of samples and 2D and 3D image acquisition, along with use of Imaris analysis; A.C. performed nephropathological scoring and categorization of all biopsy specimens and provided deidentified human tissue samples; D.K. performed the repeat mouse transfer experiments at the University of Chicago; M.M. performed manual classification of cell types; R.N.G. oversaw all experiments and data analyses relating to two-photon animal experiments; M.L.G. provided expert advice on development of CDM3, image segmentation, use of DCNN-based approaches in classification cell algorithms, and method of evaluation; M.R.C. conceived of the project, oversaw its progress, prepared of the final manuscript, and developed CDM3 as a tool to predict cognate interactions.

Corresponding authors

Correspondence to Maryellen L. Giger or Marcus R. Clark.

Ethics declarations

Competing interests

R.N.G. is a full-time employee of the National Institutes of Health.

Additional information

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

Integrated supplementary information

Supplementary Figure 1 TPEM measurements of interactions of antigen-specific and wild-type CD4+ T cells with DCs under low antigen conditions (0.01 μm PCC).

Indicated T cells (WT or 5CC7) and antigen-pulsed DCs were transferred into B10.A2 CD45.2− mice and, after 12 hours, popliteal lymph nodes were imaged by TPEM, as in Main Fig. 1. a-d, TPEM parameters plotted as mean per mouse (n=7): mean velocity (a), and mean arrest coefficient (b). The cellular interaction time for all experiments is shown in (c), and per mouse in (d). *p<0.05, **p<0.005, 2-sided Mann-Whitney U test. All center values denote the mean and error bars denote standard error of the mean. n=2 independent experiments for all panels.

Supplementary Figure 2 Overview of DCNN and performance comparison to original CDM.

(a) Five channel input is fed through 10 convolutional and 3 pooling layers with 3x3x3 kernel size. The final convolutional layer is followed by a fully connected (fc) and softmax (sftmx) layers. Shown at bottom are number of layers, number of feature maps (features), dilation factor (DF), and field of view (FOV) x,y size, where field of view refers to the total amount of information that goes into the final pixel cell-type prediction. The sparsity of convolutions is increased with the dilation factor by skipping pixels within the convolved kernel, resulting in effective kernel size (KS) of ((KS-1)*DF+1). Predictions at individual pixels resulted from a DCNN FOV of 85 pixels x 85 pixels x 5 channels. The DCNN was trained on dense 184x184 pixel patches reducing to 100x100 pixels in the final DCNN output. The fully connected layer incorporates values from all feature maps into predictions of 4 cell types. Following the fully connected layer, the softmax layer uses the softmax function to transform the cell type prediction in the range [0,1], adding up to 1.0 for all types. Each pixel is then assigned the cell type with maximum predicted probability. This produces solid cell segmentations, on which shape-based object analysis can then be performed. (b) Representative image of DCNN training as reflected by decrease in cross entropy error with increasing number of steps.

Supplementary Figure 3 Representative plots of neural networks used for analysis.

Representative plots of the simple (a), linear output (b), and final tuned (c) neural networks with input nodes, hidden layer(s), weights, and output as indicated.

Supplementary Figure 4 Scatter plot trendlines.

Trendlines for all manuscript scatter plots are included for data reference and ease of visualization. Shape parameters as indicated for (a-b) mouse cell tracker, (c-e) mouse nuclei, (f-g) human plasmacytoid dendritic cells, and (h-I) human monocytic dendritic cells. Grey lines denote 5CC7 (a-e) or CD3+CD4+ (f-i) T cells; black lines denote WT (a-e) or CD3+CD4− (f-i) T cells. Shaded regions denote 95% CI. n=2 independent experiments for all panels.

Supplementary Figure 5 Tiled image of whole biopsy section.

Representative tiled images of mDCs and pDC (green) as indicated with CD4+T cells (red) in lupus TII. Nuclei are blue. HPFs corresponding to Fig. 5e indicated by yellow boxes. n=4 (a) and 6 (b) independent tiling experiments.

Supplementary Figure 6 Distribution of T cells and DCs in lupus nephritis.

Percentage of T cells versus minimum distance in μm between CD3+CD4+ T cells and pDCs (salmon), CD3+CD4− T cells and pDCs (teal), CD3+CD4+ T cells and mDCs (blue), and CD3+CD4− T cells and mDCs (purple).

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–6, Supplementary Tables 1–6

Reporting Summary

Supplementary Movie 1

Two-photon emission microscopy Representative video of two-photon emission microscopy, performed on an adoptive mouse transfer model, reveals immune cell interactions. A single popliteal lymph node is imaged with 5CC7 T cells in red, WT T cells in green, and antigen-pulsed DCs in cyan. Length of video: 62 minutes. Scale bar as indicated. n = 3 independent transfer experiments.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liarski, V.M., Sibley, A., van Panhuys, N. et al. Quantifying in situ adaptive immune cell cognate interactions in humans. Nat Immunol 20, 503–513 (2019). https://doi.org/10.1038/s41590-019-0315-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41590-019-0315-3

This article is cited by

Search

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