Effector T-cell trafficking between the leptomeninges and the cerebrospinal fluid

Abstract

In multiple sclerosis, brain-reactive T cells invade the central nervous system (CNS) and induce a self-destructive inflammatory process. T-cell infiltrates are not only found within the parenchyma and the meninges, but also in the cerebrospinal fluid (CSF) that bathes the entire CNS tissue1,2. How the T cells reach the CSF, their functionality, and whether they traffic between the CSF and other CNS compartments remains hypothetical3,4,5,6. Here we show that effector T cells enter the CSF from the leptomeninges during Lewis rat experimental autoimmune encephalomyelitis (EAE), a model of multiple sclerosis. While moving through the three-dimensional leptomeningeal network of collagen fibres in a random Brownian walk, T cells were flushed from the surface by the flow of the CSF. The detached cells displayed significantly lower activation levels compared to T cells from the leptomeninges and CNS parenchyma. However, they did not represent a specialized non-pathogenic cellular sub-fraction, as their gene expression profile strongly resembled that of tissue-derived T cells and they fully retained their encephalitogenic potential. T-cell detachment from the leptomeninges was counteracted by integrins VLA-4 and LFA-1 binding to their respective ligands produced by resident macrophages. Chemokine signalling via CCR5/CXCR3 and antigenic stimulation of T cells in contact with the leptomeningeal macrophages enforced their adhesiveness. T cells floating in the CSF were able to reattach to the leptomeninges through steps reminiscent of vascular adhesion in CNS blood vessels, and invade the parenchyma. The molecular/cellular conditions for T-cell reattachment were the same as the requirements for detachment from the leptomeningeal milieu. Our data indicate that the leptomeninges represent a checkpoint at which activated T cells are licensed to enter the CNS parenchyma and non-activated T cells are preferentially released into the CSF, from where they can reach areas of antigen availability and tissue damage.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Figure 1: TMBP cells enter the CSF from the leptomeninges.
Figure 2: Characterization of TMBP cells in the CSF.
Figure 3: Integrins and chemokines mediate TMBP cell adhesion to leptomeningeal structures.
Figure 4: Reattachment of effector T cells from the CSF to the leptomeninges and entry into the CNS parenchyma.

Accession codes

Primary accessions

Gene Expression Omnibus

Data deposits

The data discussed in this publication have been deposited in the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus and are accessible through GEO Series accession number GSE75488.

References

  1. 1

    Alvermann, S., Hennig, C., Stuve, O., Wiendl, H. & Stangel, M. Immunophenotyping of cerebrospinal fluid cells in multiple sclerosis: in search of biomarkers. JAMA Neurol . 71, 905–912 (2014)

    PubMed  Google Scholar 

  2. 2

    Giunti, D. et al. Phenotypic and functional analysis of T cells homing into the CSF of subjects with inflammatory diseases of the CNS. J. Leukoc. Biol. 73, 584–590 (2003)

    CAS  PubMed  Google Scholar 

  3. 3

    Kivisäkk, P. et al. Human cerebrospinal fluid central memory CD4+T cells: evidence for trafficking through choroid plexus and meninges via P-selectin. Proc. Natl Acad. Sci. USA 100, 8389–8394 (2003)

    ADS  PubMed  Google Scholar 

  4. 4

    Reboldi, A. et al. C-C chemokine receptor 6–regulated entry of TH-17 cells into the CNS through the choroid plexus is required for the initiation of EAE. Nature Immunol. 10, 514–523 (2009)

    CAS  Google Scholar 

  5. 5

    Engelhardt, B. & Ransohoff, R. M. The ins and outs of T-lymphocyte trafficking to the CNS: anatomical sites and molecular mechanisms. Trends Immunol. 26, 485–495 (2005)

    CAS  PubMed  Google Scholar 

  6. 6

    Engelhardt, B., Wolburg-Buchholz, K. & Wolburg, H. Involvement of the choroid plexus in central nervous system inflammation. Microsc. Res. Tech. 52, 112–129 (2001)

    CAS  PubMed  Google Scholar 

  7. 7

    Bartholomäus, I. et al. Effector T cell interactions with meningeal vascular structures in nascent autoimmune CNS lesions. Nature 462, 94–98 (2009)

    ADS  PubMed  Google Scholar 

  8. 8

    Shechter, R. et al. Recruitment of beneficial M2 macrophages to injured spinal cord is orchestrated by remote brain choroid plexus. Immunity 38, 555–569 (2013)

    CAS  PubMed  PubMed Central  Google Scholar 

  9. 9

    Baruch, K. & Schwartz, M. CNS-specific T cells shape brain function via the choroid plexus. Brain Behav. Immun. 34, 11–16 (2013)

    CAS  PubMed  Google Scholar 

  10. 10

    Lodygin, D. et al. A combination of fluorescent NFAT and H2B sensors uncovers dynamics of T cell activation in real time during CNS autoimmunity. Nature Med. 19, 784–790 (2013)

    CAS  PubMed  Google Scholar 

  11. 11

    Kawakami, N. et al. The activation status of neuroantigen-specific T cells in the target organ determines the clinical outcome of autoimmune encephalomyelitis. J. Exp. Med. 199, 185–197 (2004)

    CAS  PubMed  PubMed Central  Google Scholar 

  12. 12

    Castellino, F. et al. Chemokines enhance immunity by guiding naive CD8 + T cells to sites of CD4+ T cell–dendritic cell interaction. Nature 440, 890–895 (2006)

    ADS  CAS  PubMed  Google Scholar 

  13. 13

    Harris, T. H. et al. Generalized Lévy walks and the role of chemokines in migration of effector CD8 + T cells. Nature 486, 545–548 (2012)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  14. 14

    Humphries, N. E. et al. Environmental context explains Lévy and Brownian movement patterns of marine predators. Nature 465, 1066–1069 (2010)

    ADS  CAS  PubMed  Google Scholar 

  15. 15

    Mues, M. et al. Real-time in vivo analysis of T cell activation in the central nervous system using a genetically encoded calcium indicator. Nature Med. 19, 778–783 (2013)

    CAS  PubMed  Google Scholar 

  16. 16

    Kivisäkk, P. et al. Localizing central nervous system immune surveillance: meningeal antigen-presenting cells activate T cells during experimental autoimmune encephalomyelitis. Ann. Neurol. 65, 457–469 (2009)

    PubMed  PubMed Central  Google Scholar 

  17. 17

    Laudanna, C., Kim, J. Y., Constantin, G. & Butcher, E. Rapid leukocyte integrin activation by chemokines. Immunol. Rev. 186, 37–46 (2002)

    CAS  PubMed  Google Scholar 

  18. 18

    Odoardi, F. et al. T cells become licensed in the lung to enter the central nervous system. Nature 488, 675–679 (2012)

    ADS  CAS  PubMed  Google Scholar 

  19. 19

    Lämmermann, T. et al. Rapid leukocyte migration by integrin-independent flowing and squeezing. Nature 453, 51–55 (2008)

    ADS  PubMed  Google Scholar 

  20. 20

    Louveau, A. et al. Structural and functional features of central nervous system lymphatic vessels. Nature 523, 337–341 (2015)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  21. 21

    Aspelund, A. et al. A dural lymphatic vascular system that drains brain interstitial fluid and macromolecules. J. Exp. Med. 212, 991–999 (2015)

    CAS  PubMed  PubMed Central  Google Scholar 

  22. 22

    Iliff, J. J. et al. A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid β. Sci. Transl. Med. 4, 147ra111 (2012)

    PubMed  PubMed Central  Google Scholar 

  23. 23

    Hatterer, E., Touret, M., Belin, M. F., Honnorat, J. & Nataf, S. Cerebrospinal fluid dendritic cells infiltrate the brain parenchyma and target the cervical lymph nodes under neuroinflammatory conditions. PLoS ONE 3, e3321 (2008)

    ADS  PubMed  PubMed Central  Google Scholar 

  24. 24

    Dreha-Kulaczewski, S. et al. Inspiration is the major regulator of human CSF flow. J. Neurosci. 35, 2485–2491 (2015)

    CAS  PubMed  PubMed Central  Google Scholar 

  25. 25

    Ley, K., Laudanna, C., Cybulsky, M. I. & Nourshargh, S. Getting to the site of inflammation: the leukocyte adhesion cascade updated. Nature Rev. Immunol . 7, 678–689 (2007)

    CAS  Google Scholar 

  26. 26

    Odoardi, F., Kawakami, N., Klinkert, W. E., Wekerle, H. & Flügel, A. Blood-borne soluble protein antigen intensifies T cell activation in autoimmune CNS lesions and exacerbates clinical disease. Proc. Natl Acad. Sci. USA 104, 18625–18630 (2007)

    ADS  CAS  PubMed  Google Scholar 

  27. 27

    Kawakami, N. et al. Autoimmune CD4+T cell memory: lifelong persistence of encephalitogenic T cell clones in healthy immune repertoires. J. Immunol. 175, 69–81 (2005)

    CAS  PubMed  Google Scholar 

  28. 28

    Lucchinetti, C. F. et al. Inflammatory cortical demyelination in early multiple sclerosis. N. Engl. J. Med. 365, 2188–2197 (2011)

    CAS  PubMed  PubMed Central  Google Scholar 

  29. 29

    Dustin, M. L. & Springer, T. A. T-cell receptor cross-linking transiently stimulates adhesiveness through LFA-1. Nature 341, 619–624 (1989)

    ADS  CAS  PubMed  Google Scholar 

  30. 30

    Flügel, A., Willem, M., Berkowicz, T. & Wekerle, H. Gene transfer into CD4+ T lymphocytes: green fluorescent protein-engineered, encephalitogenic T cells illuminate brain autoimmune responses. Nature Med. 5, 843–847 (1999)

    PubMed  Google Scholar 

  31. 31

    Gurskaya, N. G. et al. Engineering of a monomeric green-to-red photoactivatable fluorescent protein induced by blue light. Nature Biotechnol. 24, 461–465 (2006)

    CAS  Google Scholar 

  32. 32

    Eylar, E. H., Kniskern, P. J. & Jackson, J. J. Myelin basic proteins. Methods Enzymol. 32, 323–341 (1974)

    CAS  PubMed  Google Scholar 

  33. 33

    Adelmann, M. et al. The N-terminal domain of the myelin oligodendrocyte glycoprotein (MOG) induces acute demyelinating experimental autoimmune encephalomyelitis in the Lewis rat. J. Neuroimmunol. 63, 17–27 (1995)

    CAS  PubMed  Google Scholar 

  34. 34

    Flügel, A. et al. Migratory activity and functional changes of green fluorescent effector cells before and during experimental autoimmune encephalomyelitis. Immunity 14, 547–560 (2001)

    PubMed  Google Scholar 

  35. 35

    Luque, A. et al. Activated conformations of very late activation integrins detected by a group of antibodies (HUTS) specific for a novel regulatory region (355–425) of the common β1 chain. J. Biol. Chem. 271, 11067–11075 (1996)

    CAS  PubMed  Google Scholar 

  36. 36

    Siffrin, V. et al. Differential immune cell dynamics in the CNS cause CD4+ T cell compartmentalization. Brain 132, 1247–1258 (2009)

    PubMed  Google Scholar 

  37. 37

    Kawakami, N. et al. Live imaging of effector cell trafficking and autoantigen recognition within the unfolding autoimmune encephalomyelitis lesion. J. Exp. Med. 201, 1805–1814 (2005)

    CAS  PubMed  PubMed Central  Google Scholar 

  38. 38

    Issekutz, T. B. Inhibition of in vivo lymphocyte migration to inflammation and homing to lymphoid tissues by the TA-2 monoclonal antibody. A likely role for VLA-4 in vivo. J. Immunol. 147, 4178–4184 (1991)

    CAS  PubMed  Google Scholar 

  39. 39

    Sporici, R. & Issekutz, T. B. CXCR3 blockade inhibits T-cell migration into the CNS during EAE and prevents development of adoptively transferred, but not actively induced, disease. Eur. J. Immunol. 40, 2751–2761 (2010)

    CAS  PubMed  Google Scholar 

  40. 40

    Matsui, M. et al. Treatment of experimental autoimmune encephalomyelitis with the chemokine receptor antagonist Met-RANTES. J. Neuroimmunol. 128, 16–22 (2002)

    CAS  PubMed  Google Scholar 

  41. 41

    Meiron, M., Zohar, Y., Anunu, R., Wildbaum, G. & Karin, N. CXCL12 (SDF-1alpha) suppresses ongoing experimental autoimmune encephalomyelitis by selecting antigen-specific regulatory T cells. J. Exp. Med. 205, 2643–2655 (2008)

    CAS  PubMed  PubMed Central  Google Scholar 

  42. 42

    Carnevali, L. et al. Vagal withdrawal and susceptibility to cardiac arrhythmias in rats with high trait aggressiveness. PLoS ONE 8, e68316 (2013)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  43. 43

    Slobodian, I., Krassioukov-Enns, D. & Del Bigio, M. R. Protein and synthetic polymer injection for induction of obstructive hydrocephalus in rats. Cerebrospinal Fluid Res. 4, 9 (2007)

    PubMed  PubMed Central  Google Scholar 

  44. 44

    De la Calle, J. L. & Paino, C. L. A procedure for direct lumbar puncture in rats. Brain Res. Bull. 59, 245–250 (2002)

    PubMed  Google Scholar 

  45. 45

    Cabeza, R. et al. An RNA sequencing transcriptome analysis reveals novel insights into molecular aspects of the nitrate impact on the nodule activity of Medicago truncatula. Plant Physiol. 164, 400–411 (2014)

    CAS  PubMed  PubMed Central  Google Scholar 

  46. 46

    Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013)

    CAS  Article  Google Scholar 

  47. 47

    Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079 (2009)

    PubMed  PubMed Central  Google Scholar 

  48. 48

    Anders, S., Pyl, P. T. & Huber, W. HTSeq–a Python framework to work with high-throughput sequencing data. Bioinformatics 31, 166–169 (2015)

    CAS  Google Scholar 

  49. 49

    Anders, S. & Huber, W. Differential expression analysis for sequence count data. Genome Biol. 11, R106 (2010)

    CAS  PubMed  PubMed Central  Google Scholar 

  50. 50

    Durinck, S. et al. BioMart and Bioconductor: a powerful link between biological databases and microarray data analysis. Bioinformatics 21, 3439–3440 (2005)

    CAS  Google Scholar 

  51. 51

    Huang, W., Sherman, B. T. & Lempicki, R. A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nature Protocols 4, 44–57 (2009)

    CAS  Google Scholar 

  52. 52

    Odoardi, F. et al. Instant effect of soluble antigen on effector T cells in peripheral immune organs during immunotherapy of autoimmune encephalomyelitis. Proc. Natl Acad. Sci. USA 104, 920–925 (2007)

    ADS  CAS  PubMed  Google Scholar 

  53. 53

    Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nature Methods 9, 676–682 (2012)

    CAS  PubMed  Google Scholar 

Download references

Acknowledgements

The authors thank S. Hamann, A. Stas, N. Meyer, S. Mole, and M. Weig for excellent technical assistance. We thank G. Salinas-Riester for her support in performing the transcriptome analyses, T. Lingner for his help in analysing the transcriptome data and W. Lühder for contributing to the mathematical T-cell locomotion analyses. We are grateful to C. Ludwig for text editing. This work was supported by the Deutsche Forschungsgemeinschaft (TRR-SFB43 project B10, FORR 1336 project B1 and RK-Grant FL 377/3-1), the Bundesministerium für Bildung und Forschung (‘UNDERSTAND MS’), the Hertie Foundation (grants 1.01.1/11/004 and 1130072), the Ministry of Science and Culture of Lower Saxony (Niedersachsen-Research Network on Neuroinfectiology, N-RENNT) and the European Commission ERA-NET NEURON (MELTRA-BBB).

Author information

Affiliations

Authors

Contributions

C.S. performed most intravital TPLSM imaging studies. H.K. performed most of the CSF analyses, fluorescence microscopy and T-cell-CSF-transfers. M.K., E.B. and I.B. performed and analysed immunofluorescence and electron microscopic analyses. S.V. performed the mathematical cell motility analyses. M.H. performed antibody labelling and contributed to TPLSM imaging. D.M. and V.R. designed the operative strategy, D.M. performed the plexus preparation. C.C. provided HUTS4 antibody and contributed with technical advice. T.I. provided the anti-VLA-4 and anti-CXCR3 antibodies and contributed with technical advice. P.J.N. provided CCR5 blocker and contributed with scientific advice. T.Z. contributed with T-cell characterization in the CSF, D.L. designed and produced genetic retroviral sensors and contributed to the analysis of NSeq data. F.O. performed most ex vivo T-cell analyses, i.e. cytofluorometric characterizations, quantitative PCR and NSeq analyses. A.F. together with F.O. designed the study, coordinated the experimental work and wrote the manuscript with inputs from co-authors.

Corresponding author

Correspondence to Alexander Flügel.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Extended data figures and tables

Extended Data Figure 1 TMBP cells enter the CSF from the leptomeninges.

a, TMBP cell entry into the CNS compartments. TMBP–GFP cells were quantified by histological analysis in the choroid plexus of all four ventricles (1st–4th vent) and spinal cord (meninges, white matter (WM) and grey matter (GM)) and by flow cytometry in the CSF. Histology of 6–27 consecutive slices per compartment. n.d., not detectable. Representative experiment ± s.e.m. from 2 independent experiments (n = 12) from days 1.5–5 after transfer. b, Accumulation of TMBP–GFP cells in the CSF and leptomeninges occurs simultaneously. Numbers of TMBP–GFP cells (flow cytometry) in meninges and CSF. Results of 3 independent experiments. Data are mean ± s.d. of 2–4 animals per time point (n = 61). c, TMBP cells accumulate in the leptomeninges but not in the choroid plexus. Confocal laser scanning microscopy of fixed tissue sections of the lateral, third and fourth ventricles and fluorescence microscopy images of lumbar spinal cord sections. Images were recorded from the same animal for each time point. Br. st., brain stem; green, TMBP–GFP cells; blue, DAPI counterstain of nuclei. Right (I–XII), magnifications of areas of interest. Representative images of 14 different time points from days 1.5–5 after T-cell transfer. d, Localization of TMBP cells in choroid plexus explants compared to spinal cord leptomeninges. Original TPLSM images and 3D reconstructions of explanted choroid plexus of the fourth ventricle. The corresponding lumbar spinal cord leptomeninges (lumbar SC) was acquired before explantation of the choroid plexus. Green, TMBP–GFP cells; red, meningeal blood vessels; blue, collagen. Few TMBP–GFP cells can be detected within the choroid plexus tissues compared to the leptomeninges of the spinal cord. Red arrows, TMBP–GFP cells within the choroid plexus stroma; white arrows, TMBP–GFP cells outside of the choroid plexus tissue. Representative pictures of 2 independent experiments (n = 10). Scale bars, 100 μm (c, d). e, TMBP cells accumulate in the CSF before appearing in the choroid plexus. Quantification of TMBP–GFP cells from the choroid plexus of the fourth and lateral ventricles and from the CSF (flow cytometry). Means ± s.d. Representative data of 3 independent experiments including 2–4 animals per time point. f, Anti-GFP immuno-electron microscopy confirms the scarcity of TMBP cells in the choroid plexus. Upper left, choroid plexus of the fourth ventricle. The choroid plexus epithelial cells are colour-coded in yellow. The lamina propria between epithelial basal lamina (blue) and vascular basal lamina (red) is filled with loose connective tissue and meningeal fibroblasts. Arrow heads, endothelial nuclei of cross sectioned vessels (V). Top right, ependyma. The surface of the ventricular ependyma (arrows) appears smooth and continuous with numerous ciliary processes. No TMBP cells could be detected for any of the observed time points after T-cell transfer. Lower panel, leptomeninges. TMBP–GFP cells (T) are marked by typical black grains of DAB. Yellow, resident leptomeningeal cells; blue, bundles of collagen fibres; red, vascular endothelial basal laminas. MΦ, macrophages; L, vascular lumen.

Extended Data Figure 2 Interference with TMBP cell transport within the CSF does not inhibit their accumulation in the spinal cord and clinical EAE.

a, TMBP cell transport from the cisterna magna to the subarachnoidal space of the spinal cord is effectively blocked by Matrigel. Left, macroscopic image of the cisterna magna and the adjacent parts of the CNS after injection of Matrigel mixed with Evans blue dye. Cb, cerebellum; Ce, cerebrum; SC, spinal cord. Arrow, cisterna magna filled with blue-stained Matrigel. Right, PBS or Matrigel was injected in the cisterna magna of naive animals. TMBP–GFP cells were injected 24 h later intra-cisternally. TPLSM of the medulla oblongata and the cervical spinal cord was performed 6 h after the TMBP–GFP cell injection. Shown are 3D reconstructions of the TPLSM recordings. Notably, in Matrigel-injected animals the TMBP cells remained localized in the cisterna magna but did not reach the cervical spinal cord, indicating that the Matrigel efficiently blocked the migration of cells from the cisterna to the subjacent leptomeninges. In controls (PBS-injected animals), TMBP cells readily reached the leptomeninges of the cervical spinal cord. Green, TMBP–GFP cells; blue, collagen. Scale bars, 50 μm. b, Leptomeningeal T-cell infiltration and clinical disease are not impaired after Matrigel blockage. Matrigel was injected into the cisterna magna 2 days after i.v. transfer of TMBP–GFP cells. PBS i.t.-injected animals were used as control. Left, TPLSM recordings of thoracic leptomeninges 3.5 days after transfer. Representative data of 2 independent experiments. Scale bar, 200 μm. Right, clinical score assessment. Representative data of 2 independent experiments including 4 animals per treatment (n = 16). Mean ± s.e.m. c, TMBP cell transport from the cisterna magna to the subarachnoidal space of the spinal cord is effectively blocked by interrupting the CSF flow. Resting T cells (5 × 106) were injected into the cisterna magna either in control animals (upper panel) or in animals where the CSF flow was interrupted by introducing a CSF leakage at the level of the cervical spinal cord (lower panel). The efflux of the CSF induced a collapse of the subjacent thoracic subarachnoidal space (SAS) and thereby abolished the CSF flow in this area. TPLSM recordings of the thoracic spinal cord and of the medulla oblongata 6 h after i.t. injection. Depicted are 3D reconstructions of TPLSM recordings of the thoracic spinal cord leptomeninges, overviews of the same area with magnifications of areas of interest (I, II) and overviews of the medulla oblongata. Green, TMBP–GFP cells; red, blood vessels; blue, collagen. Arrows, representative TMBP–GFP cells. Scale bars, 300 μm for thoracic spinal cord and 50 μm for medulla oblongata. Note that in the animals with disrupted CSF flow, no TMBP cells were detectable at the level of the spinal cord despite the high number of TMBP cells in the medulla oblongata indicating that the leakage completely blocked the transport of cells from the cisterna magna to the subjacent spinal cord leptomeninges. d, e, Interrupting the CSF flow does not impair TMBP cell accumulation in the spinal cord leptomeninges. d, Shown are 3D reconstructions of TPLSM recordings of the thoracic spinal cord leptomeninges after i.v. transfer of TMBP–GFP cells. Images before (70 h after transfer) and after introducing a CSF leakage at the level of the cervical spinal cord (82 h after transfer). Animals with intact flow were used as control. Green, TMBP–GFP cells; red, meningeal blood vessels; blue, collagen. Scale bars, 300 μm. e, Left, TPLSM overviews of the thoracic spinal cord leptomeninges. The numbers of TMBP–GFP cells that had emigrated from the local vessels into the leptomeningeal milieu (red false colours) within the 12 h interval did not differ between controls and ‘CSF flow interrupted’ animals. Green and red marks (overlaid red dots), TMBP cells located inside or outside the vessels, respectively; blue (false colour), meningeal blood vessels. Scale bars, 200 μm. Right, relative increase of TMBP–GFP cells located in the extravascular compartment at the indicated time points during imaging either in control animals or in animals where the CSF was interrupted (leakage). Representative data of 3 independent experiments (Kruskal–Wallis ANOVA followed by Dunn’s multiple comparison test, *P < 0.05). Mean ± s.d.

Extended Data Figure 3 TMBP cells entering the leptomeningeal milieu from local vessels become detached and float within the CSF.

a, TMBP cell extravasation from the leptomeningeal vessels. Left, TPLSM during the pre-invasion phase. Representative TMBP cell diapedesis. Depicted is a very early transmigration event, that is, the majority of both TMBP cells and CD11b+myeloid cells are located inside the vascular lumina. Red, meningeal blood vessels; blue, TMBP–Lifeact–Turquoise2 cells; white, CD11b+myeloid cells labelled by i.v. administration of fluorescently conjugated anti-CD11b monoclonal antibody during imaging. Scale bars, 10 μm. Right, immuno-electron microscopy of TMBP cell extravasations observed at the indicated time points after transfer. Yellow, endothelial layer of leptomeningeal vessels; red, endothelial basal lamina. TMBP–GFP cells are marked by cytoplasmic DAB. Extravasating TMBP cells were often found to be surrounded by endothelial processes. Thereby, notable abluminal or luminal gaps (arrow heads) of the endothelial layer were frequently observed. Representative electron micrographs at the indicated time points. L, vascular lumen; T, T cells. b, Surface-rendered 3D reconstruction of the meningeal compartment. TPLSM recording in the early phase of leptomeningeal TMBP cell infiltration. Red, meningeal blood vessels; green, TMBP–GFP cells; blue, collagen. Scale bar, 300 μm. c, Visualization of a TMBP cell detaching from the leptomeninges into the CSF. Intravital video-microscopy during the established phase of leptomeningeal TMBP cell infiltration depicts a TMBP–GFP cell (false colour yellow, indicated by the white arrow) in the process of detachment from the pial surface. Yellow and blue lines, crawling and rolling steps, respectively; dotted grey arrows, direction of the CSF flow. Scale bars, 50 μm. d, In situ light exposure induces efficient and high contrast photoconversion of TMBP–Dendra2 cells without influencing T-cell motility. Images show representative TPLSM leptomeningeal overviews during established leptomeningeal TMBP cell infiltration before (0 min), 5 min, and 120 min after photoconversion. Red, meningeal blood vessels; green, non-photoconverted TMBP–Dendra2 cells; yellow, photoconverted TMBP–Dendra2 cells; blue and white circles, representative non-photoconverted or photoconverted TMBP–Dendra2 cells, respectively. Plotted are mean velocities of individual TMBP–Dendra2 cells before and 2 h after photoconversion. Data refer to 30 min time-lapse recordings. Mean values from 239 cells from at least 3 independent experiments (two-tailed Mann–Whitney U-test). e, Rapid turnover of effector T cells in the leptomeninges. Intravital TPLSM recordings on leptomeninges during the established phase of leptomeningeal TMBP cell infiltration. Images show distribution of TMBP–Dendra2 cells before or 4 h after photoconversion. For better visualization, red and green dots were overlaid onto photoconverted and non-photoconverted T cells, respectively. Grey (false colour), vessel lumen. Scale bar, 100 μm. Right, relative changes of photoconverted versus non-photoconverted TMBP–Dendra2 cells over time obtained from intravital recordings. Representative data of at least 3 independent experiments.

Extended Data Figure 4 CSF-derived TMBP cells do not represent a specialized sub-population compared to TMBP cells in the meninges and CNS parenchyma but they display a lower activation profile.

a, Similarities in transcriptome profiles between TMBP cells in blood, CSF, leptomeninges and CNS parenchyma. RNA-seq of TMBP–GFP cells sorted from blood, meninges (Men), CSF and CNS parenchyma (Par) 3 days after transfer and on in vitro cultured TMBP–GFP cells sorted 20 h and 6 days after antigenic stimulation (blast and resting T cells, respectively). Principal component analysis of the transcriptomes for all six T cell populations (left) and for the four ex vivo populations (right) show similar profiles of TMBP–GFP cells in the CSF and the other CNS compartments compared to blood and culture. Numbers in parentheses, proportion of total variability calculated for each principal component. Each data point represents a biological replicate. b, TMBP cells display similar TCR repertoires in the CNS compartments. Top, RNA-seq as in a. Normalized expression of invariant TCR complex components (left) and mean frequencies of the TCR Vβ genes (right) determined as the proportion of RNA-seq reads that map to certain Vβ segments among all reads mapping to the entire set of 24 rat Vβ segments. Shown are the 6 most abundant Vβ segments. Bottom, TMBP–GFP cells from culture (4 days after antigen stimulation), blood, or from the indicated CNS compartments on day 4 after T-cell transfer. Expression (flow cytometry) of different TCR Vβ chains. Percentages of the different Vβ chains for three different TMBP cell lines. Ex vivo cell staining containing pooled cells from the respective organs of 3–4 animals (n = 10). ce, TMBP cells within blood and CSF are not activated in contrast to TMBP cells in meninges and CNS parenchyma. c, Amounts of IFNγ and IL-17 (quantitative PCR) in TMBP cells from blood or the indicated CNS compartments at the indicated time points after transfer. Data are mean ± s.d. of duplicate measurements. Representative data of 4 independent experiments including at least 2 animals per group per time point. n.d., not determined owing to lack of cells within the CNS parenchyma at the early leptomeningeal TMBP cell infiltration phase. d, Cell surface expression of activation markers CD25 and CD134 (flow cytometry) of TMBP–GFP cells at day 4 after transfer. Black, isotype control; red, CD25 or CD134. Representative data of 4 independent experiments each combining cells from 4 or 5 animals for each compartment. e, Intracellular IFNγ and IL-17 production in TMBP cells isolated during the indicated phases of EAE. Representative data of 4 independent experiments each combining cells from 4 or 5 animals for each compartment (n = 17). f, CSF-derived TMBP cells produce cytokines upon antigenic stimulation in CSF. TMBP–GFP cells isolated at day 3 after transfer from CSF or cultured resting TMBP–GFP cells were stimulated in vitro with MBP in CSF from naive rats. IFNγ and IL-17 production (quantitative PCR). Data are mean ± s.d. of duplicate measurements representative of 2 independent experiments (n = 8). House-keeping gene, β-actin (Actb) (c, f).

Extended Data Figure 5 The trafficking of effector T cells in the CNS compartments depends on their reactivation levels and the inflammatory state in the CNS.

a, TMOG cells have a low encephalitogenic potential and display low reactivation levels in the CNS. Left, IL-17 and IFNγ expression measured by quantitative PCR in TMOG–GFP cells. House-keeping gene, β-actin (Actb) (a, c, d). Representative data of 3 independent experiments (data are mean ± s.d. of duplicate measurements). Note, TMOG cells in meninges and CNS parenchyma produce less cytokines than the highly pathogenic TMBP cells (Extended Data Fig. 4c), but more than non-pathogenic TOVA cells (Extended Data Fig. 4d). Right, clinical course after TMOG–GFP cell i.v. transfer. Cumulative data of 3 independent experiments including at least 4 animals per group. TMOG cell CNS infiltration in this model considerably precedes disease onset (Fig. 2e). b, Morphological analysis of the distribution of effector T cells with different antigen specificities and reactivation/pathogenic potentials in the CNS. Fluorescence-microscopy-derived overviews and magnified subsets (I–IV) of thoracic spinal cord sections showing distribution of fluorescently labelled T cells with the indicated antigen specificity 4 days after T-cell transfer. Open white and closed yellow arrows, representative T cells located in the meningeal compartment or in the CNS parenchyma, respectively. Scale bars: overviews, 200 μm; magnification, 100 μm. The infiltration behaviour of brain-non-reactive TOVA cells is similar to that of pathogenic TMBP cells when the CNS tissue is inflamed (after co-transfer of the TOVA cells with TMBP cells). c, TOVA cells encounter an inflammatory milieu in the meninges when co-transferred with TMBP cells. Quantitative PCR for the indicated chemokines, integrin ligands and cytokines was performed on meninges and CNS parenchyma isolated from naive animals (white columns) or from animals that 4 days previously had received either TOVA–Cherry cells (grey columns) or TOVA–Cherry cells together with TMBP–GFP cells (black columns). d, When co-transferred with TMBP cells, TOVA cells do not show signs of activation in the CNS. IFNγ and IL-17 expression (quantitative PCR) in TMBP–GFP or TOVA–Cherry cells sorted from the indicated compartments in animals transferred 4 days previously with these cells. Representative data of 3 independent experiments; data are mean ± s.d. of duplicate measurements (c, d). e, TOVA cells are released from the meningeal compartment into the CSF in higher numbers than TMBP cells. Quantification of TOVA–Cherry cells and co-injected TMBP–GFP cells during established leptomeningeal TMBP cell infiltration (flow cytometry). Representative data of 3 independent experiments including 3 animals per group. Data are mean ± s.e.m. (two-tailed Mann–Whitney U-test).

Extended Data Figure 6 TMBP cells in the leptomeningeal milieu closely interact with resident macrophages.

a, Composition of the leptomeningeal milieu and interactions of TMBP cells with resident macrophages. Upper left, representative still and time-encoded projection illustrating, via a time colour-code scale, the area that single leptomeningeal macrophages scan with their processes over a recording period of 10 min (32 s time intervals). Green, GFP+meningeal macrophages in GFP+bone marrow chimaeras7; red, blood vessels. Upper right, original TPLSM picture and surface-rendered 3D reconstructions of a representative spot within the meningeal milieu during established leptomeningeal TMBP cell infiltration. Green, extravasated motile TMBP–GFP cells; red, meningeal phagocytes (MΦ); yellow trajectories, migration paths of individual TMBP–GFP cells; blue, collagen fibres. Magnification of an individual region (white dotted rectangle) indicates contacts (yellow arrows) between TMBP–GFP cells and meningeal phagocytes. Lower left, TMBP cells in direct contact (yellow) or not in contact (green) with meningeal macrophages (red). Surface-rendered 3D reconstruction of a representative still from a 30 min TPLSM recording. Lower right, a representative TMBP–GFP effector cell (red, false colour) migrating within the leptomeningeal compartment contacting several resident meningeal macrophages (1–5, yellow contours). Orange line, migration path of the T cell; green, TMBP–GFP cells; grey, vessel lumen (false colour). Scale bars, 50 μm. b, Ultrastructural analysis confirms direct and intense contacts between T cells and meningeal macrophages. Representative overview of cells within the spinal cord leptomeninges illustrating the organization of the leptomeningeal milieu at the ultrastructural level. Lymphocytes (L) in the subarachnoidal space are located between or adjacent to collagen fibrils (arrow heads) and in direct vicinity of phagocytes (MΦ) or local resident cells (M). Right, higher magnification of a lymphocyte in close contact with a macrophage. c, TMBP cells do not follow chemotactic gradients towards meningeal phagocytes. Vector analysis of the migration steps of TMBP–GFP cells in relation to a fixed macrophage in areas with a radius (ρ) of 25, 50 or 100 μm. The cosine of the angle between the macrophage–T-cell axes (AC) and the migration vector (AB) was calculated for each step. The values 1 or −1 correspond to a TMBP–GFP cell moving towards the meningeal phagocyte or in the opposite direction, respectively. Numbers indicate average direction ± s.d. of the steps examined for each radius. The average is always close to zero and the standard deviation is very broad, excluding the presence of targeted migration. Representative data of 4 independent experiments for each radius including 4396, 15870 and 57535 steps for ρ = 25, ρ = 50 and ρ = 100 μm, respectively. d, Integrin ligands are highly expressed in meningeal phagocytes. Bar plots show quantitative PCR data for the indicated integrin ligands on meningeal phagocytes labelled with Texas Red sorted from naive animals or during the early or established phases of leptomeningeal TMBP cell infiltration. House-keeping gene, β-actin (Actb). Mean ± s.d. of duplicate measurements. Histogram plots show corresponding protein expression (flow cytometry) during the established phase of leptomeningeal TMBP cell infiltration. Bottom, representative confocal images of leptomeningeal spots acquired in naive GFP+animals or in GFP+animals with established leptomeningeal TMBP cell infiltration after in vivo staining with the indicated antibodies. Single fluorescent channels and merged pictures. The expression of the tested integrin ligands is enriched in the vicinity of the leptomeningeal blood vessels in naive tissue (left) but widely distributed during the established phase of leptomeningeal TMBP cell infiltration (right).

Extended Data Figure 7 Locomotion behaviour of TMBP cells in the leptomeningeal milieu and its regulation by VLA-4/LFA-1 integrins.

a, TMBP cells express high level of integrins. VLA-4 (Itga4) and LFA-1 (Itgal) expression (quantitative PCR) on TMBP–GFP cells from meninges or CNS parenchyma at the indicated time points of leptomeningeal T-cell infiltration. Representative data of 3 independent experiments. Data are mean ± s.d. b, VLA-4 is expressed in the active conformation on TMBP cells in the leptomeninges but not in the CSF. HUTS-4 antibody (directed against the activated conformation of β1-integrins), activation-independent anti-VLA-4 antibody or control antibody were injected i.t. into Lewis rats during the established phase of leptomeningeal T-cell infiltration. TMBP–GFP cells were isolated from meninges and CSF 4 h later; protein expression was measured by flow cytometry. Percentages of positive T cells are indicated. Representative data of 3 independent experiments. c, Integrin blockade accelerates T-cell migration. Dot plots show TMBP–GFP cell velocity before (control) and 4 h after i.t. treatment with the indicated anti-integrin monoclonal antibodies. Data from 30 min TPLSM recordings on leptomeninges during the early phase of TMBP cell leptomeningeal infiltration. Data are mean values of individual T cells from at least 3 independent experiments including 674 cells (Kruskal–Wallis ANOVA followed by Dunn’s multiple comparison). d, TMBP cell velocity is not influenced by integrins in the absence of CSF flow. Velocities of TMBP–GFP cells in the leptomeningeal milieu analysed during the established phase of leptomeningeal TMBP cell infiltration after interruption of the CSF flow in the presence or absence (control) of anti-integrin monoclonal antibodies. The interruption of the CSF flow was achieved by removal of the arachnoidea causing the efflux of the CSF from the subarachnoidal space. 30 min TPLSM recordings. Data are mean values of individual T cells from 3 independent experiments including 166 cells (two-tailed Mann–Whitney U-test). e, Integrin blockade does not change the motility pattern of TMBP cells. Mathematical analyses of at least 10 TPLSM recordings for each treatment (as in Fig. 3a, b). The Brownian walk of effector TMBP–GFP cells was not changed after i.t. application of anti-LFA-1/VLA-4 monoclonal antibodies. f, Effects of integrins on TMBP cell adhesion to the leptomeninges during the early phase of meningeal infiltration resemble those during the established phase. 30 min intravital TPLSM recordings on leptomeninges during the early phase of leptomeningeal TMBP cell infiltration. Animals were treated i.t. with PBS (control) or with the indicated anti-integrin monoclonal antibodies. Bar plots show number of TMBP–GFP cells 4 h after treatment in CSF and meninges (flow cytometry). At this time point no TMBP–GFP cells were detectable in the CNS parenchyma (Kruskal–Wallis ANOVA followed by Dunn’s multiple comparison test). Images show representative stills from 30 min TPLSM recordings and corresponding time projections of T-cell tracks before and 4 h after the indicated treatment. T-cell detachment is indicated by the reduction of TMBP cell trajectories after i.t. monoclonal antibody treatment. Blue, TMBP–Lifeact–Turquoise2 cells. Scale bar, 50 μm. Representative data of 2 independent experiments. g, Integrin blockade increases the turnover of TMBP cells in the leptomeninges. Quantification of photoconverted TMBP–Dendra2 cells after i.t. application of anti-LFA-1/VLA-4 monoclonal antibodies or control IgG (control) during TPLSM of the established phase of leptomeningeal TMBP–Dendra2 cell infiltration. Representative kinetic of at least 3 independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001 (c, f).

Extended Data Figure 8 Role of chemokines in TMBP cell motility in the leptomeningeal milieu.

a, Chemokine receptor expression profile of TMBP cells during EAE. Quantitative PCR data for the indicated chemokine receptors from TMBP–GFP cells isolated from spinal cord meninges and parenchyma during the early and established phases of leptomeningeal TMBP cell infiltration. Representative data of 3 independent experiments. Data are mean ± s.d. of replicate measurements. b, c, Inflammatory chemokines are upregulated in the CNS milieus and in meningeal macrophages during EAE. Quantitative PCR in naive animals or in animals during the early and established phases of leptomeningeal TMBP cell infiltration. Chemokine ligand expression either on meninges and CNS parenchyma (b) or on Texas-Red-labelled meningeal phagocytes isolated by flow cytometry (c). House-keeping gene, β-actin (Actb) (ac). Representative data of 3 different experiments. Data are mean ± s.d. d, TMBP cells establish direct contact with CXCL11+, MHC-II+meningeal phagocytes. Leptomeninges during early leptomeningeal TMBP cell infiltration stained for CXCL11, Iba-1 and DAPI (upper row) or CXCL11, MHC-II and DAPI (lower row). Green, TMBP–GFP cells. Scale bars, 50 μm. In the magnifications (right) yellow arrows indicate direct contacts between TMBP–GFP cells and CXCL11+meningeal phagocytes. Scale bars, 20 μm. e, Gαi signalling blockade induces a release of TMBP cells from the leptomeninges into the CSF. Left, plots show flow cytometry quantification of TMBP–GFP cells during established leptomeningeal TMBP cell infiltration in the indicated compartments 4 h after i.t. treatment with PBS (control) or PTX. Right, intravital image of spinal cord leptomeninges recorded 4 h after i.t. injection of PTX during the established phase of leptomeningeal TMBP cell infiltration. During the recording time, two TMBP–GFP cells highlighted in orange were released into the CSF. Grey (false colour), leptomeningeal blood vessels, phagocytes; green, TMBP–GFP cells; orange dotted lines, tracks of individual TMBP–GFP cells before being washed away into the CSF; blue dotted lines, contour of individual resident phagocytes; white arrows, direction of CSF flow. Scale bars, 50 μm. f, Interference with chemokine signalling affects TMBP cell contacts with leptomeningeal phagocytes but not TMBP cell velocity and meandering index. 30 min time-lapse recordings during the early (upper plots) and established phases (lower plots) of leptomeningeal TMBP cell infiltration. Bar plots show mean contact duration and contact frequencies (early and established phases: 968 and 498 motile cells, respectively) between TMBP–GFP cells and Texas-Red-labelled meningeal phagocytes before (control) and 4 h after the indicated i.t. treatments. Data are mean ± s.e.m. of 3 independent experiments. Dot plots show TMBP–GFP cell velocity and meandering index before (control) and 4 h after the indicated i.t. treatment. Data are mean values of individual T cells from at least 3 independent experiments per treatment and time point (early phase 1506 cells and established phase 860 cells; Kruskal–Wallis ANOVA followed by Dunn’s multiple comparison test). g, Gαi blockade does not change the migratory pattern of TMBP cells. Motility pattern of TMBP–GFP cells in the leptomeninges during the established phase of leptomeningeal TMBP cell infiltration before and 4 h after PTX treatment (as in Fig. 3a,b). Directional analysis and probability plots of T cell displacement for the Lévy (red), Weibull (blue) and normal (Gauss, green) distributions. The TMBP–GFP cells maintained their Brownian walk after PTX treatment. Analysis of at least eight TPLSM recordings. h, CXCR3 blockade alone recapitulates the effect of Gαi inhibition on TMBP–GFP cells released from the meninges into the CSF. Representative TPLSM overviews of spinal cord leptomeninges before and 8 h after i.v. injection of isotype control antibodies (control IgG) or anti-CXCR3 monoclonal antibody during established leptomeningeal TMBP cell infiltration. Magnification of individual regions originating from the overviews illustrate the position of the effector T cells in the z-axis in the leptomeningeal space via a colour-code scale. Note, TMBP–GFP cells are preferentially detached from the surface of the pia mater. Green, leptomeningeal blood vessels; red, phagocytes; blue, TMBP–Lifeact–Turquoise2 cells. Scale bars, 50 μm. *P < 0.05, ***P < 0.001 (e, f).

Extended Data Figure 9 Motility pattern of brain-non-reactive TOVA cells in the leptomeninges and their detachment into the CSF.

a, b, The motility pattern of TOVA cells in the leptomeninges resembles that of TMBP cells. TOVA–GFP cell motility in the leptomeninges during established inflammation (TPLSM 3.5 days after co-transfer with non-labelled TMBP cells). a, TOVA cells do not follow a preferential direction. Directional analyses of the TOVA–GFP cell tracks (as in Fig. 3a). Directions of in-plane movement (left) and associated probability for an angle varying within [0; 360] degrees in the fitting plane (right). b, TOVA cells move in a Brownian random walk. Motility pattern of TOVA–GFP cells (as in Fig. 3b). Analysis of at least 6 TPLSM recordings. c, TOVA–GFP cells are less adhesive to meningeal structures in comparison to their myelin-reactive counterparts. Mean contact duration (417 contacts per 127 cells); contact frequencies (127 cells) between MBP- or OVA-reactive T cells and meningeal phagocytes; mean velocities (304 cells) and meandering indices (236 cells) of MBP- or OVA-reactive T cells; percentage of time spent by these T cells in contact with meningeal phagocytes during the 30 min recording. Results from intravital TPLSM recordings during established meningeal inflammation on spinal cord leptomeninges, after transfer of TMBP–GFP cells or of TOVA–GFP cells together with non-labelled TMBP cells. Data are mean values ± s.e.m. of 3 independent experiments per transfer (two-tailed Mann–Whitney U-test). d, Chemokine receptors and integrin expression in TOVA–GFP cells. TOVA–GFP cells transferred into Lewis rats together with non-labelled TMBP cells were sorted from meninges and CNS parenchyma during established meningeal inflammation. Quantitative PCR. House-keeping gene, β-actin (Actb). Representative data of 3 independent experiments. Data are mean ± s.d. of duplicate measurements. e, Integrin or Gαi signalling interference induces a release of TOVA–GFP cells from the leptomeninges into the CSF. TOVA–GFP cells were co-injected with non-labelled TMBP cells. During established meningeal inflammation (3.5 days after transfer) animals were treated i.t. with PBS (control), anti-LFA-1/VLA-4 monoclonal antibodies (left) or PTX (right). Absolute numbers of TOVA–GFP cells in CSF, spinal cord meninges and parenchyma 4 h after treatment. Data are mean ± s.e.m. of representative data from at least 3 independent experiments per treatment including 3 animals per group (two-tailed Mann–Whitney U-test). fh, Gαi signalling interference does not change the motility pattern of TOVA cells. f, g, The Brownian walk of TOVA cells was not changed after PTX treatment. Directional analyses and motility pattern of TOVA–GFP cells 4 h after i.t. injection of PTX. Mathematical analyses of at least 6 TPLSM recordings (as in Fig. 3a, b) 4 days after co-transfer with TMBP cells. h, Interference with chemokine signalling does not affect TOVA cell straightness and velocity. Mean velocities (498 cells) and meandering indices (428 cells) of TOVA–GFP cells at the depicted time points of leptomeningeal T-cell inflammation. 30 min TPLSM recordings were acquired before and 4 h after i.t. treatment with PTX. *P < 0.05, **P < 0.01, ***P < 0.001 (ce).

Extended Data Figure 10 Reattachment of TMBP cells from the CSF to the leptomeninges and its regulation by integrins, chemokines and T-cell activation/CNS inflammation.

a, Distribution of reattached TMBP cells resembles that of early EAE lesions. TMBP–GFP cells are located diffusely in the meninges and the adjacent spinal cord parenchyma after reattachment from the CSF. TMBP–GFP cells retrieved from CNS tissue on day 3 after transfer were injected i.t. into naive animals. Fluorescence microscopy images of the fixed cervical spinal cord 20 h later. Insets, magnifications of areas of interest (right; I, II). Arrows, TMBP–GFP cells in the leptomeninges (white) and CNS parenchyma (yellow). Green, TMBP–GFP cells. Scale bars, 100 μm. b, Increased TMBP cell rolling and floating after forced ventilation indicate that respiration is a major driving force for the spinal cord CSF flow. Time-lapse video microscopy of TMBP–GFP cells during the established phase of leptomeningeal TMBP cell infiltration performed under 3 different conditions: (1) during standard conditions (control; respiration rate: 81 bpm; cardiac rate: 230 bpm); (2) during hyperventilation (hypervent; respiration rate: 100 bpm; cardiac rate: 200 bpm); and (3) during hyperventilation following administration of methylscopolamine (0.05 mg kg−1) to block the hyperventilation-induced vagal influence on the heart (Scopolamine/hypervent; respiration rate: 100 bpm; cardiac rate: 230 bpm). Hyperventilation induced a strong increase in rolling/floating TMBP–GFP cells that was not changed after methylscopolamine. 185 s time projections. Green, TMBP–GFP cells; white arrows, representative T cells rolling/floating in the CSF. Scale bars, 50 μm. Representative data of 3 independent experiments. c, TMBP cell transport in the CSF after i.t. injection. Distribution of TMBP–GFP cells in different levels of the spinal cord tissues after localized injection of the cells into the cisterna magna or the subarachnoidal space of the lumbar spinal cord. Flow cytometry analyses. Relative numbers of TMBP–GFP cells 6 and 24 h after i.t. transfer in the indicated CNS compartments. Combined results of 2 independent experiments for each i.t. injected site (n = 14). d, TMBP cell infiltration in the different levels of the spinal cord during transfer EAE. Quantification (flow cytometry) of TMBP–GFP cells in the meninges or parenchyma of the indicated parts of the spinal cord. Data are mean ± s.e.m. of representative data from 2 independent experiments per treatment including 2 or 3 animals per group (n = 41). e, Activated but not resting TMBP cells induce EAE after i.t. transfer. TMBP–GFP cells activated by antigenic stimulation or resting cells were i.t. transferred into naive animals. Clinical scores at the indicated time points. Representative results of 3 independent experiments with 4–6 animals per group. Data are mean ± s.d. (n = 28). f, Interference with integrin binding reduces the encephalitogenic potential of i.t. transferred TMBP–GFP cells. TMBP–GFP cell blasts were pre-treated with either an IgG control antibody or a combination of anti-LFA-1/VLA-4 monoclonal antibodies before i.t. transfer into naive animals. Clinical scores. Representative results of 3 independent experiments with 2 or 3 animals per group. Data are mean ± s.e.m. (n = 16, two-tailed Mann–Whitney U-test). g, TMBP cells invade the inflamed CNS tissue more efficiently than TOVA cells. Left, the same number of non-activated TMBP–GFP and TOVA–Cherry cells were co-injected i.t. into animals at the onset of clinical EAE, that is, 3 days after i.v. transfer with unlabelled TMBP–GFP cells. 20 h after i.t. injection, the entry of TMBP–GFP and TOVA–Cherry cells into the CNS tissue (including spinal cord meninges and parenchyma) was quantified by flow cytometry. Representative results of 2 independent experiments with 4 animals per group. Data are mean ± s.d. (n = 8). Right, resting or in vitro activated TOVA–GFP cells were i.t. injected into naive animals. In addition, a group of animals receiving resting TOVA–GFP cells were co-injected intrathecally with 25 μg of OVA antigen. 20 h after i.t. injection, TOVA–GFP cells in the spinal cord tissue were quantified. Representative of 2 independent experiments with 2 animals per group. Data are mean ± s.d. (n = 12, two-tailed Mann–Whitney U-test (left) or Kruskal–Wallis ANOVA followed by Dunn’s multiple comparison test (right)). h, Inflammation of the CNS tissue increases TMBP cell entry from the CSF. CSF-derived TMBP–GFP cells were injected i.t. either into naive animals or into animals at the onset of clinical EAE, that is, 3 days after i.v. transfer of non-labelled TMBP cells (that is, inflamed condition). Cell quantifications of TMBP–GFP cells in the CNS tissue and CSF by flow cytometry 20 h later when the re-transferred T cells had maximally infiltrated the inflamed CNS of the recipient animals. Data are mean ± s.d. of representative results of 3 independent experiments (n = 12, two-tailed Mann–Whitney U-test). i, Blocking of integrin or Gαi-signalling reduces TMBP cell migration from the CSF into the inflamed CNS tissue. CNS-derived TMBP–GFP cells were pre-treated with anti-VLA-4 or anti-LFA-1 monoclonal antibodies or with isotype control IgG antibody (control) (left) or PTX or PBS as control (right). Cells were then injected i.t. in animals at the onset of EAE and quantified as in h. Data are mean ± s.d. of representative results of 3 independent experiments (n = 18, Kruskal–Wallis ANOVA followed by Dunn’s multiple comparison test (left) or two-tailed Mann–Whitney U-test, right). jl, In active EAE induced by reactivated memory TMBP cells , trafficking of the memory TMBP cells between the distinct CNS compartments follows the same rules as the trafficking of TMBP cells during transfer EAE. j, Memory TMBP cells accumulate simultaneously in the meninges and the CSF before they occur in the choroid plexus. Quantification (flow cytometry) of GFP+TMBP-Memory cells in CSF, spinal cord meninges and parenchyma (left), or in CSF and choroid plexus of the fourth ventricle (right) at the indicated time points after immunization of 10-week-old memory animals. Data are mean ± s.d. of 3 independent experiments (n = 13). k, CSF-derived memory TMBP cells present a low activation profile. IFNγ and IL-17 expression (quantitative PCR) in GFP+memory TMBP cells 5 days after immunization. House-keeping gene, β-actin (Actb). Representative data of 3 independent experiments. Data are mean ± s.d. of duplicate measurements. l, Integrin and chemokine blockade reduces the number of TMBP-Memory cells in the leptomeninges during active EAE. Intravital TPLSM recordings of spinal cord leptomeninges from 10-week-old memory animals 5 days after immunization. Quantification of GFP+memory TMBP cells in the acquired images before (control) and 4 h after i.t. treatment of either anti-LFA-1/VLA-4 or anti-CXCR3 blocking monoclonal antibodies. Data are mean values ± s.e.m. from 3 independent experiments including 5,947 cells (two-tailed Mann–Whitney U-test). *P < 0.05, **P < 0.01, ***P < 0.001 (gi, l). m, Schematic representation of the migratory behaviour of effector T cells in the leptomeningeal milieu and the roles of chemokines, integrins and activation in controlling the T-cell migration steps. Question marks indicate unresolved points.

Supplementary information

Supplementary Table 1

T-cell activation, but not differentiation, adhesion or cell motility-related genes are differentially regulated in TMBP cells in the CSF - RNA-Seq data depicting groups of genes categorized according to their function. Red: genes differentially regulated (≥2.5-fold) in TMBP cells in the CSF compared to CNS parenchyma. Gene names in blue: master transcription factors that determine the differentiation of CD4+ T-helper cell subsets. Mean ±s.d. (XLSX 37 kb)

Supplementary Table 2

Genes differentially regulated between the CSF and the CNS parenchyma cluster in pathways related to T-cell activation - Pathway enrichment analysis using the KEGG database on the genes significantly regulated between CSF and CNS parenchyma in TMBP cells. (XLSX 16 kb)

Supplementary Table 3

Genes differentially regulated between the CSF and the leptomeninges cluster in pathways related to T-cell activation - KEGG pathway analysis on genes significantly regulated between CSF and leptomeninges in TMBP cells. (XLSX 14 kb)

TMBP cell extravasation from leptomeningeal blood vessels

Intravital TPLSM recording was performed on the thoracic SC during the pre-invasion phase of leptomeningeal TMBP cell infiltration, where the vast majority of both TMBP cells (blue) and CD11b+ myeloid cells (magenta) were located within the vascular compartment. Depicted is an example of a TMBP cell (white arrowhead) transgressing the vascular wall after having crawled within the lumen of a leptomeningeal blood vessel. Blue: TMBP-LifeAct-Turquoise2 cells. Red: blood vessels. Magenta: CD11b+ myeloid cells. Scale bars: 25 µm. Time interval: 32 sec. (AVI 7472 kb)

Detachment of TMBP cells from the leptomeninges into the CSF

Intravital TPLSM recording of the thoracic SC was performed during established leptomeningeal TMBP cell infiltration. 1st half: Representative TMBP-GFP cells (white circles) detaching from the pial surface and subsequently being dragged away within the CSF. 2nd half: Examples of TMBP-GFP cells rolling along the pial surface (white arrows). Green: TMBP-GFP cells. Red: blood vessels, meningeal phagocytes. Blue: collagen fibers. Scale bar: 50 µm. Time interval: 32 sec. Representative recording of five independent experiments. (AVI 9838 kb)

T-cell detachment from the leptomeninges verified by fast-acquisition fluorescence video microscopy

Intravital fluorescence video microscopy of the thoracic SC was performed during established leptomeningeal TMBP cell infiltration. Example of a TMBP-GFP cell detaching from the pial surface (false color: orange) and subsequently being dragged away with the CSF is depicted. Green: TMBP-GFP cells. Scale bar: 50 µm. Time interval: 5.8 sec. Representative recording of 5 independent experiments. (AVI 5504 kb)

Detachment of photoconverted TMBP-Dendra2 cells from the leptomeninges into the CSF

Intravital TPLSM recording of the thoracic SC was performed 30 min after photoconversion of TMBP-Dendra2 cells during established leptomeningeal TMBP cell infiltration. Depicted are examples of photoconverted TMBP cells (white circles) detaching from the pial surface and subsequently being dragged away with the CSF. White circle (dotted line): example of a T cell re-attaching from the CSF to the pial surface. Red (auto-fluorescence): leptomeningeal blood vessels, meningeal phagocytes. Yellow: photoconverted TMBP-Dendra2 cells. Green: non-photoconverted TMBP-Dendra2 cells. Scale bar: 50 µm. Time interval: 52 sec. Representative recording of three independent TPLSM recordings. (AVI 2062 kb)

TMBP cells are, for the majority of their crawling time, in direct contact with meningeal macrophages

Surface-rendered 3D reconstruction of an intravital TPLSM recording performed on the thoracic SC during established leptomeningeal TMBP cell infiltration. TMBP-GFP cells in contact with meningeal macrophages are depicted in yellow (magenta-colored circles), when not in contact in green (white-colored circles). Red: meningeal macrophages. Scale bar: 50 µm. Time interval: 32 sec. Representative recordings of three independent experiments. (AVI 2502 kb)

Integrin blockade results in a massive release of TMBP cells from the leptomeninges into the CSF

Intravital TPLSM recordings during established leptomeningeal TMBP cell infiltration were performed before (top) and 4 h after combined infusion of anti-VLA-4/LFA-1 mAbs (bottom). Note that after infusion of blocking antibodies a substantial fraction of TMBP-GFP cells is floating within the CSF appearing as green streaks (white circles) or rolling along the pial surface looking like pearl-strings (white arrows). Blue: collagen fibers. Scale bars: 100 µm. Time interval: 48 sec. Representative recordings of three independent experiments. (AVI 11244 kb)

Chemokine blockade via PTX induces TMBP cell detachment from the leptomeninges.

Intravital TPLSM recordings of the thoracic SC were performed during established leptomeningeal TMBP cell infiltration before (left) or 4 h after infusion of PTX (right). Orange (false color): examples of TMBP cells being released from meningeal phagocytes (Mɸ, dotted lines) into the CSF. Green: TMBP-GFP cells. Gray (false color): vessel lumen, meningeal phagocytes. Scale bars: 50 µm. Time interval: 32 sec. Representative recordings of five independent experiments. (AVI 3752 kb)

Hyperventilation induces an increase of rolling/floating TMBP cells in the CSF

Fluorescence video microscopy recordings were performed during established leptomeningeal TMBP cell infiltration under steady state conditions (control, respiratory rate: 81 bpm) and during hyperventilation (respiratory rate: 100 bpm). Green: TMBP-GFP cells. Arrows point to representative rolling/floating TMBP-GFP cells. Time interval: 8.4 sec. Scale bar: 50 µm. Representative recordings of three independent experiments. (AVI 4000 kb)

Re-attachment of TMBP cells from the CSF to the leptomeninges

Intravital TPLSM recordings on thoracic SC were performed during established leptomeningeal TMBP cell infiltration. Depicted are examples of TMBP-GFP cells re-attaching from the CSF to the pial surface (yellow circles). Note that after re-attachment T cells show either short, intermittent adhesive interactions with the pial surface (rolling; pattern 1, 1st half), or they stably adhere to the leptomeningeal structures (capture) followed by subsequent crawling (pattern 2, 2nd half). Green: TMBP-GFP cells. Red: blood vessels, meningeal phagocytes. Blue: collagen fibers. Scale bars: 50 µm. Time interval: 32 sec. Representative recordings of 5 independent experiments. (AVI 9926 kb)

Re-attachment of TMBP cells from the CSF to the leptomeninges visualized by fast-acquisition fluorescence video microscopy

Intravital fluorescence video microscopy was performed on the thoracic SC during established leptomeningeal TMBP cell infiltration. Depicted is an example of a TMBP-GFP cell (false color: orange) that after a brief period of rolling arrests (capture). Subsequently, the cell continues its way by further rolling along the pial surface. Green: TMBP-GFP cells. Scale bar: 50 µm. Time interval: 9.8 sec. Representative recording of three independent experiments. (AVI 2102 kb)

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Schläger, C., Körner, H., Krueger, M. et al. Effector T-cell trafficking between the leptomeninges and the cerebrospinal fluid. Nature 530, 349–353 (2016). https://doi.org/10.1038/nature16939

Download citation

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Search

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