The cytokine transforming growth factor-β (TGF-β) regulates the development and homeostasis of several tissue-resident macrophage populations, including microglia. TGF-β is not critical for microglia survival but is required for the maintenance of the microglia-specific homeostatic gene signature1,2. Under defined host conditions, circulating monocytes can compete for the microglial niche and give rise to long-lived monocyte-derived macrophages residing in the central nervous system (CNS)3,4,5. Whether monocytes require TGF-β for colonization of the microglial niche and maintenance of CNS integrity is unknown. We found that abrogation of TGF-β signaling in CX3CR1+ monocyte-derived macrophages led to rapid onset of a progressive and fatal demyelinating motor disease characterized by myelin-laden giant macrophages throughout the spinal cord. Tgfbr2-deficient macrophages were characterized by high expression of genes encoding proteins involved in antigen presentation, inflammation and phagocytosis. TGF-β is thus crucial for the functional integration of monocytes into the CNS microenvironment.

  • Subscribe to Nature Immunology for full access:



Additional access options:

Already a subscriber?  Log in  now or  Register  for online access.

Additional information

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

Change history

  • Update 19 April 2018

    Through publisher error, this paper was initially published incorrectly as an Article instead of a Letter. The error has been corrected for all versions of the paper.


  1. 1.

    Butovsky, O. et al. Identification of a unique TGF-β-dependent molecular and functional signature in microglia. Nat. Neurosci. 17, 131–143 (2014).

  2. 2.

    Buttgereit, A. et al. Sall1 is a transcriptional regulator defining microglia identity and function. Nat. Immunol. 17, 1397–1406 (2016).

  3. 3.

    Mildner, A. et al. Microglia in the adult brain arise from Ly-6ChiCCR2+ monocytes only under defined host conditions. Nat. Neurosci. 10, 1544–1553 (2007).

  4. 4.

    Varvel, N. H. et al. Microglial repopulation model reveals a robust homeostatic process for replacing CNS myeloid cells. Proc. Natl. Acad. Sci. USA 109, 18150–18155 (2012).

  5. 5.

    Bruttger, J. et al. Genetic cell ablation reveals clusters of local self-renewing microglia in the mammalian central nervous system. Immunity 43, 92–106 (2015).

  6. 6.

    Guilliams, M. & Scott, C. L. Does niche competition determine the origin of tissue-resident macrophages? Nat. Rev. Immunol. 17, 451–460 (2017).

  7. 7.

    Scott, C. L. et al. Bone marrow-derived monocytes give rise to self-renewing and fully differentiated Kupffer cells. Nat. Commun. 7, 10321 (2016).

  8. 8.

    van de Laar, L. et al. Yolk sac macrophages, fetal liver, and adult monocytes can colonize an empty niche and develop into functional tissue-resident macrophages. Immunity 44, 755–768 (2016).

  9. 9.

    Buch, T. et al. A Cre-inducible diphtheria toxin receptor mediates cell lineage ablation after toxin administration. Nat. Methods 2, 419–426 (2005).

  10. 10.

    Parkhurst, C. N. et al. Microglia promote learning-dependent synapse formation through brain-derived neurotrophic factor. Cell 155, 1596–1609 (2013).

  11. 11.

    Wu, S., Wu, Y. & Capecchi, M. R. Motoneurons and oligodendrocytes are sequentially generated from neural stem cells but do not appear to share common lineage-restricted progenitors in vivo. Development 133, 581–590 (2006).

  12. 12.

    Yu, X. et al. The cytokine TGF-β promotes the development and homeostasis of alveolar macrophages. Immunity 47, 903–912.e4 (2017).

  13. 13.

    Bobr, A. et al. Autocrine/paracrine TGF-β1 inhibits Langerhans cell migration. Proc. Natl. Acad. Sci. USA 109, 10492–10497 (2012).

  14. 14.

    Malipiero, U. et al. TGFbeta receptor II gene deletion in leucocytes prevents cerebral vasculitis in bacterial meningitis. Brain 129, 2404–2415 (2006).

  15. 15.

    Novitskiy, S. V. et al. Deletion of TGF-β signaling in myeloid cells enhances their anti-tumorigenic properties. J. Leukoc. Biol. 92, 641–651 (2012).

  16. 16.

    Li, J. et al. Myeloid TGF-β signaling contributes to colitis-associated tumorigenesis in mice. Carcinogenesis 34, 2099–2108 (2013).

  17. 17.

    Parsa, R. et al. TGFβ regulates persistent neuroinflammation by controlling Th1 polarization and ROS production via monocyte-derived dendritic cells. Glia 64, 1925–1937 (2016).

  18. 18.

    Nijssen, J., Comley, L. H. & Hedlund, E. Motor neuron vulnerability and resistance in amyotrophic lateral sclerosis. Acta Neuropathol. 133, 863–885 (2017).

  19. 19.

    Nagao, M., Misawa, H., Kato, S. & Hirai, S. Loss of cholinergic synapses on the spinal motor neurons of amyotrophic lateral sclerosis. J. Neuropathol. Exp. Neurol. 57, 329–333 (1998).

  20. 20.

    Wootz, H. et al. Alterations in the motor neuron–renshaw cell circuit in the Sod1G93A mouse model. J. Comp. Neurol. 521, 1449–1469 (2013).

  21. 21.

    Sekar, A. et al. Schizophrenia risk from complex variation of complement component 4. Nature 530, 177–183 (2016).

  22. 22.

    Dambuza, I. M. & Brown, G. D. C-type lectins in immunity: recent developments. Curr. Opin. Immunol. 32, 21–27 (2015).

  23. 23.

    Borrego, F. The CD300 molecules: an emerging family of regulators of the immune system. Blood 121, 1951–1960 (2013).

  24. 24.

    Krasemann, S. et al. The TREM2-APOE pathway drives the transcriptional phenotype of dysfunctional microglia in neurodegenerative diseases. Immunity 47, 566–581.e9 (2017).

  25. 25.

    Wong, K. et al. Mice deficient in NRROS show abnormal microglial development and neurological disorders. Nat. Immunol. 18, 633–641 (2017).

  26. 26.

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

  27. 27.

    Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. B 57, 289–300 (1995).

  28. 28.

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

Download references


We thank A. van Vollenhoven for flow cytometry. We thank the staff at AKM, particularly E. Qvist, for animal caretaking and M. Gustafsson for performing tail-vein injections. We thank A. Svensson, D. Sunnemark and A. Dahlstrand for help with slide scanning, performed at Offspring Biosciences, Södertälje. We thank M. Li (Sloan Kettering Institute) for Tgfbr2fl/fl mice and F. Wermeling (Karolinska Institutet) for CD45.1 mice. We acknowledge support from the Science for Life Laboratory, the National Genomics Infrastructure (NGI) and Uppmax for providing assistance in next-generation sequencing and computational infrastructure.

This work was supported by grants from the Swedish Alzheimer Foundation (Alzheimerfonden, AF-74004, R.A.H.), Swedish Research Council (Vetenskapsrådet, 2014-02087, R.A.H. and K2015-61X-20776-08-3, M.J.), Swedish Childhood Cancer Foundation (Barncancerfonden, PR2014-0154 and NCP2015-0064, X.-M.Z.), Åke Wibergs stiftelse (M14-0263, X.-M.Z.), Foundation of Swedish MS Research (MS Forskningsfonden, L.K.), Swedish Heart-Lung Foundation (Hjärt-Lungfonden, M.J.F. and D.F.J.K.), the Novo Nordisk Foundation (NNF15CC0018346, M.J.F. and D.F.J.K.), NIH-NINDS (1R01NS088137, O.B.), NIH-NIA (R01AG051812 and R01AG054672, O.B.), National Multiple Sclerosis Society (5092A1, O.B.) and Amyotrophic Lateral Sclerosis Association (ALSA2087, O.B.), and by a Nancy Davis Foundation Faculty Award (O.B.). X.-M.Z. was supported by a fellowship from the Swedish Childhood Cancer Foundation (Barncancerfonden, NC2014-0046, NBCNS). L.K. was supported by a fellowship from the Margaretha af Ugglas Foundation.

Author information

Author notes

  1. These authors contributed equally: Melanie Pieber and Roham Parsa.

  2. These authors jointly supervised this work: Xing-Mei Zhang and Robert A. Harris.


  1. Department of Clinical Neuroscience, Applied Immunology and Immunotherapy, Karolinska Institutet, Center for Molecular Medicine, Karolinska Hospital at Solna, Stockholm, Sweden

    • Harald Lund
    • , Melanie Pieber
    • , Roham Parsa
    • , David Grommisch
    • , Jinming Han
    • , Keying Zhu
    • , Xing-Mei Zhang
    •  & Robert A. Harris
  2. Department of Clinical Neuroscience, Center for Molecular Medicine, Karolinska Institutet, Stockholm, Sweden

    • Ewoud Ewing
    • , Lara Kular
    • , Maria Needhamsen
    • , Sabrina Ruhrmann
    • , André Ortlieb Guerreiro-Cacais
    • , Rasmus Berglund
    •  & Maja Jagodic
  3. Department of Neuroscience, Karolinska Institutet, Stockholm, Sweden

    • Jik Nijssen
    •  & Eva Hedlund
  4. Department of Medicine, Cardiovascular Medicine Unit, Center for Molecular Medicine, Karolinska Institutet, Stockholm, Sweden

    • Maria J. Forteza
    •  & Daniel F. J. Ketelhuth
  5. Department of Neurology, Ann Romney Center for Neurologic Diseases, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA

    • Oleg Butovsky
  6. Evergrande Center for Immunologic Diseases, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA

    • Oleg Butovsky


  1. Search for Harald Lund in:

  2. Search for Melanie Pieber in:

  3. Search for Roham Parsa in:

  4. Search for David Grommisch in:

  5. Search for Ewoud Ewing in:

  6. Search for Lara Kular in:

  7. Search for Jinming Han in:

  8. Search for Keying Zhu in:

  9. Search for Jik Nijssen in:

  10. Search for Eva Hedlund in:

  11. Search for Maria Needhamsen in:

  12. Search for Sabrina Ruhrmann in:

  13. Search for André Ortlieb Guerreiro-Cacais in:

  14. Search for Rasmus Berglund in:

  15. Search for Maria J. Forteza in:

  16. Search for Daniel F. J. Ketelhuth in:

  17. Search for Oleg Butovsky in:

  18. Search for Maja Jagodic in:

  19. Search for Xing-Mei Zhang in:

  20. Search for Robert A. Harris in:


H.L. conceived the study with R.A.H., X.-M.Z. and R.P. H.L., R.P., R.A.H., X.-M.Z. and O.B. designed experiments. H.L., M.P., R.P., D.G. and X.-M.Z. performed most of the experiments and analyzed the data. Additional experiments and data analysis or design were performed by J.H., A.O.G.C. and R.B. (flow cytometry); S.R., L.K., E.E., M.N. and M.J. (bioinformatics); and J.N., E.H., K.Z., D.F.J.K. and M.J.F. (pathology). O.B. provided reagents. H.L. wrote the paper. All authors discussed the results and contributed to the final manuscript.

Competing interests

The authors declare no competing interests.

Corresponding author

Correspondence to Robert A. Harris.

Integrated supplementary information

  1. Supplementary Figure 1 Repopulation of microglia by peripheral myeloid cells in chimeric Cx3cr1CreER/+R26DTR/+ and Cx3cr1CreER/+R26DTA/+ mice.

    (a) Gating strategy of CNS cells employed in the study. (b-c) Analysis of CNS, top panel gated on live, singlet and Ly6CLy6G cells in (b) CD45.1→Cx3cr1CreER/+R26DTR/+ and (c) CD45.1→Cx3cr1CreER/+R26DTA/+ chimeras. Values in dot plots are mean±s.d. of (b) n=5 and 6 mice from two pooled experiments and (c) n=5 mice/group from one of two independent experiments.

  2. Supplementary Figure 2 Lack of adaptive immunological changes in Cx3cr1CreER/+Tgfbr2fl/fl→DTA chimeras.

    (a) Analysis of T cell (CD11bCD45hiCD3+) and B cell (CD11bCD45hiMHCII+) numbers in the brain and spinal cord of the indicated chimeras at day 20 after TAM. Lines in graphs represent mean values of n=3 and 4 mice. The experiment was performed once. (b) Analysis of proliferating T cells, CD4+FoxP3+ regulatory T cells and IFN-γ, IL-17 and TNF-α producing CD4+ T cells in the spleens of the indicated chimeras after onset of motor symptoms (day 20 after TAM). Values in plots are mean±s.d. of n=6 and 5 mice. The experiment was performed once.

  3. Supplementary Figure 3 Analysis of lower motor neuron pathology and thalamic lesions in Cx3cr1CreER/+Tgfbr2fl/fl→DTA chimeras.

    (a) Analysis of innervation of neuromuscular junctions in the tibialis anterior muscle of Tgfbr2fl/fl→DTA and Cx3cr1CreER/+Tgfbr2fl/fl→DTA chimeras. Innervation was quantified by assessing overlay of synaptic vesicle protein (SV2) and neurofilament (NF) 165 kDA staining with α-BTX labeled endplates. Positive control is tibials anterior muscle from SODG93A mice (day 140) illustrating complete denervation. Lines in graph are mean values of n = 4, 2, 3, 4 mice/group and at least 40 motor endplates/mouse. Scale bar 50µm. (b) Enumeration of cresyl violet stained motor neurons (>200μm2) in the dorsal horns of the cervical enlargement. Lines in graph are mean values of n=4 mice/group and at least 5 sections/mouse. NS=not significant P=0.1633 Student’s two-tailed unpaired t-test. Scale bar 100µm. (c) Enumeration of VAChT+ motor neurons in the ventral horns of the spinal cord. Representative images are taken from the lumbar enlargement. Lines are mean values of n= 4, 3, 4 mice and at least 4 sections/mouse. * P= 0.0354 (cervical) and * P=0.0487 ** P=0.0034 (lumbar) 1-way ANOVA Dunnett’s Multiple Comparison test. Scale bar 200μm. (d) Hematoxylin and eosin staining visualizing lesions in the thalamus in end stage KO→DTA mice. Lesions were quantified based on the presence of giant cells (white arrow) with an area >150µm2. Lines are mean values of n=3 mice/group and 2 sections/mouse. Scale bar 100µm (top panel) and 25µm (bottom panel). (e) F4/80 and fluoromyelin staining in the thalamus. Images are representative of n=3 mice/group. Scale bar 500µm and 50µm. (f) Confocal images of F4/80 and fluoromyelin staining in the thalamus. Images are representative of n=3 mice/group. Scale bar 10µm.

  4. Supplementary Figure 4 Loss of Tgfbr2 in CX3CR1+ macrophages after niche colonization leads to onset of motor phenotype.

    (a) Experimental setup of Cx3cr1CreER/+Tgfbr2fl/fl→DTR chimeras. DTR mice were given TAM (two doses, 48h apart) to recombine CNS-resident microglia, followed by irradiation and bone marrow transplantation (BMT) from Tgfbr2fl/fl or Cx3cr1CreER/+Tgfbr2fl/fl donors one week later. After reconstitution 6 weeks later, mice were given DT to deplete microglia. Tgfbr2 expression was targeted in repopulated CX3CR1+ monocyte-derived macrophages 3 weeks later by TAM administration. (b) Clinical score assessing motor symptoms on the indicated days after TAM. Values in plot are mean±s.e.m. of n=4 and 5 mice. The experiment was performed once. (c) 4-paw hanging wire test to assess grip strength performed at onset of motor symptoms (day 20 after TAM). Lines are mean values of n=4 and 5 mice. ** P=0.0032. Student’s two-tailed unpaired t-test. (d) Analysis of CNS by flow cytometry at day 46 (end stage). Top panel gated on CD11b+CD45+Ly6C. The data is representative of n=2 mice/group. (e) Kinetic analysis of the transformation of monocyte-derived macrophages after TAM. Gated on CD11b+CD45+Ly6CLy6GF4/80hi. Lines in graphs are mean values of n=2, 2, 2, 3 mice.

  5. Supplementary Figure 5 Loss of Tgfbr2 in CNS-resident microglia leads to slowly progressing motor disease.

    (a) qRT-PCR analysis of sorted Ly6CCD11b+F4/80low (Tgfbr2fl/fl) and Ly6CCD11b+F4/80hi (Cx3cr1CreER/+Tgfbr2fl/fl) microglia day 7 after TAM. Deletion of Tgfbr2 was assessed by the expression of the floxed exon 2 divided by the unfloxed exon 1. n=2 samples/group sorted from pools of 1-5 mice/sample. (b) CNS analysis by flow cytometry 7 days after TAM. Top panel is gated on live, singlet and Ly6CLy6G cells. Dot plots are representative of three independent experiments. (c-d) Immunofluorescence for (c) F4/80 and Iba-1 or (d) P2ry12 in cortex day 7 after TAM. Images are representative of n=3mice/group. Scale bar 100μm. (e) Confocal images of one Iba-1+ microglial cell in the cortex to visualize morphology. Images are representative of n=3 mice/group. Scale bar 10μm. (f) CNS analysis of head-protected CD45.1→Cx3cr1CreER/+Tgfbr2fl/fl chimeras analyzed 3 weeks after TAM administration. Top panel is gated on live, singlet and Ly6CLy6G cells. Values in plots are mean±s.d. of n=7 and 6 mice. The experiment was performed once. (g-h) Development of motor symptoms in n=7 Tgfbr2fl/fl and n=6 Cx3cr1CreER/+Tgfbr2fl/fl mice. (g) 4-paw hanging wire test to assess grip strength (max time 180 sec) performed on the indicated days after TAM. Lines represent mean values. (h) Clinical score assessing motor symptoms. Values are mean±s.e.m. The experiment was performed once. (i) Time course analysis of MHCII surface expression on CD11b+CD45+Ly6CLy6G microglia on the indicated days after TAM. Lines represent mean values of n=15, 3, 8, 3 mice.

Supplementary information

  1. Supplementary Text and Figures

    Supplementary Figures 1–5

  2. Reporting Summary

  3. Supplementary Table 1:

    Top 20 predicted upstream regulators of the differentially expressed gene set between Tgfbr2–/– macrophages and WT macrophages, using Ingenuity Pathway Analysis. n=3 samples/group from 2-4 pooled mice/sample. Overlap p-values are calculated using Fisher’s exact test and P<0.01 are considered significant

  4. Supplementary Table 2: Primer sequences used for qRT-PCR analysis

  5. Supplementary Video 1:

    Free movement of Tgfbr2fl/fl→DTA chimeric mice day 21 after TAM administration. The video is representative of n=17 mice from three independent experiments

  6. Supplementary Video 2:

    Free movement of Cx3cr1CreER/+Tgfbr2fl/fl→DTA chimeric mice day 21 after TAM administration. The video is representative of n=26 mice from three independent experiments

  7. Supplementary Video 3:

    4-paw hanging-wire trial day 10 after TAM administration in Tgfbr2fl/fl→DTA chimeras. The video is representative of n=7 mice. The experiment was repeated from a greater height in which all mice remained suspended for 180 seconds (included in Figure 2c)

  8. Supplementary Video 4:

    4-paw hanging-wire trial day 10 after TAM administration in Cx3cr1CreER/+Tgfbr2fl/fl→DTA chimeras. The video is representative of n=11 mice. The experiment was repeated from a greater height

  9. Supplementary Video 5:

    Video of Cx3cr1CreER/+Tgfbr2fl/fl→DTA chimeric mice at end stage (day 32-40 after TAM). Of 19 mice in two independent experiments that were followed long-term for symptoms, 15 reached the end stage within 40 days of TAM administration