Commensal bacteria direct selective cargo sorting to promote symbiosis

Abstract

Mucosal immunity protects a host from intestinal inflammation and infection and is profoundly influenced by symbiotic bacteria. Here we report that in mice symbiotic bacteria directed selective cargo sorting in Paneth cells to promote symbiosis through Nod2, a cytosolic bacterial sensor, and the multifunctional protein kinase LRRK2, both encoded by inflammatory bowel disease (IBD)-associated genes. Commensals recruited Nod2 onto lysozyme-containing dense core vesicles (DCVs), which was required for DCV localization of LRRK2 and a small GTPase, Rab2a. Deficiency of Nod2, LRRK2 or Rab2a or depletion of commensals resulted in lysosomal degradation of lysozyme. Thus, commensal bacteria and host factors orchestrate the lysozyme-sorting process to protect the host from enteric infection, implicating Paneth cell dysfunction in IBD pathogenesis.

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: Lrrk2−/− mice are more susceptible to Listeria infection.
Figure 2: Lrrk2−/− Paneth cells are lysozyme deficient.
Figure 3: Lysozyme is degraded in Lrrk2−/− Paneth cells.
Figure 4: LRRK2 controls lysozyme sorting in DCVs by recruiting Rab2a.
Figure 5: Commensal bacteria regulate lysozyme sorting.
Figure 6: Nod2 mediates bacterially directed lysozyme sorting.
Figure 7: MDP restores lysozyme sorting in GF mice.

References

  1. 1

    Cash, H.L., Whitham, C.V., Behrendt, C.L. & Hooper, L.V. Symbiotic bacteria direct expression of an intestinal bactericidal lectin. Science 313, 1126–1130 (2006).

    CAS  PubMed  PubMed Central  Google Scholar 

  2. 2

    Alenghat, T. & Artis, D. Epigenomic regulation of host-microbiota interactions. Trends Immunol. 35, 518–525 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  3. 3

    Furusawa, Y., Obata, Y. & Hase, K. Commensal microbiota regulates T cell fate decision in the gut. Semin. Immunopathol. 37, 17–25 (2015).

    Google Scholar 

  4. 4

    Hugot, J.P. et al. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn's disease. Nature 411, 599–603 (2001).

    CAS  Google Scholar 

  5. 5

    Ogura, Y. et al. A frameshift mutation in NOD2 associated with susceptibility to Crohn's disease. Nature 411, 603–606 (2001).

    CAS  Google Scholar 

  6. 6

    Kobayashi, K.S. et al. Nod2-dependent regulation of innate and adaptive immunity in the intestinal tract. Science 307, 731–734 (2005).

    CAS  Google Scholar 

  7. 7

    Ramanan, D., Tang, M.S., Bowcutt, R., Loke, P. & Cadwell, K. Bacterial sensor Nod2 prevents inflammation of the small intestine by restricting the expansion of the commensal Bacteroides vulgatus. Immunity 41, 311–324 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  8. 8

    Petnicki-Ocwieja, T. et al. Nod2 is required for the regulation of commensal microbiota in the intestine. Proc. Natl. Acad. Sci. USA 106, 15813–15818 (2009).

    CAS  Google Scholar 

  9. 9

    Barrett, J.C. et al. Genome-wide association defines more than 30 distinct susceptibility loci for Crohn's disease. Nat. Genet. 40, 955–962 (2008).

    CAS  PubMed  PubMed Central  Google Scholar 

  10. 10

    Franke, A. et al. Genome-wide meta-analysis increases to 71 the number of confirmed Crohn's disease susceptibility loci. Nat. Genet. 42, 1118–1125 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  11. 11

    Anderson, C.A. et al. Meta-analysis identifies 29 additional ulcerative colitis risk loci, increasing the number of confirmed associations to 47. Nat. Genet. 43, 246–252 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  12. 12

    Zimprich, A. et al. Mutations in LRRK2 cause autosomal-dominant parkinsonism with pleomorphic pathology. Neuron 44, 601–607 (2004).

    CAS  PubMed  Google Scholar 

  13. 13

    Paisán-Ruíz, C. et al. Cloning of the gene containing mutations that cause PARK8-linked Parkinson's disease. Neuron 44, 595–600 (2004).

    Google Scholar 

  14. 14

    Liu, Z. et al. The kinase LRRK2 is a regulator of the transcription factor NFAT that modulates the severity of inflammatory bowel disease. Nat. Immunol. 12, 1063–1070 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  15. 15

    Gardet, A. et al. LRRK2 is involved in the IFN-γ response and host response to pathogens. J. Immunol. 185, 5577–5585 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  16. 16

    Clevers, H.C. & Bevins, C.L. Paneth cells: maestros of the small intestinal crypts. Annu. Rev. Physiol. 75, 289–311 (2013).

    CAS  Google Scholar 

  17. 17

    Vaishnava, S., Behrendt, C.L., Ismail, A.S., Eckmann, L. & Hooper, L.V. Paneth cells directly sense gut commensals and maintain homeostasis at the intestinal host-microbial interface. Proc. Natl. Acad. Sci. USA 105, 20858–20863 (2008).

    CAS  Google Scholar 

  18. 18

    Adolph, T.E. et al. Paneth cells as a site of origin for intestinal inflammation. Nature 503, 272–276 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  19. 19

    Wilson, C.L. et al. Regulation of intestinal α-defensin activation by the metalloproteinase matrilysin in innate host defense. Science 286, 113–117 (1999).

    CAS  PubMed  PubMed Central  Google Scholar 

  20. 20

    Kaser, A. et al. XBP1 links ER stress to intestinal inflammation and confers genetic risk for human inflammatory bowel disease. Cell 134, 743–756 (2008).

    CAS  PubMed  PubMed Central  Google Scholar 

  21. 21

    Salzman, N.H., Ghosh, D., Huttner, K.M., Paterson, Y. & Bevins, C.L. Protection against enteric salmonellosis in transgenic mice expressing a human intestinal defensin. Nature 422, 522–526 (2003).

    CAS  Google Scholar 

  22. 22

    Brandl, K., Plitas, G., Schnabl, B., DeMatteo, R.P. & Pamer, E.G. MyD88-mediated signals induce the bactericidal lectin RegIIIγ and protect mice against intestinal Listeria monocytogenes infection. J. Exp. Med. 204, 1891–1900 (2007).

    CAS  PubMed  PubMed Central  Google Scholar 

  23. 23

    Chu, H. et al. Human α-defensin 6 promotes mucosal innate immunity through self-assembled peptide nanonets. Science 337, 477–481 (2012).

    CAS  PubMed  PubMed Central  Google Scholar 

  24. 24

    Cadwell, K. et al. A key role for autophagy and the autophagy gene Atg16l1 in mouse and human intestinal Paneth cells. Nature 456, 259–263 (2008).

    CAS  PubMed  PubMed Central  Google Scholar 

  25. 25

    Sumakovic, M. et al. UNC-108/RAB-2 and its effector RIC-19 are involved in dense core vesicle maturation in Caenorhabditis elegans. J. Cell Biol. 186, 897–914 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  26. 26

    Ayabe, T. et al. Secretion of microbicidal α-defensins by intestinal Paneth cells in response to bacteria. Nat. Immunol. 1, 113–118 (2000).

    CAS  Google Scholar 

  27. 27

    Sato, T. et al. Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459, 262–265 (2009).

    CAS  Google Scholar 

  28. 28

    Bowman, G.R., Cowan, A.T. & Turkewitz, A.P. Biogenesis of Dense-core Secretory Granules (Landes Bioscience and Springer Science + Business Media, 2009).

  29. 29

    Kim, T., Gondre-Lewis, M.C., Arnaoutova, I. & Loh, Y.P. Dense-core secretory granule biogenesis. Physiology (Bethesda) 21, 124–133 (2006).

    CAS  Google Scholar 

  30. 30

    Edwards, S.L. et al. Impaired dense core vesicle maturation in Caenorhabditis elegans mutants lacking Rab2. J. Cell Biol. 186, 881–895 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  31. 31

    Sasidharan, N. et al. RAB-5 and RAB-10 cooperate to regulate neuropeptide release in Caenorhabditis elegans. Proc. Natl. Acad. Sci. USA 109, 18944–18949 (2012).

    CAS  Google Scholar 

  32. 32

    Satoh, Y., Ishikawa, K., Ono, K. & Vollrath, L. Quantitative light microscopic observations on Paneth cells of germ-free and ex-germ-free Wistar rats. Digestion 34, 115–121 (1986).

    CAS  Google Scholar 

  33. 33

    Nieuwenhuis, E.E. et al. Cd1d-dependent regulation of bacterial colonization in the intestine of mice. J. Clin. Invest. 119, 1241–1250 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  34. 34

    Shanahan, M.T. et al. Mouse Paneth cell antimicrobial function is independent of Nod2. Gut 63, 903–910 (2014).

    CAS  Google Scholar 

  35. 35

    Ogura, Y. et al. Expression of NOD2 in Paneth cells: a possible link to Crohn's ileitis. Gut 52, 1591–1597 (2003).

    CAS  PubMed  PubMed Central  Google Scholar 

  36. 36

    Barr, F.A. Review series: Rab GTPases and membrane identity: causal or inconsequential? J. Cell Biol. 202, 191–199 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  37. 37

    Aloisi, A.L. & Bucci, C. Rab GTPases-cargo direct interactions: fine modulators of intracellular trafficking. Histol. Histopathol. 28, 839–849 (2013).

    CAS  Google Scholar 

  38. 38

    VanDussen, K.L. et al. Genetic variants synthesize to produce Paneth cell phenotypes that define subtypes of Crohn's disease. Gastroenterology 146, 200–209 (2014).

    CAS  Google Scholar 

  39. 39

    Cadwell, K., Patel, K.K., Komatsu, M. & Virgin, H.W. 4th & Stappenbeck, T.S. A common role for Atg16L1, Atg5 and Atg7 in small intestinal Paneth cells and Crohn disease. Autophagy 5, 250–252 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  40. 40

    Cadwell, K. et al. Virus-plus-susceptibility gene interaction determines Crohn's disease gene Atg16L1 phenotypes in intestine. Cell 141, 1135–1145 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  41. 41

    Alegre-Abarrategui, J. et al. LRRK2 regulates autophagic activity and localizes to specific membrane microdomains in a novel human genomic reporter cellular model. Hum. Mol. Genet. 18, 4022–4034 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  42. 42

    Travassos, L.H. et al. Nod1 and Nod2 direct autophagy by recruiting ATG16L1 to the plasma membrane at the site of bacterial entry. Nat. Immunol. 11, 55–62 (2010).

    CAS  Google Scholar 

  43. 43

    Chauhan, S., Mandell, M.A. & Deretic, V. IRGM governs the core autophagy machinery to conduct antimicrobial defense. Mol. Cell 58, 507–521 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  44. 44

    Cooney, R. et al. NOD2 stimulation induces autophagy in dendritic cells influencing bacterial handling and antigen presentation. Nat. Med. 16, 90–97 (2010).

    CAS  Google Scholar 

  45. 45

    MacLeod, D.A. et al. RAB7L1 interacts with LRRK2 to modify intraneuronal protein sorting and Parkinson's disease risk. Neuron 77, 425–439 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  46. 46

    Yun, H.J. et al. An early endosome regulator, Rab5b, is an LRRK2 kinase substrate. J. Biochem. 157, 485–495 (2015).

    CAS  Google Scholar 

  47. 47

    Waschbüsch, D. et al. LRRK2 transport is regulated by its novel interacting partner Rab32. PLoS ONE 9, e111632 (2014).

    PubMed  PubMed Central  Google Scholar 

  48. 48

    Zong, J. et al. NOD2 deletion promotes cardiac hypertrophy and fibrosis induced by pressure overload. Lab. Invest. 93, 1128–1136 (2013).

    CAS  Google Scholar 

  49. 49

    Ayabe, T. et al. Activation of Paneth cell α-defensins in mouse small intestine. J. Biol. Chem. 277, 5219–5228 (2002).

    CAS  Google Scholar 

  50. 50

    Miyoshi, H. & Stappenbeck, T.S. In vitro expansion and genetic modification of gastrointestinal stem cells in spheroid culture. Nat. Protoc. 8, 2471–2482 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

The authors thank M. Cookson (National Institutes of Health, Bethesda, Maryland, USA) for the LRRK2 expression vector, T. Kufer (Universität Hohenheim, Stuttgart, Germany) for the Nod2 expression vector, H. Tang (Chinese Academy of Sciences, Beijing, China) for L. monocytogenes strain 10403s, and Z. Chang (Tsinghua University, Beijing, China) for yeast two-hybrid screening reagents. The authors thank H. Zhang (Institute of Biophysics, Chinese Academy of Sciences, Beijing, China) for helpful discussions. This work was supported by the National Natural Science Foundation of China (31271521, 31422019, 81370906), the National Basic Research Program of China (2013CB531405, 2013CB531406), the Thousand Young Talents Program of China, and the Chinese Academy of Sciences–Novo Nordisk Foundation (NNCASGWP-2012-2). Z.Q. was supported by National Science Foundation of China (81101923) and the Beijing Natural Science Foundation (13G20203).

Author information

Affiliations

Authors

Contributions

Q.Z. and Y.P. were responsible for the execution of experiments and data analysis. Y.P. and R.Y. were responsible for organoid culture. H. Wang and X.Z. were responsible for imaging analysis. B.Z., W.L. and H. Wei were responsible for the germ-free facility. Z.L. and Q.Z. wrote the manuscript with input from other authors. Z.L. supervised the study.

Corresponding authors

Correspondence to Hong Wei or Zhihua Liu.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 LRRK2 deficiency leads to enhanced susceptibility to intestinal infection and altered microbiota.

(a) Validation of LRRK2 immunostaining. Confocal images of Lrrk2−/− ileal sections immunostained with an anti-LRRK2 (clone c41-2). Multiple images are aligned to show a complete villus. Horizontal gray lines indicate where the images are aligned. Dashed yellow lines show the base of the crypts. Scale bar, 20 μm. (b) Flow cytometry analysis of CD24 and lysozyme in isolated cells from mouse ileal crypts. (c) Immunoblotting analysis of indicated proteins in isolated cell populations. Actin was used as a loading control. (d) Bacterial numbers (CFU) in spleen and liver from wild-type (WT) (n = 5) and Lrrk2−/− (n = 5) mice 24 h (left) and 48 h (right) after they were infected with 104 CFU of L. monocytogenes by tail vein injection. NS, P > 0.05, Mann-Whitney test. (e,f) The relative abundance of dominant phyla (e) and orders/families (f) identified from pyrosequencing data. Families with an average abundance greater than 1% are shown in the graph. The data regarding all the families are presented in Supplementary Table 1. (g) Principal-component analysis plot showing the clustering pattern between wild-type (n = 8) and Lrrk2−/− (n = 8) mice. Blue oval, wild-type mice; red oval, Lrrk2−/− mice. Distances were calculated using OTU abundance data. Mice of different genotypes were housed in separate cages. Numbers 1–5 of one genotype shared a cage. Numbers 6–8 of one genotype were from three different cages. Each symbol (d,g) represents an individual animal; small horizontal lines indicate the median values. Data are representative of three (ad) independent experiments.

Supplementary Figure 2 LRRK2 deficiency does not affect the expression or secretion of AMPs other than lysozyme in Paneth cells.

(a) PAS staining of crypts from wild-type and Lrrk2−/− mice. Scale bars, 50 μm. (b) Electron microscopy images of crypts from wild-type and Lrrk2−/− mice. Arrowheads indicate DCVs with enlarged halo regions. Scale bars, 2 μm. (c) Quantification of DCVs with enlarged halo regions in wild-type and Lrrk2−/− mice as shown in b. A halo region thickness > 0.2 μm was defined as enlarged. A total of approximately 500 DCVs from 2 mice of each genotype were used for quantification. Data are expressed as mean and s.e.m. *P < 0.01, Student’s t-test. (d) Immunoblotting analysis of Reg3γ and procryptdin in isolated crypts from indicated animals. Actin was used as a loading control. Quantitation of protein levels by densitometry from three independent experiments, as shown on the left. NS, P > 0.05, Student’s t-test. (e) Whole-mount images taken immediately above the ileal mucosal surface in wild-type and Lrrk2−/− mice stained with Helix pomatia lectin (Lectin-HPA) and anti-procryptdin. Scale bars, 20 μm. (f) Schematic diagram of the treatment regimen. Four groups of mice were treated with recombinant lysozyme or PBS (vehicle) by oral gavage once a day for 7 d before being infected with 109 CFU of L. monocytogenes. Fecal samples were collected 10 h after infection. Spleen and liver tissues were harvested 72 h after infection. Data are representative of three (a,b,d,e; mean and s.e.m. in d) independent experiments.

Supplementary Figure 3 Deficiency of LRRK2 or Rab2a leads to lysosomal degradation of lysozyme in Paneth cells in cultured organoids.

(a) Confocal images of lysozyme and procryptdin in cultured wild-type organoids. The boxed area is shown at higher magnification in the panels below. Scale bar, 10 μm. (b) Confocal images of lysozyme and procryptdin in cultured Lrrk2−/− organoids mock treated or treated with 1 μg/ml Brefeldin A for 24 h. Boxed areas are shown at higher magnification in the panels below. Scale bar, 10 μm. (c) Immunoblotting (upper panels) of indicated proteins and confocal images (lower panels) of lysozyme and procryptdin in cultured Lrrk2−/− crypt organoids transfected with siRNA specific for Rab27a or control. (d) Immunoblotting analyses of LRRK2 and Rab2a in cultured crypt organoids transiently transfected with siRNA specific for Rab2a or control. Actin was used as the loading control. (e) Confocal images of lysozyme and procryptdin in cultured wild-type organoids transfected with siRNA oligos against Rab2a and then treated with 100 μM leupeptin or mock treated for 24 h. Boxed areas are shown at higher magnification in the panels below. Scale bars, 10 μm. Data are representative of three (ae) independent experiments.

Supplementary Figure 4 Lysozyme in Paneth cells is degraded by lysosomes in GF mice.

(a) Quantitative RT-PCR analysis of mRNA encoding lysozyme (Lyz) among mRNA in isolated crypts from SPF and GF mice. mRNA results were calculated as described in Figure 3c. Data are expressed as the average for three individual mice ± s.e.m. NS, P > 0.05, Student’s t-test. (b) Immunohistochemical staining of lysozyme in ileal sections from SPF, GF and ex-GF mice. Scale bars, 10 μm. (c) RGB analysis of images from Figure 5c. RGB values were determined along the white dotted lines (from basal to apical) indicated in the merged images with the RGB tool in Image J. (d) Immunoblotting of lysozyme in isolated SPF and GF crypts. Actin was used as the loading control. (e) Confocal images of lysozyme and procryptdin in cultured GF organoids treated with 100 μM leupeptin or mock treated for 24 h. Boxed areas are shown at higher magnification in the panels below. Scale bars, 10 μm. (f) Immunoblotting of lysozyme in SPF and GF organoids treated as in e. Actin was used as the loading control. Data are representative of three (a,b,df) independent experiments.

Supplementary Figure 5 The presence of commensal bacteria is required for DCV localization of LRRK2 and Rab2a.

(a,b) RGB analysis of images from Figure 5e,f. RGB values were determined along the white dotted lines (from basal to apical) indicated in the merged images with the RGB tool in Image J. (c) Quantitative RT-PCR analysis of mRNA encoding LRRK2 and Rab2a among all mRNA in isolated crypts from SPF and GF mice. mRNA results were calculated as described in Figure 3c. Data are expressed as the average for three individual mice + s.e.m. *P < 0.05; NS, P > 0.05, Student’s t-test. (d) Immunoblot of LRRK2 and Rab2a in lysates prepared from isolated crypts from SPF and GF mice. Actin was used as the loading control. Data are representative of three (c,d) independent experiments.

Supplementary Figure 6 Nod2 deficiency leads to the lysosomal degradation of lysozyme in Paneth cells.

(a,b) Immunoblotting of Nod2 (a) and immunofluorescence imaging (b) of lysozyme in crypt organoids transfected with siRNA specific for Nod2 or control siRNA. Boxed areas are shown at higher magnification in the panels below. Red dashed lines outline Paneth cells. Scale bars, 10 μm. (c) Comparable lysozyme staining by colorimetric IHC in wild-type and Nod2−/− crypts. Scale bars, 20 μm. (d) Bacterial numbers in feces 10 h after infection and in livers and spleens 72 h after infection in wild-type (n = 10) and Nod2−/− (n = 10) mice infected with 109 CFU of L. monocytogenes by oral gavage. Wild-type and Nod2−/− mice were cohoused for 2 weeks prior to infection. Each symbol represents an individual mouse; small horizontal lines indicate the median values. *P < 0.01, Mann-Whitney test. (e) Confocal images of lysozyme and procryptdin in cultured Nod2−/− organoids treated with 100 μM leupeptin for 24 h or mock treated. Boxed areas are shown at higher magnification in the lower panels. Scale bars, 10 μm. (f) Immunoblot of lysozyme in Nod2−/− organoids treated as in e. Actin was used as the loading control. (g) Confocal images of SPF wild-type and Nod2−/− crypts immunostained with anti-Nod2. (h) Confocal images of Nod2 and lysozyme in SPF wild-type and Nod2−/− crypts. Nuclei are shown in blue. Scale bars, 10 μm. (i) Immunoblot of Nod2 in isolated CD24+ Paneth cells. Actin was used as a loading control. Data are representative of two (a,b,d) or three (c,ei) independent experiments.

Supplementary Figure 7 LPS treatment restores lysozyme protein expression in Paneth cells in GF mice.

(a) Confocal images of lysozyme and procryptdin in cultured crypt organoids from GF mice treated with LPS (2 μg/ml), iE-DAP (10 μg/ml), flagellin (10 μg/ml), Pam3CSK4 (2 μg/ml) or dA:dT (10 μg/ml) or mock-treated for 24 h. Boxed areas are shown at higher magnification in the panels below. Scale bars, 10 μm. (b) Confocal images of lysozyme and procryptdin in crypts from GF mice receiving LPS at a dose of 500 μg per mouse or PBS (mock) by oral gavage. Scale bars, 10 μm. (c) Percentage of Lyz+ DCVs as shown in b. DCVs from 30 Paneth cells per mouse were quantified for two mice for each treatment condition. Data are expressed as the average + s.e.m. *P < 0.01, Student’s t-test. (d) Confocal images of lysozyme and procryptdin in ileal crypts from wild-type and Tlr4−/− mice. Blue, counterstained nuclei. Scale bars, 10 μm. (eg) Confocal images of indicated proteins in ileal crypts from mock- or LPS-treated GF mice as treated in b. Blue, counterstained nuclei. Scale bars, 10 μm. Data are representatives of three (a) or two (b,dg) independent experiments.

Supplementary Figure 8 A schematic working model for bacterially directed lysozyme sorting by sequential recruitment of Nod2, LRRK2 and Rab2a to the DCV surface.

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–8 and Supplementary Tables 1 and 2 (PDF 1620 kb)

Source data

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zhang, Q., Pan, Y., Yan, R. et al. Commensal bacteria direct selective cargo sorting to promote symbiosis. Nat Immunol 16, 918–926 (2015). https://doi.org/10.1038/ni.3233

Download citation

Further reading

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