Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Letter
  • Published:

High-avidity IgA protects the intestine by enchaining growing bacteria


Vaccine-induced high-avidity IgA can protect against bacterial enteropathogens by directly neutralizing virulence factors or by poorly defined mechanisms that physically impede bacterial interactions with the gut tissues (‘immune exclusion’)1,2,3. IgA-mediated cross-linking clumps bacteria in the gut lumen and is critical for protection against infection by non-typhoidal Salmonella enterica subspecies enterica serovar Typhimurium (S. Typhimurium). However, classical agglutination, which was thought to drive this process, is efficient only at high pathogen densities (≥108 non-motile bacteria per gram). In typical infections, much lower densities4,5 (100–107 colony-forming units per gram) of rapidly dividing bacteria are present in the gut lumen. Here we show that a different physical process drives formation of clumps in vivo: IgA-mediated cross-linking enchains daughter cells, preventing their separation after division, and clumping is therefore dependent on growth. Enchained growth is effective at all realistic pathogen densities, and accelerates pathogen clearance from the gut lumen. Furthermore, IgA enchains plasmid-donor and -recipient clones into separate clumps, impeding conjugative plasmid transfer in vivo. Enchained growth is therefore a mechanism by which IgA can disarm and clear potentially invasive species from the intestinal lumen without requiring high pathogen densities, inflammation or bacterial killing. Furthermore, our results reveal an untapped potential for oral vaccines in combating the spread of antimicrobial resistance.

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

Access options

Buy this article

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

Figure 1: Enchained growth is a protective mechanism in non-typhoidal salmonellosis.
Figure 2: Enchained growth drives clonal extinction and clearance from the caecal lumen.
Figure 3: Enchained growth inhibits conjugative plasmid transfer.
Figure 4: Enchained growth functions effectively at realistic, low pathogen densities, and can be observed for other bacterial species.

Similar content being viewed by others


  1. Mantis, N. J., Rol, N. & Corthésy, B. Secretory IgA’s complex roles in immunity and mucosal homeostasis in the gut. Mucosal Immunol. 4, 603–611 (2011)

    Article  CAS  Google Scholar 

  2. Pabst, O. New concepts in the generation and functions of IgA. Nat. Rev. Immunol. 12, 821–832 (2012)

    Article  CAS  Google Scholar 

  3. Slack, E., Balmer, M. L. & Macpherson, A. J. B cells as a critical node in the microbiota–host immune system network. Immunol. Rev. 260, 50–66 (2014)

    Article  CAS  Google Scholar 

  4. WHO. Risk assessments of Salmonella in eggs and broiler chickens. (2002)

  5. Gopinath, S., Carden, S. & Monack, D. Shedding light on Salmonella carriers. Trends Microbiol. 20, 320–327 (2012)

    Article  CAS  Google Scholar 

  6. White, P. B. Further Studies of the Salmonella Group. Great Britain Medical Research Council (Her Majesty’s Stationary Office) 103, 3–160 (1926)

    Google Scholar 

  7. Le Minor, L. & Popoff, M. Y. Kauffmann–White-Scheme (Behring Diagnostika, 1988)

  8. Endt, K. et al. The microbiota mediates pathogen clearance from the gut lumen after non-typhoidal Salmonella diarrhea. PLoS Pathog. 6, e1001097 (2010)

    Article  Google Scholar 

  9. Hendrickx, A. P. et al. Antibiotic-driven dysbiosis mediates intraluminal agglutination and alternative segregation of Enterococcus faecium from the intestinal epithelium. MBio 6, e01346e15 (2015)

    Article  Google Scholar 

  10. Levinson, K. J., De Jesus, M. & Mantis, N. J. Rapid effects of a protective O-polysaccharide-specific monoclonal IgA on Vibrio cholerae agglutination, motility, and surface morphology. Infect. Immun. 83, 1674–1683 (2015)

    Article  CAS  Google Scholar 

  11. Kiørboe, T. A mechanistic approach to plankton ecology. (Princeton Univ. Press, 2008)

  12. van Schothorst, M. & Beckers, H. J. Persistent excretion of salmonellas. BMJ 2, 1301 (1978)

    Article  CAS  Google Scholar 

  13. Forbes, S. J., Eschmann, M. & Mantis, N. J. Inhibition of Salmonella enterica serovar Typhimurium motility and entry into epithelial cells by a protective antilipopolysaccharide monoclonal immunoglobulin A antibody. Infect. Immun. 76, 4137–4144 (2008)

    Article  CAS  Google Scholar 

  14. Forbes, S. J. et al. Association of a protective monoclonal IgA with the O antigen of Salmonella enterica serovar Typhimurium impacts type 3 secretion and outer membrane integrity. Infect. Immun. 80, 2454–2463 (2012)

    Article  CAS  Google Scholar 

  15. Amarasinghe, J. J., D’Hondt, R. E., Waters, C. M. & Mantis, N. J. Exposure of Salmonella enterica serovar Typhimurium to a protective monoclonal IgA triggers exopolysaccharide production via a diguanylate cyclase-dependent pathway. Infect. Immun. 81, 653–664 (2013)

    Article  CAS  Google Scholar 

  16. Michetti, P ., Mahan, M. J ., Slauch, J. M ., Mekalanos, J. J. & Neutra, M. R. Monoclonal secretory immunoglobulin A protects mice against oral challenge with the invasive pathogen Salmonella typhimurium. Infect. Immun. 60, 1786–1792 (1992)

    CAS  PubMed  PubMed Central  Google Scholar 

  17. Michetti, P. et al. Monoclonal immunoglobulin A prevents adherence and invasion of polarized epithelial cell monolayers by Salmonella typhimurium. Gastroenterology 107, 915–923 (1994)

    Article  CAS  Google Scholar 

  18. Barthel, M. et al. Pretreatment of mice with streptomycin provides a Salmonella enterica serovar Typhimurium colitis model that allows analysis of both pathogen and host. Infect. Immun. 71, 2839–2858 (2003)

    Article  CAS  Google Scholar 

  19. Kaiser, P., Diard, M., Stecher, B. & Hardt, W. D. The streptomycin mouse model for Salmonella diarrhea: functional analysis of the microbiota, the pathogen’s virulence factors, and the host’s mucosal immune response. Immunol. Rev. 245, 56–83 (2012)

    Article  CAS  Google Scholar 

  20. Hapfelmeier, S. et al. The Salmonella pathogenicity island (SPI)-2 and SPI-1 type III secretion systems allow Salmonella serovar Typhimurium to trigger colitis via MyD88-dependent and MyD88-independent mechanisms. J. Immunol. 174, 1675–1685 (2005)

    Article  CAS  Google Scholar 

  21. Moor, K. et al. Peracetic acid treatment generates potent inactivated oral vaccines from a broad range of culturable bacterial species. Front. Immunol. 7, 34 (2016)

    Article  Google Scholar 

  22. Kaiser, P., Slack, E., Grant, A. J., Hardt, W. D. & Regoes, R. R. Lymph node colonization dynamics after oral Salmonella typhimurium infection in mice. PLoS Pathog. 9, e1003532 (2013)

    Article  CAS  Google Scholar 

  23. Perfeito, L., Fernandes, L., Mota, C. & Gordo, I. Adaptive mutations in bacteria: high rate and small effects. Science 317, 813–815 (2007)

    Article  ADS  CAS  Google Scholar 

  24. Barlow, M. Methods in Molecular Biology. Horizontal Gene Transfer 532, 397–411 (2009)

    Article  CAS  Google Scholar 

  25. Grant, A. J. et al. Modelling within-host spatiotemporal dynamics of invasive bacterial disease. PLoS Biol. 6, e74 (2008)

    Article  Google Scholar 

  26. Maier, L. et al. Granulocytes impose a tight bottleneck upon the gut luminal pathogen population during Salmonella typhimurium colitis. PLoS Pathog. 10, e1004557 (2014)

    Article  Google Scholar 

  27. Stecher, B. et al. Salmonella enterica serovar typhimurium exploits inflammation to compete with the intestinal microbiota. PLoS Biol. 5, 2177–2189 (2007)

    Article  CAS  Google Scholar 

  28. Stecher, B. et al. Gut inflammation can boost horizontal gene transfer between pathogenic and commensal Enterobacteriaceae. Proc. Natl Acad. Sci. USA 109, 1269–1274 (2012)

    Article  ADS  CAS  Google Scholar 

  29. Stecher, B. et al. Like will to like: abundances of closely related species can predict susceptibility to intestinal colonization by pathogenic and commensal bacteria. PLoS Pathog. 6, e1000711 (2010)

    Article  Google Scholar 

  30. Palm, N. W. et al. Immunoglobulin A coating identifies colitogenic bacteria in inflammatory bowel disease. Cell 158, 1000–1010 (2014)

    Article  CAS  Google Scholar 

  31. Müller, A. J. et al. Salmonella gut invasion involves TTSS-2-dependent epithelial traversal, basolateral exit, and uptake by epithelium-sampling lamina propria phagocytes. Cell Host Microbe 11, 19–32 (2012)

    Article  Google Scholar 

  32. Harriman, G. R. et al. Targeted deletion of the IgA constant region in mice leads to IgA deficiency with alterations in expression of other Ig isotypes. J. Immunol. 162, 2521–2529 (1999)

    CAS  PubMed  Google Scholar 

  33. Uren, T. K. et al. Role of the polymeric Ig receptor in mucosal B cell homeostasis. J. Immunol. 170, 2531–2539 (2003)

    Article  CAS  Google Scholar 

  34. Slack, E. et al. Innate and adaptive immunity cooperate flexibly to maintain host–microbiota mutualism. Science 325, 617–620 (2009)

    Article  ADS  CAS  Google Scholar 

  35. Diard, M. et al. Antibiotic treatment selects for cooperative virulence of Salmonella typhimurium. Curr. Biol. 24, 2000–2005 (2014)

    Article  CAS  Google Scholar 

  36. Hoiseth, S. K. & Stocker, B. A. Aromatic-dependent Salmonella typhimurium are non-virulent and effective as live vaccines. Nature 291, 238–239 (1981)

    Article  ADS  CAS  Google Scholar 

  37. Maier, L. et al. Microbiota-derived hydrogen fuels Salmonella typhimurium invasion of the gut ecosystem. Cell Host Microbe 14, 641–651 (2013)

    Article  CAS  Google Scholar 

  38. Stecher, B. et al. Flagella and chemotaxis are required for efficient induction of Salmonella enterica serovar Typhimurium colitis in streptomycin-pretreated mice. Infect. Immun. 72, 4138–4150 (2004)

    Article  CAS  Google Scholar 

  39. Ilg, K. et al. O-antigen-negative Salmonella enterica serovar Typhimurium is attenuated in intestinal colonization but elicits colitis in streptomycin-treated mice. Infect. Immun. 77, 2568–2575 (2009)

    Article  CAS  Google Scholar 

  40. Diard, M. et al. Inflammation boosts bacteriophage transfer between Salmonella spp. Science 355, 1211–1215 (2017)

    Article  ADS  CAS  Google Scholar 

  41. Nutter, R. L., Bullas, L. R. & Schultz, R. L. Some properties of five new Salmonella bacteriophages. J. Virol. 5, 754–764 (1970)

    CAS  PubMed  PubMed Central  Google Scholar 

  42. Mobley, H. L. et al. Pyelonephritogenic Escherichia coli and killing of cultured human renal proximal tubular epithelial cells: role of hemolysin in some strains. Infect. Immun. 58, 1281–1289 (1990)

    CAS  PubMed  PubMed Central  Google Scholar 

  43. Drecktrah, D. et al. Dynamic behavior of Salmonella-induced membrane tubules in epithelial cells. Traffic 9, 2117–2129 (2008)

    Article  CAS  Google Scholar 

  44. Gil, D. & Bouché, J. P. ColE1-type vectors with fully repressible replication. Gene 105, 17–22 (1991)

    Article  CAS  Google Scholar 

  45. Moor, K. et al. Analysis of bacterial-surface-specific antibodies in body fluids using bacterial flow cytometry. Nat. Protocols 11, 1531–1553 (2016)

    Article  Google Scholar 

  46. Traggiai, E. et al. An efficient method to make human monoclonal antibodies from memory B cells: potent neutralization of SARS coronavirus. Nat. Med. 10, 871–875 (2004)

    Article  CAS  Google Scholar 

  47. Helaine, S. et al. Dynamics of intracellular bacterial replication at the single cell level. Proc. Natl Acad. Sci. USA 107, 3746–3751 (2010)

    Article  ADS  CAS  Google Scholar 

  48. Sellin, M. E. et al. Epithelium-intrinsic NAIP/NLRC4 inflammasome drives infected enterocyte expulsion to restrict Salmonella replication in the intestinal mucosa. Cell Host Microbe 16, 237–248 (2014)

    Article  CAS  Google Scholar 

  49. Sommer, C., Strähle, C., Köthe, U. & Hamprecht, F. A. ilastik: Interactive Learning and Segmentation Toolkit. Proc. Eighth IEEE International Symposium on Biomedical Imaging (ISBI). 2011, 230–233 (2011)

  50. Kaiser, P. et al. Cecum lymph node dendritic cells harbor slow-growing bacteria phenotypically tolerant to antibiotic treatment. PLoS Biol. 12, e1001793 (2014)

    Article  Google Scholar 

  51. Berg, H. C. Random Walks in Biology (Princeton Univ. Press, 1993)

  52. Jackson, G. A. A model of the formation of marine algal flocs by physical coagulation processes. Deep-Sea Res. A, Oceanogr. Res. Pap. 37, 1197–1211 (1990)

    Article  ADS  CAS  Google Scholar 

  53. Kiørboe, T., Andersen, K. P. & Dam, H. G. Coagulation efficiency and aggregate formation in marine phytoplankton. Mar. Biol. 107, 235–245 (1990)

    Article  Google Scholar 

  54. Lentle, R. G., Hemar, Y., Hall, C. E. & Stafford, K. J. Periodic fluid extrusion and models of digesta mixing in the intestine of a herbivore, the common brushtail possum (Trichosurus vulpecula). J. Comp. Physiol. B 175, 337–347 (2005)

    Article  Google Scholar 

Download references


The authors would like to acknowledge the support of ScopeM (ETHZ), and the staff at the RCHCI and EPIC animal facilities. They would also like to thank members of the Hardt and Hapfelmeier groups, as well as A. J. Macpherson, S. Sunagawa, M. F. Freeman, C. Mueller, B. Stecher, K. Endt, M. Stecher, M. Bajagic and M. Ackermann for technical help and/or their comments on the manuscript. E.S. is supported by the Swiss National Science Foundation (SNF, Marie Heim-Vöglin award PMPDP3_158364 and Ambizione award PZ00P3_136742 to E.S.). W.-D.H. is supported by the SNF (310030_53074; Sinergia CRSII_154414/1), ETH Zurich (ETH-33 12-2) and the Novartis Freenovation Programme. D.R.B. acknowledges support from a Human Frontier Science Program Cross-Disciplinary Fellowship. E.B. is supported by the Excellence Scholarship and Opportunity Programme (ETH). R.S. acknowledges support from a Gordon and Betty Moore Foundation Marine Microbial Initiative Award (GBMF 3783). R.R.R. acknowledges support from the Swiss National Science Foundation (31003A_149769). T.V. is supported by a Deutsche Forschungsgemeinschaft postdoctoral fellowship (grant VO 2273/1-1) and M.E.S. by the Swedish Research Council (grants 2012-262 and 2015-00635).

Author information

Authors and Affiliations



K.M., M.D., M.E.S., E.B., B.F., S.Y.W., A.T., W.-D.H and E.S. designed, performed and analysed experiments. M.A., A.D.C. and T.V. carried out image analysis. A.M., B.F.-R., G.A., S.B., L.P., C.C., D.C. and A.L. generated human monoclonal and mouse recombinant antibodies. F.B., C.L., and R.R.R. devised models and mathematically analysed barcoded-strain experiments. R.S. and D.R.B. devised and developed the model for planktonic bacteria population dynamics and wrote the extended discussion in the Supplementary Information. The paper was written by E.S. with support from W.-D.H. All authors discussed the results and commented on the manuscript.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Additional information

Reviewer Information Nature thanks A. Baumler, A. Camilli, S. Fagarasan and A. Smith for their contribution to the peer review of this work.

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

Extended data figures and tables

Extended Data Figure 1 S. Typhimurium growth in the streptomycin-treated mouse large intestine.

Naive C57BL/6 SOPF mice (n = 3 per group, representative of 3 independent experiments) were gavaged with 25 mg streptomycin and 24 h later with the indicated inoculum of S.Tmatt carrying the conditionally replicating plasmid pAM34. a, CFU g−1 faeces with fitted four-parameter logistic (sigmoidal) curves by least-squares nonlinear regression. b, Percentage of the faecal population retaining ampicillin resistance at the indicated time points. Exponential decay curves were fitted by least-squares nonlinear regression. c, Calculation of a standard curve linking percentage plasmid carriage with generation number derived from serial dilution of the inoculum and overnight cultures in vitro (least squares linear regression of the log2 percentage of plasmid carriage and calculated generation number, assuming zero death). Note that the gradient is shallower than the expected value of 1. Low levels of residual IPTG-independent plasmid replication probably explain this. The y intercept is considerably greater than log2(100) = 6.64, owing to the relatively high starting copy number (approximately n = 16) of the plasmid. d, Calculated numbers of generations in the caecum based on the data shown in b and c. Regardless of the starting inoculum, S. Typhimurium initially expands with a growth rate of approximately 2 h−1 until the caecal ‘carrying capacity’ is reached (linear regression of all values generated from a faecal S. Typhimurium density of <109 CFU g−1). e, Illustration of plasmid loss (assuming perfect segregation; the analyses in c and d did not rely on this assumption) with a starting copy number of n = 2 (red circles, pAM34 plasmid).

Extended Data Figure 2 LPS-specific secretory IgA elicited by PA-S.Tm vaccination is necessary to prevent mLN translocation, coat S. Typhimurium in the caecum, and drive clumping.

a, Wild-type and Igh-j−/− C57BL/6 mice were vaccinated with mock vaccine (PBS), an inactivated S. Choleraesuis (O-antigen mismatched) vaccine (PAS.Chl) or PA-S.Tm, as indicated. On day 21 after the first vaccination, mice were pretreated with streptomycin and challenged with 105 S.TmWT. mLN CFU were determined 18 h later. Kruskal–Wallis with Dunn’s post hoc tests shown. b, C57BL/6 mice vaccinated with PA-S.Tm or PBS were streptomycin pretreated. 23.5 h later, animals were pretreated with 1011 particles of the indicated vaccine to specifically or non-specifically out-titrate IgA in the gut lumen. 30 min later, all animals were challenged with 105 S.TmWT with (2×) or without (1×) a second 1011-dose of the indicated vaccine preparation. mLN loads were determined at 24 h after infection. Dashed lines indicate detection limits. Two-way ANOVA on log-normalized data, with Bonferroni post hoc tests. c, Wild-type or Igha−/− C57BL/6 mice were vaccinated as indicated and infected as in a. At 18 h after infection, caecal contents were stained with antibodies against IgA and against the S. Typhimurium O-antigen, and the staining intensities were analysed by flow cytometry. Representative zebra-plots (5% contours) showing S. Typhimurium O-antigen (human IgG anti-O12, Alexa647-anti-hIgG) and IgA staining (biotin anti-mouse IgA with Pacific-Blue–streptavidin) of caecal content bacteria. Non-shaded quadrant contains IgA-coated S. Typhimurium. Numbers denote the percentages of S. Typhimurium falling into each quadrant. d, e, Quantitative microscopy on intestinal content in fixed-frozen caecum sections from vaccinated wild-type (d), IgA-deficient (Igha−/−), poly Ig receptor-deficient (Pigr) and fully antibody-deficient (Igh-j−/−) C57BL/6 mice (e) vaccinated as indicated and infected as in (a). Kruskal–Wallis with Dunn’s post hoc tests shown. Total luminal S. Typhimurium loads were not different between groups (data not shown). As vaccination with PA-S.Chl yielded identical results to mock-vaccination or fully antibody-deficient vaccinated mice, PBS-vaccinated controls were subsequently used unless otherwise stated.

Extended Data Figure 3 Owing to bead-beating, CFU can be determined correctly, even when secretory IgA clumps S. Typhimurium in the gut lumen.

(See Fig. 1h for example.) Mock- (PBS) or PA-S.Tm- (vacc.) vaccinated mice were streptomycin pretreated and challenged with 105 CFU of S.Tmatt (pM965) constitutively expressing GFP. Caecal content was collected 24 h later and imaged directly after diluting 1 in 10 in PBS, or after bead-beating for 2 min at 25 Hz with a large sterile steel bead. Scale bar, 10 μm; representative images shown. Identical observations were made with S.TmWT (data not shown). Note, this verified that bead-beating (part of our routine analysis of caecum lumen, faecal and organ CFU analysis) efficiently breaks IgA-mediated clumps. Thus, plating is suitable for total CFU analysis in intestinal content of both naive and vaccinated mice.

Extended Data Figure 4 High-avidity IgA protects independently of blocking flagella-driven motility and type III secretion systems 1 and 2.

a, Representative 1-s exposure confocal microscopy images of GFP-expressing S.Tmatt in live explanted caecal content 18 h after infection with 105 CFU. Tracks in the mock-vaccinated case indicate bacterial swimming during acquisition of the image. b, Microscopy quantification of S.Tmatt swimming speed. c, Naive mice (n = 5 per group) were pretreated with streptomycin and infected with 105 CFU of the indicated S. Typhimurium strain. mLN CFU at 24 h after infection; dashed line, lower detection limit. Kruskal–Wallis with Dunn’s post hoc tests. d, e, Mock- (PBS) or PA-S.Tm- (vacc.) vaccinated mice were challenged with 105 CFU S.TmΔfliGHI strain by gavage. We analysed mLN CFU at 18 h after infection (d); faecal lipocalin-2 at 18 h after infection (e). Two-tailed Mann–Whitney U-tests. f, Competitive infection experiments assessing the role of flagella in a S.Tmatt strain background. mLN CFU at day 3 after challenging vaccinated or mock-vaccinated mice (n = 5 per group) with a 1:1 mixture of 105 CFU S.TmattΔfliGHI and S.Tmatt. Two-way ANOVA with Bonferroni post hoc tests. g, Mock- or PA-S.Tm-vaccinated VilRPF mice were streptomycin pretreated and orally infected with 105 CFU of S.Tmatt (pM965) expressing GFP constitutively. At 18 h after infection, mice were anaesthetized and artificially respirated. Caecal crypts were imaged by two-photon intravital confocal microscopy31 (scale bar, 50 μm) and we quantified the number of S. Typhimurium per crypt. Very high numbers of S. Typhimurium were designated a maximum detection-limit of 11. n = number of crypts analysed per condition. Two-tailed Mann–Whitney U-tests. Clumps were located too deeply in the caecal content to be visualized by intravital microscopy across the caecal wall. h, VilRPF mice were pretreated with streptomycin and, 24 h later, prepared for intravital confocal microscopy by injection of around 25-μm YG-fluorescent latex beads (green) directly into the caecal lumen, as an ultrabright proxy for bacterial aggregates. These beads were selected as having a similar diameter to an average S. Typhimurium clump in the gut lumen of vaccinated mice. Representative images depicting the bead location relative to the surface of the intestinal epithelium (white dashed line), and quantification (right). Scale bar, 50 μm. Note, the data shown in g and h suggest that secretory IgA-driven S. Typhimurium clusters are confined ≥100 μm away from the gut epithelium in PA-S.Tm-vaccinated mice, as cluster-sized beads are typically observed with at least this separation from the caecal epithelium.

Extended Data Figure 5 Clumping-mediated depletion of planktonic, infectious S.TmWT cells could fully account for the reduced mLN colonization in PA-S.Tm-vaccinated mice.

We fitted a stochastic birth–death model modified by immigration, describing mLN colonization22,50 in vaccinated mice to infection data with mixtures of barcoded S.TmWT strains25. a, Scheme and statement of the deterministic model22,50. μ, translocation rate from the gut lumen to the mLN per S. Typhimurium in the gut lumen (population size, G); r − c (replication minus clearance = net replication rate of each S. Typhimurium in the mLN (population size, L)). be, Mock- (PBS/irrelevant vaccine) or PA-S.Tm-vaccinated mice were pretreated with streptomycin and challenged with 105 CFU of S.TmWT, which was spiked with seven genetically barcoded strains (S.TmWT background) at 1:35 of the total population. At 18 h after infection, mLN CFU determination and analysis of barcode abundance using qPCR were used to fit the mathematical model22,50. b, The translocation rate (μG) from the gut lumen to the mLN (Student’s t-test, P < 0.0001). c, The net replication rate (r − c) in the mLN (Student’s t-test, P < 0.0001). d, A comparison of the size of the luminal planktonic population (G) either empirically determined by microscopy counting and plating (symbols, as Extended Data Fig 2d) or predicted by dividing the predicted value of μG by the predicted value μnaive (bootstrapped over 1,000 iterations to generate a median and 95% confidence interval of the mean; see ‘Modelling of translocation to the mLN’ section in Methods). There is no significant difference between the predicted and measured values by two-way ANOVA. e, Translocation rate per planktonic S. Typhimurium (μ) determined by dividing μG values determined by model-fitting by G values determined by microscopy (bootstrapping over 106 randomly sampled values from both normal distributions, used to generate mean (horizontal lines) and 95% confidence interval of the mean; Mann–Whitney U-test, P < 0.0001). Note, this data demonstrates that (1) if we assume that the translocation rate per infectious luminal S. Typhimurium is unchanged in vaccination, our model predicts an identical decrease in the size of the luminal infectious population (that is, the planktonic S. Typhimurium) as that which we can quantify by microscopy. In other words, the extent of clumping is sufficient to predict the observed decrease in mLN colonization when all other aspects of the infection are identical to naive animals. (2) Direct calculation of the translocation rate per infectious bacterium in vaccinated mice even suggests a slight increase in the predicted value of μ, consistent with a role for IgA in active sampling of luminal contents1, but inconsistent with a loss of the function of type III secretion systems 1 and 2 in vivo20. Again, this shows that secretory IgA protects by clumping S. Typhimurium deep in the gut lumen.

Extended Data Figure 6 Enchained growth and classical agglutination can complement each other at high bacterial densities.

a, Two representative plots quantifying the randomness of mCherry+ and GFP+ S. Typhimurium distribution in clusters. Blue histogram: predicted percentage of clustered mCherry+ pairs achieved when each bacterium in the images is randomly re-assigned mCherry or GFP expression (that is, simulated random agglutination) and the actual fraction determined by image analysis (red vertical line). The probability of observing this number of mCherry+ pairs in clusters if only random agglutination contributed was analysed for three mice that received a 105 CFU inoculum at 5 h after infection (P = 0.006, 0.029, 0.048); and three mice that received a 1010 CFU inoculum at 3 h after infection (P = 0.36, 0.64, 0.24). The distribution of fluorescence in clumps (Fig. 1i, j) is therefore significantly non-random after infection with realistic inocula, but not high inocula. bd, PA-S.Tm-vaccinated C57BL/6 mice were pretreated with ampicillin and infected with the indicated inoculum of S.Tmatt constitutively expressing green (GFP; pM965) or red (mCherry; pFPV25.1) fluorescent proteins. bd, Mice were killed at the indicated times after infection and caecal content explants directly imaged by confocal microscopy. Representative images and schematic diagrams depicting the corresponding processes. e, f, Mice infected as indicated were killed at 24 h after infection. Fixed, frozen 25-μm caecum sections were imaged by confocal microscopy.

Extended Data Figure 7 Modelling the effect of enchained growth and classical agglutination on clumping and the density of planktonic S. Typhimurium in the caecal lumen.

a, Equations governing the model, where C is the concentration of planktonic bacteria, t is time in hours and the subscripts CA and EG indicated ‘classical agglutination’ and ‘enchained growth’, respectively. b, Determination of cell radius from confocal fluorescence microscopy images of live explanted caecal content at 3 h post-infection with 1010 CFU (non-growing) or 5 h after infection with 105 CFU (fast-growing) S.Tmatt. c, All parameters used in the model including references to their origin51,52,53,54. d, e, Predicted density of planktonic bacteria over 24 h of exponential growth in the caecal lumen. Note that at late time points (high density of CFU), the planktonic population reaches a true equilibrium. d, Altering σ (efficiency of classical agglutination) for a fixed value of δ. Note that altering the efficiency of classical agglutination alters the size of the equilibrium planktonic population, but not the rate of increase. e, Decreasing the efficiency of enchained growth (that is, increasing δ) increases both the size of the planktonic population and the rate at which the population increases during exponential growth. f, Quantification of S. Typhimurium planktonic and clumped bacteria per high-power field in microscopy of explanted live caecal content from PA-S.Tm-vaccinated mice challenged with S.Tmatt (105 CFU). This supports the model of an equilibrium planktonic population size that remains constant, as the clumped population exponentially expands. g, The predicted percentage of the population that is clumped (solid lines) and planktonic (dashed lines) when the efficiency of enchained growth decreases (that is, increasing δ) at 3 h after infection. h, As in g, but flagella-driven swimming speed is re-introduced for fixed values of δ = 0.8 and σ = 0.2. Note that increasing swimming speed increases the efficiency of classical agglutination, potentially improving protection. i, Linking enchained growth to vaccine-mediated protection. On the basis of the stochastic model presented in Extended Data Fig. 5a, we can calculate the probability of recovering 10 CFU of S. Typhimurium from the mLN at day 1 after infection as a function of the percentage of luminal bacteria that are planktonic. The dashed lines indicate that in the case of a 100-fold reduction in the planktonic population, we expect less than 10 CFU S. Typhimurium in the mLNs in approximately 50% of the animals.

Extended Data Figure 8 Simulating the effect of IgA-driven enchained growth on clonal extinction in the caecal lumen.

In order to visualize how enchained growth leads to clonal extinction without killing, we simulated a simple scenario on the basis of the experiments of Fig. 2b, c (see Supplementary Methods). Approximately 10 CFU of each of seven barcoded clones arrive in the caecum. These bacteria get established in the caecum with probability P = 0.115 (based on the loss of evenness in unvaccinated animals). The remaining bacteria double every 30 min during the first 12 h. Growth then slows for the next 6 h as the carrying capacity is reached (typical of infection kinetics with 105 CFU S. Typhimurium in a streptomycin-pretreated mouse). 10% of the caecal content is cleared to the colon every 15 min to produce faeces. We simulated the extreme cases of no enchained growth (δ = 1, black line), or a ‘perfect’ enchained growth (δ = 0, cyan line). The ‘evenness’26, that is, the similarity of the proportions of each barcode in the population, of 300,000 simulations of this simplified model up to 18 h after infection was computed and averaged. In the naive case (that is, no enchained growth, δ = 1, black line), the evenness stabilizes after 1 h, at a value of the same order of magnitude than the median experimental value for the unvaccinated mice. In the case of perfect enchained growth (δ = 0, cyan line), the evenness continues to drop, indicating on-going clonal extinction. The experimental evenness achieved in vaccinated animals at 18 h after infection (cyan open circles represent individual mice, mean evenness of all six animals is represented as a cyan filled circle, as in Fig. 2c) is located between the simulated final values of evenness with no enchained growth and with perfect enchained growth, qualitatively consistent with the idea that enchained growth in vivo is not 100% efficient (for example, due to breaking or collision of clonal clumps, Extended Data Fig. 7).

Extended Data Figure 9 Direct toxicity cannot explain clonal extinction.

a, b, An overnight culture of S.Tmatt carrying an arabinose-inducible GFP (pBADGFPmut2) was diluted 1 in 100 into LB (1% w/v arabinose) with ampicillin selection for 3 h to fully induce the reporter. Subsequently, the bacteria were thoroughly washed and diluted 1:100 into fresh LB that was free of arabinose and contained secretory IgA purified from the small-intestinal lavages of mock- (PBS) or PA-S.Tm- (vacc.) vaccinated mice, similarly to previously described fluorescence-dilution assays47. Samples were removed and clusters disrupted by bead-beating (Extended Data Fig. 3) for plating and flow cytometry every 30 min for 4 h (n = 2 independent experiments, pooled). a, CFU ml−1, as determined by plating. b, Median GFP fluorescence intensity, as determined by flow cytometry. An additional control for the effect of agglutination on fluorescence intensity was carried out by adding purified IgA from PA-S.Tm vaccinated mice to an IgA-free sample just before flow cytometry staining. P > 0.05 by repeat-measures ANOVA on log-normalized data. This indicates no detectable effect of specific IgA on S. Typhimurium growth or viability in vitro. c, d, S.Tmatt WITS carrying a neutral kanamycin-resistance cassette and S. Enteritidis carrying a growth-neutral chloramphenicol-resistance cassette (M1513) were cultured separately overnight in LB. These cultures were mixed 1:1, then diluted 1:100 into LB without antibiotics, and supplemented with IgA purified from PA-S.Tm-vaccinated (cyan) or PBS-treated mice (black) as above. Every 30 min a sample was removed and clusters disrupted by bead-beating (Extended Data Fig. 3). CFU ml−1 of the IgA-bound strain (S. Typhimurium) (c) and IgA-non-bound strain (S. Enteritidis) (d) were determined by selective plating. P > 0.05 by repeat-measures ANOVA on log-normalized data. One representative experiment of two. The expected patterns of IgA-coating were confirmed at the end of each experiment by bacterial flow cytometry. IgA-mediated effects do not selectively disadvantage S. Typhimurium growth in the presence of rich media. eg, Classical agglutinaton does not drive clonal extinction. Mock- or PA-S.Tm-vaccinated mice were orally infected with an inoculum of 1010 CFU of S.Tmatt, spiked with approximately 10 CFU of each of seven genetically barcoded kanamycin-resistant S.Tmatt strains. e, Scheme depicting the expected effect of agglutination on clonal loss. Fill colours, barcodes; blue outlines, secretory-IgA-coated bacteria. f, The number of CFU of each barcoded strain was determined in the caecal lumen at 18 h after infection by selective plating, enrichment culture and qPCR22,50. g, Evenness26 of the resulting barcoded population in each animal analysed. Mann–Whitney U-test statistics are shown.

Extended Data Figure 10 Raw data for conjugative plasmid transfer in vaccinated mice.

(See Figs 3 and 4.) ac, Small-intestinal lavages from PA-S.Tm-vaccinated or naive mice were titrated against wild-type recipient (14028 S. Typhimurium) (a) and P2cat wild-type donor (SL1344 S. Typhimurium) (b) and compared to a monoclonal dimeric mouse IgA directed against the O12 antigen. a, b, Raw titration curves. c, Absolute titres defined relative to an O12-specific monoclonal dimeric IgA. Two-tailed Mann–Whitney U-test, not significant. df, C57BL/6 SOPF mice were orally vaccinated with PA-S.Tm (vacc.), PA-inactivated S. Choleraesuis or PBS only (irrelevant vaccine/PBS). On day 21 after the first vaccination, all mice were pretreated with ampicillin and mice were sequentially infected with 102 CFU each of P2cat wild-type donor and wild-type recipient S. Typhimurium. d, Loads of kanamycin-resistant wild-type recipient in the faeces. e, Loads of kanamycin–chloramphenicol-double-resistant transconjugants in the faeces. f, Loads of chloramphenicol-resistant P2cat wild-type donor, determined from faecal samples by selective plating (corresponding to data in Fig. 3a). gi, Naive C57BL/6 SOPF were pretreated with ampicillin. 24 h later, mice were sequentially infected with either 102 CFU each of P2cat wild-type donor or P2cat ΔfliGHI donor and wild-type recipient or ΔfliGHI recipient, respectively. g, Loads of kanamycin-resistant plasmid-negative recipient in the faeces. h, Percentage of transconjugants, of the total recipient population in faeces. i, Loads of chloramphenicol-resistant P2cat donors. S.TmWT versus S.TmΔfliGHI P > 0.05 by repeat-measures two-way ANOVA on log-normalized data. jl, Mice with low-complexity microflora were orally vaccinated as in d. On day 21 after the first vaccination, all mice were infected sequentially with 200 CFU each of P2cat attenuated donor and attenuated recipient without antibiotic pretreatment. j, Loads of kanamycin-resistant attenuated recipient in the faeces. k, Loads of kanamycin–chloramphenicol-double-resistant transconjugants in the faeces. l, Loads of chloramphenicol-resistant P2cat attenuated donor, determined from faecal samples by selective plating (corresponding to data in Fig. 4a). dl, Repeat-measures ANOVA on log-normalized data. Bonferroni post hoc tests shown.

Supplementary information

Supplementary Information

This file contains a Supplementary Discussion and Supplementary Tables 1-2. (PDF 983 kb)

PowerPoint slides

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Moor, K., Diard, M., Sellin, M. et al. High-avidity IgA protects the intestine by enchaining growing bacteria. Nature 544, 498–502 (2017).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

This article is cited by


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.


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