Checkpoint for gut microbes after birth

The route to the establishment of a beneficial microbial community in the gut after birth is not fully understood. It now emerges that a gut-cell protein in newborn mice shapes the long-term composition of this community.
Andrew J. Macpherson is in the University Clinic for Visceral Surgery and Medicine, Bern University Hospital, and in the Department for BioMedical Research, University of Bern, 3008 Bern, Switzerland.

Search for this author in:

Stephanie C. Ganal-Vonarburg is in the University Clinic for Visceral Surgery and Medicine, Bern University Hospital, and in the Department for BioMedical Research, University of Bern, 3008 Bern, Switzerland.

Search for this author in:

Although people think of divorce, bereavement and job loss as major life events, perhaps no change in a person’s life is quite as dramatic as the moment of their birth. As a baby exits the birth canal, the newborn infant loses placental support, and the respiratory system and gut must start to function. Moreover, both beneficial and pathogenic microorganisms will be encountered, and will compete to colonize the baby’s body. Writing in Nature, Fulde et al.1 report that the intestinal receptor protein TLR5 is involved in actively shaping the long-term composition of the gut microbial community, termed the microbiota, in newborn mice.

The bacterial colonization of the gut normally starts in the birth canal2. Then, successive waves of increases and decreases in microbial species occur during a period of microbiotal change, which lasts approximately 18 months in humans3. Nutritional conditions and immune-system development in early life both affect gut colonization, with far-reaching consequences for later growth and health. Tragically, more than 15 million children around the globe under the age of 5 years suffer from malnutrition and severe wasting ( This is caused by a combination of insufficient calorie intake and a type of immune dysfunction that is linked to abnormal bacterial colonization in the gut, called environmental enteropathy. The correct functioning of immune cells called B cells and T cells is partly determined by exposure during their early development to non-pathogenic microbes, which can therefore have long-term consequences for the composition of microbial species in the microbiota and subsequent resistance to pathogenic challenges4.

The immature, vulnerable immune system of a newborn can to some extent be shielded from pathogenic attack by the presence of antibodies from the infant’s mother, transferred across the placenta or in breast milk — a particularly effective health-boosting measure. Breast-milk antibodies that remain in the gut can help to determine the composition of microbes that colonize the intestine, and thereby prevent excessive immune responses to non-pathogenic microbes5,6. Yet despite this maternally provided immune protection, an infant still faces an extremely sensitive period in early life when the progressive microbial colonization of internal and external body surfaces occurs concurrently with the development and maturation of the immune system.

The cellular composition and function of almost every organ changes when a germ-free animal becomes colonized with a microbiota. Such changes are triggered by molecules from the microbes themselves, and the results can help to prevent inflammation as an animal adapts to the presence of microbes on its body surfaces.

“There does exist a mechanism in the neonate that favours the colonization of certain bacteria and disfavours the colonization of others.”

It was previously thought that the adaptation of host tissues to accommodate the presence of intestinal microbes could occur equally effectively at any age, on the basis of experiments that introduced a microbiota into adult animals raised in germ-free conditions. Yet awareness is growing that, as a newborn animal develops, an ordered, age-dependent sequence of interrelated immune and microbial checkpoints is required for proper adaptation and to ensure a healthy microbiotal composition7. Examples of adaptations in the crucial early time window include regulation of the induction of different classes of antibody called isotypes and of the numbers of immune cells called natural killer T cells in the intestine7,8. Microbial colonization of the newborn animal is not necessary for the innate branch of the immune system to develop, because molecules from maternal microbes delivered across the placenta and in milk can suffice to drive some of this process9.

Previous studies1012 of mice that are deficient in TLR5, a receptor belonging to a family that is linked to microbial recognition, reported that gut microbes in such animals have defects that trigger metabolic abnormalities such as body-weight gain and fatty changes in the liver. To investigate how postnatal development affects the establishment of the gut microbial community, Fulde and co-workers compared gene expression in gut epithelial cells in three-day-old mice with that in adult mice, and found that the gene encoding TLR5 is highly expressed in the infant mice. The authors then investigated whether this protein has a role in early postnatal gut development.

Fulde and co-workers show that the expression of TLR5 in intestinal epithelial cells in early life is an example of a checkpoint in a developmental process that is coordinated with microbial colonization to achieve healthy mutualism between the host and its microbiota. TLR5 can bind the bacterial protein flagellin, and the authors found that this drives secretion of the antimicrobial protein Reg3γ. Both TLR5 and Reg3γ help to limit early-life colonization by bacteria that express flagellin, which is a component of the flagellar structure that aids bacterial motility. Flagellin is found on some pathogenic bacteria, although not all bacteria that express flagellin are pathogenic. The authors determined when during development this TLR-5-linked effect occurred by conducting colonization experiments in which infant mice (up to 10 days old) received an equal mixture of non-flagellated and flagellated strains of Salmonella bacteria. In wild-type mice, the intestinal colonization of the strain lacking flagellin was consistently higher than that of the strain that had flagellin. This difference was not seen in TLR5-deficient mice (Fig. 1).

Figure 1 | A host protein affects the composition of the gut microbial community. Fulde et al.1 report that the expression of the protein TLR5 in gut cells during a limited time window in early postnatal development in mice helps to reduce the presence of bacteria that express the protein flagellin. This protein is a component of a bacterial structure called a flagellum, and is a hallmark of pathogenic bacteria, although not all bacteria that express flagellin are pathogenic. a, b, When the authors gave wild-type or TLR5-deficient infant mice a bacterial sample that was an equal mixture of flagellar and non-flagellar strains of Salmonella, they found that a higher proportion of flagellin-expressing bacteria were present in the gut of the TLR5-deficient mice (b) than in the wild-type mice (a). Additional experiments revealed that early expression of TLR5 can help to shape the long-term composition of microbial species residing in the gut.

In another experimental approach taken by Fulde and colleagues, germ-free, wild-type pups were intestinally colonized with either a ‘healthy’ microbiota derived from wild-type mice or a ‘dysbiotic’ microbiota from TLR5-deficient mice, which would be liable to trigger changes linked to metabolic disease1012. Fulde et al. found that wild-type pups could drive the species composition of the dysbiotic-microbiota gut residents towards that of the microbiota derived from wild-type mice. However, it seems that TLR5 is required to shape microbiotal composition only during a specific postnatal time frame, because the authors found that the dysbiotic microbiotal composition was less effectively shaped if transferred into germ-free, TLR5-deficient pups or adult, germ-free, wild-type mice. Once the microbiota composition had been shaped, it persisted long into adult life (at least 42 days). The window of opportunity for this microbiota-shaping effect is restricted to early postnatal life by the downregulation of TLR5 expression in gut epithelial cells at weaning (approximately 21 days after birth).

The gut microbial composition for a given animal strain varies considerably with the different housing facilities used for laboratory animals and even between cages in a single facility, with the microbial composition changing gradually from generation to generation even in a continuously inbred colony. External influences, such as dietary changes or environmental cage-associated effects, might bring about changes that could confound experimental results. One way around this is to breed heterozygous animals (those that have a mutation in only one of the two copies of a gene of interest), and then to compare the gut microbes of littermates that are either deficient in the gene of interest or wild type. Yet, as others have reported13, with this kind of approach, differences in gut-microbe composition are dominated by the transmission of microbiota from parents to their offspring, rather than being mainly affected by whether the offspring are deficient for a certain gene, such as the gene that encodes TLR5.

Nevertheless, in the context of the gut microbes and the stage of postnatal development that they studied, Fulde and colleagues show not only that TLR5-deficient animals develop an abnormal gut microbial community, but also that the presence of TLR5 in gut cells is sufficient to drive the community towards a more normal composition by limiting the presence of flagellated bacteria.

As with every major advance, questions remain. Determining the way in which different microbes associate and occupy niches in the gut in the presence or absence of TLR5 will require studies using mouse colonies in which the animals have a range of predefined, stable microbial compositions. This will allow researchers to discover how the presence or absence of microbial species affects the microbiota interaction with the host, and, by using a technique known as stable isotope tracing, to assess whether molecular crosstalk between microbial species affects the overall assembly of the microbiota.

Fulde and colleagues’ work provides two key messages. First, it shows that TLR5 expression in early life can have a lasting effect on the composition of the intestinal microbial community. And second, it supports the emerging idea of sequential milestones during the mutually connected postnatal development of a host and its associated microbes.

doi: 10.1038/d41586-018-05861-z
Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up


  1. 1.

    Fulde, M. et al. Nature (2018).

  2. 2.

    Dominguez-Bello, M. G. et al. Proc. Natl Acad. Sci. USA 107, 11971–11975 (2010).

  3. 3.

    Yatsunenko, T. et al. Nature 486, 222–227 (2012).

  4. 4.

    Macpherson, A. J., Gomez de Agüero, M. & Ganal-Vonarburg, S. C. Nature Rev. Immunol. 17, 508–517 (2017).

  5. 5.

    Rogier, E. W. et al. Proc. Natl Acad. Sci. USA 111, 3074–3079 (2014).

  6. 6.

    Koch, M. A. et al. Cell 165, 827–841 (2016).

  7. 7.

    Gensollen, T., Iyer, S. S., Kasper, D. L. & Blumberg, R. S. Science 352, 539–544 (2016).

  8. 8.

    Olszak, T. et al. Science 336, 489–493 (2012).

  9. 9.

    Gomez de Agüero, M. et al. Science 351, 1296–1302 (2016).

  10. 10.

    Vijay-Kumar, M. et al. Science 328, 228–231 (2010).

  11. 11.

    Chassaing, B., Ley, R. E. & Gewirtz, A. T. Gastroenterology 147, 1363–1377 (2014).

  12. 12.

    Carvalho, F. A. et al. Cell Host Microbe 12, 139–152 (2012).

  13. 13.

    Ubeda, C. et al. J. Exp. Med. 209, 1445–1456 (2012).

Download references