News & Views | Published:

Population health

Immaturity in the gut microbial community

Nature volume 510, pages 344345 (19 June 2014) | Download Citation

Undernourished children fall behind not only on growth, but also on maturation of their intestinal bacterial communities, according to a study comparing acutely malnourished and healthy Bangladeshi children. See Letter p.417

Effective assessments of child growth rely on a knowledge of underlying processes, appropriate standards and accurate measurements. These elements form a comparative framework with which trajectories can be charted and developmental milestones marked, providing 'actionable intelligence' on health and disease in individuals and populations. In this issue, Subramanian et al.1 (page 417) chart a different path — one in which the milestones are microbial — for young children living in the Mirpur urban slum of Dhaka, Bangladesh, many of whom suffer from undernutrition (Fig. 1). The authors find evidence for delays in the development of gut bacterial communities in acutely malnourished children compared with healthy children, and that these delays are only fleetingly ameliorated by standard treatment. The team's approach for classifying and tracking gut microbiota may enhance assessments of childhood health and development, and improve therapeutic strategies.

Figure 1: Children from the Mirpur slum of Dhaka, Bangladesh.
Figure 1

Growth faltering in early childhood is the hallmark of undernutrition, a pervasive condition in the developing world that arises from insufficient intake, absorption or assimilation of nutrients. Undernutrition results from scarce and nutrient-poor food, poor-quality water and unsanitary living conditions. Recurrent bouts of gastrointestinal infections exacerbate and perpetuate the problem2. Positive feedback loops can ensue both in individuals, when intestinal damage reinforces poor growth and susceptibility to infection, and over generations, when maternal undernutrition causes undernutrition in children. Over time, these cycles can impair learning, limit productivity and ultimately perpetuate poverty. Maternal and child undernutrition were a factor in 3.1 million (45% of all) deaths in children under 5 years of age in 20113. Children under 2 are particularly vulnerable to undernutrition (and infection), but are also the most responsive to treatment4.

In early childhood, assembly of gut microbial communities results from the sequential arrival of taxa from external sources and the extinction of taxa already present, in part in response to age-associated events such as weaning. Progression towards an adult-like, 'mature' state occurs over the first 2–3 years of life. These postnatal events prompt the terminal maturation of intestinal structures, stimulate immune responses and provide resistance to invasion by pathogens; aberrant or delayed assembly is associated with altered metabolism and immune function. Thus, because they affect and are affected by similar factors, undernutrition and gut-microbiota development are closely intertwined. Unravelling the two and discerning the role of host and environmental factors are daunting but important goals4.

Anthropometric indicators — physical measurements, such as weight for height, which are scored relative to a reference population — are indispensable tools in the assessment and treatment of undernutrition. Not surprisingly, equivalent international standards for gut-microbiota development are not yet available: few individuals have been followed in detail over the requisite time frame, and it also seems that microbiota composition in early childhood differs across populations5,6. In light of this, Subramanian and colleagues examined healthy and malnourished children from the same urban area, ostensibly minimizing genetic and environmental differences between the two groups.

To derive a model of gut-microbiota development in the Bangladeshi children, the researchers collected faecal samples from 50 well-nourished subjects at monthly intervals over the first 2 years of life. Next, they surveyed the bacterial communities in the samples by sequencing 16S ribosomal RNA genes, which are used to define and enumerate bacterial taxa. The taxa from 12 of the subjects were then assessed and ranked according to their ability to discriminate between different host ages. The authors found that the 24 most age-discriminatory taxa could predict the ages of the remaining 38 healthy children from their gut microbial composition.

In keeping with the tradition of anthropometric indicator scores, the authors defined two indicators of gut-microbiota maturation: relative microbiota maturity and a microbiota-for-age Z-score (MAZ). The overall gist is this: if the model classifies your gut microbiota as that of a 6-month-old when you are actually 18 months old, then your gut microbiota is probably 'immature' — its composition looks 'younger' than that of most healthy people of your age (although it may be different in other ways, too). Applying these indicators to their well-nourished cohort, the authors found that microbiota maturity decreased during diarrhoeal episodes, increased with infant-formula consumption, was unchanged by recent antibiotic use and was correlated among family members.

Subramanian and colleagues then applied their microbiota-maturation indices to 64 children aged 6–20 months at the start of the study who were sampled during and after inpatient treatment for severe acute malnutrition. The children were participating in a randomized trial comparing two therapeutic foods, in combination with supportive therapy that included antibiotics. Compared with healthy children, the malnourished children showed significant microbiota immaturity during treatment, regardless of treatment group. Notably, in the 2–3 months following treatment, the children's microbiota-maturation scores improved significantly; however, after this period, much of this catch-up maturation was lost. These patterns mirrored the anthropometric outcomes of the study: although they gained weight initially, children in both groups remained severely underweight compared with healthy children at the end of the follow-up period. The results also support previous studies of undernutrition in humanized mouse models7.

Degraded ecosystems are notoriously difficult to restore. Often, such efforts focus on restoring environmental conditions (akin to the food intervention in Subramanian and colleagues' study) and eliminating unwanted species (akin to the antibiotic therapy), then waiting for assembly processes to play out 'naturally' to restore the desired community8. But degraded communities can be resistant or resilient to change8,9, and although host health can be restored, youth cannot. The composition of mature communities may depend on the timing and order of earlier species introductions (and extinctions)10 and may prove difficult to reconstitute (by the use of probiotics, for example). Thus, an ounce of prevention is likely to be worth a pound of cure and, as with other types of developmental delays, early intervention may be crucial.

The approach presented by Subramanian et al. could be used to develop standards across the globe, and then to monitor gut colonization during early childhood, as an early-warning system for microbiotas that are falling 'off track' (and there may be many such tracks to health). A detailed analysis of microbiota maturation in well-nourished populations will complement this work, and allow further deconvolution of some of the common microbiota insults that were unavoidably layered and repeated in the current study. It is becoming clear that recognizing which features of microbiota assembly are associated with health, and understanding whether and how healthy communities bounce back after disturbance, are key requirements for future human-development roadmaps.


  1. 1.

    et al. Nature 510, 417–421 (2014).

  2. 2.

    et al. Clin. Infect. Dis. 54, 185–192 (2012).

  3. 3.

    et al. Lancet 382, 427–451 (2013).

  4. 4.

    , , & Sci. Transl. Med. 4, 137ps12 (2012).

  5. 5.

    et al. Proc. Natl Acad. Sci. USA 107, 14691–14696 (2010).

  6. 6.

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

  7. 7.

    et al. Science 339, 548–554 (2013).

  8. 8.

    , & Trends Ecol. Evol. 19, 46–53 (2004).

  9. 9.

    , , , & Science 336, 1255–1262 (2012).

  10. 10.

    in Community Ecology: Processes, Models, and Applications (eds Verhoef, H. A. & Morin, P. J.) 45–54 (Oxford Univ. Press, 2009).

Download references

Author information


  1. Elizabeth K. Costello and David A. Relman are in the Departments of Medicine and of Microbiology & Immunology, Stanford University, Stanford, California 94305-5124, USA, and at the Veterans Affairs Palo Alto Health Care System, Palo Alto, California.

    • Elizabeth K. Costello
    •  & David A. Relman


  1. Search for Elizabeth K. Costello in:

  2. Search for David A. Relman in:

Corresponding authors

Correspondence to Elizabeth K. Costello or David A. Relman.

About this article

Publication history



Further reading


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.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing