Research Article

Gut bacteria that prevent growth impairments transmitted by microbiota from malnourished children

See allHide authors and affiliations

Science  19 Feb 2016:
Vol. 351, Issue 6275, aad3311
DOI: 10.1126/science.aad3311

Microbiota and infant development

Malnutrition in children is a persistent challenge that is not always remedied by improvements in nutrition. This is because a characteristic community of gut microbes seems to mediate some of the pathology. Human gut microbes can be transplanted effectively into germ-free mice to recapitulate their associated phenotypes. Using this model, Blanton et al. found that the microbiota of healthy children relieved the harmful effects on growth caused by the microbiota of malnourished children. In infant mammals, chronic undernutrition results in growth hormone resistance and stunting. In mice, Schwarzer et al. showed that strains of Lactobacillus plantarum in the gut microbiota sustained growth hormone activity via signaling pathways in the liver, thus overcoming growth hormone resistance. Together these studies reveal that specific beneficial microbes could potentially be exploited to resolve undernutrition syndromes.

Science, this issue p. 10.1126/science.aad3311, p. 854

Structured Abstract


As we come to appreciate how our microbial communities (microbiota) assemble following birth, there is an opportunity to determine how this facet of our developmental biology relates to the healthy or impaired growth of infants and children. Childhood undernutrition is a devastating global health problem whose long-term sequelae, including stunting, neurodevelopmental abnormalities, and immune dysfunction, remain largely refractory to current therapeutic interventions.


To test the hypothesis that perturbations in the normal development of the gut microbiota are causally related to undernutrition, we first applied random forests (RF), a machine learning method, to bacterial 16S ribosomal RNA data sets generated from fecal samples that were collected serially from healthy Malawian infants and children during their first 3 postnatal years. Age-discriminatory bacterial taxa were identified with distinctive time-dependent changes in their relative abundances; they were used to construct a sparse RF-derived model describing a program of normal postnatal gut microbiota development that is shared across biologically unrelated individuals. A metric based on this model (microbiota-for-age Z-score) was used to define the state of development (maturation) of fecal microbiota from infants and children with varying degrees of undernutrition. Fecal samples obtained from 6- and 18-month-old children with healthy growth patterns or with varying degrees of undernutrition were transplanted into young germ-free mice that were fed a representative Malawian diet. The recipient animals’ rate of lean body mass gain was characterized by serial quantitative magnetic resonance, their metabolic phenotypes were determined by targeted mass spectrometry, and their femoral bone morphologic features were delineated by microcomputed tomography.


Undernourished children in the Malawian birth cohort that we studied have immature gut microbiota. Unlike microbiota from healthy children, immature microbiota transmit impaired growth, altered bone morphology, and metabolic abnormalities in the muscle, liver, and brain to recipient gnotobiotic mice. The representation of several age-discriminatory taxa in the transplanted microbiota harbored by recipient animals correlated with their growth rates. Microbiota from 6-month-old infants produced greater effects on growth than did microbiota from 18-month-old children, although in each age bin, the growth effects produced by a healthy donor’s community were greater than those produced by an undernourished donor’s community. Cohousing coprophagic mice shortly after they received microbiota from healthy or severely stunted and underweight 6-month-old infants resulted in the invasion of age- and growth-discriminatory taxa from the former into the latter microbiota in the recipient animals, with associated prevention of growth impairments. Introducing cultured members from this group of invasive species ameliorated growth and metabolic abnormalities in recipients of microbiota from undernourished donors.


These preclinical findings provide evidence that gut microbiota immaturity is causally related to childhood undernutrition. The age- and growth-discriminatory taxa that we identified should help direct studies of the effects of host and environmental factors on gut microbial community development, and they represent therapeutic targets for repairing or preventing gut microbiota immaturity.

Preclinical evidence that gut microbiota immaturity is causally related to childhood undernutrition.

(A) A model of normal gut microbial community development in Malawian infants and children, based on the relative abundances of 25 bacterial taxa that provide a microbial signature defining the “age,” or state of maturation, of an individual’s (fecal) microbiota. (Hierarchical clusterings of operational taxonomic units are indicated on the left.) (B) Fecal samples from healthy (H) or stunted and underweight (Un) infants and children were transplanted into separate groups of young germ-free mice that were fed a Malawian diet. The immature microbiota of Un donors transmitted impaired growth phenotypes to the mice. (C) Evidence that a subset of age-discriminatory taxa are also growth-discriminatory. Cohousing mice shortly after they received microbiota from 6-month-old healthy or undernourished donors resulted in the invasion of taxa from the healthy donor’s microbiota (HCH) into the undernourished donor’s microbiota (UnCH) among recipient animals and prevented growth impairments. Adding cultured invasive growth-discriminatory taxa directly to the Un donor’s microbiota (Un+) improved growth. 


Undernourished children exhibit impaired development of their gut microbiota. Transplanting microbiota from 6- and 18-month-old healthy or undernourished Malawian donors into young germ-free mice that were fed a Malawian diet revealed that immature microbiota from undernourished infants and children transmit impaired growth phenotypes. The representation of several age-discriminatory taxa in recipient animals correlated with lean body mass gain; liver, muscle, and brain metabolism; and bone morphology. Mice were cohoused shortly after receiving microbiota from healthy or severely stunted and underweight infants; age- and growth-discriminatory taxa from the microbiota of the former were able to invade that of the latter, which prevented growth impairments in recipient animals. Adding two invasive species, Ruminococcus gnavus and Clostridium symbiosum, to the microbiota from undernourished donors also ameliorated growth and metabolic abnormalities in recipient animals. These results provide evidence that microbiota immaturity is causally related to undernutrition and reveal potential therapeutic targets and agents.

Undernutrition is a leading cause of infant and childhood mortality worldwide (15). The mechanisms that underlie this disorder’s manifestations, ranging from persistent abnormalities in growth and immune function to cognitive deficits, remain obscure. Childhood undernutrition is not due to food insecurity alone; it also results from a combination of factors including diets with low nutrient density or bioavailability, pathogen burden, and gut mucosal barrier dysfunction (68).

The World Health Organization has used a cohort of 8440 healthy children living in six countries to develop anthropometric standards that define nutritional status [weight-for-height Z-score (WHZ), weight-for-age Z-score (WAZ), and height-for-age Z-score (HAZ)] (9, 10). A recent study (11) of infants and children with healthy growth phenotypes living in Mirpur, an urban slum in Dhaka, Bangladesh, involved monthly fecal collection from birth through the end of the second year of life. In this study, a 16S ribosomal RNA (rRNA) analysis of bacterial membership in the gut microbiota and the application of a machine learning method [random forests (RF) (12)] revealed 24 “age-discriminatory” taxa, whose changes in relative abundance over time define a program of normal maturation of the microbiota across biologically unrelated individuals (11). This model served as the basis for computing two related metrics, relative microbiota maturity and microbiota-for-age Z-score (MAZ), that significantly correlated with the chronological age of children with healthy growth phenotypes. Applying these metrics showed that children living in Mirpur with moderate acute malnutrition (MAM) or severe acute malnutrition (SAM) have gut microbiota that are “immature;” in other words, the representation of the age-discriminatory taxa in their gut communities was more similar to younger than to age-matched healthy individuals from the same locale. Moreover, the degree of this immaturity was greater in SAM than MAM (11). Treatment of children with SAM with either one of two ready-to-use therapeutic foods produced only incomplete and transient improvement in this immaturity and no improvement in their HAZ scores (11).

These findings raise several questions. To what extent are these age-discriminatory taxa also indicative of the normal development of the gut microbiota of infants and children elsewhere? Are the age-discriminatory taxa biomarkers of gut microbiota development, or are they mediators of healthy growth? If the latter, can their introduction into immature microbiota prevent disease?

A Malawian model of gut microbiota development

To address these questions, we first developed a RF-based model of gut microbial community development from a serially sampled cohort of Malawian twins that were concordant for healthy growth. We investigated correlations between nutritional status and microbiota maturation to ascertain whether an MAZ score could predict future growth performance.

In a previous study, we followed a cohort of 317 twin pairs and three sets of triplets, living in five rural villages in southern Malawi, from birth to 36 months of age, with periodic sampling of their fecal microbiota (13). To define healthy microbiota development in this population, we took the polymerase chain reaction amplicons generated from variable region 4 (V4) of bacterial 16S rRNA genes and sequenced them. We used 220 fecal samples collected from 27 twin pairs and two sets of triplets whose serial anthropometric measurements were indicative of consistently healthy growth [WHZ = 0.09 ± 0.93; fecal collection over an age range of 0.6 to 33.5 months; 4.0 ± 1.6 samples (mean ± SD) per individual; tables S1A and S2A provide microbiota donor characteristics and a summary of sequencing data sets]. V4 16S rRNA reads with ≥97% nucleotide sequence identity were grouped into operational taxonomic units (97% ID OTUs), and taxonomic assignments were made (14, 15).

We modeled microbiota maturation in these healthy Malawian infants and children by using RF to regress OTUs against chronological age in a subset (“training set”) of the healthy twin cohort (Fig. 1A). The application of the training set–based model to the subset excluded from model building (“test set”) yielded a positive and significant correlation between host chronological age and predicted microbiota age (coefficient of determination R2 = 0.80, P < 0.0001) (Fig. 1C and fig. S1). The model was further validated using an additional randomized nutritional intervention trial that took place in the Mangochi district of southern Malawi, in which subjects were studied at 6, 12, and 18 months of age [the iLiNS-DYAD-M study (16)]. We again saw a positive and significant correlation between chronological age and predicted microbiota age (R2 = 0.71, P < 0.0001). Multiple OTUs from the sparse Malawian model had ≥97% sequence identity to OTUs in the sparse Bangladeshi model (Fig. 1A and table S3B) (11).

Fig. 1 Sparse RF-derived model of gut microbiota maturation obtained from concordant healthy Malawian twins and triplets.

(A) A RF regression of fecal bacterial 97% ID OTUs from a training set of healthy Malawian infants and children (N = 31) on chronological age yielded a rank order of age-discriminatory taxa. RF assigns a mean squared error (MSE), or feature importance score, to each OTU; this metric indicates the extent to which each OTU contributes to the accuracy of the model. The 25 most age-discriminatory taxa, ranked by MSE, yielded a sparse model that predicted microbiota age and accounted for ~80% of the observed variance in the healthy cohort (table S3A gives a complete list of OTUs and MSE values). The top 25 most discriminatory OTUs with their taxonomic assignments are shown ranked by feature importance (mean ± SD of the increase in MSE). The inset shows the results of 10-fold cross-validation; as OTUs are added to the model in order of their feature importance rank, the model’s error decreases. Taxa highlighted in red indicate OTUs with >97% nucleotide sequence identity with an OTU present in a sparse RF-based model of microbiota maturation in healthy Bangladeshi infants and children (11). (B) A heat map of changes over time in the relative abundances of the 25 OTUs in fecal microbiota collected from healthy Malawian infants and children constituting the test set (N = 29). OTUs are hierarchically clustered according to pairwise distances determined by Pearson correlation. (C) Predictions of chronological age using the sparse 25-OTU model of microbiota age in healthy children constituting the test set. R2 was calculated by Pearson correlation.

We then used the Malawian RF-derived model to analyze the relationship between microbiota maturity and nutritional status in 259 children enrolled in iLiNS-DYAD-M who had fewer than three reported days of antibiotic consumption from 6 to 18 months of age (table S1B). There was a significant positive correlation between their MAZ scores and their WHZ (Spearman’s rank correlation coefficient ρ = 0.1664, P = 0.0073) and WAZ scores (ρ = 0.1715, P = 0.0056) but not their HAZ scores (ρ = 0.1022, P = 0.1) (table S2B). There was a significant correlation between MAZ at 12 months and anthropometry at 18 months [ρ = 0.1406 and P = 0.02 for WHZ; ρ = 0.1373 and P = 0.02 for WAZ (for HAZ, ρ = 0.1184 and P = 0.05)]. These results suggest that MAZ may be useful for predicting future (ponderal) growth. Further studies and analyses are required to discern the effects of several variables, such as the duration of prior antibiotic use, enteropathogen load, number of diarrheal days, geography, and various nutritional interventions, on the relationship of MAZ measurements at various postnatal ages to anthropometry and other metrics of healthy growth (e.g., cognitive testing and immunization responses).

Identification of age-discriminatory taxa that are also growth-discriminatory

To test whether there is a causal relationship between microbiota maturity and growth, we selected fecal samples from 19 Malawian donors representing either healthy or undernourished growth phenotypes for transplantation into 5-week-old, actively growing germ-free C57BL/6J mice. Of the 19 donor samples, nine were from 6-month-old infants, four of which were classified as healthy [WHZ = 1.45 ± 0.57 (mean ± SD), WAZ = 1.75 ± 0.56, and HAZ = 1.27 ± 0.45] and five of which were classified as moderately or severely underweight and stunted (WHZ = –1.27 ± 0.49, WAZ = –3.90 ± 1.82, and HAZ = –4.36 ± 2.07). Ten samples were from 18-month-old children, four with healthy weights (WHZ = 1.44 ± 0.08, WAZ = –0.3 ± 0.57, and HAZ = –2.84 ± 0.99) and six who were moderately or severely underweight and stunted (WHZ = –1.75 ± 0.17, WAZ = –2.88 ± 0.42, and HAZ = –3.28 ± 0.82) (table S4A includes MAZ metrics; all 18-month-old donors were members of twin pairs, in which HAZ scores are typically lower than in singletons, and all 6-month-old donors were singletons). Fecal samples from the 18-month-old children were from the Malawi twin cohort; the 6-month-old microbiota donors were participants in the iLiNS-DYAD-M study who had not yet received a nutritional supplement.

Each microbiota sample was transplanted into a separate group of 5-week-old male germ-free mice (N = 5 animals per sample). Three days before transplantation, all mice were switched onto a sterile (irradiated) Malawian diet formulated based on the results of a dietary survey of the complementary feeding practices for 9-month-old Malawian children enrolled in the iLiNS-DOSE study ( identifier: NCT00945698) that took place in the Mangochi district. We selected eight ingredients to produce a cooked diet representative of that consumed by Malawian children (M8; see the supplementary materials); its micro- and macronutrient content does not fulfill the needs of humans or mice (table S5, A and B, gives a list of ingredients and the results of a direct nutritional analysis).

After a single gavage of the donor microbiota, recipient mice were observed for 4 to 5 weeks (Fig. 2A shows the experimental design). Fecal samples were collected for bacterial V4 16S rRNA analysis. Growth was monitored by serial measurements of total body weight and body composition [lean mass and fat mass, as defined by quantitative magnetic resonance (qMR)]; after euthanasia, femurs were removed for microcomputed tomographic (micro-CT) characterization of bone morphology.

Fig. 2 Transplantation of microbiota from 6- and 18-month-old donors to young germ-free mice provides evidence of a causal relationship between gut microbiota maturity and growth phenotypes.

(A) Experimental design of the microbiota screen. Mice (4.5 weeks old) were switched to the M8 diet 3 days before gavage with the selected microbiota donor’s fecal sample (N = 5 mice per donor). Fecal samples, body weight, and body composition were defined at the indicated time. (B and C) Gnotobiotic mice colonized with fecal samples from healthy donors gain more total body weight (B) and lean mass (C) than mice colonized with microbiota from undernourished donors (the graphs show means ± SEM; P values shown for the donor status effect are based on two-way ANOVA). All recipient mice harbor microbiota that represent ≥50% of the OTU diversity that is present in the intact uncultured donor’s sample. (D) The 30 most weight gain–discriminatory OTUs and their taxonomic assignments, ranked by feature importance (the bar graph shows increase in mean MSEs ± SD). The weight gain model explained ~66% of the observed phenotypic variation (P < 0.0001, determined by a permutation test with 999 permutations). Taxa in red indicate OTUs that appear within the 30 most discriminatory OTUs for the RF-based models of both weight and lean mass gain. Taxa in purple indicate species that appear in the 25-member sparse RF-derived model of Malawian gut microbiota maturation. Colors to the right of the OTU ID numbers represent Spearman’s rank correlation of the same OTU ID with chronological age within the healthy Malawian infant and child cohort (table S9).

Mice colonized with microbiota from healthy donors (N = 8 donors) gained significantly more weight and lean body mass than mice colonized with microbiota from undernourished donors (N = 11) [P < 0.0001 and P = 0.0001, respectively, determined by two-way analysis of variance (ANOVA); table S6, A and B], yet there was no significant difference in food consumption between the groups (P > 0.05, determined by Student’s t test; table S7). 16S rRNA sequencing of fecal samples obtained from recipient mice revealed variable transplantation efficiency. Of the 19 donor samples, eight produced transplantation efficiencies of ≥50%. When we restricted our analysis to the eight donor samples producing ≥50% transplantation efficiency, the discordant growth phenotypes between recipients of healthy versus undernourished microbiota were pronounced (weight gain, P < 0.0001; lean mass gain, P = 0.0005; two-way ANOVA; Fig. 2, B and C). There was no significant difference in fat mass gain (P = 0.78, two-way ANOVA; table S6B) or food consumption between the groups (P > 0.05, Student’s t test; tables S6, A and B, and S7).

For both the 6-month and 18-month age bins, mice colonized with healthy donor microbiota gained significantly more total body weight and lean mass than those colonized with undernourished donor microbiota (weight, P = 0.0003 and P = 0.0043 for recipients of the 6- and 18-month-old donor microbiota, respectively; lean mass, P = 0.03 and P = 0.0013, respectively; two-way ANOVA; table S6, A and B). Recipients of microbiota from healthy or undernourished 6-month-old donors grew more than recipients of microbiota from 18-month-old healthy or undernourished donors (P < 0.0001 for healthy and P < 0.0001 for undernourished, based on lean mass gain; two-way ANOVA).

Growth over the course of the 5-week experiment in the 19 different groups of recipient mice ranged from 107 to 156% of the starting weight (averaged per group; table S6A). There was no significant relationship between growth phenotypes and bacterial diversity in the fecal microbiota of recipient animals (table S6A). We applied RF to regress the growth phenotypes of recipient gnotobiotic mice against 97% ID OTUs identified in their fecal microbiota. Two models were generated, one based on weight gain, the other based on lean mass gain (Fig. 2D and fig. S2, respectively, and table S8, A and B). Two of the growth-discriminatory species represented in the RF-based models of weight and lean mass gain, Bifidobacterium longum and Faecalibacterium prausnitzii, were highly age-discriminatory, ranking first and eighth, respectively, in the Malawian model shown in Fig. 1A, and fifth and first in the Bangladeshi model (11). In the healthy Malawian and Bangladeshi populations used to create these models, B. longum is an early colonizer with its highest mean relative abundance at 5 months of age, whereas F. prausnitzii becomes more prominent later (highest mean relative abundance achieved at 19 months) (Fig. 1B) (11).

The relative abundances of 13 OTUs that were significantly correlated with weight gain and seven OTUs that were significantly correlated with lean mass gain also had significant correlations with chronological age in concordant healthy Malawian twins and triplets, including three OTUs for F. prausnitzii (P < 0.0001; Fig. 2D and fig. S2; table S9 gives a complete list of OTUs with their Spearman’s rank correlation coefficients). Moreover, 15 OTUs that were positively and significantly correlated with weight gain and 11 OTUs that were positively and significantly correlated with lean mass gain were also significantly correlated with chronological age for the 259 infants and children from iLiNS-DYAD-M, including the same three OTUs for F. prausnitzii that were significantly correlated with age in the cohort of healthy Malawian twins and triplets (P < 0.0001 for all correlations; table S9).

The relationship between age- and growth-discriminatory taxa and femoral bone phenotypes

We used micro-CT to assay the morphology of trabecular and cortical regions of femurs obtained from mice harboring microbiota with ≥50% transplantation efficiency. There were strong trends toward higher cortical ratios of bone volume to tissue volume (BV/TV) and volumetric bone mineral density in recipients of gut microbial communities from undernourished donors (P = 0.05 and P = 0.07, respectively; Mann-Whitney test). Differences between mice harboring microbiota from 6- and 18-month-old donors were evident in trabecular rather than in cortical bone, regardless of donor nutritional status; mice colonized with 6-month-old donor communities had significantly higher bone mineral density, BV/TV, and trabecular connectivity and number, as well as significantly lower trabecular spacing, irrespective of their nutritional status (P < 0.001 for all bone metrics; Mann-Whitney test; fig. S3). We extended our RF and Spearman ranked correlation analyses to identify OTUs that discriminate these femoral bone phenotypes (fig. S4 and table S10). Six of the OTUs represented in both of the growth-discriminatory RF-based models, including one assigned to B. longum, were also represented among the top 20 most discriminatory features in at least three of the five RF-based models of bone metrics (fig. S4). Although our sample size was small, these results provide additional evidence for microbiota-dependent regulation of bone morphology (17), with the effects being influenced by the age and nutritional status of the donor.

Repairing impaired growth phenotypes

We next determined whether age- and growth-discriminatory bacterial species were capable of repairing the growth abnormalities associated with microbiota from stunted and underweight donors. We selected gnotobiotic recipients of microbiota from two 6-month-old donors that transmitted the most discordant growth phenotypes in the initial screen of 19 microbiota (fig. S5A). One was a healthy infant from Mangochi (HAZ = 1.49, WAZ = 1.43, and WHZ = 0.9) with age-appropriate microbiota maturity (microbiota age of 6.7 months); the other was a severely stunted and underweight infant (HAZ = –3.35, WAZ = –3.08, and WHZ = –0.79) from Malindi (located 20 km from Mangochi) with an immature microbiota (microbiota age of 4.6 months). The configurations of these transplanted microbiota were distinct: Mice colonized with the healthy donor community were dominated by F. prausnitzii (33 ± 19% relative abundance), whereas mice colonized with microbiota from the undernourished donor were dominated by Clostridium neonatale (37 ± 11% relative abundance) (N = 5 mice per donor). Of the 27 species (66 97% ID OTUs) present in recipients of the healthy donor community and the 22 species (33 97% ID OTUs) present in recipients of immature microbiota from the undernourished donor, 13 OTUs representing 13 different species were present in both (see fig. S5B and table S11).

Taking advantage of the coprophagic behavior of mice, we gavaged 5-week-old male germ-free C57BL/6J animals with microbiota from either the healthy or the stunted and underweight donor. Four days later, before phenotypic differences were apparent, we combined (dually housed) mice containing the healthy (H) donor microbiota (H-H controls). We also combined mice that received the undernourished (Un) donor community (Un-Un controls). The experimental group consisted of H and Un mice that were cohoused with one another (HCH-UnCH) (see Fig. 3A). All animals were fed the M8 diet beginning three days before colonization and throughout the course of the 3-week experiment. Animals were weighed twice a week; body composition and fecal samples were assayed once a week.

Fig. 3 Cohousing results in the transfer of species from the microbiota of cagemates colonized with the healthy donor’s community into the microbiota of cagemates colonized with the severely stunted and underweight donor’s community, resulting in the prevention of growth impairments.

(A) Experimental design for the cohousing experiments. Dually housed 4.5-week-old mice were switched to the M8 diet and colonized 3 days later with intact uncultured microbiota from either the healthy or the stunted and underweight donor. Four days post-colonization (dpc), subsets of the mice were cohoused (HCH and UnCH), while control mice colonized with microbiota from the healthy or stunted and underweight donors remained in their original isolators and were paired with a new cagemate from that isolator (H-H and Un-Un controls). Fecal samples were collected throughout the experiment; growth was assayed by changes in total body weight and body composition (the latter by qMR). Mice were euthanized three weeks after colonization. (B) HCH and UnCH mice had significantly higher lean mass gain (indicated by the asterisks) relative to the Un-Un controls 15 days after colonization (Un-Un versus HCH, P = 0.0447; Un-Un versus UnCH, P = 0.0121; Mann-Whitney test; N = 6 cages of cohoused mice and 3 cages of each dually housed control group; two independent experiments). (C) Heat map showing the results of the invasion assay. To quantify invasion further, we used the mean and SD of the null distribution of invasion scores (defined as the scores from recipients of the H or Un donor’s microbiota that had never been cohoused with each other) to calculate a z value (standard score) and a Benjamini-Hochberg adjusted P value for the invasion score of each species in HCH and UnCH mice (see the supplementary materials). We defined a taxon as a successful invader if it (i) had a Benjamini-Hochberg adjusted P ≤ 0.05 and (ii) had a relative abundance of ≤0.05% before cohousing and ≥0.5% in the fecal microbiota at the time of euthanasia. Table S12 and (C) provide information about the direction and success of invasion. Each row represents a species-level taxon, and each column represents a mouse at a given day after colonization. The rows of the heat map were hierarchically clustered according to pairwise distances using Pearson correlation. Bars at the right side of each experimental arm represent the fold change (fc) in that species’ relative abundance before and after cohousing [fold change is defined as the log2 of the average relative abundance of the species after cohousing (days 7 through 22), divided by the average relative abundance of the species before cohousing (day 4)]. Species in red represent those identified as one of the top 30 growth-discriminatory taxa by the RF-based models of weight or lean mass gain shown in Fig. 2.

HCH and UnCH cagemates both gained significantly more lean mass than Un-Un controls did (UnCH, P = 0.0121; HCH, P = 0.0447; Mann-Whitney test; Fig. 3B) but had no significant difference in lean mass gain relative to H-H controls. We characterized invasion using a previously described approach (18) that uses Microbial SourceTracker [(19) and supplementary materials]. The fecal microbiota of H-H or Un-Un controls sampled 4 days after gavage (just before cohousing) were treated as source communities. The fecal communities belonging to each HCH and UnCH mouse were then traced to these sources (see the supplementary materials). Results indicated significant invasion by members of the healthy microbiota into the microbiota of UnCH cagemates (Fig. 3C and table S12). Nine OTUs from the HCH cagemates’ microbiota were consistent invaders (a total of 12 cohoused mice). Based on the rank order of their invasion scores, the two most successful invaders were F. prausnitzii, an age- and growth-discriminatory taxon in the RF models and the most abundant OTU in the fecal microbiota of UnCH mice at the conclusion of the cohousing experiment, and Ruminococcus gnavus, a growth-discriminatory taxon. In contrast, only two OTUs from the microbial community of the undernourished donor successfully invaded the gut community of HCH mice; one was assigned as Enterococcus and the other as Eubacterium limosum, which together represented 2.7 ± 1.7% of the community after cohousing [the Enterococcus OTU 4316566 had a significant negative correlation with lean mass gain in the initial screen of 19 donor microbiota (ρ = –0.14, P = 0.0065)].

Culturing discriminatory taxa and characterizing their effects on growth

Acquisition of HCH-derived bacterial taxa in the microbiota of UnCH cagemates was accompanied by reductions in the relative abundances of 19 OTUs in the latter, six of which were below the limits of detection (<0.01%) by the end of the cohousing experiment. These observations raised two questions. Which invasive OTUs mediated the observed effects on growth phenotypes? Did the reduction in other OTUs improve growth in UnCH cagemates? Therefore, we attempted to culture F. prausnitzii and R. gnavus from the fecal microbiota of healthy Malawian infants from the iLiNS-DYAD-M cohort. These efforts yielded three additional strains—Clostridium nexile (positively correlated with lean mass gain), Clostridium symbiosum (lean mass gain–discriminatory), and Dorea formicigenerans (weight gain– and lean mass gain–discriminatory)—making for a five-member consortium.

Analogous to the cohousing experiment, male germ-free C57BL/6J mice were placed on the M8 diet at 4.5 weeks of age; 3 days later, five mice were each given a single gavage with the intact microbiota from the Un donor, with or without the five-member consortium (Fig. 4A). The five-member consortium produced a significant increase in body weight and lean mass gain compared with the untreated group (P = 0.03 and P = 0.01, respectively, at the time of euthanasia 21 days after gavage; Mann-Whitney test) (Fig. 4B). Only two members of the consortium, R. gnavus strain TS8243C and C. symbiosum strain TS8243C, successfully colonized recipient mice [relative abundances of 27 ± 3% (mean ± SD) and 2.6 ± 0.3%, respectively, in feces obtained at euthanasia] (Fig. 4C; fig. S6 and table S13 show the results of reconstructions of selected metabolic subsystems in R. gnavus TS8243C and C. symbiosum TS8243C). None of the other members were detected in recipients sampled on days 4, 7, and 14 after gavage. Moreover, the efficiency of incorporation of members of the Un donor’s microbiota into recipient mice was indistinguishable between the two treatment arms; i.e., (i) R. gnavus and C. symbiosum did not result in extirpation of any major constituents of the community [the only species that fell below 0.1% of the community’s composition was Bifidobacterium bifidum, whose relative abundance was only 0.6 ± 0.1% (mean ± SD) in the control group at the time of euthanasia], and (ii) the proportional representation of OTUs in the Un community was similar between the experimental and control groups (Fig. 4C and table S14). These results provide direct evidence that the cultured strains of R. gnavus and C. symbiosum ameliorate the impaired growth phenotype transmitted by an undernourished donor’s immature microbiota.

Fig. 4 A consortium of cultured growth-discriminatory OTUs augments the growth of mice colonized with the 6-month-old severely stunted and underweight donor’s microbiota.

(A) Experimental design, including the composition of the five-member consortium of cultured bacterial strains. (B) Weight gain and lean mass gain (21 days after gavage) of mice colonized with the donor microbiota, with (treated) or without (control) the cultured consortium. (C) Comparison of the fecal microbiota of mice belonging to untreated control and treated experimental groups, showing the establishment of OTUs from the consortium at 21 days after colonization. (D) The effects of treatment with the consortium on host metabolism (P < 0.05 for all metabolites shown, determined by Student’s t test). Each row represents a metabolite from a given tissue, and each column represents an individual mouse. Tissues were collected 21 days after colonization.

The presence of these two organisms also affected metabolic phenotypes (Fig. 4D and table S15). The metabolic features that most discriminated mice with R. gnavus and C. symbiosum from the untreated group were acylcarnitines (C5 to C22), which were significantly increased in cecal samples and decreased in the liver and serum of fed mice harboring the two taxa (P < 0.05, Student’s t test). Some of these acylcarnitines also distinguished mice harboring the H versus the Un donor microbiota (fig. S7, A and B). Among metabolites that were decreased in the livers of treated mice were the C5, C5-DC, and C6-DC acylcarnitines, all of which can be derived from branched-chain amino acid catabolism. As proposed for the comparison of mice harboring the H versus the Un donor microbiota (see the supplementary materials), the decrease in these liver metabolites suggests an impact of the presence of two growth-promoting taxa on host metabolic machinery that drives amino acids away from oxidation in favor of protein synthesis and lean mass formation. The mechanisms by which the gut microbiota communicates metabolically with other tissues remain to be defined.

Acknowledging that the generalizability of these effects needs to be directly tested using additional microbiota from other undernourished infants and children, we proceeded to assess whether the growth-discriminatory taxa identified in our preclinical model correlate with growth in Malawian infants and children. To do this, we used the 220 samples from the healthy Malawian twin cohort to generate two RF-based models. One regressed the 30 most weight gain–discriminatory OTUs shown in Fig. 2D against WHZ; the other regressed the 30 most lean mass gain–discriminatory taxa described in fig. S2 against this metric. We then evaluated how well the models were able to predict WHZ scores in the 259 members of the iLiNS-DYAD-M cohort across all time points surveyed (6, 12, and 18 months old for each individual). During this brief 12-month window, there were significant and positive correlations between predicted and observed WHZ using either the weight gain– or lean mass gain–discriminatory taxa (weight gain–discriminatory model, ρ = 0.12 and P < 0.0008; lean mass gain–discriminatory model, ρ = 0.11 and P < 0.002; P < 0.0001 for both models in a permutation test with 999 permutations). In both models, B. longum was the most discriminatory taxon, with F. prausnitzii OTUs following as the second or third most discriminatory in the weight gain and lean mass gain models, respectively (fig. S10). The increasing ability to assemble the genomes of bacterial strains from shotgun sequencing of fecal community DNA (20) should allow these types of analyses to be further developed in Malawian and other populations that are surveyed longitudinally at frequent intervals over extended periods of time. This approach would also provide direct information about the representation of the strains that we cultured and subsequently characterized in our preclinical model.


The studies reported here indicate that gut microbiota immaturity is not only associated with undernutrition but causally related to it. There are several interrelated factors that could disrupt normal gut microbiota succession in infants and children. They include, but are not limited to, (i) poor maternal nutritional status, (ii) enteropathogen invasion, (iii) the history of consumption of antibiotics, (iv) disturbances in gut mucosal immune system development (21), and/or (v) the history of complementary feeding. Phenotyped birth cohorts provide an opportunity to perform correlation analyses designed to test the significance of the relationship between microbiota maturity (and the representation of specific age- and growth-discriminatory taxa), anthropometry, and the factors listed above, as well as the enigmatic disorder currently described as environmental enteric dysfunction (22).

In this study, we showed that microbiota from 6-month-old donors produced greater effects on growth in recently weaned mice than microbiota from 18-month-old donors, although overall, the effects produced by a healthy donor’s age-appropriate community were greater than those produced by an undernourished donor’s immature microbiota. Though it will be important to extend these analyses to microbiota sampled from additional individuals representing this and other geographic sites and cultural traditions, these findings suggest that in healthy children, microbiota development is optimized to satisfy the different growth needs of the host at different ages. An immature microbiota appears to cause a form of neoteny that is not conducive to healthy growth when nutrients are limiting. How will a host adapt to therapeutic interventions that result in rapid progression to an age-appropriate microbiota? Are there mechanisms (including those involving the mucosal immune system) that feed back to regulate the rate of microbiota development? The answers have implications for designing therapeutic strategies for durable repair and prevention of stunting and neurodevelopmental and/or immunologic abnormalities associated with undernutrition.

So far, ready-to-use complementary foods with and without antibiotics have generally produced only modest effects on growth and the longer-term sequelae of undernutrition (23). Certain locally available complementary foods that are provided after the cessation of exclusive breastfeeding may have the ability to promote colonization of growth-discriminatory gut taxa in proportions that are age-appropriate, and these could be tested for their clinical value in systematic trials. In this respect, gnotobiotic mice colonized with microbiota from chronologically age-matched healthy and undernourished donors, and fed diets representative of those consumed by the microbiota donors, should permit highly controlled direct tests of the effects of antibiotics, breast milk components (e.g., bovine mimics of human breast oligosaccharides), and complementary foods on the mechanisms by which growth-promoting bacterial strains influence growth, metabolic, bone, immune, and neurologic phenotypes.

Supplementary Materials

Materials and Methods

Supplementary Text

Figs. S1 to S10

Tables S1 to S17

References (2441)


ACKNOWLEDGMENTS: We are indebted to the parents and children from Malawi for their participation in this study. We thank S. Wagoner, D. O’Donnell, M. Karlsson, and J. Serugo for their assistance with gnotobiotic mouse husbandry; D. Leib and M. Silva for their assistance with bone morphology assays; J. Guruge for help with anaerobic microbiology; B. Dankenbring for assistance with maintenance of the biospecimen repository; and M. Meier, S. Deng, and J. Hoisington-Lopez for their contribution to various facets of the DNA sequencing pipeline. This work was supported by the Bill and Melinda Gates Foundation and the NIH (grant DK30292). Additional funding was provided by the Office of Health, Infectious Diseases and Nutrition, Bureau for Global Health, U.S. Agency for International Development under the terms of cooperative agreement no. AID-OAA-A-12-00005, through the Food and Nutrition Technical Assistance III Project managed by FHI 360. Data management and statistical analysis for theiLiNS-DYAD-M clinical study were also funded by the Academy of Finland (grant 252075) and the Medical Research Fund of Tampere University Hospital (grant 9M004). Micro-CT analysis was performed in the Washington University Musculoskeletal Research Center, which is supported by NIH grant P30 AR057235. L.V.B. received stipend support from NIH predoctoral training grants NIH T32 AI007172 and T32 GM007067 and from the Lucille P. Markey Special Emphasis Pathway in Human Pathobiology. D.A.R. and S.A.L. were supported by the Russian Science Foundation (grant 14-14-00289). Collection of the human specimens included in this study was approved by the University of Malawi College of Medicine Research Ethics Committee. Specimens were provided to Washington University in St. Louis under a materials transfer agreement (MTA) between the University of Malawi and Washington University, and their collection and use for this study was approved by the Washington University Human Research Protection Office (Federalwide Assurance no. FWA00002284). 16S rRNA sequences, generated from fecal samples in raw format before post-processing and data analysis, and shotgun sequencing data sets, generated from the R. gnavus TS8243C and C. symbiosum TS8243C genomes, have been deposited in the European Nucleotide Archive under accession number PRJEB9853. J.I.G. is a cofounder of Matatu, a company characterizing the role of diet-by-microbiota interactions in animal health. A.L.O. is an adjunct vice president for research for Buffalo BioLabs. L.V.B., M.R.C., and J.I.G. designed the gnotobiotic mouse studies; L.V.B. and M.R.C. performed the experiments with gnotobiotic animals; I.T. and M.J.M. designed and implemented the clinical monitoring and sampling for the twin study and participated in patient recruitment, sample collection and preservation, and/or clinical evaluations; K.M.M., Y.F., J.M.J., K.G.D., and P.A. designed and oversaw the clinical studies, sample collection and processing, and/or clinical monitoring and evaluations in the iLiNS-DYAD-M study; L.V.B. generated the 16S rRNA data; L.V.B., M.R.C., S.V., and O.I. generated the metabolomics data; T.S. and L.V.B. cultured bacterial isolates; B.H., D.A.R., S.A.L., and A.L.O. performed metabolic reconstructions of the R. gnavus and C. symbiosum genomes; L.V.B., M.R.C., M.J.B., S.S., C.B.N., and J.I.G. analyzed the data; and L.V.B. and J.I.G. wrote the paper.

Stay Connected to Science

Navigate This Article