Bifidobacterium longum modifies a nutritional intervention for stunting in Zimbabwean infants

Ethan K Gough, Thaddeus J Edens, Lynnea Carr, Ruairi C Robertson, Kuda Mutasa, Robert Ntozini, Bernard Chasekwa, Hyun Min Geum, Iman Baharmand, Sandeep K Gill, Batsirai Mutasa, Mduduzi N N Mbuya, Florence D Majo, Naume Tavengwa, Freddy Francis, Joice Tome, Ceri Evans, Margaret Kosek, Andrew J Prendergast, Amee R Manges, the Sanitation Hygiene Infant Nutrition Efficacy (SHINE) Trial Team
doi: https://doi.org/10.1101/2024.01.18.24301438

Summary

Child stunting is an indicator of chronic undernutrition and reduced human capital. Small-quantity lipid-based nutrient supplements (SQ-LNS) has been widely tested to reduce stunting, but has modest effects. The infant intestinal microbiome may contribute to stunting, and is partly shaped by mother and infant histo-blood group antigens (HBGA). We investigated whether mother-infant fucosyltransferase status, which governs HBGA, and the infant gut microbiome modified the impact of SQ-LNS on stunting at age 18 months among Zimbabwean infants in the SHINE Trial (NCT01824940). We found that mother-infant fucosyltransferase discordance and Bifidobacterium longum modified SQ-LNS efficacy. Infant age-related microbiome shifts in B. longum subspecies dominance from infantis, a proficient human milk oligosaccharide utilizer, to suis or longum, proficient plant-polysaccharide utilizers, were partly influenced by discordance in mother-infant FUT2+/FUT3-phenotype, suggesting that a “younger” microbiome at initiation of SQ-LNS reduces its benefits on stunting in areas with a high prevalence of linear growth restriction.

Introduction

Globally, 21% of children under 5 years of age (149 million) are stunted1, defined as having a length or height >2 standard deviations below an age– and sex-matched reference population median2. Deficits in linear growth largely accrue from conception to 24 months of age3,4, corresponding to the period when normal child growth and development is most rapid. Stunting is associated with reductions in child survival, neurodevelopment, educational attainment, and adult economic productivity5–7.

A broadly tested intervention to reduce stunting has been provision of small-quantity lipid-based nutrient supplements (SQ-LNS) to improve infant and young child feeding (IYCF) starting at 6-months (mo) of age, which is the recommended time for introduction of complementary foods. Randomized controlled trials (RCT) of SQ-LNS provision to infants have shown small reductions in stunting (12% relative reduction), but effects have varied8,9. Evidence to support other nutrition interventions, which address the underlying determinants of stunting during the first 1000 days of life, starting from conception, is also limited10. Elucidating the reasons for the limited impact of nutritional interventions that are designed to alleviate stunting is crucial to the development of more effective strategies.

Recently, the human microbiome has been shown to impact infant health11, and studies suggest a role of the intestinal microbiome in child growth, particularly ponderal growth12,13. The prevailing view holds that intestinal colonization with bacteria begins at birth, after which the microbiome progresses through a succession of identifiable shifts in composition and functional capacity11 that correspond closely with infant age and developmental stage14. Deviations from an age-appropriate composition may be associated with poor health outcomes14–17, including undernutrition in low resource settings14,18–20. However, results vary21–25 and evidence for a causal effect of the intestinal microbiome on growth comes predominantly from animal models. In these experiments, a combination of commensal bacteria and a nutrient-poor diet produce a synergistic effect on growth faltering. The detrimental effect on growth was worse when certain commensal bacteria, which can act as pathobionts (e.g. Bacteroides thetaiotaomicron and Bacteroides fragilis), were present in the gut along with pathogenic species26,27.

The gut microbiome in infants is influenced by breastfeeding, with differences persisting beyond age 6mo16. Breastfeeding-associated differences are partly driven by differences in human milk oligosaccharide (HMO) composition, which are metabolized by specific commensal bacteria and thereby influence the growth and activity of bacterial populations in the infant gut, in particular, Bifidobacterium28,29. In addition, HMOs in combination with commensal gut bacteria have been shown to improve growth in animal models of undernutrition30. Active maternal α-1,2-fucosyltransferase (FUT2) and α-1,3-fucosyltransferase (FUT3) genes are key determinants of HMO composition31–33. These genes encode enzymes which catalyze addition of fucose to the disaccharides that serve as precursors to host glycan production34,35. Individuals with at least one functional FUT2 or FUT3 allele produce fucosylated histo-blood group antigens (HBGA) and are called “secretors” or “Lewis-positive”, respectively;35 by contrast, individuals lacking two functional FUT2 or FUT3 alleles don’t produce fucosylated HBGAs, and are termed “non-secretors” or “Lewis-null”, respectively35. These secretor and Lewis phenotypes act in concert to synthesize a variety of HBGAs (Figure S1)34,35. In addition to the impact of maternal FUT2 or FUT3 status on the infant gut microbiome via their influence on HMO composition, host FUT2 and FUT3 phenotypes also determine tissue surface HBGA expression in the gut, which may also affect the microbiome through the availability host glycans and competition for adhesion sites36–40.

There is growing interest in the moderating effect of the gut microbiome on nutritional interventions. Recent RCTs showed that the impact of dietary interventions for weight reduction on metabolic health outcomes is modified up to 4-fold by microbiota composition41–44. Effect modification of host diet by the gut microbiota has also been investigated in observational studies, which report that microbiome composition modified the association between diet and biomarkers of metabolic syndrome by up to 2-fold45–47. Gut microbiota composition, therefore, may also be an important modifier of dietary interventions on infant health. Only one study to date investigated the gut microbiota as an effect modifier of SQ-LNS on infant stunting, and reported limited evidence of effect-modification48. However, several studies have reported a synergistic effect of microbiome composition and diet on undernutrition phenotypes using animal models26,27,49,50.

Here we aimed to determine the effect of infant gut microbiome composition and functional capacity on efficacy of an IYCF intervention that included SQ-LNS to reduce stunting (primary outcome) and improve length-for-age z-score (LAZ) (secondary outcome) at age 18mo. We hypothesized that the efficacy of IYCF, when started at age 6mo, on infant linear growth is modified by mother and infant FUT2 and FUT3 phenotype, as an drivers of variation in gut microbiome composition, and by the infant gut microbiome. We tested this hypothesis using data from HIV-unexposed infants enrolled in the Sanitation Hygiene Infant Nutrition Efficacy (SHINE) trial conducted in rural Zimbabwe, in which the IYCF intervention modestly increased LAZ by 0.16 standard deviations and reduced stunting by 20% at age 18mo51. We found that mother-infant FUT2 and FUT3 phenotype, where the mother was FUT2+/FUT3-but the infant was not, was associated with greater IYCF reduction of stunting at 18mo. Greater impact of IYCF on stunting was also associated with infant microbiome species maturation, characterized by a shift away from Bifidobacterium longum carriage. B. longum-dominant microbiome composition was characterized by B. longum strains that were most similar to subspecies infantis, a proficient HMO utilizer. In addition, B. longum abundance and odds of detecting different B. longum strains were explained by infant age and differences between mother-infant pairs in FUT2+/FUT3-phenotype.

Results

Mother-infant FUT2+/FUT3-phenotype discordance modifies the effect of IYCF on stunting at 18mo

In a substudy of the SHINE trial, we assessed FUT2 and FUT3 status in mothers and infants from saliva samples. To determine the impact of maternal and infant HBGA on IYCF efficacy, we investigated whether concordance in secretor (FUT2) and Lewis (FUT3) phenotypes between mother and infant pairs modified the effect of IYCF on stunting or LAZ at 18mo. We defined these paired mother-infant phenotypes using combinations of FUT2 and FUT3 status (as presented in Table S1) to more precisely reflect the potential for joint activity of these enzymes in HBGA production (as presented in Figure S1). Each of these paired mother-infant FUT2/FUT3 phenotypes were classified as both (if mother and infant shared the same phenotype), none (if neither had the phenotype), infant only (if the infant had the phenotype but the mother did not) or mother only (if the infant did not have the phenotype but the mother did). We fitted multivariable regression models that included terms for the interaction of IYCF and each paired FUT2/FUT3 phenotype between mother and infant, as well as prespecified covariates from the 6mo follow-up visit when IYCF was started, with stunting status at the 18mo visit as the dependent variable. We fitted separate models for each combination of paired mother-infant FUT2/FUT3 phenotype presented in Table S1 and Figure S1. These models were restricted to 792 infants who had both maternal and infant FUT2 and FUT3 status ascertained (Figure S2). Analyses were repeated with LAZ at the 18mo visit as the dependent variable.

Amongst infants who were randomized to IYCF, those who were in the mother only FUT2+/FUT3-group (13.5%) had a probability of stunting at 18mo that was –33.0% lower (95%CI:-55.0%,-10.0%) than infants in the both FUT2+/FUT3-group (11.1%) (Table 1). The infant only (11.1%) and none (64.3%) FUT2+/FUT3-groups, and other mother-infant FUT2 and FUT3 phenotype combinations (Table S1, Figure S1), did not show evidence of effect modification of IYCF on stunting (Table S2), and there was no evidence of effect modification on LAZ at 18mo after FDR-adjustment for multiple testing (Table S3). Overall, our results indicate that discordance between mothers and infants in the FUT2+/FUT3-phenotype, whereby mothers had the phenotype and infants didn’t have the phenotype, was associated with increased efficacy of IYCF to reduce stunting but not to increase LAZ.

View this table:
Table 1. Multivariable regression model1 to estimate modification of IYCF on stunting at 18mo by mother-infant Lewis-null secretor phenotype discordance among 792 infants in whom mother and infant FUT2 and FUT3 status was ascertained

Infant gut microbiome composition modifies the effect of the IYCF intervention on stunting at 18mo

Given our finding that infants whose mothers were FUT2+/FUT3-, while the infant was not, had greater reductions in stunting by IYCF; and considering the potential importance of FUT2/FUT3 phenotypes on early infant microbiome composition via variation in maternal HMO composition and infant gut epithelial HBGA expression, we investigated whether infant gut microbiota species maturation showed evidence for modification of IYCF efficacy on stunting or LAZ at 18mo. First, we used 354 infant metagenomes from 172 infants collected from 1-18mo of age to fit a constrained Principal Coordinates Analysis (PCoA) model that included infant age at stool collection and three dummy variables for mother-infant FUT2+/FUT3-status representing the none, infant only, and mother only groups with the both group as the referent (see next section for details). Thus, we derived four PCoA axis scores from this model representing variation in microbiome composition due to infant age and explained by none, infant only, or mother only FUT2+/FUT3-status compared to both. Next, we fitted a multivariable regression model that included an IYCF-by-PCoA axis 1 score interaction term and prespecified covariates. The regression model was restricted to 53 infants who had a fecal specimen collected at the 6mo follow-up visit when IYCF started and used their covariate values at the 6mo visit (Figure S2). Stunting status at 18mo was used as the dependent variable. We repeated our analyses using PCoA axis 2 to 4 scores. Finally, we fitted regression models replacing stunting status with LAZ at 18mo as the dependent variable.

Greater PCoA axis 1 scores represented species turnover with increasing infant age (Figure S3), thus reflecting microbiota maturation, while greater PCoA axis 2 scores represented changes in species composition associated with mother-infant FUT2+/FUT3-phenotypes (Figure S3). PCoA axis 1 and 2 scores showed evidence of interaction with IYCF on stunting at 18mo (Table 2), but not LAZ (Table S4). Amongst infants randomized to IYCF, those with higher PCoA axis 1 scores at age 6mo were less likely to be stunted at 18mo compared to infants with lower PCoA axis 1 scores (difference-in-differences –76.0% [95%CI: –99.0%,-32.0%, adjusted p=0.003) (Table 2). In contrast, infants randomized to IYCF with higher PCoA axis 2 scores at age 6mo were more likely to be stunted at 18mo compared to infants who had lower PCoA axis 2 scores (difference-in-differences, 14.0% [95%CI: 7.0%,21.0%], adjusted p=0.001) (Table 2). Taken together, our findings showed that infant age-related microbiome species maturation was associated with greater reduction in the probability of stunting at 18mo by IYCF, indicating greater efficacy of the intervention; while microbiome species composition related to mother-infant FUT2+/FUT3-phenotype was associated with lesser reduction in stunting, suggesting variation in microbiome composition due to these infant characteristics is an important determinant of SQ-LNS efficacy to reduce stunting.

View this table:
Table 2. Multivariable regression models1 to estimate modification of IYCF on stunting at 18mo by infant gut microbiome species turnover in 53 infants

Infant age and mother-infant FUT2+/FUT3-phenotype explain shifts in microbiome species composition, in particular, Bifidobacterium longum

To investigate sources of variation in infant intestinal microbiome composition and derive interpretable measures of species turnover, we performed constrained PCoA of Bray-Curtis dissimilarities with permutational analysis of variance using distance matrices for hypothesis testing. Infant age at stool collection explained 12.6% (ADONIS2 p=0.001) of the variability in microbiome composition (Table S5). Notably, exclusive breastfeeding (EBF) at 3mo was not significantly associated (R2=0.346, ADONIS2 p=0.483) with microbiome composition (Table S5). However, 93.2% of mothers in these analyses reported exclusive breastfeeding (Table S6). Thus, there may not have been sufficient variability in infant breastfeeding practices in this cohort to identify EBF-associated differences between infant microbiomes. Also, inclusion of specimens collected throughout the follow-up period in our PCoA models may have obscured associations with EBF, since EBF is only recommended up to age 6mo.

We then fitted a fully adjusted multivariable constrained PCoA model that included both infant age at stool collection and mother-infant FUT2+/FUT3-phenotype. We retained the latter variable to capture variation in microbiota composition associated with mother-infant FUT2+/FUT3-status after controlling for infant age, because of our a priori aim to investigate whether microbiome variation due to mother and infant FUT2/FUT3 phenotype was a modifier of the IYCF intervention, and our finding that the mother only FUT2+/FUT3-group was associated with a modified intervention effect. In the full model, age explained 12.8% (ADONIS p=0.001) and mother-infant FUT2+/FUT3-phenotype explained 0.81% (ADONIS p=0.323) of the variation in microbiome composition (Table S5), indicating that infant age was the main driver of microbiome-wide changes in species composition in this cohort.

Age-related maturation in species composition were characterized by decreased relative abundance of B. longum, and increased relative abundance of Prevotella copri, Faecalibacterium prausnitzii, Dorea longicatena, and Dorea formicigenerans (Figure S4). Mother-infant FUT2+/FUT3-discordance was characterized by increased relative abundance of B. longum in the none and infant only groups (Figure S5-S7).

In combination with our IYCF-by-PCoA effect modification models (see previous section), these findings indicate that infant microbiota species maturation characterized by a shift away from a microbiome dominated by B. longum (i.e. greater PCoA axis 1 score) was associated with a reduction in the probability of stunting at age 18mo in infants randomized to IYCF (Table 2). In contrast, species composition associated with mother-infant FUT2+/FUT3-phenotype (i.e. greater PCoA axis 2 score), which was characterized by higher B. longum abundance, showed evidence of reduced impact of IYCF on stunting (Table 2).

Bifidobacterium longum relative abundance modifies the effect of the IYCF intervention on stunting at 18mo

To determine if differences in the relative abundances of specific microbiome species that characterized the constrained PCoA axes were also associated with significant modification of IYCF on stunting or LAZ at 18mo, we fitted multivariable regression models that included IYCF-by-species relative abundance interaction terms and prespecified covariates as described in the previous sections. These models were restricted to 53 infants who had a fecal specimen collected at the 6mo follow-up visit when IYCF started and used their covariate values at that visit (Figure S2). We fitted a separate model for each microbiome species of interest, defined as those strongly associated with PCoA axis 1 to 4 scores (i.e. loadings > 0.5 or < –0.5). Infants randomized to IYCF who had greater relative abundance of B. longum when the intervention started were more likely to be stunted at age 18mo compared to infants with lower relative abundance of B. longum (differences-in-differences 50.0% [95%CI: 26.0%,74.0%], adjusted p<0.001) (Table 3). No other species-level interactions with IYCF on stunting were identified (Table 3). Infants randomized to IYCF with greater B. longum relative abundance also had smaller LAZ at 18mo compared to infants with lower B. longum abundance; however, this association was not statistically significant after FDR-adjustment (differences-in-differences –0.88[95%CI: –1.45,-0.31], adjusted p=0.074) (Table S7). Overall, our findings indicate that greater relative abundance of B. longum in the infant gut microbiome at initiation of the IYCF intervention was associated with less efficacy of the IYCF intervention to reduce stunting at age 18mo.

View this table:
Table 3. Multivariable regression models1 to estimate modification of IYCF on stunting at 18mo by infant gut microbiome species in 53 infants

Infants randomized to IYCF in the mother only FUT2+/FUT3-group or with lower B. longum relative abundance were spared from more severe growth faltering

To investigate whether these reductions in IYCF efficacy were due to differences in the degree or timing of growth faltering, we plotted (i) LAZ growth trajectories and (ii) LAZ velocities (sd/mo) from 6-18mo of age, by both IYCF arm and paired mother-infant FUT2+/FUT3-phenotype. Infants randomized to IYCF in the mother only group had a less steep decline in LAZ trajectories on average compared to infants in the mother only group randomized to no IYCF (Figure 1). In addition, although all groups of infants had negative 6-18mo LAZ velocities on average, infants in the mother only group who were randomized to IYCF had higher LAZ velocity (−0.02 sd/mo,95%CI[-0.05,0.00]) compared to those in the no IYCF group (−0.08 sd/mo, 95%CI[-0.11,-0.05]) (Figure 1). Importantly, infants in the mother only group also benefitted more from the IYCF intervention in our interaction models (Table 1).

We repeated these exploratory analyses to compare LAZ trajectories and LAZ velocities by both IYCF arm and B. longum at the 6mo visit stratified above or below the relative abundance median. Infants randomized to IYCF who were above the median relative abundance of B. longum had declining growth trajectories on average, while infants below the median did not (Figure 2). Furthermore, the mean 6-18mo LAZ velocity was negative in both B. longum groups, but trended toward being higher in the group with relative abundance below (−0.02 sd/mo,95%CI[-0.07,0.03]) compared to above (−0.08 sd/mo 95%CI[-0.12,-0.03]) the median; however, the 95% confidence intervals overlapped. In contrast, among infants randomized to no IYCF, both those with a relative abundance above and those with a relative abundance below the median at the 6mo visit had declining growth trajectories (Figure 2) with similar average negative LAZ velocities (Figure 2).

Figure 1.
Figure 1. Infant growth trajectories and growth velocities vary by IYCF arm and mother-infant FUT2+/FUT3-phenotype.

(A) LAZ by infant age from 1-18mo (n=792), stratified by IYCF arm (light grey, no IYCF; dark grey, IYCF) and mother-infant FUT2+/FUT3-phenotype (both, none, infant only or mother only). Lines illustrate average trajectories. Shaded areas are 95% confidence bands. Infants randomized to IYCF in the mother only group had a less steep decline in LAZ trajectories on average compared to infants randomized to no IYCF.

(B) Violin plots of LAZ velocity from 6-18mo of age, stratified by IYCF arm (light grey, no IYCF; dark grey, IYCF) and mother-infant FUT2+/FUT3-phenotype (both, none, infant only or mother only). Open circles with error bars in the center of each panel indicate mean LAZ velocity and 95%CIs. All groups of infants had negative LAZ velocities on average, but infants in the mother only group who were randomized to IYCF had higher LAZ velocity on average, contributing to a less steep decline in LAZ over time and a smaller proportion of infants stunted at 18mo.

Figure 2.
Figure 2. Infant growth trajectories and growth velocities vary by IYCF arm and infant gut Bifidobacterium longum relative abundance.

(A) LAZ by infant age from 1-18mo (n=53), stratified by IYCF arm and Bifidobacterium longum relative abundance (light grey, ≤median relative abundance; dark grey, >median relative abundance). Thin lines illustrate individual trajectories. Thick lines show average trajectories. Infants randomized to IYCF who were above the median relative abundance had a declining growth trajectory on average, while infants below the median did not.

(B) Violin plots of LAZ velocity from 6-18mo of age, stratified by IYCF arm (light grey, no IYCF; dark grey, IYCF) and mother-infant FUT2+/FUT3-phenotype (both, none, infant only or mother only). Open circles with error bars indicate mean LAZ velocity and 95%CIs. All groups of infants had negative LAZ velocities on average. However, LAZ velocity trended toward being higher in the group with relative abundance below the median. The 95% confidence intervals overlap, but the higher LAZ velocity contributed to a less steep decline in LAZ over time and a smaller proportion of stunted infants at 18mo.

Overall, these results indicate that the differences in IYCF efficacy by mother-infant FUT2+/FUT3-phenotype and B. longum relative abundance can be explained by differences in LAZ velocity. Infants in the mother only FUT2+/FUT3-group and infants with lower B. longum relative abundance at 6mo were spared from more severe LAZ declines, which contributed to a lower probability of stunting at 18mo, even when the differences in velocity were not statistically significant.

Bifidobacterium longum strains most similar to subspecies infantis dominate the early infant gut microbiome

Next, we aimed to identify whether infants carried different strains of B. longum using a pangenome approach. We used PanPhlan3.052 to generate UniProt gene family profiles of dominant B. longum strains from 218 infant gut metagenomes, generated from 136 infant, with sufficient coverage of B. longum (Figure S2). To assess similarities between SHINE B. longum strains and previously characterized subspecies, we also included UniProt gene family profiles produced by PanPhlan for 118 reference strains. We performed PCoA using Jaccard dissimilarities calculated from these gene family presence or absence profiles. Three clusters of B. longum strains were identifiable by visualization of ordination plots (Figure 3 and S8). We, therefore, performed hierarchical clustering of these Jaccard dissimilarities to delineate three distinct strain clusters and place each strain into the most appropriate cluster (Figure S8). Subspecies infantis reference strains predominantly grouped with the largest cluster (hereafter called the infantis cluster) and included 15 infantis reference strains and 255 SHINE strains (Figure 3). Subspecies longum reference strains predominantly grouped with the second largest cluster (longum cluster), including 29 longum, 3 infantis, 64 unclassified subspecies strains and 14 SHINE strains. The remaining cluster included two subspecies suis, one subspecies longum, 4 unclassified subspecies reference strains and 15 SHINE strains (suis cluster) (Figure 3 and Figure S8).

Figure 3.
Figure 3. Bifidobacterium longum strains most similar to subspecies infantis dominate the early infant gut microbiome.

Ordination plots of PCoA with Jaccard dissimilarities calculated using the pangenome profiles of dominant Bifidobacterium longum strains in fecal specimens (N=284) generated using PanPhlan3.0. Three clusters of B. longum strains were identifiable. Individual strains and indicated by small circles. Those in the same cluster are enclosed by a large ellipse and are differentiated by both the color of the ellipse and the border color of the small circle (blue, infantis; dark green, longum; red, suis). Filled small circles indicate reference strains (light blue, infantis; pink, suis, light green, longum; grey, unclassified). Open small circles indicate SHINE strains. Subspecies infantis reference strains predominantly grouped with the largest cluster (blue ellipse, infantis cluster). Subspecies longum reference strains predominantly grouped with the second largest cluster (dark green ellipse, longum cluster). The remaining cluster (red ellipse, suis cluster) grouped with subspecies suis and longum.

Infant gut microbiome maturation undergoes a transition in which Bifidobacterium longum strains most similar to subspecies infantis become less prevalent

To investigate predictors of B. longum relative abundance in infants over time, we fitted a longitudinal multivariable zero-inflated mixed-effects model, using 225 specimens from 87 infants with prespecified covariate data. Covariates included infant age at specimen collection, sex, EBF at 3mo, minimum infant dietary diversity at specimen collection, and mother-infant FUT2+/FUT3-phenotypes (Figure S2). Infant age and mother-infant FUT2+/FUT3-phenotype were significant predictors of B. longum relative abundance. Each one month increase in infant age was associated with a 0.91-fold decrease (95%CI:0.90,0.92, p<0.001) in B. longum, and the mother only FUT2+/FUT3-group had the lowest relative abundance of B. longum throughout follow-up (Figure S11) with a 0.71-fold decrease in relative abundance (95%CI:0.59,0.87, p=0.001) relative to the both group (Table S9).

Also, to determine predictors of B. longum strain cluster detection in infants over time, we fitted multivariable longitudinal logistic regression models using 133 specimens from 70 infants with PanPhlan3.0 output and prespecified covariate data (Figure S2). We fitted a separate model for each strain to estimate its probability of detection. Results were consistent with predictors of B. longum relative abundance. Each one month increase in infant age was associated with a 0.85-fold decreased odds (95%CI:0.75,0.96, p=0.008) of the infantis cluster (Table S10). At the same time, the suis and longum clusters increased with age (Table S10 & Figure S12). Female infants had 8.35-fold increased odds of the infantis cluster (95%CI:3.17,21.97, p<0.001); and the infant only FUT2+/FUT3-group, which had the highest probability of the infantis cluster throughout follow-up (Figure S13A), had a 4.66-fold increased odds (95%CI:1.05,20.71, p=0.043) of the infantis cluster (Table S10). In contrast, the mother only FUT2+/FUT3-group had low probability of carrying an infantis cluster strain (Figure S13A & Table S10) and the highest probability of carrying a suis cluster strain (Figure S13C).

In summary, B. longum decreased with infant age and was lowest among infants in the mother only FUT2+/FUT3-group. The infantis cluster, which included the dominant B. longum strains, also decreased with infant age. Furthermore, infantis cluster strains were most likely to be detected in the infant only group, and were less likely to be detected in the mother only group, among whom suis cluster strains were more likely to be detected. Infant age and mother-infant FUT2+/FUT3-phenotype were important determinants of both B. longum relative abundance and strain carriage.

Bifidobacterium longum strains most similar to subspecies infantis were characterized by greater capacity for HMO degradation, uptake, siderophore and antimicrobial biosynthesis

To identify differences in metabolic potential between B. longum strain clusters, we used two-sided Fisher’s Exact tests to compare UniProt gene family presence between clusters. We restricted these analyses to the 284 SHINE Bifidobacterium pangenomes to make inferences about differences in metabolic potential between SHINE strains. We then performed overrepresentation analyses53 of differentially frequent gene families by one-sided Fisher’s Exact test to determine whether they were more likely to be involved in specific GO biological processes54, or to function as specific carbohydrate-active enzymes (CAZymes)55 or transporters56. Gene families that were more common in the infantis cluster were more likely to be involved in HMO degradation, and included genes that function as CAZyme Glycoside Hydrolase Family 20 (GH20) (β-N-acetylglucosaminidases, β-N-acetylgalactosamindase, β-6-SO3-N-acetylglucosaminidases, and lacto-N-biosidases), GH29 (fucosidases), GH95 (fucosidases), and GH33 (sialidases) (Figure 4A, Table S8). Conversely, gene families that were more common in the other clusters were more likely to be involved in degradation of plant-derived polysaccharides, including genes that function as GH42 (β-galactosidases), GH51 (L-arabinfuranosidases) and GH127 (β-L-arabinofuranosidase) CAZymes (Figure 4A, Table S8). Similarly, gene families that were more common in the infantis cluster were more likely to function as transporters involved in oligosaccharide uptake (TCID 3.A.1.1.59), while those that were less common in the infantis cluster were more likely to be involved in uptake of fructose and other sugars (TCID 3.A.1.2.23) (Figure 4A, Table S8). Other differences between strain clusters included greater carriage of gene families among infantis cluster strains that are involved in riboflavin biosynthesis, signal transduction, amino acid catabolism and uptake.

Figure 4.
Figure 4. SHINE infant Bifidobacterium longum infantis cluster strains are better adapted for human milk oligosaccharide utilization and protection against pathogens than suis and longum cluster strains.

(A) Heatmap of UniProt gene family presence in SHINE infant Bifidobacterium longum strains (N=284). UniProt gene family presence is indicated in grey. The horizontal bar at the top indicates strain cluster (blue, infantis; dark green, longum; red, suis). Vertical bars from left to right indicate: UniProt gene families that differed in prevalence between the infantis and longum cluster by two-sided Fisher’s Exact test after FDR correction (red); UniProt gene families that differed in prevalence between the infantis and suis cluster by two-sided Fisher’s Exact test after FDR correction (red); Biological Process GO groups; CAZymes; and Transporter class. Only gene families with evidence of overrepresentation in a Biological Process, CAZyme or Transporter class by one-sided Fisher’s Exact Test are presented.

(B) Heatmap of MetaCyc pathway presence in SHINE infant Bifidobacterium longum strains (N=284). MetaCyc pathway presence is indicated in grey. The horizontal bar at the top indicates strain cluster (blue, infantis; dark green, longum; red, suis). Vertical bars from left to right indicate: MetaCyc pathways that differed in prevalence between the infantis and longum cluster by two-sided Fisher’s Exact test after FDR correction (red); MetaCyc pathways that differed in prevalence between the infantis and suis cluster by two-sided Fisher’s Exact test after FDR correction (red); MetaCyc pathway type. Only pathways with evidence of overrepresentation in a pathway type by one-sided Fisher’s Exact Test are presented.

We also identified differences between strain clusters in the frequency of 115 MetaCyc57 pathways identified using MinPath58 (Figure S10). Pathways that were more common in the infantis cluster were more likely to be involved in generation of precursor metabolites and energy, and secondary metabolite biosynthesis (which predominantly included pathways for biosynthesis of siderophores and antimicrobials) (Figure 4B, Table S8). Conversely, the infantis cluster was less likely to include pathways involved in glycan metabolism and polymeric compound degradation (e.g. pectin, xylan, and arabinogalactan degradation pathways), and in fatty acid and lipid biosynthesis (Figure 4B, Table S8).

Overall, infant gut microbiomes were dominated by B. longum strains that were most similar to subspecies infantis in their UniProt gene family profiles. The primary distinction of B. longum strains in the infantis cluster was greater capacity for HMO degradation and uptake of oligosaccharides than plant-derived polysaccharides. However, strains in the infantis cluster also had greater capacity for siderophore and antimicrobial production, and displayed differences in their capacity for riboflavin, fatty acid, and lipid metabolism.

Discussion

In this study of HIV-unexposed infants enrolled in the SHINE trial in rural Zimbabwe, we tested the hypothesis that mother-infant FUT2/FUT3 phenotype and infant gut microbiome modify the effect of an intervention containing small-quantity lipid based nutrient supplements, on infant stunting and LAZ at 18mo of age. Our goal was to define mechanisms that explain the rather modest effects of SQ-LNS on linear growth, reasoning that maternal-infant HBGA phenotypes might be important given their combined role in shaping the early-life microbiome. We found the following features are associated with greater reduction in stunting at 18mo by IYCF: (i) discordance in mother-infant FUT2+/FUT3-phenotype, where the mother has the phenotype but the infant does not; (ii) changes in microbiome species composition that reflected a shift from a B. longum-dominant microbiome to a microbiome with less B. longum and a greater abundance of species characteristic of older infants; and (iii) decreased B. longum abundance that was associated with paired mother-infant FUT+/FUT3-status. These findings suggest that a persistently “younger” microbiome, characterized by a high abundance of B. longum and carriage of B. longum strains best suited to HMO metabolism, at the point when complementary foods are introduced into the infant diet, may modify the effects of an intervention that includes SQ-LNS on infant stunting.

The mother only FUT2+/FUT3-group showed evidence of a greater reduction in stunting following receipt of one year of the IYCF intervention. Mother-infant FUT2+/FUT3-phenotype was also a significant predictor of B. longum, whereby relative abundance was lower throughout follow-up in the mother only group. Active maternal FUT2/FUT3 genes are key determinants of milk oligosaccharide composition31–33. Human milk specimens form distinct groups based on maternal FUT259 and FUT3 phenotype31,33 by principal component analyses of oligosaccharide composition. In particular, FUT2+/FUT3-mothers are a subset of FUT2+ women who form an HMO cluster that is distinct from mothers with other phenotypes, including FUT2+ mothers who are also FUT3+31,33. Differences in HMO composition influence growth and activity of Bifidobacterium populations in the infant gut28,59. Host FUT2 status may also affect infant gut microbiome composition36–40. Gut Bifidobacterium39, Bacteroides36,38, Faecalibacterium40 and Roseburia40 are differentially abundant in FUT2+ compared to FUT2-individuals. However, reported differences in microbiome composition by FUT2/FUT3 status have been inconsistent, potentially due to variability in age, dietary patterns, host health, sampled gut sections and methodology between studies60,614, and studies predominantly included adults. However, CAZymes found in B. longum species that function in HMO degradation (GH29 and GH95) have also been found to function in host intestinal glycan degradation in infants62. The evidence, therefore, suggests that the FUT2/FUT3 phenotypes of mother and infant, together, can elicit a strong prebiotic selective pressure, driven by maternal milk and infant glycan composition28 that influences Bifidobacteria and broader gut microbiota composition in infants.

In contrast to our finding that mother-infant FUT2+/FUT3-phenotype predicted the presence of B. longum subspecies infantis cluster strains, a prior study reported that maternal or infant FUT2 or FUT3 status did not predict the abundance of subspecies infantis or longum in children63. However, we included infant age as a covariate in our models, while the prior study did not63. In addition, to accommodate the fact that FUT2 and FUT3 work in concert biologically to produce different HBGAs and HMOs (Figure S1), we defined combined FUT2 and FUT3 phenotypes. Another smaller study found infants born to FUT2+/FUT3-mothers had lower bifidobacterial operational taxonomic unit diversity compared to other infants39. This study provides additional evidence that maternal FUT2+/FUT3-phenotype is associated with bifidobacterial taxon composition in the infant gut. Furthermore, we considered paired maternal and infant phenotypes63. More research is needed to clarify the roles of mother-infant FUT2/FUT3, maternal HMOs and the infant glycobiome as drivers of infant gut microbiome composition and development.

In our PCoA model, infant microbiome species maturation predominantly reflected decreased abundance of B. longum and increased abundance of P. copri, F. prausnitzii, D. longicatena, and D. formicigenerans. These species were also important predictors of infant age in a microbiome-age model previously developed from this same cohort64. Age-discriminatory taxa in the same genera were also identified in previously reported microbiome-age models14. Delayed gut microbiota-age has been reported in children with severe acute malnutrition (SAM), while improvements in microbiota-age and relative abundance of age-discriminatory taxa have been correlated with better growth14. Our analyses complement this literature with evidence that infant gut microbiome species maturation increased the effect of the IYCF intervention on stunting at 18mo, while greater relative abundance of B. longum, an age-discriminatory taxon which is associated with younger age, reduced the effect. This suggests that delayed maturation of the gut microbiome at the point when complementary foods are introduced and IYCF is initiated, may impair IYCF-induced effects on growth.

Infant age was the strongest determinant of B. longum relative abundance over time. This is consistent with previous reports65. In breastfed infants, Bifidobacteria are the most abundant gut bacteria65,66. Human breast milk is rich in oligosaccharides, which Bifidobacteria preferentially utilize67. At introduction of solid foods, a wider variety of nutrients and reduced availability of HMOs correspond to a decrease in Bifidobacterium,68 greater bacterial diversity and evenness, and development of a more adult-like composition by 2–3 years of age11,69–71.

Utilization of HMOs by Bifidobacterium varies between species and strains72–78. We identified three clusters of dominant B. longum strains that also varied by infant age and mother-infant FUT2+/FUT3-phenotype. The cluster of strains with pangenomes most closely resembling subspecies infantis had the highest prevalence in early infancy. However, by age 18mo there was a marked decrease in detection of infantis cluster strains and a corresponding increase in detection of strains with pangenomes most similar to subspecies suis, which had the highest prevalence at 18mo. Strains most similar to subspecies longum increased more slowly. This finding is consistent with a recent study that also reported a pattern of succession among three distinct gut B. longum clades from birth to 24mo of age. Subspecies infantis was dominant in early infancy but peaked at age 6mo and decreased considerably thereafter. While a transitional clade that included strains most similar to subspecies suis and suillum showed a corresponding increase in abundance that peaked by 18mo, and subspecies longum expanded from 15mo-24mo of age79. In that report, the transitional suis/suillum clade harbored functional capacity to degrade both HMOs and dietary polysaccharides, suggesting it may be an adaptation of the infant gut microbiome to a period when breastfeeding may co-occur with introduction of complementary foods79. We add to this work by showing that differences between mother and infant in FUT2+/FUT3-phenotype may also play a role in driving this transition, whereby, throughout follow-up, the mother only FUT2+/FUT3-group had a low prevalence of the infantis cluster and the highest prevalence of the suis cluster; while the infantis cluster was most prevalent in the infant only group.

B. longum subspecies infantis are particularly well adapted for HMO utilization80,81. B. longum subspecies longum, on the other hand, do not grow as well on HMO, and are more specialized for utilization of diet-derived polysaccharides82. In our analyses, UniProt gene families that were more frequent in the infantis cluster were more likely to be involved in HMO degradation and uptake. Two of CAZyme families more likely to be carried by the infantis cluster (GH2983 and GH3374) were previously reported as more prevalent in subspecies infantis strains. In contrast, UniProt gene families that were less frequent in the infantis cluster were more likely to be involved in uptake of fructose and other simple sugars, including the sugar transporter 3.A.1.2.23, which was previously described in a different B. longum subspecies84. Furthermore, strains in the suis and longum clusters were more likely to carry UniProt gene families involved in such MetaCyc pathways as pectin, xylan, and arabinogalactan degradation. UniProt gene families found more frequently in infantis cluster strains were also more likely to be involved in siderophore and antimicrobial biosynthesis. Heavy metals such as iron and zinc are essential minerals for nearly all bacteria and their mammalian hosts. Strategies utilized by bacteria to acquire heavy metals include siderophore biosynthesis. Siderophores are low molecular weight iron-chelating compounds that are synthesized by bacteria to scavenge iron and other essential metals such as zinc under nutrient-restricted conditions85. For example, Bifidobacterium species isolated from iron-deficient children efficiently sequester iron via siderophore production86. Siderophores may provide a competitive advantage to Bifidobacteria87, which along with antimicrobial biosynthesis88, may also help protect the infant gut from enteric pathogens that require essential metals for colonization. However, essential metals such as iron and zinc are key components in SQ-LNS formulations89–91, and produce improvements in linear growth and reductions in stunting risk92,936. While siderophore activity varies considerably between Bifidobacterium species and strains, no research to date has investigated sequestration of essential metals by subspecies infantis strains commonly found in resource-limited settings. Overall, our analyses suggest that greater abundance of B. longum, at the time when IYCF is started, characterized by a predominance of strains with carriage of gene families which confer greater capacity for HMO utilization and siderophore biosynthesis, is associated with lower impact to reduce stunting at 18mo. These suggest potential mechanisms by which a “younger” microbiome may constrain the beneficial effects of an SQ-LNS intervention on infant stunting. However, more research is required to fully elucidate the biological mechanisms and downstream pathways to stunting that could be involved.

Our results are in contrast to a previous RCT that investigated infant microbiome composition as a modifier of SQ-LNS impact on infant growth In Malawi48. However, in the primary analyses of that RCT, there was no effect of SQ-LNS on linear infant growth94. Furthermore, SQ-LNS was provided to both mothers during pregnancy and infants starting at 6mo postpartum, active control interventions (iron-folate or multiple micronutrient supplements) were provided to mothers48,94, and 16S rRNA gene amplification and sequencing were used to characterize the microbiota which can affect both taxon detection (including Bifidobacterium) and study results95. A second RCT reported evidence for effect modification of iron supplementation on growth by maternal FUT2 status, where infants of FUT2-mothers randomized to iron supplementation showed greater declines in LAZ than infants of FUT2+ mothers96. In addition, a third, recent RCT found that suppression of B. longum by amoxicillin allowed the gut microbiota of children with SAM to better adapt to a solid-food diet by reducing the abundance of taxa specialized for breast milk utilization, resulting in improved anthropometric indicators of infant nutritional status97. However, it is essential to note that B. longum subspecies infantis is a critical early infant gut bacteria with important benefits for infant health, including protection from enteric pathogens98, immune system development99, and reducing asthma risk100. B. longum subspecies infantis also improved ponderal growth when administered to infants ∼4mo of age with SAM in two RCTs101,102. However, the effects on LAZ were not statistically significant. Our work complements this literature. Our findings suggest that a shift away from a “less mature” microbiome, characterized by a high abundance of B. longum and carriage of strains that are better suited to HMO metabolism, may be critical to support efficacy of nutrient supplements that start with the introduction of complementary feeding; and we identify maternal drivers of early infant B. longum abundance and strain carriage which, if further elucidated, may be employed to shape the infant microbiome into a more favorable composition at this critical time in infancy.

There are some limitations of this work. First, from our cohort of 1169 HIV-unexposed infants, we had microbiome data on 172 infants. However, infants included in our analyses were comparable to infants who excluded (Table S6). Also, mother-infant FUT2+/FUT3-phenotype, which was an important determinant of B. longum abundance and strain cluster detection, modified the impact of IYCF on stunting in the larger sample of 792 infants without measured microbiome data in a way that was consistent with our microbiome results. That said, our small sample size may have limited power to detect effect modification by other taxa that are indicative of a “more mature” post-weaning microbiome. Our analyses were not able to investigate effect modification by actual microbial metabolic activity, maternal HMO composition or infant gut glycobiome, which would require multi-omics approaches such as metatranscriptomics and metabolomics. Finally, it is possible that the delay in microbiome maturation we characterized may reflect some reverse causality, where sick or non-thriving children are put to the breast more with higher relative breastmilk intake. However, this behavior would need to systematically occur less often in the subset of mothers and their infants who were discordant for FUT2+/FUT3-that we describe to fully explain our results, thus we think this explanation is unlikely.

In conclusion, we present analyses of moderators of IYCF impact on infant stunting at 18mo in an RCT of IYCF using SQ-LNS in rural Zimbabwe. We report that (i) infant microbiome species maturation, characterized by a shift from B. longum dominance, particularly B. longum strains that are most similar to the proficient HMO utilizer subspecies infantis, was associated with increased IYCF reduction of stunting risk at age 18mo; (ii) discordance in mother-infant FUT2+/FUT3-phenotype, where the mother had the phenotype but the infant did not, was associated with increased IYCF reduction of stunting risk; and (iii) reduction in B. longum relative abundance was also determined by mother-infant FUT2+/FUT3-phenotype. Future work should investigate how differences between maternal HMOs and infant gut glycans determine gut Bifidobacterium species and strain composition, and investigate how to tailor interventions at the introduction of complementary feeding to balance infant nutritional needs with age-related microbiome composition and function to prevent stunting.

Methods

Study design and participants

The Sanitation Hygiene Infant Nutrition Efficacy (SHINE) trial was a 2×2 factorial cluster-randomized trial that enrolled 5280 pregnant women at a median age of 12.5 weeks gestation between November 2012 and March 2015 to test the impact of improved household water quality, sanitation, and hygiene (WASH) and improved infant and young child feeding (IYCF) via provision of SQ-LNS to the infant from age 6mo-18mo, on linear growth and anemia at age 18mo. A detailed description of the SHINE trial design and methods has been published51,103.

Briefly, research nurses made home visits twice during pregnancy and at infant ages 1, 3, 6, 12, and 18 months. At baseline, maternal education and age, household wealth104, existing water and sanitation services, and household food security105 were assessed, and mothers were tested for HIV via a rapid testing algorithm. Infant birth date, weight, and delivery details were transcribed from health facility records. Gestational age at delivery was calculated from the date of the mother’s last menstrual period ascertained at baseline. Infant weight, length, and mid-upper arm circumference were measured at every postnatal visit. Nurses were standardized against a gold-standard anthropometrist every 6 months, with retraining provided to those who failed to meet predefined criteria.

Part-way through the trial (from mid-2014 onwards) mother-infant pairs were invited to join a substudy to collect additional biological specimens. Women were informed about the substudy at their 32-week gestation visit and those with live births were enrolled at the 1-month postnatal visit, or as soon as possible thereafter106,107.

Specimen collection and processing

Mothers collected fecal specimens prior to the research nurse visit. Fecal specimens were placed in a cold box and transported to the field laboratory, where they were stored at –80°C until transfer to the central laboratory in Harare for long-term archiving at –80°C, with generator back-up. Fecal specimens were transferred via private courier on dry ice from Harare, Zimbabwe to Vancouver, British Columbia for metagenomic analyses.

The Qiagen DNeasy PowerSoil Kit was used to extract total DNA from 200mg of feces, according to manufacturer’s instructions. Paired-end libraries were constructed using the Illumina TruSeq kit and using New England Biosystem TruSeq compatible library preparation reagents. Libraries were sequenced at the British Columbia Genome Sciences Centre using the Illumina HiSeq 2500 platform. Forty-eight libraries were pooled and included per sequencing lane. Negative controls were included to capture microbial contamination in the DNA extraction and library preparation steps.

The analyses presented here utilize data and specimens from HIV-uninfected mothers and their infants enrolled in the specimen collection substudy. The fecal microbiome was characterized in 354 specimens collected from 172 HIV-unexposed infants from 1 to 18mo of age. A mean(sd) of 2.0(1.0) samples were analyzed per child. Infants included in these analyses largely resembled the population of live-born infants from the wider SHINE trial who were not included in these analyses (Table S6). However, infants in the current analyses had slightly older mothers and longer gestational ages, and fewer were born during the hungry season, but more were in a household that met the minimum dietary diversity score (Table S6). Overall, the majority were born by vaginal delivery (91.6%) in an institution (91.5%) and were exclusively breastfed (83.2% at 3 months). The prevalence of stunting was 26.9% at 18mo in this sub-study (Table S6).

Bioinformatics

Sequenced reads were trimmed of adapters and filtered to remove low-quality, short (<60 base-pairs), and duplicate reads, as well as those of human, other animal or plant origin using KneadData with default settings. Overall, 354 unique whole metagenome sequencing datasets were used from fecal specimens collected from 1mo to 18mo from 172 infants with available mother and infant FUT2 and FUT3 phenotypes (Figure S2). On average, 10.8±3.7 million paired end quality-filtered reads were generated per sample. Assessment of negative controls and technical variation have been previously reported64. Species composition was determined by mapping reads to clade-specific markers using MetaPhlAn3, while functional gene and metabolic pathway composition was determined using HUMAnN3 against the UniRef90 database, both with default settings52. Bacterial species and pathway abundance estimates were normalized to relative abundances. UniProt gene family profiles were generated for dominant B. longum strains in fecal metagenomes with sufficient coverage for pangenome analysis using PanPhlan352. To facilitate interpretation of UniProt gene family profiles, the minimum set of biological pathways sufficient to explain the gene families identified in each strain was determined using MinPath with default settings58 and the MetaCyc database57.

Assessment of FUT2 and FUT3 status

Saliva samples were collected by oral swab from mothers and infants. Available saliva from any follow-up visit was selected to assess FUT2 and FUT3 status. Secretor versus non-secretor (FUT2) and Lewis-positive versus Lewis-null (FUT3) status were ascertained for infants and their mothers using a previously reported phenotyping assay108. We defined FUT2 and FUT3 phenotype combinations as Lewis-positive non-secretors (FUT2-/FUT3+), Lewis-positive secretors (FUT+/FUT3+), or Lewis-null secretors (FUT2+/FUT3-) (Table S11) according to the histo-blood group antigen synthesis pathways defined in Figure S1. Paired mother-infant phenotype concordance or discordance was defined as presented in Table S1.

Of 1,169 mother-infant pairs, 999 (85.4%) mothers and 1104 (94.4%) infants were tested. Of these, FUT2 or FUT3 status could be determined in 889 (76.0%) and 999 (85.4%) mothers and infants, respectively (Table S11). Of those whose FUT2 or FUT3 status could be ascertained, secretors were the most frequent FUT2 phenotype (88.1% and 85.6% in mothers and infants respectively), while Lewis-positive was the most frequent FUT3 phenotype (76.8% and 74.9% in mothers and infants, respectively) (Table S11). Most mothers and infants were Lewis-positive secretors (60.5% and 64.9%, respectively), followed by FUT2+/FUT3-(25.1% and 23.2%, respectively), and Lewis-positive non-secretors (14.4% and 11.9%, respectively) (Table S1).

Statistical analyses

Infant characteristics that explain species β-diversity were evaluated using constrained principal coordinates analysis (PCoA) of Bray-Curtis dissimilarities (capscale)109. Characteristics of interest included factors that are known to be correlated with microbiome composition (infant age, sex, EBF at 3mo, dietary diversity, maternal and infant FUT2 and FUT3 phenotype). Statistical significance was tested by permutational analysis of variance using distance matrices with 1000 permutations (adonis2)110. We then developed a final multivariable constrained PCoA model that included covariates which explained a significant fraction (p<0.05) of the variance in microbiome composition (infant age at specimen collection), as well as mother-infant FUT2+/FUT3-discordance given our a prior hypothesis that FUT2 and FUT3 phenotypes are important determinants of infant microbiome composition and our finding that mother-infant FUT2+/FUT3-discordance was an important modifier of IYCF efficacy to reduce stunting at 18mo. Constrained PCoA axis scores in the final multivariable model represented changes in microbiome composition (species turnover) along gradients defined by each infant characteristic included in the full model.

Since the IYCF intervention was started at the 6mo follow-up visit, we assessed interaction between randomization to IYCF and mother-infant FUT2+/FUT3-discordance or microbiome composition using data from the 6mo visit as covariates. The primary outcome was stunting at 18mo and LAZ at 18mo was a secondary outcome. We fitted separate models for stunting and LAZ at 18mo. Interaction was assessed on the additive risk difference scale, which is most appropriate for statistical estimation of synergistic biological effects111. We used generalized linear models (glm) with a Gaussian distribution, an identity link, sandwich standard errors (sandwich)112 and an IYCF-by-mother-infant FUT2+/FUT3-phenotype interaction term. We repeated these analyses using IYCF-by-PCoA axis score interaction terms, where PCoA axis scores were derived from the final multivariable constrained PCoA model. One model was fitted per constrained PCoA axis. For species that were strongly associated with PCoA axis scores (loadings > 0.5 or < –0.5), analyses were repeated to assess IYCF-by-species interactions. We fitted a separate model for each species of interest. Models also included IYCF, infant sex, mother-infant FUT2+/FUT3-discordance, infant age at specimen collection, an indicator of whether infants met the minimum dietary diversity score at specimen collection, and LAZ at specimen collection. We did not include WASH arm because, in prior analyses, the SHINE WASH intervention did not affect stunting or LAZ at 18mo51 nor infant gut microbiome composition64. P-values were adjusted for multiple hypothesis testing to preserve the false discovery rate113.

Identification and analysis of Bifidobacterium longum strain clusters

B. longum strain profiles produced with PanPhlan3.0, which indicate whether UniProt gene families are present in a strain114, were converted to Jaccard dissimilarity matrices and visualized by PCoA to ascertain the existence of strain clusters (capscale). Three clusters were identified, and strain cluster membership was determined by hierarchical clustering of Jaccard dissimilarities and Ward’s error sum of squares algorithm (hclust)115. Hierarchical clustering dendrograms were cut at a height to obtain three clusters.

Differences in UniProt gene family profiles between B. longum strain clusters were determined by two-sided Fisher’s Exact test (fisher.test), with adjustment for multiple hypothesis testing to preserve the false discovery rate113, and were visualized using heatmaps (heatmap3). 3260 UniProt gene families were differentially present between strain clusters after FDR-adjustment. We performed overrepresentation analyses53 using one-sided Fisher’s Exact tests (fisher.test) to determine whether the differentially present gene families were more likely to function as particular CAZymes55 or transporters56, or in specific GO biological processes54. These analyses were repeated using the biological pathways determined with MinPath58. 115 pathways were differentially present between strain clusters after FDR-adjustment. We performed overrepresentation analyses of these pathways to determine whether they were more likely to have particular biological functions defined by MetaCyc pathway types.

Predictors of Bifidobacterium relative abundance and strain detection

Predictors of B. longum relative abundance over time were assessed using mixed-effects zero-inflated beta regression estimated by restricted maximum likelihood (gamlss)116. The model included infant age at specimen collection, sex, EBF at 3mo, minimum infant dietary diversity at specimen collection, and mother-infant FUT2+/FUT3-discordance, with random intercepts (re) and a first order autocorrelation structure (corCAR1).

Predictors of B. longum strain cluster were assessed by logistic regression (glm) using an indicator of strain cluster presence as the dependent variable, with the same covariates, and sandwich standard errors. An individual model was fitted separately for each cluster. Bias corrected117 logistic regression was used to facilitate stable parameter estimation due to separation resulting from the small sample size118.

All statistical analyses were conducted in R version 4.2.0. PCoA and adonis2 were performed using the vegan package119. Heatmaps were generated with heatmap3. Mixed-effects zero-inflated beta regression was performed using the gamlss package120. Bias corrected logistic regression models were fitted using the brglm2 package. Sandwich standard errors were generated with the sandwich package.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Ethan Gough (egough1{at}jh.edu).

Materials availability

This study did not generate new unique reagents.

Data and code availability

The raw metagenome sequencing data generated in this study have been deposited in the European Bioinformatics Database under accession code PRJEB51728. Final processed and annotated metagenome sequencing data files (taxa and pathways) are available at https://doi.org/10.5281/zenodo.7471082. Epidemiologic data files are available at Code for statistical analyses are available from the corresponding author upon request.

Supplemental Tables

Table S8. Overrepresentation of UniProt gene families and Metacyc pathways, which differ between strain clusters, by GO biological process, CAZyme or Transporter class or Metacyc pathway type.

Ethics approvals

All SHINE mothers provided written informed consent. The Medical Research Council of Zimbabwe (MRCZ/A/1675), Johns Hopkins Bloomberg School of Public Health (JHU IRB # 4205.), and the University of British Columbia Ethics Board (H15-03074) approved the study protocol, including the microbiome analyses. The SHINE trial is registered at ClinicalTrials.gov (NCT01824940).

Data Availability

All data produced in the present study are available upon reasonable request to the authors.

Author contributions

A.R.M., L.E.S., R.J.S., M.N.N.M., J.H.H. and A.J.P. conceptualized and designed the study. K.M., R.N., B.C., F.D.M., N.V.T., J.T., and B.M. collected data and biospecimens. H.M.G., I.B., S.K.G., R.C.R., F.F. and L.C. processed fecal specimens. M.K. conducted laboratory analyses to ascertain FUT2 and FUT3 status. C.E., J.C. and E.K.G. developed the FUT2 and FUT3 analysis plan. E.K.G. developed and conducted the microbiome statistical analysis plan. T.J.E. and E.K.G. conducted bioinformatics. A.R.M. and E.K.G. analyzed and interpreted the data. E.K.G. wrote the original manuscript draft. All authors reviewed the manuscript. A.R.M., A.J.P. and J.H.H. supervised and verified the data.

Declaration of interests

T.J.E. was paid a scientific consulting fee in relation to the analysis of the data presented here by the Zvitambo Institute for Maternal and Child Health Research. All other authors declare no competing interests.

Acknowledgments

We thank all the mothers, babies, and their families who participated in the SHINE trial and all members of the SHINE trial team (all members listed here: https://doi.org/10.1093/cid/civ844). We particularly thank the leadership and staff of the Ministry of Health and Child Care in Chirumanzu and Shurugwi districts and Midlands Province (especially environmental health, nursing, and nutrition) for their roles in operationalization of the study procedures, the Ministry of Local Government officials in each district who supported and facilitated field operations, Phillipa Rambanepasi and her team for proficient management of all the finances, Virginia Sauramba for management of compliance issues, and the programme officers at the Bill & Melinda Gates Foundation and the Department for International Development (UK Aid), who enthusiastically worked with us over a long period to make SHINE happen.

Funding was from the Bill & Melinda Gates Foundation (OPP1021542 and OPP1143707; J.H.H. and A.J.P.), with a subcontract to the University of British Columbia (20R25498; A.R.M.). United Kingdom Department for International Development (DFID/UKAID; J.H.H. and A.J.P.). Wellcome Trust (093768/Z/10/Z, 108065/Z/15/Z, 206455/Z/17/Z, 203905/Z/16/Z and 210807/Z/18/Z; A.J.P., R.C.R. and C.E.). Swiss Agency for Development and Cooperation (J.H.H. and A.J.P.). US National Institutes of Health (2R01HD060338-06; J.H.H.). UNICEF (PCA-2017-0002; J.H.H. and A.J.P.). The Nutricia Research Foundation (2021-52; E.K.G.) The funders had no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Footnotes

  • The author list is updated.

References

  1. 1.↵
    United Nations Children’s Fund, World Health Organization, and The World Bank (2019). Levels and trends in child malnutrition: key findings of the 2019 Edition of the Joint Child Malnutrition Estimates – UNICEF Regions. (United Nations Children’s Fund (UNICEF), World Health Organization, International Bank for Reconstruction and Development/The World Bank).
  2. 2.↵
    Onis, M.D., Garza, C., Onyango, A.W., and Martorell, R. (2006). WHO child growth standards. Acta Paediatrica 95, 1–104.
    OpenUrl
  3. 3.↵
    Benjamin-Chung, J., Mertens, A., Colford, J.M., Hubbard, A.E., van der Laan, M.J., Coyle, J., Sofrygin, O., Cai, W., Nguyen, A., Pokpongkiat, N.N., et al. (2023). Early-childhood linear growth faltering in low– and middle-income countries. Nature 621, 550–557.
    OpenUrlCrossRefPubMed
  4. 4.↵
    Gough, E.K., Moodie, E.E.M., Prendergast, A.J., Ntozini, R., Moulton, L.H., Humphrey, J.H., Manges, A.R., Stephens, D.A., Moodie, E.E.M., Prendergast, A.J., et al. (2016). Linear growth trajectories in Zimbabwean infants. American Journal of Clinical Nutrition 104, 1616–1627.
    OpenUrlAbstract/FREE Full Text
  5. 5.↵
    Adair, L.S.L.L.S., Fall, C.H.D.D., Osmond, C., Stein, A.D.A.A.D., Martorell, R., Ramirez-Zea, M., Sachdev, H.S., Dahly, D.L., Bas, I., Norris, S.A., et al. (2013). Associations of linear growth and relative weight gain during early life with adult health and human capital in countries of low and middle income: Findings from five birth cohort studies. The Lancet 382, 525–534.
    OpenUrl
  6. 6.
    Victora, C.G., Adair, L., Fall, C., Hallal, P.C., Martorell, R., Richter, L., and Sachdev, H.S. (2008). Maternal and child undernutrition: consequences for adult health and human capital. Lancet 371, 340–357.
    OpenUrlCrossRefPubMedWeb of Science
  7. 7.↵
    Vollmer, S., Harttgen, K., Subramanyam, M.A., Finlay, J., Klasen, S., and Subramanian, S.V. (2014). Association between economic growth and early childhood undernutrition: evidence from 121 Demographic and Health Surveys from 36 low-income and middle-income countries. The Lancet Global Health 2, e225–234.
    OpenUrl
  8. 8.↵
    Dewey, K.G., Wessells, K.R., Arnold, C.D., Prado, E.L., Abbeddou, S., Adu-Afarwuah, S., Ali, H., Arnold, B.F., Ashorn, P., Ashorn, U., et al. (2021). Characteristics that modify the effect of small-quantity lipid-based nutrient supplementation on child growth: an individual participant data meta-analysis of randomized controlled trials. Am J Clin Nutr 114, 15S–42S.
    OpenUrl
  9. 9.↵
    Dewey, K.G., Arnold, C.D., Wessells, K.R., Prado, E.L., Abbeddou, S., Adu-Afarwuah, S., Ali, H., Arnold, B.F., Ashorn, P., Ashorn, U., et al. (2022). Preventive small-quantity lipid-based nutrient supplements reduce severe wasting and severe stunting among young children: an individual participant data meta-analysis of randomized controlled trials. Am J Clin Nutr 116, 1314–1333.
    OpenUrl
  10. 10.↵
    Ruel, M.T., Alderman, H., and Maternal Child Nutrition Study Group (2013). Nutrition-sensitive interventions and programmes: how can they help to accelerate progress in improving maternal and child nutrition? Lancet 382, 536–551.
    OpenUrlCrossRefPubMedWeb of Science
  11. 11.↵
    Arrieta, M.-C., Stiemsma, L.T., Amenyogbe, N., Brown, E.M., and Finlay, B. (2014). The intestinal microbiome in early life: health and disease. Frontiers in Immunology 5, 427.
    OpenUrl
  12. 12.↵
    Jones, H.J., Bourke, C.D., Swann, J.R., and Robertson, R.C. (2023). Malnourished Microbes: Host– Microbiome Interactions in Child Undernutrition. Annual Review of Nutrition 43, 327–353.
    OpenUrl
  13. 13.↵
    Robertson, R.C., Manges, A.R., Finlay, B.B., and Prendergast, A.J. (2019). The Human Microbiome and Child Growth – First 1000 Days and Beyond. Trends Microbiol 27, 131–147.
    OpenUrlCrossRefPubMed
  14. 14.↵
    Subramanian, S., Huq, S., Yatsunenko, T., Haque, R., Mahfuz, M., Alam, M.A., Benezra, A., DeStefano, J., Meier, M.F., Muegge, B.D., et al. (2014). Persistent gut microbiota immaturity in malnourished Bangladeshi children. Nature 510, 417–421.
    OpenUrlCrossRefPubMedWeb of Science
  15. 15.
    Zhao, G., Vatanen, T., Droit, L., Park, A., Kostic, A.D., Poon, T.W., Vlamakis, H., Siljander, H., Härkönen, T., Hämäläinen, A.M., et al. (2017). Intestinal virome changes precede autoimmunity in type I diabetes-susceptible children. Proceedings of the National Academy of Sciences of the United States of America 114, E6166–E6175. doi:10.1073/pnas.1706359114.
    OpenUrlAbstract/FREE Full Text
  16. 16.↵
    Ho, N.T., Li, F., Lee-Sarwar, K.A., Tun, H.M., Brown, B.P., Pannaraj, P.S., Bender, J.M., Azad, M.B., Thompson, A.L., Weiss, S.T., et al. (2018). Meta-analysis of effects of exclusive breastfeeding on infant gut microbiota across populations. Nature Communications 9, 4169. doi:10.1038/s41467-018-06473-x.
    OpenUrlCrossRefPubMed
  17. 17.↵
    Reyes, A., Blanton, L.V., Cao, S., Zhao, G., Manary, M., Trehan, I., Smith, M.I., Wang, D., Virgin, H.W., Rohwer, F., et al. (2015). Gut DNA viromes of Malawian twins discordant for severe acute malnutrition. Proceedings of the National Academy of Sciences of the United States of America 112, 11941–11946. doi:10.1073/pnas.1514285112.
    OpenUrlAbstract/FREE Full Text
  18. 18.↵
    Gough, E.K., Stephens, D.A., Moodie, E.M., Prendergast, A.J., Stoltzfus, R.J., Humphrey, J.H., and Manges, A.R. (2016). Linear growth faltering in infants is associated with Acidaminococcus sp. and community-level changes in the gut microbiota. Microbiome 3, 24.
    OpenUrl
  19. 19.
    Ghosh, T.S., Gupta, S.S., Bhattacharya, T., Yadav, D., Barik, A., Chowdhury, A., Das, B., Mande, S.S., and Nair, G.B. (2014). Gut Microbiomes of Indian Children of Varying Nutritional Status. PLoS ONE 9, e95547.
    OpenUrlCrossRefPubMed
  20. 20.↵
    Gupta, S.S., Mohammed, M.H., Ghosh, T.S., Kanungo, S., Nair, G.B., and Mande, S.S. (2011). Metagenome of the gut of a malnourished child. Gut Pathog 3, 7.
    OpenUrlCrossRefPubMed
  21. 21.↵
    Huey, S.L., Jiang, L., Fedarko, M.W., McDonald, D., Martino, C., Ali, F., Russell, D.G., Udipi, S.A., Thorat, A., Thakker, V., et al. (2020). Nutrition and the Gut Microbiota in 10– to 18-Month-Old Children Living in Urban Slums of Mumbai, India. mSphere 5, e00731–20.
    OpenUrl
  22. 22.
    Digitale, J., Sié, A., Coulibaly, B., Ouermi, L., Dah, C., Tapsoba, C., Bärnighausen, T., Lebas, E., Arzika, A.M., Glymour, M.M., et al. (2020). Gut Bacterial Diversity and Growth among Preschool Children in Burkina Faso. Am J Trop Med Hyg 103, 2568–2573.
    OpenUrl
  23. 23.
    Collard, J.-M., Andrianonimiadana, L., Habib, A., Rakotondrainipiana, M., Andriantsalama, P., Randriamparany, R., Rabenandrasana, M. a. N., Weill, F.-X., Sauvonnet, N., Randremanana, R.V., et al. (2022). High prevalence of small intestine bacteria overgrowth and asymptomatic carriage of enteric pathogens in stunted children in Antananarivo, Madagascar. PLoS Negl Trop Dis 16, e0009849.
    OpenUrl
  24. 24.
    Kamng’ona, A.W., Young, R., Arnold, C.D., Kortekangas, E., Patson, N., Jorgensen, J.M., Prado, E.L., Chaima, D., Malamba, C., Ashorn, U., et al. (2019). The association of gut microbiota characteristics in Malawian infants with growth and inflammation. Scientific Reports 9.
  25. 25.↵
    Perin, J., Burrowes, V., Almeida, M., Ahmed, S., Haque, R., Parvin, T., Biswas, S., Azmi, I.J., Bhuyian, S.I., Talukder, K.A., et al. (2020). A Retrospective Case-Control Study of the Relationship between the Gut Microbiota, Enteropathy, and Child Growth. The American journal of tropical medicine and hygiene.
  26. 26.↵
    Brown, E.M., Wlodarska, M., Willing, B. p, Vonaesch, P., Han, J., Reynolds, L.A., Arriet, M.-C., Uhrig, M., Scholz, R., Partida, O., et al. (2015). Diet and specific microbial exposure trigger features of environmental enteropathy in a novel murine model. Nature Communications 6, 7806.
    OpenUrl
  27. 27.↵
    Kau, A.L., Planer, J.D., Liu, J., Rao, S., Yatsunenko, T., Trehan, I., Manary, M.J., Liu, T.-C., Stappenbeck, T.S., Maleta, K.M., et al. (2015). Functional characterization of IgA-targeted bacterial taxa from undernourished Malawian children that produce diet-dependent enteropathy. Science Translational Medicine 7, 276ra24-276ra24.
    OpenUrlAbstract/FREE Full Text
  28. 28.↵
    Salli, K., Hirvonen, J., Siitonen, J., Ahonen, I., Anglenius, H., and Maukonen, J. (2021). Selective Utilization of the Human Milk Oligosaccharides 21-Fucosyllactose, 3-Fucosyllactose, and Difucosyllactose by Various Probiotic and Pathogenic Bacteria. Journal of Agricultural and Food Chemistry 69.
  29. 29.↵
    Davis, J.C.C., Totten, S.M., Huang, J.O., Nagshbandi, S., Kirmiz, N., Garrido, D.A., Lewis, Z.T., Wu, L.D., Smilowitz, J.T., German, J.B., et al. (2016). Identification of Oligosaccharides in Feces of Breast-fed Infants and Their Correlation with the Gut Microbial Community*. Molecular & Cellular Proteomics 15, 2987–3002.
    OpenUrl
  30. 30.↵
    Charbonneau, M.R., O’Donnell, D., Blanton, L.V., Totten, S.M., Davis, J.C.C., Barratt, M.J., Cheng, J., Guruge, J., Talcott, M., Bain, J.R., et al. (2016). Sialylated Milk Oligosaccharides Promote Microbiota-Dependent Growth in Models of Infant Undernutrition. Cell 164, 859–871.
    OpenUrlCrossRefPubMed
  31. 31.↵
    Mank, M., Hauner, H., Heck, A.J.R., and Stahl, B. (2020). Targeted LC-ESI-MS2 characterization of human milk oligosaccharide diversity at 6 to 16 weeks post-partum reveals clear staging effects and distinctive milk groups. Anal Bioanal Chem 412, 6887–6907.
    OpenUrl
  32. 32.
    Castanys-Muñoz, E., Martin, M.J., and Prieto, P.A. (2013). 2’-fucosyllactose: an abundant, genetically determined soluble glycan present in human milk. Nutr Rev 71, 773–789.
    OpenUrlCrossRefPubMed
  33. 33.↵
    Blank, D., Dotz, V., Geyer, R., and Kunz, C. (2012). Human milk oligosaccharides and Lewis blood group: individual high-throughput sample profiling to enhance conclusions from functional studies. Adv Nutr 3, 440S–9S.
    OpenUrlAbstract/FREE Full Text
  34. 34.↵
    Schneider, M., Al-Shareffi, E., and Haltiwanger, R.S. (2017). Biological functions of fucose in mammals. Glycobiology 27, 601–618.
    OpenUrlCrossRefPubMed
  35. 35.↵
    Cooling, L. (2015). Blood groups in infection and host susceptibility. Clinical Microbiology Reviews 28.
  36. 36.↵
    Tong, M., McHardy, I., Ruegger, P., Goudarzi, M., Kashyap, P.C., Haritunians, T., Li, X., Graeber, T.G., Schwager, E., Huttenhower, C., et al. (2014). Reprograming of gut microbiome energy metabolism by the FUT2 Crohn’s disease risk polymorphism. ISME J 8, 2193–2206.
    OpenUrlCrossRefPubMed
  37. 37.
    Rausch, P., Rehman, A., Künzel, S., Häsler, R., Ott, S.J., Schreiber, S., Rosenstiel, P., Franke, A., and Baines, J.F. (2011). Colonic mucosa-associated microbiota is influenced by an interaction of crohn disease and FUT2 (Secretor) genotype. Proceedings of the National Academy of Sciences of the United States of America 108.
  38. 38.↵
    Gampa, A., Engen, P.A., Shobar, R., and Mutlu, E.A. (2017). Relationships between gastrointestinal microbiota and blood group antigens. Physiol Genomics 49, 473–483.
    OpenUrlCrossRefPubMed
  39. 39.↵
    Wacklin, P., Mäkivuokko, H., Alakulppi, N., Nikkilä, J., Tenkanen, H., Räbinä, J., Partanen, J., Aranko, K., and Mättö, J. (2011). Secretor genotype (FUT2 gene) is strongly associated with the composition of bifidobacteria in the human intestine. PLoS ONE 6, e20113.
    OpenUrlCrossRefPubMed
  40. 40.↵
    Rühlemann, M.C., Hermes, B.M., Bang, C., Doms, S., Moitinho-Silva, L., Thingholm, L.B., Frost, F., Degenhardt, F., Wittig, M., Kässens, J., et al. (2021). Genome-wide association study in 8,956 German individuals identifies influence of ABO histo-blood groups on gut microbiome. Nat Genet 53, 147–155.
    OpenUrl
  41. 41.↵
    Christensen, L., Vuholm, S., Roager, H.M., Nielsen, D.S., Krych, L., Kristensen, M., Astrup, A., and Hjorth, M.F. (2019). Prevotella Abundance Predicts Weight Loss Success in Healthy, Overweight Adults Consuming a Whole-Grain Diet Ad Libitum: A Post Hoc Analysis of a 6-Wk Randomized Controlled Trial. J Nutr 149, 2174–2181.
    OpenUrlCrossRefPubMed
  42. 42.
    Eriksen, A.K., Brunius, C., Mazidi, M., Hellström, P.M., Risérus, U., Iversen, K.N., Fristedt, R., Sun, L., Huang, Y., Nørskov, N.P., et al. (2020). Effects of whole-grain wheat, rye, and lignan supplementation on cardiometabolic risk factors in men with metabolic syndrome: a randomized crossover trial. Am J Clin Nutr 111, 864–876. doi:10.1093/ajcn/nqaa026.
    OpenUrlCrossRef
  43. 43.
    Shin, J.-H., Jung, S., Kim, S.-A., Kang, M.-S., Kim, M.-S., Joung, H., Hwang, G.-S., and Shin, D.-M. (2019). Differential Effects of Typical Korean Versus American-Style Diets on Gut Microbial Composition and Metabolic Profile in Healthy Overweight Koreans: A Randomized Crossover Trial. Nutrients 11, E2450.
    OpenUrl
  44. 44.↵
    Hjorth, M.F., Roager, H.M., Larsen, T.M., Poulsen, S.K., Licht, T.R., Bahl, M.I., Zohar, Y., and Astrup, A. (2018). Pre-treatment microbial Prevotella-to-Bacteroides ratio, determines body fat loss success during a 6-month randomized controlled diet intervention. Int J Obes (Lond) 42, 580–583.
    OpenUrl
  45. 45.↵
    Li, Y., Wang, D.D., Satija, A., Ivey, K.L., Li, J., Wilkinson, J.E., Li, R., Baden, M., Chan, A.T., Huttenhower, C., et al. (2021). Plant-Based Diet Index and Metabolic Risk in Men: Exploring the Role of the Gut Microbiome. J Nutr 151, 2780–2789.
    OpenUrl
  46. 46.
    Li, J., Li, Y., Ivey, K.L., Wang, D.D., Wilkinson, J.E., Franke, A., Lee, K.H., Chan, A., Huttenhower, C., Hu, F.B., et al. (2022). Interplay between diet and gut microbiome, and circulating concentrations of trimethylamine N-oxide: findings from a longitudinal cohort of US men. Gut 71, 724–733.
    OpenUrlAbstract/FREE Full Text
  47. 47.↵
    Wang, D.D., Nguyen, L.H., Li, Y., Yan, Y., Ma, W., Rinott, E., Ivey, K.L., Shai, I., Willett, W.C., Hu, F.B., et al. (2021). The gut microbiome modulates the protective association between a Mediterranean diet and cardiometabolic disease risk. Nat Med 27, 333–343.
    OpenUrlCrossRefPubMed
  48. 48.↵
    Hughes, R.L., Arnold, C.D., Young, R.R., Ashorn, P., Maleta, K., Fan, Y.-M., Ashorn, U., Chaima, D., Malamba-Banda, C., Kable, M.E., et al. (2020). Infant gut microbiota characteristics generally do not modify effects of lipid-based nutrient supplementation on growth or inflammation: secondary analysis of a randomized controlled trial in Malawi. Sci Rep 10, 14861.
    OpenUrl
  49. 49.↵
    Smith, M.I., Yatsunenko, T., Manary, M.J., Trehan, I., Mkakosya, R., Cheng, J., Kau, A.L., Rich, S.S., Concannon, P., Mychaleckyj, J.C., et al. (2013). Gut microbiomes of Malawian twin pairs discordant for Kwashiorkor. Science 339, 548–554.
    OpenUrlAbstract/FREE Full Text
  50. 50.↵
    Schwarzer, M., Makki, K., Storelli, G., Machuca-Gayet, I., Srutkova, D., Hermanova, P., Martino, M.E., Balmand, S., Hudcovic, T., Heddi, A., et al. (2016). Lactobacillus plantarum strain maintains growth of infant mice during chronic undernutrition. Science 351, 854–857.
    OpenUrlAbstract/FREE Full Text
  51. 51.↵
    Humphrey, J.H., Mbuya, M.N.N., Ntozini, R., Moulton, L.H., Stoltzfus, R.J., Tavengwa, N.V., Mutasa, K., Majo, F.D., Mutasa, B., Mangwadu, G., et al. (2019). Independent and combined effects of improved water, sanitation, and hygiene, and improved complementary feeding, on child stunting and anaemia in rural Zimbabwe: a cluster-randomised trial. The Lancet Global Health 7, e132– e147.
    OpenUrlCrossRef
  52. 52.↵
    Beghini, F., McIver, L.J., Blanco-Míguez, A., Dubois, L., Asnicar, F., Maharjan, S., Mailyan, A., Manghi, P., Scholz, M., Thomas, A.M., et al. (2021). Integrating taxonomic, functional, and strain-level profiling of diverse microbial communities with bioBakery 3. Elife 10, e65088.
    OpenUrlCrossRefPubMed
  53. 53.↵
    Wieder, C., Frainay, C., Poupin, N., Rodríguez-Mier, P., Vinson, F., Cooke, J., Lai, R.P., Bundy, J.G., Jourdan, F., and Ebbels, T. (2021). Pathway analysis in metabolomics: Recommendations for the use of over-representation analysis. PLoS Comput Biol 17, e1009105.
    OpenUrl
  54. 54.↵
    The Gene Ontology Consortium, Aleksander, S.A., Balhoff, J., Carbon, S., Cherry, J.M., Drabkin, H.J., Ebert, D., Feuermann, M., Gaudet, P., Harris, N.L., et al. (2023). The Gene Ontology knowledgebase in 2023. Genetics 224, iyad031.
    OpenUrlCrossRefPubMed
  55. 55.↵
    Drula, E., Garron, M.-L., Dogan, S., Lombard, V., Henrissat, B., and Terrapon, N. (2022). The carbohydrate-active enzyme database: functions and literature. Nucleic Acids Res 50, D571–D577.
    OpenUrlCrossRef
  56. 56.↵
    Saier, M.H., Reddy, V.S., Moreno-Hagelsieb, G., Hendargo, K.J., Zhang, Y., Iddamsetty, V., Lam, K.J.K., Tian, N., Russum, S., Wang, J., et al. (2021). The Transporter Classification Database (TCDB): 2021 update. Nucleic Acids Res 49, D461–D467.
    OpenUrlCrossRefPubMed
  57. 57.↵
    Caspi, R., Billington, R., Fulcher, C.A., Keseler, I.M., Kothari, A., Krummenacker, M., Latendresse, M., Midford, P.E., Ong, Q., Ong, W.K., et al. (2018). The MetaCyc database of metabolic pathways and enzymes. Nucleic Acids Research 46, D633–D639.
    OpenUrlCrossRefPubMed
  58. 58.↵
    Ye, Y., and Doak, T.G. (2009). A parsimony approach to biological pathway reconstruction/inference for genomes and metagenomes. PLoS Comput Biol 5, e1000465.
    OpenUrlCrossRefPubMed
  59. 59.↵
    Bazanella, M., Maier, T.V., Clavel, T., Lagkouvardos, I., Lucio, M., Maldonado-Gòmez, M.X., Autran, C., Walter, J., Bode, L., Schmitt-Kopplin, P., et al. (2017). Randomized controlled trial on the impact of early-life intervention with bifidobacteria on the healthy infant fecal microbiota and metabolome. American Journal of Clinical Nutrition 106, 1274–1286.
    OpenUrlAbstract/FREE Full Text
  60. 60.
    Davenport, E.R., Cusanovich, D.A., Michelini, K., Barreiro, L.B., Ober, C., and Gilad, Y. (2015). Genome-wide association studies of the human gut microbiota. PLoS ONE 10. doi:10.1371/journal.pone.0140301.
    OpenUrlCrossRefPubMed
  61. 61.
    Turpin, W., Bedrani, L., Espin-Garcia, O., Xu, W., Silverberg, M.S., Smith, M.I., Guttman, D.S., Griffiths, A., Moayyedi, P., Panaccione, R., et al. (2018). FUT2 genotype and secretory status are not associated with fecal microbial composition and inferred function in healthy subjects. Gut Microbes 9.
  62. 62.↵
    Bell, A., and Juge, N. (2021). Mucosal glycan degradation of the host by the gut microbiota. Glycobiology 31, 691–696.
    OpenUrlCrossRef
  63. 63.↵
    Colston, J.M., Taniuchi, M., Ahmed, T., Ferdousi, T., Kabir, F., Mduma, E., Nshama, R., Iqbal, N.T., Haque, R., Ahmed, T., et al. (2022). Intestinal Colonization With Bifidobacterium longum Subspecies Is Associated With Length at Birth, Exclusive Breastfeeding, and Decreased Risk of Enteric Virus Infections, but Not With Histo-Blood Group Antigens, Oral Vaccine Response or Later Growth in Three Birth Cohorts. Front Pediatr 10, 804798.
    OpenUrl
  64. 64.↵
    Robertson, R.C., Edens, T.J., Carr, L., Mutasa, K., Gough, E.K., Evans, C., Geum, H.M., Baharmand, I., Gill, S.K., Ntozini, R., et al. (2023). The gut microbiome and early-life growth in a population with high prevalence of stunting. Nat Commun 14, 654.
    OpenUrl
  65. 65.↵
    Stewart, C.J., Ajami, N.J., O’Brien, J.L., Hutchinson, D.S., Smith, D.P., Wong, M.C., Ross, M.C., Lloyd, R.E., Doddapaneni, H.V., Metcalf, G.A., et al. (2018). Temporal development of the gut microbiome in early childhood from the TEDDY study. Nature 562, 583–588.
    OpenUrlCrossRefPubMed
  66. 66.↵
    Yatsunenko, T., Rey, F.E., Manary, M.J., Trehan, I., Dominguez-Bello, M.G., Contreras, M., Magris, M., Hidalgo, G., Baldassano, R.N., Anokhin, A.P., et al. (2012). Human gut microbiome viewed across age and geography. Nature 486, 222–228.
    OpenUrlCrossRefPubMedWeb of Science
  67. 67.↵
    Marcobal, A., and Sonnenburg, J.L. (2012). Human milk oligosaccharide consumption by intestinal microbiota: Human milk oligosaccharide consumption. Clinical Microbiology and Infection 18, 12– 15.
    OpenUrlCrossRefPubMed
  68. 68.↵
    Bäckhed, F., Roswall, J., Peng, Y., Feng, Q., Jia, H., Kovatcheva-Datchary, P., Li, Y., Xia, Y., Xie, H., Zhong, H., et al. (2015). Dynamics and stabilization of the human gut microbiome during the first year of life. Cell Host and Microbe 17, 690–703. doi:10.1016/j.chom.2015.04.004.
    OpenUrlCrossRefPubMed
  69. 69.↵
    Collado, M.C., Cernada, M., Baüerl, C., Vento, M., and Pérez-Martínez, G. (2012). Microbial ecology and host-microbiota interactions during early life stages. Gut Microbes 3, 352–365.
    OpenUrlCrossRefPubMed
  70. 70.
    Matamoros, S., Gras-Leguen, C., Vacon, F.L., Potel, G., and Cochetiere, M.-F. de L. (2013). Development of intestinal microbiota in infants and its impact on health. Trends in Microbiology 21, 167–173.
    OpenUrlCrossRefPubMedWeb of Science
  71. 71.↵
    Laursen, M.F. (2021). Gut Microbiota Development: Influence of Diet from Infancy to Toddlerhood. Annals of Nutrition and Metabolism.
  72. 72.↵
    Lewis, Z.T., Totten, S.M., Smilowitz, J.T., Popovic, M., Parker, E., Lemay, D.G., Van Tassell, M.L., Miller, M.J., Jin, Y.-S., German, J.B., et al. (2015). Maternal fucosyltransferase 2 status affects the gut bifidobacterial communities of breastfed infants. Microbiome 3, 13.
    OpenUrlCrossRefPubMed
  73. 73.
    Ojima, M.N., Jiang, L., Arzamasov, A.A., Yoshida, K., Odamaki, T., Xiao, J., Nakajima, A., Kitaoka, M., Hirose, J., Urashima, T., et al. (2022). Priority effects shape the structure of infant-type Bifidobacterium communities on human milk oligosaccharides. ISME J 16, 2265–2279.
    OpenUrl
  74. 74.↵
    Li, M., Zhou, X., Stanton, C., Ross, R.P., Zhao, J., Zhang, H., Yang, B., and Chen, W. (2021). Comparative Genomics Analyses Reveal the Differences between B. longum subsp. infantis and B. longum subsp. longum in Carbohydrate Utilisation, CRISPR-Cas Systems and Bacteriocin Operons. Microorganisms 9, 1713.
    OpenUrl
  75. 75.
    Gotoh, A., Katoh, T., Sakanaka, M., Ling, Y., Yamada, C., Asakuma, S., Urashima, T., Tomabechi, Y., Katayama-Ikegami, A., Kurihara, S., et al. (2018). Sharing of human milk oligosaccharides degradants within bifidobacterial communities in faecal cultures supplemented with Bifidobacterium bifidum. Scientific Reports 8, 13958.
    OpenUrl
  76. 76.
    Sakanaka, M., Gotoh, A., Yoshida, K., Odamaki, T., Koguchi, H., Xiao, J.-Z., Kitaoka, M., and Katayama, T. (2019). Varied Pathways of Infant Gut-Associated Bifidobacterium to Assimilate Human Milk Oligosaccharides: Prevalence of the Gene Set and Its Correlation with Bifidobacteria-Rich Microbiota Formation. Nutrients 12, 71. doi:10.3390/nu12010071.
    OpenUrlCrossRef
  77. 77.
    Garrido, D., Ruiz-Moyano, S., Lemay, D.G., Sela, D.A., German, J.B., and Mills, D.A. (2015). Comparative transcriptomics reveals key differences in the response to milk oligosaccharides of infant gut-associated bifidobacteria. Sci Rep 5, 13517.
    OpenUrlCrossRefPubMed
  78. 78.↵
    Garrido, D., Ruiz-Moyano, S., Kirmiz, N., Davis, J.C., Totten, S.M., Lemay, D.G., Ugalde, J.A., German, J.B., Lebrilla, C.B., and Mills, D.A. (2016). A novel gene cluster allows preferential utilization of fucosylated milk oligosaccharides in Bifidobacterium longum subsp. longum SC596. Sci Rep 6, 35045.
    OpenUrlCrossRef
  79. 79.↵
    Vatanen, T., Ang, Q.Y., Siegwald, L., Sarker, S.A., Le Roy, C.I., Duboux, S., Delannoy-Bruno, O., Ngom-Bru, C., Boulangé, C.L., Stražar, M., et al. (2022). A distinct clade of Bifidobacterium longum in the gut of Bangladeshi children thrives during weaning. Cell 185, 4280–4297.e12.
    OpenUrlCrossRef
  80. 80.↵
    Sela, D.A., Chapman, J., Adeuya, A., Kim, J.H., Chen, F., Whitehead, T.R., Lapidus, A., Rokhsar, D.S., Lebrilla, C.B., German, J.B., et al. (2008). The genome sequence of Bifidobacterium longum subsp. infantis reveals adaptations for milk utilization within the infant microbiome. Proceedings of the National Academy of Sciences 105, 18964–18969.
    OpenUrlAbstract/FREE Full Text
  81. 81.↵
    Thomson, P., Medina, D.A., and Garrido, D. (2018). Human milk oligosaccharides and infant gut bifidobacteria: Molecular strategies for their utilization. Food Microbiology 75, 37–46.
    OpenUrl
  82. 82.↵
    LoCascio, R.G., Desai, P., Sela, D.A., Weimer, B., and Mills, D.A. (2010). Broad conservation of milk utilization genes in Bifidobacterium longum subsp. infantis as revealed by comparative genomic hybridization. Appl Environ Microbiol 76, 7373–7381.
    OpenUrlAbstract/FREE Full Text
  83. 83.↵
    Fushinobu, S., and Abou Hachem, M. (2021). Structure and evolution of the bifidobacterial carbohydrate metabolism proteins and enzymes. Biochem Soc Trans 49, 563–578.
    OpenUrl
  84. 84.↵
    Wei, X., Guo, Y., Shao, C., Sun, Z., Zhurina, D., Liu, D., Liu, W., Zou, D., Jiang, Z., Wang, X., et al. (2012). Fructose Uptake in Bifidobacterium longum NCC2705 Is Mediated by an ATP-binding Cassette Transporter. Journal of Biological Chemistry 287, 357–367. doi:10.1074/jbc.M111.266213.
    OpenUrlAbstract/FREE Full Text
  85. 85.↵
    Johnstone, T.C., and Nolan, E.M. (2015). Beyond Iron: Non-Classical Biological Functions of Bacterial Siderophores. Dalton Trans 44, 6320–6339.
    OpenUrlCrossRefPubMed
  86. 86.↵
    Vazquez-Gutierrez, P., Lacroix, C., Jaeggi, T., Zeder, C., Zimmerman, M.B., and Chassard, C. (2015). Bifidobacteria strains isolated from stools of iron deficient infants can efficiently sequester iron. BMC Microbiol 15, 3.
    OpenUrlCrossRefPubMed
  87. 87.↵
    Vazquez-Gutierrez, P., de Wouters, T., Werder, J., Chassard, C., and Lacroix, C. (2016). High Iron-Sequestrating Bifidobacteria Inhibit Enteropathogen Growth and Adhesion to Intestinal Epithelial Cells In vitro. Front Microbiol 7, 1480.
    OpenUrlCrossRef
  88. 88.↵
    Gibson, G. r., and Wang, X. (1994). Regulatory effects of bifidobacteria on the growth of other colonic bacteria. Journal of Applied Bacteriology 77, 412–420.
    OpenUrlCrossRefPubMedWeb of Science
  89. 89.↵
    Arimond, M., Zeilani, M., Jungjohann, S., Brown, K.H., Ashorn, P., Allen, L.H., and Dewey, K.G. (2015). Considerations in developing lipid-based nutrient supplements for prevention of undernutrition: experience from the International Lipid-Based Nutrient Supplements (iLiNS) Project. Matern Child Nutr 11, 31–61.
    OpenUrl
  90. 90.
    Beal, T., White, J.M., Arsenault, J.E., Okronipa, H., Hinnouho, G.-M., Murira, Z., Torlesse, H., and Garg, A. (2021). Micronutrient gaps during the complementary feeding period in South Asia: A Comprehensive Nutrient Gap Assessment. Nutr Rev 79, 26–34.
    OpenUrl
  91. 91.↵
    White, J.M., Beal, T., Arsenault, J.E., Okronipa, H., Hinnouho, G.-M., Chimanya, K., Matji, J., and Garg, A. (2021). Micronutrient gaps during the complementary feeding period in 6 countries in Eastern and Southern Africa: a Comprehensive Nutrient Gap Assessment. Nutr Rev 79, 16–25.
    OpenUrl
  92. 92.
    Park, J.J.H., Harari, O., Siden, E., Dron, L., Zannat, N.-E., Singer, J., Lester, R.T., Thorlund, K., and Mills, E.J. (2020). Interventions to improve linear growth during complementary feeding period for children aged 6-24 months living in low– and middle-income countries: a systematic review and network meta-analysis [version 2; peer review: 3 approved]. 3, 1660.
    OpenUrl
  93. 93.
    Imdad, A., and Bhutta, Z.A. (2011). Effect of preventive zinc supplementation on linear growth in children under 5 years of age in developing countries: a meta-analysis of studies for input to the lives saved tool. BMC Public Health 11, S22.
    OpenUrlCrossRefPubMed
  94. 94.↵
    Ashorn, P., Alho, L., Ashorn, U., Cheung, Y.B., Dewey, K.G., Gondwe, A., Harjunmaa, U., Lartey, A., Phiri, N., Phiri, T.E., et al. (2015). Supplementation of Maternal Diets during Pregnancy and for 6 Months Postpartum and Infant Diets Thereafter with Small-Quantity Lipid-Based Nutrient Supplements Does Not Promote Child Growth by 18 Months of Age in Rural Malawi: A Randomized Controlled Trial. J Nutr 145, 1345–1353. doi:10.3945/jn.114.207225.
    OpenUrlAbstract/FREE Full Text
  95. 95.↵
    Abellan-Schneyder, I., Matchado, M.S., Reitmeier, S., Sommer, A., Sewald, Z., Baumbach, J., List, M., and Neuhaus, K. (2021). Primer, Pipelines, Parameters: Issues in 16S rRNA Gene Sequencing. mSphere 6, e01202–20.
    OpenUrlCrossRefPubMed
  96. 96.↵
    Paganini, D., Uyoga, M.A., Kortman, G.A.M., Boekhorst, J., Schneeberger, S., Karanja, S., Hennet, T., and Zimmermann, M.B. (2019). Maternal Human Milk Oligosaccharide Profile Modulates the Impact of an Intervention with Iron and Galacto-Oligosaccharides in Kenyan Infants. Nutrients 11, 2596.
    OpenUrl
  97. 97.↵
    Schwartz, D.J., Langdon, A., Sun, X., Langendorf, C., Berthé, F., Grais, R.F., Trehan, I., Isanaka, S., and Dantas, G. (2023). Effect of amoxicillin on the gut microbiome of children with severe acute malnutrition in Madarounfa, Niger: a retrospective metagenomic analysis of a placebo-controlled trial. The Lancet Microbe 4, e931–e942.
    OpenUrl
  98. 98.↵
    Ruiz, L., Flórez, A.B., Sánchez, B., Moreno-Muñoz, J.A., Rodriguez-Palmero, M., Jiménez, J., Gavilán, C.G. de los R., Gueimonde, M., Ruas-Madiedo, P., and Margolles, A. (2020). Bifidobacterium longum subsp. infantis CECT7210 (B. infantis IM-1®) Displays In Vitro Activity against Some Intestinal Pathogens. Nutrients 12, 3259.
    OpenUrl
  99. 99.↵
    Henrick, B.M., Rodriguez, L., Lakshmikanth, T., Pou, C., Henckel, E., Arzoomand, A., Olin, A., Wang, J., Mikes, J., Tan, Z., et al. (2021). Bifidobacteria-mediated immune system imprinting early in life. Cell 184, 3884–3898.e11.
    OpenUrlCrossRef
  100. 100.↵
    Dai, D.L.Y., Petersen, C., Hoskinson, C., Del Bel, K.L., Becker, A.B., Moraes, T.J., Mandhane, P.J., Finlay, B.B., Simons, E., Kozyrskyj, A.L., et al. (2023). Breastfeeding enrichment of B. longum subsp. infantis mitigates the effect of antibiotics on the microbiota and childhood asthma risk. Med 4, 92–112.e5.
    OpenUrl
  101. 101.↵
    Barratt, M.J., Nuzhat, S., Ahsan, K., Frese, S.A., Arzamasov, A.A., Sarker, S.A., Islam, M.M., Palit, P., Islam, M.R., Hibberd, M.C., et al. (2022). Bifidobacterium infantis treatment promotes weight gain in Bangladeshi infants with severe acute malnutrition. Sci Transl Med 14, eabk1107.
    OpenUrl
  102. 102.↵
    Nuzhat, S., Hasan, S.M.T., Palit, P., Islam, M.R., Mahfuz, M., Islam, M.M., Alam, M.A., Flannery, R.L., Kyle, D.J., Sarker, S.A., et al. (2023). Effects of probiotic and synbiotic supplementation on ponderal and linear growth in severely malnourished young infants in a randomized clinical trial. Sci Rep 13, 1845.
    OpenUrl
  103. 103.↵
    Humphrey, J.H., Jones, A.D., Manges, A., Mangwadu, G., Maluccio, J.A., Mbuya, M.N.N., Moulton, L.H., Ntozini, R., Prendergast, A.J., Stoltzfus, R.J., et al. (2015). The sanitation hygiene infant nutrition efficacy (SHINE) Trial: Rationale, design, and methods. Clinical Infectious Diseases 61, S685–702.
    OpenUrlCrossRefPubMed
  104. 104.↵
    Chasekwa, B., Maluccio, J.A., Ntozini, R., Moulton, L.H., Wu, F., Smith, L.E., Matare, C.R., Stoltzfus, R.J., Mbuya, M.N.N., Tielsch, J.M., et al. (2018). Measuring wealth in rural communities: Lessons from the sanitation, hygiene, infant nutrition efficacy (SHINE) trial. PLoS ONE 13, 1–19.
    OpenUrlCrossRefPubMed
  105. 105.↵
    Maxwell, D., Watkins, B., Wheeler, R., and Collins, G. (2003). The Coping Strategy Index: a tool for rapid measurement of household food security and the impact of food aid programs in humanitarian emergencies. (CARE and WFP).
  106. 106.↵
    E.K, G., L.H, M., K, M., R, N., R.J, S., F.D, M., L.E, S., G, P., N, G., M, J., et al. (2020). Effects of improved water, sanitation, and hygiene and improved complementary feeding on environmental enteric dysfunction in children in rural Zimbabwe: A cluster-randomized controlled trial. PLoS Neglected Tropical Diseases 14, e0007963.
    OpenUrl
  107. 107.↵
    Mutasa, K., Ntozini, R., Mbuya, M.N.N., Rukobo, S., Govha, M., Majo, F.D., Tavengwa, N., Smith, L.E., Caulfield, L., Swann, J.R., et al. (2021). Biomarkers of environmental enteric dysfunction are not consistently associated with linear growth velocity in rural Zimbabwean infants. The American journal of clinical nutrition 113.
  108. 108.↵
    Colston, J.M., Francois, R., Pisanic, N., Yori, P.P., Mccormick, B.J.J., Olortegui, M.P., Gazi, M.A., Svensen, E., Ahmed, M.M.M., Mduma, E., et al. (2019). Effects of Child and Maternal Histo-Blood Group Antigen Status on Symptomatic and Asymptomatic Enteric Infections in Early Childhood. Journal of Infectious Diseases 220, 151–162.
    OpenUrlCrossRef
  109. 109.↵
    Legendre, P., and Anderson, M.J. (1999). Distance-Based Redundancy Analysis: Testing Multispecies Responses in Multifactorial Ecological Experiments. Ecological Monographs 69, 1–24.
    OpenUrlCrossRefWeb of Science
  110. 110.↵
    Anderson, M.J. (2001). A new method for non-parametric multivariate analysis of variance. Austral Ecology 26, 32–46.
    OpenUrl
  111. 111.↵
    VanderWeele, T.J. (2012). Sample Size and Power Calculations for Additive Interactions. Epidemiol Methods 1, 159–188.
    OpenUrl
  112. 112.↵
    Naimi, A.I., and Whitcomb, B.W. (2020). Estimating Risk Ratios and Risk Differences Using Regression. Am J Epidemiol 189, 508–510.
    OpenUrlCrossRefPubMed
  113. 113.↵
    Benjamini, Y., and Hochberg, Y. (1995). Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society. Series B (Methodological) 57, 289–300.
    OpenUrlCrossRefPubMedWeb of Science
  114. 114.↵
    Scholz, M., Ward, D.V., Pasolli, E., Tolio, T., Zolfo, M., Asnicar, F., Truong, D.T., Tett, A., Morrow, A.L., and Segata, N. (2016). Strain-level microbial epidemiology and population genomics from shotgun metagenomics. Nature Methods 13, 435–438.
    OpenUrl
  115. 115.↵
    Murtagh, F., and Legendre, P. (2014). Ward’s Hierarchical Agglomerative Clustering Method: Which Algorithms Implement Ward’s Criterion? Journal of Classification 31, 274–295.
    OpenUrlCrossRef
  116. 116.↵
    Chen, E.Z., and Li, H. (2016). A two-part mixed-effects model for analyzing longitudinal microbiome compositional data. In Bioinformatics, pp. 2611–2617. doi:10.1093/bioinformatics/btw308.
    OpenUrlCrossRefPubMed
  117. 117.↵
    Kosmidis, I., Kenne Pagui, E.C., and Sartori, N. (2020). Mean and median bias reduction in generalized linear models. Stat Comput 30, 43–59.
    OpenUrl
  118. 118.↵
    Mansournia, M.A., Geroldinger, A., Greenland, S., and Heinze, G. (2018). Separation in Logistic Regression: Causes, Consequences, and Control. Am J Epidemiol 187, 864–870.
    OpenUrl
  119. 119.↵
    Dixon, P. (2003). VEGAN, a package of R functions for community ecology. Journal of Vegetation Science 14, 927–930.
    OpenUrlCrossRefWeb of Science
  120. 120.↵
    Stasinopoulos, D.M., and Rigby, R.A. (2007). Generalized additive models for location scale and shape (GAMLSS) in R. Journal of Statistical Software 23, 1–46. doi:10.18637/jss.v023.i07.
    OpenUrlCrossRef
Back to top
Posted March 04, 2024.
Download PDF

Supplementary Material

Data/Code
Email
Bifidobacterium longum modifies a nutritional intervention for stunting in Zimbabwean infants
Ethan K Gough, Thaddeus J Edens, Lynnea Carr, Ruairi C Robertson, Kuda Mutasa, Robert Ntozini, Bernard Chasekwa, Hyun Min Geum, Iman Baharmand, Sandeep K Gill, Batsirai Mutasa, Mduduzi N N Mbuya, Florence D Majo, Naume Tavengwa, Freddy Francis, Joice Tome, Ceri Evans, Margaret Kosek, Andrew J Prendergast, Amee R Manges, the Sanitation Hygiene Infant Nutrition Efficacy (SHINE) Trial Team
medRxiv 2024.01.18.24301438; doi: https://doi.org/10.1101/2024.01.18.24301438
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area