**Background:** Childhood stunting affects an estimated 149 million children under 5 globally. The gut microbiome is increasingly implicated in malnutrition, but longitudinal data from resource-limited settings during the first two years of life remain scarce. This study aimed to evaluate the effect of age and residential location on intestinal microbiota composition and diversity in children under 24 months in urban and rural Sindh, Pakistan.
**Methods:** This was a planned prospective pilot sub-study nested within a cluster-randomized trial (NCT00705445) comparing two micronutrient powder formulations with or without zinc. From 2746 children enrolled at 3–6 months and followed to 24 months, 60 children (10 per intervention arm per site) with fecal samples available at all six time points (3, 6, 9, 12, 18, and 24 months) were randomly selected. DNA was extracted from 360 fecal samples using QIAamp Stool Mini kits, and the V4 hypervariable region of the 16S rRNA gene was sequenced on a Roche 454 GS FLX platform. Computational analyses used QIIME 1.5; OTUs were clustered at 97% identity. Alpha diversity (Chao1, equitability, Shannon, phylogenetic diversity) and beta diversity (weighted UniFrac PCoA) were assessed. Linear mixed-effects models adjusted for gender, residential location, and diarrheal episodes were used to estimate age-specific changes. LEfSe identified differentially abundant taxa.
**Key Results:** At baseline, 24/60 (40%) children were wasted, 13/22 (22%) stunted, and 21/35 (35%) underweight. Urban children had a significantly higher median age at recruitment (101 vs. 91 days, p=0.0011). Beta diversity showed significant clustering by age: weighted UniFrac distances differed significantly at 9 (p<0.05), 12 (p<0.01), 18 (p<0.001), and 24 (p<0.001) months compared to 3 months. No clustering by residential location or intervention arm was observed. All four alpha diversity indices increased significantly with age (p<0.0001 for Chao1, equitability, Shannon, and PD). The most abundant phyla were Actinobacteria, Firmicutes, Proteobacteria, and Bacteroidetes. Firmicutes and Bacteroidetes increased significantly with age (age-specific effect +0.030, p<0.0001 and +0.015, p<0.0001, respectively), while Actinobacteria and Proteobacteria decreased (−0.039, p<0.0001 and −0.015, p=0.0001). Among genera, Bifidobacterium (−0.0372, p<0.0001), Escherichia/Shigella (−0.0157, p<0.0001), and Streptococcus (−0.0077, p=0.0003) decreased significantly; Lactobacillus showed no significant change (p=0.795). LEfSe analysis confirmed enrichment of Actinobacteria and Proteobacteria in the first year and Firmicutes, Bacteroidetes, and Cyanobacteria in the second year. Urban-rural comparisons revealed sporadic differentially abundant taxa at various time points (e.g., Clostridiaceae enriched in rural children at 3 and 6 months; Bifidobacterium enriched in urban children at 6 months). No significant differences in alpha or beta diversity were found between urban and rural children at any time point. Intervention arms showed very few differentially abundant taxa.
**Clinical Implications:** This study confirms that age is the dominant driver of gut microbiota maturation in Pakistani children from 3 to 24 months, with a trajectory similar to that seen in other low- and middle-income settings. The lack of significant differences by residential location or micronutrient intervention suggests that environmental and nutritional interventions may need to be more targeted or of longer duration to measurably alter microbiota composition. The small number of malnourished children per subgroup precluded robust analysis of microbiota differences by nutritional status. Future studies should employ larger sample sizes, metagenomic sequencing, and tools like the microbiota-for-age Z-score to better understand the relationship between gut microbiota and linear growth stunting.