**Background:** The Asian elephant (Elephas maximus) is an endangered species with only approximately 300 individuals remaining in China. Insufficient maternal milk is a major contributor to the low survival rate of young elephants—approximately 25.6% of elephant calves in Myanmar die before age 5, with a quarter of these deaths attributed to inadequate milk intake. At the Xishuangbanna Asian Elephant Sanctuary, goat milk is currently used to supplement rescued infants and young elephants. Diet, particularly early nutrition, strongly influences gut microbiota composition and metabolic activity, which in turn affects host nutrient acquisition, immune regulation, and development. This study aimed to evaluate the association between different milk-containing diets and the gut microbiomes of young Asian elephants to identify optimal breast milk supplementation strategies.
**Methods:** In March 2019, fresh fecal samples were collected from eight young Asian elephants at Wild Elephant Valley in Xishuangbanna, divided into three groups: (1) elephant milk-only diet (BF1, BF2, BF3; healthy, ~6 months old), (2) elephant milk–plant mixed-feed diet (BPM1, BPM2, BPM3; healthy, >1 year old), and (3) goat milk–plant mixed-feed diet (GPM1, GPM2; healthy, >3 years old). Microbial DNA was extracted using the EZNA Soil DNA Kit. The V3-V4 hypervariable region of the 16S rRNA gene was amplified with primers 338F/806R and sequenced on the Illumina MiSeq platform. Raw reads were processed using fastp v0.20.0 and FLASH v1.2.7. OTUs were clustered at 97% similarity using UPARSE v7.1, and taxonomy was assigned via RDP Classifier v2.2 against the Silva v138 database. Alpha diversity indices (Chao1, Shannon, Pielou) were calculated using mothur v1.30.2, with between-group differences tested using Welch's t-test. Differential taxa were identified using the Kruskal–Wallis H test. Functional predictions were generated using PICRUSt2, with significance assessed by Kruskal–Wallis H test.
**Key Results:** UPGMA clustering based on Bray-Curtis distances clearly separated the elephant milk-only group from the milk–plant mixed-feed groups, and further subdivided the mixed-feed groups by milk type (elephant milk vs. goat milk). Alpha diversity analysis showed that Chao1 and Shannon indices differed significantly between the three groups (p < 0.05). The milk–plant mixed-feed diet groups had significantly higher richness and diversity than the elephant milk-only group (p < 0.05). The Shannon and Pielou indices were significantly higher in the elephant milk–plant mixed-feed group than in the goat milk–plant mixed-feed group (p < 0.05). At the phylum level, Firmicutes and Bacteroidetes were dominant across all groups. The elephant milk-only group had a high abundance of Proteobacteria (average ~17.3%). The elephant milk–plant mixed-feed group showed higher abundances of Spirochaetae (~8.8%), Fibrobacteria (~3.8%), and Verrucomicrobia (~3.6%). The goat milk–plant mixed-feed group contained nearly no Proteobacteria, Spirochaetae, or Fibrobacteria, but had abundant Synergistetes. At the family level, the elephant milk–plant mixed-feed group had significantly enriched Rikenellaceae, Spirochaetaceae, Fibrobacteraceae, and Bacteroidales_UCG-001 (p < 0.05), while the goat milk–plant mixed-feed group had significantly enriched Prevotellaceae, Synergistaceae, and Christensenellaceae (p < 0.05). Functional prediction revealed that membrane transport and cell motility pathways were significantly enriched in the elephant milk–plant mixed-feed group compared to the goat milk–plant mixed-feed group (p = 0.044 for both), while energy metabolism (p = 0.044), amino acid metabolism (p = 0.044), and signal transduction (p = 0.025) pathways were significantly enriched in the goat milk–plant mixed-feed group. Nutrient composition comparison showed that Asian elephant milk contains higher total solids (17.56–19.60%), protein (3.30–5.23%), and milk fat (7.70–8.30%) compared to goat milk (11.53–13.00%, 3.17–3.75%, and 3.95–4.25%, respectively). Yak milk composition (water 83.74%, total solids 16.60–18.52%, protein 4.68–5.41%, milk fat 6.72–8.18%, ash 0.72–1.19%, lactose 4.40–5.10%) was noted to be similar to that of Asian elephant milk.
**Clinical Implications:** This study provides the first description of the gut microbiota composition and function in young Asian elephants fed a goat milk-supplemented diet. The lower microbial diversity and enrichment of Prevotellaceae (associated with obesity) in the goat milk–plant mixed-feed group suggest that goat milk may not be the optimal supplement for young elephants. The higher abundance of lignocellulose-degrading bacteria (Rikenellaceae, Spirochaetaceae, Fibrobacteraceae) in the elephant milk–plant mixed-feed group indicates that elephant milk supplementation better facilitates the transition to a plant-based diet. Based on compositional similarity, yak milk is proposed as a potentially more suitable alternative to goat milk for supplementing rescued young Asian elephants. The authors acknowledge the small sample size (n=8) as a limitation but note that these represent all available samples from the Xishuangbanna region at the time. This study demonstrates the utility of gut microbiome analysis for evaluating milk source suitability in endangered wildlife conservation.