Altitude, Not Ethnicity, Shapes the Oral Microbiome of Preschool Children

The community of bacteria living in the mouths of preschool children appears to shift dramatically with elevation, according to a new cross-sectional study of 3- to 5-year-olds living on the plains and the plateau of western China. Researchers report that children living at roughly 4,000 meters carried salivary microbial communities with markedly higher diversity than peers of the same ethnic background living near sea level, and that this altitude signal was far stronger than any signal tied to whether the children were Han Chinese or Tibetan.

The study, published in Frontiers in Oral Health, enrolled 87 children and divided them into three groups: 27 Han Chinese children living at low altitude, around 400 meters, and 30 Han Chinese and 30 Tibetan children living long term at high altitude, around 4,000 meters. All participants were healthy, with no carious or missing teeth, no pulpitis or chronic periapical periodontitis, no systemic diseases, and no recent use of antibiotics, corticosteroids, immunosuppressants, probiotics, or antibacterial mouthwash. Guardians completed questionnaires covering demographics, diet, oral hygiene, and family socioeconomic conditions, and each child provided two milliliters of unstimulated saliva for sequencing.

The team sequenced the V3–V4 region of the bacterial 16S rRNA gene on an Illumina NovaSeq platform and processed the data with QIIME2, using the DADA2 pipeline for quality filtering, denoising, and chimera removal. After rarefaction to a depth of 45,620 reads per sample, the researchers classified sequence variants against the Greengenes database. To predict what the communities might be doing metabolically, they applied PICRUSt2 with reference to the KEGG database, an approach that infers potential gene functions from the taxa present rather than measuring them directly.

The diversity results were striking. All four alpha-diversity indices used in the study—Chao-1 and Observed Species, which estimate richness, and the Shannon and Simpson indices, which capture diversity—were significantly lower in the low-altitude Han group than in either high-altitude group, with no significant difference between the two high-altitude groups. In multiple linear regression models adjusted for daily sugar intake, dietary pattern, residential area, maternal education, and maternal employment status, altitude showed positive standardized coefficients of 0.74 to 0.76 across all four indices, and every one of those associations survived Benjamini-Hochberg false discovery rate correction with q values below 0.01. Ethnicity, by contrast, had no independent statistical effect on any alpha-diversity measure.

Community composition told a parallel story. Principal coordinate analysis based on Bray–Curtis distances showed samples clustering by elevation, with the first axis explaining 28.4 percent of community variation and the second 8.8 percent; the low-altitude Han group separated most clearly from the two high-altitude groups, which overlapped partially but still differed from each other. PERMANOVA confirmed that overall community structure differed significantly across the three groups, with altitude as the primary driver. Counts of amplicon sequence variants reinforced the pattern: the low-altitude group yielded 3,781 ASVs, the high-altitude Han group 8,127, and the high-altitude Tibetan group 9,027, with only 846 core ASVs shared by all three groups. The two high-altitude groups shared far more ASVs with each other than either did with the low-altitude group.

At the phylum level, altitude was independently and positively associated with Bacteroidetes after FDR correction, with a standardized coefficient of 0.80 and q below 0.001. A nominally significant negative association with Firmicutes, at a coefficient of −0.35 and an unadjusted P of 0.028, did not survive correction, reaching a q of 0.056. At the species level, altitude was positively associated with Prevotella melaninogenica, Haemophilus parainfluenzae, Veillonella parvula, and Veillonella atypica, with coefficients ranging from 0.39 to 0.66 and all q values below 0.01, while showing a negative association with Granulicatella adiacens. An association with Fusobacterium periodonticum did not reach significance after correction. None of the six species-level associations tested for ethnicity remained significant after FDR correction.

The predicted functional profiles diverged along the same lines. Low-altitude children’s communities showed higher predicted enrichment of glycolysis and gluconeogenesis, pentose-glucuronic acid interconversion, ascorbic acid and aldonic acid metabolism, fatty acid biosynthesis, and fatty acid metabolism, while the high-altitude groups showed predicted enrichment of NOD-like receptor signaling and protein digestion and absorption pathways. Streptococcus, Prevotella melaninogenica, Gemella, Streptococcus salivarius, and Fusobacterium periodonticum emerged as the major contributing genera to these differential pathways. The authors note that these are taxon-based predictions from 16S data, not direct measurements of gene content, and that metagenomic sequencing would be needed to confirm actual functional profiles.

The findings contrast with earlier work on adults. Previous studies of plateau populations, focused mainly on adults, reported that Han Chinese adults showed greater salivary microbial diversity than Tibetan adults and that adult microbiomes tend to show reduced diversity under chronic high-altitude stress. The authors speculate that the reversal in young children may reflect the developmental plasticity of the oral microbiome between ages 3 and 5, a window in which environmental factors become the dominant influence and microbial communities remain highly responsive to external pressures such as hypoxia and low temperature. They also point out that many high-altitude children in the study came from rural areas, where more complex living environments have been linked to greater microbial diversity.

Baseline questionnaires revealed imbalances among the groups that the regression models had to contend with: daily sugar intake, dietary pattern, residential area, maternal education, and maternal employment all differed significantly across groups, with the low-altitude group showing the highest proportion of high-frequency sugar consumers. After adjustment, however, none of these confounding factors independently influenced microbial diversity or the abundance of core species in the models, leaving altitude as the dominant statistical correlate. The authors emphasize that because the study enrolled only children free of active caries and obvious periodontal inflammation, the links drawn between microbial patterns and disease risk are hypothesis-generating inferences based on known pathogenic mechanisms, not direct predictions of disease outcomes.

The limitations are considerable and the authors lay them out explicitly. As a cross-sectional observational study, it can establish only correlations, not causation. Saliva samples cannot fully represent the dental plaque or subgingival biofilms where caries and periodontal disease actually develop. The study lacked a low-altitude Tibetan control group, so altitude and ethnicity could not be fully separated in a factorial design, and residual confounding from unmeasured dietary, cultural, or lifestyle factors may remain. The authors also caution that ethnicity here represents a composite of cultural, dietary, socioeconomic, and environmental factors and should not be read as a genetic effect. Future work, they write, should include children with caries and gingivitis, sample plaque and gingival crevicular fluid, add low-altitude Tibetan controls, apply metagenomic and multi-omics methods, and follow children longitudinally to test whether these salivary microbial signatures predict oral disease as it develops.

Subject of Research: Associations of altitude and ethnicity with the salivary oral microbiome of preschool children in China

Article Title: A study on differences in the structure and function of the oral microbiome in 3- to 5-year-old children from plateau and plain regions’

Article References: Bai, X., Yang, L., Zhou, Z., Sun, X., Liu, F., & Dong, N. (2026). A study on differences in the structure and function of the oral microbiome in 3- to 5-year-old children from plateau and plain regions’. Frontiers in Oral Health, 7, Article 1983860. https://doi.org/10.3389/froh.2026.1983860

Image Credits: AI Generated

DOI: 10.3389/froh.2026.1983860

Keywords: oral microbiome, preschool children, altitude, 16S rRNA sequencing, Tibetan Plateau, alpha diversity, PICRUSt2, dental caries, periodontal disease, saliva microbiota, cross-sectional study, Bacteroidetes