Skin bacterial community dynamics of hands and forearms before and after military field exercise

Abstract

The human skin microbiome is crucial for health and immunity, especially under the extreme conditions military personnel face. Soldiers often encounter unique stressors and hygienic challenges that can alter their skin's microbial composition, particularly in field environments. In this study, we aimed to investigate the impact of military field exercises on the diversity and composition of the skin bacterial microbiota using 16S rRNA sequencing. We conducted a longitudinal study of Norwegian soldiers (n = 19) participating in outdoor training operations during the NATO winter exercise Cold Response 2022. Skin swabs were taken from soldiers' hands and forearms before and after the 10-day military exercise, and following a 3-week post-exercise leave. Our results reveal hand- and forearm-specific shifts in bacterial populations associated with the exercise, likely influenced by environmental exposure, reduced hygiene, and heightened social contact. Alpha diversity increased on forearms while remaining stable on hands, which appeared more resilient to perturbations. Both sites exhibited temporal changes in composition, with soil- and water-associated bacteria enriched post-exercise; most being transient on hands but more sustained on forearms. The soldiers' microbiomes converged during the exercise, then diverged in the post-exercise leave period, and neither skin site returned to baseline composition at follow-up. Our findings highlight the impact of collaborative outdoor activities on microbial communities and suggest that resilience and stability differ between skin sites.IMPORTANCEOptimizing soldier health and resilience is critical for maintaining military readiness and operational effectiveness. The skin, as the body's first line of defense, is subjected to numerous challenges in military environments. Unique environmental and hygiene challenges can disrupt the skin microbiome and increase susceptibility to skin and soft tissue infections. This longitudinal research provides valuable insights into the effects of military service on the bacterial dynamics of the skin microbiome but can also inform hygiene management and disease prevention in comparable situations.

Keywords: 16S rRNA; bacteria; environment; metagenomics; microbiota; military personnel; sequencing; skin microbiome.

Fig 1

Study design. Soldiers (n = 19) participating in a 10-day field exercise were swabbed on their hands and forearms for skin microbiome samples. Two samples (both hands and both forearms were pooled individually together) were taken for each subject in three rounds: (1) The day before the start of exercise (baseline), (2) immediately after the end of exercise, and (3) Three weeks (3W) after the end of exercise.

Fig 2

Baseline taxonomic composition. Relative abundance (%) of top five genera (colors) in each subject at baseline (before the start of exercise) for (A) hand samples (n = 19) and (B) forearm samples (n = 8). Subjects are ordered by decreasing Staphylococcus abundance, the most dominant genus. Other: all other genera combined, each having less than 2.5% relative abundance.

Fig 3

Alpha diversity. Box plots of (unadjusted) bacterial evenness represented by Shannon and Inverse Simpson (InvSimpson) metric across the three sampling rounds for each skin site. (A) Hand samples (n = 51), (B) forearm samples (n = 30). Colors indicate sample rounds (gray = Baseline, blue = After exercise, gold = 3 weeks post-exercise). Boxes display interquartile range (IQR), the line is the median, and whiskers extend to 1.5 × IQR, with outliers shown beyond. Each black dot represents an individual subject. Statistical difference between rounds found by pairwise comparison of estimated marginal means after adjusting for the batch in a linear mixed model is marked by asterisks (significance levels: *: 0.05, **: 0.01, ***: 0.001).

Fig 4

Temporal beta diversity. PCoA plot of Bray-Curtis dissimilarity between the three sampling rounds for hand (A) and forearm (B) samples, with 95% confidence interval ellipses. Each dot is colored according to sample round (gray = Baseline, blue = After 10-day exercise, gold = 3 weeks after end of exercise), and dashed lines connecting samples from the same individual. The two axes show the principal components explaining the highest percentage of variances [denoted in brackets] in the communities. ANOSIM R values (range −1 to 1) indicate the magnitude of community variation attributable to sample rounds, with higher values reflecting greater differences between than within rounds. Significant differences between rounds were tested with ANOSIM (999 permutations).

Fig 5

Inter- and intra-individual variation. Box plots of distances within (A and B) and between (C and D) individuals over time. Boxes display interquartile range (IQR), the line is the median, and whiskers extend to 1.5 × IQR, with outliers shown beyond. Colors indicate sample rounds if not otherwise specified (gray = Baseline, blue = After exercise, gold = 3 weeks post-exercise). Statistical significance: *0.05, **0.01, ***0.001). (A) Bray-Curtis distance between samples from the same individual over time (intra-individual), within hands (pink) and forearms (blue). (B) Bray-Curtis distance between hands and forearms (H-to-F) within each individual for each round. (C) Bray-Curtis distances between individuals in each round. (D) Beta dispersion, distance to centroid for each individual in each round.

Fig 6

Differentially abundant taxa. (A) and (B) Differential abundance comparing sample rounds for (A) hands (n = 51) and (B) forearms (n = 30). The top 15 differentially abundant ASVs for each skin site are shown, colored by phylum, and ordered by the effect size (log2 fold change ±SE) for the first contrast (post-exercise vs baseline). ASVs with absolute log2 fold change >0.2 and FDR < 0.1 are included. Opaque points indicate significantly different ASVs in each comparison. (C) and (D) Heatmaps of the abundance of ASVs (matched by row with A and B) with log10 +1 transformed counts (from low gray to high red) across samples (columns), hierarchically clustered (Bray-Curtis, Ward-D2) by sample round (gray = baseline, blue = post-exercise, gold = 3 weeks post-exercise) for (C) hands and (D) forearms.