Response of the human gut and saliva microbiome to urbanization in Cameroon

Abstract

Urban populations from highly industrialized countries are characterized by a lower gut bacterial diversity as well as by changes in composition compared to rural populations from less industrialized countries. To unveil the mechanisms and factors leading to this diversity loss, it is necessary to identify the factors associated with urbanization-induced shifts at a smaller geographical scale, especially in less industrialized countries. To do so, we investigated potential associations between a variety of dietary, medical, parasitological and socio-cultural factors and the gut and saliva microbiomes of 147 individuals from three populations along an urbanization gradient in Cameroon. We found that the presence of Entamoeba sp., a commensal gut protozoan, followed by stool consistency, were major determinants of the gut microbiome diversity and composition. Interestingly, urban individuals have retained most of their gut eukaryotic and bacterial diversity despite significant changes in diet compared to the rural areas, suggesting that the loss of bacterial microbiome diversity observed in industrialized areas is likely associated with medication. Finally, we observed a weak positive correlation between the gut and the saliva microbiome diversity and composition, even though the saliva microbiome is mainly shaped by habitat-related factors.

Figure 1

Description of the urbanization gradient. (A) Map of the sampling area and sampling design; numbers in brackets show the number of gut (GM) and saliva (SM) microbiomes included in the analysis. FAMD (Factor analysis of mixed data) of non-dietary contextual information showing (B) individuals colored by urbanization level (rural, semi-urban, urban) and (C) variables that significantly contributed to the construction of depicted axes; ordered factors were coded both as numerical (num.) and categorical (cat.) variable. Corresponding plots showing the 1^st and the 3^rd axis and the ordinations based on dietary can be found in Supplementary Figs. 1–3.

Figure 2

Alpha diversity of the gut and saliva microbiome along the urbanization gradient. Effective number of ASVs (A,C,E,G; non phylogenetic indices) and lineages (B,D,F; phylogenetic indices). Asterisk denotes significant directional changes along the gradient. ⁺Indicates the group mean.

Figure 3

The most important predictors of ASV diversity for the gut and saliva microbiome selected by random forest regression. The values represent the means of 10 model replicates. Depicted variables were selected by kmeans clustering of the variable importance values as described in Methods section, only for the models with R-squared > 0.1. Both non-phylogenetic (D) and phylogenetic (D_ph) diversity indices are shown. The order (⁰, ¹, ² and ^Inf) represents the weight given to taxa abundances. Detailed description of diversity indices can be found in Table 1 and Methods. Corresponding plots showing the alpha diversity predictors for the rural-urban data subsets can be found in Supplementary Fig. 4. Complete results are in Supplementary Table 3.

Figure 4

Gut microbiome composition. PCA showing: (A) individuals colored by urbanization level (rural, semi-urban, urban) and with shapes corresponding to individual enterotypes (F1–F4); (B) contextual variables selected by ordistep and ASVs associated either with these variables or with the ordination axes (only the ASVs and variables within the lower or upper 2.5% quantile are shown, for all variables see Supplementary Tables 7 and 8); arrows denote ASVs or numerical variables, factors are represented by their levels’ names placed at the centroid of individuals of the given category; (C) variation in community composition explained by different factors selected by envfit (right, each variable tested independently) and ordistep (left, the factors retained in the best non-redundant model); dr = 24 h-recall. Corresponding plots for the 1^st and 3^rd axis and for the rural-urban data subsets are in Supplementary Fig. 6.

Figure 5

Saliva microbiome composition. PCA showing: (A) individuals colored by urbanization level (rural, semi-urban, urban) and with shapes corresponding to individual stomatotypes (S1 & S2); (B) contextual variables selected by ordistep and ASVs associated either with these variables or with the ordination axes (only the ASVs and variables within the lower or upper 2.5% quantile are shown, for all variables see Supplementary Tables 7 and 8); arrows denote ASVs or numerical variables, factors are represented by their levels’ names placed at the centroid of individuals of the given category; (C) variation in community composition explained by different factors selected by envfit (right, each variable tested independently) and ordistep (left, the factors retained in the best non-redundant model); dr = 24 h-recall. Corresponding plots for the 1^st and 3^rd axis and for the rural-urban data subsets are in Supplementary Fig. 6.

Figure 6

Relationship between gut and saliva microbiome. (A) Correlation between ASV alpha diversity measures (summarized in Table 1) between gut and saliva microbiomes of the same individual (n = 95); (B) average Bray-Curtis dissimilarities and (C) average ASV turnover (corrected for differences in species richness, see Methods) between gut and saliva microbiome of the same individual, for each population separately. Cross denotes mean.