The skin microbiota of the axolotl Ambystoma altamirani is highly influenced by metamorphosis and seasonality but not by pathogen infection

Abstract

Background: Microbiomes have been increasingly recognized as major contributors to host health and survival. In amphibians, bacterial members of the skin microbiota protect their hosts by inhibiting the growth of the fungal pathogen Batrachochytrium dendrobatidis (Bd). Even though several studies describe the influence of biotic and abiotic factors over the skin microbiota, it remains unclear how these symbiotic bacterial communities vary across time and development. This is particularly relevant for species that undergo metamorphosis as it has been shown that host physiology and ecology drastically influence diversity of the skin microbiome.

Results: We found that the skin bacterial communities of the axolotl A. altamirani are largely influenced by the metamorphic status of the host and by seasonal variation of abiotic factors such as temperature, pH, dissolved oxygen and conductivity. Despite high Bd prevalence in these samples, the bacterial diversity of the skin microbiota did not differ between infected and non-infected axolotls, although relative abundance of particular bacteria were correlated with Bd infection intensity.

Conclusions: Our work shows that metamorphosis is a crucial process that shapes skin bacterial communities and that axolotls under different developmental stages respond differently to environmental seasonal variations. Moreover, this study greatly contributes to a better understanding of the factors that shape amphibian skin microbiota, especially in a largely underexplored group like axolotls (Mexican Ambystoma species).

Keywords: Amphibians; Metamorphosis; Seasonality; Skin microbiota.

Fig. 1

Bacterial diversity of A. altamirani skin and environmental samples. A Upset plot illustrating the number of unique and shared ASVs. Numbers aside the color bars indicate how many ASVs were present on each sample type (color bars) and shared between sample types (gray bars). B Alpha Faith’s Phylogenetic diversity (PD) across sample types. C Principal coordinate analysis (PCoA) based on weighted UniFrac distances across sample types. D Beta dispersion using Analysis of multivariate homogeneity of groups dispersions. Letters a–d indicate statistically significant comparisons

Fig. 2

ANCOM results showing differentially abundant bacterial families between metamorphic and pre-metamorphic axolotls. Left panel shows ANCOM W values, the middle panel shows the relative proportion for each bacterial family, and the right panel shows the best taxonomic assignments according to the SILVA database at order (O), class (C) or family (F) level. Circles and bars are color-coded according to the host metamorphic status

Fig. 3

Seasonal influence over metamorphic and pre-metamorphic skin bacterial diversity. A Phylogenetic diversity (PD) across seasons in metamorphic samples. Letters a–d indicate statistically significant comparisons. B PD across seasons in pre-metamorphic samples. C Seasonal variation of pH, delta temperature and mean temperature of the stream water. D Principal coordinate analysis (PCoA) based on weighted UniFrac distances across seasons of metamorphic samples. E PCoA based on weighted UniFrac distances across seasons in pre-metamorphic samples. Circles in D and E panels are color-coded by season

Fig. 4

ANCOM results showing differentially abundant bacterial families in metamorphic and pre-metamorphic axolotls across consecutive seasons: autumn to winter seasons in pre-metamorphic axolotls, and winter to spring seasons for metamorphic and pre-metamorphic axolotls. From left to right: ANCOM comparisons color-coded by season, ANCOM W values, the relative bacterial family proportion and the best taxonomic assignment according to SILVA at order (O), class (C) or family (F) level. Circles and bars are color-coded by season. Shared bacterial families between metamorphic and pre-metamorphic axolotls between winter and spring seasons are shown in bold

Fig. 5

Distance based redundancy analysis of A. altamirani skin bacterial communities. Distances in the PCA are based on a weighted UniFrac distance matrix. Vector directions indicate the type of correlation of each predictor variable. Distance of each sample with respect to vectors highlight the weight of the correlation with a given predictor variable. Non quantitative variables are represented as centroids (outlined circles larger). Circles are color-coded by host metamorphic status

Fig. 6

A. altamirani skin bacterial diversity with respect to Bd infection status. A Alpha phylogenetic diversity (PD) between infected and non-infected in metamorphic axolotls. B Principal coordinate analysis (PCoA) based on weighted UniFrac distances in infected vs non-infected metamorphic axolotls. C PD between infected and non-infected in pre-metamorphic axolotls. D PCoA based on weighted UniFrac distances in infected vs non-infected in pre-metamorphic axolotls. Circles are color-coded by Bd infection status