Altitude and fungal diversity influence the structure of Antarctic cryptoendolithic Bacteria communities

Abstract

Endolithic growth within rocks is a critical adaptation of microbes living in harsh environments where exposure to extreme temperature, radiation, and desiccation limits the predominant life forms, such as in the ice-free regions of Continental Antarctica. The microbial diversity of the endolithic communities in these areas has been sparsely examined. In this work, diversity and composition of bacterial assemblages in the cryptoendolithic lichen-dominated communities of Victoria Land (Continental Antarctica) were explored using a high-throughput metabarcoding approach, targeting the V4 region of 16S rDNA. Rocks were collected in 12 different localities (from 14 different sites), along a gradient ranging from 1000 to 3300 m a.s.l. and at a sea distance ranging from 29 to 96 km. The results indicate Actinobacteria and Proteobacteria are the dominant taxa in all samples and defined a 'core' group of bacterial taxa across all sites. The structure of bacteria communities is correlated with the fungal counterpart and among the environmental parameters considered, altitude was found to influence bacterial biodiversity, while distance from sea had no evident influence.

Figure 1.

Relative abundances of the dominant bacterial OTUs in the cryptoendolithic communities in Victoria Land, Antarctica. Abundances based upon sequence taxonomy classified at the rank of Phylum.

Figure 2.

Heat map of the ‘core’ taxa relative abundance and UPGMA hierarchical clustering of sites. Values are scaled (log-transformed) by OTUs relative abundances across all sites. Abundances are indicated by the color intensity: dark and light red indicate higher relative abundances; orange and pale-orange indicate lower relative abundances. Yellow indicates a frequency < −1. Both the ‘core’ OTUs and sites were clustered using a Bray-Curtis index.

Figure 3.

Pareto-Lorenz distribution curves based on the number of OTUs and their frequencies. The dashed vertical line at the 0.2 x-axis level is plotted to evaluate the range of the Pareto values. Each line represents a sampling site.

Figure 4.

Spearman’s correlation ranks of biodiversity indices (Richness, Shannon’s diversity and Simpson’s dominance indices) correlated to the altitudinal gradient.

Figure 5.

Linear regression correlation between bacterial and fungal biodiversity (Fig. 5a) and fungal and bacterial composition distances (Bray-Curtis distances, usign Hellinger transformed OTUs tables). Correlation was also tested with Mantel test (<0.01).