Unique Geothermal Chemistry Shapes Microbial Communities on Mt. Erebus, Antarctica

Abstract

Mt. Erebus, Antarctica, is the world's southernmost active volcano and is unique in its isolation from other major active volcanic systems and its distinctive geothermal systems. Using 16S rRNA gene amplicon sequencing and physicochemical analyses, we compared samples collected at two contrasting high-temperature (50°C-65°C) sites on Mt. Erebus: Tramway Ridge, a weather-protected high biomass site, and Western Crater, an extremely exposed low biomass site. Samples were collected along three thermal gradients, one from Western Crater and two within Tramway Ridge, which allowed an examination of the heterogeneity present at Tramway Ridge. We found distinct soil compositions between the two sites, and to a lesser extent within Tramway Ridge, correlated with disparate microbial communities. Notably, pH, not temperature, showed the strongest correlation with these differences. The abundance profiles of several microbial groups were different between the two sites; class Nitrososphaeria amplicon sequence variants (ASVs) dominated the community profiles at Tramway Ridge, whereas Acidobacteriotal ASVs were only found at Western Crater. A co-occurrence network, paired with physicochemical analyses, allowed for finer scale analysis of parameters correlated with differential abundance profiles, with various parameters (total carbon, total nitrogen, soil moisture, soil conductivity, sulfur, phosphorous, and iron) showing significant correlations. ASVs assigned to Chloroflexi classes Ktedonobacteria and Chloroflexia were detected at both sites. Based on the known metabolic capabilities of previously studied members of these groups, we predict that chemolithotrophy is a common strategy in this system. These analyses highlight the importance of conducting broader-scale metagenomics and cultivation efforts at Mt. Erebus to better understand this unique environment.

Keywords: Antarctica; Mt. Erebus; geothermal; microbial community analysis; thermal gradient.

Figure 1

(A) Satellite image of Antarctica (top left inset, red rectangle indicates Victoria Land); main image shows Victoria Land and Ross Island, with Mt. Erebus (−77.53 and 167.17 DD) identified in the red rectangle. (B) The sampling sites on Mt. Erebus (red rectangles), Tramway Ridge (−77.519444 and 167.116389 DD) and Western Crater (−77.519167 and 167.187778 DD), and other main geothermal features on the Mt. Erebus summit. Satellite image in (A) is from the Antarctic Digital Database Map Viewer https://www.add.scar.org/, Open Source. Satellite image in (B) was purchased from DigitalGlobe Incorporated, Longmont CO United States (2019).

Figure 2

Comparison of soil physicochemistry between Tramway Ridge (TR) and Western Crater (WC). (A,C) Boxplots for (A) each environmental parameter and (C) elements of interest that were significantly different between the two sites, showed significant trends with temperature, or are potential substrates for microbial metabolism. (B) Line plots showing the variation with temperature of environmental parameters and elements that were significantly correlated (Spearman’s correlation coefficient, p < 0.05) with temperature at one of the two sites; all significant correlations were found at WC. Elements were grouped by their patterns of difference between sites. * indicates significant difference between samples from the Western Crater transect and both Tramway Ridge transects (two-sided t-test, p < 0.05). Ppb: concentration in parts per billion. GWC: gravitational water content. TC: total carbon; TN: total nitrogen.

Figure 3

A distance-based Redundancy Analysis (dbRDA) of 16S rRNA gene ASV profiles, which were center log-ratio transformed and constrained by the physicochemical parameters identified as significantly (Mantel test, p < 0.05) affecting the microbial community composition either (A) between Western Crater and Tramway Ridge, or (B) within Tramway Ridge. Ordination was constructed using Unifrac distances among samples. The shape represents the site/transect of origin for the sample, color represents the pH of the sample as pH was identified as being most important for distinguishing samples between sites. The percentages on the axes correspond to the variance in biological samples explained by the combination of the selected environmental parameters. The arrow direction for each environmental factor represents the variance direction for that parameter. TR: Tramway Ridge site; WC: Western Crater site; TR1: Tramway Ridge transect 1; TR2: Tramway Ridge transect 2; GWC: gravitational water content; EC: electroconductivity; TC: total carbon; TN: total nitrogen.

Figure 4

Relative abundance of either (A) Archaea or Bacteria and (B) most abundant phyla (> 1%) at the three transects studied. Samples are ordered on the x-axis in order of increasing temperature within transect. TR1: Tramway Ridge transect 1; TR2: Tramway Ridge transect 2; WC: Western Crater transect.

Figure 5

Center log-ratio transformed (CLR) abundance data for the ASVs identified by the Random Forest algorithm as being most important (mean decrease in Gini value, MDG, > 0.06) for discriminating between the microbial communities at the three transects. ASVs are ordered by domain, then phylum, in ascending order. The lowest assignable taxonomic assignments are also given. The bottom four ASVs are Archaea, the rest are Bacteria. On the x-axis, samples are ordered by increasing temperature within each transect. TR1: Tramway Ridge transect 1; TR2: Tramway Ridge transect 2; WC: Western Crater transect; The full taxonomy of these ASVs is given in Supplementary Table 4.

Figure 6

(A) Visual representation of the undirected co-variation network of the trimmed ASV data set from all samples. Nodes and lines are color-coded by module, as defined by the igraph fast greedy modularity optimization algorithm. Node size reflects the mean center log-ratio abundance of the ASV across all samples. (B) Heat map of the relative contribution of ASVs from each of the modules to the total microbial community in each sample, clustered by transect. The x-axis is ordered by increasing sample temperature within transect. TR1: Tramway Ridge 1 transect; TR2: Tramway Ridge 2 transect; WC: Western Crater transect. (C) Microbial composition of each module from the network analysis ordered alphabetically. Size of the bars reflects the relative abundance of that class within the module. Only classes with a relative abundance above 2% are plotted for ease of visualization.