Stratified Bacterial Diversity along Physico-chemical Gradients in High-Altitude Modern Stromatolites

Abstract

At an altitude of 3,570 m, the volcanic lake Socompa in the Argentinean Andes is presently the highest site where actively forming stromatolite-like structures have been reported. Interestingly, pigment and microsensor analyses performed through the different layers of the stromatolites (50 mm-deep) showed steep vertical gradients of light and oxygen, hydrogen sulfide and pH in the porewater. Given the relatively good characterization of these physico-chemical gradients, the aim of this follow-up work was to specifically address how the bacterial diversity stratified along the top six layers of the stromatolites which seems the most metabolically important and diversified zone of the whole microbial community. We herein discussed how, in only 7 mm, a drastic succession of metabolic adaptations occurred: i.e., microbial communities shift from a UV-high/oxic world to an IR-low/anoxic/high H₂S environment which force stratification and metabolic specialization of the bacterial community, thus, modulating the chemical faces of the Socompa stromatolites. The oxic zone was dominated by Deinococcus sp. at top surface (0.3 mm), followed by a second layer of Coleofasciculus sp. (0.3 to ∼2 mm). Sequences from anoxygenic phototrophic Alphaproteobacteria, along with an increasing diversity of phyla including Bacteroidetes, Spirochaetes were found at middle layers 3 and 4. Deeper layers (5-7 mm) were mostly occupied by sulfate reducers of Deltaproteobacteria, Bacteroidetes and Firmicutes, next to a high diversity and equitable community of rare, unclassified and candidate phyla. This analysis showed how microbial communities stratified in a physicochemical vertical profile and according to the light source. It also gives an insight of which bacterial metabolic capabilities might operate and produce a microbial cooperative strategy to thrive in one of the most extreme environments on Earth.

Keywords: 16S rRNA amplicon sequencing; UV radiation; extremophiles; high-altitude lakes; stromatolites.

FIGURE 1

Environmental setting of Socompa’s Stromatolites. (A) Sampling site at the shore of the alkaline Socompa Lake showing domical and columnar stromatolite morphologies. (B) Transversal cut of the sample with clear visible laminations. (C) Magnification on stratification showing a pinkish-white layer in the surface (ly1, orange dot), a dark green layer (ly2, green dot), a light brownish (ly3, red dot), a dark brownish layer (ly4, brown dot), a light brownish (ly5, yellow dot), and a dark brown layer (ly6, dark brown dot).

FIGURE 2

Diversity and Dominance Indexes. Observed OTUs, Chao 1 estimator, evenness, dominance, and Shannon indexes for the six layers, at 97 and 80% OTU identity, normalized with the number of sequences of the smaller dataset (17653).

FIGURE 3

Depth profiles of physico-chemical parameters and bacterial diversity. (A) Light profile; (B) oxygen, pH, and H₂S profile; (C) pigment profile adapted from Farías et al., 2013). (D) Heatmap of the 29 most abundant clusters of sequences; (E) phyla abundances for the combined layers. The clusters, represented in a dendrogram according to their pattern of distribution, displays three main branches; the dark-blue branch with key functional groups that were present virtually in all layers, although they clearly concentrate in great abundance at three different levels; the red branch showing clusters that were mostly widespread in the middle; and the gray branch with clusters located at dark anoxic layers. The phylum assignation of the clusters ID are listed down, and the ‘best hit’ taxonomic assignations are listed in the Supplementary Figure S2. Relative abundances (>0.1%) in the heatmap are indicated within the boxes in numbers and with a color key. A dendrogram of the layers shows pairs that match according to their different biological and physicochemical microenvironment.

FIGURE 4

Layer-by-layer taxonomical profile. Bacterial taxonomic assignations of all six layers expressed in percentage of the relative abundances obtained from the 16S (HV4) pyrotags clustered at 97% OTU identity level and classified against the green-genes database. Only OTUs with values ≥ 0.4% were plotted.

FIGURE 5

Scanning electron microphotography of layer 2. (A) Filamentous Cyanobacteria. (B) Magnified view of one filamentous with attached carbonate crystals and diatoms. (C–E) EDS images from the filamentous Cyanobacteria green layer with diatoms and crystals attached. Gray: Original Image 600X; Cyan: Silicium, representing diatoms; White: Calcium, representing aragonite crystals. Thick arrows: Cyanobacteria filaments; Point arrows: diatoms; Thin arrows: aragonite crystals.