Thermophilic microorganisms possess several adaptations to thrive in high temperature, which is reflected as biosynthesis of proteins and thermostable molecules, isolation and culture represent a great methodological challenge, therefore High throughput sequencing enables screening of the whole bacterial genome for functional potential, providing rapid and cost-effective information to guide targeted cultures for the identification and characterization of novel natural products. In this study, we isolated two thermophilic bacterial strains corresponding to Bacillus LB7 and Streptomyces LB8, from the microbial mats in the Atacama Desert. By combining genome mining, targeted cultures and biochemical characterization, we aimed to identify their capacity to synthesize bioactive compounds with antimicrobial properties. Additionally, we determined the capability to produce bioactive compounds under controlled in vitro assays and detected by determining their masses by Thin-Layer Chromatography/Mass Spectrometry (TLC/MS). Overall, both isolates can produce antimicrobial (e.g., Myxalamide C by-product) and antioxidants (e.g. Dihydroxymandelic Acid, Amide biotine and Flavone by-products) compounds. Bacillus LB7 strain possesses a more diverse repertoire with 51.95% of total metabolites unmatched, while Streptomyces LB8 favors mainly antioxidants, but has over 70% of unclassified compounds, highlighting the necessity to study and elucidate the structure of novel compounds. Based on these results, we postulate that the uncultured or rare cultured thermophiles inhabiting high-altitude hydrothermal ecosystems in the Atacama Desert offer a promising opportunity to the study of novel microbial bioactive compounds.

Microbial community structures are shaped by geochemical factors and their interactions with the lithosphere, hydrosphere, and atmosphere through the processes of chemical mobilisation and mineralisation. High-altitude wetlands and salt flats in the central Andes are characterised by pronounced physicochemical gradients and extreme climatic conditions, representing hotspots of microbial diversity. We here hypothesise about the existence of direct relationships between the local microbiology and the climate cyclicity variables based on meteorological and biogeochemical patterns that develop over a short time scale (five years). We have here analysed the interactions between hydrometeorological and biogeochemical variables and the microbial communities of the Salar de Huasco. These results were obtained by correlating 16S cDNA and DNA gene Illumina sequences with meteorological/satellite data collected both at monitoring stations and by remote sensing between the years 2015 and 2020. The precipitation levels and flooded areas (i.e., areas covered and/or saturated with permanent water) detected in the Salar de Huasco revealed a marked hydric cyclicity that correlated seasonally with intra-annual wet and dry seasons. Overall, at this site, wet periods occurred from December to April, and dry periods from May to November. Meteorological variables such as solar radiation, air temperature, relative humidity, wind speed, atmospheric pressure, and wind direction were well-defined, showing a potential association with the hydrogeology of the area, which is directly related to the wetlands' flooded areas. Finally, the microbial presence and potentially active microbial communities were determined through the sequencing of the 16S gene (DNA and cDNA, respectively), this were associated with climatic seasonality and spatially distributed physical and chemical heterogeneity. Other non-local inter-annual scale processes, such as El Niño-Southern Oscillation (ENSO) events, modify the physical and chemical context of the wetland, thus forming unique ecological niches in the Andean Mountains.

Salar de Huasco is a wetland in the Andes mountains, located 3800 m above sea level at the Chilean Altiplano. Here we present a study aimed at characterizing the viral fraction and the microbial communities through metagenomic analysis. Two ponds (H0 and H3) were examined in November 2015. Water samples were processed using tangential flow filtration to obtain metagenomes from which the DNA fraction of the sample was amplified and sequenced (HiSeq system, Illumina). The ponds were characterized by freshwater and the viral-like particles to picoplankton ratio was 12.1 and 2.3 for H0 and H3, respectively. A great number of unassigned viral sequences were found in H0 (55.8%) and H3 (32.8%), followed by the family Fuselloviridae 20.8% (H0) and other less relatively abundant groups such as Microviridae (H0, 11.7% and H3, 3.3%) and Inoviridae (H3, 2.7%). The dominant viral sequences in both metagenomes belong to the order Caudovirales, with Siphoviridae being the most important family, especially in H3 (32.7%). The most important bacteria phyla were Proteobacteria, Bacteroidetes and Firmicutes in both sites, followed by Cyanobacteria (H0). Genes encoding lysogenic and lytic enzymes (i.e., recombinases and integrases) were found in H0 and H3, indicating a potential for active viral replication at the time of sampling; this was supported by the presence of viral metabolic auxiliary genes at both sites (e.g., cysteine hydrolase). In total, our study indicates a great novelty of viral groups, differences in taxonomic diversity and replication pathways between sites, which contribute to a better understanding of how viruses balance the cycling of energy and matter in this extreme environment.

Hydrothermal systems are ideal to understand how microbial communities cope with challenging conditions. Lirima, our study site, is a polyextreme, high-altitude, hydrothermal ecosystem located in the Chilean Andean highlands. Herein, we analyze the benthic communities of three nearby springs in a gradient of temperature (42–72 °C represented by stations P42, P53, and P72) and pH, and we characterize their microbial diversity by using bacteria 16S rRNA (V4) gene metabarcoding and 16S rRNA gene clone libraries (bacteria and archaea). Bacterial clone libraries of P42 and P53 springs showed that the community composition was mainly represented by phototrophic bacteria (Chlorobia, 3%, Cyanobacteria 3%, at P42; Chlorobia 5%, and Chloroflexi 5% at P53), Firmicutes (32% at P42 and 43% at P53) and Gammaproteobacteria (13% at P42 and 29% at P53). Furthermore, bacterial communities that were analyzed by 16S rRNA gene metabarcoding were characterized by an overall predominance of Chloroflexi in springs with lower temperatures (33% at P42), followed by Firmicutes in hotter springs (50% at P72). The archaeal diversity of P42 and P53 were represented by taxa belonging to Crenarchaeota, Diapherotrites, Nanoarchaeota, Hadesarchaeota, Thaumarchaeota, and Euryarchaeota. The microbial diversity of the Lirima hydrothermal system is represented by groups from deep branches of the tree of life, suggesting this ecosystem as a reservoir of primitive life and a key system to study the processes that shaped the evolution of the biosphere.

High Throughput DNA Sequencing (HTS) revolutionized the field of paleomicrobiology, leading to an explosive growth of microbial ancient DNA (aDNA) studies, especially from environmental samples. However, aDNA studies that examine environmental microbes routinely fail to authenticate aDNA, examine laboratory and environmental contamination, and control for biases introduced during sample processing. Here, we surveyed the available literature for environmental aDNA projects—from sample collection to data analysis—and assessed previous methodologies and approaches used in the published microbial aDNA studies. We then integrated these concepts into a case study, using shotgun metagenomics to examine methodological, technical, and analytical biases during an environmental aDNA study of soil microbes. Specifically, we compared the impact of five DNA extraction methods and eight bioinformatic pipelines on the recovery of microbial aDNA information in soil cores from extreme environments. Our results show that silica-based methods optimized for aDNA research recovered significantly more damaged and shorter reads (<100 bp) than a commercial kit or a phenol–chloroform method. Additionally, we described a stringent pipeline for data preprocessing, efficiently decreasing the representation of low-complexity and duplicated reads in our datasets and downstream analyses, reducing analytical biases in taxonomic classification.