Phylogenetic diversity and metabolic potential of prokaryotic communities in permafrost and brine pockets of perennially frozen Antarctic lakes (Northern Victoria Land)

Permafrost can be defined soil material which lies at temperature below 0 °C for at least two years due to a continuously frozen state. The depth of permafrost table is variable and changes depending on the seasonally temperature. Permafrost table should be not considered as a unique layer, as it is stratified in active layer, talik and brine. Active layer is the portion of soil above the permafrost table. It is called active because its physic status is seasonally modified by thawing and freezing changes, thus becoming active in interconnection with the atmosphere. This layer plays an important role in cold regions because most ecological, hydrological, biogeochemical activities take place within it. This depth varies during different seasons and locations, from 2 cm in the coldest area to 100 cm in the warmer area. Talik is a layer of unfrozen ground in a permafrost area. Talik may have temperatures above 0 °C or below 0 °C. Brines are amounts of liquid water, which is a salt water basin within permafrost table formed during winter season. The brine formation depends on the increase of salt concentration in groundwaters under or behind permafrost, combined to evaporation and halite dissolution processes that generate the formation of veins and pockets of salt liquid that does not frozen under 0 °C. Permafrost is considered as an extreme environment due to its physiochemical features, namely low temperature, oligotrophic nature of sediments and water availability. Microorganisms that live in this environment are generally psychrophiles, which had successfully colonized all cold environments adapting their enzymes to permit them to survive and function in extreme habitats. The study of the permafrost and brine is therefore important to understand microorganism adaptations to extreme environments. For these reasons, the aim of the present Thesis was the study of structure and functions of the prokaryotic communities inhabiting Antarctic permafrost and brines. Permafrost samples were collected during different Antarctic campaigns from three sites (i.e. Edmonson Point, Boulder Clay and Dry Valleys) at different depths, while brine samples derived from lakes at Tarn Flat and Boulder Clay (in the inland and on the coast, respectively). Culture-independent (e.g. NGS sequencing of 16S rRNA genes and CARD-FISH) and culture-dependent (e.g. evaluation of best isolation method using basal and rich media at different concentrations of nutrient, identification and phenotypic characterization of bacterial isolates) approaches were used to investigate for the first time the composition, activity and adaptation of microbial assemblages in permafrost samples collected in Antarctica. The study was divided in two different chapters: the first one dealing with permafrost samples and the second one on brine samples. The first objective of this work was the analysis of the prokaryotic community inhabiting permafrost samples and the comparison of the different kind of samples among the communities. For this purpose, permafrost samples of different geomorphological eras were selected, from active layer to the oldest samples to ice permafrost. The samples were collected at Edmonson Point (active layer, EP), Boulder Clay (BC-1, BC-2 and BC-3) and Dry Valleys (oldest permafrost, DY). The culture-independent approach was applied for the sole active layer (EP). Such layer was chosen because it is continually in inter-connection with the atmosphere and changes seasonally due to the alternation between the thawed and frozen states, becoming an ecological niche colonized by diverse and functionally cold-adapted microbial assemblages, which adapt themselves to the seasonally changes of temperature, water availability and ice presence. The results showed a total of 330 OTUs distributed in 10 different bacterial phyla, with the predominance of Proteobacteria and Actinobacteria, followed by the Acidobacteria, Nitrospirae, Chloroflexi, Firmicutes and Bacteroidetes. The Gemmatimonadetes, Chlorobi and Cyanobacteria constituted a minor component in the sample EP. Differences in the relative abundances were observed for sequences affiliated to proteobacterial classes as they were mainly referred to the Alpha-, Beta-, and Gammaproteobacteria, whereas the Delta- and Epsilonproteobacteria were less represented. A number of sequences were not identified at phylum level. At genus level, Lactobacillus (among Firmicutes), Nitrospira (among Nitrospirae), Marmoricola and Propionibacterium (among Actinobacteria), were the more represented. The culture-dependent approach was used with two strategies of cultivation, direct plating of diluted cell suspensions on agar media and enrichment in liquid media before plating. All the isolates were phylogenetically affiliated. Enumeration of cultivable heterotrophic bacteria showed an order of 103 CFU g-1 by direct plating, while after enrichment in TSB at different strengths, viable counts were three to six orders of magnitude higher than those obtained by direct plating. Overall, the cultivable bacteria were distributed within five different taxa, with the predominance of Firmicutes, followed by Actinobacteria, Gammaproteobacteria, Alphaproteobacteria and Betaproteobacteria. Members in these two latter groups were isolated only after enrichment, with the Alphaproteobacteria that were obtained only from BC samples and Betaproteobacteria only from sample EP. At genus level, the strains were affiliated mainly to Bacillus, Arthrobacter and Sporosarcina. The cultivation on R2 Agar medium, an oligotrophic medium, was the best method for the recovery of bacteria from Antarctic permafrost samples. Diluted media (i.e. TSA50 and TSA1) generally yielded lower numbers of bacterial colonies than full strength media (i.e. TSA100 and R2A). The molecular approaches were useful to recovery a high number of sequences. The phyla retrieved were typical of soil, and some sequences were related to unknown or unclassified bacteria. This could suggest that they are unique to Antarctic soils or that similar environments globally have not been microbiologically well characterized. The second part of work was the analysis of prokaryotic communities inhabiting brines using culture-independent and culture-dependent approaches. The samples were collected from two Boulder Clay lakes (BC1, BC2 and BC3) and in a Tarn Flat lake (TF4 and TF5), at different depths. TF4 and TF5 brines were collected from the same borehole, while BC1 and BC2 from the same lake and BC3 in a second lake in the Boulder Clay site. The abundance and diversity, using microscopy, was detected using DAPI staining and CARD-FISH methods and was in the range of 106-107 cells mL-1. The CARD-FISH analysis showed that Proteobacteria and Bacteroidetes were predominant in similar percentage, except for BC2 that presented Bacteroidetes as the only predominant phylum. The NGS sequencing method was used for the analysis of the total bacterial community, the active community (BC1, BC2 and BC3 only) and the Archaeal community. For the total community, TF5 was the highest diverse sample (Shannon index of 1.981) followed by BC1 (1.783), whereas BC2 was the lowest diverse sample (1.087). Total bacterial community composition was similar in phyla percentage and was represented by Proteobacteria followed by Bacteroidetes and Actinobacteria. At genus level, Flavobacterium and Algoriphagus were best represented in Boulder Clay samples, while Ulvibacter, Marichromatium, Marinobacter and Shewanella were best represented in Tarn Flat samples. The active bacterial community was represented by Bacteroidetes, followed by Proteobacteria, Actinobacteria and Firmicutes. At genus level, Caldimicrobium (among Thermodesulfobacteria), Flavobacterium and Algoriphagus (among Bacteroidetes) and Lebetimonas (among Epsilonproteobacteria) were predominant. The Archaeal community structure was highest in BC1 (Shannon index 1.178) al lowest in BC2 (0.4084) and was represented by Euryarchaeota, followed by Crenarchaeota and Ancient_Archaeal_Group. The predominant order were Methanopyrales, Methanobacteriales and Methanosarcinales, anaerobic methanogens often isolated from hydrothermal vents. The statistical analysis showed a similarity between TF4 and TF5 and BC2 and BC3, while BC1 was totally distant from the other samples. The culture-dependent methods showed a CFU abundance of 103 and 223 isolates that were shared between Boulder Clay and Tarn Flat samples. The phylogenetic affiliation of isolates showed a high presence of Proteobacteria followed by Actinobacteria and Firmicutes. The genera mainly represented were Pseudomonas, Psychrobacter, Marinobacter, Leifsonia, Carnobacterium and Sporosarcina. These strains were assayed for different capabilities and potentials, different growth conditions (pH, NaCl percentage, temperature), production of extracellular enzymes (catalase, oxidase, hemolysis, DNAse), hydrolysis of complex substances (agarase, lipase, gelatinase, amylase, chitinase), antibiotic susceptibility, pollutant tolerance and biotechnological potential (EPS production, pollutant degradation, inhibitory activity). Numerous isolates grew in the presence of NaCl 19 %, mainly affiliated to Psychrobacter, all strains were able to growth between 6-7 to 9 pH range and all strains were able to growth at 4 and 15 °C (only Marinobacter sp. at 4 °C), 38 strains from TF and all strains from BC were able to grow at 25 °C. Few strains were able to tolerate antibiotics, in particular ampicillin was the best tolerated, while for heavy metals nickel and copper were the best tolerated metals, instead cadmium were no tolerated. EPS production was showed by seven strains, mainly belonging to Pseudomonas sp., while ten strains were able to degrade Aroclor 1242 at 4 and 15 °C (mainly belonging to Rhodobacter sp.). Two strains were further assayed for biodegradative efficiency and about 80-90 % of congeners were removed. One aim of this work was to analyze the viable bacterial community in brine samples in order to understand the life in such extreme and harsh environment, which has been poorly studied. In conclusion, this study allowed to compare extreme habitats, such as Antarctic permafrost and brines, using culture-dependent and culture-independent approaches. The use of different approaches was helpful to have a more complete profile of the present situation in these environments and to understand the past matters through more detailed analysis. These environments represent the terrestrial model of such exobiological niches with their unique halotolerant and aerobic psychrophilic community, that is a possible model for extraterrestrial life.