Patterns in the distribution of soil bacterial 16S rRNA gene sequences from different regions of Antarctica
Highlights
► We surveyed 16S rRNA gene data from 35 Antarctic soil environments. ► We examine taxonomic patterns in relation to geographic regions and soil parameters. ► Most Antarctic soil environments contained a narrow range of bacterial species. ► Soil from different Antarctic regions show different bacterial composition. ► The observed patterns coincide with Gressit Line biogeographical boundary.
Introduction
Bacteria are a major component of the soil biosphere, and it has also been estimated that 1 g of soil might contain up to ~ 104 different bacterial species (Torsvik et al., 2002). These bacteria are heavily involved in biogeochemical cycles (Beare et al., 1995, Overmann and van Gemerden, 2000, Rosswall, 1982, White et al., 1998) and are vital in maintaining soil ecological function (Beare et al., 1995, Torsvik et al., 2002). The functional roles of bacteria are particularly significant in Antarctica, where terrestrial trophic interactions are generally dominated by microorganisms (Davis, 1981, Hogg et al., 2006, Vincent, 2000) due to harsh environments restricting the development of higher plant and animal communities.
Prior to the advent of molecular techniques, the study of soil bacterial diversity was dependent on culturing approaches and morphological characterization (e.g. Baker, 1970, Bölter, 1992, Smith and Tearle, 1985, Williams, 1969). Despite providing valuable information about morphology and physiological factors such as enzymatic characteristics, the community patterns and levels of diversity inferred from culture-dependent methods were inevitably underestimated and inaccurate, being skewed towards the few easily cultivated species and, hence, unlikely to be representative of the natural communities from which they were obtained (Coleman and Whitman, 2005, Torsvik and Ovreas, 2002). However, over the last two decades, PCR-based 16S rDNA molecular approaches have become the main analytical methods used in the study of soil bacterial ecology and diversity (Kirk et al., 2004, Nocker et al., 2007, O'Donnell and Görres, 1999). These methods require very little starting material and are relatively faster than culture-based methods. More importantly, the PCR amplification strategy can be carried out without the need to obtain pure isolates. Even though these methods still do not address well the issues of relative abundance or activity in natural communities (unless used in conjunction with Quantitative-PCR and Reverse Transcriptase-PCR) (Freeman et al., 1999), they provide a better estimation of the overall diversity, as uncultured members can be incorporated into the diversity assessments.
To date, although only a fraction of the molecular assessments of soil have provided 16S rRNA gene sequence information (while there are an increasing number of molecular Antarctic studies, few have been based on 16S rDNA), availability of such data does allow a basic understanding of the phylogenetic diversity and composition of soil bacterial communities. For example, increasing evidence suggests that, at a higher taxonomic level (i.e. phylum and class), the soil bacterial community composition can be highly stable (Janssen, 2006, Youssef and Elshahed, 2008). In contrast, at lower taxonomic levels (i.e. genus and species), community composition might be sensitive to both spatial and environmental gradients (Green and Bohannan, 2006, Martiny et al., 2006).
In comparison to the traditional culture-dependent approach, most 16S rDNA based diversity studies have suggested that the Antarctic soil bacterial community is more diverse than was previously thought (e.g. Aislabie et al., 2008, Cowan et al., 2002, de la Torre et al., 2003, Niederberger et al., 2008, Yergeau et al., 2009). Nevertheless, little is known about the patterns in distribution of soil bacteria across Antarctica. In particular, it remains unclear whether regionalisation within Antarctica is an important feature of soil bacterial distributions, as their taxonomic composition is largely dependent on habitat characteristics regardless of geographical position. For example, consistent findings that Antarctic dry mineral soils are dominated by members of Acidobacteria and Actinobacteria have been reported from Signy Island (maritime Antarctic), Alexander Island (maritime/continental transition zone) and the McMurdo Dry Valleys (continental Antarctic) (Chong et al., 2010, Chong et al., 2012a, Pointing et al., 2010). However, such simple taxonomic comparisons are inconclusive, as one or two selected phyla may not be appropriate to reveal the true phylogenetic variation across different regions. With the increasing sequencing effort applied to Antarctic soil bacterial studies over the past few years, we are now at a stage where it is possible to initiate detailed phylogenetic analysis to search for taxonomic patterns in the sequence data obtained from different regions of Antarctica.
In this study, we integrated sequence information reported from 13 studies that, together, obtained samples from 35 sites ranging from 51°S to 78°S, representing a more or less linear transect down the Scotia arc/Antarctic Peninsula, through the continental interior, and up along the Transantarctic Mountains to Victoria Land as far as Cape Hallett (Fig. 1). The range of Victoria Land locations effectively encompasses the area of interest of the international ‘Latitudinal Gradient Project’ (LGP, most recently overviewed by Howard-Williams et al., 2010), with these authors specifically recognising the emergent observation ‘that studies covering very broad ranges, extending from continental through maritime and sub-Antarctic regions show clear patterns of change in biological communities’. The current study therefore provides an apposite microbiological extension of the approach of the LGP. Ideally, this approach would be extended to include the rest of the Antarctica, but this is currently not possible due to the lack of available sequence data that that are consistent with our selection criteria, other than from the single site at Schimarcher Oasis.
Most previous studies of Antarctic soil bacterial community composition have been based on a quantitative species-based approach (sensu Lozupone and Knight, 2008), where the diversity of a particular site was enumerated based on the number of detected taxa, while the relationships between sites were compared based on the overlap of well-defined taxonomic units (e.g. Aislabie et al., 2008, Cannone et al., 2008, Yergeau et al., 2007b). As our database was constructed from multiple studies that utilized different assessment methods, this type of quantitative species-based comparison was not possible. Therefore, we opted for a qualitative, divergence-based, approach to detect the possible biogeographic relationships of the soil bacterial communities studied. Generally, sites which harboured a higher fraction of distantly related taxa were regarded as more diverse, while those which shared greater similarity in their bacterial lineages were considered more similar. Our first objective was to assess if a compilation of the currently available Antarctic soil 16S rRNA gene sequences would be sufficient to reveal general ecological patterns, for example, addressing the question of whether the taxonomic distribution of bacteria across Antarctica is random or non-random. Second, we examined if the detected distribution pattern could be correlated with previously recognised geographical regions from studies of other groups, latitude or the underlying soil chemical parameters associated with the various study sites.
Section snippets
Antarctic soil bacterial sequence selection
We obtained GenBank accession numbers (Benson et al., 2010) of soil bacterial sequences from 13 published Antarctic articles (Aislabie et al., 2008, Aislabie et al., 2009, Chong et al., 2012a, Chong et al., 2012b, Foong et al., 2010, Ganzert et al., 2011, Niederberger et al., 2008, Pointing et al., 2010, Shivaji et al., 2004, Shravage et al., 2007, Smith et al., 2006, Xiao et al., 2007, Yergeau et al., 2007b). These studies were selected based on the criteria that (a) the study samples were
Results
In this study, we included 13 surveys (Table 1, Fig. 1) that examined the molecular diversity of soil bacterial communities in Antarctica. We employed a strict filter to exclude studies that only focused on a specific taxonomic group or were based on culture-dependent methods.
We tested the “completeness” of our sequence data by searching the GenBank database (http://www.ncbi.nlm.nih.gov) using the keywords “Antarctic AND 16S AND soil” (accessed 08 Nov 2011). A total of 30,721 hits were
Environmental influences
Although the spatial coverage of Antarctic soil studies remains patchy and limited to date (Chown and Convey, 2007, Frenot et al., 2005, Hogg et al., 2006, Tindall, 2004, Wall, 2005), most have suggested that soil bacterial distribution in Antarctica is not random, but is highly sensitive to differences in environmental conditions. It has been suggested that variations in Antarctic soil bacterial community diversity and composition are mainly attributable to differences in soil pH, heavy metal
Conclusion
In conclusion, we have demonstrated that 16S rRNA gene sequences currently available from the Antarctic soil environment are sufficient to show some pattern in relation to both small-scale local soil parameters and large-scale geographical position. Although limited by spatial coverage and number of available sequences, our data provide the first suggestion that Antarctic bacterial diversity might be structured in a similar large-scale manner to that of Antarctic terrestrial eukaryotes.
The
Acknowledgements
The authors would like to acknowledge Dr. Katrin Zwirglmaier for advice on the ARB software. We thank two anonymous reviewers for constructive and helpful comments. This work is supported by the National Antarctic Research Center, University of Malaya, and also contributes to the British Antarctic Survey's ‘Polar Science for Planet Earth’ and the Scientific Committee on Antarctic Research's ‘Evolution and Biodiversity in Antarctica’ research programmes.
References (87)
- et al.
Relation between soil classification and bacterial diversity in soils of the Ross Sea region, Antarctica
Geoderma
(2008) - et al.
Biotic and abiotic factors influencing soil properties across a latitudinal gradient in Victoria Land, Antarctica
Geoderma
(2008) - et al.
High levels of spatial heterogeneity in the biodiversity of soil prokaryotes on Signy Island, Antarctica
Soil Biology and Biochemistry
(2010) - et al.
Linking species richness, biodiversity and ecosystem function in soil systems
Pedobiologia
(2005) - et al.
Metagenomic analyses of the dominant bacterial community in the Fildes Peninsula, King George Island (South Shetland Islands)
Polar Science
(2010) - et al.
Spatial scaling of microbial biodiversity
Trends in Ecology & Evolution
(2006) - et al.
Biotic interactions in Antarctic terrestrial ecosystems: are they a factor?
Soil Biology and Biochemistry
(2006) - et al.
Methods of studying soil microbial diversity
Journal of Microbiological Methods
(2004) - et al.
16S rDNA methods in soil microbiology
Current Opinion in Biotechnology
(1999) - et al.
Microbial interactions involving sulfur bacteria: implications for the ecology and evolution of bacterial communities
FEMS Microbiology Reviews
(2000)