Authors: Janina C. Vogt aff001; Raeid M. M. Abed aff002; Dirk C. Albach aff001; Katarzyna A. Palinska aff003 Authors place of work: Institute for Biology and Environmental Science (IBU), Biodiversity and Evolution of Plants, Carl-von-Ossietzky University of Oldenburg, Oldenburg, Germany aff001; Biology Department, College of Science, Sultan Qaboos University, Al Khoud, Muscat, Sultanate of Oman aff002; Department of Marine Biology and Ecology, Institute of Oceanography, University of Gdansk, al. Marszałka Józefa Piłsudskiego 46, Gdynia, Poland aff003 Published in the journal: PLoS ONE 14(11) Category: Research Article doi: https://doi.org/10.1371/journal.pone.0224444
Latitudinal diversity gradients are well-known for plants and animals, but only recently similar patterns have been described for some specific microbial communities in distinct habitats. Although microbial diversity is well-investigated worldwide, most of the studies are spatially too restricted to allow general statements about global diversity patterns. Additionally, methodological differences make it hard and often impossible to compare several studies. This study investigated the cyanobacterial diversity in tidal flats along geographical and ecological gradients based on high-throughput sequencing of 16S rRNA gene fragments (Illumina MiSeq) and environmental data on a large spatial scale from the subtropics to the Arctic. Latitude and strongly correlated environmental parameters (e.g. temperature) were identified as important drivers of cyanobacterial diversity on global scale resulting in a latitudinal diversity gradient similar to that known from plants and animals. Other non-correlated parameters (e.g. grain size) were shown to be more important on local scales, although no consistent pattern occurred across different locations. Among a total number of 989 operational taxonomic units (OTUs) only one cosmopolitan (classified as Coleofasciculus chthonoplastes), but many location-specific and putative endemic ones (78%) were detected. High proportions of rare members of the community (up to 86%) were found in all samples. Phylogenetic beta diversity was shown to be influenced by the developmental stage of the mat community becoming increasingly similar with increasing stabilization.
Phylogenetics – Community structure – Phylogeography – Biogeography – Sequence databases – Sediment – Latitude – Shannon index
Detailed analyses of microbial diversity have become common during the last years based on cost-efficient high-throughput sequencing (HTS) technologies, which allow deep insights into the community structure of environmental samples including the previously undetected ‘rare biosphere’. Rare microbial species have been shown to be extremely diverse and significantly contributed to the increase in richness and diversity of communities [1]. Whereas studies focusing on local communities increased our knowledge of the microbial diversity worldwide [2; 3], biogeographical patterns only start to emerge with an increasing use of HTS technologies.
In the beginning of the 20th century, Beijerinck [4] and Baas-Becking [5] established the hypothesis that ‘Everything is everywhere, but the environment selects’, which is commonly assumed for free-living microorganisms. Thus, microorganisms are supposed to be ubiquitous and disperse freely without geographical barriers or isolation as supported by studies of benthic and planktonic protists [6; 7; 8; 9]. Within the phylum Cyanobacteria such cosmopolitan distribution patterns were shown for some species e.g. Coleofasciculus (Microcoleus) chthonoplastes [10], Mastigocladus laminosus [11; 12], Cylindrospermopsis raciborskii [13; 14; 15], and Microcystis aeruginosa [16; 17]. Recent high-resolution studies of microbial diversity found evidence for unique biogeographical distribution patterns of microorganisms on different spatial scales [18; 19; 20; 21] and several molecular studies observed them for specific cyanobacterial taxa [22] e.g. in hot springs [23] and ice-covered lakes in the Antarctica [24].
Although biogeographical patterns and the related gradients in biodiversity have been well investigated in plants and animals for centuries [25; 26; 27; 28; 29; 30; 31], there is still less known about microbial and especially prokaryotic biogeography [22; 32; 33; 34; 35].
One of the oldest and most commonly observed patterns of worldwide biodiversity is the latitudinal gradient, which is characterized by decreasing species richness from the tropics to the poles. Latitudinal gradients have been demonstrated to exist in diverse higher eukaryotic taxa [36; 37; 38; 39; 40; 41; 42; 43]. Latitude and correlated parameters, such as climate (temperature), are commonly supposed to determine biodiversity for historical (age/area), evolutionary (speciation/diversification rates), and ecological reasons (ecological/physiological opportunities, distribution ranges) [43; 44]. Based on that the tropics are commonly considered to be a “cradle and museum of biodiversity” that spread out more or less successfully towards the poles [39; 44]. However, recent studies showed that high ecological opportunity and high speciation rates are not necessarily restricted to the tropics [45; 46] with one study having found evidence for habitat-specific variations within this general latitudinal diversity gradient [47].
Due to the immense population size, a tremendous dispersal potential, and diverse physiological properties of microorganisms, not all mechanisms shaping biogeography of macroorganisms, such as speciation, extinction, dispersal, and longevity, are directly adaptable to microorganisms, even if they are not fundamentally different. Environmental gradients and habitat types are supposed to become much more important for their geographical distribution than distances [32; 34; 48].
Recent studies verified latitudinal diversity gradients for terrestrial and marine/planktonic bacteria [49; 50; 51; 52] and suggested biogeographical patterns for prokaryotes similar to those described for eukaryotes according to their physiological properties [11; 32; 52]. However, most studies on microbial diversity are spatially too restricted to allow general statements about global patterns [24; 53; 54].
Tidal flats occur worldwide along low-sloped coastlines with non-vegetated mud/sand plains in sheltered/low-energy marine environments (bays, lagoons, estuaries, back-barrier of islands), or open coasts [55; 56; 57]. Strongly fluctuating environmental conditions due to tidal activity with periodic inundation and desiccation are characteristics of these highly dynamic coastal ecosystems at the transition from terrestrial to marine environments. Tidal flats provide extreme, but comparable environmental conditions at geographically distant sites differing mainly in climate conditions. Microbial communities in tidal flats usually form thin biofilms or more or less stabilized, laminated microbial mats, where microorganisms are arranged according to internal physicochemical gradients [58; 59; 60]. Cyanobacteria are important members of these benthic microbial mats as pioneer organisms, main primary producers, and stabilizers [58; 59; 60; 61]. Diverse environmental parameters influence the successive formation of such mat communities and their composition, e.g. temperature, grain size, salinity, and nutrient composition [58; 62; 63]. Since tidal flats are spatially separated without strict boundaries and inhabited by a highly diverse microbial community [61; 64; 65], they are a promising habitat for an investigation of biodiversity patterns of cyanobacterial populations.
Based on the ecological importance and variability of this habitat as well as its worldwide distribution, we investigated cyanobacterial diversity patterns along geographical and ecological gradients in different climate zones ((sub)arctic to subtropical-arid) within a defined and comparable habitat. We compared diversity measures based on high-throughput sequencing of 16S rRNA gene fragments from tidal flats at five geographically distant sites (Iceland, France, Germany, Croatia, and Oman), including samples that differed in grain size, salinity, alkalinity, and nutrient composition. This data allowed us to test the two most commonly inferred hypotheses about microbial biogeography: the “Everything is everywhere”-hypothesis and the latitudinal biodiversity gradient hypothesis, with unprecedented precision for such organisms based on comparable sampling and sequencing methodologies.
Specifically, we examined the following hypotheses: 1) cyanobacterial species from tidal flats are cosmopolitan across latitude rather than forming climate zone-specific communities; 2) the diversity of cyanobacterial communities decrease with latitude similar to that of plants and animals; 3) environmental parameters such as grain size, salinity, total alkalinity and nutrients show similar effects on community diversity across different climate zones.
Study sites were located in tidal flats in different climate zones: (sub)arctic (Iceland), temperate (Germany, France (Brittany)), subtropical-dry-summer (Croatia), and subtropical-arid (Oman) (Fig 1, S1 Table). The necessary permissions to access the study sites and collect samples there were acquired from the Nationalparkverwaltung “Niedersächsisches Wattenmeer” for the study sites in Germany and from Sultan Qaboos University within the framework of a joint scientific collaboration project for the study sites in Oman. For study sites in Iceland, France, and Croatia, the national focal points were contacted and no specific permits were required. The field studies did not involve any endangered or protected species. As previously described by Vogt et al. [62; 63], the upper part of the sediment (ca. 1 cm) including the entire phototrophic part of the microbial community was sampled at low tide using sterile Petri dishes (2–3 replicates per sample within an area of about 1m2). Pore water samples were collected directly at the same positions using Rhizons (Rhizosphere Research Products, Wageningen, NL) that were placed horizontally in the upper 5 mm of the sediment and extracted pore water through a filter membrane with the help of vacuum syringes. Open water samples were collected at incoming tide with 500 ml plastic bottles and were sterile filtered afterwards. All sediment and water samples were taken once at each study site between summer 2013 and summer 2015 (S1 Table).
Open and pore water salinity was measured directly in the field using a hand-held refractometer (arcarda Handelsgesellschaft, Reichelsheim, Germany) and pH of the open water samples as soon as possible in the lab with a pH/mV hand-held meter pH330 (WTW, Weilheim, Germany). Total alkalinity (modified according [66]) and nutrient concentrations were measured spectrophotometrically (partially modified and adapted after: [67] (NH4); [68; 69] (NO2, NOx); [70; 71; 72] (PO4)) using a Fluostar Optima Reader (BMG LABTECH GmbH, Ortenberg, Germany).
For grain size measurements, the sediment samples were dried in a compartment drier at 40°C for at least 8 h, ground with mortar and pestle and sieved through a 2 mm screen. Particles larger than 2mm in diameter were sorted out and disregarded for further measurements. Sieved sediment was pre-treated with 10% (v/v) HCl to remove carbonates and 30% H2O2 to remove organic matter. Grain size was measured with a laser particle sizer Analysette 22 MicroTec plus with Wet Disperser Unit navigated by the MaS Control Software (Fritsch, Idar-Oberstein, Germany) in three replicates per sample and twice per replicate. The mean grain size of 90% of all particles was used to classify four grain size fractions according to the DIN EN ISO 14688–1 norm: CSa = coarse sand (0.63–2.0mm), MSa = medium sand (0.2–0.63mm), FSa = fine sand (0.063–0.2mm), CSi = coarse silt (0.02–0.063mm).
The stabilization level of all samples was qualitatively classified into three classes: high (well-stabilized and laminated microbial mats that can be peeled off the sediment in large pieces (> 25 cm2)), medium (slightly stabilized sediment without clear laminations, that cannot be peeled off the sediment in large, but smaller pieces), and low (lose sediment without notably amounts of stabilizing compounds).
Climate data (air temperature at 10 m above the surface of the earth (°C), precipitation (mm/day) as averaged monthly values over a 22-years period (Jul 1983—Jun 2005)) was obtained from the Atmospheric Science Data Center/NASA Langley Research Center (https://eosweb.larc.nasa.gov/sse/, accessed 09/2017), seawater temperatures from www.seatemperature.org (accessed 08/2017). Annual mean values were calculated based on the monthly averaged values (Table 1).
Linear distances between sampling sites were calculated based on the measured coordinates using google maps (https://www.google.com/maps).
Total DNA of sediment samples (pooled triplicates) was isolated using the UltraClean Microbial DNA Isolation Kit (MO BIO Laboratories, Inc., Carlsbad, CA, USA) for all samples except the ones collected in Oman. It was not possible to reach sufficient DNA concentrations for the Oman samples due to their hypersalinity using the UltraClean Microbial DNA Isolation Kit, despite additional dilution steps. Thus, we had to switch to the Power Biofilm DNA Isolation Kit (MO BIO Laboratories, Inc., Carlsbad, CA, USA), which used alternative buffers, but the same filter columns. In both cases, the manufacturers protocol was extended by 10 freeze and thaw cycles (liquid nitrogen/60°C water bath) and an overnight incubation with Proteinase K during the lysis process. Pre-selective PCR was performed with cyanobacteria-specific CYA106F [73] and universal bacterial 16S1494R [74] primers and the REDTaq Ready Mix (Sigma-Aldrich Chemie GmbH, Munich, Germany)). PCR products were purified using the QIAquick PCR Purification Kit (QIAgen, Hilden, Germany). Details of the entire DNA preparation process were previously described by Vogt et al. [62; 63].
Sequencing of 16S rRNA gene fragments with cyanobacteria specific primers (CYA359F/CYA781R [73]) was conducted in two runs (samples from Oman and all other samples) at the Research and Testing Laboratory (Lubbock, Texas, USA) using the Illumina MiSeq technology.
All sequences were checked for quality by trimming off primer sequences, removing all sequences with mismatches in primer sequences, ambiguous bases, homopolymers longer than 8 bp, and/or average quality scores below 25 in a 50 bp sliding window and screening for a defined fragment length (250–450 bp) using MOTHUR v. 1.36.1 [75] as described by Vogt et al. [62; 63]. The remaining quality checked sequences were aligned against the SILVA reference dataset (Release 102, http://www.arb-silva.de/, [76]), modified for bacterial sequences (silva.bacteria reference alignment, http://www.mothur.org/wiki/Silva_reference_files (accessed 04/2012)) and combined with an additional cyanobacterial rRNA dataset from Herdman (http://cyanophylo.blogspot.de/, downloaded June 2014). Identical sequences were condensed and sequences with up to 4 bp difference to each other were additionally pre-clustered (MOTHUR v.1.36.1 [75]). Chimeric sequences were detected using the VSEARCH command (de novo, dereplicated) based on the UCHIME algorithm [77; 78] and removed (MOTHUR v.1.39.1 [75]). The VSEARCH distance-based greedy clustering method (de novo) was used to create operational taxonomic units (OTUs) at a 97% similarity level (MOTHUR v.1.39.1 [75; 78; 79]). OTUs with less than 11 sequences within the entire dataset were excluded. Representative consensus sequences of each OTU (> 10 sequences) were classified using NCBI Nucleotide Blast and the ‘16S ribosomal RNA sequences (Bacteria and Archaea)’ database (https://blast.ncbi.nlm.nih.gov, [80]). The taxonomic information of the best match (highest identity, max/total score, query cover) was adopted in a resolution depending on the identity values: 85%, 90% and 95% identity for phyla, orders, and genera, respectively. Only cyanobacterial sequences (at least 85% similarity) were analyzed further. Representative sequences of all cyanobacterial OTUs were deposited in the NCBI GenBank database (https://www.ncbi.nlm.nih.gov/Genbank/) with accession numbers MF507041-MF508244.
Relative sequence abundances were calculated based on the total number of sequences and OTUs per sample. OTUs that included less than 0.1% of all quality checked sequences per sample were considered as rare OTUs.
OTU-based alpha and beta diversity estimates were calculated in MOTHUR v.1.39.1 [75] based on all quality checked sequences per sample and described as richness (observed (sobs) and chao1 (chao) estimated numbers of OTUs) and diversity (Shannon diversity index (shannon)), as well as shared richness (observed (sharedsobs) numbers of shared OTUs) and dissimilarity coefficients (Bray-Curtis (braycurtis) and Jaccard (jclass) dissimilarity coefficients). The shared observed richness values per sample (sharedsobs calculated in MOTHUR v.1.39.1 [75]) were summarized per location as presence/absence matrix and a VENN diagram was constructed (manually) based on these summarized values. Rarefaction curve data for observed OTUs per sample were calculated in MOTHUR v.1.39.1 [75] with 1000 randomizations using all available quality checked sequences and a re-sampling without replacement approach.
A maximum likelihood tree with rapid bootstraps (100 replicates, GTRGAMMA) was constructed in RAxML Version 8 [81] (raxmlGUI version 1.5b1 [82]) for the representative sequences of all 989 OTUs. Based on this tree the phylogenetic diversity was analyzed in R version 3.3.1 [83] using the ‘picante’ package (version 1.6–2) [84]. Faith’s phylogenetic diversity (PD [85] = total of the unique branch length in the tree), mean pairwise distance (MPD [86], sequence abundance weighted), and MPD between pairs of communities (phylogenetic beta diversity) were calculated according to the developer’s instructions (http://picante.r-forge.r-project.org/picante-intro.pdf (accessed 04/2010)).
To identify potential linear correlations between different environmental/geographical parameters and diversity values, regression analyses were performed in R version 3.3.1 using the linear model function for linear regression [83].
Redundancy analyses (RDA) based on sequence abundances per OTU and standardized environmental parameters were performed in R version 3.6.0 [87] using the ‘vegan’ package (version 2.5–5) [88]. Sequence numbers per sample and OTU (+ 0.5 to avoid zero values) were centered log-ratio (clr) transformed using the ‘compositions' package (version 1.40–2) [89]. Non - and less-correlated environmental parameters were selected according to their collinearity (pairwise correlation (Pearson) < 0.8 or > - 0.8). Dependent parameters (e.g. NO2 + NO3 = NOX) were removed. Variance inflation factors of the selected parameters were checked to be below 10. Forward model selection using ordiR2step method of the R package ‘vegan’ [88] was performed to identify the best environmental predictors for OTU distribution/occurrence. Total and pure effects of the environmental parameters were calculated using permuRDAv1.6.R (https://github.com/chassenr/ARISA) to quantify the significance of selected environmental parameters explaining the variation in cyanobacterial community composition.
Principal Coordinates Analyses (PCoAs) based on Jaccard and Bray-Curtis dissimilarity coefficients (OTU-based) as well as the phylogenetic beta diversity were calculated in MOTHUR v.1.39.1 [75] to illustrate the dissimilarity of the sampling sites and identify potential distribution patterns based on different diversity estimates considering the presence (Jaccard), abundance (Bray-Curtis), and phylogenetic relationship (phylogenetic beta diversity) of the detected OTUs.
All statistical tests (Shapiro-Wilk normality test, Kruskal-Wallis test, ANOVA) were performed in R version 3.3.1 [83] or version 3.6.0 [87]. P-values < 0.05 were considered as significant.
A total number of 425,730 sequences from 24 samples were clustered into 989 cyanobacterial OTUs (> 10 sequences, 97% similarity). OTU - and phylogeny-based richness (sobs, Chao1, PD) and diversity (Shannon, MPD) estimates per sample, as well as the relative abundance of rare OTUs (< 0.1% of all sequences per sample) varied strongly among all samples (Table 2, S1 Fig). All richness and diversity estimates were calculated based on the complete dataset, including all available quality-checked sequences per sample without normalization that would have removed 90% of the final sequences and 10% of the observed OTUs. Detailed sequence counts during sequence analysis including a putative normalization (S2 Table) and rarefaction curves of all samples (S2 Fig) are shown as Supplementary Information.
A latitudinal diversity gradient with significant linear correlations (negative) of richness (OTU - and phylogeny-based, p < 0.001) and OTU-based diversity values (Shannon index, p < 0.05) was verified here (Fig 2A). However, phylogenetic diversity (MPD) showed no significant correlation to latitude. As expected, the latitudinal gradient of sampling sites was strongly correlated (p < 0.001) to climate-related parameters (i.e. annual mean temperature, precipitation, water temperature, and the length of the vegetation period (> 10°C) (S3 Fig)). Therefore, similar trends in alpha diversity were observed for these climate-related parameters as shown here for annual mean temperatures with increasing richness (sobs, PD, p < 0.001) and OTU-based diversity (Shannon index, p < 0.05) with increasing temperature (Fig 2B, positive correlations) and decreasing precipitation (S5 Table, negative correlation). No significant correlations to temperature were observed for phylogenetic diversity (MPD). Only a minimal trend of increasing alpha diversity (including the phylogenetic diversity (MPD)) with increasing grain size could be detected (Fig 2C).
Apart from different climate conditions, the results demonstrated that tidal flats differ clearly in environmental parameters such as salinity (from 0.2 to 18%), grain size (from 33.2 to 790.7 μm), TA (from 0.4 to 2.9 mM), and sediment stabilization (Table 3). Detailed results of nutrient measurements in pore and seawater samples are available as supplementary data (S3 Table). Significant linear correlations (p < 0.05) between different environmental parameters were detected by regression analyses e.g. for latitude and climatic parameters, as well as for seawater salinity, TA, NH4 NO3, NOX, and pore water salinity (S4 Table). Expected correlations of nutrient availability and grain size [58; 90] could not be verified here (S4 Table). As shown by linear regression, analyses richness values (sobs, PD) were best correlated to latitude, temperature, precipitation, salinity, and alkalinity (p < 0.001), while the correlations of diversity estimates (Shannon, MPD) to all measured abiotic parameters were much weaker or were nearly absent (0.01 < p < 1) (Fig 2, S5 Table).
The influence of environmental parameters on the distribution of OTUs (sequence numbers per OTU and sample, centered log-ratio transformed) was shown via RDA analysis (Fig 3) based on a subset of non - or less-correlated abiotic parameters (standardized values of annual mean temperature, mean grain size, and pore water NH4/NOX/PO4). The forward model selection revealed the annual mean temperature and mean grain size as the most important factors influencing cyanobacterial OTUs distribution. Slight taxon-specific patterns could be observed. Whereas Oscillatoriales OTUs were scattered along both RDA axes, the predominantly unicellular Chroococcales and Pleurocapsales OTUs were enriched in samples with larger grain size, while Synechococcales OTUs were abundant in samples with higher temperature. No clear patterns were distinguishable for OTUs classified as Chroococcidiopsidales, Nostocales, Spirulinales, and unclassified cyanobacteria.
The taxonomic classification on order level revealed a high variability in community composition (based on sequence abundance) within and between locations. However, within some locations (FR, IC, OM) unique grain size patterns were detected (Fig 4A).
Summarized groups per location and sediment type showed differences in community composition and increased abundances of some orders on specific locations or sediment types, however, with wide ranges within each group (Fig 4B). Predominantly unicellular cyanobacteria, such as Chroococcales and Pleurocapsales, were more frequently found on coarse sand sediment, with an exception of the Pleurocapsales-rich Croatian sample that was classified as coarse silt. The order Synechococcales, which includes unicellular and filamentous species, was mainly detected on fine sand to coarse silt in Iceland or on fine to medium sand in Oman. Oscillatoriales dominated samples of all locations and all sediment types, while Spirulinales were mainly found in France and Germany and Nostocales in Iceland, both on fine to medium sand.
A large number of OTUs were location specific, with the maximum number present in the Oman samples (481 OTUs). The percentage of location specific OTUs was highest in Oman (86%) and Croatia (72%) and clearly lower in France (44%), Iceland (34%) and Germany (25%). Only one OTU, classified as Coleofasciculus (97–100% similarity to a Coleofasciculus chthonoplastes sequence (accession number NR_125521.1 in the NCBI database)), was shared by all locations (Fig 5A). Pairwise comparisons of the locations revealed highest numbers of shared OTUs for France—Iceland (107) and France—Germany (88), lowest for Iceland—Oman (20) and Germany—Oman (27), except the single Croatian sample with only 5 to 9 shared OTUs with every other location (Fig 5B). These results were also supported by Jaccard dissimilarity coefficients showing lowest dissimilarity between French—Icelandic (0.75) and French—German (0.78) samples and highest dissimilarity for Croatian samples and all other locations (0.96–0.99), as well as for Oman—Icelandic (0.97) and Oman—German (0.96) samples (Fig 5B). Interestingly, lowest phylogenetic beta diversity was observed for samples from Germany and Oman (0.75) close to that of French and German samples (0.78) (Fig 5B). However, all beta diversity estimates had a rather high dissimilarity/diversity (> 0.75) between the investigated locations.
Although linear regression analyses between beta diversity estimates and the (logarithmized) linear distances between paired samples revealed significant linear correlations (p < 0.001 (OTU-based), p < 0.01 (phylogenetic) as shown in S4 Fig) and indicate a distance-decay relationship, no clear similarity patterns concerning locations and distances could be verified by PCoAs based on different beta diversity estimates (Fig 6A–6C).
Based on Jaccard dissimilarity coefficients (OTU presence/absence) the PCoA clearly separated the hypersaline samples from Oman (O1-5), three samples from France (T4, G, N1-3) and two Icelandic samples (5, 6) together with one French sample (T3), but the remaining samples built a loose cluster (Fig 6A). PCoA revealed one large cluster that contained most of the samples from France, Iceland, and Croatia and again a separate cluster of two Icelandic (5, 6) and one French sample (T3), based on Bray-Curtis dissimilarity coefficients (including sequence abundance per OTU) (Fig 6B). The hypersaline samples from Oman (1–5) and the German samples (sa, cg, si) formed a tail from the main cluster with the most distant ones (DE_si, OM_4, OM_5) clustering with one French sample (N4-5). Based on the phylogenetic beta diversity (MPD between samples) the PCoA (Fig 6C) revealed one main cluster including samples from all locations. Some hypersaline samples from Oman (1–3), two samples from Iceland (1–2), and the Croatian sample were clearly separated. Three French samples (N1-3, T4, G) formed a tail from the main cluster and were most similar to the less saline sample from Oman (OM_X) and one more Icelandic sample (3).
While all three PCoAs did not reveal clear location or grain size patterns, the stabilization level of the community occurred as potential predictor of at least phylogenetic beta diversity. Comparing the mean pairwise distances between samples (phylogenetic beta diversity) according to their level of stabilization revealed the lowest phylogenetic distances between well-stabilized samples (high—high) and the highest ones between less stabilized ones (low—low) (Fig 6D).
To investigate (global) patterns of cyanobacterial diversity, our study focused here on tidal flats as a highly dynamic habitat type that provides similar environmental conditions at distant sites, but also includes variability in environmental parameters on different spatial scales (e.g. climate, grain size, and nutrients). Thus, tidal flats offer the opportunity to investigate diversity patterns along geographical and ecological gradients within a distinct habitat.
Previous molecular/high-resolution studies on cyanobacterial biogeography and diversity were mainly performed either in island-like environments, like hot springs with strong geographical barriers [23] or in open oceans without or with at least very weak geographical barriers [91; 92]. Opposed to these extremes in terms of putative geographical isolation, tidal flats are delimited habitats, but they are widely distributed along all coastlines without strict boundaries towards the flanking terrestrial and marine habitats. Although each sampling site showed a unique composition of environmental parameters (Table 3, S3 Table), they all have the special tidal character with its periodic fluctuations of environmental conditions in common, which make this habitat an extreme one in terms of wide ranges of diverse abiotic parameters.
Benthic communities from tidal flats and their diversity have been already intensively studied at different locations all over the world [60; 61; 62; 63; 64; 65; 93; 94; 95; 96], but it remains challenging to compare or assemble the results of different studies due to methodological differences. Therefore, it is more than difficult to deduce global patterns of diversity from several datasets of different spatially restricted studies (but see [2; 3]). The novelty of this study is its spatial coverage from the subtropics (Oman, 20°N/58°E) to the subarctic (Iceland, 65°N/22°W), including a set of 24 different samples from five locations combined with the use of a high-throughput sequencing method (Illumina MiSeq) to investigate cyanobacterial diversity patterns in one distinct habitat. So far, there are no comparable studies/datasets available describing cyanobacterial diversity patterns in such a large area based on the same methodologies. Previous studies often used either different, often less resolved methods (e.g. clone libraries, fingerprinting methods, 454 sequencing), and/or were spatially much more restricted (e.g. [24; 49; 50; 51; 53; 62; 63; 97; 98]). In general, most of the studies about global diversity of prokaryotes are restricted to marine and mainly planktonic environments [20; 49; 51; 97], which are likely to facilitate wide distribution ranges of small organisms due to the simple dispersal capabilities in open waters with few geographical barriers. These differences in dispersal capabilities are most probably also one reason why Zinger et al. [97] observed a higher similarity with increasing distance between pelagic communities than between benthic ones.
Latitudinal diversity gradients as observed here for cyanobacteria in tidal flats (Fig 2) are one of the most widely recognized patterns in ecology [37; 38; 39; 43] since Humboldt & Bonpland [25], Wallace [26; 27] and Darwin [28] described an increased species richness in tropical regions compared to that in temperate ones. There are diverse hypotheses about reasons for these latitudinal differences in species richness. Historically the tropics are supposed to provide more time and opportunity for diversification than temperate regions due to their greater age, historically larger area, and their more stable climate, while the dispersal of organisms out of the tropics is often limited. Therefore, a plentitude of species have accumulated in tropical regions during Earth’s history making them a ‘cradle’ and a ‘museum’ of biodiversity [39; 44]. Due to higher speciation and lower extinction rates caused by environmental/ecological influences as e.g. climatic conditions (e.g. temperature, climate stability), biotic interactions (e.g. competition, predation), ecological opportunity (e.g. tolerance ranges/niche width, dispersal, metabolic rates, generation times) and area effects (e.g. isolation, population size), the diversification rates and evolutionary speed in the tropics are also supposed to be higher compared to temperate regions [44; 99]. However, recent studies have found evidence for an equalization of diversification rates over latitudes and nowadays highest speciation rates in temperate regions [45].
Latitudinal diversity gradients were intensively studied for a wide variety of taxonomic groups in different environments [37; 38; 44; 47; 100; 101; 102]. Clear relationships between the overall diversity on community level, latitude, and temperature were shown to be a result of interfering, distinct biogeographical distribution patterns of different individual taxa across latitudes according to their unique ecological/physiological properties [38; 42]. Therefore, the gradients become clearer with increasing numbers of taxa and habitat types [38; 42; 47], even though they may vary between different habitat types [47]. Although, focusing on cyanobacterial communities in tidal flat regions in this study restricted both the taxonomic coverage and the habitat type, decreasing richness and diversity values (except MPD) from subtropical (20°N) to subarctic regions (65°N) (Fig 2A) verified the hypothesis of a latitudinal diversity gradient for cyanobacteria in tidal flats. This is similar to what has been observed for many other organisms in different habitats [38; 50; 51]. These results support ecological/environmental reasons for the observed latitudinal gradient of cyanobacterial diversity in tidal flats, although additional historical/evolutionary reasons for the present distribution and diversity could not be excluded.
The observed latitudinal diversity gradients (Fig 2) are a clear evidence for biogeographical distribution patterns of cyanobacterial taxa that are not only determined by the presence or absence of taxa within an area, but also by their composition/proportion within a community, which may strongly influence diversity estimates [34]. Microbial diversity is known to be highly variable on temporal scales and associated ecological parameters as shown e.g. for seasonality [34; 62]. Therefore, observed gradients of diversity and biogeographical distribution pattern are snapshots of the current microbial community and may change with time. However, the sampling of microbial communities near the climax state of annual mat development in summer months at the subarctic to subtropical/dry summer locations (IC, DE, FR, CR) compared to perennial microbial mats sampled in winter at the subtropical-arid location (OM) ensure the investigation of a most comparable, complex and diverse microbial community. Our previous seasonal studies at the German location [62] previously demonstrated that alpha diversity of cyanobacterial mat communities in temperate regions is highest in summer to autumn and that it is mainly the composition of the community, that changed during the course of a year, rather than the presence or absence of season-specific taxa.
Our results obtained from samples collected along geographical and ecological gradients, clearly indicate that not everything is evenly distributed everywhere (Fig 5A). However, our dataset is not sufficient to answer the question if it’s only the environment that selects, or if there are additional drivers of biodiversity, since present-day diversity and distribution patterns are always a result of both historical events and the actual environmental conditions. Hanson et al. [34] proposed that (micro-) evolutionary and ecological influences on microbial distribution and diversity are often inseparable, which might also be true here. They predicted that four processes create and maintain biogeographical patterns (and diversity gradients) of microorganisms: selection, drift, dispersal and mutation. These all may be influenced by evolutionary and ecological factors. Considering the tremendous dispersal potential of microorganisms due to their small cell sizes, large population sizes, and short generation times resulting in high dispersal rates, it is highly likely that ecological properties and environmental selection are much more important than overcoming large distances [32; 34; 48; 103; 104]. However, recent transplantation studies of salt marsh sediment communities at the US East Coast demonstrated that transplanted communities tend to maintain their original community structure, despite changed environmental conditions [105].
In line with this, distance-decay effects were shown by significant correlations between beta diversity estimates and (logarithmized) linear distances between samples (S4 Fig), as also observed for bacterial communities in other studies (e.g. along rivers in France and New Zealand [106]). However, such distance gradients are often strongly correlated to ecological gradients such as climate conditions, which potentially also explains the patterns we received. The expected high ecological similarity between France, Germany, and partially Iceland, based on their close geographic location and similar climate conditions, is supported by them having the highest shared richness (107 OTUs FR/IC, 88 OTUs FR/DE, 35 OTUs FR/DE/IC) and lowest dissimilarity values (Jaccard dissimilarity coefficients 0.75–0.84) between the summarized groups of these locations (Fig 5B). Dispersal of microorganisms between France and Germany is facilitated by a direct connection along the coastline. Iceland is at least directly connected via ocean currents, while the distances towards Croatia and Oman along coastlines and via ocean currents as putative dispersal routes, are much longer than the measured linear distances between the sampling sites. Lowest shared richness and similarity between Croatia and all other locations is almost certainly not explained by climate, but more probable by different coastal properties (sandy rock pools vs. sandy beaches), although the lowest sample number for Croatia (1) may have influenced the results as well. Lower shared richness and higher dissimilarity coefficients between the other samples (Fig 5B) were indicative for different ecological conditions and climate, as well as for the large distance to each other, as was observed for Iceland and Croatia, which had only five shared OTUs and a Jaccard dissimilarity coefficient of 0.98.
Beta diversity estimates of the summarized locational groups are highly likely to be influenced by the number of samples per location, that were not even within our study (Table 2). Additional influences on alpha and beta diversity estimates due to the use of different DNA extraction kits and different sequencing runs for the hypersaline samples from Oman and that of all other locations could also not be excluded, since total sequence numbers and the overall richness and diversity in samples from Oman was shown to be highest (Table 2). Additionally, sequence counts during sequence analyses also suggest highest quality (less low quality loss) of Oman sequences (S2 Table).
Our dataset of 24 samples from 5 locations allowed no general statement about cosmopolitanism or endemism of detected taxa, but instead high amounts of location-/sample-specific OTUs indicated at least (micro-) habitat-specific taxa, since the unique composition of different environmental conditions between the sampled sites clearly influenced community composition.
Despite the common assumption that microorganisms disperse freely around the world without geographical barriers [6; 9; 107], such cosmopolitan distribution patterns could only be verified for one single OTU (classified as Coleofasciculus chthonoplastes) across our sampling locations (Fig 5A). This OTU occurred in each location (but not in each sample) and was predominantly abundant/dominant confirming observations of Nemergut et al. [108] that cosmopolitan taxa tend to be the abundant members in individual assemblages within one habitat type. The occurrence of shared OTUs between samples from distant locations with differing environmental conditions as the 20 shared OTUs between samples from Oman and Iceland (Fig 5) indicated wide dispersal and wide tolerance ranges of the respective taxa, even if they were not detected at all sampled locations. Shared OTUs that were abundant in samples of different locations could be predominantly classified as common oscillatorialean genera like Coleofasciculus and Lyngbya (S7 Table), which are known to have wide ecological tolerance and distribution ranges [10; 109]. Proportions of sequences per sample/location within a shared OTU could for sure strongly vary and there were many OTUs dominant at one location, but rare in the other ones. These differences in community composition are again likely to be a result of the environmental conditions and biotic interactions within the community.
The great majority of detected OTUs here were specific for distinct locations (78% of all OTUs, Fig 5A) or even sampling sites (39% of all OTUs). Environmental parameters and their unique composition were supported as important factors for community composition at each location. OTUs specific to single samples or groups makes them unique and often indicates potentially endemic or habitat-specific taxa. These taxa are commonly supposed to have closer tolerance ranges and are therefore dispersal-limited, but specialized or adapted to special/extreme environmental conditions where they are competitively stronger than other taxa. This matched the highest number of specific OTUs (481) at the extreme location in Oman (Fig 5A), which was characterized by hot and hypersaline conditions.
Although most location-specific OTUs were rare members of the community, there were single samples that reached high proportions of sequences within location specific OTUs. For example, up to 73% and 63% of all sequences per sample in samples from Iceland (Otu22, IC_2) and Oman (Otu3, OM_1), respectively. These likely represent the occurrence of specialized (e.g. halophilic) and warm/cold adapted taxa that are able to outcompete less specialized taxa in such extreme locations, as has been previously shown for the study site in Oman [63]. Interestingly, many specific OTUs were affiliated with genera that are usually known to have wide tolerance ranges (e.g. Coleofasciculus, Lyngbya), which could be either a hind for the development of ecotypes or genetic differences beyond the investigated 16S rRNA gene fragment (insufficient resolution).
Rare members of a community are supposed to be important keystone species and distinguishing marks between communities from different locations and potential indicators for differing environmental influences [3]. They might serve as dormant seedbank and important hidden drivers of ecosystem function improving the metabolic flexibility and physiological tolerance of a community to changing and/or harsh/extreme conditions [1; 110; 111], but their small numbers of individuals make them putatively also more vulnerable to these conditions compared to well-established and abundant taxa [110; 112].
Temperature and grain size of the inhabited sediment were shown here as most important factors for the spatial distribution of OTUs (Fig 3). Environmental parameters that are strongly correlated to latitude are known as important ecological reasons for latitudinal biodiversity patterns on large spatial/global scales [43; 44]. Such correlations were shown here for temperature and other climatic parameters (S3 Fig). Temperature is commonly assumed to influence speciation rates and natural selection by its impact on metabolic rates and generation times [44; 99]. As previously shown for marine planktonic bacteria from the tropics to the poles [49; 51] and diverse eukaryotic and prokaryotic taxa along elevation gradients [42, 113], temperature was confirmed here as important driver of cyanobacterial biodiversity in tidal flats (Fig 2B, S5 Table). However, although an increasing phylogenetic clustering was previously described for bacterial communities along elevation (temperature) gradients towards mountaintops [114], such a significant impact of temperature on MPD values could not be verified here for cyanobacterial communities along the latitudinal gradient (Fig 2).
Strong linear correlations of climatic and (dissolved) chemical parameters as shown here for salinity and alkalinity (p < 0.001) (S3 Table) were supposed to be mainly caused by climate-related evaporation/desiccation or dilution. The local variability of nutrient concentrations, on the contrary, was rather indicative for putative sampling site specific anthropogenic or environmental influences (e.g. eutrophication) and physiological processes within the respective communities [58;59]. Such (independent) environmental parameters have already been reported by others to modify/shift the temperature-biodiversity relationships (e.g. [113]). We could show this effect in our study for grain size and nutrient concentrations, which were variable across latitudes and additionally influenced diversity (Table 3, S3 Table). However, these parameters are likely to be more important on smaller local scales and for community composition, as shown by Vogt et al. [62; 63]. Grain size has previously been reported to be an important factor influencing microbial community development and composition [58; 62; 115] and our results confirmed this parameter as an important driver of cyanobacterial diversity (Fig 3). The observed variability of diversity estimates within locations (Fig 2A) indicated the importance of variable parameters across latitudes for present communities on smaller spatial/local scale. However, there were no consistent (linear) diversity gradients within locations based on the measured parameters in this study as shown e.g. for grain size patterns with inverse trends at different locations (Fig 2V, S5 Fig).
Temporal, especially seasonal dynamics of abiotic and (related) biotic parameters additionally influence the current biodiversity (e.g. [62]). Thus, composition and (functional) complexity of microbial communities are known to change during succession influenced by environmental parameters [58; 61]. The developmental stage and complexity of each community could be roughly estimated according to the degree of stabilization and lamination, which usually increases successively with the number of individuals and members with different physiological properties/niches comprising in the end all major element cycles [61]. However, discontinuities in mat growth and development or even the destruction of benthic microbial mats may be caused by extreme environmental conditions as shown for our study site in Germany during winter season [62]. Furthermore, varying physical factors such as wind and wave activity according to local characteristics of the coastline, and factors like (hyper-) salinity and grain size of the colonized sediment may either facilitate, decelerate, or even change succession and consequently composition of microbial communities [60, 115]. Due to the small number of samples per location in this study and their different ranges of ecological variability within each location, the observed correlations of diversity and abiotic parameters can only be understood as putative trends. In general, a high variability of environmental conditions and sampling sites per location was desired to cover a broad range of intertidal (micro) habitats, but unfortunately it was not possible to sample broad ranges of all measured parameters in each location.
The ability of organisms to cope with or adapt to extreme environmental conditions (e.g. high or low temperatures, hypersalinity), plays an important role for their distribution across latitudes [11]. Due to the potentially limited dispersal of organisms according to their physiological tolerances/plasticity towards environmental conditions, diversity is likely to be restricted at more extreme sites [50; 93]. However, since cyanobacteria are known as important primary producers and pioneer organisms that tolerate a wide range of extreme/harsh environmental conditions, they are expected to show high diversity in extreme environments and early stage mat communities, as has been observed in previous studies [61; 116; 117; 118; 119]. This same trend was shown here for hypersaline samples from Oman and early stage communities from France, Germany and Iceland (Table 2). These results are in line with previous observations by Rothrock and Garcia-Pichel [93], which observed that cyanobacterial diversity along an intertidal desiccation gradient was less impacted than bacterial and archaeal ones, or the study of Pommier et al. [49] detecting lowest correlations of cyanobacterial OTU richness to latitude and temperature compared to all other bacterial taxa, and finally the studies of Abed et al. [64] and Vogt et al. [63], which observed that cyanobacterial diversity even increased along a hypersaline intertidal gradient with increasing distance to lower tidal line. The frequently high tolerance ranges of cyanobacteria for most diverse environmental conditions and their high diversity in extreme environments [120] may be one reason for all these observations.
Taxonomic classification based on 16S rRNA gene fragments as shown here (Fig 4) can only approximate the actual taxonomy, since perfect matches with reference datasets were usually rare. In this study we had only 3 OTUs (<0.5%) with 100% and 88 OTUs (9%) with at least 97% similarity matches within the ‘16S ribosomal RNA sequences (Bacteria and Archaea)’ database. Although, clustering sequences at a 97% similarity level and classifying OTUs on order level in this study (Fig 4A) was appropriate to get an overview of community composition of the samples, one should be aware that gradients of diversity and biogeographical distribution patterns were also shown to be influenced by taxonomic resolution [34].
Contrary trends of OTU-based (Shannon) and phylogeny-based (MPD) diversity estimates across latitudes (Fig 2A) highlighted the importance of phylogenetic/taxonomic information comparing communities and interpreting results. Including phylogenetic distances for diversity estimates allows deeper insights into community structure and function based on the assumption that closely related species/lineages share similar niches (niche conservatism) and their metabolic abilities are conserved [121]. The observed trends of decreasing phylogenetic diversity with increasing mat stabilization (S6 Fig) suggested a shift and restriction of ecological niches of cyanobacteria during succession. The more niches are inherited successively by non-cyanobacterial members of the microbial community, the fewer niches remain for cyanobacterial ones. Moreover, stabilizing compounds within microbial mats may serve as a protective shield against environmental influences cushioning their fluctuations [58; 59; 116]. Thus, the internal physicochemical gradients within the nearly closed small-scaled ecosystem might become more important and are supposed to be more similar across latitudes than the outer environmental conditions. The stronger influences of changing environmental conditions on less stabilized communities could be one reason why they were shown to be phylogenetically much more diverse among the studied locations than well-stabilized communities, which become increasingly similar across latitudes (Fig 6D).
Although annual mean temperature (latitude) and mean grain size were identified as the most important (quantitative) factors influencing sequence abundance (Fig 3), no general latitudinal/temperature related or consistent grain size pattern of taxonomic community composition could be observed on a global scale (Fig 4). However, distinct grain size patterns on smaller spatial scales within each location (Fig 4A) were observed, indicating changing (competitive) success of different taxonomic groups on specific grain sizes dependent on other environmental conditions (e.g. climate). This suggested an overall stronger impact of climate and latitude than granulation of the sediment.
OTUs classified as Oscillatoriales occurred in all locations as dominant member of the community (except Croatia) (Fig 4, S7 Table) confirming earlier studies of Stal et al. [58] and Garcia-Pichel et al. [10] and indicating a high physiological plasticity and wide distribution range of this order. Their distribution along both RDA axes (Fig 3) indicates that some taxa of the order are rather influenced by changing temperatures whereas others are more influenced by different grain sizes. Oscillatoriales (e.g. Coleofasciculus, Lyngbya) are known as widespread marine/benthic cyanobacteria and main components of well-stabilized microbial mats [109].
The increased abundance of sequences classified as Spirulinales (Fig 4, S7 Table) matched their description as frequent member of marine coastal communities [109].
The distribution of OTUs classified as Synechococcales was mainly correlated to temperature, as indicated by RDA (Fig 3). This order includes unicellular and filamentous taxa commonly detected in marine and terrestrial environments all over the world [64; 94; 119; 122; 123]. In line with other studies in extreme environments (e.g. [119; 123]) they were here mainly detected in samples from Iceland and Oman in all sediment types with the highest abundances in less developed mats (Fig 4, S7 Table).
Unicellular taxa (Chroococcales, Chroococcidiopsidales, Pleurocapsales, and Synechococcales) are often described in marine and freshwater environments, and are also known to dominate harsh and unfavorable environments from deserts to rocky shores and tidal flats across latitudes [64; 117; 123; 124; 125; 126; 127; 128; 129; 130]. RDA indicated a strong influence of grain size for the distribution of OTUs classified as Chroococcales or Pleurocapsales (Fig 3). As coarse-grained sediment was suggested to be unfavorable for the development of well-stabilized mats with high amounts of filamentous taxa [58], it is not surprising that the unicellular orders reached high sequence abundances in early stage communities on coarse-grained sediment (Fig 4, S7 Table). High production rates for extracellular polymeric substances (EPS) were described for unicellular cyanobacteria [131], privileging them as important autotrophic pioneer organisms at low nutrient sites. Putative input and sedimentation of planktic unicellular taxa from seawater to the benthic communities could be one more reason for their increased abundance in early stage communities.
Although all of the Croatian sequences classified as Pleurocapsales stayed unclassified on genus level (96%), their high sequence abundance (Fig 4, S7 Table) was in line with other studies at this limestone coast [129; 130]. The nearly complete absence of sequences classified as filamentous Oscillatoriales and Nostocales in the well-stabilized Croatian sample is surprising and may be explained by the extraordinary local settings (limestone coast with sandy rock pools) that may favor unicellular species [129; 130]. Also biases during sample preparation could not be excluded, since microscopic investigations (not shown here) detected high amounts of filamentous species (with and without heterocysts) wrapped within thick sheaths, which may have withstood the lysis process during the DNA isolation.
Only two mats from Iceland (IC_1, IC_2) were dominated by OTUs classified as Nostocales (Fig 4, S7 Table), which are known to be able to fix nitrogen and to produce akinetes which help to overcome unfavorable conditions such as frost events [132]. Thus, these taxa are expected to occur at unfavorable and low nutrient sites and also survive at high latitudes with low temperatures and low light intensities in the winter season [132].
In summary our results show that microorganisms’ area of distribution and their distance-decay similarities tend to be significant, but weaker than that for macroorganisms. Our study could confirm finding of Soininen [133] that latitudinal patterns of cyanobacterial distribution are affected by species traits and ecosystem characteristics.
This study observed cyanobacterial diversity pattern in microbial tidal flat communities from the subtropics to the subarctic along geographical and ecological gradients.
Strong distinctions in alpha diversity and community composition, as well as high dissimilarity values between samples due to high numbers of location-specific taxa (beta diversity), gives the first evidence that a true cosmopolitan distribution of cyanobacterial communities does not exist. However, the increased phylogenetic similarity of well-stabilized samples across latitudes supported at least wide distribution patterns of a few closely related cyanobacterial taxa that are predominantly found in well-stabilized microbial mat systems.
Latitudinal diversity gradients of cyanobacterial communities in tidal flats similar to those known from plants and animals were identified here for OTU - and phylogeny-based alpha diversity estimates (especially richness values). Latitude and related environmental parameters (e.g. climate, temperature) were supported as important drivers of diversity on global scale, while environmental parameters that are not correlated to latitude/climate (e.g. grain size, nutrients) were shown to be more important on a local scale. However, these local patterns were not consistent across latitudes or different locations. Environmental parameters such as temperature/latitude and grain size were also shown to be the most important factors influencing the distribution of detected cyanobacterial taxa.
So far, our dataset of 24 samples from 5 locations provides the best spatial coverage within a single study using high-throughput sequencing methods. However, further studies including more samples and locations could help to generalize the results. Higher coverage of similar ecological conditions at distant sites and different ecological conditions at nearby sites could give more convincing inferences about biogeography of specific cyanobacteria taxa.
S1 Table [pdf] Overview of sampling sites with sampling date, location and geographic position.
S2 Table [pdf] Sequence counts (nseqs) during sequence analysis, OTU counts of the final dataset, and normalized values based on the final dataset.
S3 Table [ta] Abiotic parameters of the sampling sites.
S4 Table [sal] Significance values (p-values) of linear regression analyses between different abiotic parameters.
S5 Table [sobs] Significance values (p-values) of linear regression analyses between abiotic parameters and alpha diversity.
S6 Table [clr] Effect summary of selected environmental parameters used for RDA.
S7 Table [ic] Taxonomic classification and relative sequence abundances of each detected taxon per sample on order and genus level.
S1 Fig [pd] OTU- and phylogeny-based alpha diversity.
S2 Fig [ic] Rarefaction curves for observed OTUs of all samples per location.
S3 Fig [ic] Linear regression analyses of climatic parameters and mean grain size in relation to latitude.
S4 Fig [mpd] Linear regression analyses of beta diversity estimates and distances between sample pairs.
S5 Fig [ic] Linear correlations between Shannon diversity estimates and mean grain sizes within locations.
S6 Fig [sobs] Relationship between stabilization level and mean grain size as well as diversity estimates.
1. Sogin ML, Morrison HG, Huber JA, Welch DM, Huse SM, Neal PR, et al. Microbial diversity in the deep sea and the underexplored “rare biosphere”. Proc Natl Acad Sci U S A. 2006;103(32):12115–20. doi: 10.1073/pnas.0605127103 16880384
2. Lozupone CA, Knight R. Global patterns in bacterial diversity. Proc Natl Acad Sci U S A. 2007;104(27):11436–40. doi: 10.1073/pnas.0611525104 17592124
3. Ramirez KS, Knight CG, de Hollander M, Brearley FQ, Constantinides B, Cotton A, et al. Detecting macroecological patterns in bacterial communities across independent studies of global soils. Nat Microbiol. 2017;3(2):189–96. doi: 10.1038/s41564-017-0062-x 29158606
4. Beijerinck MW. De infusies en de ontdekking der backteriën. In: Jaarboek van de Koninklijke Akademie voor Wetenschappen. Amsterdam, The Netherlands: Müller; 1913. p. 119–140.
5. Baas-Becking LGM. Geobiologie of inleiding tot de milieukunde. W.P. Van Stockum & Zoon; 1934. (Diligentia-voordrachten).
6. Finlay BJ, Clarke KJ. Ubiquitous dispersal of microbial species. Nature. 1999;400 : 828.
7. Patterson DJ, Lee WJ. Geographic distribution and diversity of free living heterotrophic flagellates. In: Leadbetter BSC, Green JC, editors. The Flagellates—Unity, Diversity and Evolution. London: Taylor & Francis; 2000. p. 269–287.
8. Darling KF, Wade CM, Stewart IA, Kroon D, Dingle R, Brown AJL. Molecular evidence for genetic mixing of Arctic and Antarctic subpolar populations of planktonic foraminifers. Nature. 2000;405(6782):43–7. doi: 10.1038/35011002 10811211
9. Finlay BJ. Global dispersal of free-living microbial eukaryote species. Science. 2002;296 : 1061–3. doi: 10.1126/science.1070710 12004115
10. Garcia-Pichel F, Prufert-Bebout L, Muyzer G. Phenotypic and phylogenetic analyses show Microcoleus chthonoplastes to be a cosmopolitan cyanobacterium. Appl Environ Microbiol. 1996;62(9):3284–91. 8795218
11. Hoffmann L. Geographic distribution of freshwater blue-green algae. Hydrobiologia. 1996;336 : 33–40.
12. Miller SR, Castenholz RW, Pedersen D. Phylogeography of the thermophilic cyanobacterium Mastigocladus laminosus. Appl Environ Microbiol. 2007;73(15):4751–9. doi: 10.1128/AEM.02945-06 17557856
13. Padisák J. Cylindrospermopsis raciborskii (Woloszynska) Seenayya et Subbu Raju, an expanding, highly adaptive cyanobacterium: worldwide distribution and review of its ecology. Arch fur Hydrobiol Suppl Monogr Stud. 1997;107(4):563–93.
14. Gugger M, Molica R, Le Berre B, Dufour P, Bernard C, Humbert J-F. Genetic diversity of Cylindrospermopsis strains (Cyanobacteria) isolated from four continents. Appl Environ Microbiol. 2005;71(2):1097–100. doi: 10.1128/AEM.71.2.1097-1100.2005 15691973
15. Antunes JT, Leão PN, Vasconcelos VM. Cylindrospermopsis raciborskii: Review of the distribution, phylogeography, and ecophysiology of a global invasive species. Front Microbiol. 2015;6(473):1–13.
16. van Gremberghe I, Leliaert F, Mergeay J, Vanormelingen P, van der Gucht K, Debeer AE, et al. Lack of phylogeographic structure in the freshwater cyanobacterium Microcystis aeruginosa suggests global dispersal. PLoS One. 2011;6(5):e19561. Available from: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.001956117. doi: 10.1371/journal.pone.0019561 21573169
17. Moreira C, Spillane C, Fathalli A, Vasconcelos V, Antunes A. African origin and europe-mediated global dispersal of the cyanobacterium Microcystis aeruginosa. Curr Microbiol. 2014;69(5):628–33. doi: 10.1007/s00284-014-0628-2 24952206
18. Bik HM, Sung W, De Ley P, Baldwin JG, Sharma J, Rocha-Olivares A, et al. Metagenetic community analysis of microbial eukaryotes illuminates biogeographic patterns in deep-sea and shallow water sediments. Mol Ecol. 2012;21(5):1048–59. doi: 10.1111/j.1365-294X.2011.05297.x 21985648
19. Bates ST, Clemente JC, Flores GE, Walters WA, Parfrey LW, Knight R, et al. Global biogeography of highly diverse protistan communities in soil. ISME J. 2013;7(3):652–9. doi: 10.1038/ismej.2012.147 23235291
20. Salazar G, Cornejo-Castillo FM, Benítez-Barrios V, Fraile-Nuez E, Álvarez-Salgado XA, Duarte CM, et al. Global diversity and biogeography of deep-sea pelagic prokaryotes. ISME J. 2016;10(3):596–608. doi: 10.1038/ismej.2015.137 26251871
21. Boenigk J, Wodniok S, Bock C, Beisser D, Hempel C, Grossmann L, et al. Geographic distance and mountain ranges structure freshwater protist communities on a European scalе. Metabarcoding and Metagenomics. 2018;2:e21519. Available from: https://mbmg.pensoft.net/articles.php?id=21519
22. Ribeiro KF, Duarte L, Oliveira L. Everything is not everywhere: a tale on the biogeography of cyanobacteria. Hydrobiologia. 2018;820(1):23–48.
23. Papke RT, Ramsing NB, Bateson MM, Ward DM. Geographical isolation in hot spring cyanobacteria. Environ Microbiol. 2003;5(8):650–9. 12871232
24. Taton A, Grubisic S, Balthasart P, Hodgson DA, Laybourn-Parry J, Wilmotte A. Biogeographical distribution and ecological ranges of benthic cyanobacteria in East Antarctic lakes. FEMS Microbiol Ecol. 2006;57(2):272–89. doi: 10.1111/j.1574-6941.2006.00110.x 16867145
25. Humboldt A von, Bonpland A. Ideen zu einer Geographie der Pflanzen: nebst einem Naturgemälde der Tropenländer. Tübingen, Paris: F. G. Cotta, F.Schoell; 1807. Available from: https://www.biodiversitylibrary.org/item/37871
26. Wallace AR. Palm Trees of the Amazon and Their Uses. London: John Van Voorst; 1853.
27. Wallace AR. The geographical distribution of animals. New York: Harper & Brothers, Publishers, Franklin Sqare; 1876.
28. Darwin C. The Origin of Species: And, The Voyage of the Beagle. Alfred A. Knopf; 1859. Available from: https://books.google.de/books?id=arm7fmLePg8C
29. Schall JJ, Pianka ER. Geographical Trends in Numbers of Species. Science. 1978;201(4357):679–86. doi: 10.1126/science.201.4357.679 17750221
30. Currie DJ, Paquin V. Large-scale biogeographical patterns of species richness of trees. Nature. 1987;329 : 326–7.
31. Mutke J, Barthlott W. Patterns of vascular plant diversity at continental to global scales. Biol Skr. 2005;55 : 521–37.
32. Martiny JBH, Bohannan BJM, Brown JH, Colwell RK, Fuhrman JA, Green JL, et al. Microbial biogeography: putting microorganisms on the map. Nat Rev Microbiol. 2006;4(2):102–12. doi: 10.1038/nrmicro1341 16415926
33. Ramette A, Tiedje JM. Biogeography: An emerging cornerstone for understanding prokaryotic diversity, ecology, and evolution. Microb Ecol. 2007;53(2):197–207. doi: 10.1007/s00248-005-5010-2 17106806
34. Hanson CA, Fuhrman JA, Horner-Devine MC, Martiny JBH. Beyond biogeographic patterns: Processes shaping the microbial landscape. Nat Rev Microbiol. 2012;10(7):497–506. doi: 10.1038/nrmicro2795 22580365
35. Thompson LR, Sanders JG, McDonald D, Amir A, Ladau J, Locey KJ, et al. A communal catalogue reveals Earth’s multiscale microbial diversity. Nature. 2017;551(7681):457–63. doi: 10.1038/nature24621 29088705
36. Gaston KJ. Global patterns in biodiversity. Nature. 2000;405(6783):220–7. doi: 10.1038/35012228 10821282
37. Willig MR, Kaufman DM, Stevens RD. Latitudinal Gradients of Biodiversity: Pattern, Process, Scale, and Synthesis. Annu Rev Ecol Evol Syst. 2003;34(1):273–309.
38. Hillebrand H. On the generality of the latitudinal diversity gradient. Evaluation. 2004;163(2):192–211.
39. Jablonski D, Roy K, Valentine JW. Out of the Tropics: Evolutionary dynamics of the latitudinal diversity gradient. Science. 2006;314 : 102–6. doi: 10.1126/science.1130880 17023653
40. Kreft H, Jetz W. Global patterns and determinants of vascular plant diversity. Proc Natl Acad Sci U S A. 2007;104(14):5925–30. doi: 10.1073/pnas.0608361104 17379667
41. Buckley LB, Davies TJ, Ackerly DD, Kraft NJB, Harrison SP, Anacker BL, et al. Phylogeny, niche conservatism and the latitudinal diversity gradient in mammals. Proc R Soc B Biol Sci. 2010;277(1691):2131–8.
42. Peters MK, Hemp A, Appelhans T, Behler C, Classen A, Detsch F, et al. Predictors of elevational biodiversity gradients change from single taxa to the multi-taxa community level. Nat Commun. 2016;7(13736). Available from: https://www.nature.com/articles/ncomms13736
43. Willig MR, Presley SJ. Latitudinal Gradients of Biodiversity. In: Reference Module in Life Sciences. Elsevier; 2017. Available from: http://www.sciencedirect.com/science/article/pii/B9780128096338021749
44. Mittelbach GG, Schemske DW, Cornell H V., Allen AP, Brown JM, Bush MB, et al. Evolution and the latitudinal diversity gradient: Speciation, extinction and biogeography. Ecol Lett. 2007;10(4):315–31. doi: 10.1111/j.1461-0248.2007.01020.x 17355570
45. Schluter D. Speciation, Ecological Opportunity, and Latitude. Am Nat. 2016;187(1):1–18. doi: 10.1086/684193 26814593
46. Cutter AD, Gray JC. Ephemeral ecological speciation and the latitudinal biodiversity gradient. Evolution (NY). 2016;70(10):2171–85.
47. Kinlock NL, Prowant L, Herstoff EM, Foley CM, Akin-Fajiye M, Bender N, et al. Explaining global variation in the latitudinal diversity gradient: Meta-analysis confirms known patterns and uncovers new ones. Glob Ecol Biogeogr. 2018;27(1):125–41.
48. Astorga A, Oksanen J, Luoto M, Soininen J, Virtanen R, Muotka T. Distance decay of similarity in freshwater communities: Do macro - and microorganisms follow the same rules? Glob Ecol Biogeogr. 2012;21(3):365–75.
49. Pommier T, Canbäck B, Riemann L, Boström KH, Simu K, Lundberg P, et al. Global patterns of diversity and community structure in marine bacterioplankton. Mol Ecol. 2007;16(4):867–80. doi: 10.1111/j.1365-294X.2006.03189.x 17284217
50. Yergeau E, Newsham KK, Pearce DA, Kowalchuk GA. Patterns of bacterial diversity across a range of Antarctic terrestrial habitats. Environ Microbiol. 2007;9(11):2670–82. doi: 10.1111/j.1462-2920.2007.01379.x 17922752
51. Fuhrman JA, Steele JA, Hewson I, Schwalbach MS, Brown M V., Green JL, et al. A latitudinal diversity gradient in planktonic marine bacteria. Proc Natl Acad Sci U S A. 2008;105(22):7774–8. doi: 10.1073/pnas.0803070105 18509059
52. Andam C, Doroghazi J, Campbell A, Kelly P, Choudoir M, Buckley D. A Latitudinal Diversity Gradient in Terrestrial Bacteria of the Genus Streptomyces. mBio. 2016;7(2):e02200–15. Available from: https://mbio.asm.org/content/7/2/e02200-15 doi: 10.1128/mBio.02200-15 27073097
53. Crump BC, Hopkinson CS, Sogin ML, Hobbie JE. Microbial Biogeography along an Estuarine Salinity Gradient: Combined Influences of Bacterial Growth and Residence Time. Appl Environ Microbiol. 2004;70(3):1494–505. doi: 10.1128/AEM.70.3.1494-1505.2004 15006771
54. Namsaraev Z, Mano MJ, Fernandez R, Wilmotte A. Biogeography of terrestrial cyanobacteria from Antarctic ice-free areas. Ann Glaciol. 2010;51(56):171–7.
55. Dyer KR, Christie MC, Wright EW. The classification of intertidal mudflats. Coast Shelf Res. 2000;20 : 1039–60.
56. Stutz ML, Pilkey OH. Global distribution and morphology of deltaic barrier island systems. J Coast Res. 2002;707(36):694–707.
57. Daidu F, Yuan W, Min L. Classifications, sedimentary features and facies associations of tidal flats. J Palaeogeogr. 2013;2(1):66–80.
58. Stal LJ, van Gemerden H, Krumbein WE. Structure and development of a benthic marine microbial mat. FEMS Microbiol Ecol. 1985;31(2):111–25.
59. van Gemerden H. Microbial mats: A joint venture. Mar Geol. 1993;113(1–2):3–25.
60. Stal LJ. Cyanobacterial mats and stromatolites. In: Whitton BA, editor. Ecology of cyanobacteria II: Their diversity in space and time. Dordrecht: Springer Netherlands; 2012. p. 65–125.
61. Bolhuis H, Cretoiu MS, Stal LJ. Molecular ecology of microbial mats. FEMS Microbiol Ecol. 2014;90(2):335–50. doi: 10.1111/1574-6941.12408 25109247
62. Vogt JC, Albach DC, Palinska KA. Cyanobacteria of the Wadden Sea: seasonality and sediment influence on community composition. Hydrobiologia. 2018;811(1):103–17.
63. Vogt JC, Abed RMM, Albach DC, Palinska KA. Bacterial and archaeal diversity in hypersaline cyanobacterial mats along a transect in the intertidal flats of the Sultanate of Oman. Microb Ecol. 2018;75 : 331–47. doi: 10.1007/s00248-017-1040-9 28736793
64. Abed RMM, Kohls K, Schoon R, Scherf AK, Schacht M, Palinska KA, et al. Lipid biomarkers, pigments and cyanobacterial diversity of microbial mats across intertidal flats of the arid coast of the Arabian Gulf (Abu Dhabi, UAE). FEMS Microbiol Ecol. 2008;65(3):449–62. doi: 10.1111/j.1574-6941.2008.00537.x 18637042
65. Allen MA, Goh F, Burns BP, Neilan BA. Bacterial, archaeal and eukaryotic diversity of smooth and pustular microbial mat communities in the hypersaline lagoon of Shark Bay. Geobiology. 2009;7(1):82–96. doi: 10.1111/j.1472-4669.2008.00187.x 19200148
66. Sarazin G, Michard G, Prevot F. A rapid and accurate spectroscopic method for alkalinity measurements in sea water samples. Water Res. 1999;33(1):290–4.
67. Benesch R, Mangelsdorf P. Eine Methode zur colorimetrischen Bestimmung von Ammoniak in Meerwasser. Helgoländer wissenschaftliche Meeresuntersuchungen. 1972;23 : 365–75.
68. Miranda KM, Espey MG, Wink DA. A rapid, simple spectrophotometric method for simultaneous detection of nitrate and nitrite. Nitric Oxide Biol Chem. 2001;5(1):62–71.
69. Schnetger B, Lehners C. Determination of nitrate plus nitrite in small volume marine water samples using vanadium(III)chloride as a reduction agent. Mar Chem. 2014;160 : 91–8.
70. O’Dell JW. Method 365.1, Revision 2.0: Determination of phosphorus by semi-automated colorimetry. EPA—United States Environ Prot Agency. 1993;1–15.
71. Itaya K, Ul M. A new micromethod for the colorimetric determination of inorganic phosphate. Clin Chim Acte. 1966;14(3):361–6.
72. Altmann HJ, Fürstenau E, Gielewski A, Scholz L. Photometrische Bestimmung kleiner Phosphatmengen mit Malachitgrün. Fresenius’ Zeitschrift für Anal Chemie. 1971;256 : 274–6.
73. Nübel U, Garcia-Pichel F, Muyzer G. PCR primers to amplify 16S rRNA genes from cyanobacteria. Appl Environ Microbiol. 1997;63(8):3327–32. 9251225
74. Wilmotte A, Demonceau C, Goffart A, Hecq J-H, Demoulin V, Crossley AC. Molecular and pigment studies of the picophytoplankton in a region of the Southern Ocean (42–54°S, 141–144°E) in March 1998. Deep Sea Res Part II Top Stud Oceanogr. 2002;49(16):3351–63.
75. Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, et al. Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol. 2009;75(23):7537–41. doi: 10.1128/AEM.01541-09 19801464
76. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2013;41(Database issue):D590–6. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3531112/ doi: 10.1093/nar/gks1219 23193283
77. Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics. 2011;27(16):2194–200. doi: 10.1093/bioinformatics/btr381 21700674
78. Rognes T, Flouri T, Nichols B, Quince C, Mahé F. VSEARCH: a versatile open source tool for metagenomics. PeerJ. 2016;4:e2584. Available from: https://peerj.com/articles/2584/ doi: 10.7717/peerj.2584 27781170
79. Westcott SL, Schloss PD. De novo clustering methods outperform reference-based methods for assigning 16S rRNA gene sequences to operational taxonomic units. PeerJ. 2015;3:e1487. Available from: https://peerj.com/articles/1487/ doi: 10.7717/peerj.1487 26664811
80. Zhang Z, Schwartz S, Wagner L, Miller W. A Greedy Algorithm for Aligning DNA Sequences. J Comput Biol. 2000;7(1/2):203–14.
81. Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics. 2014;30(9):1312–3. doi: 10.1093/bioinformatics/btu033 24451623
82. Silvestro D, Michalak I. raxmlGUI: a graphical front-end for RAxML. Org Divers Evol. 2012;12 : 335–7.
83. R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2016. Available from: https://www.r-project.org/
84. Kembel SW, Cowan PD, Helmus MR, Cornwell WK, Morlon H, Ackerly DD, et al. Picante: R tools for integrating phylogenies and ecology. Bioinformatics. 2010;26(11):1463–4. doi: 10.1093/bioinformatics/btq166 20395285
85. Faith DP. Conservation evaluation and phylogenetic diversity. Biol Conserv. 1992;61 : 1–10.
86. Webb CO, Ackerly DD, McPeek MA, Donoghue MJ. Phylogenies and Community Ecology. Annu Rev Ecol Syst. 2002;33(1):475–505.
87. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2019. Available from https://www.r-project.org/
88. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. vegan: Community Ecology Package (R package version 2.5–5); 2019. Available from https://cran.r-project.org/package=vegan
89. van den Boogaart KG, Tolosana-Delgado R. “compositions”: A unified R package to analyze compositional data. Comput. Geosci. 2008;34(4):320–338.
90. Billerbeck M, Røy H, Bosselmann K, Huettel M. Benthic photosynthesis in submerged Wadden Sea intertidal flats. Estuar Coast Shelf Sci. 2007;71(3–4):704–16.
91. Martiny AC, Tai APK, Veneziano D, Primeau F, Chisholm SW. Taxonomic resolution, ecotypes and the biogeography of Prochlorococcus. Environ Microbiol. 2009;11(4):823–32. doi: 10.1111/j.1462-2920.2008.01803.x 19021692
92. Mazard S, Ostrowski M, Partensky F, Scanlan DJ. Multi-locus sequence analysis, taxonomic resolution and biogeography of marine Synechococcus. Environ Microbiol. 2012;14(2):372–86. doi: 10.1111/j.1462-2920.2011.02514.x 21651684
93. Rothrock MJ Jr, Garcia-Pichel F. Microbial diversity of benthic mats along a tidal desiccation gradient. Environ Microbiol. 2005;7(4):593–601. doi: 10.1111/j.1462-2920.2005.00728.x 15816936
94. Bolhuis H, Stal LJ. Analysis of bacterial and archaeal diversity in coastal microbial mats using massive parallel 16S rRNA gene tag sequencing. ISME J. 2011;5(11):1701–12. doi: 10.1038/ismej.2011.52 21544102
95. Bolhuis H, Fillinger L, Stal LJ. Coastal microbial mat diversity along a natural salinity gradient. PLoS One. 2013;8(5):e63166. Available from: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0063166%0A%0A 23704895
96. Scholz B, Einarsson H. Microphytobenthic community composition of two sub-Arctic intertidal flats in Huna Bay (Northern Iceland). Eur J Phycol. 2015;50(2):182–206.
97. Zinger L, Amaral-Zettler LA, Fuhrman JA, Horner-Devine MC, Huse SM, Welch DBM, et al. Global patterns of bacterial beta-diversity in seafloor and seawater ecosystems. PLoS One. 2011;6(9):e24570. Available from: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0024570 21931760
98. Lacap-Bugler DC, Lee KK, Archer S, Gillman LN, Lau MCY, Leuzinger S, et al. Global diversity of desert hypolithic cyanobacteria. Front Microbiol. 2017;8(867):1–13.
99. Fine PVA. Ecological and evolutionary drivers of geographic variation in species diversity. Annu Rev Ecol Evol Syst. 2015;46(1):369–92.
100. Fischer AG. Latitudinal variations in organic diversity. Evolution (NY). 1960;14 : 64–81.
101. Pianka ER. Latitudinal gradients in species diversity: A review of concepts. Am Nat. 1966;100(910):33–46.
102. Currie DJ. Energy and large-scale patterns of animal - and plant-species richness. Am Nat. 1991;137(1):27–49.
103. Green J, Bohannan BJM. Spatial scaling of microbial biodiversity. Trends Ecol Evol. 2006;21(9):501–7. doi: 10.1016/j.tree.2006.06.012 16815589
104. Dvořák P, Poulíčková A, Hašler P, Belli M, Casamatta DA, Papini A. Species concepts and speciation factors in cyanobacteria, with connection to the problems of diversity and classification. Biodivers Conserv. 2015;24(4):739–57.
105. Angermeyer A, Crosby SC, Huber JA. Salt marsh sediment bacterial communities maintain original population structure after transplantation across a latitudinal gradient. PeerJ. 2018;6:e4735. Available from: https://peerj.com/articles/4735/ doi: 10.7717/peerj.4735 29736349
106. Echenique-Subiabre I, Zancarini A, Heath MW, Wood SA, Quiblier C, Humbert JF. Multiple processes acting from local to large geographical scales shape bacterial communities associated with Phormidium (cyanobacteria) biofilms in French and New Zealand rivers. Sci. Rep. 2018;8(1):1–12. doi: 10.1038/s41598-017-17765-5
107. Fenchel T, Finlay BJ. The ubiquity of small species: Patterns of local and global diversity. Bioscience. 2004;54(8):777.
108. Nemergut DR, Costello EK, Hamady M, Lozupone C, Jiang L, Schmidt SK, et al. Global patterns in the biogeography of bacterial taxa. Environ Microbiol. 2011;13(1):135–44. doi: 10.1111/j.1462-2920.2010.02315.x 21199253
109. Komárek J, Anagnostidis K. Süßwasserflora von Mitteleuropa, Bd. 19/2: Cyanoprokaryota. 2. Teil / 2nd part: Oscillatoriales. Vol. 49. Springer Spektrum; 2007.
110. Lennon JT, Jones SE. Microbial seed banks: The ecological and evolutionary implications of dormancy. Nat Rev Microbiol. 2011;9(2):119–30. doi: 10.1038/nrmicro2504 21233850
111. Sjöstedt J, Koch-Schmidt P, Pontarp M, Canbäck B, Tunlid A, Lundberg P, et al. Recruitment of members from the rare biosphere of marine bacterioplankton communities after an environmental disturbance. Appl Environ Microbiol. 2012;78(5):1361–9. doi: 10.1128/AEM.05542-11 22194288
112. Jousset A, Bienhold C, Chatzinotas A, Gallien L, Gobet A, Kurm V, et al. Where less may be more: How the rare biosphere pulls ecosystems strings. ISME J. 2017;11(4):853–62. doi: 10.1038/ismej.2016.174 28072420
113. Wang J, Pan F, Soininen J, Heino J, Shen J. Nutrient enrichment modifies temperature-biodiversity relationships in large-scale field experiments. Nat. Commun. 2016;7 : 1–9.
114. Wang J, Soininen J, He J, Shen J. Phylogenetic clustering increases with elevation for microbes. Environ. Microbiol. Rep. 2012;4(2):217–26. doi: 10.1111/j.1758-2229.2011.00324.x 23757276
115. Stal LJ. Microphytobenthos, their extracellular polymeric substances, and the morphogenesis of intertidal sediments. Geomicrobiol J. 2003;20(5):463–78.
116. Krumbein WE, Paterson DM, Stal LJ. Biostabilization of Sediments. Krumbein WE, Paterson DM, Stal LJ, editors. Oldenburg: Bibliotheks - und Informationssystem der Universität Oldenburg (bis) Verlag; 1994.
117. Komárek J. 3—Coccoid and Colonial Cyanobacteria. In: Wehr JD, Sheath RG, editors. Freshwater Algae of North America. Burlington: Academic Press; 2003. p. 59–116. (Aquatic Ecology).
118. Oren A. Cyanobacteria in hypersaline environments: biodiversity and physiological properties. Biodivers Conserv. 2015;24(4):781–98.
119. Pessi IS, Maalouf P de C, IV HDL, Baurain D, Wilmotte A. On the use of high-throughput sequencing for the study of cyanobacterial diversity in Antarctic aquatic mats. J Phycol. 2016;52 : 356–68. doi: 10.1111/jpy.12399 27273529
120. Whitton BA. Ecology of cyanobacteria II: Their diversity in space and time. Whitton BA, editor. Springer Netherlands; 2012. XV, 760.
121. Wiens JJ, Graham CH. Niche Conservatism: Integrating Evolution, Ecology, and Conservation Biology. Annu Rev Ecol Evol Syst. 2005;36(1):519–39.
122. Komárek J, Kaštovský J, Mareš J, Johansen JR. Taxonomic classification of cyanoprokaryotes (cyanobacterial genera) 2014, using a polyphasic approach. Preslia. 2014;86(4):295–335.
123. Abed RMM, Klempová T, Gajdos P, Certík M. Bacterial diversity and fatty acid composition of hypersaline cyanobacterial mats from an inland desert wadi. J Arid Environ. 2015;115 : 81–9.
124. Garcia-Pichel F, Nübel U, Muyzer G. The phylogeny of unicellular, extremely halotolerant cyanobacteria. Arch Microbiol. 1998;169(6):469–82. doi: 10.1007/s002030050599 9575232
125. Silva SMF, Pienaar RN. Some Benthic Marine Cyanophyceae of Mauritius. Bot Mar. 2000;43(1):11–27.
126. Büdel B, Weber B, Kühl M, Pfanz H, Sültemeyer D, Wessels D. Reshaping of sandstone surfaces by cryptoendolithic cyanobacteria: bioalkalization causes chemical weathering in arid landscapes. Geobiology. 2004;2(4):261–8.
127. Komárek J, Elster J, Komárek O. Diversity of the cyanobacterial microflora of the northern part of James Ross Island, NW Weddell Sea, Antarctica. Polar Biol. 2008;31(7):853–65.
128. de Los Ríos A, Valea S, Ascaso C, Davila A, Kastovsky J, McKay CP, et al. Comparative analysis of the microbial communities inhabiting halite evaporites of the Atacama Desert. Int Microbiol. 2010;13(2):79–89. doi: 10.2436/20.1501.01.113 20890842
129. Brandes M, Albach DC, Vogt JC, Mayland-Quellhorst E, Mendieta-Leiva G, Golubic S, et al. Supratidal extremophiles—Cyanobacterial diversity in the rock pools of the Croatian Adria. Microb Ecol. 2015;70(4):876–88. doi: 10.1007/s00248-015-0637-0 26048370
130. Palinska KA, Abed RMM, Vogt JC, Radtke G, Golubic S. Microbial endoliths on East Adriatic Limestone Coast: morphological vs. molecular diversity. Geomicrobiol J. 2017;34(10):903–15.
131. Pereira S, Zille A, Micheletti E, Moradas-Ferreira P, De Philippis R, Tamagnini P. Complexity of cyanobacterial exopolysaccharides: Composition, structures, inducing factors and putative genes involved in their biosynthesis and assembly. FEMS Microbiol Rev. 2009;33(5):917–41. doi: 10.1111/j.1574-6976.2009.00183.x 19453747
132. Komárek J. Süßwasserflora von Mitteleuropa, Bd. 19/3: Cyanoprokaryota. 3. Teil / 3rd part: Heterocytous Genera. Vol. 49. Springer Spektrum; 2013. XVIII, 1131.
133. Soininen J. Macroecology of unicellular organisms—patterns and processes. Environ Microbiol Rep. 2012;4(1):10–22. doi: 10.1111/j.1758-2229.2011.00308.x 23757224
Zvyšte si kvalifikaci online z pohodlí domova
Nemáte účet? Registrujte se
Zadejte e-mailovou adresu, se kterou jste vytvářel(a) účet, budou Vám na ni zaslány informace k nastavení nového hesla.
