• polar cyanobacteria;
  • Phormidium;
  • 16S rRNA gene;
  • restriction fragment length polymorphism;
  • internal transcribed space;
  • morphotypes


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Selected strains of filamentous Phormidium-like cyanobacteria isolated from two Arctic regions (Ellesmere Island, High Canadian Arctic and Svalbard) and from Antarctica (Antarctic peninsula, South Shetland Islands and South Orkney Islands) were studied. The polyphasic approach used included phenotypic observations of morphological features and genotypic analyses (restriction fragment length polymorphism of 16S rRNA gene, internal transcribed space, 16S rRNA gene sequence analysis). Although genotypes generally correspond to observed morphotypes, the genetic analyses revealed a high degree of biodiversity that could not be unveiled using solely morphological evaluations. According to the phylogenetic analysis, the three clones were divided into two major clades, indicating that the phylogenetic distance between Arct-Ph5/Ant-Ph68 and Ant-Ph58 was so large they belonged to different genera. The polyphyletic position of strains of the genus Phormidium was confirmed by this study, attesting the need to entirely revise classification in this taxon in the future.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Cyanobacteria occur in very diverse ecological niches including extreme environments such as desert rocks, cold or hot springs, and very hostile cold circumpolar areas (Castenholz & Waterbury, 1989; Whitton & Potts, 2000). The biota of polar regions is dominated by microorganisms, of which cyanobacteria are the most abundant phototrophic component (for examples, see Vincent, 2000; Elster, 2002). They are well adapted to prolonged freezing (Davey, 1989; Šabackáet al., 2006) with the capability of metabolizing even at −20°C (Vincent et al., 2004). The filamentous types are the most common cyanobacterial groups that occur in various polar freshwater and terrestrial habitats. They belong to the division Oscillatoriales, mainly to genera Leptolyngbya, Phormidium and Oscillatoria (Komárek, 1999; Elster, 2002; Mueller et al., 2004).

Until recently, cyanobacterial taxonomy was based predominantly on morphological features evaluated by light microscope observations and culture techniques. Cyanobacteria from extreme environments were divided into species using reference literature designed for temperate areas (Broady & Kibblewhite, 1991; Komárek, 1999).

Nevertheless, the determination of the most abundant group, narrow filamentous types of cyanobacteria from the order Oscillatoriales, remains particularly challenging and uncertain. In many cases, these organisms are identified as Phormidium sp. (Stal & Krumbein, 1985) or classified under the genera Leptolyngbya (Nadeau et al., 2001), Schizothrix (Sage & Sullivan, 1978) or less frequently as Oscillatoria and Lyngbya.

In the past decade, molecular approaches have brought powerful new technology to the study of cyanobacterial diversity, including extreme polar environments (Garcia-Pichel et al., 2001). Currently, most phylogenetic analyses rely on amplification and sequencing of the ribosomal 16S rRNA gene (Lyra et al., 2001; Iteman et al., 2002; Casamatta et al., 2003), or restriction fragment length polymorphism analysis (RFLP) of 16S rRNA gene amplicons (Viti et al., 1997). However, the high level of conservation at this locus among cyanobacteria may limit its use for determination at the species level (Rossello-Mora & Amann, 2001). Therefore, the ITS (internal transcribed spacer) of the ribosomal operon and RFLP-ITS (Scheldeman et al., 1999; Iteman et al., 2002; Taton et al., 2003) are frequently used as complementary analyses and generally allow for identification at the species level (Lu et al., 1997; Iteman et al., 2002).

Despite the increasing interest in the ecological roles played by cyanobacteria in the polar regions and their biotechnological potential, our knowledge concerning their geographical distribution is very limited, even when considering the most common types (Vincent, 2000).

For some authors (Vincent, 2000), Antarctica, unlike any other region on the planet (including the Arctic), was relatively isolated from the rest of the world since its separation from Gondwanaland and the formation of the Polar Front >10 million years ago. Therefore, if there exist endemic species among microorganisms, it is very likely that they will be found in Antarctica. Many cosmopolitan species of microorganisms have been observed in various Antarctic marine and terrestrial environments; however, only a few of these microorganisms have been examined using molecular techniques. Moreover, their genetic relatedness to other populations remains to be confirmed.

The aim of this research was to study the relationships between nine Phormidium-like strains isolated from two Arctic regions (Svalbard archipelago Norway and Ellesmere Island, Canada) as well as several sites in maritime Antarctica (Antarctic peninsula), by a polyphasic (morphological, phylogenetic and genotypic) approach. RFLP analysis was carried out for both the 16S rRNA gene and ITS regions. Three complete 16S rRNA gene sequences were determined and compared to the available large sequences from various cyanobacterial taxa. The ubiquity and/or endemism and circumpolar distribution of Arctic and Antarctic filamentous cyanobacteria are discussed.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Sampling sites

Samples of cyanobacterial mats from mineral and ornithogenic soils, from shallow wetlands (seepages) and wetted soils, were collected from Ellesmere island and Svalbard archipelago in the Arctic area (Fig. 1a) and from localities near the western coast of the Antarctic peninsula, including King George and Signy Islands (Fig. 1b). Samples were frozen immediately after collection and maintained at −20°C during transportation to the laboratory.


Figure 1.  Location of sampling sites in polar regions. (a) Arctic sites; (b) Antarctic sites.

Download figure to PowerPoint

Strain preparation

Each cyanobacterial strain was purified by isolation of a single trichome, which was grown on agar plates with solidified BG-11 medium (Rippka et al., 1979). The plates were streaked with a small amount of field material or precultivated in tubes containing liquid BG-11 medium. After a few days of cultivation at 12 or 18°C, visible trichomes were observed and separately transferred to sterile agar tubes. Nine uni-algal strains have thus been isolated and cultured at a temperature of 6°C and light of 30 μmol m−2 s−1. Each isolated strain was placed in sterile tubes containing liquid medium (the Arct-Ph5 strain was thawed from liquid nitrogen storage) and sent to the Pasteur Culture Collection, where they were immediately placed in 100 mL Erlenmeyer flasks with 40 mL of BG 11 liquid media and incubated at 18°C, receiving white light (Osram White universal White) with a photosynthetic photon flux density of 10 μmol m−2 s−1, under a light/dark cycle of 14 h/10 h.

Microscopic observations

Isolated strains were regularly observed with a light microscope Olympus CX 40 and the life forms and life stages were recorded. The strains were determined according to the identification literature on both temperate regions (Geitler, 1932; Komárek & Anagnostidis, 2005) and Antarctic habitats (Broady & Kibblewhite, 1991; Komárek, 1999). All important qualitative and quantitative features (cell width, cell length, attenuation of apical cells, sheaths morphology) were evaluated.

DNA extraction and amplification of the 16S rRNA gene and ITS region

The DNA extraction was performed from 9 mL of mature culture, using the Nucleobond AXG100 kit (Macherey-Nagel, Düren, Germany). The lysis treatment (proteinase K and lysozyme) was completed with a supplementary incubation of 2 h at 80°C to improve disruption of the cells and inactivation of DNases. The lysate was then loaded on a flow-through cartridge according to the manufacturer's protocol. The 16S rRNA gene and ITS were amplified with the primer sets listed in Table 1. Each PCR reaction contained 5 μL Taq commercial buffer (10 × Promega), 2 mM MgCl2, 250 μM of each deoxynucleotide triphosphate (dATP, dCTP, dGTP, dTTP), 500 ng of each primer, 2.5 U Taq polymerase (Promega, Madison, WI) and 1 μL of template DNA. The volume was adjusted to 50 μL with sterile high purity water. The PCR reactions were performed in a 9700 Perkin Elmer thermocycler (Applied Biosystems, Foster City, CA) and initiated using a 95°C step for 5 min, followed by 35 cycles of 95°C (for 1.5 min), 55°C (for 2 min), 72°C (for 3 min) and a final elongation step of 72°C for 7 min. Storage of the reactions was at 4°C.

Table 1.   Primer sequences used for PCR amplification and their target sites
PrimersTarget*DirectionSequence (5′–3′)PositionReferences
  • *

    Direction of primer with respect to orientation of RNA: Forward (F) or Reverse (R).

  • Targeted region of amplification genes and spacer.

  • Numbering of the target sites corresponds to the nucleotide positions in the rrn operon of Escherichia coli.

A216S rRNA geneFAGAGTTTGATCCTGGCTCAG8–27Wilmotte et al. (1993); 16S27F, Taton et al. (2003)
567ITS-16S rRNA geneRGGTCTCCCTAAAAGGAGGTG1550–1531Iteman (2002)
75116S rRNA gene-ITSFCACCTCCTTTTAGGGAGACC1531–1550This study
34023S rRNA gene-ITSRCTCTGTGTGCCTAGGTATCC2054–2034Iteman et al. (2002)

An aliquot of 10 μL of each reaction was separated using gel electrophoresis (21 × 14.5 cm) in 1 × TBE (Tris-borate-EDTA) buffer at 120 V for 4 h. The bands on the gels were visualized by staining with ethidium bromide (0.5 μL mL−1) and photographed under UV light with an Image Master VDS-Pharmacia-Biotech.-Thermal Imaging System FTI-500.

RFLP screening of the two PCR products

The PCR products (10 μL), corresponding to the 16S rRNA genes and ITS regions of these nine strains were digested with 5 U of the following restriction enzymes: EcoRI/MboI/RsaI/StyI/XbaI for the 16S rRNA gene and AluI/HinfI/MspI for the ITS regions, according to the manufacturer's instructions (Gibco-BRL, Life Technologies). The DNA fragments were separated in a 1% (w/v) agarose gel (Litex FMC) and visualized as described above. A picture of the gel was saved as a TIF file and recorded with gelcompar II version 1.0 software (Applied Maths, Kortrijk, Belgium) to draw a dendogram based on the similarity matrix.

Cloning and sequencing of the 16S rRNA gene

One strain of each different RFLP-16S rRNA gene banding profile was selected for cloning and sequencing. To reduce quantitative PCR biases in single reactions (Polz & Cavanaugh, 1998), 10 separate PCR reactions were pooled. The combined PCR products were purified using the Wizard PCR Prep. DNA Purification System kit (Promega), prior to cloning into the pGEM-T-Easy vector (Promega). The ligation mixtures were transformed into competent Escherichia coli JM109 cells, as suggested by the manufacturer's instructions. Transformants were screened by colour selection on Luria agar plates containing IPTG, ampicillin (100 μg mL−1; Sigma, St Louis, MO) and X-Gal (Sigma) as an indicator. Putative positive colonies were picked from the plates and resuspended in 50 μL of sterile water. An aliquot of cell suspensions (1 μL) was used as template in a PCR reaction, using the two M13 universal primers (Promega). The amplification program consisted of one cycle of 95°C (for 5 min), 30 cycles of 94°C (for 45 s), 58.5°C (for 45 s), 72°C (for 60 s) and an elongation step of 72°C (10 min). Three clones from each library were randomly selected and sequenced on both strands of the inserts (ABI PRISM Dye terminator Cycle sequencing Kit, Foster City, CA) by GenomeExpress (Meylan, France).

Phylogenetic analyses

A first phylogenetic tree was constructed from a binary matrix where the characters were the presence/absence of RFLP-ITS bands. The neighbor-joining method (Saitou & Nei, 1987) was performed with the software package treecon for Windows 1.3b (Van de peer & De Wachter, 1994).

For the phylogenetic analyses, 16S rRNA gene sequences were aligned using genedoc v. 2.4 software package (Nicholas & Nicholas, 1997) and a final pairwise alignment was done using the clustalw program (Thompson et al., 1994). The secondary structure of the 16S rRNA gene (i.e. loop structure) was taken into account and manually checked in genedoc.

The second phylogenetic tree was constructed using large (1128 positions) 16S rRNA gene sequences with the paup version 4.0 software (Swofford, 1998). Two methods – the Neighbor-Joining Distance (NJ) and Maximum-Likelihood (ML) – were selected after evaluation of the best-fit model of nucleotide substitution with ModelTest 3.5 (Posada & Crandall, 1998). Only the NJ model is shown. The Evolutionary Model for the NJ method was the general time reversible model GTR+I+G (log-likelihood score LnL=−3914.8906) with a proportion of invariable sites (I) of 0.4663 and a Gamma distribution shape parameter of 0.6536 (for a number of gamma rate categories of 8). The nonparametric bootstrap analysis (500 replicates) was performed on the consensus tree. Only the bootstrap values higher than 60% were mentioned at the node level.

Nucleotide sequence accession numbers

Nucleotide sequences have been deposited in the GenBank-EMBL database, under the accession numbers DQ493872DQ493874.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Morphological characterization

The microscopic observations of nine polar strains (Table 2) revealed their relationship to the order Oscillatoriales, specifically to the genus Phormidium, based on recent taxonomic literature (Komárek & Anagnostidis, 2005). Strains were divided into three distinct morphotypes (Table 3) based on major phenotypic features.

Table 2.   Taxonomic assignment and origin of the cyanobacterial strains examined in this study
Taxonomic assignment1MorphotypeStrainPolar region latitude/longitudeLocationCollection yearHabitatReferences/isolator
Phormidium autumnale group Fig. 4a and bAAnt-Ph10Antarctic 67°34′S 68°08′WRothera-Adelaide island (Antarctic peninsula)2003Mineral soilŠabackáet al. (unpublished)
AAnt-Ph38Antarctic 64°46′S 64°03′WPalmer-Anvers island (Antarctic peninsula)2003Ornithogenous soilŠabackáet al. (unpublished)
AAnt-Ph68Antarctic 60°43′S 45°36′WSigny-Coronation island (South Orkney Islands)2003SeepageŠabackáet al. (unpublished)
AArct-Ph5Arctic 79°08′N 80°30′WEllesmere island, Sverdrup Pass (Canada)1991Mineral soilElster et al. (1997)
AArct-Ph34Arctic 79°57.8′N 11°21.2′ENy-Alesund (Spitzbergen-Svalbard)2002Mineral soilKaštovská (unpublished)
Phormidium murrayiCAnt-Ph16Antarctic 67°34′S 68°08′WRothera-Adelaide island (Antarctic peninsula)2003Ornithogenous soilŠabackáet al. (unpublished)
Table 3.   Morphological classification of the nine Phormidium-like strains compared to genotypic clustering from RFLP-ITS and 16S rRNA gene
MorphotypeMorphological traitsStrainsGenotypic cluster from RFLP analysis
16S rRNA geneITS
A Ph. autumnale– type Fig. 4a and bThick and straight trichomes, 4.5–8.5 μm wide, slowly attenuated and with curved end cells. Terminal hook over 2–8 (more) cells, cell w/l ratio = 0.9/4.86. Sheath present, colorless, calyptra present. Cells blue-green.Arct-Ph5 Arct-Ph34 Ant-Ph38 Ant-Ph10 Ant-Ph68IA IA IA IB IBIA IC IA IB IB
B Phormidium sp. Fig. 4cStraight trichomes 5–7.5 μm wide, dense keritomization inside the cells, slightly curved end cells. Rare terminal hook over 2–5 cells, calyptra present. Sheaths present, colorless. Cells olive green with intense granulationArct-Ph18IAIA
C Ph. murrayiStraight to flexous trichomes, no terminal attenuation of trichome 2.2–4.0(5) μm wide, cell w/l ratio =0.36/2.14Ant-Ph16 Ant-Ph53 Ant-Ph58IIIIA IIB
Fig. 4dSheath often absent, calyptra absent. Cells blue-green, with fairly homogeneous content.   

Morphotype A (Fig. 4a and b) was characterized by generally straight trichomes (width 4.5–8.5 μm) with frequently curved cell ends that usually contained calyptra. Cell length/width ratio ranged from 0.9 to 4.86. These main morphological characteristics are typical of the morphotype of the Phormidium autumnale group (Komárek & Anagnostidis, 2005, group VII). Five strains – Arct-Ph5, Ant-Ph38, Arct-Ph34, Ant-Ph10 and Ant-Ph68 – are included under this designation (Table 3). Further division within the Phormidium group was not possible for these strains at the morphological level.


Figure 4.  Light microscopy photomicrographs showing, in longitudinal section, the diversity of the cyanobacterial strains belonging to each morphotype (a, b, c). Additionally, drawings illustrated the various shape of trichomes and the main morphological characteristics for each strain. Scale bars, 20 μm. (a) Strain Arct-Ph5 (cluster IA –Ph. autumnale type); orig., b and c. (b) Strain Arct-Ph34 (cluster IA –Ph. autumnale type); orig., b and c. (c) Strain Arct-Ph18 (cluster IA –Phormidium sp.); orig., b, c and d. (d) Strain Ant-Ph58 (cluster II –Ph. murrayi); a – after Šabacká, (unpublished), b to d – orig. from natural material.

Download figure to PowerPoint

Strain Arct-Ph18 (Fig. 4c), originally identified as Phormidium sp. cf. (Table 2), was morphologically similar to the strains from Morphotype A, but with a very short attenuation at the end of the trichome, and with different structured cell content (more distinct keritomy caused by thylakoid arrangement, and more intense granulation). Consequently, strain Arct Ph18 was designated as Morphotype B (Table 3).

Three strains Ant-Ph 16, Ant-Ph53, and Ant-Ph 58 were morphologically distinct from the other strains, with narrow trichomes no more than 5 μm wide, a variable cell ratio, and a total absence of calyptra (Table 3). These strains were designated as Morphotype C (Fig. 4d) and assigned to taxon Phormidiummurrayi [(West and West) Anagnostidis & Komárek 1988 (=Lyngbya murrayiWest & West 1911)]. The species Ph. murrayi, however, was still noted as an unrevised taxon in the last taxonomic review of Komárek & Anagnostidis (2005), and supposedly belongs to a different genus from previously described clusters.

Genotypic analyses and comparison with morphological designations

The size of the 16S rRNA gene PCR amplicons was similar (∼1480 bp) for all strains when visualized by gel electrophoresis. Therefore, several restriction enzymes were used to further separate the strains. The first group of enzymes (MboI, StyI, XbaI) separated the Ant-Ph16, Ant-Ph53 and Ant-Ph58 from the others, whereas EcoRI and RsaI revealed an additional subgroup, by distinguishing the Ant-Ph10 and Ant-Ph68 (IB) to the Arct-Ph5, Arct-Ph18, Arct-Ph34 (IA) and Ant-Ph38 (Table 4).

Table 4.   Banding profiles of the 16S rRNA gene amplicons obtained with the nine cyanobacterial strains examined after hydrolysis with five restriction enzymes
IsolateEnzyme and length (bp) of restriction fragments16S rRNA gene-RFLP clusters
  1. Restriction patterns obtained for each restriction endonuclease; with EcoRI: A (613, 330, 555), B (NT=Not cut), C (620); MboI: A (9, 24, 235, 380, 827), B (15, 24, 227, 235, 377, 600); RsaI: A (157, 266, 282, 340, 407), B (157, 161, 266, 340, 409), C (8, 71, 121, 161, 340, 352, 405); StyI: A (232, 1118), B (50, 190, 196, 921); XbaI: A (1095), B (Not cut). The 16S rRNA gene-RFLP clusters are obtained from the combination of all restriction patterns.

  2. The 16S rRNA gene of strains indicated in bold type has been cloned and sequenced.


The combination of four restriction enzymes, performed on the ITS fragment (420 bp size for Ant-Ph16/Ant-Ph53 and Ant-Ph58; and 560 bp for other strains) allowed a higher level of resolution. Two additional subclades were detected. One subclade (IC, Fig. 2) showed a genetic divergence between the strain Arct-Ph34 isolated from ornithogenic soils and the other strains of cluster I, and the second one (IIA) differentiated the Ant-Ph16 strain from both Ant-Ph53 and Ant-Ph58 of group II (Fig. 2).


Figure 2.  Schematic RFLP-ITS banding profiles obtained from three restriction enzymes (AluI, HinfI, MspI). All restriction bands were taken into consideration for the binary matrix and construction of the Neighbor-Joining dendogram in treecon software. The scale bar indicates a Juke–Cantor distance of 0.1.

Download figure to PowerPoint

As a consequence, the comparison between the phenotypic and genotypic analyses revealed two major discordances (Table 3, Fig. 2): the first concerned the morphological distinction of Arct-Ph18 as a transient form of the Phormidium autumnale group, which was not congruent with the genotypic analyses; the second was the significant genetic divergence between the strains Ant-Ph10, Ant-Ph68 (cluster IB in Fig. 2), from the other strains of the group I, which was not morphologically detected within morphotype A.

Phylogenetic analyses

To generate the most accurate positions among studied strains, the phylogenetic tree in Fig. 3 was constructed using large 16S rRNA gene sequences (1128 positions) from diverse cyanobacterial taxa.


Figure 3.  Phylogenetic tree based on large 16S rRNA gene sequences (1128 positions) and obtained by the Neighbor-Joining method using paup 4.0 software. Bootstraps values (>60%) are noted above the nodes. The scale bar indicates 0.01 mutations per nucleotide positions. The three clones of the study are in bold type; the Arctic strains are underlined and the Antarctic ones are noted with Ant in superscript. Asterisks indicate psychrophilic strains.

Download figure to PowerPoint

Three major clades (I–III) were distinguished based on phylogenetic analysis (Fig. 3). Two clones (Arct-Ph5 and Ant-Ph68) were included in branch 1 of Clade I, and shared a high relatedness with nine other strains, of which seven originated from polar areas. The Arct-Ph5 was very closely related (Bootstrap value =100%) to another Arctic strain, Oscillatoria sp. E17, collected in the Norwegian region by Nadeau et al. (2001). Indeed, it shared 100% similarity with E17 on larger sequences, with no base substitution, even when considering the complete genomic region (data not shown).

Additionally, the Arct-Ph5 was closely related (>98% of similarity) to four Antarctic strains, including our clone Ant-Ph68, AntG16 and two Oscillatoria sp. strains Lunch and Orange (Fig. 3). This tight lineage was closely related to a group composed of various genera, such as Oscillatoria, Phormidium and Microcoleus (with Microcoleus rushforthii, UTCC296, and M. antarcticus UTCC474, sharing 97% of similarity). Within the cluster I.1, five strains (marked with asterisks in Fig. 3) possessed the particular physiological property to grow at low temperatures (<10°C) and were determined as psychrophilic (Nadeau et al., 2001; Casamatta et al., 2005).

The clone Ant-Ph58 was included in Clade II, within branch 5 (Fig. 3). It shared high similarity (98%) with a psychrophilic Antarctic strain originally named Ph. murrayi UTCC475 (Bootstrap values =100%) and recently renamed Microcoleus glaciei (Casamatta et al., 2005). Other strains of this sister branch were phylogenetically more distant, sharing <93% similarity (Fig. 3). In addition, the Ant-Ph68 sequence matched on shorter 16S rRNA gene regions and shared only 93.2% similarity with three nonpolar Microcoleus strains (data not shown).

Finally, the phylogenetic analysis has shown that the Phormidium strains (Arct-Ph5/Ant-Ph68) clustered more closely to the Arthrospira/Planktothrix strains (results congruent with Komárek & Kastovsky, 2003) than to the Ant-Ph58 strain (Fig. 3) and revealed such a large phylogenetic distance between the three clones that they must belong to different genera. The polyphyletic position of strains of the genus Phormidium was confirmed by this study. This was the case also for the strains of Microcoleus genus, which were located throughout the major clades (Fig. 3).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The morphological investigations undertaken in the present study have revealed three different morphotypes that were previously assigned to the genus Phormidium, according to the current taxonomic literature (Komárek & Anagnostidis, 2005). Morphotype A (Table 3, Fig. 4a and b) consists of strains that belong to the Ph. autumnale group (Komárek & Komárek, 1999). The Ph. autumnale group has frequently been observed to occur in polar habitats (Skulberg, 1996; Elster et al., 1997) and is often the dominant filamentous cyanobacterium on the Antarctic continent (Broady & Kibblewhite, 1991). Phormidium autumnale is also frequently reported from different latitudes and environments worldwide (Komárek & Anagnostidis, 2005).

Morphotype B (Table 3, Fig. 4c) differed from the typical Ph. autumnale group mainly by the presence of large keritomization inside cells. The distinct keritomization is known in various genera of cyanobacteria (Geitler, 1932) and can be caused by widening of thylakoids, as observed in the genus Tychonema (Komárek & Albertano, 1994). It may also be due to arrangement of radial thylakoids (as is the case for our strain Arct-Ph18). This second type of keritomization occurs commonly in the genus Phormidium.

The third morphotype, Morphotype C (Table 3, Fig. 4d), differed from the others by narrow trichomes, absence of terminal attenuation of the trichome and absence of calyptra. Morphotype C, designated as Ph. murrayi, has frequently been observed in various aquatic systems of continental Antarctica (Broady & Kibblewhite, 1991; Mataloni & Tell, 2002), but has not yet been recorded from King George Island, the site where this strain was collected (Komárek & Komárek, 1999).

Within the three morphotypes, no further subgroups were identified based on the morphological criteria. The only differences observed concerned slight variations in cell length or filament coloration, which could not be considered valid phenotypic characters due to their variability correlated to environmental parameters.

The 16S rRNA gene sequences of the three strains Arct-Ph5, Ant-Ph58 and Ant-Ph68, when compared to available sequences in databases, revealed their close relatedness with two other polar strains assigned to Oscillatoria sp. (Ant-G16, E17) by Nadeau et al. (2001). Both clones were related to the same strains within the same tightly defined lineage (Fig. 3), whatever the method used, and this was also supported by bootstrap values, indicating that this group of strains has real phylogenetic significance.

To date, we have not been able to entirely resolve the taxa in this study; however, we can assume by careful phenotypic analysis combined to two genotypic methods that Arct-Ph5 and Ant-Ph68 belong to the Ph. autumnale group but not to the same species (based on differences of 13 nucleotides in the 16S rRNA gene and in the RFLP-ITS banding profiles).

A more complex assignment concerns Ant-Ph58. This strain was related solely to one strain (Ph. murrayi UTCC475) collected from Antarctica, in a pond on McMurdo Ice Shelf (Casamatta et al., 2005), sharing a high similarity of 98% on large 16S rRNA gene sequences. It shared only low similarity (<93%) with other cyanobacterial genera available in international databases.

The alignments of small 16S rRNA gene sequences (<600 bases, data not shown) revealed also its relationship to four Microcoleus strains, which led Casamatta and collaborators to combine the phylogenetic and electronic microscopy analyses to the new designation of the Ph. murrayi in Microcoleus glaciei (Casamatta et al., 2005). Our results confirmed the separate generic position of Ph. murrayi from typical Phormidium species (e.g. from Ph. autumnale group). Nevertheless, the genetic coincidence of our strains with the genus Microcoleus remains to be tested in the future. Indeed, several polar strains of cyanobacteria have been revised, such as Oscillatoria sp. UTCC296 or Phormidium subfuscum UTCC474, renamed M. rushforthii and M. antarcticus by Casamatta et al. (2005). But they cannot belong to the same genus as Microcoleus glacei UTCC475 as there is such a large phylogenetic distance between them (shown in Fig. 3).

Our results show that the current genus Phormidium (in the sense of Komárek & Anagnostidis, 2005) is still very heterogeneous, and there is a need to entirely revise the Phormidium and Microcoleus taxa in the future. The polyphasic approach combining thorough evaluation of morphological and genetic features is very useful for addressing taxonomic questions, but has not yet been applied extensively (Boyer et al., 2002; Komárek & Kaštovsky, 2003).

The close position of Arct-Ph5 and Ant-Ph68 into the tight lineage (I.1, Fig. 3) composed of both polar strains, revealed that this clustered group was not geographically tied to one specific polar region, and brought new insight into the bipolar distribution of the Ph. autumnale group. These results are congruent with previous works, which detected a ‘bipolar distribution’ between the Antarctic and Arctic cyanobacteria (Darling et al., 2000) and within the Ph. autumnale group (Komárek & Anagnostidis, 2005).

Until now, very little rigorous evidence of the existence of either cosmopolitan or endemic distribution of cyanobacterial species has been found (Staley & Gosink, 1999), mainly because of the absence of molecular analyses (Castenholz, 1996) or the use of small 16S rRNA gene sequences (Garcia-Pichel et al., 1996), which could lead to inaccurate or biased interpretations. Indeed, it will be more and more difficult to find cosmopolitan species in the future with the increased use of powerful molecular methods, which can detect a wider diversity of cyanobacteria at a higher level of resolution. An intriguing case can be noted in this study, among the RFLP (Fig. 2, Table 4) banding profiles. Despite the relatively low number of restriction enzymes tested here, the RFLP on ITS sequences showed an identical profile for both Arctic strains Arct-Ph5 and Arct-Ph18 and the Antarctic strain Ant-Ph38, collected from ornithogenic soil (Fig. 2). The examination of a less conserved region (ITS) revealed the close relatedness of these three strains, and led us to derive two hypotheses about the unclear geographical origin of this strain, collected from ornithogenic substrate. The first one is that Ant-Ph38 was already locally present in Antarctica. Thus, other local samplings would most likely reveal the presence of this strain and open new debate about the existence of cosmopolitan species in the polar areas. The second explanation brings into focus the possibility of dispersal agents. Such dispersers could be marine migratory birds, which are known to cross the two hemispheres (Schlichting et al., 1978), and may accidentally introduce alien strains in Antarctica.

This study has revealed the efficiency of a polyphasic approach to determine the relatedness of different polar strains, but also highlights the lack of data available in databanks to continue further phylogenetic investigations on the dominant genus of filamentous cyanobacteria in extreme cold areas. Indeed, there is a crucial need to deposit new 16S rRNA gene sequences of Phormidium strains to allow for better resolution of the taxonomic problems and the still unclear phylogenetic position of this genus, and to permit new investigations about the existence of endemic species at each pole.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors wish to thank Dr Isabelle Iteman, Mrs Therese Coursin and Anne-Marie Castets for helpful experimental advice; Sabrina Cadel and Dr Mike Herdman for valuable discussions and advice with the phylogenetic analysis. We express our gratitude to the three reviewers who provided constructive comments to improve the original manuscript. K. Comte was a recipient of the European project fellowship ‘COBRA’ (QLRI-2001-01645). The study was supported by the Institut Pasteur and the Centre National de la Recherche Scientifique (CNRS, URA 2172) in France. The Czech coauthors were supported by the grant GA CR 206/05/0253. We are grateful to Drs Christine C. Foreman, Alyssa Carre-Mlouka and Andy Tolonen for language corrections.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Anagnostidis K & Komárek J (1988) Modern approach to the classification system of cyanophytes 3. Oscillatorialese Arch Hydrobiol 50: 327472.
  • Boyer SL, Johansen JR & Flechtner VR (2002) Phylogeny and genetic variance in terrestrial Microcoleus (Cyanophyceae) species based on sequence analysis of the 16S rRNA gene and associated 16S-23S ITS region. J Phycol 38: 12221235.
  • Broady PA & Kibblewhite AL (1991) Morphological characterization of Oscillatoriales (cyanobacteria) from Ross Island and southern Victoria Land Antarctica. Antarct Sci 3: 3545.
  • Casamatta DA, Vis ML & Sheath RG (2003) Cryptic species in cyanobacterial systematics: a case study of Phormidium retzii (Oscillatoriales) using RAPD molecular markers and 16S rDNA sequence data. Aquat Bot 77: 295309.
  • Casamatta DA, Johansen JR, Vis ML & Broadwater ST (2005) Molecular and ultrastructural characterization of ten polar and near-polar strains within the Oscillatoriales (Cyanobacteria). J Phycol 41: 421438.
  • Castenholz RW (1996) Endemism and biodiversity of thermophilic cyanobacteria. Nova Hedwigia 112: 3347.
  • Castenholz RW & Waterbury JB (1989) Oxygenic photosynthetic bacteria (sect. 19), Group I. Cyanobacteria. Bergey's Manual of Systematic Bacteriology (StaleyJT, BryantMP, PfennigN & HoltJG, eds), pp. 17101728. Williams & Wilkins, Baltimore, MD.
  • Darling KF, Wade CM, Stewart IA, Kroon D, Dingle R & Brown AJL (2000) Molecular evidence for genetic mixing of Arctic and Antarctic subpolar populations of planktonic foraminifers. Nature 405: 4347.
  • Davey MC (1989) The effect of freezing and dessication on photosynthesis and survival of terrestrial antarctic algae and cyanobacteria. Polar Biol 10: 2936.
  • Elster J (2002) Ecological classification of terrestrial algal communities in polar environments. Geoecology of Antarctic Ice-Free Coastal Landscapes (BeyerL & BölterM, eds), pp. 303326. Springer Verlag, Berlin, Heidelberg.
  • Elster J, Svoboda J, Komarek J & Marvan P (1997) Algal and cyanoprocaryote communities in a glacial stream, Swerdrup Pass 79°N, central Ellesmere Island, Canada. Arch Hydrobiol (Suppl) 85: 5793.
  • Garcia-Pichel F, Prufert-Bebout L & Muyzer G (1996) Phenotypic and phylogenetic analyses show Microcoleus chthonoplastes to be a cosmopolitan cyanobacterium. Appl Environ Microbiol 62: 32843291.
  • Garcia-Pichel F, Lopez-Cortes A & Nübel U (2001) Phylogenetic and morphological diversity of cyanobacteria in soil desert crusts from the Colorado Plateau. Appl Environ Microbiol 67: 19021910.
  • Geitler L (1932) Cyanophyceae. Kryptogamen-Flora von Deutschland, Österreich der Schweitz, Vol. 14 (RabenhorstL, ed), pp. 11196. Akad. Verlagsgesellschaft, Leipzig.
  • Iteman I, Rippka R, Tandeau De Marsac N & Herdman M (2002) rDNA analyses of planktonic heterocystous cyanobacteria, including members of the genera Anabaenopsis and Cyanospira. Microbiology 148: 481496.
  • Komárek J (1999) Diversity of cyanoprokaryotes (cyanobacteria) of King George Island, maritime Antarctica – a survey. Arch Hydrobiol (Suppl) 94: 181193.
  • Komárek J & Albertano P (1994) Cell structure of a planktic cyanoprokaryote, Tychonema bourrellyi. Arch Hydrobiol/Algolog Stud 75: 157166.
  • Komárek J & Anagnostidis K (2005) Cyanoprokaryota-2. Teil/2nd Part: Oscillatoriales. (BudelB, KrienitzL, GartnerG & SchagelM, eds) Süsswasserflora von Mitteleuropa 19/2, Elsevier Spektrum, Heidelberg.
  • Komárek J & Kastovšky J (2003) Adaptability in diversification processes of cyanobacteria; the example of Synechococcus bigranulatus. Arch Hydrobiol/Algol Stud 109: 299304.
  • Komárek O & Komárek J (1999) Diversity of freshwater and terrestrial habitats and their oxyphototroph microflora in the Arctowski Station region, South Shetlands Islands. Pol Polar Res 20: 259282.
  • Lu W, Evans H, McColl SM & Saunders VA (1997) Identification of cyanobacteria by polymorphisms of PCR-amplified ribosomal DNA spacer region. FEMS Microbiol Lett 153: 141149.
  • Lyra C, Suomalainen S, Gugger M, Vezie C, Sundman P, Paulin L & Sivonen K (2001) Molecular characterization of planktonic cyanobacteria of Anabaena, Aphanizomenon, Microcystis and Planktothrix genera. Int J Syst Evol Microbiol 51: 513526.
  • Mataloni G & Tell G (2002) Microalgal communities from ornithogenic soils at Cierva Point, Antarctic Peninsula. Polar Biol 25: 488491.
  • Mueller DR, Vincent WF, Bonilla S & Laurion S (2004) Extremotrophs, extremophiles and broadband pigmentation strategies in a high arctic ice shelf ecosystem. FEMS Microbiol Ecol 53: 7387.
  • Nadeau TL, Milbrandt EC & Castenholtz RW (2001) Evolutionary relationships of cultivated Antarctic Oscillatorians (Cyanobacteria). J Phycol 37: 650654.
  • Nicholas KB & Nicholas HB (1997). Genedoc: a tool for editing and annoting multiple sequence alignments.
  • Polz MF & Cavanaugh CM (1998) Bias in template-to-product ratios in multitemplate PCR. Appl Environ Microbiol 64: 37243730.
  • Posada D & Crandall KA (1998) ModelTest: testing the model of DNA substitution. Bioinformatics 14: 817818.
  • Rippka R, Deruelles J, Waterbury J, Herdman M & Stanier RY (1979) Genetic assignments, strain histories and properties of pure cultures of cyanobacteria. J Gen Microbiol 111: 161.
  • Rossello-Mora R & Amann R (2001) The species concept for prokaryotes. FEMS Microbiol Revi 25: 3967.
  • Šabacká M & Elster J (in press) Response of cyanobacteria and algae from Antarctic wetland habitats to freezing and desiccation stress. Polar Biol DOI DOI: 10.1007/s00300-006-0156-z.
  • Sage WW & Sullivan MJ (1978) Distribution of blue-green algae in a Mississippi gulf coast salt marsh. J Phycol 14: 3337.
  • Saitou N & Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4: 406425.
  • Scheldeman P, Baurain D, Bouhy R, Scott M, Muhling M, Whitton BA, Belay A & Wilmotte A (1999) Arthrospira (‘Spirulina’) strains from four continents are resolved into only two clusters, based on amplified ribosomal DNA restriction analysis of the internally transcribed spacer. FEMS Microbiol Lett 172: 213222.
  • Schlichting HEJ, Speziale BJ & Zink RM (1978) Dispersal of algae and protozoa by Antarctic flying birds. Antarctic J US 13: 147149.
  • Skulberg OM (1996) Toxins produced by cyanophytes in Norwegian inland waters – health and environment. Chemical data as a basis of geomedical investigations (LagJ, ed), pp. 197216. The Norwegian Academy of Science and Letters, Oslo.
  • Swofford DL (1998) PAUP*, Phylogenetic Analysis Using Parsimony (* and other methods). Version 4.0 Beta 10. Sinauer, Sunderland, MA.
  • Stal LJ & Krumbein WE (1985) Isolation and characterization of cyanobacteria from a marine microbiol mat. Bot Mar 28: 351365.
  • Staley JT & Gosink JJ (1999) Poles apart: biodiversity and biogeography of sea ice bacteria. Annu Rev. Microbiol 53: 189215.
  • Taton A, Grubisic S, Brambilla E, De Wit R & Wilmotte A (2003) Cyanobacterial diversity in natural and artificial microbial mats of lake Fryxell (McMurdo dry valleys, Antarctica): a morphological and molecular approach. Appl Env Microbiol 69: 51575169.
  • Thompson JD, Higgins DG & Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 46734680.
  • Van de Peer Y & De Wachter R (1994) treecon for Windows: a software package for the construction and drawing of evolutionary trees for the Microsoft Windows environment. Comput Appl Biosci 10: 569570.
  • Vincent WF (2000) Evolutionary origins of antarctic microbiota: invasion, selection and endemism. Antarct Sci 12: 374385.
  • Vincent WF, Mueller DR & Bonilla S (2004) Ecosystems on ice: the microbial ecology of Markham ice shelf in the high arctic. Cryobiology 48: 103112.
  • Viti C, Ventura S, Lotti F, Capolino E, Tomaselli L & Giovannetti L (1997) Genotypic diversity and typing of cyanobacterial strains of the genus Arthrospira by very sensitive total DNA restriction profile analysis. Res Microbiol 148: 605611.
  • West W & West GS (1911) Freshwater algae. British Antarctic Expedition 1907–1909, Reports on the Scientific Investigations. Biology, 1: 263298.
  • Whitton BA & Potts M (2000) The ecology of cyanobacteria. Their Diversity in Time and Space. Kluwer Academic Publishers, UK, London.
  • Wilmotte A, Vanderauwera G & Dewachter R (1993) Structure of the 16S ribosomal-RNA of the thermophilic cyanobacterium Chlorogloeopsis HTF (Mastigocladus-laminosus HTF) strain PCC7518 and phylogenetic analysis. FEBS Letters 317: 96100.