Elsevier

Aquatic Botany

Volume 77, Issue 4, December 2003, Pages 295-309
Aquatic Botany

Cryptic species in cyanobacterial systematics: a case study of Phormidium retzii (Oscillatoriales) using RAPD molecular markers and 16S rDNA sequence data

https://doi.org/10.1016/j.aquabot.2003.08.005Get rights and content

Abstract

The genetic variability of the cosmopolitan, ubiquitous freshwater cyanobacterium Phormidium retzii was assessed using random amplified polymorphic DNA (RAPD) markers and nearly complete (ca. 95%) 16S rDNA sequences. Strains consistent with the morphological species circumscription were utilized from geographically distant locations ranging from British Columbia, Canada, in the north to Rio Claro, Costa Rica in the south, and from Rhode Island, USA in the east coast to Washington State, USA on the west coast. In addition, some strains were from geographically close (<35 km apart) locales. To assess biogeographic trends, the nine RAPD primers utilized yielded 133 distinct bands in total from the 12 strains. Non-metric multidimensional scaling (NMDS) analyses revealed that the strains isolated from Ohio, Connecticut, and Rhode Island were closely related with the Costa Rica strain and one Washington State strain associated with this group. The two strains from Canada, British Columbia and Ontario, were closely related and clustered with the two other strains from Washington State. The remaining two strains from Mexico and Missouri were loosely associated with this cluster and genetically distinct. Geographic proximity did not correlate with genetic similarity (Mantel test, r=0.235,P>0.05). A 1340 bp region of the 16S rRNA gene was sequenced for 8 of the 12 strains used in the RAPD analysis. Sequence homology among strains ranged from 88.4 to 98.4%, implying the presence of cryptic species within this group of strains. Given the lack of sequence similarity, P. retzii as presently circumscribed most likely represents several cryptic species not clearly distinguishable with light microscopy of morphological features. This conclusion may explain the lack of correlation between geographic proximity and genetic similarity. This research provides further strength for the case of species of cyanobacteria being more localized rather than having global distributions.

Introduction

Cyanobacteria are among the most widespread, morphologically distinct, and abundant prokaryotes known (Whitton, 1992). However, phylogenetic relationships among the cyanobacteria are relatively poorly understood. Traditionally, they have been classified using morphological and ecological characters (Geitler, 1932, Desikachary, 1959). Stanier et al. (1978) and Rippka et al. (1979) argued that the systematic treatment of cyanobacteria should not be based on traditional botanical methods, but rather on type cultures deposited in permanent culture collections as with other bacteria. However, morphological changes induced by culture conditions and environmental plasticity can be problematic for cyanobacterial taxonomy based solely on morphology (Stanier et al., 1971, Palinska et al., 1996).

In total, very few cyanobacteria have been analyzed using new molecular techniques such as DNA sequencing, random amplified polymorphic DNA (RAPD), and DNA polymorphisms (Mullins et al., 1995, Neilan, 1995, Neilan et al., 1997, Otsuka et al., 1999). For broad phylogenetic studies, sequence data from the 16S rRNA gene are most commonly employed due to its utility for distinguishing higher level taxonomic groups as well as traditional species (Neilan, 1995, Nelissen et al., 1995). Of the sequenced taxa, most have been those responsible for potable water problems (Neilan et al., 1997, Otsuka et al., 1999) or taxa primarily from marine environments (e.g., Moore et al., 1998, Urbach et al., 1998). In terms of molecular data, one particularly understudied section of cyanobacteria is the Oscillatoriales, which ostensibly falls within Section 3 of the cyanobacteria (filamentous members that never form heterocysts, Rippka, 1988). This grouping has traditionally included the well known genera Oscillatoria, Phormidium, and Lyngbya, which are primarily differentiated by sheath properties (Geitler, 1932). However, sheath production is only one of the typical diagnostic features subject to the direct effects of both environmental and culture conditions and as such may not be phylogenetically informative (Rippka et al., 1979, Whitton, 1992). This section has been shown to be polyphyletic, and some authors have suggested that any species designations at this point would be premature (e.g., Castenholz, 2001), although others would disagree (i.e., Anagnostidis and Komárek, 1988).

Phormidium retzii (C.A. Agardh) Gomont is a cosmopolitan cyanobacterium (Geitler, 1932). Sheath and Cole (1992) reported that it is the most widespread macroalgal (forming macroscopic mats) species in North America and collected in all biomes sampled. This cyanobacterium also has been recorded in Europe and Asia (e.g., Desikachary, 1959, Drouet, 1968). In North America alone, it has been recorded from a wide range of habitats, including epiphytic on mosses in Newfoundland, springs in Wisconsin, ponds in Minnesota, as well as collections from Florida, Alabama and Texas, warm mud in North Carolina and streams in south-central Alaska (Tilden, 1910, Whitford and Schumacher, 1973, Sheath et al., 1986). Further, P. retzii was reported as the most widespread taxon encountered from lotic ecosystems of São Paulo State, Brazil (Branco et al., 1999). This taxon was also found in a survey of cyanobacterial diversity of King George Island, Antarctica, which as the author points out, had a similar ecology with populations outside of Antarctica (Komárek, 1999). Thus, this species is truly a widespread, ecologically amorphous taxon. In addition, this taxon has an unusually broad species description encompassing samples with occasionally slightly constricted cell walls, straight cells 4.5–12 μm wide by 4–9 μm long with a thin sheath (Geitler, 1932). Due to the potentially great degree of environmentally induced morphological variability, molecular markers may be invaluable for resolving questions pertaining to the systematics and biogeography of this organism. Therefore, the present study was initiated investigating 12 strains of P. retzii using RAPD analysis and 16S rDNA sequence data to examine genetic variation across North America in this cosmopolitan cyanobacterium.

Section snippets

Field and tissue collection

A total of 12 P. retzii strains were collected from across North America (Fig. 1). All strains morphologically fit the species concept for P. retzii, even after recent systematic revision (sensu Anagnostidis and Komárek, 1988). Motile trichomes were 4.5–12 μm in diameter, more or less isodiametric and granular, with thin, diffluent sheaths and truncated apical cells. After collection, some strains (Table 1) were grown on 1.5% agar plates containing BG-11 medium (Stanier et al., 1971). Unialgal

Results

A total of 133 RAPD bands were generated using 9 primers. The total number of bands per strain ranged from 36 to 55 (mean=47.25,S.D.=±7.4). Seven of the nine primers generated bands in all strains, although OPAA-06 and OPAA-09 did not. The removal of the bands generated by these primers did not greatly alter the patterns of grouping (data not shown). Bands generated from all primers were utilized, as the Jaccard coefficient is suitable for data which may include double-negative values. Since

Discussion

The species circumscription of P. retzii is one of the most broadly delimited in the cyanobacteria. Geitler (1932) describes the species as: filaments±straight, mostly not or sometimes constricted at the cross-walls, end cells not capitate nor tapering, 4.5–12 μm wide with a thin sheath, cells shorter than wide, 4–9 μm long, non-granular cross-walls, in flowing or standing water on rocks and cosmopolitan. Given the wide range of morphology attributed to this taxon, Drouet (1968) combined this

Conclusion

This analysis of P. retzii strains from across North America was initially undertaken to be an assessment of the biogeography of this taxon. Microscopic examination of the strains indicated that they all correspond to the species description sensu stricto Geitler (1932). However, 16S rDNA sequencing to confirm taxonomic assignment revealed 16S rDNA similarity scores often <97%, which has typically been used to infer different species. At this time, it is premature to assign new species to this

Acknowledgments

The authors extend their thanks to Melissa Hall, Mary Koske, Kirsten Müller, G. Lemon, Tara Rintoul and Alison Sherwood for assistance in collection of sample materials. We thank Nanda Filkin for translation of German texts. Jeff Johansen provided valuable discussion on species concepts. In addition, this work was partially supported by a Sigma-Xi grant in aid of Research to DAC. Portions of this manuscript were written while DAC was supported by an Ohio University Graduate Fellowship.

References (59)

  • Castenholz, R.W., Waterbury, J.B., 1989. Oxygenic photosynthetic bacteria. Group 1. Cyanobacteria. In: Staley, J.T.,...
  • F.M Cohan

    The effects of rare but promiscuous genetic exchange on evolutionary divergence in prokaryotes

    Am. Nat.

    (1994)
  • F.M Cohan

    Bacterial species and speciation

    Syst. Biol.

    (2001)
  • Desikachary, T.V., 1959. Cyanophyta. Indian Council of Agricultural Research, New...
  • Drouet, F., 1968. Revision of the Oscillatoriaceae. Academy of Natural Sciences of Philadelphia, Philadelphia,...
  • Eldredge, N., Cracraft, J., 1980. Phylogenetic Patterns and the Evolutionary Process. Columbia University Press, New...
  • V.R Flechtner et al.

    Spirirestis rafaelensis gen. et sp. nov. (Cyanophyceae), a new cyanobacterial genus from arid soils

    Nova Hedwigia

    (2002)
  • G.E Fox et al.

    How close is close: 16S rRNA sequence identity may not be sufficient to guarantee species identity

    Int. J. Syst. Bacteriol.

    (1992)
  • F Garcia-Pichel et al.

    Phenotypic and phylogenetic analyses show Microcoleus chthonoplastes to be a cosmopolitan cyanobacterium

    Appl. Environ. Microbiol.

    (1996)
  • Geitler, L., 1932. Cyanophyceae. In: Rabenhorst, L. (Ed.), Kryptogamen-flora von Deutschland, Österreich und der...
  • P Jaccard

    Etude comparative de la distribution florale dans une portion des Alpes et des

    J. Bull. Soc. Vaud. Sci. Nat.

    (1901)
  • M Kimura

    A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide substitutions

    J. Mol. Evol.

    (1980)
  • J.P Kociolek et al.

    Freshwater diatom biogeography

    Nova Hedw.

    (2000)
  • J Komárek

    Do all cyanophytes have a cosmopolitan distribution? Survey of the freshwater cyanophyte flora of Cuba

    Arch. für Hydrobiol. Suppl.

    (1985)
  • J Komárek

    Diversity of cyanoprokaryotes (Cyanobacteria) of King George Island, maritime Antarctica—a survey

    Arch. für Hydrobiol. Suppl.

    (1999)
  • Komárek, J., Anagnostidis, K., 1999. Cyanoprokaryota 1. Teil: Chroococcales. In: Ettl, H., Gerloff, J., Heynig, H.,...
  • J Lehtimäki et al.

    Characterization of Nodularia strains, cyanobacteria from brackish waters, by genotypic and phenotypic methods

    Int. J. Syst. Evol. Microbiol.

    (2000)
  • Liedloff, A., 1999. Mantel ver. 2.02. Non-parametric Mantel Test calculator....
  • Maddison, W.P., Maddison, D.R., 1992. MacClade, ver. 3.0. Sinauer,...
  • Cited by (73)

    • Cyanobacterial diversity and taxonomic uncertainty: polyphasic pathways to improved resolution

      2021, Advances in Phytoplankton Ecology: Applications of Emerging Technologies
    • Metabarcoding reveals a more complex cyanobacterial community than morphological identification

      2019, Ecological Indicators
      Citation Excerpt :

      In fact, phylogeny inferred from molecular data often disagrees with traditional taxonomy based on morphological features (Komárek, 2013). Molecular tools more readily detect cryptic species (Casamatta et al., 2003), and numbers of genotypes found in a sample typically exceed the number of morphotypes identified (Taton et al., 2003). Thus, metabarcoding is likely to detect more cyanobacterial taxa, however, to our knowledge, such data have not been compared to morphological identifications.

    • Cyanobacterial Taxonomy: Morphometry to Molecular Studies

      2018, Cyanobacteria: From Basic Science to Applications
    • Description of new genera and species of marine cyanobacteria from the Portuguese Atlantic coast

      2017, Molecular Phylogenetics and Evolution
      Citation Excerpt :

      The molecular characterization of these isolates, based on partial 16S rRNA gene sequences, showed that 15% of these sequences had less than 97% similarity compared to the sequences available in the databases emphasizing the presence of novel cyanobacterial diversity (Brito et al., 2012). Despite all these strains have been previously identified at the genus or species level (Brito et al., 2012), it is well known that traditional cyanobacterial taxa may encompass several cryptic species or even genera, i.e. morphologically (almost) identical organisms that fit the same current taxon but that belong to phylogenetically distinct clades/lineages (Casamatta et al., 2003; Komárek, 2016; Komárek et al., 2014). Therefore, the main goal of this study was to further characterize seven of those isolates (two unicellular and five filamentous) and to re-evaluate their classification according to the recent scheme proposed by Komárek et al. (2014), and the subsequent taxonomic recommendations and principles (Komárek, 2016).

    View all citing articles on Scopus
    View full text