Molecular phylogeny, scale evolution and taxonomy of centrohelid heliozoa
Introduction
Heliozoa as traditionally conceived are now known to be polyphyletic (Nikolaev et al., 2004), as earlier strongly suggested by their very different microtubule arrangements in the axonemes of the slender, radiating axopodia by which they feed (Smith and Patterson, 1986). Superficially similar minor groups formerly classified in phylum Heliozoa (Cavalier-Smith, 1993) have been removed one by one to their proper evolutionary homes within heterokont chromists (Cavalier-Smith et al., 1995, Nikolaev et al., 2004, Cavalier-Smith and Chao, 2006) or Cercozoa (Nikolaev et al., 2004). Only centrohelid heliozoa (order Centrohelida Kühn, 1924, as refined by Dürrschmidt and Patterson, 1986–1987) remain in the thus purified phylum Heliozoa, provisionally placed in the Chromista, because of a weak grouping of centrohelids with haptophytes on distance and maximum likelihood 18S rRNA trees (Cavalier-Smith and Chao, 2003a), shown also more strongly on Bayesian trees (Nikolaev et al., 2006). With around 85 described species, centrohelids or ‘centroheliozoa’ formed the largest part of the originally heterogeneous phylum Heliozoa. Siemensma (1991) recognised seven genera and Mikrjukov (2002) 15. Unlike the minor groups now removed from Heliozoa, centrohelids have flat mitochondrial cristae and axonemes with numerous microtubules arrayed in hexagons or triangles (Febvre-Chevalier, 1990).
In most centrohelids the axopodia (bearing numerous kinetocysts used for prey capture) project through a thick coat, often double, of silica scales (Fig. 1a, b, d and e), which are well characterised by electron microscopy. Coat features and scale morphology have been used to partition them into three families: Acanthocystidae (most have plate-like tangential silica scales with a well-developed central sternum; many also have pointed outer radial scales with a well-developed shaft); Raphidiophryidae (most have tangential rod-shaped scales; some also have outer radial trumpet/tube/funnel-like scales); and Heterophryidae (without silica scales; Chlamydaster has a mucous coat (Fig. 1c) and Oxnerella is naked; Heterophrys and Sphaerastrum have organic spicules embedded in a mucous coat).
To examine centrohelid phylogeny and test existing classifications, we cultured 35 new strains from seven of the 15 genera (11 species) recognised by Mikrjukov (2002) and sequenced their 18S rRNA genes. Scales of 18 were studied by electron microscopy, which enabled eight to be identified, but revealed ten new species. We conclude that the ancestral centrohelid heliozoan had both tangential and radial silica scales, and that the contrasting outer scale types that characterise Pterocystis/Choanocystis and Acanthocystis, respectively, probably diverged very early. Our findings require radical revision of centrohelid classification and several new species and higher taxa (effected in the Appendix A).
Centrohelid heliozoa are conspicuous by light microscopy and it has been assumed that most are cosmopolitan (Siemensma, 1991, Mikrjukov, 2001, Mikrjukov, 2002). Centrohelids are predominantly freshwater organisms, many fewer having been recorded from the sea or brackish water. To characterise their overall genetic diversity, to test whether discrete marine and freshwater lineages exist, and to begin exploring their global distribution, we used centrohelid-specific PCR primers to construct numerous 18S rRNA libraries from environmental samples. Such group-specific primers reveal the biodiversity of a group much more efficiently than do general eukaryotic primers (Bass and Cavalier-Smith, 2004, von der Heyden and Cavalier-Smith, 2005), such as those used in most recent environmental surveys which revealed an immense variety of novel and well-known lineages (López-Garcı´a et al., 2001, Moon-van der Staay et al., 2001, López-Garcı´a et al., 2003, Berney et al., 2004, Cavalier-Smith, 2004). No environmental sequences published before our present work finished grouped within centrohelids or with the, possibly related, unknown marine microheliozoan of Cavalier-Smith and Chao (2003a).
Our results confirm that centrohelids are predominantly freshwater, which we infer was their ancestral state. However, we discovered four exclusively marine, relatively ancient, multi-species clades and three other well-separated marine lineages. This refutes assumptions that there are no truly marine centrohelids or of free interchange between marine and freshwater habitats since marine morphotypes resemble freshwater ones (Siemensma, 1991, Mikrjukov, 2001).
Section snippets
Culturing centrohelid heliozoa
Table 1 lists the strains, their sources, and sequence accession numbers. Centrohelids were isolated by repeated serial dilutions of 50 μl of environmental samples, previously microscopically confirmed to contain centrohelids, into sterile microtitre wells (Nunclon) containing 230 μl of growth medium. Pipetting was very gentle to reduce damage to the fragile axopodia.
Freshwater strains were grown in ‘SES’ medium (UKNCC, 2001); marine strains were grown in ‘Plymouth Erdschreiber’ medium (UKNCC,
Phylogeny, species identification, and polyphyletic silica scale loss
The combination of a centrohelid-specific and a eukaryote-specific primer amplified 18S rRNA genes from 35 new cultures of centrohelid heliozoa. PCR amplification yielded products of 1.5–2.2 kb (Table 1). Larger products were from Polyplacocystis, Acanthocystis and Marophrys (=Heterophrys) marina. Length differences are caused by insertions in 18S rRNA of Acanthocystis and raphidiophryids (13 or 12 insertions, respectively). These insertions vary in length and sequence composition between the
Discussion
Even though the basal branching order among centrohelids is poorly resolved, as in most eukaryote phyla (Cavalier-Smith and Chao, 2003b), our phylogenetic analyses and ultrastructural studies support our revision of centrohelid familial classification (see Appendix A) and yield novel conclusions about centrohelid evolution. Our first conclusion is that members of clade B yield much longer branches on the 18S rRNA tree than those of clade A, indicating much more rapid evolution of their
Acknowledgments
We are indebted to Ken Clarke at the CEH/FBA, Windermere for the electron microscopic examinations. S.v.d.H. thanks the Natural Environment Research Council (NERC UK: Freshwater and Marine Microbial Biodiversity Programme) for financial support. T.C.-S. thanks NERC UK for research grants and professorial fellowship support and the Canadian Institute for Advanced Research (Evolutionary Biology Programme) for fellowship support.
References (52)
- et al.
Basic local alignment search tool
J. Mol. Biol.
(1990) Protist phylogeny and the high-level classification of Protozoa
Eur. J. Protistol.
(2003)- et al.
Ribosomal RNA evidence for chloroplast loss within Heterokonta: pedinellid relationships and a revised classification of ochristan algae
Arch. Protistenkd.
(1995) - et al.
Cosmopolitan metapopulations of free-living microbial eukaryotes
Protist
(2004) - et al.
Biodiversity of terrestrial protozoa appears homogeneous across local and global spatial scales
Protist
(2001) Protist diversity: estimates of the near-imponderable
Protist
(1999)- et al.
Morphological and molecular investigations of Paramecium schewiakoffi sp. nov. (Ciliophora, Oligohymenophorea) and current status of distribution and taxonomy of Paramecium spp
Eur. J. Protist.
(2004) Observations of centroheliozoa from Moscow marine aquaria
Arch. Protistenkd.
(1994)- et al.
Global eukaryote phylogeny: combined small- and large-subunit ribosomal DNA trees support monophyly of Rhizaria, Retaria and Excavata
Mol. Phyl. Evol.
(2007) - et al.
Phylum-specific environmental DNA analysis reveals remarkably high global biodiversity of Cercozoa (Protozoa)
Int. J. Syst. Evol. Microbiol.
(2004)