Validation and justification of the phylum name Cryptomycota phyl. nov.

Abstract

The recently proposed new phylum name Cryptomycota phyl. nov. is validly published in order to facilitate its use in future discussions ofthe ecology, biology, and phylogenetic relationships ofthe constituent organisms. This name is preferred over the previously tentatively proposed “Rozellida” as new data suggest that the life-style and morphology of Rozella is not representative of the large radiation to which it and other Cryptomycota belong. Furthermore, taxa at higher ranks such as phylum are considered better not based on individual names of included genera, but rather on some special characteristics - in this case the cryptic nature of this group and that they were initially revealed by molecular methods rather than morphological discovery. If the group were later viewed as a member of a different kingdom, the name should be retained to indicate its fungal affinities, as is the practice for other fungal-like protist groups.

Introduction

The designation “cryptomycota” was introduced by Jones et al. (2011) to accommodate a well-supported clade (using ribosomal DNA (rDNA) phylogenies) of organisms putatively branching deep within the fungal radiation. The rank of phylum is the most appropriate for this group as current results show that it has fungal characteristics but is distinct from other fungi in not having a chitin-rich cell wall in the major stages of its life-cycle so far identified, including putative trophic interactions. However, Cryptomycota was not validly published as a scientific name in that work as no Latin diagnosis was provided (McNeill et al. 2006: Art. 36). A Latin diagnosis is provided here in order to formally establish the name. In addition, comments are made on our decision to introduce this name rather than take up the earlier informal name “Rozellida”, and on the distinctive features of the phylum and its position.

Taxonomy

Cryptomycota M. D. M. Jones & T. A. Richards, phyl. nov.

MycoBank MB563383

Etymology: crypto- — hidden; and -mycota, a phylum of fungi.

Fungi unicellulares, zoosporis unicellularis, uniflagellatibus, flagellis microtubularis, cystes sine tunica chitinosa vel cellulosa. Consortia epibiontica formata.

Fungi unicellular, zoospores single-celled with a single microtubular flagellum, and cysts without a chitin/cellulose cell wall. Forming epibiontic associations.

Representatives: GenBank accession nos AJ130857, AJ130849.1, AJ130850, FJ687265, FJ687267 and FJ687268, and Rozella.

Illustrations: Jones et al. (2011: figs 1d, 2a-e).

Discussion

It has been demonstrated that Rozella occupies a deep branching position in phylogenetic analyses of kingdom Fungi (James et al. 2006a, b), although bootstrap support for this relationship is inconsistent and often weak in the most comprehensively sampled phylogenies (James et al. 2006a, b, Jones et al. 2011). The name “Rozellida” was coined by Lara et al. (2010) to accommodate Rozella and a number of environmental sequences that form a distinct clade, but we refer to this group henceforth as Cryptomycota for reasons indicated below. Jones et al. (2011) showed that Cryptomycota are more diverse than previously recognised and that the molecular diversity of this group may be as diverse as the rest of the known Fungi according to rDNA gene markers.

Members of Cryptomycota are found in freshwater, soil, sediment, and some marine habitats. Jones et al. (2011) used lineage-specific fluorescence in situ hybridization (FISH), cell wall stains, and immuno-fluorescence staining to show two distinct lineages within Cryptomycota, which comprised ovoid cells of ca. 5 µn diam, existing in at least three morphologies in freshwater environments: uniflagellate zoospores, more variably-shaped cells without flagella attached to other eukaryotic microscopic organisms (e.g. diatom hosts), and non-flagellate cysts. None of these stages were shown to possess a chitin or cellulose wall, although other life-cycle phases with a chitin and/or cellulose cell wall may remain undetected. A chitin cell wall is sometimes cited as defining feature of kingdom Fungi, although we note that this is not a reliable diagnostic feature as distantly related protist groups also possess chitin on their cell surface (e.g. Kneipp et al. 1998).

The name “Rozellida” was applied to this phylum by Lara et al. (2010) but in an informal way between inverted commas and with no formal diagnosis. The ICZN (International Code of Zoological Nomenclature; International Commission on Zoological Nomenclature 1999) does not apply to names above the rank of family-group, but if it were in those ranks it would be viewed as unavailable as a conditional name (Art. 15.1). For names introduced under the ICZN which later are found to belong to Fungi, the ICN (International Code of Nomenclature for algae, fungi, and plants) now accepts them as available under Art. 45.4 (as revised at the Melbourne Congress in July 2011; McNeill et al. 2011). Thus, no Latin diagnosis was required, as it was for fungal names introduced between 1935 and 1 January 2012). However, we are inclined not to accept “Rozellida” because of the use of the inverted commas suggesting the usage was a tentative suggestion and in any case note that it is not mandatory to follow the principle of priority of publication for names above the rank of family (ICN) or family group (ICZN). Indeed, the ICZN does not cover ranks higherthan family group.

We decided that it would be better not to definitely establish a name based on Rozella for several reasons:

• (1) The fungal termination to be used for names in the rank of phylum is “-mycota” under the ICN (McNeill et al. 2006; Art 16.4), and that termination has also been used for phyla traditionally studied by mycologists but which are no longer considered Fungi but placed in other kingdoms. Examples include Hyphochytriomycota R.H. Whittaker (Whittaker 1969: 154) now placed in Straminipila M.W. Dick 2001, Myxomycota Bold (Bold 1957: 152) for slime moulds in the Protozoa, and Oomycota Arx (Arx 1967: 16) for fungal analogues in the Straminipila. This practice has been employed in standard reference works (e.g. Kirk et al. 2001) and also the most recent textbooks (e.g. Moore et al. 2011).

• (2) Cryptomycota represent a very diverse radiation, potentially equivalent to or larger than the rest of the known fungi. Of the three lineages within the radiation for which morphological data exist, Rozella appears to be exceptional in that it is primarily an intracellular parasite; indeed the possession of intracellular sporangia is included in the generic description of Rozella species (Held 1981). To extend the implication of this life-cycle characteristic across the rest of the radiation — where there is no evidence of this life-cycle characteristic — would be misleading. Lara et al. (2010) were also hesitant commending the use of the proposed name “between quotation marks until morphological and/or ultrastructural synapomorphies are defined to diagnose and validate this entire group”. Jones et al. (2011) demonstrate that this key characteristic of Rozella does not seem to extend across the whole group and therefore the name “Rozellida” is not representative of the group as a whole.

• (3) It is important to recognize that our current knowledge of the life stages of the newly discovered Cryptomycota and of Rozella is very incomplete. As Jones et al. (2011) suggest, chitin may be present in the walls of some currently unknown Cryptomycota life-cycle stage(s) and/or present in uncharacterized lineages within Cryptomycota, and even in currently unknown stages in Rozella. It would be premature, therefore, to separate Cryptomycota from the kingdom Fungi on the single character that they do not possess chitin walls (which, as mentioned above is not diagnostic for Fungi).

• (4) Cryptomycota have some strong resemblances to Chytridiomycota (‘chytrids’) in both structure (e.g. flagellar apparatus) and ecology, if not in cell wall chemistry. There is no agreed defining non-molecular characteristic for identifying the boundaries of kingdom Fungi. Therefore, as several other key characteristics are shared by Cryptomycota and some Fungi, the former are most sensibly and parsimoniously considered as belonging to the latter as they form the closest branches on phylogenetic trees (James et al. 2006a, b, Lara et al. 2010, Jones et al. 2011). This stance is entirely consistent with the historical position regarding Rozella: for the last 40 years leading mycologists have classified this genus within Fungi (e.g. Held, 1981; Kirk et al. 2008).

• (5) Cryptomycota (including Rozella) consistently branch with Fungi in all phylogenies so far constructed. However, their position as the primary branch within fungi is much weaker (e.g. James et al. 2006; Jones et al. 2011). Indeed, they could actually occupy a higher branching position within Fungi. If this is the case, their lack of some traditionally diagnostic fungal features such as a chitin cell wall may be the result of secondary losses, which would not preclude them from being considered Fungi. In this case, excluding Cryptomycota from the Fungi could potentially make the rest of fungi paraphyletic — a highly undesirable and not logically sustainable situation. In the absence of a strong morphological argument to exclude this group from the fungal kingdom — we must therefore look to the only available data, which is phylogenetic, and argues that Cryptomycota are most reasonably considered to be within Fungi.

• (6) Consequently, we agree with Lara et al. (2010) that there are sound reasons for considering Rozella (and now we suggest other Cryptomycota) as Fungi. Whether or not Cryptomycota other than Rozella prove to be phagocytotic (which in itself would not be a sufficiently strongly deterministic trait for inclusion in — or exclusion from — Fungi, as some plant lineages and oomycetes have also lost phagotrophy), theirchytrid-like uniflagellate zoospore stage and particularly their phylogenetic position argue most parsimoniously for a fungal affiliation.

• (7) The names used for taxa at the highest ranks, such as phylum, are better not based on names of included genera, but rather on some special characteristic, as is the case with, for example, the phyla Ascomycota and Basidiomycota. In this way the names immediately convey some feature of the taxon. In this case, we highlight the cryptic nature of Cryptomycota in that they were hidden from science until revealed by molecular methods rather than morphological discovery.

In conclusion, we consider the formal validation of the name Cryptomycota to be justified, and commend it for use for this group of organisms as it emphasises the fungal affinity and attributes of the organisms so far known within this group. Even if in some future classification these organisms were placed outside the Fungi, we consider the name should be retained to reflect their nature as fungal analogues.