Parameciom boetschlii, Krenek & Berendonk & Fokin, 2015

Krenek, Sascha, Berendonk, Thomas U. & Fokin, Sergei I., 2015, New Paramecium (Ciliophora, Oligohymenophorea) congeners shape our view on its biodiversity, Organisms Diversity & Evolution (New York, N. Y.) 15 (2), pp. 215-233 : 226-231

New morphospecies P. boetschlii sp. nov.

Among the four herein described and discussed Parameciom spp. , P. boetschlii sp. nov. can be clearly accepted as a new valid morphospecies. The set of morphological features of this ciliate unmistakably separates it from the rest of all known paramecia (Figs. 1–8, 9–16, 17–22, and 38). It has unusual combinations of morphological characters. While the cell size of P. boetschlii sp. nov. is comparable to members of the Parameciom subgenus (almost the same as P. jenningsi , P. schewiakoffi or some P. caodatom stocks ( Fig. 38 ; Fokin et al. 2004; Fokin and Chivilev 2000; 1999)), the general cell shape did not exactly fit the cigar-shaped form typical for Parameciom subgenus representatives. Further, the micronucleus (MI) structure is different from all of the abovementioned species. It is quite large and belongs to the compact morphological type, but has no achromatin cape, as has P. caodatom , and neither looks like the chromosomal type of

0.5

P. jenningsi and P. schewiakoffi . Additionally, the MI chromatin is very fine and homogenous (Figs. 9, 10, 12), resembling the structure of the germinal nucleus of P. potrinom .

According to its cortex peculiarities, P. boetschlii sp. nov. fits better within the Cypriostomom or Helianter subgenera. The cytoproct starts halfway between the buccal overture (BO) and the posterior end of the cell and terminates very close to the cell’ s end (Figs. 3, 4, 8). Each contractile vacuole (CV) has several pores like in P. nephridiatom (Figs. 2, 5, 7, 14). The position of the BO is shifted a bit forward to the anterior end of the cell, but not as much as in the case of P. potrinom , rather similar to P. polycaryom (Figs. 3, 4). The structure of the collecting canals-free contractile vacuoles (CV) (Figs. 15, 16) fits very well with both of the Helianter representatives, P. doboscqoi and P. potrinom . Further peculiarities could be found within the nuclear reorganization process of P. boetschlii sp. nov.; neither a crescent MI stage nor a clear parachote stage (for MI progamic phase) could be detected, while only a small number of old macronucleus (MA) fragments are unique as well ( Fokin et al. 2001).

Multivariate morphometric analysis affiliated P. boetschlii sp. nov. with the Cypriostomom subgenus ( Fig. 35a ), while in the MDS analysis its position is close to the Helianter representative P. doboscqoi ( Fig. 35b ). However, from previous analyses, we know that cell length is a very weighty characteristic ( Fokin and Chivilev 2000; 1999). This may support P. boetschlii sp. nov. being associated with P. woodroffi (almost the same cell size) from the Cypriostomom subgenus, but not with the Helianter representatives, which are much small- er (cf. Fig. 38 ).

Molecular analyses using two different markers, 18SrRNA and mtCOI gene sequences, confirmed that P. boetschlii sp. nov. is a member of the Helianter subgenus, either as a sister species of P. potrinom (18S-rDNA, Fig. 36 ) or at the very basal position of the genus with a close relationship

P. nephridiatom , i P. borsaria, j P. calkinsi, k P. doboscqoi , l P. potrinom , m P. polycaryom . Bar 100 μm. © 2010/2011 by Protozoological Society affiliated with Russian Academy of Sciences to P. potrinom and P. doboscqoi strains (COI, Fig. 27). However, as shown in our tree analyses, neither the 18SrDNA nor the mtCOI sequence data can sufficiently resolve the relationships between the different Parameciom subgenera. This is indicated by partially low support values for some basal branch nodes, which are prone to changes due to the inclusion of new sequences. Here, the addition of P. boetschlii sp. nov. into the COI data set placed the Helianter subgenus at a quite basal position within the genus, while Chloroparameciom, previously been sister to Helianter or at the base (see, e.g., Boscaro et al. 2012 or Tarcz et al. 2013), now represents the sister group to subgenus Parameciom . Accordingly, multi-gene analyses with concatenated sequence data can be an effective method to improve phylogenetic accuracy, and multi-gene trees are more reliable than single-gene trees ( Yi et al. 2014). However, due to often insufficient sampling and/or the use of 18S-rDNA as the gene marker in ciliate phylogeny, complete datasets of potentially more appropriate marker genes are rare and often do not allow for a proper tree reconstruction. In the era of nextgeneration sequencing (NGS) technologies now available (even for ciliatologists) at a reasonable price, future studies in this field can hopefully more often take advantage of NGS to unravel phylogenetic ambiguities.

Cryptic diversity in Parameciom

For describing a new morphospecies, it should be sufficient to have one obvious morphological character that differs from relative ciliates ( Foissner et al. 1999). In herein investigated additional three Parameciom spp. (besides P. boetschlii sp. nov.), we could not identify such distinct morphological features. Nevertheless, since all of these three newly identified species could be clearly separated from the rest of all known paramecia by using morphometric and molecular analysis, we currently prefer to refer them as cryptic Parameciom species. Future studies on additional eco-physiological parameters and/or phylogeographical patterns, as well as new morphological studies or techniques, may be able to reveal presently unknown diagnostic structural differences and allow rapid and practical diagnosis of their taxonomic statuses.

This cryptic status should therefore only be provisional, representing a necessary step to establishing formal scientific names for these newly discovered and described species ( de Leon and Nadler 2010; Jörger and Schrödl 2013). While there is some debate on the general practice on how to describe and delimit new cryptic species (e.g., reviewed in Bickford et al. 2007), there is no official requirement by the International Code of Zoological Nomenclatore (ICZN) on how to name these species in order to indicate their provisional cryptic status and to distinguish them from valid biological species. Specifically, if DNA sequence data serve as the primary character set for species delineation, the Linnaean system is under question ( Tautz et al. 2003), because there is no common practice on how to include such molecular data in formal species descriptions ( Goldstein and DeSalle 2011). While some authors argue for a formal scientific naming of cryptic species to avoid merging morphologically undistinguishable, but genetically divergent lineages (e.g., de Leon and Nadler 2010; Kadereit et al. 2012), others prefer to use clade numbers when studying cryptic species (e.g., Leliaert et al. 2009; Thomas et al. 2014; Skaloud and Rindi 2013). Here, we propose the systematic term B Eocandidatos^ as a component of the taxonomic name to make a distinction between valid biological species or taxonomic units and the provisional cryptic species status. We have chosen this fictional term referring to the provisional status Candidatos (L. candidatos, a candidate) used in prokaryotic nomenclature for formally described bacteria that are lacking characteristics required for a valid description according to the International Code of Nomenclatore of Bacteria ( Murray and Schleifer 1994; Murray and Stackebrandt 1995). The term was altered by prefixing Eo to indicate that the newly described candidate species represents a eukaryote. Since the International Code of Nomenclatore of Algae, Fongi, and Plants (ICN) also do not provide an official requirement on how to treat a cryptic species status, we recommend the use of the term Eocandidatos when describing any new cryptic eukaryotic species. In accordance with the prokaryotic term Candidatos, we propose the term Eocandidatos preceding a species specific epithet consisting of the genus name followed by a single specific epithet according to ICN or ICZN rules. Further, we suggest the term Eocandidatos to be italicized, while the succeeding species name should be in roman type and the entire name in quotation, e.g., B Eocandidatos Paramecium germanicum^. This would allow an instant differentiation of cryptic species from valid species. This indefinite rank Eocandidatos might be established as an Appendix to the nomenclatural codes of ICN and ICZN, according to the ICSP policy for the nomenclature of bacteria.

However, caution is called for when describing new cryptic species from DNA-based characters as long as genetic information of closely related species is not available. Therefore, taxonomists still must carefully check the type material of all known morphologically related species and may eventually compile the missing sequence information to avoid the establishment of duplicate species under separate names. Thorough genetic information of all species in one genus, including data on intraspecific variability, is therefore necessary to unveil the degree of cryptic diversity and finally declare species valid. On the other hand, there is controversial discussion among protistologists as to whether observed DNA variation only reflects neutral mutation accumulation and that the phenotype therefore comprises the only proper characters by which to define valid protist species ( Fenchel and Finlay 2006). Therefore, we would like to encourage discussion not only among protistologists about a change in current taxonomic practice. Granting DNA-based characters a similar weight as morphological features, at least in these instances in which molecular data are informative enough to justify the absence of obvious morphological differences, could not only lead to a more practical and straightforward integrative taxonomy, but also to a better understanding of species diversity and richness. Appropriately, this procedure then certainly needs a common means to present and indicate the diagnostic molecular markers, which will in turn need further in-depth discussion among scientist to ensure stability and traceability of future taxonomy ( Jörger and Schrödl 2013).

New cryptic Parameciom species

As stated above, the further three herein investigated paramecia are rather difficult to discriminate from similar members of the genus by morphology. Therefore, they currently cannot be accepted as clearly separated morphospecies, due to common formal requirements not fulfilled and are thus treated as cryptic species. More precise, B Eocandidatos P. hungarianum^ actually has no morphological features that deviate significantly from those of P. polycaryom ( Table 3; Fokin and Chivilev 1999), a member of the Cypriostomom subgenus. Classification and ordination of the morphometric data of B Eocandidatos P. hungarianum^ also indicated similarity to P. polycaryom . However, they are not identical, as shown by rather large Euclidian distances and non-overlapping coordinates in UPGMA and MDS analyses, respectively ( Figs. 35a, b ). Further, molecular data clearly separated B Eocandidatos P. hungarianum^ from P. polycaryom , showing a closer relationship to other Cypriostomom members, namely P. nephridiatom , P. woodroffi and/or P. calkinsi ( Figs. 36 , 37).

A similar situation exists for comparisons between B Eocandidatos P. brazilianum^ and its close relative P. moltimicronocleatom . On the one hand, they are quite similar in general morphology, but can be separated by morphometric analysis ( Figs. 35a, b ). On the other hand, molecular data based on 18S-rDNA sequences clearly assigned B Eocandidatos P. brazilianum^ to P. moltimicronocleatom within the Parameciom subgenus ( Fig. 36 ). Within this tree reconstruction, P. moltimicronocleatom , however, appears as two well supported, separate clades with an averaged sequence similarity of only 98.0 %. At the same time, sequence similarity between P. primaorelia and P. tetraorelia, for example, which are valid biological species, is 99.6 % ( Strüder-Kypke et al. 2000). This suggests that the second clade may represent a distinct biological entity, which, besides the herein described B Eocandidatos P. brazilianum^, consists of several P. moltimicronocleatom strains and one unspecified Parameciom species ( Parameciom sp. para200) that have been all isolated from waste water in Pakistan or Italy, respectively. Moreover, the size and number of the micronuclei as well as the presence of fragmented macronuclei in B Eocandidatos P. brazilianum^ deviated from the classical P. moltimicronocleatom phenotype ( Table 3). Unfortunately, no other sequence data of any strain of these clade representatives are available nor could we produce own data due to lack of sufficient DNA material or a viable cell culture. Hence, we currently have no additional molecular evidence to separate these two clades as genetic and potentially valid biologically distinct species.

B Eocandidatos P. germanicum^, according to its general constitution and size, is rather similar to P. moltimicronocleatom too, but the former has the compact MI morphological type instead of the vesicolar generative nuclei of the latter (Figs. 25, 26). The MI morphological type, hence, might be the one morphological character that allows differentiation between species, which is unfortunately not an easy or very practical character to perceive, since it requires excellent staining preparations and practice. Regarding the MI morphological type, B Eocandidatos P. germanicum^ could be weakly related to P. caodatom , as shown by the UPGMA analysis ( Fig. 35a ). Based on its molecular characteristics, B Eocandidatos P. germanicum^ is positioned inconsistently within the genus according to the genetic marker used ( Figs. 36 , 37). In any case, this species does not group with P. moltimicronocleatom , but rather with P. caodatom (based on 18S-rDNA), or it is located basally relative to the Parameciom subgenus (based on COI). As stated above, further multi-gene analyses may be able to help clarify its phylogenetic position too.

Current and future perspective of Parameciom biodiversity

Parameciom species have been investigated for over 250 years. Hence, B it is very unlikely that in Europe or North America taxonomists would have missed such obvious ciliates as paramecia during checking of water samples^ ( Fokin et al. 2004). The redescription of P. chlorelligerom ( Kreutz et al. 2012) and the discovery of the new morphospecies P. boetschlii sp. nov. herein described prove the opposite. Still, correct identification and not simply finding paramecia in water samples is crucial, and protistologists (or hydrobiologists of any kind) must have an adequate taxonomic knowledge to accomplish this task. In fact, the classical taxonomy based on traditional morphological examinations is declining, which means that very few people can do this kind of investigation professionally indeed. The redescriptions of P. doboscqoi and P. nephridiatom ( Gelei 1938; 1925; Chatton and Brachon 1933; Shi et al. 1997; Fokin et al. 1999a, b) are excellent examples of this decline. Both species are widely distributed in brackish and (sometimes) freshwater habitats almost ubiquitously in the Northern Hemisphere and particularly in Europe ( Fokin et al. 1999a, b). However, for decades, they were listed as doubtful species and nobody recorded these ciliates in nature ( Wichterman 1953, 1986), as they were either confused with P. calkinsi and P. woodroffi or just ignored as strange paramecia ( Fokin et al. 1999a, b).

While traditional taxonomists may unfortunately disappear, we have faith that protist taxonomy is not dying but evolving in the way it is done. Molecular approaches will eventually make a major contribution to modern taxonomy, but future taxonomists will still need classical methods for verification and validation of the most reliable morphological characters needed for integrative taxonomy. For example, P. boetschlii sp. nov. at first glance could be treated as P. caodatom , which may be a reason it is not checked precisely. Yet, some morphological peculiarities and genetic data convinced us to investigate it in more detail. The same holds true for the other herein studied Parameciom spp.

On the other hand, molecular tools alone have the potential to unveil morphologically hidden diversity. As shown in our tree analyses based on 18S-rRNA and COI gene sequences ( Figs. 36 , 37), there seems to be a higher current stock diversity in Parameciom than expected. As mentioned above, the two separated P. moltimicronocleatom clades (18S-rDNA, Fig. 36 , cf. Buosi et al. 2014) may represent distinct species with B Eocandidatos P. brazilianum^ as potential type species of the sister clade to P. moltimicronocleatom . Moreover, as shown in Fig. 36 , there also exists another 18S-rDNA-based B P. caodatom^ clade sister to B Eocandidatos P. germanicum^ and P. caodatom (strain Isn4), which consists of strains isolated from waste water in Pakistan. These strains (typed with FT) exhibit a rather large genetic distance from all other thus far described and investigated P. caodatom species ( Table S1). Therefore, we consider it to represent a new Parameciom congener as well. Further studies on morphology and genetics are needed to clarify its taxonomic status and phylogeographical patterns.

Within our comprehensive analyses based on current Parameciom COI gene sequences available, we may have revealed some potential misidentifications (e.g., P. calkinsi BM1- 11 and BOB130-7; P. woodroffi PrS; P. moltimicronocleatom Kr 113-3, V105-2, TRB 101–1 and TRB 101–4; and P. chlorelligerom GC ; cf. Fig. 37), which should be carefully checked again, if the respective strains are still in stock. In particular, the sequence of P. chlorelligerom strain GC clustered within the outgroup, thereby compromising the monophyly of genus Parameciom . As already stated by Boscaro et al. (2012), we also urgently call for the use of comprehensive data sets when analyzing and publishing new sequence data of single or new species. Further, the P. moltimicronocleatom strains V105-2, TRB 101–1 and TRB 101–4 investigated by Tarcz et al. (2012) and originating from the Vologda or Irkutsk regions of Russia may also represent new cryptic or even true biological species. These specimens form a separate cluster with a comparatively large genetic distance (cf. Fig. 37, Table S2) to other Parameciom subgenus representatives (see also Boscaro et al. 2012). Here, further studies are needed as well to elucidate phylogenetic and taxonomic statuses. The morphological type of micronuclei, for example, may provide clues on the relation to B Eocandidatos P. germanicum^ as the most basal congener to these specimens.

Our molecular analyses further suggest that P. caodatom may represent a complex of cryptic species too, comprising potentially distinct genetic entities with different geographic ranges and environmental demands. As shown in Fig. 37, there are five different COI haplogroups that do not form one well-supported clade, but rather show uncertain affiliations with close relationships. While haplogroup PcCOI_a mainly consists of European strains, its sister clade comprises representatives from different continents. Another cluster includes strains from tropical Indonesia (INL-1, INP-3, INK-1) that show an altered eco-physiology thereby indicating local adaptation and potential geographic constraints ( Krenek et al. 2012). While former studies using RAPD and ARDRA fingerprinting methods have rejected a potential sibling species concept in P. caodatom ( Stoeck et al. 2000) , further and more comprehensive analyses on the different varieties and subpopulations will shed light on these clades relationships and this potential cryptic species complex.

Apparently, the species richness of genus Parameciom seems to be higher than our current knowledge suggests and could be further increased and revised by more careful investigations that use a combination of classical taxonomic and modern molecular approaches. Moreover, future studies should not only investigate remote and exotic territories. Considering less frequently sampled habitats in the Old World will conceivably increase our knowledge as well. Such studies can generate more accurate species lists and species richness estimations, as our present study indicates.

Figures

Fig. 38 Schematical images of major Parameciom morphospecies made according to its morphometric data (reprint of Fig. 1 from Fokin 2010/11). a P. moltimicronocleatom, b P. caodatom, c P. jenningsi, d P. schewiakoffi, e P. boetschlii sp. nov., f P. woodroffi, g P. aorelia, h

Fig. 35 Dendrogram (a) for hierarchical clustering (UPGMA) and topogram (b) for non-metric multidimensional scaling (MDS) of morphometric and morphobiological characteristics of 16 Parameciom species. AU P. aorelia, BU P. borsaria, CA P. caodatom, CL P. calkinsi, DU P. doboscqoi, JE P. jenningsi, MU P. moltimicronocleatom, NR P. nephridiatom, PB BEocandidatos P. brazilianum^, PG BEocandidatos P. germanicum^, PH BEocandidatos P. hungarianum^, PN P. boetschlii sp. nov., PO P. polycaryom, PU P. potrinom, SH P. schewiakoffi, WO P. woodroffi

Fig. 36 SSU rDNA tree topology. Phylogenetic reconstruction of the genus Parameciom based on 18S-rDNA sequences inferred by Bayesian Inference analysis. The alignment contained 49 taxa and 1640 sites including gaps. Sequences of new or cryptic species characterized within this study are shown in boldface. Sequences of other Peniculida served as outgroup. Nombers at nodes (occasionally indicated by an arrow) represent support values for the Bayesian Inference and Maximum Likelihood