Shedding light on the polyphyletic behavior of the genus Sterkiella: The importance of ontogenetic and molecular phylogenetic approaches

Present study, investigates a poorly known species of the genus Sterkiella, i.e., S. tricirrata, based on two populations isolated from soil samples collected from the Colfiorito Regional Park, Umbria Region, Italy and from the Silent Valley National Park, India. Both populations showed a highly similar morphology, however different ontogenetic pattern in between. The study confirms the validity of the species S. tricirrata which was considered to be a species within the Sterkiella histriomuscorum complex. The main ontogenetic difference between S. tricirrata and other species of the genus Sterkiella is the different mode of formation of anlagen V and VI of the proter in the former. In the phylogenetic analyses, Sterkiella tricirrata clusters with Sterkiella sinica within the stylonychine oxytrichids, in a clade away from the type species (Sterkiella cavicola) of the genus Sterkiella. The study highlights the importance of ontogenetic as well as molecular data in shedding light on the polyphyletic behavior of the genus Sterkiella. A detailed description of S. tricirrata based on morphology, ontogenesis and molecular phylogenetic methods is presented. Further, the improved diagnosis has been provided for the genus Sterkiella and the poorly known species S. tricirrata.


Introduction
Recent studies, among the hypotrich and spathidiid ciliates, have shown that detailed observations of characters often resolve the discrepancy between the morphological and molecular analyses [1][2][3][4]. This reiterates the need for an integrated approach to investigate in-depth ciliate diversity [5]. The identification of cryptic characters among hypotrich ciliates (e.g. cyst structures, morphology, details on the mode of division) has justified the separation of morphologically similar species reflecting distant relationships in molecular phylogeny [3,[6][7][8][9]. The structure of resting cyst has provided the support for separation of morphologically similar ciliate species, i.e., cyst species [3,4,[7][8][9] [3,10]. Similarly, when ontogenesis within a genus is compared, most congeners have a rather similar pattern with some minor variations [6]. Kumar et al. [2] erected a new genus, i.e., Metasterkiella , for a species isolated from the petroleum contaminated soil, which was morphologically similar to species of the genus Sterkiella Foissner, Blatterer, Berger & Kohmann, 1991. However, this species possesses a variant character in ontogenetic pattern, i.e., it is the only known stylonychid ciliate, thus far, where cirrus V/3 was incorporated during the anlagen formation. The relatedness between genera within the subfamilies Oxytrichinae and Stylonychinae is rather difficult to understand, despite being strongly reflected as monophyletic groups in the phylogenetic analyses [11][12][13][14][15]. A probable explanation could be the insufficient availability and interpretation of morphological and ontogenetic data for the known species. Previous studies have shown that the stylonychid genus Sterkiella is a non-monophyletic assemblage [1,2,16,17]. Sterkiella histriomuscorum, a sibling species complex according to Berger [11], include populations which have not been studied in detail and thus considered to be synonyms under the complex.
In this study, we describe a poorly known species Sterkiella tricirrata (Buitkamp, 1977) Berger, 1999, which is probably a synonym of one of the species within the Sterkiella histriomuscorum complex together with S. terricola, according to the remarks of Berger [11]. The Indian and Italian populations of S. tricirrata were studied and found to be highly similar in morphology. Ontogenetic stages of both populations showed difference in the anlagen formation during the early divisional stages with respect to that of Sterkiella species [11,18,19], and thus suggesting its possible separation at the genus level. Detailed data on the morphometry and ontogenesis for both populations and molecular analyses based on SSU rRNA gene of Italian population is presented. Furthermore, this study highlights the relevance of combining ontogenetic and molecular phylogenetic approaches in identifying species in polyphyletic assemblages such as that represented by the stylonychid genus Sterkiella.

Description of the sampling site and sample processing
Soil samples were collected from the core zone of the Silent Valley National Park, India (11˚08' 40.72"N; 129˚20' 38"E) in January, 2008 and from the plains of the Colfiorito and Plestini uplands, Umbria region, Central Italy (43˚01' 40.72"N; 12˚52' 39.46"E) in July, 2009. Vegetative cells were excysted from resting cysts from two-weeks-dried soil samples (approximately 200 g) by employing the non-flooded Petri dish method [20]. A clonal culture of Sterkiella tricirrata was established for both the populations as described in Kumar et al. [1], i.e., using Pringsheim's medium for culturing and the green alga Chlorogonium elongatum as food source. Observations on the live specimens were made using a microscope with bright-field and differential interference contrast illuminations at a magnification of 100-1000×. The protargol staining method described by Kamra and Sapra [21] was used with some modification to reveal the ciliature. Measurements of impregnated specimens were performed at a magnification of 1000× using an ocular micrometer for Italian population and Leica software IM50 image manager for the Indian population. An Optika microscope camera was employed for photomicrography for the Italian population and Leica camera DFC320 for the Indian. The illustration of the live specimen was prepared using free-hand sketches, while those of impregnated specimens were made with the drawing device. Terminology is according to Berger [11] and Wallengren [22].

DNA extraction, PCR amplification, and sequencing
Unfortunately, we could not perform the DNA extraction for the Indian population. Thus the methods described here, including the phylogenetic analyses of the SSU rRNA gene, exclusively refer to the Italian population. Five cells were collected from a clonal culture with the help of glass micropipettes and washed three times with autoclaved distilled water (same culture was used for live observation and protargol staining to study morphology and ontogenesis). Genomic DNA was extracted using the Norgen DNA Kit (Elettrofor Scientific Instruments, Borsea, Italy), following the manufacturer's instruction [23]. Extracted DNA (5 μl) was dispensed into a PCR tube containing 5 μl of autoclaved distilled water, and amplifications were carried out using high-fidelity Pfx50 DNA polymerase (Invitrogen, Italy) in a total volume of 50 μl with the universal eukaryotic primers Euk A (FW 5'-AACCTGGTTGAT CCTGCCAGT-3') and Euk B (RV 5'-TGATCCTTCTGCAGGTTCACCTAC-30) [24]. Additionally, nested primer pairs Eup 18S (FW 5'-TAG AGG GAC TTT GTG TGC AAC C-3') and Eup 18S (RV 5'-ATC TCC CTG AAA CAC ACG TTG G-3') were used in combination with the universal primers for amplification and sequencing. The PCR program for 18S rDNA amplification included an initial denaturation at 94˚C for 3 min, followed by 35 cycles of 94˚C for 1 min, 55˚C for 45 s and 72˚C for 80 s, with a final extension step at 72˚C for 10 min. After confirmation of the appropriate size, the PCR products were purified using the Nucleospin gel extraction kit (Qiagen) and were then directly sequenced on both strands at StarSEQ GMBH, Germany.

Phylogenetic analyses
For phylogenetic analyses, the SSU rRNA gene sequence of Sterkiella tricirrata was aligned with 55 SSU rRNA gene sequences of hypotrich ciliates from GenBank using the MAFFT software v. 7.047 (choosing the iterative refinement methods Q-INS-I that considers the secondary structure of the SSU rRNA molecules) [25].
Ambiguously aligned regions were identified and excluded from the phylogenetic analyses with GBlocks v.0.91b [26] using parameters optimized for rRNA alignments (minimum length of A block = 5, allowed gap positions = with half), leaving 1,644 unambiguous positions. The final alignment was then used for subsequent phylogenetic analyses after converting the FASTA (.fas) file to NEXUS (.nex) format using the open web-based tool ALTER (Alignment Transformation EnviRonment) [27]. A Bayesian inference (BI) analysis was performed using MrBayes v.3.2.1 [28] and the GTR+I+G model, as selected by the jModel Test v.2.1.3 software [29] under the Akaike Information Criterion corrected (AICc). Markov chain Monte Carlo (MCMC) simulations were run, with two sets of four chains using the default settings, for 1,000,000 generations with trees sampled every 100 generations and discarding the first 25% of the sampled trees as burn-in. The remaining trees were used to generate a consensus tree and to calculate the posterior probabilities (PP) of all branches using the majority-rule consensus approach. The previous alignment was also used to perform a Maximum Likelihood (ML) tree by means of the Molecular Evolutionary Genetic Analysis (MEGA) software, v.5.2.2 [30] using the default parameters and the GTR+I+G model. The reliability of tree topology was assessed by 1,000 bootstrap replicates and was expressed as a percentage. Phylogenetic trees were visualized using the free software package FigTree v1.4 by A. Rambaut at http://tree.bio.ed.ac.uk/ software/figtree/.

Data availability
The newly obtained SSU rRNA gene sequence of Sterkiella tricirrata is available from the Gen-Bank/EMBL databases (accession number: MG805314). Two neotype slides of the Italian population containing the protargol stained neotype specimen and relevant morphostatic specimens have been deposited at the Natural History Museum, London, UK, with registration numbers NHMUK 2014.

Nomenclatural acts
The electronic edition of this article conforms to the requirements of the amended International Code of Zoological Nomenclature, and hence the new names contained herein are available under that Code from the electronic edition of this article. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix "http://zoobank.org/. The LSID for this publication is: urn:lsid:zoobank.org:pub:DB29FEE1-22B6-48CC-9E8D-661AD15BBB06. The electronic edition of this work was published in a journal with an ISSN, and has been archived and is available from the following digital repositories: PubMed Central, LOCKSS.

Description of Sterkiella tricirrata
Morphometric data of the Indian and Italian population of Sterkiella tricirrata highly overlap (Table 1). Thus only a detailed description of the Italian population is provided below; minor differences with the Indian population in some characters include:  Table 1).

Notes on ontogenesis
The ontogenetic stages of Italian and Indian population show a common origin of anlagen II, III, V, and VI for the proter and the opisthe (Figs 2H, 5A-5E, 6A-6C, 7A-7C, 8A-8G and 9A-9K). Difference in the anlagen formation was observed in the Indian population, i.e., a Wshaped formation for the anlagen IV, V, and VI of the proter during the late-early stage, similar to type species of the genus Sterkiella [18,19] (Fig 9E and 9F).
The oral primordium originates close to transverse cirri IV/1 and extends towards the buccal vertex (  The marginal anlagen arise at each of two levels by "within-row" anlagen formation utilizing one or two of the parental cirri at each level. The marginal anlagen elongate deploying four or five parental cirri and differentiate into new marginal rows. The remaining parental marginal cirri are resorbed (Figs 5D, 5E, 6A, 8C-8E, 8G, 8H and 8J).
On the dorsal surface, three anlagen are formed within row from dorsal kineties 1, 2 and 3 at two levels, (one set for the proter and one for the opisthe) (Figs 6B, 6C, 8F, 8G and 9K). The  Nuclear division proceeds in the usual manner, i.e., in mid-dividers the macronuclear nodules fuse to form a single mass which divides twice to produce the typical four nodules in late dividers (Figs 6B, 6C and 8C-8G). The micronuclei undergo mitotic division.

SSU rRNA gene sequence and phylogeny
The SSU rRNA gene sequence of Sterkiella tricirrata Italian population is 1,628 bp in length and has a GC content of 45.15%. It has been deposited in the NCBI database under the accession number MG805314. Phylogenetic trees inferred from the SSU rRNA gene sequences using ML and BI present similar topologies; thus, only the BI tree is shown here membranelles; AZM, adoral zone of membranelles; E, endoral membrane; FC3, frontal cirrus 3; FVC, frontoventral cirri; LM, left marginal row; MA, macronuclear nodules; OP, oral primordium; P, paroral membrane; RM, right marginal row; TC, transverse cirri; V/3, postoral ventral cirrus. Numerals denote cirral anlagen. Scale bars = 40 μm.
The Indian population of Sterkiella tricirrata shows minor differences in size and ciliature with the Italian population as mentioned in the description section. The resting cyst of Indian population appears to be smooth (vs. wrinkled in Italian population); however, additional data on the resting cyst of the Indian population is required to confirm this feature. The original population described by Buitkamp [32] could not be meaningfully compared since most of the morphometric data are lacking. Main differences observed (data from the single image of a protargol stained specimen provided in Buitkamp [32] rely in the number of cirri in left (10 vs. 16 and 20 in Indian and Italian populations, respectively) and right (12 vs. 14 and 16 in Indian and Italian populations, respectively) marginal rows. Further, the original description of S. tricirrata mentioned the presence of five (instead of six recorded in the present study) dorsal kineties. We agree with Berger [11] since the dorsal kinety 6 is rather short it could have been easily missed by Buitkamp [32]. A reinvestigation of the Ivory Coast population will further clarify if it requires separation at the species/subspecies level.

Notes on the ontogenesis of the genus Sterkiella
Berger and Foissner [15] reported that the anlagen V and VI of the opisthe originate de novo in species of the genus Sterkiella. Later, Foissner et al. [19] provided a detailed ontogenetic data on Sterkiella cavicola (Kahl, 1935) Foissner, Blatterer, Berger & Kohmann, 1991, correcting the previous observations that the anlagen V and VI originate by disaggregation of the cirrus V/4. The same pattern is observed also for Sterkiella tricirrata where cirrus V/4 generates anlagen V and VI of the opisthe; however the anterior patches of both the anlagen move anteriorly and later form the proter anlagen V and VI (Table 2). On the contrary, anlagen V and VI of the proter originate from a disaggregation of cirrus IV/3 in S. cavicola [19]. Further, the ontogenetic data of S. tricirrata shows that the anterior portions of anlagen II and III of the opisthe proliferate anterior and merge with the disaggregating cirri II/2 and III/2 respectively to form anlagen II and III of the proter. Recently, Kumar et al. [2] erected a novel genus, Metasterkiella, for a species having similar morphological features as that of Sterkiella histriomuscorum; however, the former not only showed difference in the anlagen formation but also the involvement of cirrus V/3 in anlagen formation, a feature never reported for any stylonychid ciliate. Possibly the involvement of cirrus V/3 during anlagen formation and the semirigid body indicate that the M. koreana might have recently evolved from an Oxytricha-like ancestor. As mentioned above, Sterkiella tricirrata also shows differences with Sterkiella cavicola in the formation of anlagen II, V and VI, i.e., confluent anlagen II and anlagen V and VI of the opisthe give rise to anlagen V and VI of the proter by enlargement and then splitting, though cirrus V/3 remains intact during ontogenesis. The ontogenetic difference between the Indian and Italian populations, i.e., formation of a W-shaped pattern (vs. separate) by the anlagen IV, V, and VI of the proter in late-early divider, indicates that the Indian population may represent a separate subspecies/species if the pattern mentioned is found to be stable in other populations with consensus of molecular data. As of now, we do not perform its separation from the Sterkiella histriomuscorum complex and wait for further data to resolve the phylogenetic status of the species within the complex. However, the different morphogenetic patterns, within the genus Sterkiella, as seen in the present study and Foissner et al. [19] needs to be reflected in the generic characteristics, thus we have provided an improved diagnosis of the genus Sterkiella. The Austrian population of Sterkiella histriomuscorum shows some similarity in anlagen formation with S. tricirrata [18,33]; however, a detailed investigation of its morphogenesis is required for a reliable comparison.

Phylogenetic position of Sterkiella tricirrata
Sterkiella tricirrata clusters with S. sinica (1.00 BI and 99% ML; Fig 9) within the stylonychine oxytrichids, in a clade away from the type species (Sterkiella cavicola) of the genus Sterkiella; we assume that the molecular relatedness of S. tricirrata and S. sinica could be because of similarity in the formation of anlagen. However, a detailed investigation of the ontogenesis of S. sinica is needed to properly compare these genetically similar species. Our phylogenetic  [2] suggested that it probably belongs to the genus Metasterkiella due to highly similar morphology and gene sequence; this interpretation is also supported by our phylogenetic analyses. The classification of Sterkiella nova has been widely debated among classical taxonomists and molecular biologists who established this species as model organism for analyzing various biological phenomena such as epigenetic inheritance, genome rearrangement, somatic differentiation and many others [11,18]. In this regards, Foissner and Berger [18] described S. histriomuscorum and S. nova in great detail from viable genetic systems (via frozen resting cysts) established by molecular biologists. They mentioned that both species are inseparable based on the morphological characters; though based on the differences in molecular sequences of actin I and DNA pol α genes, they proposed them as different species. Considering the complexity of identification it is unclear whether the gene sequences provided by Hewitt et al. [35], which is used in the present study and remains the only sequence available for S. nova, is of same species described by Foissner and Berger [18].
As of now, only differences which seem most suitable to solve the polyphyletic behavior of the genus Sterkiella is the data on the ontogenetic pattern on the ventral and dorsal surface. In Sterkiella cavicola anlagen V and VI of the proter originate from cirrus IV/3 forming W-shaped anlagen [19], whereas it forms from opisthe's anlagen V and VI during the early ontogenetic stages in Sterkiella tricirrata and the genus Metasterkiella. In our phylogenetic tree, Metasterkiella forms a distant clade away from that of Sterkiella tricirrata the involvement of cirrus V/3 (vs. intact) during anlagen formation possibly justifies this distant relationships. Nonetheless, several examples exists like, Parasterkiella thompsoni, Fragmospina depressa, which would have been easily identified as Sterkiella species but separated based on detailed investigations on morphology and cyst structure. Parasterkiella thompsoni shows a different ontogenetic pattern on the dorsal surface and acquires a place distant from Sterkiella species [17], for the species of the genus Fragmospina no gene sequence is available thus far. We believe that addition of related molecular sequences, e.g., Fragmospina, S. histriomuscorum populations, and gene sequences from other loci will further support the monophyly of the genus Sterkiella.

Soil ciliate diversity and species identification: A contribution
Ciliated protists are a highly diverse group of microbial eukaryotes that play a key role in soil microbial food webs by mediating the fluxes of nutrients and energy between different trophic levels [36]. Nevertheless, ciliate diversity in the soil is a still largely neglected research topic and this taxon is significantly less studied than other soil microbial taxa such as bacteria and fungi [37,38]. Since 2009, our group has made a significant contribution to in-depth knowledge about the diversity of soil ciliates across two continents, i.e., Europe (Italy) and Asia (India and South Korea) [1,9,[39][40][41][42][43][44][45]. Numerous faunistic surveys performed in the framework of several projects, allowed us to isolate and describe several novel species and genera, as well as re-describe poorly known or even misidentified species [9,[39][40][41][42][43][44][45]. According to Foissner [46], more than 70-80% of the soil ciliate diversity is still unexplored and a single soil sample can host new species/genera such as in the case of the soil sample collected from the regional Park of Colfiorito described in this study, in which one new and one poorly known species were identified, i.e., Pseudouroleptus plestiensis [45] and Sterkiella tricirrata (present study).
In the end, our sampling effort has allowed us to contribute to strengthen the knowledge about soil ciliate diversity, providing hints about their biogeographic distributions and new distinguishing characters (i.e., cyst morphology, molecular data, ontogenetic processes, arrangement and number of cirri, etc.) among hypotrich ciliates that can be helpful in species identification within problematic (cryptic) species "complexes".

Improved diagnosis
Body semi-rigid. Eighteen or less frontal-ventral-transverse cirri arranged in typical oxytrichid pattern. One right and one left row of marginal cirri. Six dorsal kineties including dorsomarginal rows, kinety 3 with simple fragmentation; caudal cirri present. Undulating membranes in Oxytricha pattern. Opisthe's anlage II may contribute to proter's anlage II. Anlagen V and VI of the proter originate from cirrus IV/3 forming W-shaped anlagen or from anlagen V and VI of the opisthe.

Improved diagnosis (averages are from the populations of India, Italy, and Ivory Coast)
Size about 80 × 40 μm in vivo; body elongate to broadly ellipsoidal. Nuclear apparatus composed of two macronuclear nodules and two micronuclei on average. Invariably, 16 frontalventral-transverse cirri, including three transverse cirri. Right and left marginal rows composed of an average of 15 and 14 cirri, respectively. Adoral zone 37% of body length and composed of an average of 23 membranelles. Three narrowly spaced, inconspicuous caudal cirri. Resting cyst with wrinkled surface. Soil habitat.

Neotype material
Since the original description is incomplete and no type material is available thus according to the Article 75.  Buitkamp [32] isolated Sterkiella tricirrata from the soil collected from the burnt savannah in the Ivory Coast. The Italian population was identified from the 'Molinaccio' site during the summer (dry season), where it was moderately abundant in non-flooded Petri dish culture. For details on the soil physico-chemical parameters and other ciliate species identified in the same soil sample, refer to Bharti et al. [45]. The Indian population was identified from the soil sample collected from the tracts of the tropical rain forest of the Silent Valley National Park, India. For details on other ciliates species identified from the soil samples collected, refer to Kumar et al. [47]. Feeds on bacteria, small amoeba, and flagellates; clonal cultures can be raised as mentioned in materials and methods section.