Cymbopleuranatellia – a new species from Transbaikal area (Russia, Siberia) described on the basis of molecular and morphological investigation

Abstract A new cymbelloid diatom species from the genus Cymbopleura (Krammer) Krammer is described on the basis of molecular and morphological investigations. Cymbopleuranatellia Glushchenko, Kulikovskiy & Kociolek, sp. nov. is, on the basis of results with molecular data, close to C.naviculiformis (Auerswald ex Heiberg) Krammer. The two species differ both by molecular distance and morphological features. Morphologically, C.natelliasp. nov. is compared with several other species in the genus. This work is a pioneer investigation of cymbelloid taxa using molecular tool from Transbaikal area.


Introduction
Cymbelloid diatoms have unusually high diversity in ancient Lake Baikal but we almost have no information about them in waterbodies surrounding this lake. Historically, cymbelloid diatoms have been referred to taxa with dorsiventral symmetry (Round et al. 1990; Kulikovskiy et al. 2016). However, results from molecular investigations and a better un-derstanding of morphological features have shown that genera with naviculoid symmetry also belong to this monophyletic group. New diatom genera from Lake Baikal, such as Ochigma Kulikovskiy, Lange-Bertalot & Metzeltin (Kulikovskiy et al. 2012), Paraplaconeis Kulikovskiy, Lange-Bertalot & Metzeltin (Kulikovskiy et al. 2012) and Khursevichia Kulikovskiy, Lange-Bertalot & Metzeltin (Kulikovskiy et al. 2012) are good examples of naviculoid diatoms that are more closely related to cymbelloid taxa (Kulikovskiy et al. 2012(Kulikovskiy et al. , 2016. Lake Baikal includes many "typical" dorsiventral genera and species that were comprehensively investigated by Kulikovskiy et al. (2012). In that work we described many new species from the genera Cymbella C.A. Agardh, Encyonema Kützing and Cymbopleura (Krammer) Krammer (Kulikovskiy et al. 2012) based on valve morphology. However, these new taxa were described without molecular investigation. Our understanding of the diversity of algae in ancient lake systems will also be enhanced by using molecular methods.
Members of the genus Cymbopleura were previously considered to be members of the genus Cymbella. Cymbopleura was separated from Cymbella based on the features of absence of apical pore fields and central raphe ends turned towards the ventral side of valve (Krammer 1982;Kulikovskiy et al. 2012). The new genus Karthickia Kociolek, Glushchenko & Kulikovskiy is very similar to Cymbopleura except that it differs by the presence of a single external stigma opening that has two elongated slit-like openings internally, residing within two longer central striae (Glushchenko et al. 2019). This new genus was described from Southeast Asia. Another new genus that is similar to Cymbopleura is Yasnitskya Rodionova & Pomazkina (Pomazkina and Rodionova 2014). This genus is characterized by interesting internal areolar morphology. Yasnitskya was also described on the basis morphological investigation only.
The aim of the present report is to begin a series of studies using molecular methods to study the identity and systematic placement of cymbelloid diatoms from Lake Baikal and surrounding watrebodies, with the description of new species from the genus Cymbopleura. We plan that this work as our pioneer investigation of dorsiventral species from this area and that this work will be continued in future with new molecular data for other known taxa in this lineage.

Culturing
A subsample of each collection was added to WC liquid medium (Guillard and Lorenzen 1972). A monoclonal strain was established by micropipetting a single cell under an inverted microscope. Non-axenic unialgal cultures were maintained in WC liquid medium at 22-25 °C in a growth chamber with a 12:12 h light:dark photoperiod. The strain investigated here was designated B209.

Preparation of slides and microscope investigation
The culture was treated with 10% hydrochloric acid to remove carbonates and washed several times with deionized water for 12 h. Afterwards, the sample was boiled in concentrated hydrogen peroxide (≈ 37%) to remove organic matter. It was washed again with deionized water four times at 12 h intervals. After decanting and filling with deionized water up to 100 ml, the suspension was pipetted onto coverslips and left to dry at room temperature. Permanent diatom preparations were mounted in Naphrax. Light microscopic (LM) observations were performed with a Zeiss Axio Scope A1 microscope equipped with an oil immersion objective (× 100, n.a. 1.4, differential interference contrast [DIC]) and Axiocam ERc 5s camera (Zeiss). Valve ultrastructure was examined by means of scanning electron microscopes JEOL JSM-6510LV (JEOL Ltd., Japan) operating at 15 kV and 8 mm of working distance (IBIW, Institute for Biology of Inland Waters RAS, Borok, Russia). For scanning electron microscopy (SEM), part of the suspensions was fixed on aluminum stubs after air-drying. The stubs were sputter-coated with 50 nm of Au by means of a Eiko IB 3 (Eiko Engineering, Japan).
The cleaned material, sample and slide are deposited in the collection of MHA, Main Botanical Garden Russian Academy of Science, Moscow, Russia. The type slide was designated B209.

Molecular investigation
Total DNA from the studied strain was extracted using Chelex 100 Chelating Resin, molecular biology grade (Bio-Rad Laboratories, USA), according to the manufacturer's protocol 2.2. Partial 18S rDNA (435 bp, including the highly variable V4 region) gene was amplified using primers D512for and D978rev from Zimmermann et al. (2011).
Amplification was carried out using premade polymerase chain reaction (PCR) mastermixes (ScreenMix by Evrogen, Russia). Amplification conditions for 18S rDNA gene were as follows: initial denaturation for 5 min at 95 °C followed by 35 cycles of 30 s denaturation at 94 °C, 30 s annealing at 52 °C, and 50 s extension at 72 °C, with the final extension for 10 min at 72 °C. PCR products were visualized by horizontal electrophoresis in 1.0% agarose gel stained with SYBRTM Safe (Life Technologies, USA). The products were purified with a mixture of FastAP, 10×FastAP Buffer, Exonuclease I (Thermo Fisher Scientific, USA), and water. The sequencing was performed using a Genetic Analyzer 3500 instrument (Applied Biosystems, USA).
Editing and assembling of the consensus sequences were carried out by processing the direct and reverse chromatograms in Ridom TraceEdit (ver. 1.1.0) and Mega7 software (Kumar et al. 2016). The reads were included in the alignments along with corresponding sequences of 74 diatom species downloaded from GenBank (taxa names and Accession Numbers are given in Fig. 5). Five diatom species from Rhopalodiaceae were chosen as the outgroups.
The nucleotide sequences of the 18S rDNA gene were aligned separately using the Mafft v7 software and the E-INS-i model (Katoh and Toh 2010). The final alignments were then carried out: unpaired sites were visually determined and removed from the beginning and the end of the resulting matrices. The resulting alignments had lengths of 439 characters.
The data set was analyzed using the Bayesian inference (BI) method implemented in Beast ver. 1.10.1 software (Drummond and Rambaut 2007) to construct phylogeny. For the alignment partition the most appropriate substitution model, shape parameter α and a proportion of invariable sites (pinvar) were estimated using the Bayesian information criterion (BIC) as implemented in jModelTest 2.1.10 (Darriba et al. 2012). This BIC-based model selection procedure TrN+I+G model, α = 0.5130 and pinvar = 0.4620 for 18S rDNA gene. We used the HKY model of nucleotide substitution instead of TrN given that it was the best matching model available for BI. A Yule process tree prior was used as a speciation model. The analysis ran for 5 million generations with chain sampling every 1000 generations. The parameters-estimated convergence, effective sample size (ESS) and burn-in period were checked using the Tracer ver. 1.7.1 software (Drummond and Rambaut 2007). The initial 25% of the trees were removed, the rest were retained to reconstruct a final phylogeny. The phylogenetic tree and posterior probabilities of its branching were obtained on the basis of the remaining trees, having stable estimates of the parameter models of nucleotide substitutions and likelihood. The Maximum Likelihood (ML) analysis was performed using RAxML program (Stamatakis et al. 2008). The nonparametric bootstrap analysis with 1000 replicas was used. The programs FigTree ver. 1.4.4 and Adobe Photoshop CC (19.0) were used for viewing and editing of the trees. Description. LM (Figs 1 and 2). Cells solitary. A single chloroplast is present per cell. The chloroplast has two lobes, that underlie the valve face, and they are connected by a wide isthmus (Fig. 1). Valves subelliptical, dorsiventral with moderately convex dorsal margin and slightly convex ventral margin, often almost straight near the valve centre. Ends are bluntly rostrate. Length 17.6-23.5 μm (20.5 ± 1.6; n = 30), width 8.7-9.5 μm (9.1 ± 0.2; n = 30). Length-to-width ratio 1.97-2.52. Central area more or less pronounced, rounded, 1/3 to the valve breadth. Axial area narrow, more often weakly expands to central area, less often -almost not expanded beyond the median line of the valve. Raphe filiform near the center, lateral towards the apices, with proximal raphe ends deflected ventrally, tipped with weakly inflated pores. Striae finely punctate, radiate, becoming subparallel and condensing towards to the ends, 13-15 in 10 μm (14 ± 0.7; n = 30) at the central part, 18-20 in 10 μm (19 ± 0.7; n = 30) near the ends. Areolae difficult to resolve in LM (Fig. 2). SEM, external view (Fig. 3). Valve face is flat. Central area formed by shortened striae. Striae uniseriate, extending to valve face towards mantle on both the dorsal and ventral margins, composed of very small, elongate, lineolate areolae, 30-35 in 10 μm (32.5 ± 1.3; n = 30). Proximal raphe endings are slightly expanded, unilaterally curved. Distal raphe fissures unilaterally curved opposite the proximal raphe ends, extending onto the valve mantle.

Cymbopleura natellia
SEM, internal view (Fig. 4). The raphe slits close to proximal endings are arcuate. Proximal raphe endings weakly deflected to the dorsal margin. Distal raphe ends terminated small helictoglossae. Areolae arranged in a series with narrow vimines, compared to the wide interstriae (virgae), occlusions absent but tectullae present.  Etymology. New species is dedicated to our friend Natella Otarovna Gabuadze. Distribution. As yet known only from the type locality. Sequence data. Partial 18S rDNA gene sequence comprising V4 domain sequence (GenBank accession number MZ503642).

Molecular investigation
The phylogenetic analyses were conducted using a single gene dataset (Fig. 5). Sequences of Cymbopleura species as well as Cymbella, Gomphonema, Placoneis and another pennate diatom species were included in the phylogenetic analyses. According to the maximum likelihood (ML) and Bayesian Inference (BI) phylogenetic analyses, Cymbopleura natellia sp. nov. (the strain B209) appeared most closely related to the strain 22 vi092D of Cymbopleura naviculiformis (Auerswald ex Heiberg) Krammer 2003 with high statistical support (ML 94; BI 0.99) and other Cymbopleura strains (Fig. 5). We should point out, however, that Cymbopleura is non-monophyletic in these analyses, with two strains of Cymbopleura inaequalis and an unidentified Cymbopleura strain (D213 001 MK300893) having a closer relationship with some Cymbella species than Cymbopleura strains.

Discussion
According to the data of molecular analysis, the species Cymbopleura natellia sp. nov. is most closely related to C. naviculiformis. According to morphological data, the new species is distinguished by bluntly rostrate, not cuneate valve ends, whereas C. naviculiformis has rostrate to subcapitate, narrow-protracted valve ends. The central area of the new species is generally smaller (1/3 of the valve width) than the central area of C. naviculiformis (1/2-2/3 of the valve width). The new species differs from C. naviculiformis in a shorter valve length (17.6-23.5 μm in the new species versus   26-50 μm in C. naviculiformis), as well as, in general, a smaller valve width (8.7-9.5 μm in the new species versus 9.0-13.0 μm in C. naviculiformis). The density of striae overlaps in species (13-15 in 10 μm in the central part, 18-20 towards the ends in the new species versus 10-15 in 10 μm in the central part, 16-19 in 10 μm towards the ends in C. naviculiformis) and areolae (30-35 in 10 μm in the new species versus 27-35 in C. naviculiformis) ( Table 1).
Cymbopleura natellia sp. nov. is morphologically similar to Cymbopleura designata (Krammer) Bahls (in Bahls 2019). The two species are close in terms of density of striae (13-15 in the central part, 18-20 at the ends in the new species versus 13-15 in the central part, 18-20 at the ends in C. designata). The main difference between these two species is the shape of valves. The new species has subelliptical valve outlines while C. designata has a wide lanceolate shape of the valves. The central area of the new species is more differentiated than that of C. designata, the central area of which is slightly wider than the axial area. The ends of C. designata are apiculate, extending laterally from valve ends. There is also a difference in areolae density (30-35 in 10 μm in the new species versus 26-30 in 10 μm in C. designata).
Cymbopleura natellia sp. nov. is also morphologically similar to C. frequens Krammer 2003. These two species have a similar number of areolae (30-35 in 10 μm in the new species and 30-36 in 10 μm in C. frequens). At the same time, the species differ in the outline of the valves, with the valves of C. frequens being almost naviculoid or slightly dorsiventral, while in C. natellia sp. nov. the valves are subelliptic and evidently dorsiventral. The ends of C. frequens are protracted, apiculate to rostrate.
Cymbopleura hercynica (A.W.F. Schmidt) Krammer 2003 is another species that is morphologically similar to the new species, in terms of stria density (12-15 in the central part, up to 20 in 10 μm toward to the ends in C. hercynica versus 13-15 in the central part, 18-20 in 10 μm to the ends in the new species) and density of areolae (32-36 in 10 μm in C. hercynica versus 30-35 in 10 μm in the new species). Also, the two species are generally similar in valve shape. The species differ in the shape of the valve ends, in the new species the ends are wide, in C. hercynica they are apiculate to subrostrate protracted. The width of the central area of C. hercynica reaches 1/2 of the total valve width; in the new species, this ratio does not exceed 1/3. It should be noted that the new species was isolated from the Zagza River, where the water conductivity is low, while C. hercynica prefers brackish-water habitats or reservoirs with increased salinity (Krammer 2003;Li et al. 2007).