Anteholosticha sigmoidea ( Foissner, 1982 ) Berger, 2003

Kim, Kang-San, Ji, Su-Jung, Kim, Sanghee & Min, Gi-Sik, 2021, Antarctic population of Anteholosticha sigmoidea (Foissner, 1982) Berger, 2003 (Ciliophora: Urostylidae) with notes on its phylogenetic position, Zootaxa 4942 (2), pp. 290-300 : 291-298

Anteholosticha sigmoidea ( Foissner, 1982) Berger, 2003 ( Figs 1 , 2 ; Table 1)

Voucher slides. Three slides (NIBRPR0000110238–NIBRPR0000110240) with protargol-stained specimens were deposited in the National Institute of Biological Resources (NIBR), South Korea.

Gene sequence. The 18S rRNA gene sequence of the present Antarctic population was deposited in the Gen-Bank under the accession number MN864520 .

Morphological description of the Antarctic population. Size in vivo 90–150 × 25–35 μm, 67–95 × 9–19 μm after protargol preparations ( Figs. 1A, B ; 2A, H ). Body outline usually elongated elliptical, left and right cell margins slightly convex or almost straight in vivo ( Figs. 1A , 2A ). Cell colorless to grayish at low magnification. Contractile vacuole left of mid-body, about 10 μm in diameter during diastole, with collecting canals ( Figs. 1A ; 2 B–E). Colorless cortical granules roughly spherical, 0.5–1.0 μm in diameter, distributed irregularly on both the ventral and dorsal surface, stain with protargol ( Figs. 1E, F ; 2G, L, M ). Cytoplasm colorless, containing many food vacuoles and lipid droplets (ca. 1–4 μm across). Feeds on algae and bacteria. Macronuclear nodules size 4.0–10.0 × 1.5–4.0 μm in protargol preparations, 13–18 in number, arranged in left body portion. Two to five micronuclei sized 1.5–4.0 × 1.0–3.0 μm protargol impregnation ( Figs. 1A, D ; 2H, K ).

Adoral zone approximately 26.5% of cell length in protargol preparations, the base of the largest membranelles about 8 μm long, composed of 22–28 membranelles. Paroral and endoral membranes crossing at the level of buccal cirrus ( Figs. 1B ; 2H, J ). All cirri relatively fine, mostly 11–15 μm long in vivo. Three frontal cirri arranged in the oblique row near the anterior body end; two frontoterminal cirri behind the rightmost frontal cirrus; buccal cirrus near the middle of the paroral membrane; two pretransverse cirri arranged almost at the same level; midventral complex composed of 17–29 midventral cirri, arranged in a zigzag pattern, commencing near the level of the posterior frontoterminal cirrus and terminating around the mid-body; four to eight transverse cirri protruding beyond the posterior body end. One left (25–40 cirri) and one right (25–37 cirri) marginal row, both marginal rows non-confluent posteriorly ( Figs. 1A, B ; 2H, J ).

Invariably four bipolar dorsal kineties; cilia about 3 μm in vivo; dorsal kineties 1–4 composed of 10–17, 13–20, 12–19, and 14–19 bristles, respectively; caudal cirri lacking.

Morphological comparison with other A. sigmoidea populations. Foissner (1982) described three populations of A. sigmoidea that were highly similar, indicating that they belong to a single species ( Berger 2006). However, another population of A. sigmoidea observed later differs from the previous three populations by the length of the midventral complex (52.5% vs. 64.7% of the cell length; measured from illustration), number of adoral membranelles (25–28 vs. 16–24), and number of dorsal bristles in kinety 1 (16 bristles vs. nine bristles) ( Foissner 1982; 1984). Although they were considered to be local variations ( Berger 2006; 1984), the Antarctic population was compared with both groups.

Group I differs from the Antarctic population by the presence/absence of collecting canals of the contractile vacuole (absence vs. presence), and the number of macronuclear nodules (sic to 12 vs. 13–19) ( Foissner 1982). The Antarctic population differs from group II by the number of macronuclear nodules (13–19 vs. five to nine), the arrangement of cortical granules (distributed irregularly vs. in longitudinal rows), length of the midventral complex (64.7% vs. 53.8% of cell length; measured from illustration), and the position of the posteriormost transverse cirrus (90.8% vs. 96.9% of cell length; measured from the illustration) ( Foissner 1984). The Antarctic population has more than three differences from each group of A. sigmoidea . A comparison with both types showed that the variability of the Antarctica population falls within the range of A. sigmoidea except for the number of macronuclear nodules ( Table 1, 2).

Morphological comparison with congeners. The Antarctic population of A. sigmoidea is morphologically similar to four congeners, A. australis ( Blatterer and Foissner, 1988) Berger 2003 , A. distyla ( Buitkamp, 1977) Berger, 2003 , A. rectangula Jung et al., 2016 , and A. verrucosa ( Foissner, 2000) Berger, 2008 , in body shape (elongate), nuclear apparatus (arranged along the left cell margin), presence of cortical granules, and terrestrial habitat ( Berger 2006; 2008; Jung et al. 2016).

Anteholosticha distyla differs from the Antarctica population of A. sigmoidea by the body length in vivo (150– 180 μm vs. 90–150 μm), number of adoral membranelles (30–33 vs. 22–28), number of midventral cirri (33 or 34 vs. 17–29), number of transverse cirri (two vs. four to eight), and length of the dorsal cilia (5 μm vs. 3 μm). In addition, the midventral complex of A. distyla terminates slightly ahead of the pretransverse cirri, whereas that of the present population terminates around the mid-body ( Berger 2006). The Antarctic population of A. sigmoidea can be distinguished from A. verrucosa by the number of adoral membranelles (22–28 vs. 14–17), midventral pairs (17–29 vs. 7–15), left marginal cirri (5–40 vs. 12–18), right marginal cirri (25–37vs. 13–21), and transverse cirri (four to eight vs. two to four) ( Berger 2008). Another Antarctic ciliate, A. rectangula , can be separated from A. sigmoidea by the body length in vivo (67–95 μm vs. 90–150 μm) as well as by the number of macronuclear nodules (four to eight vs. 13–19), adoral membranelles (19–22 vs. 22–28), and dorsal kineties (three vs. four) ( Jung et al. 2016). Anteholosticha australis has a long midventral complex that terminates ahead of the pretransverse cirri. In contrast, the midventral complex of the Antarctic population terminates around the mid-body of the cell. In addition, the Antarctica population of A. sigmoidea differs from A. australis by having irregularly arranged cortical granules (vs. in longitudinal rows) ( Berger 2006).

Anteholosticha australis is distinguished from A. sigmoidea by shape of the cortical granules as well as by the number of adoral membranelles and macronuclear nodules ( Berger 2006). However, if A. australis is compared with the pooled A. sigmoidea populations, most quantitative values overlap. The Tulnerfeld population of A. sigmoidea (from group I) has 11 or 12 macronuclear nodules (vs. 10–16 in A. australis ), and the Alpine population of A. sigmoidea (from group II) has 25–28 adoral membranelles (vs. 27–33 in A. australis ) ( Berger 2006; Foissner 1982; 1984). Although more adoral membranelles and macronuclear nodules were observed in A. australis , the values overlap and do not show any significant difference. Including the Antarctic population of A. sigmoidea , these two species differ only in the shape of cortical granules ( Table 2) ( Berger 2006). Molecular data are needed to test whether this morphological difference reflects the intraspecific variability or differentiation at the species-subspecies level. Unfortunately, the 18S rRNA gene sequence is available only for the Antarctic population of A. sigmoidea .

Molecular analyses ( Fig. 3 ). The 18S rRNA gene sequence of A. sigmoidea is 1,770 bp in length and has a GC content of 45.1%. The sequence was deposited in GenBank with accession number MN864520 . BI and ML analyses produced similar topologies; thus, only the ML tree is presented in Fig. 3 . The gene tree indicates the previously reported polyphyly of the genus Anteholosticha ( Fig. 3 ) ( Huang et al. 2014, Jung et al. 2016; Zhao et al. 2015).

Phylogenetic position of A. sigmoidea . Molecular analyses based on the 18S rRNA gene sequences indicate that the genus Anteholosticha is polyphyletic and scattered in nine separate lineages ( Jung et al. 2016; Zhao et al. 2015). This heterogeneity was confirmed later by molecular data ( Huang et al. 2014). Although low supporting (ML 21%), the Antarctic population of A. sigmoidea is sister to the clade including five Anteholosticha species, A. marimonilata Xu et al., 2011 , A. monilata (Kahl, 1928) Berger, 2003 (type species of Anteholosticha ), A. pulchra (Kahl, 1932) Berger, 2003 , A. pseudomonilata Li et al., 2011 , and A. rectangula Jung et al., 2016 . Among them, four species, A. monilata , A. marimonilata , A. pulchra , and A. pseudomonilata , can be distinguished from A. sigmoidea by having a long (vs. comparatively shortened) midventral complex. In addition, A. marimonilata , A. pulchra , and A. pseudomonilata were discovered in marine habitats (vs. terrestrial) ( Berger 2006). Another Antarctic terrestrial ciliate, A. rectangula , is similar to A. sigmoidea in terms of the macronuclear apparatus (arranged along the left cell margin), cirral pattern (shortened midventral complex and a similar number of cirri on the ventral surface), and habitat (Antarctic soil) ( Table 2). In the gene trees, A. sigmoidea and A. rectangula were distinctly separated, supporting the morphological differences discussed above.

Anteholosticha sigmoidea did not cluster with any other Anteholosticha species, which might be due to the insufficient molecular data for the genus Anteholosticha . In previous studies, groups I and II of A. sigmoidea were classified as a single species, and the Antarctic group shared many features (body shape elongated elliptical, macronuclear nodules arranged along the left cell margin, cortical granules present, four dorsal kineties, and terrestrial habitat) with the groups. Nevertheless, the three populations have morphological differences mentioned above, particularly the Antarctic population, which are geographically isolated. Therefore, not all three groups may represent a single species. However, DNA sequence information of groups I and II are not available to support this proposal. Additional molecular data of other known Anteholosticha species (e.g., A. australis , A. distyla , European populations of A. sigmoidea , A. sphagni , and A. verrucosa ) and of new species will be needed for more in-depth studies including the issue of A. australis and A. sigmoidea -complex.

Figures

FIG. 1. A–E. Anteholosticha sigmoidea, an Antarctic population from life (A) and after protargol preparation (B–F). A. Ventral view, a representative specimen. B, C. Infraciliature of ventral (B) and dorsal (C) side of the same specimen, the arrows indicate basal bodies at the top of the right marginal row. D. Nuclear apparatus. E, F. Arrangement of cortical granules on ventral (E) and dorsal (F) side. AZM, adoral zone of membranelles; FTC, frontoterminal cirri; Ma, macronuclear nodules; Mi, micronucleus; PTC, pretransverse cirri; TC, transverse cirri; 1–4, dorsal kineties 1–4. Scale bars, 50 μm.

FIG. 2. A–M. Anteholosticha sigmoidea, an Antarctic population from life (A–G) and after protargol preparation (H–M). A, B. Ventral and dorsal view of a representative specimen. C–E. Contractile vacuole with collecting canals. F. Arrows indicate macronuclear nodules, arrowheads mark micronuclei. G. Dorsal cilia (arrows) and cortical granules (arrowheads). H, I. Infraciliature of ventral (H) and dorsal (I) side, arrows indicate the basal bodies at the top of the right marginal row. J. Ventral view of the ciliature in an anterior body portion. K. Macronuclear nodules and micronuclei. L, M. Arrangement of cortical granules on ventral (L) and dorsal (M) surface. AZM, adoral zone of membranelles; BC, buccal cirrus; CV, contractile vacuole; FTC, frontoterminal cirri; LMR, left marginal cirral row; Ma, macronuclear nodules; Mi, micronucleus; MP, midventral pairs; RMR, right marginal cirral row; PTC, pretransverse cirri; TC, transverse cirri; 1–4, dorsal kineties 1–4. Scale bars, 50 μm.

FIG. 3. Maximum likelihood (ML) and Bayesian inference (BI) analyses based on the 18S rRNA gene sequences (majority consensus tree of ML). The sequence of Anteholosticha sigmoidea provided in the present work is indicated by bold face and an arrow. The numbers at nodes indicate the ML bootstrap values of 1,000 replicates and the posterior probability of BI. Fully supported (100/1.00) branches are marked with solid circles. The scale bar corresponds to two substitutions per 100 nucleotide positions. The asterisks mark the Anteholosticha sequences.