Taxonomically and Functionally Distinct Ciliophora Assemblages Inhabiting Baltic Sea Ice

Abstract

Ciliophora is a phylum of unicellular eukaryotes that are common and have pivotal roles in aquatic environments. Sea ice is a marine habitat, which is composed of a matrix of solid ice and pockets of saline water in which Ciliophora thrive. Here, we used phylogenetic placement to identify Ciliophora 18S ribosomal RNA reads obtained from wintertime water and sea ice, and assigned functions to the reads based on this taxonomic information. Based on our results, sea-ice Ciliophora assemblages are poorer in taxonomic and functional richness than under-ice water and water-column assemblages. Ciliophora diversity stayed stable throughout the ice-covered season both in sea ice and in water, although the assemblages changed during the course of our sampling. Under-ice water and the water column were distinctly predominated by planktonic orders Choreotrichida and Oligotrichida, which led to significantly lower taxonomic and functional evenness in water than in sea ice. In addition to planktonic Ciliophora, assemblages in sea ice included a set of moderately abundant surface-oriented species. Omnivory (feeding on bacteria and unicellular eukaryotes) was the most common feeding type but was not as predominant in sea ice as in water. Sea ice included cytotrophic (feeding on unicellular eukaryotes), bacterivorous and parasitic Ciliophora in addition to the predominant omnivorous Ciliophora. Potentially mixotrophic Ciliophora predominated the water column and heterotrophic Ciliophora sea ice. Our results highlight sea ice as an environment that creates a set of variable habitats, which may be threatened by the diminishing extent of sea ice due to changing climate.

Keywords: DNA metabarcoding; Mixotrophy; Phylogenetic placement; Predator–prey interactions; Winter ecology.

Fig. 1

Baltic Sea ice associated Ciliophora placed in an unrooted 18S ribosomal RNA gene reference tree [29]. The clades of the reference tree that do not contain Baltic Sea ice associated Ciliophora are collapsed. The classes that include Baltic reads are written in bold. The size of the circle is proportional to the number of unique reads placed at the given node

Fig. 2

Number (N) and evenness of 98% Ciliophora operational taxonomic units (OTUs) and functions in the time-series samples, based on normalized number of reads (39,861 reads/sample). (a) Sea-ice samples (***, 15 samples) had significantly lower richness but significantly higher evenness than under-ice water (15 samples) and water-column samples (16 samples) according to one-way analysis of variance (ANOVA) (richness: F = 22.09, p = 2.70E-6; evenness: F = 54.26, p = 1.74E-12) and following Tukey’s pairwise comparisons (richness: ice–under-ice water Tukey’s Q = 8.18, p = 0.00013; ice–water column: Q = 5.93, p = 0.00050; evenness: ice–under-ice water Tukey’s Q = 14.59, p = 0.0001; ice–water column: Q = 9.82, p = 0.0001). In addition, under-ice water samples (*) had significantly lower evenness than water-column samples (Tukey’s Q = 4.77, p = 0.005). (b) Sea-ice samples (***) had significantly fewer but significantly more evenly distributed functions than under-ice water and water-column samples according to one-way ANOVA (richness: F = 4.94, p = 0.012; evenness: F = 67.68, p = 5.21E-14) and following Tukey’s pairwise comparisons (richness: ice–under-ice water: Tukey’s Q = 3.74, p = 0.03; ice–water column: Q = 3.98, p = 0.02; evenness: ice–under-ice water: Tukey’s Q = 14.30, p = 0.0001; ice–water column: Q = 14.34, p = 0.0001)

Fig. 3

Differences in Ciliophora assemblages and their functions. (a, d) Non-metric multidimensional scaling (NMDS) plots based on Bray–Curtis dissimilarity indices of the Ciliophora assemblages in the different samples. (a) Taxonomic composition, (d) functional composition. (b, e) Distance-based redundancy analysis plots based on binomial distribution of the Ciliophora assemblages in the different samples. Significance was tested with a permutation test (999 permutations, significance level p < 0.05) and following pairwise Adonis with Holm-corrected p-values. (b) Taxonomic composition. All sample types differed significantly (p = 0.003). (e) Functional composition. Functions in ice differed significantly from functions in water and under-ice water (UIW, p = 0.003) and functions in UIW differed significantly from functions in the water column (p = 0.015). (c, f) Venn diagrams showing variation partitioning for Ciliophora assemblages

Fig. 4

Ciliophora succession. (a) Relative abundance of 98% Ciliophora operational taxonomic units (OTUs) in sea ice, under-ice water (UIW) and the water column at two time-series sampling stations. (b) Relative abundance of Ciliophora functional groups in sea ice, under-ice water and the water column at two time-series sampling stations. The abbreviated functions include: (B) bacteria filtration, (P*) histophagous, (O) omnivorous, (C) cytotrophic, (X) parasitic, (D) occasionally detritivorous, (s) surface dwelling, (p) planktonic, (h) heterotrophic and (m) potentially mixotrophic. OTUs and functions were merged to show the most abundant groupings

Fig. 5

Local contribution to beta diversity (LCBD) values at the different sampling sites (Kr, Krogarviken; St, Storfjärden) and sample types over time of Ciliophora assemblages (a) and Ciliophora functions (b). The circle surface areas are proportional to the LCBD values. Circles with a black rim indicate significant LCBD values at the 0.05 level. UIW, under-ice water