Diversity of Bacterial Photosymbionts in Lubomirskiidae Sponges from Lake Baikal

Sponges are permanent benthos residents which establish complex associations with a variety of microorganisms that raise interest in the nature of sponge-symbionts interactions. A molecular approach, based on the identification of the 16S rRNA and ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit genes, was applied to investigate diversity and phylogeny of bacterial phototrophs associated with four species of Lubomirskiidae in Lake Baikal. The phylogeny inferred from both genes showed three main clusters of Synechococcus associated with Baikalian sponges. One of the clusters belonged to the cosmopolitan Synechococcus rubescens group and the two other were not related to any of the assigned phylogenetic groups but placed as sister clusters to S. rubescens. These results expanded the understanding of freshwater sponge-associated photoautotroph diversity and suggested that the three phylogenetic groups of Synechococcus are common photosynthetic symbionts in Lubomirskiidae sponges.


Introduction
Sponges are an important component of the marine and freshwater benthos ecosystems that establish associations with a great diversity of unicellular and multicellular organisms [1]. At the photosynthetic zone, sponges can benefit from phototrophic symbionts which fix carbon using the Calvin-Benson cycle and provide products of photosynthesis to the host [2][3][4].
Photosynthetic symbionts are prevalent in marine sponges of coastal regions worldwide where they contribute significantly to net primary production [5,6]. From onethird to more than half of the sponges of tropical and temperate regions harbor a high level of photosynthetic symbionts [7,8]. In Lake Baikal, sponges are necessary components of the benthos and ubiquitous on rocky grounds in the littoral zone. Sponges from the endemic family Lubomirskiidae are widely distributed in Lake Baikal and often harbor photosynthetic symbionts. From 14 described species of Lubomirskiidae [9], there are three common species among which photosynthetic Lubomirskia baicalensis (L. baicalensis) and Baikalospongia bacillifera (B. bacillifera) are widely distributed in the photic zone of Lake Baikal. In contrast to marine sponges, there is not a lot of data on photosynthetic symbionts of freshwater sponges, although associations with unicellular green algae, including Chlorella spp., Choricystis minor, yellow-green algae, and Chloroflexi have been shown in cosmopolitan sponges [4,[10][11][12][13] and cyanobacterial sequences detected in L. baicalensis [14]. Nevertheless, the identification and diversity of sponge-associated phototrophs in Lake Baikal are undetermined to date.
2 International Journal of Biodiversity of Lake Baikal at depths of 10-20 m by scuba diving (Table S1, see Supplementary Material available online at http://dx.doi.org/10.1155/2014/152097). Sponge tissue samples (3-5 cm 3 ) were rinsed twice in 96% ethanol and stored at 4 ∘ C until DNA extraction. Species identification was based on external morphological characteristics and morphology of spicules according to the guides of Rezvoi (1936) and Efremova (2001) [22,23].
Total DNA was extracted using the RiboSorb kit (AmpliSens, Russia) according to the manufacturer's protocol. The RbcL gene was amplified in 15 L of PCR reaction mix (Screen Mix, Evrogen, Russia) with 10 pmol of each primer, cbbL 595f (5 -GACTTCACCAAAGACGACGA-3 ) and 1387r (5 -TCGAACTTGATTTCTTTCCA-3 ), as described by Elsaied and Naganuma [16]. Touchdown PCR was done in four repeats and mixed for each tissue sample. The amplification profile consisted of an initial denaturation at 95 ∘ C for 3 min followed by 5 touchdown cycles of 94 ∘ C for 20 s, 55 ∘ C for 20 s, and 72 ∘ C for 1 min and then 20 cycles of 94 ∘ C for 20 s, 52 ∘ C for 20 s, and 72 ∘ C for 1 min and 10 cycles of 94 ∘ C for 20 s, 50 ∘ C for 20 s, and 72 ∘ C for 1 min. 35 PCR cycles were followed by a final extension at 72 ∘ C for 15 min. The 16S rRNA gene was amplified with primers CYA106Fc (5 CGGACGGGTGAGTAACGCGTGA 3 ) and CYA781R (5 GACTACWGGGGTATCTAATCCCWTT 3 ) [26]. The amplification profile consisted of an initial denaturation at 95 ∘ C for 3 min followed by 30 cycles of 94 ∘ C for 20 s, 60 ∘ C for 20 s, 72 ∘ C for 1 min and a final extension at 72 ∘ C for 15 min.
The PCR products were detected by electrophoresis in 0.8% agarose gel and purified using a DNA purification kit (Cytokine, Russia) and cloned into the pAL-TA vector (Evrogen, Russia). Inserts were detected by amplification with M13 primers and digested with endonucleases HhaI and HaeIII. Clone inserts with different restriction profiles were sequenced using ABI 3130xl Genetic Sequencer (Applied Biosystem, USA) (Table S1). Sequences were checked for chimeras with the Decipher tool [27].
For phylogenetic tree reconstruction maximum parsimony (MP), maximum likelihood (ML), and Bayesian analysis were performed. Kimura 2-parameter model was used for estimating of genetic distances [31]. The first two codon positions were used for analysis of rbcL dataset (MP, ML). Bootstrap analysis was performed with 1000 replicates. In the Bayesian analysis, the MCMC chain was run for 1,000,000, sampled every 100th step.

Results
Clone library screening was performed to identify phototrophs associated with Baikalian sponges. The 16S rRNA and rbcL gene fragments were analysed in L. baicalensis, L. abietina, B. bacillifera and B. martinsoni. The chlorophyll level was estimated for three of these species and chlorophyll A was detected in L. baicalensis (122 ± 32 g/g), B. bacillifera (37 ± 12 g/g), and L. abietina (20 ± 0.5 g/g).
In total, 71 and 126 inserts were sequenced from the 16S rRNA and rbcL clone libraries, respectively. Identical sequences were removed from further analysis and 38 16S rRNA and 54 rbcL gene sequences were deposited in the GenBank database. Sequences obtained from 16S rRNA gene were 95-100% identical, while sequences obtained from the rbcL gene were more heterogeneous (83-100%). RbcL sequences with nucleotide identities greater than 90% were combined into operational taxonomic units (OTUs). There were three -cyanobacteria OTUs resolved from analysis of the 733 bp rbcL gene alignment. The first OTU was 98% identical to Synechococcus rubescens (AM701775) while the maximum nucleotide sequence identities from two other OTUs were 87% and 89% similar to Cyanobium gracile 6307 (CP003495), and closer relatives could not be found in the GenBank database.
The phylogenetic analysis inferred from 16S rRNA and rbcL genes showed three main clusters for Synechococcus which included sequences from all species of sponges and lake water analysed (Figures 1 and 2). Better resolved trees were based on rbcL in comparison with 16S rRNA based phylogeny. The phylogenetic reconstruction inferred from both genes showed that phylogenetic positions for two of three clusters were similar. There was a fully resolved cluster of S. rubescens that combined highly identical (99%) sequences from Lake Baikal, as well as Synechococcus strains from European lakes. The two other groups were placed as sister clusters to S. rubescens (Figure 1). Clusters BL1 and BS1 included sequences from Lake Baikal only (Figures 1 and 2).
The phylogenetic positions of identified clusters were slightly different from the 16S rRNA based phylogeny. Using 16S rRNA data, there was a cluster related to the S. rubescens phylogenetic lineage that included the clade of 23 Synechococcus sequences (intragroup identity 98-99%) from Lake Baikal, sequences from lakes in Mongolia, and high altitude oligotrophic lakes of the Tibetan plateau and the Pyrenees, while rbcL-based phylogeny showed a specific cluster of 25 sequences that was sister to S. rubescens cluster (Figures 1  and 2). The rest of the sequences (16S rRNA tree) formed single branches or were clustered within Baikalian clades of picocyanobacteria from lake water (Figure 2).
The only rbcL sequence from B. bacillifera (JX570973) that showed a low-level nucleotide identity (<83%) with other  sequences from the GenBank database formed a distinct branch within the -cyanobacteria ( Figure 1). Thus, the majority of sequences from sponges and water belonged to -cyanobacteria with form IA RuBisCO. Sequences from Baikalian sponges did not form distinct phylogenetic clusters and no major sponge-specific clusters have been found.

Discussion
The phylogeny inferred from 16S rRNA and rbcL genes generally agrees that three main phylogenetic clusters for chlorophyll IA-containing picocyanobacteria are present in Baikalian sponges. The only one sequence belonging to Oscillatoria was detected in B. bacillifera. The finding of Synechococcus and Oscillatoria groups in freshwater sponges is similar to the results obtained from marine sponges [7,8,[17][18][19]. In contrast with marine sponges, no specific phylogenetic groups for sponge species or sponges with various levels of chlorophyll A were found, and sequences derived from sponges (as well as from lake water) were dispersed across the clusters described.
The different levels of chlorophyll A in sponges are also demonstrated in marine environments, where the major role for sponge-specific Synechococcus in ecology and primary productivity of sponges with high level of chlorophyll A has been described [7]. The highest level of chlorophyll A we found in L. baicalensis may be dependent on better light condition available to branched structures that grow up to one meter high in comparison to fully attached cushionand crust-like sponges (L. abietina, B. bacillifera, and B. martinsoni), though the presence of eukaryotic photosymbionts could also be significant.
The S. rubescens cluster detected in Baikalian sponges was affiliated with widely distributed Subalpine cluster I (group B) [20] proving high ecological plasticity of this group. The presence of phylogenetic clusters that included only Baikalian sequences, as well as clusters with no attribution to the Synechococcus phylogenetic group described, confirms the results of Pommier et al. [21] suggesting that endemic clusters are formed in the local bacterioplankton community in response to their adaptation to unique ecological conditions. These results give a new insight into biodiversity of phototrophs associated with Baikalian sponges that could be further used for purpose of ecosystem monitoring. This is especially important with regard to increasing anthropogenic pressure as a result of tourism development in the Baikalian region.

Conclusions
Molecular detection of 16S rRNA and rbcL genes from clone libraries were performed for the identification of phototrophs associated with four species of Lubomirskiidae from Lake Baikal. The phylogeny inferred from both genes showed that S. rubescens and two specific clusters of Synechococcus were associated with sponges and also found in lake water. Three major clusters of Synechococcus that did not form sponge-specific groups were detected in all studied species of Baikalian sponges. These results added to the understanding of symbiotic associations in freshwater photosynthetic sponges and could be further applied to the assessment of ecological impacts on the ecosystems in Lake Baikal.