sallentina and R. sp. SWK7 (112 shared sulfatases). The close relationship between R. baltica and R. europaea was also confirmed by phylogenies

based on 16S rRNA genes, DNA–DNA hybridization and multi locus sequence analysis ( Winkelmann et al., 2010). The vast majority of sulfatase genes in the dataset were found to be single copy genes in their respective genomes. This suggests an immensely diverse range of application for the encoded proteins. Sulfatases being identified as involved in cellular mechanisms apart from carbohydrate degradation in previous studies (Wecker et al., 2009 and Wecker et al., 2010) were in any case conserved in at least three OTUs. Phylogenetic analysis on the protein sequences was carried

out with both Neighbor Joining selleck and Maximum Likelihood methods in order to reveal evolutionary relations and functional capabilities. Sulfatase sequences representing one gene per species and cluster, in total 708 sequences, were selected and aligned with 67 sequences of reviewed sulfatases from UniProt, resulting in an alignment with 6429 positions. The sequence lengths varied between 264 and 1829 amino acids (the latter ABT-263 mouse one being a fusion enzyme with two sulfatase domains and an additional domain of unknown function (DUF1680) exclusively found in the genome of R. sallentina). Several other orthologous genes featured a multi-domain structure with genes sizes above 1000 residues, but the vast majority of all sequences ranged between 450 and 550 residues in length. Both obtained trees showed the same topology. Fig. 4 Ureohydrolase depicts the Maximum Likelihood

tree as unrooted and circular. The early stages of the sulfatase evolution showed low confidence values in general. The tree revealed 22 distinct branches with at least two clustered sequences, with three additional single Rhodopirellula sp. sequences being unclustered and possibly representing distinct functionality. Of the 22 branches, 19 branches contained sequences of Rhodopirellula origin, while the remaining three branches were consisting of reference sequences only: (i) glucosamine (N-acetyl)-6-sulfatase (GNS) together with mammalian sulfatases 1 and 2, (ii) two Chlostridium sulfatases (SULF_CLOP1 and SULF_CLOPE), and (iii) eukaryotic arylsulfatases (arsK) were not clustered to any Rhodopirellula sequence, respectively. Two reference sequences from Bacteria represented single sequence lineages: the E. coli gene yidJ and the choline sulfatase betC from Sinorhizobium meliloti. All Rhodopirellula spp. contained sequences of all 19 branches. Five of the major branches contained both known and Rhodopirellula sequences (Clusters G, H, I, M, and N, respectively; Table 2), leaving 14 clusters of just Rhodopirellula sp. genes, which are not closely related to any sulfatase sequence with known activity.