The Evolution of Silicon Transport in Eukaryotes

Abstract

Biosilicification (the formation of biological structures from silica) occurs in diverse eukaryotic lineages, plays a major role in global biogeochemical cycles, and has significant biotechnological applications. Silicon (Si) uptake is crucial for biosilicification, yet the evolutionary history of the transporters involved remains poorly known. Recent evidence suggests that the SIT family of Si transporters, initially identified in diatoms, may be widely distributed, with an extended family of related transporters (SIT-Ls) present in some nonsilicified organisms. Here, we identify SITs and SIT-Ls in a range of eukaryotes, including major silicified lineages (radiolarians and chrysophytes) and also bacterial SIT-Ls. Our evidence suggests that the symmetrical 10-transmembrane-domain SIT structure has independently evolved multiple times via duplication and fusion of 5-transmembrane-domain SIT-Ls. We also identify a second gene family, similar to the active Si transporter Lsi2, that is broadly distributed amongst siliceous and nonsiliceous eukaryotes. Our analyses resolve a distinct group of Lsi2-like genes, including plant and diatom Si-responsive genes, and sequences unique to siliceous sponges and choanoflagellates. The SIT/SIT-L and Lsi2 transporter families likely contribute to biosilicification in diverse lineages, indicating an ancient role for Si transport in eukaryotes. We propose that these Si transporters may have arisen initially to prevent Si toxicity in the high Si Precambrian oceans, with subsequent biologically induced reductions in Si concentrations of Phanerozoic seas leading to widespread losses of SIT, SIT-L, and Lsi2-like genes in diverse lineages. Thus, the origin and diversification of two independent Si transporter families both drove and were driven by ancient ocean Si levels.

Keywords: Lsi2; SIT; convergent evolution; eukaryotes; silicon; transporter..

Fig. 1

Si biomineralization across the Eukaryotes. The eukaryotic phylogeny is based on Adl et al. (2012). Major eukaryotic supergroups are named in boxes. Those taxa with highlighted names contain one or more biosilicifying species. Taxa with widespread biosilicification and extensively silicified lineages are in bold and underlined.

Fig. 2

Summary table of SIT, SIT-L, and Lsi2-like genes detected in selected species. Si transporters are widely distributed across the eukaryotic supergroups and are found in species that are extensively silicified (black circle), partially silicified (gray circle), and nonsiliceous (white circle). Hatched circles signify species where no biosilica has been reported but where the wider class is extensively silicified. A filled square indicates that the gene was detected (dark red for SIT family, light green for Lsi2-like). The Florenciella sp. SIT and Collozoum SIT-L were classified on the basis of alignments and preliminary phylogenies (see Materials and Methods). White squares indicate absence from fully sequence genomes; blank space denotes that the relevant gene was not detectable but only transcriptomic data was available. Taxonomic classifications and degrees of silicification match those from fig. 1.

Fig. 3

A generalized structure for SITs and SIT-Ls. (A) SIT structure schematic. (B) SIT-L structure schematic. Transmembrane helixes (gray) are based on predictions from TMPred. Conserved motifs (boxes, proposed binding sites highlighted in yellow, other conserved motif in bright green) were determined from the alignment in supplementary figure S1, Supplementary Material online. Circles are conserved residues: hydroxylated residues = dark green, positive residues= bright red, negative residues = magenta.

Fig. 4

Phylogenetic tree of SIT and SIT-L sequences. Bacterial SIT-Ls form a distinct clade, and eukaryotic SITs can be divided into two main groups. Group 1 SITs and Group 1 SIT-Ls are connected by a branch with good support (93/90/98), whereas the basal branching order of Group 2 has poor statistical support. SITs and SIT-Ls largely follow the species phylogeny, with the exception of the paraphyletic stramenopile, dinoflagellate, and rhizarian sequences in Group 2. Brown = bacteria, dark green = haptophyte, gray = rhizarian (light gray = foraminiferan), Bright red = choanoflagellate, magenta = metazoan, orange = dinoflagellate, dark blue = diatoms, light blue = other stramenopiles. SIT-L sequences are in boxes. This unrooted radial tree is based on the RaxML maximum likelihood analysis with the best-fitting LG + G4 model from an alignment of 485 amino acid residues. Numbers at nodes are a percentage of 100 bootstrap or 1000 ultrafast bootstrap replicates in the format RaxML/PhyML/IQ-TREE support value, with */*/* signifying nodes with 100% support for all methods (for clarity only support values for major nodes are shown, see supplementary figure S6, Supplementary Material online for full trees). Scale bar indicates average number of amino acid substitutions per site.

Fig. 5

Phylogenetic tree of SIT N-terminal halves, SIT C-terminal halves and SIT-Ls. Phylogenetic analysis of halved SITs and SIT-Ls reflects that of the full-length sequences (fig. 4). The N-terminal and C-terminal halves of the Group 1 SITs (diatom, haptophyte, and choanoflagellate) are, respectively, most related to each other. In contrast, the N- and C-terminal halves of the nondiatom stramenopiles SITs of Group 2 are more closely related to the other half of the same gene. This indicates that these SITs arose by independent duplication, and this may have even happened multiple times in the chrysophytes/synurophytes. Arrows show different inferred duplication-inversion-fusion events, with solid arrows indicating a duplication event, open arrows the resulting N-terminal halves and hatched arrows the resulting C-terminal halves. The various arrow angles denote individual inferred SIT-producing events; arrows of the same angle are inferred to be part of the same event. Brown = bacteria, dark green = haptophyte, gray = rhizarian (light gray = foraminiferan), bright red = choanoflagellate, magenta = metazoan, orange = dinoflagellate, dark blue = diatom, light blue = other stramenopiles. Tree based on RaxML maximum likelihood analysis with the best-fitting LG + G4 model from an alignment of 166 amino acid residues. Note that the tree topology is that of an unrooted tree; the bacterial SIT-L clade was arbitrarily designated as an outgroup for presentation purposes. Numbers at nodes are a percentage of 100 bootstrap or 1000 ultrafast bootstrap replicates in the format RaxML/PhyML/IQ-TREE value. Scale bars indicate average number of amino acid substitutions per site. Full phylogenetic trees are given in supplementary figure S7, Supplementary Material online.

Fig. 6

Phylogenetic tree of Lsi2-like sequences from a taxonomically diverse range of eukaryotes and prokaryotes. The sequences divide into two main groups: one containing Pink-Eyed Dilution P-protein (PED) sequences and the other containing the active Si transporter Low Si 2 (Lsi2). Among opisthokonts most metazoans and all choanoflagellates investigated had at least one PED-like sequence; conversely only loricate choanoflagellates, siliceous sponges and three lophotrochozoans (Lottia gigantea, C. teleta, and Lingula anatina) possessed Lsi2-like sequences. Plant Lsi2 sequences form a strongly supported (86% RaxML, 99% PhyML, 100% IQ-TREE bootstrap) monophyletic clade. Sequences from multiple siliceous and nonsiliceous eukaryotes branch in the Lsi2-like clade, including a Thalassiosira pseudonana Si-responsive gene, which falls within a well-supported (100% RaxML, 100% PhyML, 100% IQ-TREE bootstrap) diatom branch. Brown = bacteria, dark green = haptophyte, gray = rhizarian (light gray = foraminiferan), bright red = choanoflagellate, magenta = metazoan, orange = alveolate (light orange = dinoflagellate), dark blue = diatom, light blue = other stramenopiles, bright green = archaeplastids, turquoise = cryptophyte, yellow = amoebozoan (note contamination of MMETSP sequences is accounted for, see supplementary table S4, Supplementary Material online). The tree was produced using RaxML maximum likelihood analysis with the best-fitting LG + G4 model from an alignment of 247 amino acid residues. Nodes with <20% bootstrap support were collapsed to give a topology agreed across all methods (see supplementary text S1–S5, Supplementary Material online for full uncollapsed tree files). Scale bar indicates average number of amino acid substitutions per site; slashes indicate very long branches that were clipped for display purposes.

Fig. 7

Schematic model of SIT evolution. This interpretation is based on the phylogenetic analyses presented in figs. 4 and 5. SITs likely originated as a gene encoding a 5-TMD protein (circled), and diversified into the Group 1 SITs, Group 1 SIT-Ls and Group 2 SIT/SIT-Ls. Independent duplication (black arrows), inversion (dashed lines), and fusion (white arrows) events gave rise to 10-TMD SITs in both the Group 1 SITs and in Group 2. SIT protein structure diagrams are adapted from fig. 3, with N and C terminals shown, TMDs in gray crossing the membrane (in blue). Conserved GRQ and EGXQ motifs are highlighted in yellow.

Fig. 8

Geological history of Si. The graph plots approximate seawater silica concentration (at 25 °C) over the past 600 My, based on Racki and Cordey (2000). Indicated are the approximate age ranges of the three biosilicification phases, the Precambrian (black), Palaeozoic (hatching), Mesozoic (white), and Cenozoic (gray) eras and major events involving biosilicifying organisms. The first phase extends from at least the Archean (3000 Ma) until the Precambrian/Cambrian boundary. This is characterized by high seawater Si concentrations and witnessed the origin of the eukaryotic supergroups (a, ∼2700 Ma) (Parfrey et al. 2011). The second phase occurred between the Cambrian and the Mesozoic and saw a fall in seawater Si concentrations. Here, we find the first fossil evidence for biosilicifying organisms (b, spicules from ∼540 Ma) and widespread biogenic sedimentary silica deposits from sponges and radiolarians (Maldonado et al. 1999; Knoll 2003; Knoll and Kotrc 2015). The third phase covers the further reduction in seawater Si concentrations to modern levels (∼10 μM in surface waters). This phase is marked out by the appearance of new siliceous groups in the fossil record (c, approximately 200 Ma), the rise to dominance of the diatoms (d, ∼33 Ma) and reduced biosilicification across several taxa.

Fig. 9

Distribution of biomineralization and transporter genes in eukaryotic groups. Summary table of the presence of silica biomineralization, calcium biomineralization and Si-related transporter genes herein examined in selected eukaryotic groups (taxonomic classification based on Adl et al. 2012, italics signify incertae sedis). Black circle = extensive/widespread biosilicification, gray circle= minor/limited biosilicification, white circle = biosilicification absent. Gray shading denotes where genomic or transcriptomic data was available for analysis; asterisks mark where data from that lineage was only available from non-siliceous species. Symbols are placed in taxa where relevant genes were detected; blanks indicate gene not detected and are not a definitive statement of absence. 1 = Group 1, 2 = Group 2; double for SIT, single for SIT-L (see fig. 4). L = Lsi2-like, P = pink-eye dilution-like, Int = intermediate type (see fig. 6). The simplified phylogeny (based on fig. 1) is annotated with our hypothesis for the SIT repertoire (circled) of the eukaryotic LCA, and at the base of each eukaryotic supergroup. This interpretation is based on the ancestral origin hypothesis that assumes vertical inheritance in major groups rather than HGT (see Discussion). Question marks identify supergroups where no SITs or SIT-Ls have been confirmed thus far. Note that Group 2 SIT-Ls are today only found in the SAR supergroup; however, the lack of robust phylogenetic support for the branches separating Group 2 SITs/SIT-Ls from the eukaryotic LCA and the uncertain position of the root (fig. 4) means we cannot rule out a Group 2 origin in the eukaryotic LCA (marked by the 2? symbol) based on our present results. Silicified structures: CW = cell wall, Cys = cysts, Gr = granules, Lor = lorica, Mp = mouthparts, Phy = phytoliths, Sc = scales, Sp = spicules, T = test, Tb = tablets,? = uptake evidence for unknown Si utilization; brackets indicate that is a Si minor component of composite biomineralized structure. The “Ca?” column reports instances of calcification with a tick (Faber and Preisig 1994; Knoll 2003; Knoll and Kotrc 2015). References: 1. Lahr et al. 2013; 2. Sperling et al. 2010; 3. Monniot et al. 1992; 4. Carlisle 1981; 5. Karlson and Bamstedt 1994; 6. Bone et al. 1983; 7. Hua and Li 2007; 8. Williams 1998; 9. Desouky et al. 2002; 10. Leadbeater 2015; 11. Nicholls and Durrschmidt 1985; 12. Patterson and Durrschmidt 1986; 13. Patterson and Durrschmidt 1988; 14. Conforti et al. 1994; 15. Preisig 1994; 16. Chopin et al. 2004; 17. Fuhrman et al. 1978; 18. Millington and Gawlik 1967; 19. Hodson et al. 2005; 20. Durak et al. 2016; 21. Drescher et al. 2012; 22. Yoshida et al. 2006; 23. Foissner et al. 2009; 24. Sen Gupta 2003; 25. Anderson 1986; 26. Ogane et al. 2010; 27. Anderson 1994; 28. Febvre-Chevalier 1973; 29. Mizuta and Yasui 2012.