SIT1

Summary: InterPro2GO mapping from the diatom silicon transporter family signature (IPR004693). The process is supported for SIT1 by its plasma-membrane localization and by knockdown of SIT1 together with SIT2, which slows silicic acid uptake at low external concentrations.

Reason: The family mapping is sound: SITs are the only proteins of the IPR004693/PF03842 family, and purified family members transport silicic acid into proteoliposomes (PMID:27305972). For SIT1 itself, eGFP-tagged SIT1 is in the plasma membrane and a construct that knocks down SIT1 and SIT2 together delays silicic acid depletion from 20 uM medium (PMID:25380754). The knockdown cannot separate SIT1 from SIT2, but plasma-membrane location, family activity and the knockdown phenotype together support import across the plasma membrane. The process holds only for low silicic acid; at high concentrations uptake is diffusional (PMID:18162598).

Summary: Proposed new location. SIT1-eGFP expressed from its native promoter localizes to the plasma membrane of silicon-starved T. pseudonana cells.

Reason: Direct localization of a C-terminal eGFP fusion of this gene product under native promoter control; the plasma membrane signal is strong under silicon starvation, with only faint ER labelling in G1 interpreted as transit to new plasma membrane. The location agrees with the accepted import process and with the predicted polytopic topology.

Summary: Proposed new molecular function, inferred from SIT-family members whose transport activity was measured directly in proteoliposomes. No GO term exists for silicic-acid-specific transporter activity or for sodium-coupled silicic acid symport (the accepted GO:0015708 covers the process, not the activity), so this is the most specific available term.

Reason: SIT1 has not itself been purified or assayed (it expressed very poorly in yeast), but it belongs to the SIT family, whose members PtSIT1 and NaSIT transport silicic acid with classical kinetics in reconstituted proteoliposomes driven by an inward sodium gradient, and whose T. pseudonana paralog TpSIT3 was reconstituted in the same assay. TpSIT3 is a weaker donor than PtSIT1: it gave only a modest signal just discernible above background, so the kinetics were measured on PtSIT1 and NaSIT. B7T1F4 (PtSIT1) therefore carries the inference and B8C4N0 (TpSIT3) is a supporting conspecific donor rather than an equal one. SIT1 is 88% identical to SIT2 and is plasma-membrane localized, and its knockdown together with SIT2 slows uptake at low silicic acid. GO defines this term on silicates as salts of silicic acids, which covers the Si(OH)4 substrate; GO has no silicic-acid-specific transporter activity term (see proposed_new_terms). The sodium coupling is a family-level inference, so it is left out of the annotation and raised as a question. With/From donors (UniProt REST and NCBI lookups): PtSIT1 is UniProtKB:B7T1F4 (gene SIT1, EMBL EU879093, 512 aa). TpSIT3 is UniProtKB:B8C4N0, a 437-residue fragment gene model that is an exact internal substring of the full-length 565-residue TpSIT3 cDNA protein (GenBank ABB81827), whose own UniProtKB entry Q0QVM6 was deleted as not part of a reference proteome. NaSIT is omitted: the only full-length Nitzschia alba SIT in NCBI (GenBank ABB81817, 549 aa) maps to UniProtKB Q0QVN6, also deleted, and no live UniProtKB entry exists. The paper's cached text does not list accessions, so the donor sequences are matched by gene name and length.