Enterohepatic bile salt cycling transport relay; BSEP/ASBT/FABP6/OSTalpha-beta/NTCP

Conjugated bile salts are not consumed by the digestion they enable: they are secreted into bile, act as detergents in the intestinal lumen, and are then reclaimed and returned to the liver, so that the whole bile salt pool recirculates many times a day and only a small fraction is lost in faeces. The relay is carried out by a small set of membrane transporters acting in a fixed order around a closed loop, each at a defined pole of a polarised epithelium: ATP-driven export across the hepatocyte canalicular (apical) membrane into bile (BSEP/ABCB11); sodium-coupled uptake across the apical brush border of the terminal ileal enterocyte (ASBT/SLC10A2); cytosolic shuttling across the enterocyte by a soluble bile acid binding protein (FABP6); basolateral efflux into portal blood by an obligate heteromeric facilitative transporter (OSTalpha/OSTbeta, SLC51A/SLC51B); and sodium-coupled reuptake across the hepatocyte basolateral (sinusoidal) membrane (NTCP/SLC10A1), closing the cycle. The hepatic uptake step has two parallel implementations - the sodium-dependent SLC10A1 route for conjugated bile salts and the sodium-independent OATP route for unconjugated and some conjugated species - which is modelled here as a variant set. Two paralogous members of one family (PANTHER:PTHR10361) supply the two sodium-coupled steps at opposite epithelial poles, and PAINT resolves this as a sub-branch split: the apical ileal node and the basolateral hepatic node are distinct ancestral nodes within the family. Loss of any single step produces a recognisable human disorder, which is the clearest evidence that the steps are required parts rather than redundant routes.

Scope: Vertebrate enterohepatic transport relay for conjugated bile salts, bounded at the membrane transport steps that move bile salts between hepatocyte, bile, intestinal lumen, enterocyte and portal blood. Bile acid synthesis from cholesterol, side-chain conjugation, bacterial deconjugation and dehydroxylation in the colon, and FXR/FGF19 transcriptional feedback are adjacent modules and are deliberately outside this boundary; they are referred to here only as the source, modification and regulation of the cargo this module moves. No module-level deep research report accompanies this module: the deep-research providers configured in this checkout were unavailable when it was written (Falcon returned 402 Payment Required, OpenAI 401), so the module was assembled from Reactome, the cached publications and the SLC10 gene reviews in this repository. Five participants - ABCB11, SLC51A, SLC51B, FABP6 and SLCO1B1 - do not yet have gene reviews here, so module QC will report them as coverage gaps; they are grounded on UniProt and PANTHER identifiers only. Bacterial modification of the cargo (deconjugation, 7-alpha-dehydroxylation) changes which species each transporter sees but adds no step performed by a host gene product, so it is left out of the parts list. Sodium stoichiometry is written inconsistently across the sources cited here, and the module deliberately lists sodium(1+) as a substrate without a coefficient rather than picking a side. Reactome:R-HSA-194187 writes the ASBT reaction with one Na+ per bile salt, while voltage-clamp measurement of the human transporter gives 2:1 (a 3-fold drop in extracellular Na+ shifted the flux-voltage relationship by -52 mV, PMID:9856990) and UniProt's curated RHEA reactions for both SLC10A1 and SLC10A2 (e.g. RHEA:71875) use two. GO:0008508 does not fix the stoichiometry, so the annotations stand either way. This is recorded here rather than as a knowledge gap because the quantity is settled experimentally; what differs is how two databases wrote it down, which a curator resolves and no experiment would. ABCB11 has two conflicting PANTHER assignments: PANTHER's own classification file places it in PTHR24221:SF165 "BILE SALT EXPORT PUMP", which this module uses, while UniProt's PANTHER cross-reference points to PTHR43394:SF23, in a family named for the mitochondrial MDL1 permease. The member index keeps both. A reasoned row in interpro/panther/panther-members-overrides.tsv would record the adjudication permanently; it has not been added because the disagreement is about family boundaries in the wider ABCB clade rather than about this module. That is a curator writing a row, not an open question about the biology, which is why it sits here and not in knowledge_gaps.