MDR1 (PfMDR1 / Pgh1, PF3D7_0523000, UniProt Q7K6A5) — curation notes
Provenance note. Automated deep research was not available in this environment
(Falcon auth failure, invalid OpenAI key), so no *-deep-research-*.md file was generated.
What follows is a manual literature review carried out by reading the cached UniProt
record, the cached publications in publications/, and targeted web searching. Every
assertion below carries inline provenance as [PMID:NNNN "verbatim quote"] or a pointer
to the UniProt flat file.
Identity
- UniProt accession Q7K6A5; the entry carries no gene symbol, so the review's
gene_symbol stays as the accession and the folder is the alias MDR1.
- Commonly called PfMDR1 (P. falciparum multidrug resistance protein 1) or, in the
older literature, Pgh1 (P-glycoprotein homologue 1); gene id PF3D7_0523000.
- Family: UniProt — "Belongs to the ABC transporter superfamily. ABCB family.
Multidrug resistance exporter (TC 3.A.1.201) subfamily."
PANTHER: PTHR43394 / PTHR43394:SF27 (ATP-DEPENDENT TRANSLOCASE ABCB1-LIKE).
Architecture: two ABC_membrane (PF00664) TMDs + nucleotide-binding domains (PF00005),
i.e. a full-length type-1 ABC exporter (InterPro IPR039421).
Localization — digestive (food) vacuole membrane
The founding localization paper is unambiguous:
Confirmed repeatedly since:
UniProt records SUBCELLULAR LOCATION: Vacuole membrane {ECO:0000269|PubMed:1674943,
ECO:0000269|PubMed:8097560}; Multi-pass membrane protein.
Curation consequence. The organelle is the apicomplexan food vacuole (GO:0020020),
and the protein sits in its membrane. GO has GO:0005774 vacuolar membrane but (checked
via OLS) no "food vacuole membrane" term, so the best available pair is
GO:0020020 + GO:0005774. The ARBA-derived GO:0005737 cytoplasm is true only in the
trivial sense that the vacuole is a cytoplasmic organelle; it is uninformative for an
integral DV-membrane protein.
Nucleotide binding and ATPase activity
- Photoaffinity labelling of purified digestive vacuoles with 8-N3-ATP labels PGH1:
PMID:8097560
and the labelling is nucleotide-specific:
PMID:8097560
The authors already read this as a transport function at the DV membrane:
PMID:8097560
- Direct biochemistry on purified protein:
PMID:37527341
PMID:37527341
UniProt lists KM 0.33 mM for ATP and Vmax 198.3 nmol/min/mg (ECO:0000269|PubMed:37527341).
- Cryo-EM confirms a canonical ABC catalytic cycle (inward-facing apo → ATP-bound,
NBD-dimerized, outward-facing toward the DV lumen) and an unusual N-terminal
autoinhibitory element:
PMID:37527341
PMID:37527341
Transport activity and direction
The decisive functional work is the Xenopus oocyte system of Shafik et al.:
This matters for GO. PfMDR1 is not a plasma-membrane efflux pump; terms framing it as
cellular drug efflux (e.g. xenobiotic detoxification by transmembrane export across the
plasma membrane) would be wrong. Substrate range per UniProt: "Transports lumefantrine,
mefloquine, chloroquine, quinine, quinidine, amodiaquine, piperaquine, dihydroartemisinin
and quinacrine (PubMed:35507548)", plus human P-gp substrates rhodamine B and vinblastine,
and not amantadine.
Drug resistance — what the IDA reference actually shows
The GOA row derived from PMID:10706290 is GO:0009410 response to xenobiotic stimulus (IDA).
The experiment is allelic exchange changing drug susceptibility — i.e. it demonstrates that
the protein handles the drugs, not that the gene product executes a cellular
"response to a xenobiotic stimulus" program. The accurate process term for what PfMDR1
itself does is GO:0006855 xenobiotic transmembrane transport (verified via OLS: a
descendant of both GO:0042908 xenobiotic transport and GO:0055085 transmembrane transport).
- No animal/plant-only terms are present in the GOA. The IBA rows (GO:0016020,
GO:0042626, GO:0055085) come from a broad ABCB/MDR node (PANTHER:PTN000657189,
PTN002517083) with plant, fungal, nematode, fly and mammalian donors; the claims they
make (membrane, ATPase-coupled transport, transmembrane transport) are generic to the
family and correct for PfMDR1 — they are simply less specific than the direct evidence.
- Organelle checks: the DV/food vacuole is a real apicomplexan compartment and is
correctly used here; nothing is mis-assigned to apicoplast, rhoptry, microneme or the
parasitophorous vacuole.
- Host-vs-symbiont: no host-interaction process terms are asserted, and none are
warranted — PfMDR1 acts inside the parasite on the parasite's own DV membrane.
- Paralog over-annotation: P. falciparum has several ABC transporters (PfMDR2,
PfMRP1/2, PfABCI3 etc.); none of the annotations reviewed here transfer a paralog-specific
function onto PfMDR1. The EC 7.6.2.2 mapping (GO:0008559) is directly confirmed by
UniProt's experimental CATALYTIC ACTIVITY entry for this accession.
Knowledge gaps
- The endogenous (non-drug) substrate is unknown. PMID:35507548 Targeted searching found no published
natural substrate and no clean knockout; the drug substrates are almost certainly
adventitious cargo of a transporter whose physiological job is still undefined. PfCRT,
its DV-membrane neighbour, turned out to export haemoglobin-derived peptides
PMID:35507548, which is the obvious model to test for PfMDR1.
- Essentiality is asserted but not, in the sources read here, backed by a conditional
knockout: PMID:35507548
- Physiological trigger for the N-terminal R domain is unknown; unlike CFTR's R domain it
lacks phosphorylation sites, so what relieves autoinhibition in vivo is open.
Ontology gap noted
GO has no food vacuole membrane cellular-component term (OLS search returns nothing;
GO:0020020 has no children). A DV-membrane protein can currently only be annotated to the
generic GO:0005774 vacuolar membrane plus GO:0020020 food vacuole. Recorded as a
proposed_new_terms entry in the review.