Halofuginone
Halofuginone. A febrifugine derivative that inhibits the prolyl-tRNA synthetase domain of human glutamyl-prolyl-tRNA synthetase. PMID:22327401 verbatim: "Here we show…
A febrifugine derivative that inhibits the prolyl-tRNA synthetase domain of human glutamyl-prolyl-tRNA synthetase. PMID:22327401 verbatim: "Here we show that HF binds glutamyl-prolyl-tRNA synthetase (EPRS), inhibiting prolyl-tRNA synthetase activity; this inhibition is reversed by the addition of exogenous proline or EPRS." The target is the human enzyme — PMID:24100331 verbatim, "Halofuginone (HF), a coccidiostat used in veterinary medicine, exerts its effects by acting as a high-affinity inhibitor of the enzyme glutamyl-prolyl-tRNA synthetase (EPRS)". Blocking prolyl-tRNA charging triggers the amino-acid response, which is the proposed basis of its anti-fibrotic and TH17-suppressive activity; the anti-fibrotic arm is described as "Inhibition of collagen synthesis by halofuginone is achieved by inhibiting transforming growth factor beta-dependent Smad3 phosphorylation" (PMID:12869955, a review).
THE EPRS AFFINITY IS NOW AUTHORED as a receptor_occupancy row; see that row for the verbatim quote and the arithmetic. IT IS AN AFFINITY WITHOUT A TIME COURSE, deliberately: this record has no authored PK, so no Ce(t) is ever computed and pd.needs-solvable-pk says so. An earlier pass could not author it at all, because receptor.needs-keo was a flat error. The rule was narrowed instead (see 4007e83): it still errors where a route IS solvable and the curve is computable. NO keo IS AUTHORED HERE AND AN ESTIMATE WOULD BE EMPTY, not merely soft: approximated keos are permitted without a source_pmid and 196 records carry one, but every such estimate is anchored to an observed time-to-peak effect. This record has neither a PD time course nor any plasma curve for a keo to describe the approach to. PubChem free base CID 456390 (C16H17BrClN3O3, 414.7), re-queried 2026-09-26. THE SALT MATTERS: halofuginone hydrobromide is CID 11591339 (C16H18Br2ClN3O3, 495.6), and the veterinary material and much "halofuginone hydrobromide" dosing is the salt, so any molar work must first establish which species a dose refers to (HYGIENE R10). The stored mass is the free base. THE ORAL ROUTE IS STATED, not inferred: the phase I studied halofuginone "when administered once or twice daily orally to patients with advanced solid tumours" (PMID:16815702). DOSES are verbatim from the phase I (PMID:16815702): "Patients were treated with escalating doses of halofuginone at doses ranging from 0.5 to 3.5 mg/day", "The 'acute' MTD was reached at 3.5 mg/day, with nausea, vomiting, and fatigue as DLT", and "The recommended dose for phase II studies of halofuginone is 0.5mg administered orally, once daily" — which is why typical is 0.5 and not the midpoint of the escalation range. ANIMAL PK, RECORDED BUT DELIBERATELY NOT AUTHORED AS A CURVE. Cattle, oral 1.2 mg/kg repeated at 48 h (PMID:2590824): "The apparent terminal elimination half-life ranged from 24.2 to 28.9 h with a harmonic mean of 27.3 h" — 27.3 h is the paper's own harmonic mean, a published point value, not a midpoint. Mouse and rat (PMID:11761455): mouse IV 1.5 mg/kg gave "an AUC of 19,874 ng/ml min, which corresponded to a total body clearance (CLtb) of 75 ml/min per kg", and critically "The bioavailability of HF after i.p. and oral delivery to mice was 100% and 0%, respectively". A 0% oral F in mouse against a drug given orally to humans is a species discrepancy large enough that carrying any animal parameter onto this record would mislead, which is why the record stays uncharacterized — the R8 dofetilide lesson. THE TOPICAL ROUTE IS SUPPORTED BY A NEGATIVE: PMID:10609960, one patient, topical ointment for cutaneous cGVHD, "No adverse effects were observed, and no plasma levels of halofuginone could be detected." THE TARGET CONSTANT CAME FROM FULL TEXT, NOT AN ABSTRACT: no abstract states a Ki, Kd or IC50 for halofuginone against EPRS, which is why this cell sat open for two passes. PMID:24100331 says only "high-affinity", qualitatively — the ixekizumab defect shape in R2. Numerals that must NOT be imported: PMID:36394909's IC50 0.18 microM and Kd 30.3 nM belong to its designed compound 3, not to halofuginone; PMID:31821989's IC50 1.4 microM is compound 30d against Salmonella ThrRS. See the GAPS rows. THE PHASE I FULL TEXT IS UNREACHABLE, AND THE CHASE IS NOW CLOSED RATHER THAN OPEN. Every route was tried on 2026-09-27: no PMC record; Europe PMC reports isOpenAccess N with the paywalled DOI as its only full-text URL and mines the ABSTRACT SECTION ONLY; the Elsevier article API is reachable but returns AUTHENTICATION_ERROR for this key on every view including META, and also on an OPEN-ACCESS control article, so the refusal is the key’s account entitlement rather than this paper’s paywall; Unpaywall, OpenAlex and CORE are outside this environment’s egress policy. ALL 42 PMC ARTICLES CITING THE PHASE I were fetched and searched and NOT ONE quotes its parameters. Fifteen further human-flagged halofuginone studies state no PK value either. What would unlock it: institutional Elsevier entitlement, or the numbers quoted in some future paper.
Sources: PMID:16815702, PMID:22327401, PMID:24100331, PMID:2590824, PMID:11761455, PMID:12869955