br Funding This work was supported by the

Funding This work was supported by the Deutsche Forschungsgemeinschaft (SFB766 and GRK1708).
Conflict of interest
Acknowledgements
Introduction Hepatitis C virus (HCV) infection is a global problem affecting about 3% of the world’s population [1], [2]. Mother-to-child transmission of HCV is the major route of the infection in children occurring in 5–10% [3], [4] and up to 20% in HIV co-infected pregnant women [5], [6]. High proportions (73–92%) of vertically infected children suffer from chronic disease [3]; liver biopsies typically show liver inflammation and fibrosis [3]. Multiple studies have suggested an association between risk of transmission and HCV viral loads [7]. Therefore, treatments that decrease HCV viremia in pregnant women are expected to reduce rates of vertical HCV transmission [8]. Ribavirin, a purine analogue with broad-spectrum anti-viral activity [9], [10], is a WHO essential medicine for adults and children [11], [12] that is used (inter alia) as the backbone of various HCV therapeutic regimens [10], [13]. Due to its mechanism of action (interference with biosynthesis of guanine nucleotides) and teratogenicity observed in animal studies [14], ribavirin has been assigned to FDA Pregnancy Category X [15]. However, in humans, single and series case reports have documented normal pregnancies with no birth defects [14], [16], [17], [18], [19], [20], [21] unless concomitant teratogen was administered [22]. Importantly, preliminary findings obtained from an interim analysis of potential teratogenicity at the mid-point of enrolment (ClinicalTrials.gov identifier: NCT00114712) “do not suggest a clear signal of human teratogenicity for ribavirin” [15]. However, knowledge of ribavirin’s placental transfer mechanisms is also important for robust evaluation of the safety of its use in pregnancy [23], [24]. Ribavirin is a highly hydrophilic drug (log P −1.85), reaching maximal plasmatic concentrations (Cmax) in healthy volunteers of 2.6 µM after a 400 mg single oral dose. It accumulates strongly in the plasma, reaching 4-fold Cmax at steady-state [25], and is metabolised via two pathways with no participation of cytochrome P450 enzymes: reversible phosphorylation or deribosylation and amide hydrolysis. Following a single dose, it has a relatively short plasma half-life (2 h), but its active metabolite ribavirin triphosphate is eliminated slowly, with a plasma half-life of 120–170 h [14], [25]. Ribavirin does not bind plasma proteins and is extensively distributed (Vd = 4500–6000 L), especially to Piroxicam of skeletal muscles, liver, and erythrocytes, indicating involvement of specific membrane transporters [25]. In vitro and in vivo experiments have indicated that ribavirin may cross the placenta, and/or distribute into milk in humans [26], [27], but these possibilities have not been unequivocally demonstrated [28]. Pharmacokinetics of nucleoside-derived antiviral drugs are frequently modulated by activities of nucleoside transporters (NTs) and ATP-binding cassette (ABC) efflux pumps [27], [29], [30], [31], [32]. There are two subfamilies of NTs: equilibrative (ENTs, mediating facilitated diffusion) and concentrative (CNTs, mediating Na+-dependent active transport) [27], [33], [34], [35], [36]. ENTs are sensitive to S-(4-nitrobenzyl)-6-thioinosine (NBMPR). NBMPR is considered a specific ENTs inhibitor [37]: at a concentration of 0.1 µM NBMPR inhibits ENT1/Ent1 selectively, while at 100 µM NBMPR blocks activities of both ENT1/Ent1 and ENT2/Ent2 [38], [39]. Up to date, no selective inhibitor of CNTs has been identified. As cellular uptake of nucleosides is believed to be mediated predominantly by nucleoside transporters [30], [40], inhibitory effect of Na+ depletion on nucleosides cellular uptake has been established as sufficient evidence of involvement of CNTs in nucleosides membrane transport [27], [30], [32], [33], [35], [36], [37], [40], [41], [42], [43], [44], [45]. Both ENTs and CNTs reportedly control maternal-to-foetal transfer of nucleosides and nucleoside-derived drugs [27], [31], [32], [33], [34]. ABC transporters are a vast superfamily of proteins that participate in diverse processes. Potentially important ABC transporters in the context of this study include p-glycoprotein (ABCB1), breast cancer resistance protein (ABCG2), and multidrug resistance-associated protein 2 (ABCC2), all localized in apical membrane of human placental syncytiotrophoblast. These efflux pumps protect the developing foetus against potentially harmful xenobiotics by limiting transfer of their substrates from maternal to foetal circulation [46], [47]. It has been suggested that ribavirin may be a substrate of human ENT1 [27], [44], [48], [49], CNT2 [50] and CNT3 [27], [44], [51], and mouse Ent1 [52]. In vitro studies have indicated that ENT1 and CNT3 are important for placental uptake of ribavirin [27] and subsequent analysis showed that Ent1 is required for transfer of ribavirin into foetal circulation in mice [26]. However, possible roles of NTs in ribavirin pharmacokinetics in the human placenta have not been directly studied. Moreover, the possibility that ABCB1, ABCG2, or ABCC2 may interact with ribavirin and reduce its placental maternal-to-foetal transfer rates has not been rigorously tested in previous investigations.