Characterization, modification and applicability of an in vitro placental transfer assay for screening purposes P92 Caroline Gomes*; Catharina W. van Dongen; Barbara Birk; Eric Fabian; Julian Doersam; Bennard van Ravenzwaay; Robert Landsiedel Experimental Toxicology and Ecology, BASF SE *Contact: caroline.a.gomes@basf.com Ø Alternative methods to animal testing are being developed to address developmental toxicity and the placental transfer is essential to determine the potential embryo toxic effects of a substance. Ø The objective of this study was to improve and characterize the methodology of this model, and to assess the placental transfer of 7 substances to assess the model applicability. Introduction Figure 1. Schematic overview of in vitro BeWo transfer assay. Ø An in vitro placental transfer assay using the human trophoblastic cell line, BeWo b30 clone, has been developed using a transwell system (Figure 1) and showed good correlations with ex vivo data 1,2 Methodology BeWo cells b30 clone (Addexbio, USA) were cultivated in the “regular condition” which refers to the methodology described in literature 1 . An optimization was tested to avoid refreshing the cell culture medium every day (Figure 2). The parameters used to check the barrier integrity, cell growth or cytotoxicity of the control substances are summarized in Figure 2. The Papp values were calculated by the equation: !"## (%&/() = ∆,/∆- .∗0 1 where ΔQ is the amount of compound entered in the basolateral compartment (nmol), Δt is the time (seconds) of the transfer experiment, A is the cell surface area in cm 2 of the insert and C0 the exposure concentration (µM). Figure 2. Methodology scheme of the in vitro placental transfer using the BeWo b30 cell line. The parameters assessed overtime or at the testing transfer day (day 6) are numbered from 1 to 4. Fluorescein was measured by spectrophotometer. Amoxicillin and antipyrine were determined by LC-MS analysis at Pharmacelsus GmbH, we are grateful to Dr. Ursula Müller-Vieira. TEER: trans-epithelial electrical resistance. WPI: World Precision Instruments (USA). Results: characterization and optimization Figure 3. TEER measurements of BeWo b30 cell layer from day 3 to day 7 after seeding (A). Fluorescein transfer in the in vitro placental transfer model from day 3 to 7 after seeding (B). Data are shown as mean ± standard deviation of triplicates per experiment. Substance Ex vivo placental transfer index In vitro placental transfer Relative Papp Papp (10 -6 cm/s) Recovery (%) Acyclovir 0.32 7 0.41 18 ± 1.2 128 Caffeine 0.95 8 1.34 59 ± 11 103 Cimetidine 0.46 9 0.30 13 ± 0.9 109 0 10 20 30 40 50 60 70 2 3 4 5 6 7 8 TEER (Ω·cm 2 ) Days (post-seeding) 0 2 4 6 8 10 12 14 16 18 20 2 4 6 8 Papp value (x10 -6 cm/s) Days (post-seeding) Run 1 - Regular Run 1 - Optimized Run 2 - Regular Run 2 Optimized Run 3 - Optimized 15 25 35 45 55 65 Papp value (x10 -6 cm/s) Antipyrine Li et al., 2013 (1) Li et al., 2015 (3) Li et al., 2016 (4) Kloet et al., 2015 (5) Strikwold et al., 2017 (6) Run 1 - Regular Run 1 - Optimized Run 2 - Regular Run 2 - Optimized Run 3 - Optimized Results: model applicability Methodology The applicability of the model was assessed by: a) Comparing the in vitro relative Papp values of 3 substances (caffeine, acyclovir and cimetidine) with data from the ex vivo human placental transfer assay in the optimized condition b) Assessing the in vitro placental transfer of 4 azoles (difenoconazole, flusilazole, miconazole and triadimefon) in the regular condition R² = 0,9913 R² = 0,5991 R² = 0,9945 R² = 0,9742 0 2 4 6 8 10 0 20 40 60 80 100 Basolateral amount (nmol) Time (minutes) Flusilazole Miconazole Difenoconazole Triadimefon 0 5 10 Papp value (x10 -6 cm/s) Amoxicillin Regular Optimized Whole insert Diverse layers Uniform layers Figure 4. Papp values of the permeability controls, Antipyrine and Amoxicillin, determined at day 6 (at 60 minutes) in the in vitro placental transfer model compared to literature. Data are shown as mean ± standard deviation of triplicates. The substances recoveries were higher than 90%. Figure 5. Photomicrograph of transwell insert with BeWo b30 cells at day 1 expressing tight junction protein ZO-1 (green, immunostaining). Nuclei staining (blue) with Hoechst 33342. Magnification of 400x. Figure 6. Photomicrographs of transwell inserts with BeWo b30 cells at day 6. Bar scales of 2.5 µm for the whole insert pictures and 50 µm for the others. Supported by Maike Huisinga et al., BASF SE. A B Conclusion and future perspectives Ø The permeability controls in the optimized model were in accordance with literature data. Moreover, the optimized method was a less time-consuming methodology. The assay is robust and reproducible. The histology of the cell barrier showed that cell multilayers expressing tight junctions are the structure of the cell barrier in both experimental conditions. Ø Data from this in vitro placental transfer model were in the same order of magnitude as the data from the ex vivo human perfusion model, corroborating the good in vitro and ex vivo correlation previous shown 1,2 . Ø The results from the azoles demonstrate that even compounds from the same chemical class can have different transfer rates. Ø This model can be combined with other in vitro or in silico methods as part of an animal-free risk-assessment approach to address developmental toxicity. Further characterization in respect of the effects of active transporters in this model remains to be assessed. Table 2. Papp values determined after 60 minutes of exposure to the 4 azoles. Data from two independent biological experiments (n=3) presented as mean  SD. Data published by Dimopoulou et al. 2018 (10). Substance Run 1 Run 2 Papp value (10 -6 cm/s) Recovery (%) Papp value (10 -6 cm/s) Recovery (%) Difenoconazole 11 ± 1.1 45 8 ± 0.7 35 Flusilazole 26 ± 2.0 63 22 ± 2.4 58 Miconazole 1.3 ± 0.7 14 0.5 ± 0.3 11 Triadimefon 42 ± 4.5 89 31 ± 5.5 71 Figure 7. In vitro placental transfer of 4 azoles into the basolateral compartment. Data from two run experiments (n=3) presented as mean  SD. Miconazole showed no linear transfer. Table 1. In vitro and ex vivo placental transfer comparison. Data from one experiment (n=3) presented as mean  SD. 23"4(563 74869 = (:;,=>?@>ABCD E ∗:?,:BEF@GHFBI E) :;,:BEF@GHFBI E ∗:?,=>?@>ABCD E when Af is amount of substance in the fetal compartment, t is sampling time, and Am is the amount in the maternal compartment. R6K"L7M6 !"## = NO@@ =>?@>ABC D NO@@ :BEF@GHFBI 1 Li et al., 2013. Archives of Toxicology. 2 Poulsen et al., 2009. Toxicology in vitro. 3 Li et al., 2015. Toxicology in Vitro. 4 Li et al., 2016. Archives of Toxicology. 5 Kloet et al., 2015. Toxicology In Vitro. 6 Strikwold et al., 2017. Archives of Toxicology. 7 Henderson et al., 1992, J Lab Clin Med. 8 Mose et al., 2008, J Toxicol Environ Health A. 9 Ching et al., 1987, JPET. 10 Dimopoulou et al. 2018, Toxicol Lett.