Environmental Toxin Biliatresone-Induced Biliary Atresia-like Abnormal Cilia and Bile Duct Cell Development of Human Liver Organoids

Abstract

Biliary atresia (BA) is a poorly understood and devastating obstructive bile duct disease of newborns. Biliatresone, a plant toxin, causes BA-like syndrome in some animals, but its relevance in humans is unknown. To validate the hypothesis that biliatresone exposure is a plausible BA disease mechanism in humans, we treated normal human liver organoids with biliatresone and addressed its adverse effects on organoid development, functions and cellular organization. The control organoids (without biliatresone) were well expanded and much bigger than biliatresone-treated organoids. Expression of the cholangiocyte marker CK19 was reduced, while the hepatocyte marker HFN4A was significantly elevated in biliatresone-treated organoids. ZO-1 (a tight junction marker) immunoreactivity was localized at the apical intercellular junctions in control organoids, while it was markedly reduced in biliatresone-treated organoids. Cytoskeleton F-actin was localized at the apical surface of the control organoids, but it was ectopically expressed at the apical and basal sides in biliatresone-treated organoids. Cholangiocytes of control organoids possess primary cilia and elicit cilia mechanosensory function. The number of ciliated cholangiocytes was reduced, and cilia mechanosensory function was hampered in biliatresone-treated organoids. In conclusion, biliatresone induces morphological and developmental changes in human liver organoids resembling those of our previously reported BA organoids, suggesting that environmental toxins could contribute to BA pathogenesis.

Keywords: biliary atresia; biliatresone; organoids.

Figure 1

Biliatresone-induced aberrant growth of human liver organoids. (A) Schematic diagram of the establishment of human liver organoids for biliatresone treatment. (B) Representative bright-field images; Hematoxylin & Eosin staining of sections of control and Day-5 biliatresone-treated organoids. (C) Growth curve of control and biliatresone-treated human liver organoids. The diameters of organoids (25 randomly chosen organoids were measured in each culture) were measured for the growth curve. *, a statistically significant difference between biliatresone-treated and untreated cultures (p < 0.05). Error bars indicate standard deviation.

Figure 2

Biliatresone-induced hepatocytic differentiation of human liver organoids. (A) Representative images of control and Day-2 biliatresone-treated human liver organoids stained for HNF4A (green). Nuclei were stained with DAPI (blue). The percentage of HNF4A-positive cells (white arrow) in control and Day-2 biliatresone-treated organoids was determined by counting the total number of cells and HNF4A+ cells of the organoids. The number of organoids counted in each group = 3; *, p < 0.05, Student’s t-test; error bars indicated the standard deviation. (B) Representative images of control and Day-2 biliatresone-treated human liver organoids stained for MDR1 (red). Nuclei were stained with DAPI (blue). (C) MDR1 activity assay of organoids. (C) Representative images demonstrating the fluorescent substrate Rhodamine 123 (R123) localization in the lumen of control organoids at 30 min. In contrast, luminal R123 localization of Day-2 biliatresone-treated organoids was minimal during the 30-min incubation. The fluorescent intensity along the white line of images of control and Day-2 biliatresone-treated organoids at 30 min of incubation were plotted.

Figure 3

Biliatresone-induced cytoskeleton defects in human liver organoids. (A) Immunostaining of control and Day-2 biliatresone-treated human liver organoids for cytoskeleton protein F-actin (green). Nuclei were stained with DAPI (blue). The number of organoids examined in each group was 20. (B) (Left panel). Representative images demonstrating the accumulation of FITC-dextran localization in the lumen of Day-2 biliatresone-treated human liver organoids but not of control organoids at 10 and 30 min of incubation. (Right panel), the fluorescent intensity along the white line of images of control and Day-2 biliatresone-treated organoids at 10 and 30 min of incubation were plotted. The number of organoids examined in each group was 10.

Figure 4

Biliatresone decreased the number of ciliated cholangiocytes in human organoids. Representative images of control and Day-1, -2, -3 and -5 biliatresone-treated human liver organoids stained for pericentrin (PCNT; green) and acetylated α-tubulin (Tubulin; red). Nuclei were stained with DAPI (blue). Boxed regions were magnified and shown as insets. The percentages of ciliated cells (white arrow) in control and biliatresone-treated organoids were determined by counting the total number of cells and ciliated cells of the organoids. The number of organoids counted in each group was 20.

Figure 5

Reduction in cholangiocyte cilia mechanosensory response in biliatresone-treated organoids. (A) Representative images of control and biliatresone-treated human cholangiocytes stained with Calbryte 520 after application of shear flow with HBSS at 1 dyne/cm². (B) Plot profile of fluorescent intensity changes of selected control and biliatresone-treated human cholangiocytes (highlighted with dotted line and numbered 1, 2 and 3) for 20 s. (C) The percentages of evoked cells after perfusion stimulation of control (Ctrl) and biliatresone-treated (BTS) human cholangiocytes were determined by counting the total number of cells and stimulated cells (green) at three randomly chosen fields of the cultures. *, p < 0.05, Student’s t-test; error bars indicated the standard deviation.