Bile Acid Conjugation on Solid Nanoparticles Enhances ASBT-Mediated Endocytosis and Chylomicron Pathway but Weakens the Transcytosis by Inducing Transport Flow in a Cellular Negative Feedback Loop

Abstract

Bile acid-modified nanoparticles provide a convenient strategy to improve oral bioavailability of poorly permeable drugs by exploiting specific interactions with bile acid transporters. However, the underlying mechanisms are unknown, especially considering the different absorption sites of free bile acids (ileum) and digested fat molecules from bile acid-emulsified fat droplets (duodenum). Here, glycocholic acid (GCA)-conjugated polystyrene nanoparticles (GCPNs) are synthesized and their transport in Caco-2 cell models is studied. GCA conjugation enhances the uptake by interactions with apical sodium-dependent bile acid transporter (ASBT). A new pathway correlated with both ASBT and chylomicron pathways is identified. Meanwhile, the higher uptake of GCPNs does not lead to higher transcytosis to the same degree compared with unmodified nanoparticles (CPNs). The pharmacological and genomics study confirm that GCA conjugation changes the endocytosis mechanisms and downregulates the cellular response to the transport at gene levels, which works as a negative feedback loop and explains the higher cellular retention of GCPNs. These findings offer a solid foundation in the bile acid-based nanomedicine design, with utilizing advantages of the ASBT-mediated uptake, as well as inspiration to take comprehensive consideration of the cellular response with more developed technologies.

Keywords: apical sodium dependent bile acid transporter (ASBT); bile acid-modified nanoparticles; chylomicron pathway; transcytosis; transport feedback.

Figure 1

a) CLSM analysis of cellular uptake of CPN and GCPN in Caco‐2 cells in 0.5 h. GCPN demonstrated much higher uptake efficiency. b) Competitive effect of free GCA on the uptake of CPN and GCPN. Free GCA inhibited the uptake of GCPN by ≈40%, while slightly inducing the CPN uptake. **, p < 0.01.

Figure 2

Endocytosis mechanisms of CPN and GCPN in Caco‐2 cells. Effects of a) temperature and b) pharmacological inhibitors on the endocytosis of CPN and GCPN in Caco‐2 cells. Endocytosis of both NPs was reduced at lower temperatures. Nystatin, genistein, dynasore, and CytD inhibited the uptake of both CPN and GCPN, indicating the participation of caveolae, dynamin, and actin filaments in the endocytosis. CPZ inhibited CPN uptake but not GCPN, suggesting only CPNs were internalized by clathrin‐mediated pathway. c–e) RT‐PCR analysis of expression of c) Flot‐1, d) Arf6, and e) RhoA after knocking down by siRNA. All of the mRNA expression decreased by over 80% compared with nonspecific siRNA (siRNA nc). f) Effect of Flot‐1, Arf6, and RhoA on the endocytosis of CPN and GCPN. Only knocking down of Arf6 showed inhibition of GCPN uptake. g) Effect of Cdc42 on the endocytosis of CPN and GCPN. ZCL278 induced the uptake of CPN but did not influence GCPN. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.

Figure 3

CLSM analysis of ASBT distribution and colocalization with CPN and GCPN at 2, 8, and 24 h in Caco‐2 cells: a) X–Y images; b) X–Z images. CPN did not colocalize with ASBT and showed little influence on its distribution. GCPN colocalized with ASBT at 2 h and gradually separated from it in 8 h. ASBT was internalized into the cytoplasm in 8 h and recycled back to the cell membrane in 24 h.

Figure 4

Colocalization images and Pearson's coefficient of CPN and GCPN (red) with ER, Golgi apparatus, late endosomes, and lysosomes (green) in Caco‐2 cells. The white square parts were shown in higher magnification. Both CPN and GCPN showed colocalization with ER and Golgi (higher Pearson's coefficient) but little with Rab 7 (late endosomes) and lysosomes. Compared with CPN, GCPN demonstrated less involvement in the maturation process.

Figure 5

a) Colocalization of CPN/GCPN (red) with IBABP (green) in Caco‐2 cells. The white arrow indicates colocalization. b) RT‐PCR analysis of IBABP expression after CPN and GCPN treatment.

Figure 6

a) TEER of Caco‐2 monolayers with the treatment of chitosan, CPN, and GCPN. No significant TEER change occurred with the treatment of CPN and GCPN in 24 h. b) Cumulative permeation of Na‐Flu across the Caco‐2 monolayers with the treatment of chitosan, CPN, and GCPN. CPN and GCPN did not influence Na‐Flu permeation. c) Quantitative analysis of transcytosis of CPN and GCPN across the Caco‐2 monolayers. d–f) Effect of brefeldin, bafilomycin, and 17α‐ethinylestradiol on the transcytosis of CPN and GCPN across the Caco‐2 monolayers. Brefeldin reduced the transcytosis of CPN and GCPN, while bafilomycin and 17α‐ethinylestradiol did not disturb their transport. *, p < 0.05; ***, p < 0.001.

Figure 7

Genomics study of CPN and GCPN transport in Caco‐2 cells. a) Compared with control group, numbers of up‐ and downregulated genes after treatment with CPN and GCPN in 24 h. b) Compared with CPN group, number of up‐ and downregulated genes in GCPN group. c) PCA plot of the sample distribution. d) Sample Euclidean distance among the control, CPN, and GCPN. e) Heat map of the top 484 up‐ and downregulated genes in GCPN versus CPN. f–h) The statistical ORT based on the GO classification between GCPN versus CPN. The p‐values of the listed genes classes were all <0.05. i) Top upregulated (red) and downregulated (blue) genes in the corresponding pathways.

Figure 8

Schematic illustration of transport pathway regulation effect of GCA conjugation in Caco‐2 cells. Both NPs transported through the monolayer by transcytosis rather than paracellular pathway. Endocytosis of GCPN was mediated by ASBT and thus demonstrated stronger efficacy, which was assisted by caveolae. The uptake of CPN was nonspecific and mediated by clathrin and caveolae. After internalization, ASBT and GCPN were separated, with GCPN recycled to the membrane while GCPNs were bound to IBABP. Neither the two NPs was largely transported to lysosomes, and GCPN exhibited even less involvement in acidification. Both NPs transported through the ER‐Golgi route, while GCPN demonstrated lower efficacy in this pathway. After ER‐Golgi packaging, GCPN shared the chylomicron pathway and entered the lacteal after exocytosis. Neither of the NPs was transported via OSTα/β. With GCA conjugation, the cells responded to the GCPN transport flow in a feedback mode, with cytoskeleton‐related pathway upregulated, while most transport progresses including the acidification, endocytosis, intracellular transport, and exocytosis were downregulated, which finally formed into a negative feedback loop.