• 2019-10
  • 2019-11
  • 2020-03
  • 2020-07
  • 2020-08
  • 2021-03
  • NCT-501 br Fig Migration capacity of the implanted MIA


    Fig. 6. Migration capacity of the implanted MIA NCT-501 and PSCs in zebrafish. MIA cells and PSCs were co-cultured for 48 h, and injected in zebrafish embryos at 4 hpf. Cell migration in the representative embryos was tracked from 24 to 72 hpf. MIA-PSC cells derived from TCPS and CS-HA substrates are respectively shown in panels (A) and (B). BF, bright field. Scale bar, 200 μm. (C) The average migration distance of MIA cells in each zebrafish was determined at 72 hpf as illustrated in the left panel and summarized in the right panel. Asterisks indicate statistically significant differences, **, p < 0.01.
    these two models were generated on round-bottom microplates that did not mimic the tumor HA-rich microenvironment.
    Chitosan and hyaluronan are common natural products with good biocompatibility and many biomedical applications [38,39]. Recently, we have demonstrated that lung cancer cells or lung cancer cells co-cultured with mesenchymal stem cells (MSCs) rapidly self-assembled 
    into 3D tumor spheroids on CS-HA substrates [29,40]. The MSC-tumor spheroids had core-shell structure and showed an up-regulation in tu-morigenic gene expression [40]. These previous works were on lung tumor that had different morphology (without stroma). The colon cancer cell lines HT29 formed 3D tumor spheroids on CS-HA substrates. However, hepatocellular carcinoma Huh7 only aggregated to form
    colonies on CS-HA membrane [41]. In the current study, PSCs stroma co-cultured with pancreatic cancer cells MIA showed unique core-shell features on CS-HA plates but not on the other polymers (Fig. 1). CS-HA coated plates are suitable for generation of PDAC tumor-spheroids.
    The tumor microenvironment for PDAC is HA-rich, which is a un-ique feature. Comparing with other cancers in human, HA is more abundant in the stromal compartment of the PDAC, and the abundance of HA is an indicator for the lower prognosis [24,27]. One of the major advantages of CS-HA plates is the HA component employed in the fabrication of CS-HA plates that can mimic the tumor microenviron-ment. Second, the CS-HA plates are easy to produce and no special procedure is required for cell culture. Third, CS-HA plates are trans-parent and it is possible to visualize spheroid formation and movement on the flat well, which is not feasible on the round bottom well. The number and size of the spheroids in the CS-HA plates can be modified and manipulated in large scale for drug screening.
    The commonly used agarose cannot replace CS in this method be-cause we found pancreatic cancer cells only fully interact with PSCs when the CS and HA are present in the system. Therefore, we believe HA plays an important role in the microenvironment. HA is known for its strong expression in 80% of primary PDAC tissues, combined with positive staining being detected both in tumor and stromal components [42,43]. Moreover, HA expression was independent of prognostic factor in patients undergoing resection for PDAC [24,44]. Therefore, we be-lieve that 3D culturing of pancreatic cancer cells and PSCs on the CS-HA plates is most likely to mimic the microenvironment of pancreatic cancer.
    HA is a linear biopolymer with repeating disaccharide units of N-acetylglucosamine and D-glucuronic acid, which are linked by glyco-sidic linkages. The carboxyl group of HA forms a stable covalent bond to the amine group of chitosan via a carbodiimide chemistry. In our previous study, the glucuronic acid assay was used to measure the conjugation between CS and HA (the surface amount of conjugated HA) on the surface [45]. The procedure consistently proved HA-grafted chitosan with a very limited release (< 2%) to the culture medium in 28 days [45]. Moreover, the surface contact angle (hydrophobicity) decreased and the surface charge became more negative with in-creasing the amount of grafted HA. These measurements demonstrated the different surface characteristics of the substrates [45].
    CS-HA is more similar to the ECM composition of PDAC than CS and PVA substrates. The process of spheroid formation involves both cell adhesion and migration. The cell adhesion molecule N-cadherin plays a key role in 3D spheroid formation of MSCs [46]. CS-HA compared to CS and PVA could promote more N-cadherin expression in adipose-derived adult stem cells [47]. In our study, even though co-cultured cells on CS-HA showed the greater tendency to form PDAC tumor-like spheroids, there was no obvious up-regulation of N-cadherin protein expression for co-cultured cells compared to mono-cultured PSCs on CS-HA (Fig. S10, SI). This trend was consistent with that of the mRNA expression (Fig. 4A). The protein lysate of the MIA-PSC group was from both tumor cells and PSCs, not PSCs alone. The strong cell-cell interaction among PSCs may give rise to high N-cadherin expression in the mono-cultured PSCs. Meanwhile, N-cadherin expression might be affected by the het-erotypic cell-cell interaction between MIA and PSCs. Taken together, these results suggested that the interaction between tumor cells and PSCs might influence the N-cadherin expression.