br Migration of MIA cells and PSCs in the zebrafish
3.5. Migration of MIA NCT501 and PSCs in the zebrafish xenograft model
In vivo migration capacity of the in vitro co-cultured MIA cells and PSCs was evaluated by the zebrafish xenograft. Co-cultured cells from the TCPS (Fig. 6A) or CS-HA (Fig. 6B) gradually migrated and were dispersed in the embryos from 24 to 72 hpf, while the number of PSCs obviously decreased from 48 to 72 hpf. At 72 hpf, the migratory dis-tance of MIA cells derived from the spheroid (CS-HA) group was en-hanced by more than 150% as compared to that of the regular co-cul-ture (TCPS) group (Fig. 6C). In addition, a part of co-migrated MIA-PSC cells were observed and represented by the yellow dots shown in the merged fluorescent images of both groups. Meanwhile, we determined the migration capacity of the TCPS mono-cultured MIA cells in zebra-fish (Fig. S9, SI), where the distribution was close to that in the TCPS co-culture group. Regarding the survival of transplanted MIA cells in zebrafish, apparent cell reduction was only observed in the embryos injected with TCPS mono-cultured MIA cells (Fig. S9, SI), but not in either co-culture group (Fig. 6A). r> Fig. 4. The gene expression profile for cells cultured on TCPS (2D) or the CS-HA coated plates (3D) by quantitative RT-PCR. MIA cells and PSCs were seeded in diﬀerent ratios as indicated. The gene expression was determined at 48 h after seeding. The expression level was normalized to 18sRNA and expressed as the relative ratio to that of mono-cultured MIA cells on TCPS (2D). The gene expression of typical epithelial-mesenchymal transition (EMT) marker (E-cad, N-cad, Vimentin, TWIST1, and SNAIL1) and the cell mobility associated marker (MMP2) is shown in the panel (A). The gene expression of the extracellular matrix (ECM) markers COL1, LUMICAN, and SNED1 gene is shown in the panel (B). The gene expression of stemness phenotype markers OCT4 and CD44 is shown in the panel (C). The expression of the chemo-resistance related genes ABCC1 and DARPP-32 is shown in the panel (D). Asterisks indicate statistically significant diﬀerences, *, p < 0.05, **, p < 0.01, ***, p < 0.005, ****, p < 0.001.
Fig. 5. The eﬀect of combined drug, Abraxane (Abr) and gemcitabine (GEM), on the cell viability of pancreatic tumor-like spheroids. (A) The morphology of MIA cells and PSCs in a ratio 1:9 on CS-HA coated plates (3D). The cells were treated with GEM (100 nM), or GEM (100 nM) in combination with diﬀerent concentration of Abraxane for 48 h. Red arrows indicate the margin of the spheroids. Blue arrows represent where the spheroids started to lose the integrity. (B) Cell viability on the TCPS plates (2D) after the co-cultured MIA cells and PSCs (1:9) were exposed to GEM (100 nM) or GEM in combination with Abraxane for 48 h. (C) Cell viability on the CS-HA coated plates (3D) after spheroid formation (48 h) and treatment with the drug for 48 h. PSC spheroids or MIA:PSC (1:9) spheroids were exposed to GEM (100 nM) or GEM in combination with Abraxane on the CS-HA coated plates (3D) for 48 h. The cell viability was analyzed by CCK-8 assay. Each point indicate the mean ± SD of multiple determinations (n = 3–6). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
PDAC is composed of a majority of stromal cells that surround a minority of malignant cells. The specific tumor organization and unique cancer microenvironment play critical roles in failures of chemotherapy [33,34] and radiotherapy . To overcome the hurdle, one main task is to establish a PDAC-mimicking microenvironment in vitro that can recapitulate the structural organization of cancer cells and PSCs in vivo. Recently, a couple of 3D multicellular models mimicking tumors were reported. In one of the models, the pancreatic cancer cells (PANC-1), fibroblasts (MRC-5), and endothelial cells (HUVEC) were co-cultured on
the round-bottomed well plates . The MRC-5 cell line was derived from lung tissue but not from the pancreas. The culture process was complicated because the fibroblasts needed to be coated with 9 layers of fibronectin before use. The structure of tumor spheroids was diﬀerent from that in our study. In their spheroids, fibroblasts and endothelial cells were rich in core while tumor cells were on the outer layer . In another model, hepatocellular carcinoma (HCC) cells were co-cultured with human hepatic stellate cells (HSCs), lung fibroblast cells, or en-dothelial cells on the round-bottom ultra-low attachment microplates to form spheroids. The spheroids promoted HCC chemoresistance and migration phenotypes . Nevertheless, the tumor-like spheroids in