Page 15 - IJB-9-3

International Journal of Bioprinting Bioprinting of PDAC microtissues for drug screening

over 7 days of culture, and the pancreatic cancer cells were rich and stroma-poor models, the area of α-SMA-derived
embedded in the fibroblastic network, confirming a dense fluorescent signal was calculated and is shown in Figure 5B.
stromal microenvironment created in the prepared PDAC Obviously, after 7 days of culture, the areas of α-SMA
+
microtissues. fibroblasts in stroma-poor and stroma-rich models were
To characterize cell proliferation of different pancreatic significantly increased, which were approximately 7- and 5.7-
cancer models, we selected three time points: days 1, 4, and folds, respectively, larger than those at the beginning (day 1).
7 of culture and calculated the changes in microtissue area On the other hand, to distinguish cell types in the
in different models (Figure 4B). During the culture process, co-cultures and assess the amount of cancer cells, we
the area of tumor microtissue in different models increased stained cancer cells with CK19 (in red), considering that
due to spontaneous cell proliferation and aggregation. We CK19 is positively expressed in pancreatic cancer cells .
[33]
also noticed that cells in the mono-culture PDAC model We then measured the area of CK19 cells in three PDAC
+
proliferated slowly, and the microtissue gained a mere models at different time points (Figure 5C). As shown in
0.6-fold increase in area when cultured for 7 days. However, the results, there was no significant difference in the area of
for both co-culture models, cell proliferation rates were CK19 cells in the three models at the early stage of culture
+
significantly faster, especially for the stroma-rich PDAC (day 1). As co-culture period increased, we found that
model, and the area of microtissue was 3.5-fold higher than cancer cells in the stroma-rich model grew at the fastest
it was on day 1. The results showed that the fibroblasts in rate when compared to the other two groups.
the cancer microenvironment promote tumor growth, and
crosstalks between tumor and stroma may be involved. 3.5. Drug response in different 3D PDAC models
To further evaluate the density of the produced PDAC Stromal cells in the pancreatic tumor microenvironment as
microtissues, we calculated the ratio of the total cell area to well as the ECM are recognized as the important causes of
the entire hydrogel bead area (Figure 4C). After 7 days of increased tumor drug resistance [34-36] . Hence three different
cultivation, the density of the mono-tumor microtissue was PDAC models including stoma-rich, stoma-poor, and
not significantly increased, with the percentage increasing mono-tumor microtissues were dealt with gemcitabine
from 8.6 ± 0.5% at the beginning (day 1) to 13.6 ± 0.7%. at different concentrations, a standardized drug for the
[37]
However, the ratio of total cell area to GelMA hydrogel treatment of pancreatic cancer . Gemcitabine solutions of
bead area in the stroma-rich model was 21.9 ± 0.6% at day 50, 75, and 100 μM/mL were tested when three 3D PDAC
1 of culture, and the model achieved a high density up to microtissues have been cultured for 1 week. Meanwhile,
92.8 ± 1.6% after culture for 1 week. The results verified three kinds of PDAC microtissues treated with non-drug
that the cells in the stroma-rich model were more tightly medium and medium containing 0.1% dimethyl sulfoxide
connected to each other and formed a tight 3D fibroblastic (DMSO) were, respectively, analyzed as controls. In this
network. process, the produced PDAC microtissues were incubated
with drug medium for 72 h before cell viability was
3.4. Imaging of tumor-stroma interactions in PDAC determined. From the results (Figure 6A and Figure S2),
microtissues three PDAC microtissues treated with non-drug medium
After demonstrating the morphological and structural and medium containing 0.1% DMSO displayed good
advantages of the printed 3D PDAC microtissues, we cell states during the subsequent 72-h cultivation. While
then tried to investigate the tumor-stoma crosstalk within the mono-tumor microtissue showed poor resistance to
the models based on specific markers immunostaining. drug and cell death occurred at a drug concentration of
As reported, α-SMA is the most common biomarker 50 μM/mL, the co-culture models demonstrated significant
of CAFs [30,31] . We then evaluated α-SMA expression in cell death when the concentration of drug increased
co-culture models to explore whether the normal fibroblasts up to 75 μM/mL. Notably, the dead cells in the mono-
were activated over the co-culture period. Figure 5A showed tumor microtissue were the larger cell spheroids within
that there were fewer α-SMA cells in both co-culture the hydrogel beads, indicating that the drug molecules
+
models at day 1 of culture. However, the area of α-SMA diffused well in the GelMA hydrogel beads. While in both
+
cells significantly increased when cultured for 7 days. co-culture models, the dead cells were mainly distributed
Furthermore, the fibroblasts changed cellular morphology around the hydrogel beads, and this might be due to the
and acquired elongated spindle shapes. These features fibrous barrier provided by the fibroblasts that prevents
[38]
confirmed that NHDFs in both co-culture models were drug diffusion . On the other hand, over the period of
gradually activated and transformed to CAF-like phenotype drug incubation, the 3D fibroblastic network involved in
through the interactions with pancreatic cancer cells . For the co-culture models collapsed as the concentration of the
[32]
quantitative evaluation of α-SMA expression of stroma- drug increased.

Volume 9 Issue 3 (2023) 7 https://doi.org/10.18063/ijb.v9i3.676