However, whereas this medium allowed Capan 2 cell proliferation in monolayer culture, it was not able to sustain Capan 2 cell growth in spheroid in 96 well plates. Consequently, different growth media composition were evaluated and we found that defined DMEM F12 medium supplemented with EGF and B27 induced AZD1152-HQPA Capan 2 spheroid growth up to 16 fold between day 1 and day 10. Determination of cell viability by measurement of cell ATP content confirmed that Capan 2 spheroids grown faster in the defined medium. Intraand inter assay precision of spheroid volume and ATP measurement was found to be suitable to ensure robust pharmacological studies . To confirm the dependence on EGF, Capan 2 spheroids were cultured in defined medium supplemented with EGF.
Four days later, EGF was washed out and Capan 2 spheroids were maintained in LY404039 10 serum. In this condition, we observed that Capan 2 spheroid growth was inhibited . The spheroid internal structure depends on a nutrient and oxygen gradient which controls a decreasing gradient of cell proliferation from the periphery to the center of spheroid. A central necrotic area is generally observed in spheroids larger than 500 m due to critical O2 concentration in the central zone. We determined the repartition of proliferative and apoptotic cells in Capan 2 spheroids of various sizes cultured in defined medium supplemented with EGF and B27. Formalinfixed tissue teck embedded Capan 2 spheroid sections were immuno stained for the proliferation and apoptotic markers Ki 67 and cleaved PARP respectively.
We found that proliferative and non proliferative cells were distributed throughout the 400 m size Capan 2 spheroid and a gradient of proliferation appears on spheroid measuring 600 m and more in diameter. While apoptosis was not detected in 400 m spheroids, apoptotic cells were observed in the center of the spheroid of larger diameters. Consequently, this model allows the investigation of drug response taking into account cell heterogeneity. Resistance to gemcitabine treatment of the Capan 2 spheroid model Considering increase in spheroid size, change in proliferation gradient and the occurrence of a necrotic core, we applied cytotoxic treatment between days 4 and 7, thus avoiding overlapping effects. Indeed, we did not observe significant difference in gemcitabine EC50 between 4, 5, 6 and 7 days spheroids .
As a consequence we cultured spheroids for four days before treatment as this protocol is compatible with automated HTS application. We first compared the effect of gemcitabine on Capan 2 cells growing as monolayer and as spheroid. Figure 3 shows the effect of different gemcitabine concentrations on spheroid culture compared to the monolayer culture. We observed that a 3 day treatment with gemcitabine exerted a similar efficiency but gemcitabine potency was found to be much higher in monolayer culture compared to spheroids indicating that gemcitabine effect could be correlated to multicellular growth condition. To evaluate if this resistance is linked to the presence of quiescent cells in the Capan 2 spheroid, we tested the response to gemcitabine treatment of quiescent spheroids. Capan 2 spheroid need for EGF was used to induce a quiescent state. As a