Introduction
Pancreatic cancer remains a formidable challenge, with 3D cell culture offering a more physiologically relevant model than traditional 2D systems. This study compares how two common scaffold-free 3D platforms— Poly(2-hydroxyethyl methacrylate) (Poly-HEMA or PH) coated plates and ultra-low attachment (ULA) plates—influence spheroid formation, drug sensitivity to gemcitabine, and invasion dynamics in three pancreatic cancer (PCa) cell lines: PANC-1, SU.86.86, and BxPC-3. The work highlights platform-dependent differences in morphology, proliferation, and adhesion signaling that can drive distinct phenotypes and responses to therapy.
Spheroid Morphology and Basal Proliferation Across Platforms
All three PCa cell lines formed spheroids in both PH and ULA systems, but their architectures varied. PH-coated surfaces produced smaller, less cohesive spheroids, whereas ULA plates yielded larger, more compact structures. Basal ATP-based viability after five days was consistently lower in ULA-cultured cells, suggesting reduced metabolic activity under non-adhesive conditions. This physical reorganization underpins downstream differences in drug response and invasion patterns.
Gemcitabine Response Is Platform-Dependent
Gemcitabine treatment (0.125–4000 μg/mL for 48 h) revealed clear platform-driven differences. PANC-1 and SU.86.86 spheroids showed modest to moderate viability changes across platforms, with PH-cultured PANC-1 spheroids being slightly more sensitive at the highest dose, while SU.86.86 on ULA plates demonstrated notable gemcitabine resistance. BxPC-3 was less variable and excluded from the most divergent comparisons. Annexin-V staining confirmed apoptotic activity, while PH-cultured spheroids tended to disrupt at high drug doses, a contrast to the more intact ULA spheroids that persisted under gemcitabine challenge. These findings align with broader evidence that 3D environments frequently confer greater chemoresistance than 2D cultures, with platform-specific nuances shaping the extent of resistance.
Invasion Potential and Adhesion Marker Expression
Invasion assays showed pronounced platform effects. PANC-1 spheroids largely remained enclosed in ECM with limited degradation, though PH-derived groups exhibited minor invasion. SU.86.86 spheroids displayed the most striking platform-dependent invasion: PH-derived spheres favored single-cell invasion, while ULA-derived spheroids caused broader ECM degradation, indicating a shift from single-cell to more collective or matrix-altering invasion modes depending on the culture surface.
Adhesion molecules mirrored these functional differences. In PANC-1, E-Cadherin protein was higher in ULA cultures despite lower mRNA, suggesting post-transcriptional regulation. N-Cadherin and MMP-7 showed platform-specific shifts, with higher N-Cadherin mRNA in PH cultures and varied MMP-7 expression contributing to different invasion programs. SU.86.86 cells showed a more complex pattern: some markers favored PH at the protein level, while transcript levels rose in ULA conditions, highlighting post-transcriptional control and proteostasis as key modulators of phenotype.
Quantitative immunofluorescence supported these trends. E-Cadherin and Integrin α1 intensities were elevated in ULA spheroids for PANC-1, reinforcing epithelial-like traits and stronger cell–ECM adhesion in this platform. SU.86.86 displayed subtler shifts, with PH-driven samples showing modest increases in E-Cadherin and Integrin α1 signals. The data underscore how platform context can redefine adhesion landscapes and, consequently, invasion strategies.
<h2Interpretation and Implications
These findings reveal that culture platform—not just cell line—shapes spheroid architecture, drug resistance, and invasion modes in pancreatic cancer models. ULA conditions favor cohesive, mechanically resilient spheroids with greater gemcitabine resistance, aligning with epithelial-like phenotypes. In contrast, PH surfaces promote smaller, more migratory spheroids and single-cell invasion tendencies, particularly in SU.86.86, suggesting EMT-associated behavior can be captured more readily on these substrates. Importantly, mRNA and protein levels do not always align, highlighting the need for integrated transcript–protein analyses to translate mechanistic insights into actionable targets.
<h2Practical takeaways for researchers
- Choose the culture platform based on the research question: ULA is advantageous for epithelial-like, chemoresistant phenotypes; PH is informative for EMT-related invasion patterns.
- Expect platform-dependent discrepancies between mRNA and protein levels for adhesion/invasion markers; multi-omics validation is recommended.
- Invasion assessments should consider both single-cell and collective behaviors, as platform context governs the dominant mode.
Limitations and Future Directions
The study analyzed two PCa lines in depth, which may not capture full disease heterogeneity. Broader omics profiling and inclusion of stromal/immune components could deepen mechanistic understanding and improve translatability to in vivo contexts.
