Poster R12 - Unveiling Hypoxia Patterns in a Breast Cancer Progression Model: Insights from multimodal chemical and structural imaging

Introduction: Hypoxia, characterized by oxygen deprivation, provides a selective advantage to Triple Negative Breast Cancer (TNBC) in solid tumors by influencing key metabolic, developmental, and survival pathways. To comprehensively study hypoxia, unbiased chemical imaging methodologies like Fourier Transform Infrared (FT-IR) spectroscopic imaging can prove useful. Combining IR imaging with label-free structural imaging such as quantitative phase imaging (QPI) can measure cellular biochemistry and morphology without dyes and stains, enhancing our understanding of changes under hypoxia. The presented study aims to use this combination to understand a model for TNBC by examining four isogenic cell lines under normal and hypoxic conditions, thereby mimicking spatial and temporal tissue conditions during cancer progression. 2D cell culture experiments showed both changes in morphology (widened cells, granularity) as well as chemical changes in protein, nucleic acid and lipid composition. Cultured in 3D, tumor cells were observed to displace benign cells to the core, where they perished, resulting in enriching the co-culture with more aggressive tumor cells over time.

Materials and

Methods: Cell culture: We used the MCF10AT series cells representing different stages of triple-negative breast cancer (TNBC)/basal-like progression: M1 (MCF10A), M2 (MCF10AT1k.cl2), M3 (MCF10Ca1h), and M4 (MCF10CA1a.cl1). Cells were cultured in 2D and exposed to various levels of physical hypoxia over 1 to 5 days to characterize hypoxia chemically and morphologically. Since 3D cell cultures are shown to be more representative of tumor heterogeneity and behavior in vivo, GFP-labeled M1 (GFP-M1) cells were co-cultured with M2, M3, orM4 cells in 1:1 ratio on ultra-low attachment plates to form spheroids of 200 μm allowing for a hypoxic gradient.

Chemical Characterization: IR imaging was employed to record full vibrational spectra at each pixel using the Perkin-Elmer Spotlight 400 imaging spectrometer, in reflection mode with a spatial pixel size of 6.25 µm and a spectral resolution of 4 cm-1, with 64 coadditions averaged for each pixel. By calculating the area under the curve (AUC) for significant bands within the IR spectra, we identified statistically significant bands that indicated chemical changes associated with hypoxia.

Morphological Characterization: QPI measurements were conducted by using Spatial light interference microscopy (SLIM) using a system constructed in-house under 20X magnification.

Spatial Characterization: Confocal fluorescence microscopy was conducted using Leica SP8 UV/Visible Laser Confocal Microscope under 40X magnification. Cross-sectional images of the spheroids were taken to visualize spatial rearrangement of the initially mixed cells to assess each cell line’s preference for either the proliferative surface or the hypoxic core.

Results, Conclusions, and Discussions: Results and discussion

Preliminary assessment: Our initial assessment of hypoxia treatment was done over 5 days, providing sufficient time for all changes to manifest. IR analysis revealed that hypoxia responses in MCF10AT cells peaked at 3 days, so subsequent analyses focused on this period.

IR imaging: In tumor cells (M3 and M4), the area for 1060 cm⁻¹ band (commonly assigned to nucleic acids) decreased with hypoxia, suggesting cell-cycle arrest and persisted cell traits. Moreover, new peaks at 1020 cm⁻¹, 1080 cm⁻¹, and 1150 cm⁻¹, linked to elevated glycogen, were noted in tumor cells. Conversely, M1 and M2 cells displayed spectral shifts indicative of DNA damage under hypoxia. Shifts in Amide I and II peaks are associated with secondary structural changes. In M1 and M2 cells increased α-helix content was observed while M4 cells exhibited more β-structure proteins. In contrast, M3 cells maintained consistent peak positions. Hypoxia-induced lipid accumulation was observed across all cell lines, most notably in M3 cells, indicating enhanced hypoxia adaptation.

QPI: Under hypoxia, both M1 and M2 cells were wider with smoother edges and reduced cytoplasmic extensions. M3 cells appear less widened, with the presence of high-phase-value granules adjacent to the nucleus, potentially indicative of lipid droplets. In contrast, M4 cells show minimal changes under hypoxic conditions.

3D co-culture: Observations revealed that M2 cells outcompete GFP-M1 cells in co-culture, pushing GFP-M1 cells into the hypoxic core where they mostly perish. In GFP-M1 and M3 co-culture, GFP-M1 cells were found to remain on the surface, while M3 cells, adapted to hypoxia, populate the core. In GFP-M1 and M4 co-culture, a heterogeneous structure formed, with M4 cells invading the GFP-M1 layer, forming branched structures possibly correlating with M4's invasive, metastatic nature.

Conclusion

The study utilized advanced imaging techniques to explore how TNBC progression responds to hypoxia at the molecular level. 2D mono-culture analysis with IR and QPI imaging identified key biomarkers and morphological changes. In 3D co-culture systems, the spatial arrangement of TNBC model cells within a tumor was observed. This integrated approach enables visualization of chemical variations during hypoxia within a tumor-like 3D environment.

Acknowledgements (Optional):