PhD candidate McGill University Montreal, Quebec, Canada
Introduction:: In this study, we investigated the role of active and passive mechanisms of cancer cell migration in 3D cultures. In particular, we focused on the phenomenon of sedimentation, whereby cancer cells encapsulated in a hydrogel matrix migrate towards the bottom of the space. Cell migration is central to cancer growth and metastasis yet remains poorly understood. Significant insights into this phenomenon can be gained with the use of 3D culture techniques as they provide more realistic conditions in terms of tumor morphology, gene expression, and related biochemical processes than 2D cultures. Despite the fact that 3D cell culture experiments have yielded significant insights into cancer growth, the phenomenon of sedimentation remains incompletely understood. A more rigorous understanding of these mechanisms would enable us to interpret migration patterns in 3D cultures and relate them to in-vivo conditions. Our study employed a highly multidisciplinary approach combining 3D culture experiments, multiscale spatiotemporal modeling to investigate the mechanisms underlying the phenomenon of sedimentation.
Materials and Methods:: The mechanisms were classified into passive and active categories (see attached Figure). Passive mechanisms were considered to be gravitational force and compression of the hydrogel matrix. On the other hand, active mechanisms involved the use of the cellular migration machinery that moves a cell with respect to a stimulus.
The 3D cell culture experiments included mono-cultures of triple-negative breast cancer (TNBC) cells during the presence (different doses) and absence of the migrastatic drug Paclitaxel. The 3D scaffolds consisted of TNBC cell laden Matrigel semi-ellipsoids which were tracked for 15 days using confocal microscopy. In addition, we examined sedimentation in co-cultures of TNBC/cancer-associated fibroblasts, and TNBC mono-cultures with different adhesion properties of the surrounding environment. Furthermore, we performed an RNA-sequencing experiment which included 3D mono-cultures in presence and absence of Paclitaxel. The RNA-sequencing data were examined by performing differential gene expression analysis between untreated and Paclitaxel treated samples. Overall, the collected data consisted of 3D imaging samples and bulk RNA-sequencing samples taken at various time points and treatment conditions.
To further examine our hypothesis on the reason behind migration we designed a hybrid discrete-continuum mathematical model which was validated using a recently developed computational framework [1]. This framework utilized the continuum part of the model to perform calibration with the experimental data, and the discrete part to perform validation using spatial statistical analysis techniques, as well as relate the biophysical mechanisms to the spatial patterns.
Results, Conclusions, and Discussions:: Results
The Paclitaxel allowed us to inhibit any possible active migration, and isolate the effects of passive mechanisms such as gravity and compression of the hydrogel. Our results suggested that passive mechanisms are insufficient to bring the cells at the bottom. The transcriptomic analysis revealed evidence of active migration, and collective migration patterns regulated by TGF-β, MAPK, and Adherens Junctions pathways [2].
We mathematically examined the hypothesis of chemotactic migration based on self-generated signaling gradients produced by cells attached at the glass bottom. The calibrated model suggested a good agreement with the data yielding average errors < 40% for the control, and < 10% for the treatment data. The spatial analysis revealed clustered patterns that become more pronounced as treatment dose increases [3].
Further validation of our hypothesis was provided by the co-culture and the adhesion experiments. Both co-cultures and low adhesion experiments showed reduced downward migration of cancer cells. This, in turn, suggested that the signaling stimulated by the fibroblasts antagonized the signals stemming from the bottom. In addition, it indicated the significance of adhesion sites in migration [3].
Discussion
We investigated the mechanisms underlying sedimentation in 3D cell cultures. The computational framework allowed us to calibration and validation into the two scales of the model. In addition, it allowed to preserve the data-points for both calibration and validation. This technique will greatly benefit studies with limited amount of data – a typical case in biology.
Our results suggested that sedimentation is a result of active migration [3]. This behavior presents similarities to the 2D wound healing assays. However, the introduction of the 3rd dimension presents many advantages compared to the 2D cultures, as significant differences in the behavior of cells have been observed between them [4]. Therefore, the presented setup provides opportunities for investigating cell migration in more realistic conditions.
Conclusion
Overall, this study suggests that sedimentation is regulated by self-generated chematactic gradients and adhesion sites. Future extensions include the increase of the complexity of the tumor microenvironment by introducing different cell types, e.g. immune cells, as well as to investigate the relationship between drug resistance and migration.
Acknowledgements (Optional): :
References (Optional): : [1] Dimitriou, N. M., Flores-Torres, S., Kinsella, J. M., & Mitsis, G. D. (2022). Quantifying the Morphology and Mechanisms of Cancer Progression in 3D in-vitro environments: Integrating Experiments and Multiscale Models. IEEE Transactions on Biomedical Engineering. https://doi.org/10.1109/TBME.2022.3216231
[2] Friedl, P., & Mayor, R. (2017). Tuning collective cell migration by cell--cell junction regulation. Cold Spring Harbor Perspectives in Biology, 9(4), a029199.
[3] Dimitriou, N. M., Flores-Torres, S., Kyriakidou, M., Kinsella, J. M., & Mitsis, G. D. (2023). Cancer cell sedimentation in 3D cultures reveals active migration regulated by self-generated gradients and adhesion sites. BioRxiv. https://doi.org/10.1101/2023.02.15.528731
[4] Hickman, J. A., Graeser, R., de Hoogt, R., Vidic, S., Brito, C., Gutekunst, M., van der Kuip, H., & consortium, I. M. I. P. (2014). Three-dimensional models of cancer for pharmacology and cancer cell biology: Capturing tumor complexity in vitro/ex vivo. Biotechnology Journal, 9(9), 1115–1128. https://doi.org/10.1002/biot.201300492
