(E-184) High-throughput cell motility assay for systemic analysis of cancer metastasis

Introduction:: Cancer metastasis represents a major obstacle for achieving long-term remission in patients suffering from non-hematopoietic malignancy and is the leading cause of cancer-associated death [1]. A fundamental contributor to metastasis is cancer cell motility which is regulated by many components within the tumor microenvironment (TME) [2,3]. Though the effects of individual components on cell motility have been extensively characterized, it remains difficult to recapitulate the complexity of the entire system [4,5]. For systemic analysis, a high-throughput cell motility model would be needed to characterize multiple configurations of components. A better understanding of how these components jointly regulate cancer motility could lead to identification of molecular targets and therapeutic strategies to inhibit metastasis.

Conventional methods to assess cancer cell motility in vitro are built on 2D platforms which have limitations in terms of cost, reproducibility, and throughput [4]. A key limitation to throughput of cell motility assays is lack of automated analysis tools. Manual analysis is time consuming, making it difficult to test multiple conditions quickly. To overcome these limitations, we established a workflow that allows imaging real-time cancer cell motility in a 96 well-plate format using brightfield microscopy. Importantly, we established a machine-vision and deep learning analysis method to automatically track the migration process of individual cells from time-lapse videos without additional molecular labeling. This allows us to rapidly examine cancer cell motility behaviors from ~100 different conditions within a day as opposed to months using conventional methods.


Materials and Methods:: Breast cancer cells were first seeded at 1000-2000 cells/well in 96- well plates and incubated at cell culture conditions (37°C and 5% CO₂). Following incubation, various treatment conditions were added to each well accordingly. The plates were then placed on a microscope (Nikon TI) with an on-stage incubator to maintain cell culture conditions. Time-lapse brightfield images were acquired at 10-minute intervals over a minimum of 16 hours (4 positions per well; 384 positions per plate). The live nuclei in the brightfield images were detected using a custom deep learning model with DeepLabv3+ network architecture, and the trajectories of individual live nuclei were than tracked using the previously established tracking methods.


Results, Conclusions, and Discussions:: Our developed deep-learning model was able to detect cells within brightfield images with a high overall accuracy for various cell lines and densities. Using the proposed method, we can track ~4000 individual cells per condition across ~100 frames in a 16+ hour video. We demonstrated our assay by studying how breast cancer cells, MDA-MB-231 and MCF-7, respond to various cytokines at different concentrations. Our results show that EGF, TNF-α, IL-6, and HGF increase breast cancer cell motility which is consistent with previous findings. We further demonstrated our motility assay by applying it to a set of 100 FDA-approved compounds and identified a few novel motility regulators.

Our results demonstrate that the proposed deep learning-based workflow can accurately and effectively track cells within brightfield images in high-throughput without additional molecular labels. Our workflow can be used to identify novel motility regulator targets and provides necessary tools for systemic motility analysis.


Acknowledgements (Optional): :

References (Optional): :

1. 
   Siegel, R.L., Miller, K.D. & Jemal, A. Cancer Statistics, 2017. CA: a cancer journal for clinicians 67, 7-30 (2017).
   
   


2. 
   Ridley, A.J. et al. Cell migration: Integrating signals from front to back. Science 302, 1704-1709 (2003).
   
   


3. 
   Jimenez Valencia, A.M. et al. Collective cancer cell invasion induced by coordinated contractile stresses. Oncotarget 6, 43438-43451 (2015).
   
   


4. 
   Fraley, S.I. et al. A distinctive role for focal adhesion proteins in three-dimensional cell motility. Nat. Cell Biol. 12, 598-604 (2010).
   
   


5. 
   Fraley, S.I., Feng, Y., Giri, A., Longmore, G.D. & Wirtz, D. Dimensional and temporal controls of three-dimensional cell migration by zyxin and binding partners. Nature communications 3, 719 (2012).
   
   


6. 
   Arihiro, K., Oda, H., Kaneko, M., & Inai, K. (2000). Cytokines facilitate chemotactic motility of breast carcinoma cells. Breast Cancer, 7(3), 221–230.
   
   


7. 
   Wolczyk, D., Zaremba-Czogalla, M., Hryniewicz-Jankowska, A., Tabola, R., Grabowski, K., Sikorski, A. F., & Augoff, K. (2016). TNF-α promotes breast cancer cell migration and enhances the concentration of membrane-associated proteases in lipid rafts. Cellular Oncology, 39(4), 353–363.
   
   


8. 
   Lu, Z., Jiang, G., Blume-Jensen, P., & Hunter, T. (2001). Epidermal growth factor-induced tumor cell invasion and metastasis initiated by Dephosphorylation and downregulation of focal adhesion kinase. Molecular and Cellular Biology, 21(12), 4016–4031.