The transition to an era of functional glycomics has created a need for the development of novel techniques to assess complex glycoconjugate cellular adhesion molecules (e.g., selectin ligands) that may be involved in capturing and mediating the adhesion circulating tumor cells to distant sites. Cell line, knockout, and selectin antagonist evidence suggests that selectin/selectin-ligand interactions are involved in metastasis. However, identification and characterization of functional selectin ligand expression on cancer tissue (i.e., CTC origin site) remains challenging due to the inherent heterogeneity of tissues magnified by the "catch-slip" and stochastic nature of selectin/selectin-ligand bonds. We have developed a flow-based assay known as dynamic biochemical tissue analysis (DBTA) that involves the perfusion of selectin-coated microprobes over tissues at physiologic wall shear rates. Using DBTA, the expression of functional selectin ligands has been detected (i.e., specific binding and rolling of selectin-coated microprobes) on tissue from multiple solid tumors at primary and metastatic sites. Although DBTA provides evidence that multiple types of cancer can express complicated glycoconjugate structures (selectin ligands) that are able to bind to selectins in a shear-dependent manner, the challenge of the assay is in the analysis of the large amounts of image sequence data generated. This analysis is further complicated by the dynamic behavior of selectin-coated microprobe adhesion to selectin ligands expressed on a heterogeneous cancer tissue surface. To extract the biophysical underpinnings towards the development of this prospective in vitro diagnostic, a mathematical model was devised to refine the design space of this flow assay. The model is semi-empirical and consists of a coupled set of stochastic, 4D partial differential equations that account for the transport of probes to the tissue surface as well as the dynamics of probe adsorption to a heterogeneous tissue surface. Empirical data were obtained by processing image sequences with machine-learning, template-matching algorithms to conduct image segmentation, followed by probe tracking over time using a program written in Python. The empirical data displayed a stochastic adsorption profile and showed agreement with the model equations that were solved numerically using a finite difference scheme written in C++. Altogether, this approach is able to successfully bridge the gap between the seemingly random rolling interactions observed in the flow assay and the underlying physical mechanisms.
Citation Format: Eric Martin, Douglas Goetz, Monica Burdick. Modeling selectin-coated microprobe adhesion to selectin ligands heterogeneously expressed on cancer tissues [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4277.