A high-content image-based method for quantitatively studying context-dependent cell population dynamics

Abstract

Tumor progression results from a complex interplay between cellular heterogeneity, treatment response, microenvironment and heterocellular interactions. Existing approaches to characterize this interplay suffer from an inability to distinguish between multiple cell types, often lack environmental context, and are unable to perform multiplex phenotypic profiling of cell populations. Here we present a high-throughput platform for characterizing, with single-cell resolution, the dynamic phenotypic responses (i.e. morphology changes, proliferation, apoptosis) of heterogeneous cell populations both during standard growth and in response to multiple, co-occurring selective pressures. The speed of this platform enables a thorough investigation of the impacts of diverse selective pressures including genetic alterations, therapeutic interventions, heterocellular components and microenvironmental factors. The platform has been applied to both 2D and 3D culture systems and readily distinguishes between (1) cytotoxic versus cytostatic cellular responses; and (2) changes in morphological features over time and in response to perturbation. These important features can directly influence tumor evolution and clinical outcome. Our image-based approach provides a deeper insight into the cellular dynamics and heterogeneity of tumors (or other complex systems), with reduced reagents and time, offering advantages over traditional biological assays.

Computational investigation of biological and technical variability in high throughput phenotyping and cell line identification

“…To simulate these practices, we choose intended observation times 0 < T1 < T2 < T3; three observations allows us to distinguish between exponential growth (log(N) vs. time appears linear) and growth near the confluent limit (log(N) vs. time appears sub-linear). Typically for cancer cell culture experiments, T1 ~ 1-2 days, giving cells time to adhere to the plate and re-enter the cell cycle (i.e., to start measuring past the "lag" phase), and the measurements are on the order of 1 day apart [1,2,18]. To simulate timing variability (e.g., an experimentalist needed to wait to access an instrument), we choose T1 * ~ N(T1, 1 hr), T2 * ~ N(T2, 1 hr), and T3 * ~ N(T3, 1 hr).…”

“…As high-throughput experimental systems advance, novel experiments are using them to measure key cell behaviors (e.g., migration, proliferation, death, and metabolic activity) across a broad range of microenvironmental conditions (e.g., [1][2][3][4][5]). To handle these multicellular phenotypic data and interface them with statistical, mathematical, and computational models, projects such as MultiCellDS [6], PharmML [7], and the Cell Behavior Ontology [8] are developing the consistent data models necessary to analyze, compare, and share experimental data.…”

PhysiCell: an Open Source Physics-Based Cell Simulator for 3-D Multicellular Systems

“…29 Timothy was recently developed to simulate large-scale colonies of 10 9 cells on high- 30 performance supercomputers. [12,13] Most of these platforms offer a general-purpose 31 pre-compiled "client" that can load models and settings from an XML file; this helps 32 overcome difficulties stemming from complex dependencies. See the supplementary 33 materials for a detailed software comparison.…”

“…(See Cell 401 cycling.) We used a simpler "live cells" cycle model [14] Hanging drop spheroids (HDS)-a 3-D cell culture model where a small cluster or 422 aggregate of tumor cells is suspended in a drop of growth medium by surface tension-are 423 increasingly used to approximate 3-D in vivo growth conditions [31]. Unlike traditional 424 2-D monolayer experiments, HDSs allow scientists to investigate the impact of substrate 425 gradients on tumor growth, particularly oxygen gradients.…”

The emergence of dynamic phenotyping

“…Experiments could be carried out under varying temperature and dynamic media composition (e.g., using microfluidic platforms). Recent experimental advances have enabled the measurement of singlecell dynamics in complex microenvironments, such as co-culture (Garvey et al 2016) and tissue explants (Lande-Diner et al 2015) and can advance our understanding of circadian phenotypes under more realistic conditions.…”