ESC-Track: A computer workflow for 4-D segmentation, tracking, lineage tracing and dynamic context analysis of ESCs

Laura Fernndez-de-Manel1, Covadonga Daz-Daz2, Daniel Jimnez-Carretero1, Miguel Torres2, and Mara C. Montoya1

1Cellomics Unit 2Cardiovascular Developmental Program, Cell & Developmental Biology Area, Centro Nacional de Investigaciones Cardiovasculares CNIC, Madrid, Spain

BioTechniques, Vol. 62, No. 5, May 2017, pp. 215222

Abstract

Embryonic stem cells (ESCs) can be established as permanent cell lines, and their potential to differentiate into adult tissues has led to widespread use for studying the mechanisms and dynamics of stem cell differentiation and exploring strategies for tissue repair. Imaging live ESCs during development is now feasible due to advances in optical imaging and engineering of genetically encoded fluorescent reporters; however, a major limitation is the low spatio-temporal resolution of long-term 3-D imaging required for generational and neighboring reconstructions. Here, we present the ESC-Track (ESC-T) workflow, which includes an automated cell and nuclear segmentation and tracking tool for 4-D (3-D + time) confocal image data sets as well as a manual editing tool for visual inspection and error correction. ESC-T automatically identifies cell divisions and membrane contacts for lineage tree and neighborhood reconstruction and computes quantitative features from individual cell entities, enabling analysis of fluorescence signal dynamics and tracking of cell morphology and motion. We use ESC-T to examine Myc intensity fluctuations in the context of mouse ESC (mESC) lineage and neighborhood relationships. ESC-T is a powerful tool for evaluation of the genealogical and microenvironmental cues that maintain ESC fitness.

Stem cells provide essential functions during embryonic development and tissue regeneration. Mouse embryonic stem cells (mESCs) are derived from pluripotent cells of the early mouse embryo and can be maintained as stable cell lines with a high self-renewal capacity. They provide a versatile in vitro model for understanding differentiation of human tissues, and their study has led to major advances in cell and developmental biology (1,2). A key challenge in the field is to understand the mechanisms involved in guiding stem cell fate (3-5), which have broad applications in biomedicine, from elucidating the causes of cancer to the use of stem cells in regenerative medicine. Thus, the biological properties of ESCs are a matter of great scientific, commercial, and medical interest.

ESC-Track (ESC-T) is a computational tool for automated segmentation and tracking of single mouse embryonic stem cells (mESCs) from live-cell 4-D confocal image data sets. The ESC-T workflow enables the extraction of parameters related to fluorescence signal localization and dynamics, cellular morphology, and cell motion for individual cells in the context of lineage and neighborhood relationships.

Optical imaging advances have led to the emergence of powerful live imaging tools with individual cell resolution in three-dimensional (3-D) space and in time (3-D + time or 4-D) (6,7). Moreover, a new generation of fluorescent proteins and dyes allows biochemical characterization of signaling pathways in intact living cells (8). Tagging by fluorescent proteins enables positional tracking of any given cell over time, which is easily achieved when the population of tagged cells is distributed among non-expressing cells by virtue of lineage or in experimental mosaics, but it becomes challenging when a fluorescent protein label is widely expressed (9). The ability to track and analyze live cells in time-lapse 4-D microscopy images is a matter of intense research (10,11) since visual inspection and analysis are insufficient to extract meaningful insights, making automated tracking and quantitative analysis of cells an absolute requirement. This is such a challenging task that several competitions have been carried out in order to evaluate cell segmentation and tracking algorithms (12,13). Computational tools are essential for extracting quantitative measurements from stem cell populations growing in 3-D physiological conditions and to translate the measurements into biological knowledge, allowing the study of a range of cell behaviors, such as motility, cell division, death, phagocytosis, etc. Most of these methods have been applied to Drosophila (14-18) and zebrafish (19-21) embryogenesis, or plant morphogenesis (22) studies. Of special relevance to the field of stem cell biology is the ability to integrate the cell behavior analysis with information about lineage (parentprogeny) and contextual (neighborhood) cellular relationships (9,11). In the last decade, several generic processing and tracking packages, such as Icy (23), Cell Profiler (24), tTt (25), qTfy (25), or the Fiji plugin TrackMate (26,27) have been reported. Some complex methods have been developed for specific applications, such as MARS (22), ACME (21), EDGE-4-D (17), and RACE (18) for particle (28), nuclear (16,29) or cellular (17,18,20-22) segmentation, and STARRYNITE (29), U-TRACK (28), ALT (22), EDGE-4-D (17), and TGMM (16) for tracking.

Here, we present a computational workflow that allows the automated segmentation and tracking of individual mESCs from live-cell 4-D confocal image data sets based on the combination of membrane and non-homogeneous nuclear signals, allowing lineage and neighborhood reconstruction. The workflow enables the extraction of parameters for fluorescence signal localization and dynamics, cellular morphological characteristics, and motion-related aspects of individual cells in the context of lineage and neighborhood relationships. ESC-T was used to study Myc dynamics in mESC cultures, and it proved to be a very valuable computational tool for stem cell research as it allowed the evaluation of genealogical and microenvironmental cues during mouse ESC culture in an unprecedented manner.

Automatic cell and nuclei segmentation and cell tracking. The proposed pipeline (Figure 1) uses images obtained from ESCs expressing tdTomato and GFP-MYC signals as described in the Supplementary Material. The pre-processing step consists of median filtering combining both nuclear and membrane signals (mycGFP median minus tdTomato median) (Figure 1, Steps 14) and is followed by application of a 2-D watershed segmentation algorithm, rendering 2-D sets (cell portions) (Figure 1, Step 5). Spatiotemporal (3-D + t) association rules based on the overlap of sets are applied to connect sets in 3-D space and time for automatic segmentation and tracking through the following pipeline:

View original post here:
ESC-Track: A computer workflow for 4-D segmentation, tracking, lineage tracing and dynamic context analysis of ESCs - BioTechniques.com