Myotubes, which are multi-nucleated fibers, are formed through the fusion of myoblasts. In cell culture, differentiated myoblasts that are produced following ceasing their growth align with each other and fuse into a myotube. The efficiency of these processes, i.e. myogenesis, thus affects the rate of myotube formation. Quantitative analysis of myotubes formation is, however, often difficult as the various size of myotubes that contain a variety of the numbers of nuclei are formed over time. Therefore, we developed a single-cell tracking approach to quantitatively evaluate the process of myotube formation.
The single-cell tracking analysis is performed by generating live-cell imaging videos to record behaviors of myoblasts, tracking of individual myoblasts recorded in the videos, and generate lineage data to characterize individual myoblasts. As this analysis can obtain spatiotemporal information of individual myoblasts, e.g. motility of myoblasts, events occurred in a myoblast, and interaction of a myoblast with other myoblasts, it has been considered as a powerful cell biological research technique. However single-cell tracking analysis has not frequently been used, as the analysis was extremely laborious. We thus developed a computerized single-cell lineage tracking analysis method (microscope and software) that creates fully automated live-cell imaging videos (1–4-week-long observations); performs image segmentation (an imaging technique to extract information from a video) and automatic cell tracking and cellular event identification; creates a database, and performs data analysis. We used near-infrared differential interference contrast imaging to minimize phototoxicity, and to compare the behavior of a group of cells and control cells, we simultaneously observed multiple conditions and created as many as 16 videos. The development of automated software to handle images increased the analytical capacity to characterize individual cells or cell populations through statistical analysis. We used this technique to investigate the process of myotube formation of primary myoblasts isolated from wile type, galectin 3 -/- (galectin 3: a muscle lectin, which promotes myogenesis) and mdx mice.
We video-recorded the process of myotube formation using myoblasts isolated from wile type, galectin 3 -/- and mdx mice. Myogenesis of myoblasts isolated from wile type mice was initiated by replacing growth medium to a differentiation medium, and after 20 hrs of cell culture, alignment of myoblasts occurred, and some myoblasts started to fuse, forming bi-nuclear myoblasts. These bi-nuclear myoblasts were either fused with another bi- and multi-nuclear myoblast or fused with a mono-nuclear myoblast, resulting in the formation of myoblasts with the various number of nuclei. Multi-nuclear myoblasts were then fused with each other, forming large myotubes with multiple nuclei as large as ones have more than 0.5 mm diameter. On the other hand, myoblasts isolated from galectin 3 -/- mice were more motile compared with wiled type myoblasts. Although the alignment of myoblasts occurred after 20 hrs of cell culture, these myoblasts had reduced the chance to fuse due to the high motility. Thus, the formation of multi-nuclear myoblast was delayed compared with that of wild type myoblasts, resulting in the formation of smaller myotubes. Myoblasts isolated from mdx mice were also highly motile. However, different from myoblasts isolated from galectin 3 -/- mice, alignment of myoblasts was hindered possible due to the weak attachment of myoblasts to the culture surface that was coated with matrigel, resulting in the poor formation of myotubes. We used these videos to trace-back the process of myotube formation by individual cell tracking, as such tracing back allowed to convert the video imaging data into quantitative values, which include the motility of myoblasts, the rate of myoblast fusion and the myotube formation. These results suggest that live-cell imaging videos, which were generated by simultaneous monitoring of myogenesis of myoblasts isolated from wild, galectin-3 -/- and mdx mice, highlight the difference of myogenesis of these myoblasts.
During the establishment of a single-cell tracking approach, we developed various image processing software, one of which can be used to characterize muscle fibers stained with hematoxylin and eosin (HE). For example, although a fluorescent image is often used to determine the size of muscle fibers, HE stained images are also useful for the determination, as it allows to evaluate the status of fibers, e.g. the presence or absence of ice crystals in the fibers. To use the HE stained images for the characterization, we developed a color picker that extracts a certain range of color from the image, and an image segmentation and quantitation tools. These tools allow characterizing muscles containing 3000-5000 fibers. We will introduce these tools together with the single-cell tracking system.
In conclusion, we developed a live-cell imaging microscope and software that can be used to analyze the process of myogenesis. As the analysis allows to quantitatively evaluate the spatiotemporal change of individual myoblasts, we believe that the approach can be used to investigate the response of myoblasts to various drugs for the treatment of diseases caused by an abnormality of muscle.