Why do we need computer-assisted motion analysis systems to understand how cells locomote? It is now clear that every mechanism that cells have evolved for locomotion is complex, and that nonquantitative, anecdotal descriptions are no longer sufficient for answering the questions now being posed to elucidate these mechanisms. This point has been brought home in a particularly poignant fashion in recent years in the analysis of cytoskeletal mutants. In order for animal cells to crawl, they must extend pseudopods driven by actin polymerization. Although actin alone will polymerize in vitro, more that fifty proteins interact with actin in vivo during the extension of a pseudopod. Biochemical experiments have suggested roles for many of these proteins, such as capping, cross-linking, sequestering, force generation, localization and membrane attachment. However, when null mutants of several of these proteins were generated in the model amoeboid cell Dictyostelium discoideum, the mutant cells still extended pseudopods and crawled. These results were interpreted to mean either that there was functional redundancy (i.e., two or more proteins had evolved for each function) or that many of these actin-associated proteins served no role in locomotion. However, when computer-assisted methods were used to examine the behavior of a number of these mutants, neither of the above two explanations proved correct. In each case, computer-assisted analysis revealed a very specific behavioral defect , suggesting that each actin-associated protein plays a specific role in fine-tuning the process of cell locomotion. In most cases, the motility defects of a mutant cell could only have been identified by the quantitative data provided by computer analysis, and this was most evident in the abnormal 3D behavior of pseudopods formed by ponticulin-minus cells ).

Interpreting the locomotion of animal cells qualitatively, therefore, can not provide the resolution necessary to assess abnormalities in the complex and highly regulated processes of pseudopod extension and retraction, especially when a cell is responding to extracellular cues, as in the process of chemotaxis . In amoebae and white blood cells, a number of rules govern the localization and 3D dynamics of pseudopod extension and retraction, turning, the maintenance of cellular polarity and the behavioral responses to extracellular signals. Because most cells move slowly and a large portion of their behavior is three dimensional, simply staring at the 2D images of normal and abnormal cells in most cases will not reveal behavioral differences. The abnormal formation by a mutant cell of pseudopods at twice the frequency of normal cells and the resulting reduction in velocity and chemotactic efficiency can not be readily distinguished by qualitative, real time observations. One must apply computer-assisted technologies. The W.M. Keck Dynamic Image Analysis Facility was established to develop and provide access to these technologies.