Statistical image analysis reveals features affecting fates of Myxococcus xanthus developmental aggregates

Starving Myxococcus xanthus bacteria use their motility systems to self-organize into multicellular fruiting bodies, large mounds in which cells differentiate into metabolically inert spores. Despite the identification of the genetic pathways required for aggregation and the use of microcinematography to observe aggregation dynamics in WT and mutant strains, a mechanistic understanding of aggregation is still incomplete. For example, it is not clear why some of the initial aggregates mature into fruiting bodies, whereas others disperse, merge, or split into two. Here, we develop high-throughput image quantification and statistical analysis methods to gain insight into M. xanthus developmental aggregation dynamics. A quantitative metric of features characterizing each aggregate is used to deduce the properties of the aggregates that are correlated with each fate. The analysis shows that small aggregate size but not neighbor-related parameters correlate with aggregate dispersal. Furthermore, close proximity is necessary but not sufficient for aggregate merging. Finally, splitting occurs for those aggregates that are unusually large and elongated. These observations place severe constraints on the underlying aggregation mechanisms and present strong evidence against the role of long-range morphogenic gradients or biased cell exchange in the dispersal, merging, or splitting of transient aggregates. This approach can be expanded and adapted to study self-organization in other cellular systems.