| Cell-to-cell variability is a critical issue in biomedical science and plays a role in drug resistance of bacteria and cancer, as well as in the formation of biofilms and differentiation of stem cells. Despite its broad relevance, quantitatively understanding cellular heterogeneity from its sources to its consequences is still an open challenge. The chemotaxis system of the bacterium Escherichia coli is an ideal model for such investigations since it is very well characterized genetically and biochemically. Chemotaxis, which is cellular movement toward favorable chemical conditions, is an important behavior across microbiology and human physiology, but the impact of cell-to-cell heterogeneity on chemotaxis is mostly unexplored.;Despite sharing the same genetically-encoded machinery, individual E. coli cells exhibit substantial differences in chemotactic behaviors. Here, we characterize the consequences of this behavioral heterogeneity for population performance by using modeling, simulations, and experiments. We find that there exist trade-offs in which a single chemotactic behavior cannot perform optimally in all environmental challenges. We show that, in some cases, optimal populations can benefit from non-genetic diversity to provide specialists for different tasks. We demonstrate that this is mechanistically possible through mutations in genetic regulatory sequences that alter the distribution of protein levels in the population. We therefore propose a framework for analyzing the sources, consequences, and adaptability of non-genetic diversity wherein noisy distributions of proteins give rise to heterogeneous phenotypes, which in turn give rise to differential performance and ultimately collective fitness. |
