Predicting the Dynamics and Heterogeneity of Genomic DNA Content within Bacterial Populations across Variable Growth Regimes

• Published in: ACS synthetic biology 2017-07, Vol.6 (7), p.1131-1139
• Main Authors: du Lac, Melchior, Scarpelli, Andrew H, Bates, Declan G, Leonard, Joshua N
• Format: Article
• Language: English
• Subjects: Bacteria - genetics; DNA - genetics; Escherichia coli - genetics; Gene Dosage - genetics; Genomics; Synthetic Biology - methods
• ISSN: 2161-5063; 2161-5063
• DOI: 10.1021/acssynbio.5b00217

For many applications in microbial synthetic biology, optimizing a desired function requires careful tuning of the degree to which various genes are expressed. One challenge for predicting such effects or interpreting typical characterization experiments is that in bacteria such as E. coli, genome copy number varies widely across different phases and rates of growth, which also impacts how and when genes are expressed from different loci. While such phenomena are relatively well-understood at a mechanistic level, our quantitative understanding of such processes is essentially limited to ideal exponential growth. In contrast, common experimental phenomena such as growth on heterogeneous media, metabolic adaptation, and oxygen restriction all cause substantial deviations from ideal exponential growth, particularly as cultures approach the higher densities at which industrial biomanufacturing and even routine screening experiments are conducted. To meet the need for predicting and explaining how gene dosage impacts cellular functions outside of exponential growth, we here report a novel modeling strategy that leverages agent-based simulation and high performance computing to robustly predict the dynamics and heterogeneity of genomic DNA content within bacterial populations across variable growth regimes. We show that by feeding routine experimental data, such as optical density time series, into our heterogeneous multiphasic growth simulator, we can predict genomic DNA distributions over a range of nonexponential growth conditions. This modeling strategy provides an important advance in the ability of synthetic biologists to evaluate the role of genomic DNA content and heterogeneity in affecting the performance of existing or engineered microbial functions.