Gene regulatory architecture in antibiotic resistance is optimized for specific dynamic regimens

Cellular responses to antibiotics present a fundamental trade-off: the need for rapid resistance deployment to prevent drug accumulation, while resistance gene expression is costly in its absence. Consequently, antibiotic responses are regulated, determining not just the expression of resistance but also its timing, intensity, and duration.

This regulation is achieved through the regulation of promoters and operators' strength and organization. However, it remains unclear how drug exposure dynamics select for particular regulatory strategies. To address this, a dynamical model of a typical antibiotic response was created, involving a transcriptional repressor sensing the drug and controlling resistance protein expression, while considering the drug's impact on cell growth and gene expression.

A cost function was devised to weigh the various strains on the cell during the response, and in silico optimization of regulatory parameters was performed under diverse selective pressures for three increasing complexity architectures: constitutive resistance expression, repression by a constitutively expressed repressor, and self-regulation of the repressor.

Strong drug exposures favored constitutive resistance expression, high basal expression costs promoted repression by a constitutive repressor, and self-regulation of the repressor emerged only when protection was required while curbing overexpression of resistance. Stochastic simulations revealed that self-regulation of the repressor shields cells from fluctuations in repressor levels, preventing unnecessary resistance induction prior to exposure and resistance induction failure post-exposure.

This framework connects environmental dynamics to the evolution of gene regulation, offering a means to predict which regulatory strategies arise under specific selective pressures.