Microbial biocontainment is an essential goal for engineering safe, next-generation living therapeutics. However, the genetic stability of biocontainment circuits, including kill switches, is a challenge that must be addressed. Kill switches are among the most difficult circuits to maintain due to the strong selection pressure they impart, leading to high potential for evolution of escape mutant populations. Here we engineer two CRISPR-based kill switches in the probiotic Escherichia coli Nissle 1917, a single-input chemical-responsive switch and a 2-input chemical- and temperature-responsive switch. We employ parallel strategies to address kill switch stability, including functional redundancy within the circuit, modulation of the SOS response, antibiotic-independent plasmid maintenance, and provision of intra-niche competition by a closely related strain. We demonstrate that strains harboring either kill switch can be selectively and efficiently killed inside the murine gut, while strains harboring the 2-input switch are additionally killed upon excretion. Leveraging redundant strategies, we demonstrate robust biocontainment of our kill switch strains and provide a template for future kill switch development.

The composite members of the microbiota face a range of selective pressures and must adapt to persist in the host. We highlight recent work characterizing the evolution and transfer of genetic information across nested scales of host-associated microbiota, which enable resilience to biotic and abiotic perturbations. At the strain level, we consider the preservation and diversification of adaptive information in progeny lineages. At the community level, we consider genetic exchange between distinct microbes in the ecosystem. Finally, we frame microbiomes as open systems subject to acquisition of novel information from foreign ecosystems through invasion by outsider microbes.

Graphical Abstract Highlights d Carbohydrate availability in the gut drives E. coli Nissle adaptation in vivo d Gut monocolonization selects for glycosyl hydrolases enabling population cross-feeding d Mutations that enhance mucin utilization are enriched in lowdiversity guts d Prior antibiotic exposure in conventional guts can lead to evolved probiotic resistance In Brief E. coli Nissle is a probiotic and chassis for engineered biotherapies, but its adaptive behavior in the gut is unclear. Crook et al. report host-mediated selective pressures modulating carbohydrate utilization and metabolism of E. coli Nissle. This in-host evolution also promotes probiotic survival by enabling effective stress responses during colonization. SUMMARYProbiotics are living microorganisms that are increasingly used as gastrointestinal therapeutics by virtue of their innate or engineered genetic function. Unlike abiotic therapeutics, probiotics can replicate in their intended site, subjecting their genomes and therapeutic properties to natural selection. We exposed the candidate probiotic E. coli Nissle (EcN) to the mouse gastrointestinal tract over several weeks, systematically altering the diet and background microbiota complexity. In-transit EcN accumulates genetic mutations that modulate carbohydrate utilization, stress response, and adhesion to gain competitive fitness, while previous exposure to antibiotics reveals an acquisition of resistance. We then leveraged these insights to generate an EcN strain that shows therapeutic efficacy in a mouse model of phenylketonuria and found that it was genetically stable over 1 week, thereby validating EcN's utility as a chassis for engineering. Collectively, we demonstrate a generalizable pipeline that can be applied to other probiotics to better understand their safety and engineering potential. ing of GFP-Tagged EcN, B.W.; Engineering of EcN:PAL2 Strains, Z.C.; Creation, Sharing, and Guidance on Pah enu2 -Mouse Model, S.D.

Poultry contaminated with Salmonella enterica subsp. enterica are a major cause of zoonotic foodborne gastroenteritis. Salmonella Heidelberg is a common serotype of Salmonella that has been implicated as a foodborne pathogen associated with the consumption of improperly prepared chicken. To better understand the effectiveness of common antimicrobial disinfectants (i.e., peroxyacetic acid [PAA], acidified hypochlorite [aCH], and cetylpyridinium chloride [CPC]), environmental isolates of nontyphoidal Salmonella were exposed to these agents under temperature, concentration, and contact time conditions consistent with poultry processing. Under simulated processing conditions (i.e., chiller tank and dipping stations), the bacteriostatic and bactericidal effects of each disinfectant were assessed against biofilm and planktonic cultures of each organism in a disinfectant challenge. Log reductions, planktonic MICs, and mean biofilm eradication concentrations were computed. The biofilms of each Salmonella isolate were more resistant to the disinfectants than were their planktonic counterparts. Although PAA was bacteriostatic and bactericidal against the biofilm and planktonic Salmonella isolates tested at concentrations up to 64 times the concentrations commonly used in a chiller tank during poultry processing, aCH was ineffective against the same isolates under identical conditions. At the simulated 8-s dipping station, CPC was bacteriostatic against all seven and bactericidal against six of the seven Salmonella isolates in their biofilm forms at concentrations within the regulatory range. These results indicate that at the current contact times and concentrations, aCH and PAA are not effective against these Salmonella isolates in their biofilm state. The use of CPC should be considered as a tool for controlling Salmonella biofilms in poultry processing environments.

The development of robust engineered probiotic therapies demands accurate knowledge of genetic construct expression in the gut. However, the monetary and ethical costs of testing engineered strains in vertebrate hosts are incompatible with current high-throughput design-build-test cycles. To enable parallel measurement of multiple construct designs, we placed unique DNA barcodes in engineered transcripts and measured barcode abundances via sequencing. In standard curve experiments, the barcode sequences exhibited consistent relationships between input and measured abundances, which allowed us to use transcript barcoding to measure expression levels of 30 GFP-expressing strains of E. coli Nissle in parallel. Applying this technology in culture and in the mouse gut, we found that GFP expression in the gut could often be predicted from expression levels in culture, but several strains exhibited gut-specific expression. This work establishes the experimental design parameters and advantages of transcript barcoding to measure the performance of many engineered probiotic designs in mammalian hosts.