Search

The expanding knowledge of the health impacts of the metabolic activities of the gut microbiota reinforces the current interest in engineered probiotics. Tryptophan metabolites, in particular indole lactic acid (ILA), are attractive candidates as potential therapeutic agents. ILA is a promising compound with multiple beneficial effects, including amelioration colitis in rodent models of necrotizing enterocolitis, as well as improved infant immune system maturation. In this work, we engineered and characterized in vitro and in vivo an Escherichia coli Nissle 1917 strain that produces ILA. The 2-step metabolic pathway comprises aminotransferases native of E. coli and a dehydrogenase introduced from Bifidobacterium longum subspecies infantis. Our results show a robust engineered probiotic that produces 73.4 ± 47.2 nmol and 149 ± 123.6 nmol of ILA per gram of fecal and cecal matter, respectively, three days after colonization in a mouse model. In addition, hereby is reported an engineered-probiotic-related increase of ILA in the systemic circulation of the treated mice. This strain serves as proof of concept for the transfer of capacity to produce ILA in vivo and as ILA emerges as a potent microbial metabolite against gastrointestinal inflammation, further development of this strain offers efficient options for ILA-focused therapeutic interventions in situ. [ABSTRACT FROM AUTHOR]

Background: Metabolic engineering projects often require integration of multiple genes in order to control the desired phenotype. However, this often requires iterative rounds of engineering because many current insertion approaches are limited by the size of the DNA that can be transferred onto the chromosome. Consequently, construction of highly engineered strains is very time-consuming. A lack of well-characterised insertion loci is also problematic. Results: A series of knock-in/knock-out (KIKO) vectors was constructed for integration of large DNA sequences onto the E. coli chromosome at well-defined loci. The KIKO plasmids target three nonessential genes/operons as insertion sites: arsB (an arsenite transporter); lacZ (β-galactosidase); and rbsA-rbsR (a ribose metabolism operon). Two homologous 'arms' target each insertion locus; insertion is mediated by λ Red recombinase through these arms. Between the arms is a multiple cloning site for the introduction of exogenous sequences and an antibiotic resistance marker (either chloramphenicol or kanamycin) for selection of positive recombinants. The resistance marker can subsequently be removed by flippase-mediated recombination. The insertion cassette is flanked by hairpin loops to isolate it from the effects of external transcription at the integration locus. To characterize each target locus, a xylanase reporter gene (xynA) was integrated onto the chromosomes of E. coli strains W and K-12 using the KIKO vectors. Expression levels varied between loci, with the arsB locus consistently showing the highest level of expression. To demonstrate the simultaneous use of all three loci in one strain, xynA, green fluorescent protein (gfp) and a sucrose catabolic operon (cscAKB) were introduced into lacZ, arsB and rbsAR respectively, and shown to be functional. Conclusions: The KIKO plasmids are a useful tool for efficient integration of large DNA fragments (including multiple genes and pathways) into E. coli. Chromosomal insertion provides stable expression without the need for continuous antibiotic selection. Three non-essential loci have been characterised as insertion loci; combinatorial insertion at all three loci can be performed in one strain. The largest insertion at a single site described here was 5.4 kb; we have used this method in other studies to insert a total of 7.3 kb at one locus and 11.3 kb across two loci. These vectors are particularly useful for integration of multigene cassettes for metabolic engineering applications. [ABSTRACT FROM AUTHOR]