Synthetic biology and metabolic engineering has aided the production of chemicals using renewable resources, thus offering a solution to our dependence on the dwindling petroleum resources. While a major portion of petroleum resources go towards production of fuels, a significant fraction also goes towards production of specialty chemicals. There has been a growing interest in recent years in commercializing bio-based production of such high value compounds. In this thesis the biosynthesis of aromatic esters has been explored, which have typical application as flavor and fragrance additive to food, drinks and cosmetics. Recent progress in pathway engineering has led to the construction of several aromatic alcohol producing pathways, the likes of which can be utilized to engineer aromatic ester biosynthesis by addition of a suitable enzyme from the acyltransferase class. Enzyme selection and screening done in this work has identified chloramphenicol O-acetyltransferase enzyme(CAT) as a potential candidate to complete the biosynthetic pathways for each of 2-phenethyl acetate, benzyl acetate, phenyl acetate and acetyl salicylate. In the end, E. coli strains capable of producing up to 60 mg/L 2-phenethyl acetate directly from glucose were successfully constructed by co-expressing CAT in a previously engineered 2-phenylethanol producing host.

Emergence of multidrug resistant (MDR) bacteria is a major concern to global health. One of the major MDR mechanisms bacteria employ is efflux pumps for the expulsion of drugs from the cell. In Escherichia coli, AcrAB-TolC proteins constitute the major chromosomally-encoded drug efflux system. AcrB, a trimeric membrane protein is well-known for its substrate promiscuity. It has the ability to efflux a broad spectrum of substrates alongside compounds such as dyes, detergent, bile salts and metabolites. Newly identified AcrB residues were shown to be functionally relevant in the drug binding and translocation pathway using a positive genetic selection strategy. These residues—Y49, V127, D153, G288, F453, and L486—were identified as the sites of suppressors of an alteration, F610A, that confers a drug hypersensitivity phenotype. Using site-directed mutagenesis (SDM) along with the real-time efflux and the classical minimum inhibitory concentration (MIC) assays, I was able to characterize the mechanism of suppression.

Three approaches were used for the characterization of these suppressors. The first approach focused on side chain specificity. The results showed that certain suppressor sites prefer a particular side chain property, such as size, to overcome the F610A defect. The second approach focused on the effects of efflux pump inhibitors. The results showed that though the suppressor residues were able to overcome the intrinsic defect of F610A, they were unable to overcome the extrinsic defect caused by the efflux pump inhibitors. This showed that the mechanism by which F610A imposes its effect on AcrB function is different than that of the efflux pump inhibitors. The final approach was to determine whether suppressors mapping in the periplasmic and trans-membrane domains act by the same or different mechanisms. The results showed both overlapping and distinct mechanisms of suppression.

To conclude, these approaches have provided a deeper understanding of the mechanisms by which novel suppressor residues of AcrB overcome the functional defect of the drug binding domain alteration, F610A.

This dissertation focuses on the biosynthetic production of aromatic fine chemicals in engineered Escherichia coli from renewable resources. The discussed metabolic pathways take advantage of key metabolites in the shikimic acid pathway, which is responsible for the production of the aromatic amino acids phenylalanine, tyrosine, and tryptophan. For the first time, the renewable production of benzaldehyde and benzyl alcohol has been achieved in recombinant E. coli with a maximum titer of 114 mg/L of benzyl alcohol. Further strain development to knockout endogenous alcohol dehydrogenase has reduced the in vivo degradation of benzaldehyde by 9-fold, representing an improved host for the future production of benzaldehyde as a sole product. In addition, a novel alternative pathway for the production of protocatechuate (PCA) and catechol from the endogenous metabolite chorismate is demonstrated. Titers for PCA and catechol were achieved at 454 mg/L and 630 mg/L, respectively. To explore potential routes for improved aromatic product yields, an in silico model using elementary mode analysis was developed. From the model, stoichiometric optimums maximizing both product-to-substrate and biomass-to-substrate yields were discovered in a co-fed model using glycerol and D-xylose as the carbon substrates for the biosynthetic production of catechol. Overall, the work presented in this dissertation highlights contributions to the field of metabolic engineering through novel pathway design for the biosynthesis of industrially relevant aromatic fine chemicals and the use of in silico modelling to identify novel approaches to increasing aromatic product yields.

Biomass synthesis is a competing factor in biological systems geared towards generation of commodity and specialty chemicals, ultimately limiting maximum titer and yield; in this thesis, a widely generalizable, modular approach focused on decoupling biomass synthesis from the production of the phenylalanine in a genetically modified strain of E. coli BW25113 was explored with the use of synthetic trans-encoded small RNA (sRNA) to achieve greater efficiency. The naturally occurring sRNA MicC was used as a scaffold, and combined on a plasmid with a promoter for anhydrous tetracycline (aTc) and a T1/TE terminator. The coding sequence corresponding to the target binding site for fourteen potentially growth-essential gene targets as well as non-essential lacZ was placed in the seed region of the of the sRNA scaffold and transformed into BW25113, effectively generating a unique strain for each gene target. The BW25113 strain corresponding to each gene target was screened in M9 minimal media; decreased optical density and elongated cell morphology changes were observed and quantified in all induced sRNA cases where growth-essential genes were targeted. Six of the strains targeting different aspects of cell division that effectively suppressed growth and resulted in increased cell size were then screened for viability and metabolic activity in a scaled-up shaker flask experiment; all six strains were shown to be viable during stationary phase, and a metabolite analysis showed increased specific glucose consumption rates in induced strains, with unaffected specific glucose consumption rates in uninduced strains. The growth suppression, morphology and metabolic activity of the induced strains in BW25113 was compared to the bacteriostatic additives chloramphenicol, tetracycline, and streptomycin. At this same scale, the sRNA plasmid targeting the gene murA was transformed into BW25113 pINT-GA, a phenylalanine overproducer with the feedback resistant genes aroG and pheA overexpressed. Two induction times were explored during exponential phase, and while the optimal induction time was found to increase titer and yield amongst the BW25113 pINT-GA murA sRNA variant, overall this did not have as great a titer or yield as the BW25113 pINT-GA strain without the sRNA plasmid; this may be a result of the cell filamentation.

With the aid of metabolic pathways engineering, microbes are finding increased use as biocatalysts to convert renewable biomass resources into fine chemicals, pharmaceuticals and other valuable compounds. These alternative, bio-based production routes offer distinct advantages over traditional synthesis methods, including lower energy requirements, rendering them as more "green" and "eco-friendly". Escherichia coli has recently been engineered to produce the aromatic chemicals (S)-styrene oxide and phenol directly from renewable glucose. Several factors, however, limit the viability of this approach, including low titers caused by product inhibition and/or low metabolic flux through the engineered pathways. This thesis focuses on addressing these concerns using magnetic mesoporous carbon powders as adsorbents for continuous, in-situ product removal as a means to alleviate such limitations. Using process engineering as a means to troubleshoot metabolic pathways by continuously removing products, increased yields are achieved from both pathways. By performing case studies in product toxicity and reaction equilibrium it was concluded that each step of a metabolic pathway can be optimized by the strategic use of in-situ adsorption as a process engineering tool.