PROGRAM | Chemical Engineering

By: Neil Butler Chair: Aditya Kunjapur

ABSTRACT

Proteins are the molecular workhorses of most biological processes, ranging from cellular metabolism to immune response to molecular transport. However, protein function can be limited by the narrow range of chemical functional groups present in the side chains of the twenty standard amino acids, which are the building blocks that make up natural proteins. In recent years, efforts to introduce alternative, non-standard amino acids (nsAAs) into proteins via genetic code expansion has afforded engineers the ability to incorporate functionalities previously absent from proteins. Of particular interest is the incorporation of nitroaromatic functionalities into proteins, as this moiety is more electron deficient than the standard phenyl or 4-hydroxyphenyl side chains encoded by phenylalanine and tyrosine, respectively. Notably, the substitution of the nitroaromatic nsAA para-nitro-L-phenylalanine (pN-Phe) into proteins has been of interest given its applications in enzyme engineering and immunology. However, these demonstrations relied on the supplementation of chemically synthesized pN-Phe to bacterial cultures for production of nitrated proteins, limiting the contexts where these proteins or cells can be used. In this thesis, I present a platform for biosynthesis of pN-Phe and subsequent incorporation of pN-Phe into proteins for the autonomous production of nitrated proteins in bacteria.

As an initial stage in development, we engineered a strain of Escherichia coli capable of synthesizing pN-Phe from central carbon metabolism. Prior to this work, metabolic engineering efforts toward de novo nitro-product biosynthesis and investigations into nitro-forming enzymes have been limited. Here, I utilized previously characterized genes for the biosynthesis of an amine precursor of pN-Phe, para-aminophenylalanine (pA-Phe). Through a bioprospecting campaign of putative diiron monooxygenases, I identified an enzyme capable of N-oxidation that converted pA-Phe to pN-Phe. Optimization of the E. coli chassis through manipulation of the native aromatic amino acid biosynthetic pathways and modifications to both plasmid constructs and media conditions enabled improvement in pN-Phe biosynthesis to near millimolar levels in relevant culture conditions.

To enable production of nitrated proteins from bio-synthesized pN-Phe, I next identified an orthogonal translation system (OTS) capable of the site-specific incorporation of pN-Phe. Through fluorescence-based screening of orthogonal aminoacyl-tRNA synthetase and tRNA pairs, we obtained an OTS that selectively incorporates pN-Phe, with minimal crosstalk with natural amino acids or pathway intermediates (pA-Phe). Then, with minor genetic engineering and coupling the OTS to the pathway for pN-Phe synthesis, we obtained a strain capable of in situ use of an expanded, nitroaromatic amino acid-containing genetic code, produced directly from glucose.

A more generalized strategy for production of non-standard building blocks was then investigated using the enzyme class L-threonine transaldolases (TTAs). TTAs are PLP-dependent enzymes which synthesize β-hydroxy-non-standard amnio acids (β-OH-nsAA) from L-threonine (L-Thr) and aromatic aldehydes. At the onset of this work, only a single TTA, ObiH, had been previously characterized and was relatively limited in substrate scope and poor L-Thr affinity. To address these challenges, laboratory collaborators and I characterized 12 additional candidate TTAs identified by bioprospecting. While initial expression of many of these enzymes was poor, we were able to significantly improve protein production by appending a solubility tag. Subsequent in vitro and in vivo testing identified TTAs with higher L-Thr affinity, more rapid reaction kinetics, and a broader substrate scope including but not limited to nitroaromatic substrates.

Lastly, I explore engineering of E. coli for bio-catalysis focused on an additional functional group difficult to manipulate in biological systems, aromatic aldehydes. Here, I identify and address unexpected oxidation of a model collection of aromatic aldehydes, including many that originate from biomass degradation within bio-catalytically relevant resting cell conditions. By performing combinatorial inactivation of six candidate aldehyde dehydrogenase genes in the E. coli genome using multiplexed automatable genome engineering (MAGE), I demonstrate that this oxidation can be substantially slowed, with greater than 50% retention of 6 out of 8 aldehydes when assayed 4 h after addition. Applications of this strain for resting cell biocatalysis substantially improved titers in reactions that feature aldehydes as substrates or products.

In summary, this thesis focuses on expansion of the chemical repertoire available to microbes, primarily focused on distributed production of nitrated proteins, but also initiating means to access broader chemical diversification of proteins in engineered bacteria.

Back >