This item is under embargo and not available online per the author's request. For access information, please visit


Date of Award

Summer 8-15-2018

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Plant & Microbial Biosciences)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Unlike humans and other metazoans, organisms such as fungi, bacteria, and plants have retained the enzymatic machinery necessary to synthesize their aromatic amino acids de novo. Chorismate, the final product of the shikimate pathway, is the precursor to the three aromatic amino acidsѠtryptophan, tyrosine, and phenylalanineѠand is upstream of a number of plant growth hormones, including auxins and benzoates. Phenylalanine and tyrosine both stem from the precursor prephenate, which is formed from chorismate by chorismate mutase, and use dehydrogenases, aminotransferases, and dehydratases in their biosynthetic pathways. Although aromatic amino acid biosynthesis is important for protein synthesis, secondary metabolism, and human health, much of what is known about plant aromatic amino acid biosynthesis and its regulation has been inferred from microbial investigations of the pathway.

L-Tyrosine is essential for protein synthesis and is a precursor of numerous specialized metabolites crucial for plant and human health. Tyrosine can be synthesized via two alternative routes by a key regulatory TyrA family enzyme, prephenate or arogenate dehydrogenase (PDH/TyrAp or ADH/TyrAa), representing a unique divergence of primary metabolic pathways. The molecular foundation underlying the evolution of the alternative Tyr pathway is currently unknown. Here we characterized recently-diverged plant PDH and ADHs, obtained the x-ray crystal structure of soybean PDH, and identified a single amino acid residue that defines TyrA substrate specificity and regulation. Structures of mutated PDHs co-crystallized with Tyr indicate that substitutions of Asn222 confers ADH activity and Tyr-sensitivity. Subsequent mutagenesis of the corresponding residue in divergent plant ADHs introduced PDH activity and relaxed Tyr sensitivity, highlighting the critical role of this residue in TyrA substrate specificity that underlies the evolution of alternative Tyr biosynthetic pathways in plants.

The three-dimensional structure of soybean PDH1 allowed for the identification of both the cofactor- and ligand-binding sites. Here, we present steady-state kinetic analysis of twenty site-directed active site mutants of the soybean (Glycine max) PDH compared to wild-type. Molecular docking of the substrate, prephenate, into the active site of the enzyme reveals its potential interactions with the active site residues and makes a case for the importance of each residues in substrate recognition and/or catalysis, most likely through transition state stabilization. Overall, these results suggest that the active site of the enzyme is highly sensitive to any changes, as even subtle alterations substantially reduced the catalytic efficiency of the enzyme.

Chorismate mutase catalyzes the branch point reaction of phenylalanine and tyrosine biosynthesis to generate prephenate. In Arabidopsis thaliana, there are two plastid-localized chorismate mutases that are allosterically regulated (AtCM1 and AtCM3) and one cytosolic isoform (AtCM2) that is unregulated. Previous analysis of plant chorismate mutases suggested that the enzymes from early plants (i.e., bryophytes/moss, lycophytes, and basal angiosperms) formed a clade distinct from the isoforms found in flowering plants; however, no biochemical information on these enzymes is available. To understand the evolution of allosteric regulation in plant chorismate mutases, we analyzed a basal lineage of plant enzymes homologous to AtCM1 based on sequence similarity. The chorismate mutases from the moss/bryophyte Physcomitrella patens (PpCM1 and PpCM2), the lycophyte Selaginella moellendorffii (SmCM), and the basal angiosperm Amborella trichopoda (AmtCM1 and AmtCM2) were characterized biochemically. Tryptophan was a positive effector for each of the five enzymes examined. Histidine was a weak positive effector for PpCM1 and AmtCM1. Neither tyrosine nor phenylalanine altered the activity of SmCM; however, tyrosine was a negative regulator of the other four enzymes. Phenylalanine down-regulates both moss enzymes and AmtCM2. The 2.0 x-ray crystal structure of PpCM1 in complex with the tryptophan identified the allosteric effector site and reveals structural differences between the R- (more active) and T-state (less active) forms of plant chorismate mutases. Molecular insight into the basal plant chorismate mutases guides our understanding of the evolution of allosteric regulation in these enzymes.

Plants make tyrosine and phenylalanine by a different pathway from many microbes, which requires prephenate aminotransferase (PAT) as the key enzyme. PAT produces arogenate, the unique and immediate precursor for both tyrosine and phenylalanine in plants, and also has aspartate aminotransferase (AAT) activity. The molecular mechanisms governing the substrate specificity and activation or inhibition of PAT are currently unknown. Here we present the x-ray crystal structures of wild-type and various mutants of PAT from Arabidopsis thaliana (AtPAT). Steady-state kinetic and ligand binding analyses identified key residues, such as Glu108, that are involved in both keto acid and amino acid substrate specificities and likely contributed to the evolution of PAT activity among class Ib AAT enzymes. Structures of AtPAT mutants co-crystalized with either _-ketoglutarate or pyridoxamine 5խphosphate (PMP) and glutamate further define the molecular mechanisms underlying keto acid and amino acid substrate recognition. Furthermore, cysteine was identified as an inhibitor of PAT from A. thaliana and Antirrhinum majus plants as well as Chlorobium tepidum bacterium, uncovering a potential new effector of PAT.

As a way of regulating the concentration of a given plant hormone within the cell, plants contain a class of enzymes known as GH3 proteins that conjugate amino acids to acyl acid hormones to activate or inactivate the hormone molecule. Grouping of the GH3 proteins by substrate preference shows that the largest class of GH3 proteins use the growth-promoting hormone auxin (indole-3-acetic acid, or IAA), and the second largest class uses the defense hormone jasmonic acid; however, some of the auxin-specific GH3 proteins, notably Arabidopsis thaliana GH3.5, have been found to use multiple acyl acid substrates, including the auxins IAA and phenylacetic acid and benzoic acid, with roughly the same catalytic efficiencies. Despite the promiscuity in acyl acid substrate preference, the active site residues that are in contact with a ligand in the crystal structure of AtGH3.5 do not vary from those found in the AtGH3.2 active site. Using single-turnover kinetics, the rate data suggests that both proteins prefer IAA as the substrate. In conducting these experiments, we found that the adenylated reaction intermediate was being hydrolyzed into acyl acid and AMP in the absence of the amino acid, a typical feature of pre-transfer editing. Here we show that a non-cognate acyl acid-adenylate intermediate is more easily hydrolysable than the cognate acyl acid-adenylate, likely due to a slowed structural switch that provides a checkpoint for fidelity before the full reaction proceeds.

There are 19 GH3 proteins in the model plant Arabidopsis, but only 9 of those proteins have confirmed substrates. One clade of GH3 proteins is predicted to use benzoate as substrate and includes AtGH3.7 and AtGH3.12. Previously, AtGH3.12 was identified as a 4-hydroxybenzoic acid-glutamate synthetase and was found to influence pathogen defense responses through the plant defense hormone salicylic acid. Here, we screened 15 plant hormones with 20 amino acids to find that AtGH3.7 is a chorismate-cysteine synthetase. Rescreening of AtGH3.12 found that it too uses chorismate and is a chorismate-glutamate synthetase. These results were confirmed using a combination of x-ray crystallography, site-directed mutagenesis, single-turnover kinetics, and mass spectrometry. Because chorismate is a precursor to the three aromatic amino acids in plants and a precursor to salicylic acid, these proteins are modulating the concentration of a pathway intermediate. Interestingly, gh3.7 mutants did not have the same pathogen susceptibility phenotype as gh3.12 mutants, but the gh3.7/gh3.12 double mutant was more susceptible to pathogen infection than gh3.12 alone. This study reveals how GH3 proteins, which traditionally regulate plant hormones, can also use hormone intermediates to regulate multiple hormones.

The overarching goal of my thesis research was to understand how plants regulate the biosynthesis of aromatic amino acids, which are essential to both protein synthesis and the production of downstream secondary metabolites, including phytohormones. This was accomplished by structural and biochemical studies of prephenate dehydrogenase, arogenate dehydrogenase, chorismate mutase, and prephenate aminotransferase. Also, biochemical assays were used to unveil the basis of substrate fidelity in the GH3 proteins. Finally, the role of aromatic amino acid biosynthesis pathway intermediates was further explored when AtGH3.7 and AtGH3.12 were found to conjugate amino acids to the aromatic branchpoint metabolite chorismate. This research as a whole has advanced our understanding of the regulation of plant aromatic amino acid and hormone metabolic pathways, unveiling how plants have evolved specific regulatory mechanisms and networks that are more complex than previously thought.


English (en)

Chair and Committee

Joseph M. Jez

Committee Members

Barbara N. Kunkel, Lucia C. Strader, Richard D. Vierstra, Timothy A. Wencewicz,


Permanent URL: 2020-06-19

Available for download on Friday, June 19, 2020