Lindsay Eltis

Members of the genus Rhodococcus have considerable potential as biocatalyst to transform aromatic by-product and waste streams into sustainable chemicals. Such streams include lignin of woody biomass and polyethylene terephthalate (PET), a plastic. Both are abundant polymers that are difficult to valorize due to their recalcitrance. Herein, I investigated the potential of Rhodococcus jostii RHA1 to transform lignin and PET. First, I tested if RHA1’s lignin degrading ability can be enhanced by secreting bacterial lignin-modifying enzymes via the twin arginine translocation (Tat) system. I studied the growth of RHA1 on lignin in combination with small laccase (sLac) secretion and analyzed the sLac reaction products. The results indicate that the Tat system can be exploited to efficiently secrete lignin-modifying enzymes but that this strategy may not yield enhanced growth on lignin as toxic 2,6-dimethoxybenzoquione was produced. Second, I investigated the metabolism of RHA1 on a lignin-enriched stream of corn stover. A time-resolved transcriptomic study showed that RHA1 co-expressed various aromatic and organic acid pathways during exponential growth. I tested the requirement of relevant pathways by growth experiments with mutant strains. When entering the stationary phase, RHA1 highly up-regulated the propane, the alkylphenol, and the alkylguaiacol pathways suggesting that these could serve an unknown role other than the catabolism of their associated compounds. Third, I examined the PET-transform potential of RHA1 by secreting the PET-hydrolyzing enzymes, PETase and MHETase, using the Sec secretion system. While translocation of active enzyme was limited, expression of the Sec secretion signal MHETase construct allowed RHA1 to grow on two PET breakdown products: mono-(2-hydroxyethyl) terephthalate and bis(2-hydroxyethyl) terephthalate. Furthermore, I analyzed the metabolism of RHA1 on ethylene glycol, another PET breakdown product, using transcriptomics and described new gene candidates for the degradation of short chain alcohols and their dependence on the cofactor mycofactocin. Taken together, my results provide insights into utilizing RHA1 as a lignin and PET transforming biocatalyst and reveal remaining barriers. The data will inform engineering strategies to develop efficient RHA1 strains that yield high-value oleochemicals from biomass and plastic waste streams.

View record

Tuberculosis (TB), a disease caused by the airborne bacterial pathogen Mycobacterium tuberculosis (Mtb) claims the lives of ~1.3 million people every year. Control of Mtb has been hindered by the inefficiency of the current treatment for the disease, which involves several months on a multi-drug regimen with extensive side-effects. The ability of Mtb to enter a state of metabolic quiescence further complicates the development of antimycobacterial agents. During infection, Mtb relies on host-produced cholesterol as a source of carbon and energy. The essentiality of cholesterol catabolism for Mtb during infection has been repeatedly demonstrated and the pathway has thus emerged as a potential target for the development of new TB therapeutics. Catabolism of the cholesterol sidechain and rings A/B leads to the production of propionyl-CoA, pyruvate, and the rings C/D containing metabolite 3aα-H-4α(3'-propanoate)-7aβ-methylhexahydro-1,5-indanedione (HIP). Many of the enzymatic steps involved in the degradation of HIP yield CoA thioester metabolites, the accumulation of which have been shown to cause cholesterol-mediated toxicity in deletion mutants. The findings presented in Chapter 3.1 demonstrate that HIP catabolism is important for the pathogenesis of Mtb and that the cholesterol catabolic pathway is a valid drug target. Chapter 3.2 reveals a novel mechanism involved in the regulation of HIP catabolism in which the first enzyme in its catabolism is reversibly acetylated. In Chapter 3.3, I have shown that deletion of certain HIP catabolizing enzymes leads to the accumulation of cholesterol-derived CoA thioesters, sequestration and depletion of free CoA resulting in a toxic phenotype. These findings suggest that HIP catabolizing enzymes are intriguing candidates for the development of novel TB antibiotics. Together, these data build on previous studies and provide additional insights into cholesterol catabolism in Mtb.  

View record

Bacteria play a central role in degrading aromatic compounds, including those derived from lignin, a heterogeneous aromatic polymer and a major component of woody biomass. These catabolic pathways and enzymes play a critical role in the global carbon cycle and have biocatalytic potential in the valorization of biomass. Nevertheless, many of these pathways and enzymes remain poorly understood. This thesis describes the characterization of three such catabolic enzymes, LsdA, LigZ and LigY, each of which has a metal cofactor. LsdA from Sphingomonas paucimobilis TMY1009 is an Fe²⁺-dependent oxygenase that catalyzes the cleavage of stilbenoids into aldehydes. In kinetic anal-yses, LsdA only transformed 4-hydroxystilbenes and was inhibited by phenylazophenol. Crystallo-graphic and mutagenesis analyses established that Lys-134 is essential for catalysis, consistent with a role in deprotonating the stilbene’s 4-hydroxyl. LigZ and LigY catalyze successive reactions in 2,2’-dihydroxy-3,3’-dimethoxy-5,5’-dicarboxy-biphenyl (DDVA) catabolism by Sphingobium sp. SYK-6. In this study, highly active preparations of LigZ, an Fe²⁺-dependent extradiol dioxygenase, afforded an accurate description of the enzyme’s substrate specificity and mechanism-based inactivation. 4,11-Dicarboxy-8-hydroxy-9-methoxy-2-hydroxy-6-oxo-6-phenyl-hexa-2,4-dienoate (DCHM-HOPDA) was identified as the meta-cleavage product (MCP) of the LigZ reaction and was shown to undergo a rapid, reversible non-enzymatic transformation under physiological conditions (t½ ~5 min). The enolate form of the MCP is the substrate of LigY, which catalyzed the hydrolysis of DCHM-HOPDA to 5-carboxyvanillate (5CVA) and 4-carboxy-2-hydroxypenta-2,4-dienoate (CHPD). Phy-logenetic, biochemical and structural analyses identified LigY as a Zn²⁺-dependent amidohydrolase, in contrast to all known MCP hydrolases, which are serine-dependent enzymes. Nevertheless, transient-state kinetic analyses revealed a catalytic intermediate, ESred, with a red-shifted spectrum similar to that observed in serine-dependent MCP hydrolases. Further, 4-methyl HOPDA inhibited LigY and yielded a complex with a spectrum similar to that of ESred. Crystallographic analyses of this complex revealed the MCP binds the Zn²⁺ in a bidentate manner. Together, the data support a mechanism in which the nucleophile is activated in the same substrate-assisted manner as in the serine-dependent enzymes, but that in LigY, the nucleophile is water, such that C-C fission is preceded by a gem-diol intermediate. Overall, this thesis provides important insights into several lignin-degrading enzymes and the superfamilies to which they belong.

View record

Rhodococcus is a genus of soil bacteria that are among the best-studied oleaginous bacteria, and have considerable potential for the sustainable production of lipid-based chemicals. Herein, I characterized the biosynthesis of wax esters (WEs) in Rhodococcus jostii RHA1 and created tools to develop the biocatalytic potential of rhodococci. In Section 3.1, I established that RHA1 produces WEs and identified key enzymes involved in this production. Specifically, RHA1 produced WEs to 0.0002% of cellular dry weight (CDW) during exponential growth on glucose. These WEs contained 31 to 34 carbon atoms and were saturated. Bioinformatics revealed that RHA1 contains a putative fatty acyl-CoA reductase (FcrA). Purified FcrA catalyzed the NADPH-dependent transformation of stearoyl-CoA to stearyl alcohol with a specific activity of 45±3 nmol/mg∙min and dodecanal to dodecanol with a specific activity of 5300±300 nmol/mg∙min. A strain of RHA1 overproducing FcrA accumulated WEs to ~13% CDW. In Section 3.2, I expanded the genetic tools available in rhodococci, creating pSYN, a modular integrative-vector. I employed this vector to identify and characterize P₁₀, a strong, potentially constitutive rhodococcal promoter. Various strength promoters were created from P₁₀, resulting in the PT₂, PT₁, and PM₆ promoters, which were 1.3-, 2.2-, and 6-fold stronger, respectively, than Pnit, a constitutive promoter previously characterized in Rhodococcus. RHA1 transformed with a single copy of fcrA under the control of these various promoters accumulated WEs. In Section 3.3, I further developed RHA1 as a WE biocatalyst. Screening a variety of enzymes identified WS2 of Marinobacter hydrocarbonoclasticus DSM 8798 as an effective wax synthase in RHA1. Cassettes for the co-expression of chromosomally integrated fcrA and ws2 were created and transformed into RHA1; resulting in a biocatalyst that accumulated WEs to greater than 15% CDW, at yields of 0.05 g/g glucose, while maintaining 80% of the specific growth rate of WT. Accumulated WEs were 29 to 38 carbon atoms in length, of which 75% were unsaturated, with a ~2:1 mix of mono- and diunsaturated species. Overall, this thesis provides insight into the biosynthesis of WEs in rhodococci and facilitates the development of this genus for biocatalytic applications, including the production of high-value neutral lipids.

View record

Cholesterol, a four-ringed steroid with an alkyl side chain, is an important growth substrate for Mycobacterium tuberculosis (Mtb) during infection. Many aspects of this catabolism remain unknown although steroid catabolism is a defining feature of mycobacteria and the related rhodococci. Using a variety of approaches, I elucidated key aspects of cholesterol catabolism in Mtb and other bacteria, particularly with respect to 3aα-H-4α(3'-propanoate)-7aβ-methylhexahydro-1,5-indane-dione (HIP), a metabolite that contains the last two steroid rings (C/D). Chapter 2 demonstrates that the first two steroid rings (A/B) are degraded prior to the side chain in mycobacteria and rhodococci. This was established by targeting HsaD, the final ings A/B-degrading enzyme. Thus, a ΔhsaD mutant of Rhodococcus jostii RHA1 accumulated cholesterol-derived catabolites with partially degraded side-chains. Moreover, HsaD from Mtb had 100-fold higher specificity (kcat/KM) for a metabolite with a partially-degraded side chain. Chapter 3 presents a mechanism for KstR2, a TetR family transcriptional repressor that regulates the HIP catabolic genes, including ipdABCF and echA20. The KstR2 dimer bound two equivalents of HIP-CoA with high affinity (KD = 80±10 nM). Crystallographic analyses revealed that HIP-CoA binding induces conformational changes in the dimer that preclude DNA-binding. Mutagenesis substantiated the roles of Arg162 and Trp166 in HIP-CoA binding. In Chapter 4, key HIP catabolic steps are elucidated. Two previously undescribed metabolites, 3aα-H-4α(carboxyl-CoA)-5-hydroxy-7aβ-methylhexahydro-1-indanone (5-OH-HIC-CoA) and (R)-2-(2-carboxyethyl)-3-methyl-6-oxocyclohex-1-ene-1-carboxyl-CoA (COCHEA-CoA) were identified using deletion mutants of ipdC and ipdAB, respectively, combined with novel metabolomics approaches. Together, purified IpdC, IpdF and EchA20 transformed 5-OH-HIC-CoA to COCHEA-CoA. These data, along with those from additional mutants, were used to formulate a HIP catabolic pathway and to predict that cholesterol catabolism yields four propionyl-CoA, four acetyl-CoA, one pyruvate, and one succinate. Chapter 5 establishes that IpdAB catalyzes a retro-Claisen-like ring-opening of COCHEA-CoA (kcat/KM = 2±0.7 × 10⁵ M⁻¹s⁻¹) despite structural similarity with Class I CoA transferases. Based on crystal structures of IpdAB and biochemical data, a mechanism for ring-cleavage is proposed in which conserved Glu105 acts as a catalytic base. Overall, this work significantly advances our understanding of bacterial steroid catabolism and facilitates the development of novel therapeutics to treat TB.

View record

The meta-cleavage product (MCP) hydrolases are members of the α/β-hydrolase superfamily that utilize a Ser-His-Asp triad to catalyze the hydrolysis of a C-C bond. The catalytic mechanism of the MCP hydrolases is poorly defined and particularly interesting due to a requisite substrate ketonization that precedes hydrolysis. To resolve the catalytic mechanism of the MCP hydrolases, two enzymes were studied: tetrameric BphDLB400 from Burkholderia xenovorans LB400 and dimeric DxnB2 from Sphingomonas witichii RW1. Both efficiently hydrolyze 2-hydroxy-6-oxo-6-phenylhexa-2,4-dienoic acid (HOPDA) to produce 2-hydroxypenta-2,4-dienoic acid (HPD) and benzoic acid.A series of experiments established that BphDLB400 uses an histidine-independent nucleophilic mechanism of catalysis and is half-site reactive. Benzoylation of Ser112 was demonstrated by LC ESI/MS/MS analyses and a pre-steady-state kinetic burst of HPD formation indicated the reactivity. While acylation during HOPDA hydrolysis by BphDLB400 occurred on a similar timescale for the WT and H265Q variant, esterase activity was abrogated in the histidine variant. Thus, alternative mechanisms of nucleophile activation are employed for C-C and C-O bond cleavage.A covalent mechanism of catalysis was inferred for DxnB2, however, the turnover of HOPDA was 1:1 with respect to enzyme concentration. A solvent kinetic isotope effect suggested that a proton transfer, and therefore, substrate ketonization determines the rate of acylation in the MCP hydrolases. Substrate ketonization, and therefore acylation, can be indirectly observed as consumption of ESred, an intermediate named for its bathochromically-shifted absorption spectrum. A proton transfer to ESred allowed the assignment of this species to an enzyme-bound HOPDA dianion. An extended Brønsted analysis revealed a linear correlation between substrate basicity and the rate constant determined for the ketonization reaction. Finally, the MCP hydrolase P-subsite, which contacts the MCP dienoate moiety, was definitively linked to substrate ketonization. In DxnB2 Asn43 and Arg180 variants, ESred formation was found to limit this proton transfer reaction.A substrate-assisted nucleophilic mechanism of catalysis has been proposed for the MCP hydrolases. Therein, the electron-rich dienoate moiety substitutes for the His-Asp pair as the general base for nucleophile activation. Overall, definition of the chemical mechanism of the MCP hydrolases has implications for environmental bioremediation strategies and the rational design of therapeutics.

View record

Mycobacterium tuberculosis (Mtb) is the human pathogen that causes tuberculosis. A gene cluster encoding a cholesterol degradation pathway plays a role in Mtb virulence. Two iron-containing monooxygenases in this pathway were characterized with respect to their roles in bacterial cholesterol catabolism: the Rieske oxygenase (RO) KshAB, and the cytochrome P450 (P450) Cyp125. These enzymes are predicted to catalyze the first ring-opening step and the first transformation of the steroid side chain, respectively. Cyp125A1 (Mtb) and Cyp125A14P (Rhodococcus jostii RHA1) were expressed in R. jostii RHA1 and characterized in vitro using the Mtb reductase KshB. Both enzymes were purified with the heme iron in a predominantly high spin state and exhibiting thiolate ligation of the heme iron. Both P450s bound cholesterol and 4-cholesten-3-one with apparent submicromolar affinity. Cyp125A1 was demonstrated to catalyze C26-monohydroxylation of both steroids. KshA (a terminal oxygenase) and KshB (an oxygenase reductase) of Mtb were produced in Escherichia coli and characterized in vitro. KshAB had over twenty times the apparent substrate specificity for steroid substrates with isopropionyl-CoA side chains than for the corresponding 17-keto steroids. The apparent KMO₂ with a CoA thioester-bearing steroid was 90 ± 10 μM whereas that for the corresponding 17-keto steroid was in excess of 1.2 mM. These results suggest that the physiological substrate(s) for KshAB is likely a CoA thioester intermediate of cholesterol side chain degradation.A comprehensive phylogenetic analysis was undertaken to consolidate the available RO literature. Six hundred fifty enzymes that are fully representative of the RO terminal oxygenase (RO-O) sequences in the NCBI database were collected and aligned to a structure-based sequence template. The structure-based alignment was also used to objectively define the structurally conserved positions that were included in phylogenetic reconstruction. The resulting analysis revealed a level of RO-O diversity that has been unrecognized in previous literature and that necessitates a different approach to RO-O classification. A classification scheme based on the system currently in use for P450s was proposed.This work provides significant insight into the cholesterol degradation pathway of Mtb and the RO-O protein family and contributes to potential commercial applications in bioremediation, biocatalysis, and Mtb therapeutics.

View record

Mycobacterium tuberculosis (Mtb) is the leading cause of mortality from bacterial infection. A cholesterol degradation pathway identified in Mtb is implicated in the pathogen’s survival in the host. This pathway includes an Fe(II)-containing extradiol dioxygenase, HsaC, and a meta-cleavage product (MCP) hydrolase, HsaD, which are predicted to catalyze the cleavage of DHSA (3,4-dihydroxy-9,10-seconandrost-1,3,5(10)-triene-9,17-dione) and subsequent hydrolysis of DSHA (4,5-9,10-diseco-3-hydroxy-5,9,17-trioxoandrosta-1(10),2-diene-4-oic acid), respectively. HsaC and HsaD were expressed in E. coli, purified, and characterized. Substrates were obtained by biotransformation of cholesterol using a ΔhsaC mutant of Rhodococcus jostii RHA1. From steady-state kinetic studies, purified HsaC efficiently cleaved the proposed steroid metabolite, DHSA (kcat/Km = 15 ± 2 μM-¹s-¹), better than the biphenyl catechol, DHB, or a synthetic analogue, DHDS. Two halogenated substrates, 2’,6’-diCl DHB and 4-Cl DHDS, inactivated HsaC with partition coefficients
View record

Rhodococcus jostii RHA1 grows on various nitriles. Proteomic studies revealed that RHA1 utilizes an nth pathway, which includes an Fe³⁺-dependent nitrile hydratase (NHase), to catabolize phenylacetonitrile. By contrast, RHA1 utilizes the anh pathway, which includes a novel acetonitrile hydratase (ANHase), a αβ-heterodimeric metalloenzyme, to catabolize acetonitrile.To better characterize ANHase, a Rhodococcus-E. coli shuttle vector was constructed using a benzoate-inducible promoter. ANHase transformed nitriles containing up to four carbons, with highest specificity for acetonitrile and propionitrile. The enzyme contained cobalt, copper and zinc, and lacked sequence identity with known NHases. Accordingly, ANHase is proposed to belong to a previously unknown class of NHases. The α subunit of ANHase possesses two potential metal-binding sequences: an N-terminal sequence rich in His and acidic residues; and a Cys-rich motif (CLLGCAC). The latter is reminiscent of the conserved catalytic motif in Co³⁺-dependent NHases, suggesting that it coordinates Co³⁺ in ANHase. The N-terminal sequence is similar to that found in some other cobaltoenzymes. Spectrophotometric analysis of a synthetic N-terminal peptide (MPDHGHDHGHNHDACDSE) demonstrated that it formed a 2:1 complex with Co²⁺ and that Cys is a probable ligand. Isothermal titration calorimetry (ITC) studies demonstrated that the peptide binds Co²⁺, Zn²⁺ and Ni²⁺ with similar affinities (Kd = 3.2–4.3 µM).Finally, AnhE, an 11-kDa protein whose gene occurs between the ANHase structural genes, anhA and anhB, was characterized. A ΔanhE mutant grew on acetonitrile only if anhE was provided in trans. Co-expression of anhE with anhA enabled reconstitution of ANHase in vitro in the presence of low concentrations of Co²⁺. Co²⁺ and other divalent metal ions stabilized a dimeric form of AnhE. ITC studies demonstrated that the dimer binds two Co²⁺ ions with different affinities (Kd1 = 0.12 nM, Kd2 = 110 nM) but only one Zn²⁺ (Kd = 11 nM) and one Ni²⁺ (Kd = 49 nM). Together, these data suggest that AnhE acts as a dimeric metallochaperone to deliver cobalt to ANHase. Overall, these studies provide insight into a novel NHase and its maturation. The findings have potential applications in the bacterial transformation of nitriles as well as implications for metal-trafficking in biological systems.

View record

Bacteria are found in almost all conceivable environments, and some species can survive many different conditions. The ability to detect environmental conditions and respond with appropriate changes to gene expression is essential to survival. Bacteria sometimes express genes involved in horizontal gene transfer when encountering a stressful environment. Horizontal gene transfer has an important role in the evolution of prokaryotic genomes. Rhodobacter capsulatus produces a mediator of horizontal gene transfer called the gene transfer agent (GTA). The R. capsulatus GTA is a bacteriophage-like particle that transfers ~4 kb of double stranded genomic DNA using a transduction-like mechanism. Previously, two proteins encoded outside the GTA gene cluster, GtaI and CtrA, were found to regulate GTA expression. GtaI and GtaR are LuxI-type and LuxR-type quorum sensing proteins, respectively. CtrA and CckA are homologues of the response regulator and sensor kinase, respectively, of the Caulobacter crescentus CtrA/CckA signal transduction system. In this thesis, I studied the interactions between these regulatory proteins, environmental conditions and GTA in R. capsulatus. I found that growth conditions had opposite effects on GTA and ctrA expression, but no effect on gtaR expression, and phosphate limitation decreased expression of ctrA. Knockout experiments revealed that GtaI and GtaR affect ctrA, gtaR and GTA expression. Results from GtaR-DNA binding experiments were consistent with a model in which GtaR directly regulates its own expression but indirectly regulates ctrA and GTA expression. These studies also identified GtaR binding sequences. I found that in R. capsulatus CtrA did not regulate its own transcription, contrary to what occurs in C. crescentus. My research also showed that GTA expression was affected by at least one other unidentified system. Promoter deletion studies of ctrA, gtaR and GTA genes identified sequences that may be involved in GtaI-, GtaR-, CtrA-, and/or growth condition-based regulation. Overall, these studies contribute to the understanding of how bacteria detect multiple environmental signals and respond with changes to gene expression.

View record

3C proteinases (3Cpros) are a family of essential cysteine proteinases found in various viruses of medical and agricultural significance. Three lines of research related to the characterization of 3Cpros were pursued. In the first, biological selections and screens were developed to evaluate proteinase activity in a high-throughput fashion. A selection system based on the cleavage of an engineered transcriptional regulator, XylR, by hepatitis A virus (HAV) 3Cpro failed. However, this strategy facilitated the development of a screen based on the cleavage of fused green fluorescent protein variants, CyPet and YPet. The screen was used to demonstrate that HAV 3Cpro prefers Ile, Val or Leu at the P₄ position of the cleavage sequence, and Gly, Ser or Ala at the P₁’ position. In the second project, the 3Cpro from Israeli acute paralysis virus (IAPV), a dicistrovirus, was investigated. IAPV has been associated with the recent colony collapse disorder afflicting commercial hives. A portion of the replicase including 3Cpro was heterologously produced. The resulting autoprocessed fragments were analyzed using mass spectrometry to identify the 3Cpro cleavage sequence PIVIE/AQT. This cleavage sequence likely occurs between the 3A/3B proteins of the polypeptide and is the first within the replicase to be described in this family of viruses. Thirdly, inhibitors of HAV 3Cpro and SARS 3CLpro were developed. The keto- glutamine analogue (HIP2-171-2) competitively inhibited SARS 3CLpro with a Kic = 0.17 ± 0.03 μM and is among the most potent peptide inhibitors developed against this proteinase. The azapeptide epoxide (APE) KAE-3-91 irreversibly inhibited SARS 3CLpro with a kinact/Ki of 1900 ± 400 M−¹s−¹, which is similar in magnitude to that of the first generation APE’s produced to inhibit caspases. Finally, both SARS 3CLpro and HAV 3Cpro were screened against a library of inhibitory halopyridinyl esters. Each of three halopyridinyl esters inhibited 3Cpro with apparent Kic’s of 120-240 nM. However, further study revealed that the inhibitors were slowly hydrolyzed by both proteinases. Overall, the described screens and inhibitors should facilitate the further characterization of 3C and related proteinases as well as the development of novel antivirals.

View record