• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Synthesis and reactivity of synthetic analogs for nickel redox enzymes: Superoxide dismutase and acetyl coenzyme A synthase

Dissertation
Author: Molly J. O'Hagan
Abstract:
Synthetic model complexes of nickel-containing enzyme active sites are important tools for understanding the structure-function properties of the active sites as well as for investigating the enzyme mechanisms. The synthesis and reactivity of structural mimics and catalytic-intermediate analogs can provide insight into how the diverse chemistry of nickel-containing enzymes occurs. Reported herein are two synthetic models of redox active nickel-dependant metalloproteins, nickel-containing superoxide dismutase and acetyl coenzyme A synthase. Nickel-containing superoxide dismutase, Ni SOD, contains a mononuclear nickel active site found at the N-terminus with active site ligands found within the first six residues of the protein sequence.1 The active site sequence is unstructured in the absence of Ni2+ providing a novel target for synthetic modeling. A Ni SOD model system was designed utilizing short peptides as ligands based on the native active site amino acid sequence, NH2 -His-Cys-Gly-Gly-Pro-Cys-COOH. The system is used to investigate the minimum construct necessary to establish the native peptide fold and metal coordination. The [Ni(HCGGPC)]- complex was determined to be monomeric by mass spectrometry and optical spectral titrations with K d = 4.7 ± 0.4 x 10-12 M. One- and two-dimensional NMR spectral data suggests the nickel coordination sphere is the same as found in the enzyme. Metal selectivity of the peptide was measured comparing Ni 2+ , Zn2+ , and Co2+ . Ni2+ is preferred two orders of magnitude over Zn2+ and three orders of magnitude over Co2+ . The reactivity of the model complex with O2·- is similar to that observed for the enzyme.1 Reaction of the complex with substoichiometric KO2 results in a Ni3+ species with a similar EPR spectrum as that observed in the enzyme. The superoxide dismutase activity of the nickel complex was measured to be, IC50 =191 ± 9 μM much reduced compared to the native enzyme. The decreased activity is due to the product-catalyzed decomposition of the complex suggesting substrate channeling and the active site pocket of the enzyme provide essential oxidative protection of the active site thiols. Acetyl coenzyme A synthase, ACS, catalyzes the production of acetyl coenzyme A from one-carbon units, CO and CH3 . A recently proposed ACS catalytic mechanism involves a zero valent nickel as the complex to accept a methyl group from methylcobalt species during catalytic turnover.2 Synthetic analog studies were designed to probe the feasibility of Ni(0) as the methyl acceptor. Reaction of [triphos]Ni(PPh3 ) (triphos = bis(2-diphenylphosphinoethyl)phenyl phosphine) with CH3 Co(dmgBF2 )2 py proceeds quantitatively yielding [Ni(triphos)(Me)]+ and Co(I). 3 Kinetic data of methyl transfer and the rate of other alkyls (Et, i Pr) suggests that the mechanism of alkyl transfer proceeds via an SN 2 pathway similar to the ACS enzyme. The nickel-alkyls complexes possessing β-hydrogen undergo β-hydrogen elimination forming a nickel-hydride complex characterized by NMR, FT-IR spectroscopies and X-ray crystallography. This system is also used to model another intermediate along the catalytic pathway, the Ni-acyl complex, to help determined the importance of the order of substrate binding to the ACS active site during catalysis.

TABLE OF CONTENTS LIST OF FIGURES..................................................................................................xii LIST OF SCHEMES..............................................................................................xviii LIST Of ABBREVIATIONS...................................................................................xix ABSTRACT.............................................................................................................xx CHAPTER 1: INTRODUCTION................................................................................1 1.1 Nickel in Biology....................................................................................1 1.1.1 Nickel Bioavailability.................................................................1 1.1.2 Nickel-dependent Enzymatic Pathways.......................................3 1.2 Nickel-containing Enzyme Active Sites..................................................4 1.3 Model Complexes of Nickel-containing Active Sites..............................6 CHAPTER 2: PEPTIDE MODELS OF THE ACTIVE SITE OF NICKEL SUPEROXIDE DISMUTASE.........................................................................9

2.1 Introduction............................................................................................9 2.1.1 Biochemical Significance..........................................................10 2.1.2 Ni SOD Enzyme Structure........................................................13 2.1.3 Proposed Mechanism of Disproportionation..............................17 2.1.4 Ni SOD Model Complexes........................................................20 2.2 Experimental Procedures......................................................................25 2.2.1 General Procedures...................................................................25

vii 2.2.2 Materials...................................................................................25 2.2.3 Physical Methods......................................................................25 2.2.4 Peptide Synthesis and Purification.............................................26 2.2.5 Nickel-Peptide Complex Synthesis............................................27 2.2.6 Dissociation Constant Determination........................................27 2.2.7 NMR Experiments....................................................................30 2.2.8 X-ray absorption spectroscopy..................................................30 2.2.9 Computational methods.............................................................31 2.2.10 SOD activity assays..................................................................32 2.3 Model Complex Design, Synthesis and Structural Characterization......32 2.3.1 Peptide Design, Synthesis and Characterization.........................32 2.3.2 Nickel-Peptide Complex Synthesis and Structural Characterization........................................................................47

2.3.3 Density functional theory investigation of isomers and protonation states......................................................................70

2.4 Stoichiometry and metal selectivity of Ni SOD peptide model complexes.............................................................................................73

2.6 Reactivity of Ni SOD peptide model complex.......................................80 2.5 Conclusions..........................................................................................85 CHAPTER 3: MECHANISM OF ALKYL TRANSFER FROM ALKYLCOBALOXIME TO (TRIPHOS)NI(PPH 3 ): MODEL FOR ACETYL COENZYME A SYNTHASE.......................................................88

3.1 Introduction..........................................................................................88 3.1.1 ACS/COdH Enzyme and the Global Carbon Cycle...................89

viii 3.1.2 ACS/COdH Enzyme Structure..................................................92 3.1.3 Proposed Mechanisms...............................................................95 3.1.4 Synthetic model studies...........................................................100 3.2 Experimental Section..........................................................................104 3.2.1 General Procedures.................................................................104 3.2.2 Materials.................................................................................104 3.2.3 Physical Methods....................................................................104 3.2.4 Synthetic Methods...................................................................105 3.2.3 Reactivity Studies...................................................................108 3.2.5 Kinetic Studies of alkyl transfer..............................................109 3.2.6 X-ray Crystallography.............................................................109 3.3 Results and Discussion.......................................................................116 3.3.1 Kinetic Characterization of Methyl Transfer............................116 3.3.2 -hydrogen elimination decomposition pathway.....................123 3.3.2.1 Ni-alkyl decomposition.............................................123 3.3.2.2 -hydrogen elimination: nickel-hydride complex......129 3.3.3 Alkyl transfer from (5-hexenyl)cobaloxime to (triphos)Ni(PPh 3 ): Radical probe.............................................140

3.3.4 Importance of the substrate binding order in ACS catalysis.....147 3.4 Conclusions........................................................................................151 References………….…………………………………….............………… 154 Appendix: Permission Licenses……………………………………………..164

i x

x LIST OF TABLES Table 2.1: NMR resonance assignments for NH 2 -HCGGPC-CONH 2 , 100 mM NaH 2 PO 4 pH 6.8, 277 K........................................................................40 Table 2.2: NMR resonance assignments for NH 2 -HCGGPC-CONH 2 , 100 mM NaH 2 PO 4 pH 4.3, 277 K........................................................................44 Table 2.3: NMR resonance assignments for NH 2 -GCGGPC-CONH 2 , 100 mM NaH 2 PO 4 pH 6.8, 298 K........................................................................47 Table 2.4: Maxima energies and corresponding  values for the model complexes compared to the reduced protein, and previously reported model complexes, [Ni II (SOD M1 )] 62 and K 2 [Ni(CGC)]. 60 ............49 Table 2.5: NMR resonance assignments for [Ni(HCGGPC)] - , 100 mM NaH 2 PO 4 pH 6.8, 277 K........................................................................56 Table 2.6: NMR resonance assignments [Ni(GCGGPC)] - 100 mM NaH 2 PO 4

pH 6.8, 298 K.......................................................................................61 Table 2.7: XANES parameters of [Ni(HCGGPC)] - compared to K 2 [Ni(CGC)] and the reduced enzyme........................................................................65 Table 2.8: Summary of low energy isomer DFT calculations.....................................73 Table 2.9: SOD activity of [Ni(HCGGPC)] - relative to the enzyme and other reported peptide models........................................................................83 Table 3.1 Crystallographic data for (5-hexenyl)Co(dmgBF 2 ) 2 (py) and (cyclopentylmethyl)Co(dmgBF 2 ) 2 (py)................................................112 Table 3.2 Crystallographic data for (neopentyl)Co(dmgBF 2 ) 2 (py)......................113 Table 3.3: Crystallographic data for [(triphos)Ni(PPh 3 )(H)]PF 6 and [(triphos)Ni(PPh 3 )]OTf.......................................................................114 Table 3.4 Crystallographic data for (triphos)Ni(P(m-tolyl)) 3 ...............................115 Table 3.5: Observed rate constants of alkyl transfer for the methyl, ethyl, and isopropyl-Co(dmgBF 2 ) 2 py..................................................................123

xi Table 3.6: The bond lengths of the nickel-triphos complexes of different oxidation states. Ni(0) = (triphos)Ni(PPh 3 ), Ni(I) = [(triphos)Ni(PPh 3 )] + , Ni(II) = [(triphos)Ni(PPh 3 )(H)] + ........................139

xii

LIST OF FIGURES Figure 1.1: Active site structure of urease with urea bridging the two nickel centers, a proposed mechanistic intermediate. 24 .....................................5 Figure 1.2: A) Active site structure of NiFe hydrogenase with H ligand modeled as a bridging hydride which has not been spectroscopically observed. 28 B) Cofactor F 430 found in the active site of methyl- CoM reductase. 26 ...................................................................................6 Figure 2.1: (A) Ni SOD subunit and (B) hexamer structure looking down the by-pyramidal axis. Reprinted with permission from Biochemistry. Getzoff and coworkers. 1 ......................................................................14 Figure 2.2: A) Ni SOD active site structure. The superoxide anion is proposed to bind directly to the nickel in the axial position with hydrogen bonding to the Y9 side chain and the peptide backbone stabilizing the interaction. B) The space-filling model showing substrate channel that directs the substrate to the open coordination site of the nickel. Reprinted with permission from Biochemistry. Getzoff and coworkers. 1 ...................................................................................15 Figure 2.3: Proposed mechanism of Ni SOD catalysis. 1 ............................................19 Figure 2.4: Acyl formation and transfer using the Ni(CGC)Ni(dppe)(Me) model system.................................................................................................22 Figure 2.5: Synthetic Ni SOD models reported by Harrop. 66 .....................................24 Figure 2.6: NH 2 -HCGGPC-CONH 2 peptide sequence...............................................33 Figure 2.7: NH 2 -GCGGPC-CONH 2 peptide sequence...............................................33 Figure 2.8: Analytical HPLC trace of NH 2 -HCGGPC-CONH 2 . Linear gradient 0-30% Buffer B over 30 minutes. NH 2 -HCGGPC- CONH 2 elutes at 11-12% Buffer B......................................................34 Figure 2.9: High Resolution ESI Mass Spectrum of [NH 2 -HCGGPC-CONH 2 ] +

detected in positive ion mode. Left shows experimentally observed isotope pattern, right calculated spectrum..............................35

xiii Figure 2.10: 1-D 1 H spectrum of NH 2 -HCGGPC-CONH 2 . 100 mM NaH 2 PO 4

pH 6.8, 277 K. *denotes node due to water resonance suppression.......36 Figure 2.11: NH 2 -HCGGPC-CONH 2 TOCSY spectrum. 100 mM NaH 2 PO 4 pH 6.8, 277 K............................................................................................37 Figure 2.12: NH 2 -HCGGPC-CONH 2

13 C-HSQC spectrum. 100 mM NaH 2 PO 4

pH 6.8, 277 K. Inset shows the imidazole resonances..........................38 Figure 2.13: NH 2 -HCGGPC-CONH 2

15 N-HSQC spectrum. 100 mM NaH 2 PO 4

pH 6.8, 277 K......................................................................................39 Figure 2.14: 1 H NMR spectrum of the amide region of the NH 2 -HCGGPC- CONH 2 peptide at pH 4.3 in red, at pH 6.8 spectrum in blue................41 Figure 2.15: TOCSY spectrum, amide region, of NH 2 -HCGGPC-CONH 2 at pH 4.3.......................................................................................................42 Figure 2.16: 15 N-HSQC spectrum of NH 2 -HCGGPC-CONH 2 at pH 4.3....................43 Figure 2.17: NH 2 -GCGGPC-CONH 2 TOCSY spectrum: 100 mM NaH 2 PO 4 pH 6.8, 298 K............................................................................................45 Figure 2.18: NH 2 -GCGGPC-CONH 2

13 C-HSQC of spectrum. 100 mM NaH 2 PO 4 pH 6.8, 298 K......................................................................46 Figure 2.19: High Resolution ESI Mass spectrometry of [Ni(HCGGPC)] -

detected in negative ion mode. Left, observed spectrum, right shows the calculated spectrum.............................................................48 Figure 2.20: Absorption spectra showing ligand field transitions of Ni(HCGGPC), Ni(GCGGPC), K 2 [Ni(CGC)]. 60 ....................................49 Figure 2.21: One dimensional NMR of apo in red and holo peptide in blue. 100 mM NaH 2 PO 4 pH 6.8, 277 K. * denotes node due to water resonance suppression.........................................................................50 Figure 2.22: Overlay of NH 2 -HCGGPC-CONH 2 and [Ni(HCGGPC)] - TOCSY spectra. 100 mM NaH 2 PO 4 pH 6.8, 277 K...........................................51 Figure 2.23: [Ni(HCGGPC)] -

13 C-HSQC spectrum. 100 mM NaH 2 PO 4 pH 6.8, 277 K. Inset shows the resonances for the His1 side chain..................52 Figure 2.24: [Ni(HCGGPC)] -

15 N-HSQC spectrum. 100 mM NaH 2 PO 4 pH 6.8, 277 K. Inset shows the N-terminal resonance......................................53

xiv Figure 2.25: Aliphatic region of the NOESY spectrum of [Ni(HGCGGPC)] -

Lines highlight the cross-peaks between H’s of Gly5 and H’s of Pro5 consistent with trans proline conformation..............................55 Figure 2.26: [Ni(GCGGPC)] - TOCSY spectrum. 100 mM NaH 2 PO 4 pH 6.8, 277 K. Lines highlight N-terminal cross-peaks...................................57 Figure 2.27: [Ni(GCGGPC)] -

13 C-HSQC spectrum. 100 mM NaH 2 PO 4 pH 6.8, 277 K..................................................................................................58 Figure 2.28: pH-dependent change in chemical shift for the imidazole C-H’s. Curves show Henderson-Hasselbach fit to the data..............................63 Figure 2.29: XANES spectrum of [Ni(HCGGPC)] - ...................................................65 Figure 2.30: Structure of model [Ni(HCGGPC)] - as determined by UV/Vis, NMR, and XAS spectroscopies...........................................................66 Figure 2.31: Electronic absorption spectra showing ligand field transitions of Ni(HCALPC), Ni(HCGGPC), Ni(GCGGPC), and K 2 [Ni(CGC)]. 60 ......68 Figure 2.32: Apo NH 2 -HCALPC-CONH 2 peptide in red and holo NH 2 - HCALPC-CONH 2 in blue. * denotes node due to water resonance suppression.........................................................................69 Figure 2.33: DFT constructed model of the lowest energy isomer structure: trans proline with protonated C6 thiolate.............................................72 Figure 2.34: Ni 2+ titration of NH 2 -HCGGPC-CONH 2 monitored by 1 H NMR spectroscopy. Only the amide region is shown for clarity....................74 Figure 2.35: Direction titration of NH 2 -HCGGPC-CONH 2 with Ni 2+ . Curvature of the titration data is highlighted with log[Ni 2+ ] plot shown on the lower right.....................................................................................75 Figure 2.36: Back-titration of [Ni(HCGGPC)] - with NTA........................................77 Figure 2.37: Titration of [Ni(HCGGPC)] - with Zn 2+ . The decrease in absorbance for [Ni(HCGGPC)] - upon addition of Zn 2+ .........................79 Figure 2.38: Titration of [Ni(HCGGPC)] - with Co 2+ monitoring the decrease in absorbance for the NiS charge transfer band at 265 nm. The increasing absorbance at 518 nm is due to aqueous Co 2+ ......................80

xv Figure 2.39: EPR spectrum of [Ni(III)HCGGPC] 2- recorded at 77 K with 100 mM NaH 2 PO 4 pH 11...........................................................................81 Figure 2.40: Superoxide dismutase activity measured by KO 2 system. The percent NBT reduction is shown as a function of sample concentration, Apo peptide, Ni 2+ , and [Ni(HCGGPC)] 2+ .....................83 Figure 3.1: Wood-Lungdahl pathway of autotrophic carbon fixation. 107 ...................91 Figure 3.2: Crystal structure of ACS shown in the closed conformation (a) where the yellow and green domains swing inward to cover the A cluster, and the open conformation (b). Reprinted with permission. 109 ......................................................................................93 Figure 3.3: Structure of the C cluster active site.......................................................94 Figure 3.4: A Cluster Active Site Structure. The ligand denoted by L has not been crystallographically identified and is proposed be the site of substrate binding during catalysis........................................................95 Figure 3.5: Ragsdale’s proposed mechanism of ACS................................................97 Figure 3.6: Proposed mechanism with Ni(0) as a catalytic intermediate....................98 Figure 3.7: Acyl formation and transfer using the Ni(CGC)Ni(dppe)(Me) model system. 63 ............................................................................................101 Figure 3.8: Methyl transfer using cobaloxime model system. 127 .............................102 Figure 3.9: Linear relationship of k obs vs. [Ni] indicating a first order dependence on [Ni]. Error bars indicate standard error in k obs ............118 Figure 3.10: Plot of k obs vs. 1/[PPh 3 ] indicating an inverse first order dependence on the [PPh 3 ]. Error bars indicate standard error in k obs .....................................................................................................119 Figure 3.11: Plot of 1/k obs vs. [PPh 3 ]/[CoMe] where y-intercept = 1/k 1 and the slope = k 2 K eq. Error bars indicate standard error in 1/k obs ...................120 Figure 3.12: Solvent dependence on the rate of methyl transfer. Plot of k obs vs. [Ni] in THF, in toluene......................................................................122 Figure 3.13: The time course of the reaction of (isopropyl)Co(dmgBF 2 ) 2 py and (triphos)Ni(PPh 3 ) monitored by 1 H NMR spectroscopy......................124

xvi Figure 3.14: Molecular structure of (neopentyl)Co(dmgBF 2 ) 2 (py). Crystallographic details are given in Table 3.2...................................125 Figure 3.15: 1 H NMR spectrum of (neopentyl)Co(dmgBF 2 ) 2 py in CD 3 CN. * denotes residual solvent resonances, H 2 O and CH 3 CN.......................126 Figure 3.16: 1 H NMR spectra of neopentyl transfer reaction. Red spectrum shows the transfer reaction after 15 minutes and the blue shows the reaction after 18 hours, ~50% conversion observed......................127 Figure 3.17: 31 P NMR spectrum of neopentyl transfer reaction. * denotes [Co(PPh 3 )(dmgBF 2 ) 2 ] - .......................................................................128 Figure 3.18: High resolution ESI spectrum of [(triphos)Ni(Np)] + . Inset shows theoretically calculated spectrum. Isotope pattern consistent with molecular formula.............................................................................129 Figure 3.19: The 1 H resonance of the Ni-H complex observed in the ethyl transfer reaction in red and the isopropyl transfer reaction in blue......130 Figure 3.20: 1 H NMR spectrum of the [(triphos)Ni(PPh 3 )(H)]PF 6 in the hydride region with coupling outlined using a splitting tree diagram. The coupling constants were determined using a global fitting option of Mestrec software to determine peak positions and intensities.........132 Figure 3.21: 31 P NMR spectrum of the [(triphos)Ni(PPh 3 )(H)]PF 6 ..........................133 Figure 3.22: Molecular structure of [(triphos) Ni(PPh 3 )(H)]PF 6 . Thermal ellipsoid are drawn to 50% probability. Counter ion and hydrogen atoms, except the hydride ligand, have been omitted for clarity. Crystallographic details are given in Table 3.3.......................134 Figure 3.23: High resolution ESI mass spectrum of [(triphos)Ni(PPh 3 )(H)] +

shown on the left and the theoretically calculated spectrum on the right. Isotope pattern consistent with molecular formula...................135 Figure 3.24: 31 P{ 1 H} NMR spectra of [(triphos)Ni(PPh 3 )(H)]PF 6 with heating to 60 °C for 18 hours. Red spectrum denotes time 0 and blue spectrum measured after 18 hours. * denoted PF 6 resonance remains unchanged............................................................................137 Figure 3.25: Molecular structure of [(triphos)Ni(PPh 3 )]OTf. Counter ion and hydrogen atoms have been omitted for clarity. Crystallographic details are given in Table 3.3.............................................................138

xvii Figure 3.26: Molecular structure of (5-hexenyl)Co(dmgBF 2 ) 2 (py). Hydrogen atoms have been omitted for clarity. Crystallographic details are given in Table 3.1..............................................................................142 Figure 3.27: Molecular structure of (cyclopentylmethyl)Co(dmgBF 2 ) 2 (py). Hydrogen atoms have been omitted for clarity. Crystallographic details are given in Table 3.1.............................................................143 Figure 3.28: UV/Vis spectrum of the reaction between (5- hexenyl)Co(dmgBF 2 ) 2 (py) and (triphos)Ni(PPh 3 ) after 36 hours at room temperature. Inset shows Job plot verifying 1:1 reaction stoichiometry.....................................................................................144 Figure 3.29: ESI MS of 5-hexenyl transfer reaction in positive mode......................145 Figure 3.30: [(triphos)Ni(CH 3 )]OTf, [(triphos)Ni(COCH 3 )]OTf observed after addition of excess CO. * denotes solvent peaks: pentane, ethyl ether. [(triphos)]Ni(CH 3 )OTf reformed after removal of CO by vacuum..............................................................................................149 Figure 3.31: 31 P{ 1 H} NMR spectra of the [(triphos)Ni(CH 3 )]OTf in red and the [(triphos)Ni(COCH 3 )]OTf in blue......................................................150

xviii

LIST OF SCHEMES Scheme 2.1: Fenton reaction generating ROS…………………………………….…11 Scheme 2.2: Reaction catalyzed by SOD……………………………………….……11 Scheme 3.1: Reactions catalyzed by COdH/ACS………..………………………….91

Scheme 3.2: Chiral methyl transfer in ACS system as determined by Floss and coworkers………………………………………………………….99

Scheme 3.3: Methyl transfer from MeCo(dmgBF 2 ) 2 py to Ni(triphos)(PPh 3 )…..…..103

Scheme 3.4: Proposed mechanistic scheme for methyl transfer……………………117

Scheme 3.5: Synthesis of [(triphos)Ni(PPh 3 )(H)]PF 6 ………………………….…..131

Scheme 3.6: Decomposition of the hydride complex via homolytic bond cleavage to form Ni(I)………………………………………………...136

Scheme 3.7: Decomposition pathway yielding the [Ni(triphos)(PPh 3 )] + ……….….138

Scheme 3.8: Products of the 5-hexenyl transfer via an S N 2 or electron transfer mechanism…………………………………………………...140

xix LIST OF ABBREVIATIONS 18-crown-6 1,4,7,10,13,16-hexaoxacyclooctadecane ACS acetyl coenzyme A synthase COdH carbon monoxide dehydrogenage

CFeSP corrinoid iron-sulfur protein DIEA diisopropylethylamine DMF N,N-dimethylformamide dmgBF 2 (difluoroboryl)dimethylglyoximato DMSO dimethylsulfoxide dppe diphenylphosphinoethane Et ethyl Fmoc fluorenylmethyloxycarbonyl HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid HSQC heteronuclear single quantum coherence spectroscopy i Pr isopropyl KO 2 potassium superoxide Me methyl NBT nitrobluetetrazolium chloride NOESY nuclear overhauser spectroscopy NTA nitrilotriacetic acid OTf trifluoromethanesulfonate PPh 3 triphoenylphosphine py pyridine ROS reactive oxygen species SOD superoxide dismutase O 2 • - superoxide radical TCEP tris-(2-carboxyethyl)phosphine TFA trifloroacetic acid THF tetrahydrofuran TOCSY total correlation spectroscopy TSP trimethylsilylpropionic acid triphos bis(2-diphenylphosphinoethyl)phenyl phosphine XAS X-ray absorption spectroscopy XO xanthine oxidase

xx ABSTRACT Synthetic model complexes of nickel-containing enzyme active sites are important tools for understanding the structure-function properties of the active sites as well as for investigating the enzyme mechanisms. The synthesis and reactivity of structural mimics and catalytic-intermediate analogs can provide insight into how the diverse chemistry of nickel-containing enzymes occurs. Reported herein are two synthetic models of redox active nickel-dependant metalloproteins, nickel-containing superoxide dismutase and acetyl coenzyme A synthase. Nickel-containing superoxide dismutase, Ni SOD, contains a mononuclear nickel active site found at the N-terminus with active site ligands found within the first six residues of the protein sequence. 1 The active site sequence is unstructured in the absence of Ni 2+ providing a novel target for synthetic modeling. A Ni SOD model system was designed utilizing short peptides as ligands based on the native active site amino acid sequence, NH 2 -His-Cys-Gly-Gly-Pro-Cys-COOH. The system is used to investigate the minimum construct necessary to establish the native peptide fold and metal coordination. The [Ni(HCGGPC)] - complex was determined to be monomeric by mass spectrometry and optical spectral titrations with K d = 4.7 ± 0.4 x 10 –12 M. One- and two-dimensional NMR spectral data suggests the nickel coordination sphere is the same as found in the enzyme. Metal selectivity of the peptide was measured comparing Ni 2+ , Zn 2+ , and Co 2+ . Ni 2+ is preferred two orders of magnitude over Zn 2+

and three orders of magnitude over Co 2+ . The reactivity of the model complex with O 2 .

is similar to that observed for the enzyme. 1 Reaction of the complex with substoichiometric KO 2 results in a Ni 3+ species with a similar EPR spectrum as that

xxi observed in the enzyme. The superoxide dismutase activity of the nickel complex was measured to be, IC 50 =191 ± 9 μM much reduced compared to the native enzyme. The decreased activity is due to the product-catalyzed decomposition of the complex suggesting substrate channeling and the active site pocket of the enzyme provide essential oxidative protection of the active site thiols. Acetyl coenzyme A synthase, ACS, catalyzes the production of acetyl coenzyme A from one-carbon units, CO and CH 3 . A recently proposed ACS catalytic mechanism involves a zero valent nickel as the complex to accept a methyl group from methylcobalt species during catalytic turnover. 2 Synthetic analog studies were designed to probe the feasibility of Ni(0) as the methyl acceptor. Reaction of [triphos]Ni(PPh 3 ) (triphos = bis(2-diphenylphosphinoethyl)phenyl phosphine) with CH 3 Co(dmgBF 2 ) 2 py proceeds quantitatively yielding [Ni(triphos)(Me)] + and Co(I). 3 Kinetic data of methyl transfer and the rate of other alkyls (Et, i Pr) suggests that the mechanism of alkyl transfer proceeds via an S N 2 pathway similar to the ACS enzyme. The nickel-alkyls complexes possessing -hydrogen undergo -hydrogen elimination forming a nickel-hydride complex characterized by NMR, FT-IR spectroscopies and X-ray crystallography. This system is also used to model another intermediate along the catalytic pathway, the Ni-acyl complex, to help determined the importance of the order of substrate binding to the ACS active site during catalysis.

1 1 CHAPTER 1 INTRODUCTION 1.1 Nickel in Biology In 1926, James Sumner published the first crystallization of an enzyme, the nickel-containing enzyme, urease. 4 However, nickel was not identified as the active site metal at the time. It was not until 1975 that nickel was discovered as the cofactor for urease and therefore, a biologically significant metal. 5 In the 35 years since the latter discovery, nickel has been identified as an essential nutrient for plants, bacteria and archea. 6 Nickel has been proposed as an essential element for animals as well, although to date, the functional role of nickel in these organisms is unknown. 7

Currently, there are nine known nickel containing enzymes: urease, NiFe hydrogenase, CO dehydrogenase, acetyl-CoA synthase, methyl-coenzyme M reductase, glyoxalase I, aci-reductone dioxygenase, methylenediurease, and nickel-containing superoxide dismutase. 8

Nickel is used in Nature to catalyze a wide range of chemical transformations in a wide variety organisms. 8,9 There are many unanswered questions regarding why nickel is chosen as the active site metal above other more abundant metals and what the structure-function relationships are for nickel-containing active sites. In the field of nickel biochemistry, synthetic model studies are essential tools for understanding the diverse chemistry of nickel-containing enzymes. 1.1.1 Nickel Bioavailability The environmental concentrations of nickel (0.0018% of the earth’s crust), are below other biologically important metals such as zinc (0.0054%), manganese

2 2 (0.053%) and iron (3.09%). 10 However, in the proposed environment where life originated, concentrations of nickel are thought to have been much higher, with the metal found primarily in the form of Ni-Fe-S rich minerals. Nickel-containing enzymes are prominent in autotrophic pathways, i.e. CO 2 , CH 4 and H 2 metabolism, found in very primitive organisms suggesting the evolution of nickel bio-catalysis might have been among the earliest metabolic pathways. 11

The concentrations of nickel necessary for dependent organisms, plants, and bacteria, are very low. 6 The different ecospheres have sufficient nickel concentrations for these organisms’ needs. As a result, symptoms of nickel deficiency are rare and were unidentified until recently. 6,12 Nickel is present in the earth’s crust at a reported concentration of 86 mg/kg. 13 It is most abundant in soil samples where concentrations are found up to 1 mg/g. Seawater has the lowest concentration ranging from 0.1-0.6 μgL -1 (2 μM). 6,13 The seawater nickel concentrations are higher than that of iron (<<1 nM), a fact that is proposed to be causative in the evolution of the nickel-containing superoxide dismutase in marine organisms as a replacement of the iron-containing superoxide dismutase. 14

Nature has developed intricate nickel uptake, storage, and chaperone systems for enzyme active assembly to specify nickel incorporation into proteins. Numerous proteins have been identified that function in nickel homeostasis including nickel transport proteins, NikABCDE, an ATP-dependent membrane transport protein, and NiCoT, permease proteins. 15 Four nickel chaperone proteins have been identified from the urease gene cluster and have been shown to be essential to urease activation. 16 A chaperone system is also proposed to aid in peptide folding as well as nickel insertion in the active site of nickel-containing superoxide dismutase. 1

3 3 1.1.2 Nickel-dependent Enzymatic Pathways Nickel is involved in diverse number of enzymatic pathways depending on the organism. 6,13 In plants, nickel dependence results mainly from the enzyme urease. 17 Urease catalyzes the degradation of urea to ammonia and carbon dioxide, thus providing nitrogen for amino acid synthesis. 5,9 Bacteria contain the highest number of nickel-containing enzymes. Various species of bacteria use nickel ions in the active sites of enzymes essential to the uptake of cellular carbon and production of cellular energy. 18 These enzymes are acetyl coenzyme A synthase, carbon monoxide dehydrogenase, methyl-coenzyme M reductase and hydrogenase. The pathogen Helicobacter pylori uses the production of ammonia by urease to neutralize the stomach acids of its host making the extreme environment liveable. Helicobacter pylori colonization can cause gastric ulcers for the host. 6 The soil bacteria Streptomyces and cyanobacteria also contain a nickel-containing superoxide dismutase, which protects the organism from oxidative damage. 19

The nickel dependency of animals has been suggested, but the biological function of nickel is not identified. 6 Nickel is proposed to be involved in the methionine-folate cycle based on several studies in rats, pigs and humans. 20 In a study with pigs, nickel supplementation was shown to partially recover the symptoms of B 12

deficiency including decreasing homocysteine levels and reversing iron accumulation in the liver. 21 Recently, a human study yielded similar results where patients receiving hemodialysis treatment were shown to have an inverse relationship between nickel concentrations in blood plasma and concentrations of homocysteine. 22 These studies suggest nickel is involved in homocysteine regulation possibly via the methionine- folate cycle. However, the proteins involved in such regulation are unknown. Hyperhomocysteinemia, increased levels of homocysteine in the blood, is a serious

4 4 health problem associated with increased risk of coronary artery disease. 23 Identifying the nickel pathway of involvement in homocysteine regulation could provide a novel treatment target. 1.2 Nickel-containing Enzyme Active Sites The nine known nickel-containing enzymes catalyze a diverse range of reactions that utilize nickel’s varied coordination numbers, geometries and oxidation states. 8 The most common oxidation state found in Nature is Ni(II) but Ni(I) and Ni(III) are also observed. The Ni(0) oxidation state has not been spectroscopically observed, but is proposed as a catalytic intermediate in acetyl coenzyme A synthesis. 2

Nickel is a moderately polarizable metal, which makes it suitable for complexing a variety of amino acid side chain and main chain heteroatoms, thus tuning the reactivity of the metal centers. Nickel can act as a Lewis acid or a redox active center depending on the coordination sphere. Although the bioavailability of nickel is much less than other biologically relevant divalent metals such as Zn 2+ or Mn 2+ , nickel is still used as a Lewis acid cofactor for metalloproteins. 9 For example, the active site of urease contains two nickel ions that are proposed to be bridged via the urea substrate during catalysis, Figure 1.1. 24 The exact positions of the active site residues and water molecules are important for catalysis. This hypothesis is supported by the crystal structure of the inactive Mn-containing enzyme where the only differences observed from the active enzyme are that the water molecules were found at longer distances from the metal, 0.2 Å longer, and disorder in the ligated aspartic acid. 25 The nickel ions are required for catalysis possibly because nickel is the exact size necessary to orient the active site structure.

5 5 Ni N N O O NH Ni O O N N O N N N N NH O NH 3 +

Figure 1.1: Active site structure of urease with urea bridging the two nickel centers, a proposed mechanistic intermediate. 24

In the enzymes NiFe hydrogenase, ACS, COdH, methyl-CoM reductase, and Ni SOD, the nickel ion is redox active. 6,9 The redox potentials required for these various enzymes range from +0.89 to -0.60 V. 8 These redox potentials are tuned by the identity of the ligands and the coordination sphere. For example, electron-rich thiolates and/or anionic amide nitrogens are found in the active sites of Ni SOD 1 and hydrogenase, 9 Figure 1.2 (A). The active site of methyl-CoM reductase contains the coenzyme F 430 , an electron rich hydrocorphin macrocycle, Figure 1.2 (B). 26 These functionalities help to stabilize high oxidation state, nickel(III) intermediates, proposed in their catalytic mechanisms. Relatively electron poor ligands such as bridging thiolates, are used to stabilize lower oxidation states, i.e. nickel(I) and perhaps nickel(0), in the active site of acetyl coenzyme A. 27

6 6

Ni N N N N COOH COOH COOH HN O HOOC H 2 N HOOC H H CH 3 H 3 C H O O

Figure 1.2: A) Active site structure of NiFe hydrogenase with H ligand modeled as a bridging hydride which has not been spectroscopically observed. 28 B) Cofactor F 430 found in the active site of methyl-CoM reductase. 26

1.3 Model Complexes of Nickel-containing Active Sites Synthetic model studies of metalloenzyme active sites have provided much insight into enzyme structure-function relationships. 29 Small molecule model studies can help to understand the properties of the ligand sets that diversify the chemical transformations catalyzed by nickel-containing enzymes. Small synthetic models provide a strategy for systematic changes in structure and electronic properties of ligands to determine these structure-function relationships. Model complexes provide a framework for investigating enzyme mechanisms by the synthesis and characterization of proposed catalytic intermediates analogs. These model studies provide insight for the development of industrial catalysts based on biocatalysis, especially with the current focus on green methodologies. They are also important in the development of novel therapies. In the future, these sorts of studies could aid in novel protein engineering, gene therapies and well as small molecule therapeutics. A) B) Ni S Fe S CN CO C N S S H

Full document contains 189 pages
Abstract: Synthetic model complexes of nickel-containing enzyme active sites are important tools for understanding the structure-function properties of the active sites as well as for investigating the enzyme mechanisms. The synthesis and reactivity of structural mimics and catalytic-intermediate analogs can provide insight into how the diverse chemistry of nickel-containing enzymes occurs. Reported herein are two synthetic models of redox active nickel-dependant metalloproteins, nickel-containing superoxide dismutase and acetyl coenzyme A synthase. Nickel-containing superoxide dismutase, Ni SOD, contains a mononuclear nickel active site found at the N-terminus with active site ligands found within the first six residues of the protein sequence.1 The active site sequence is unstructured in the absence of Ni2+ providing a novel target for synthetic modeling. A Ni SOD model system was designed utilizing short peptides as ligands based on the native active site amino acid sequence, NH2 -His-Cys-Gly-Gly-Pro-Cys-COOH. The system is used to investigate the minimum construct necessary to establish the native peptide fold and metal coordination. The [Ni(HCGGPC)]- complex was determined to be monomeric by mass spectrometry and optical spectral titrations with K d = 4.7 ± 0.4 x 10-12 M. One- and two-dimensional NMR spectral data suggests the nickel coordination sphere is the same as found in the enzyme. Metal selectivity of the peptide was measured comparing Ni 2+ , Zn2+ , and Co2+ . Ni2+ is preferred two orders of magnitude over Zn2+ and three orders of magnitude over Co2+ . The reactivity of the model complex with O2·- is similar to that observed for the enzyme.1 Reaction of the complex with substoichiometric KO2 results in a Ni3+ species with a similar EPR spectrum as that observed in the enzyme. The superoxide dismutase activity of the nickel complex was measured to be, IC50 =191 ± 9 μM much reduced compared to the native enzyme. The decreased activity is due to the product-catalyzed decomposition of the complex suggesting substrate channeling and the active site pocket of the enzyme provide essential oxidative protection of the active site thiols. Acetyl coenzyme A synthase, ACS, catalyzes the production of acetyl coenzyme A from one-carbon units, CO and CH3 . A recently proposed ACS catalytic mechanism involves a zero valent nickel as the complex to accept a methyl group from methylcobalt species during catalytic turnover.2 Synthetic analog studies were designed to probe the feasibility of Ni(0) as the methyl acceptor. Reaction of [triphos]Ni(PPh3 ) (triphos = bis(2-diphenylphosphinoethyl)phenyl phosphine) with CH3 Co(dmgBF2 )2 py proceeds quantitatively yielding [Ni(triphos)(Me)]+ and Co(I). 3 Kinetic data of methyl transfer and the rate of other alkyls (Et, i Pr) suggests that the mechanism of alkyl transfer proceeds via an SN 2 pathway similar to the ACS enzyme. The nickel-alkyls complexes possessing β-hydrogen undergo β-hydrogen elimination forming a nickel-hydride complex characterized by NMR, FT-IR spectroscopies and X-ray crystallography. This system is also used to model another intermediate along the catalytic pathway, the Ni-acyl complex, to help determined the importance of the order of substrate binding to the ACS active site during catalysis.