xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader ntm 2200385Ka 4500
controlfield tag 001 002324023
008 101007s2009 xx a ob 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003045
b T931p 2009
Turbeville, Tracy D.
PLP-dependent [alpha]-oxoamine synthases :
phylogenetic analysis, structural plasticity, and structure-function studies on 5-aminolevulinate synthase /
by Tracy D. Turbeville.
x, 183 leaves :
Dissertation (Ph.D.)--University of South Florida, 2009.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
ABSTRACT: 5-Aminolevulinate synthase (ALAS) and 8-amino-7-oxononanoate synthase (AONS) are two of four homodimeric members of the alpha-oxoamine synthase family of pyridoxal 5-phosphate (PLP)-dependent enzymes. The evolutionary relationships among alpha-oxoamine synthases representing a broad taxonomic and phylogenetic spectrum have been examined to help identify residues that may regulate substrate specificity. The structural plasticity of ALAS has been documented in studies of functional circularly permuted ALAS variants and the single polypeptide chain ALAS dimer (ALAS/ALAS) exhibiting a greater turnover number than wild-type ALAS. An examination of the contribution of each ALAS/ALAS active site to the enzymatic activity shows that each active site makes distinct contributions to the steady-state activity of the enzyme. Chimeric ALAS/AONS proteins exhibited an oligomeric structure with two sites having ALAS activity and two sites having AONS activity.Remarkably, the steady-state rates for both the ALAS and AONS activities were lower than that observed in the parent enzymes, while the reactivity of the ALAS sites in ALAS/AONS was similar to that of wild-type ALAS. We propose that the different contribution of each active site to the steady-state activity of ALAS/ALAS and the reduced steady-state activities of the ALAS/AONS chimera, compared to the parent enzymes, relate to different extents of conformational changes associated with product release due to the strain caused with the linking the two ALAS (or ALAS and AONS) subunits. Thus, the extensive plasticity seen in ALAS extends to another member of the alpha-oxoamine family, AONS. In the alpha-oxoamine synthase family a conserved histidine hydrogen bonds with the phenolic oxygen of PLP and may be significant for substrate-binding, PLP-positioning, and maintaining the pKa of the imine nitrogen.The replacement of this conserved histidine, H282, with alanine in murine erythroid ALAS has multiple effects on the spectral, binding, and kinetic properties of the enzyme and supports the conclusion that H282 plays multiple roles in the enzymology of ALAS. Altogether, these results imply that amino acid H282 coordinates the movement of the pyridine ring with the reorganization of the active-site hydrogen bond network and acts as a hydrogen bond donor to the phenolic oxygen to maintain the protonated Schiff base and enhance the electron sink function of the PLP cofactor.
Advisor: Gloria C. Ferreira, Ph.D.
Pyridoxal 5 phosphate
x Molecular Medicine
t USF Electronic Theses and Dissertations.
PLP-Dependent -Oxoamine Synthases: Phylogenetic Analysis, Structural Plasticity, and Structure-Function Studies on 5-Aminolevulinate Synthase by Tracy D. Turbeville A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Molecular Medicine College of Medicine University of South Florida Major Professor: Gloria C. Ferreira, Ph.D. W. Lee Adair, Jr., Ph.D. R. Kennedy Keller, Ph.D. James Garey, Ph.D. Huntington Potter, Ph.D. Ann Smith, Ph.D. Date of Approval: June 29, 2009 Keywords: heme; pyridoxal; pyridoxa l 5Â’ phosphate; biotin; ALAS; AONS Copyright 2009, Tracy D. Turbeville
Dedication Â… to God who gave me the strength and perseve rance to continue when I wanted to give up and who has filled my life with the lo ve and support of many wonderful people. Â…to my grandma, Anna Turbeville, who taught me the value of love and compassion. Her memory continues to be a sour ce of inspiration and strength. Â… and to my fiance, Steve Colafrancesco, fo r his love, care, encouragement and support without which I would not have made it across the finish line.
Acknowledgements This dissertation would neve r have come to fruition without the support of many individuals, and it is with pleasure and gratitude that I acknowledge their efforts. I would like to express my utmost gratitude to my major professor, Dr. Gloria Ferreira, for her support as both a mentor and a friend. I also wish to thank the other members of my committee, Dr. Lee Adair, Dr. Ken Ke ller, Dr. James Garey and Dr. Huntington Potter, as well as, Dr. Greg Hunter for thei r advice and support ove r the course of my Ph.D. study. I also am very appreciative fo r the encouragement and counsel given by Dr. Clark Craddock, Dr. Peter Neame and Dr. Patricia Kruk. Over my extended tenure as a graduate stude nt, I have had the honor of working with a variety of persons and personalities that sa ved me from the tediousness that sometimes occurs in the pursuit of science. I tha nk you allÂ—Dr. Dave Chappell, Dr. Junshun Zhang, Matt Sampson, Chris Adams, Anna Fomi na, Meena Reddy, Arianna Mangravita, Michelle Grigsby, Matt Lopata, and Zhen Shi. In particular, I am most indebted to Dr. Anton Cheltsov and Dr. Thomas Lendrihas who made everyday an adventure. I am also thankful for the love, encouragem ent and support of my friends and family. I am forever grateful to my mother, Judy Tu rbeville, for the patience and understanding that she has demonstrated even when I did not make it easy, to my dad, Dewey Turbeville, and Andrea Turbeville for being my biggest fans, and to my siblings, Kim Fleeman, Dewey Turbeville, and Hunter Tu rbeville, for keeping me grounded. Of course, there are many more aunts, uncles, cousins, grandparents, nieces and nephews that have always been and continue to be a source of encouragement.
i Table of Contents List of Tables iii List of Figures iv List of Schemes vi List of Abbreviations vii Abstract ix Chapter One: Introduction 1 Historical perspectives on vitamin B6 1 Discovery of vitamins 2 Unraveling the vitamin B complex 6 Characterization of vitamin B6 as a cofactor and coenzyme 7 Medical significance of vitamin B6 9 Vitamin B6-dependent enzymes 15 Biotin 20 Biotin biosynthesis 20 Medical significance of biotin 21 Heme 27 Heme biosynthesis 27 Medical significance of hemoprot eins and heme biosynthesis 33 The medical significance of other -oxoamine synthases 41 Research Aims 42 References 44 Chapter Two: Sequence and phylogene tic analysis of PLP-dependent -oxoamine synthases and their role in identifyi ng residues regulating enzyme specificity 50 Introduction 50 Experimental Details 55 Results 59 Discussion 75 References 85
ii Chapter Three: Functional asymmetry fo r active sites of single chain homoand chimeric dimers of 5-aminolevulin ate synthase and 8-amino-7-oxononanoate synthase 90 Abstract 90 Introduction 91 Experimental Details 96 Results 106 Discussion 127 References 134 Chapter Four: Histidine-282 in 5-aminol evulinate synthase affects substrate binding and catalysis 138 Abstract 138 Introduction 139 Experimental Details 142 Results 149 Discussion 156 Supporting Information 172 References 173 Chapter Five: Summary and Conclusions 177 References 183 About the Author End Page
iii List of Tables Table 1.1 Vitamin B components 11 Table 1.2 PLP-dependent Enzymes 14 Table 1.3 The role of some hemoprot eins in biological processes 38 Table 1.4 Classification of porphyrias 40 Table 2.1 Aligned sequences 61 Table 2.2 Percent identity of aligned sequences 64 Table 2.3 Role of selected residues in ALAS structure 82 Table 3.1 Nomenclature defining plasmi ds and enzymes described in this report 99 Table 3.2 Growth of transformed E. coli strains on selective media 108 Table 3.3 Summary of steady-stat e kinetic parameters for ALAS, ALAS/ALAS, ALAS/ALASK313A and ALASK313A/ALAS 111 Table 3.4 ALAS activity: Summary of steady-state kinetic parameters for ALAS and ALAS/AONS chimera 125 Table 3.5 AONS activity: Summary of st eady-state kinetic parameters for AONS and ALAS/AONS chimera 125 Table 4.1 Summary of st eady-state kinetic parameters and dissociation constants 152 Table 4.2 Summary of p K values obtained from pH dependence of kinetic parameters 152
iv List of Figures Figure 1.1 Ribbon diagrams of representa tive enzymes of fold types I Â– IV 18 Figure 1.2 PLP-dependent enzymes of the -family of are homodimers 19 Figure 1.3 Biotin-depe ndent carboxylases 25 Figure 1.4 Biotin-dependent carboxylase in metabolism 26 Figure 1.5 Overview of te trapyrrole biosynthesis 31 Figure 2.1 Alignment of representati ve sequences of ALAS and KBL with AONS sequences. 62 Figure 2.2 Phylogenetic trees for the -oxoamine synthase family of PLPdependent enzymes 67 Figure 2.3 Growth of R872 cel ls in minimal media 69 Figure 2.4 Growth of R872 cells expressing ALAS variants 72 Figure 2.5 Comparison of cell growth of freshly transformed R872 cells 73 Figure 1.6 Succinyl-CoA binding pocket of R. capsulatus ALAS 83 Figure 2.7 Active site interactions with PLP cofactor and glycine substrate in R. capsulatus ALAS 84 Figure 3.1 Schematic representation of the expression plasmids 100 Figure 3.2 Absorption spectra of ALAS, ALAS/ALAS, ALASK313A/ALAS and ALAS/ALASK313A 109 Figure 3.3 Kinetics of a pre-steadystate burst of ALA product in the ALAS/ALASK313A and ALASK313A/ALAS reactions 114 Figure 3.4 Schematic representation illustrating the possible active site arrangements for ALAS/AONS 119
v Figure 3.5 Determination of the molecular mass of the ALAS/AONS chimera by gel filtration chromatography 121 Figure 3.6 Absorption and fluoresce nce spectra of ALAS, AONS and ALAS/AONS 122 Figure 3.7 Kinetics of a pre-steadystate burst of ALA product in the ALAS/AONS 126 Figure 4.1 Spatial position of active site residues in the R. capsulatus ALAS holoenzyme crystal structure 164 Figure 4.2 Absorption and fluorescence sp ectra of ALAS and H282A variant 165 Figure 4.3 The pH dependence of fluorescence emission 166 Figure 4.4 pH dependence of log kcat, log kcat/ Km Gly and log 1/ Km Gly for ALAS and H282A 167 Figure 4.5 Reaction of 60 M H282A variant with glycine 168 Figure 4.6 pH dependence of the Kd for glycine for ALAS and H282A 169 Figure 4.7 UV-visible absorption spectra of H282A in the presence of ALA and pH-dependence of ALA-quinonoid intermediate formation 170 Figure 4.8 Circular dichroism sp ectra of ALASand H282A-ligand complexes 171 Figure 4.9 Reaction of H282A variant with glycine 172
vi List of Schemes Scheme 1.1 Reactions of PLP-dependent enzymes 13 Scheme 1.2 Biotin 20 Scheme 1.3 Biotin biosynthesis 24 Scheme 1.4 Heme 27 Scheme 1.5 Heme biosynthesis 31 Scheme 2.1 ALASand AONScatalyzed reactions 54 Scheme 3.1 Reactions cataly zed by ALAS and AONS 92 Scheme 4.1 ALAS-catalyzed reaction 143 Scheme 4.2 ALAS glycine binding model 1 146 Schame 4.3 ALAS glycine binding model 2 146
vii List of Abbreviations AAT Aspartate Aminotransferase AD-P Aminolevulinate dehydratase porphyria AIP Acute Intermittent Porphyria ALAS 5-Aminolevulinate Synthase ALA 5-Aminolevulinate AONS 8-Amino-7-Oxononanoate Synthase AON 8-Amino-7-Oxononanoate AMPSO 3-([1,1-Dimethyl-2-Hydroxye thyl]amino)-2Hydroxypropane Sulfonic Acid BSA Bovine Serum Albumin CD Circular Dichroism CoA Coenzyme A CEP Congenital Erythr oblastic Porphyria CYPs Cytochrome P450s DAPA 7, 8-Diaminope largonic Acid DEAE Diethylaminoethyl DNA Deoxyribonucleic acid dNTP Deoxyribonuc leotide triphosphate EPP Erythropoieti c Protoporphyria EDTA Ethylenediaminetetraacetic acid GABA Gamma-Aminobutyric Acid
viii GluTR Glutamyl-tRNA Reductase GSAM Glutamate-1-Semialdehyde-2,1-Aminomutase HEPES N-(2-Hydroxyethyl) Pipera zine-NÂ’-(2-Ethane Sulfonic Acid) HPLC High Performance Liquid Chromatography JBC Journal of Biological Chemistry KBL 2-Amino-3-Ketobutyrate Ligase MCD Multiple Carboxylase Deficiency ME Minimum Evolution MEGAWHOP Megaprimer PCR of Whole Plasmid MOPS 4-Morpholinepropanesulfonic Acid MP Maximum Parsimony NAD+ -Nicotinamide Adenine Dinucleotide NJ Neighbor-Joining PBG Porphobilinogen PCR Polymerase Chain Reaction PLP Pyridoxal 5Â’-Phosphate PMSF Phenylmethylsulphonyl Fluoride PCoA Pimeloyl-Coenzyme A SCoA Succinyl-CoA SDS-PAGE Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis SPT Serine Palmitoyltransferase WT Wild-Type
ix PLP-Dependent -Oxoamine Synthases: Phylogenetic Analysis, Structural Plasticity, and Structure-Function St udies on 5-Aminolevulinate Synthase Tracy Dawn Turbeville Abstract 5-Aminolevulinate synthase (ALAS) and 8-amino-7-oxononanoate synthase (AONS) are two of four homodimeric members of the -oxoamine synthase family of pyridoxal 5Â’-phosphate (PLP)-dependent enzymes. The evolutionary relationships among oxoamine synthases representing a broad taxonomic and phylogenetic spectrum have been examined to help identify residues that may regulate substrate specificity. The structural plasticity of ALAS has been documented in studies of functional circularly permuted ALAS variants an d the single polypeptid e chain ALAS dimer (ALAS/ALAS) exhibiting a greater turnove r number than wild-type ALAS. An examination of the contribution of each ALAS /ALAS active site to th e enzymatic activity shows that each active site makes distinct c ontributions to the steady-state activity of the enzyme. Chimeric ALAS/AONS proteins exhi bited an oligomeric structure with two sites having ALAS activity and two site s having AONS activity. Remarkably, the steady-state rates for both the ALAS and AONS activities were lower than that observed in the parent enzymes, while the reactiv ity of the ALAS sites in ALAS/AONS was similar to that of w ild-type ALAS. We propose that th e different contribution of each active site to the steady-state activity of ALAS/ALAS and the reduced steady-state
x activities of the ALAS/AONS chimera, compared to the parent enzymes, relate to different extents of conformational changes associated with product release due to the strain caused with th e linking the two ALAS (or ALAS and AONS) subunits. Thus, the extensive plasticity seen in ALAS extends to another member of the -oxoamine family, AONS. In the -oxoamine synthase family a conserved histidine hydrogen bonds with the phenolic oxygen of PLP and ma y be significant for substr ate-binding, PLP-positioning, and maintaining the p K a of the imine nitrogen. The replacement of this conserved histidine, H282, with alanine in murine er ythroid ALAS has multip le effects on the spectral, binding, and kinetic properties of the enzyme and supports the conclusion that H282 plays multiple roles in the enzymology of ALAS. Altogether, these results imply that amino acid H282 coordinates the movement of the pyridine ring with the reorganization of the activ e-site hydrogen bond networ k and acts as a hydrogen bond donor to the phenolic oxygen to maintain th e protonated Schiff base and enhance the electron sink function of the PLP cofactor.
1 Chapter One Introduction Historical Perspectives on Vitamin B6 Cellular metabolism involves a series of multi-step reaction pathways which are necessary for the maintenance of life. Cata bolic pathways break down substances to yield energy, whereas anabolic pathways assemble complex products. The diverse reaction pathways are intertwined with each pa thway requiring a unique set of enzymes. Enzymes allow reactions to proceed rapidly and are essential to push thermodynamically unfavorable reactions by coupl ing them to favorable ones. While many of the enzymes involved in carrying out this vast array of ch emical reactions do not need any additional components, some require a non-protein constitu ent or cofactor for activity. Cofactors can be inorganic or organi c and many are essential for optimal health and disease prevention. Most vitamins are precursors fo r metabolically essent ial cofactors with diverse biochemical functions. The wide availability of vitamins in the form of inexpensive pills has contributed to a modern day fascination with vitami n supplementation by many, making it hard to imagine that a century ago the existence of such Â“accessory food factorsÂ” was heavily debated. The first vitamin was isolated in 1913; by the mid 1930Â’s the first multivitamin products were available in pharmacies and gr ocery stores and milk was fortified with vitamin D. Today an average of 35% of Am ericans report the recent use of multivitamins
2 (Panel, 2006). Many diseases associated with vitamin deficiencies that were commonplace a century ago are virtually unknown to todayÂ’s generation. Vitamin B6 is among the many vitamins in wh ich the nutritional importance has been recognized and the metabolic and biochemical significance has been well characterized. Vitamin B6 is a collective term for a group of pyridin e derivatives that serve as precursors for the coenzyme pyri doxal phosphate, which has been recognized as the cofactor for over 100 enzyme catalyzed reactions (Percudani and Peracchi, 2003). The scientific discoveries and developments leading to the isolation of vitamin B6 are common to all vitamins. Discovery of vitamins Â“Hard work on interesting problems is enjoyable and preferable to aimless wasting of leisure time. It may also lead to unexpected findings that give insights into important related problems. Such unexpected findingsÂ—sometimes calle d Â‘luckÂ’Â—frequently happen to the active researcher, but only rarely to those who prefer talk to study and work. So one should study and work hard, on interesting problems of any nature, with the purpose of explaining nature and helping others (Miles and Metzler, 2004).Â” Esmond Snell In the last days of the 19th century, when Christian Eijkman was sent to the Dutch colony in Java to identify the microbe responsible for beribe ri, a nervous system ailment that would eventually be linked to a deficiency of thiamine (vitamin B1) in the diet the link between diet and disease was almost unimaginable (Car penter, 2003). During this period, the recent sucesses of Robert Koch in identifying microorganisms responsible for
3 several diseases had scientist around the globe searching for other links between disease and microorganisms (Carpenter, 2003). Eijkma n attempted to replicate the disease and isolate the microbe by injecting blood from sick soldiers into animals, yet he was unable to replicate the disease or identify any asso ciated microorganism. While EijkmanÂ’s work did not find the source of beriberi, his ra ther serendipitous obs ervation in the 1890Â’s would eventually link beriberi to dietary insu fficiencies. He observed that Â“uninfectedÂ” chickens began exhibiting leg weakness (pol yneuritis), a characteristic observed in beriberi patients. He carefully characte rized the chickensÂ’ illness, including their unexpected spontaneous recove ry (Carpenter, 2003). Eijkman soon began an inquiry to unde rstand the chickensÂ’ extraordinary recovery. He would learn that, prior to th e appearance of the polyneuritis symptoms, the chickensÂ’ diet was changed from unpolished brown rice to polishe d white rice and the recovery of the chickens coin cided with the addition of brow n rice back into the diet. After a series of carefully designed experime nts, replicating the symptoms and recovery of the chickens, he confirmed that the observe d polyneuritis was direct ly attributable to diet (Carpenter, 2003). Although Eijkman mi stakenly reasoned that the white rice contained a toxin and the brown rice the antidot e or Â“the anti-beriberi factorÂ”, his inquiry into his chance observation would begin the tr ansformation of the nutritional paradigm of the period. EijkmanÂ’s health would prevent him from continuing hi s research and force him to return to Holland; however, he had laid the groundwork that led his successor and colleague, Gerret Grijn s, to theorize in 1901 that white rice "lacked a certain substance of importance in the metabolism of the central nervous system "(Rosenfeld, 1997; Carpenter, 2003).
4 Sir Frederick G. Hopkins had begun his ca reer during the same period as Eijkman when the prevailing nutritional paradigm wa s that calories alone were sufficient for animals to survive. His 1901 observation th at mice lacking the amino acid tryptophan died within a Â“fortnightÂ” (two weeks) would be Hopkins firs t challenge to this theory (Hopkins, 1912). Through a classic series of experiments, Hopkins unveiled evidence that there was an unknown constituent found in normal diets that was not found in a diet of purified proteins, carbohydrates, fa ts, and salts, in 1906 and 1907. However, it would be five years before Hopkins published his unorthodox findings in the 1912 article, "Feeding Experiments Illustrating the Importa nce of Accessory Food Factors in Normal Dietaries (Hopkins, 1912).Â” By the time Hopkins research was fina lly published, other sceintists had begun working to find the elusive Â“anti-beriberiÂ” fa ctor contained in the rice polishings. Among them was London chemist Casmir Funk, who believed he had isolated a nitrogencontaining factor that cured beriberi and assumed it to be an amine. Thus, Funk Â“suggested the name vitamine for it as being one of those nitrogenous substances, minute quantities of which are essential in the diet of birds, man and some other animals (Funk, 1912).Â” Although the anti-beriberi factor wa s only a contaminant and was eventually demonstrated to not be an amine, the name stuck (Carpenter, 2003). As long as the nature of Hopkins and Funk dietary components remained elusive, the existence of Â“vitaminesÂ” w ould remain in doubt. Fortunatel y, a great deal of attention was focused on the Â“vitamineÂ” question l eading to the two groups simultaneously reporting the discovery of the first acce ssory food substance in the Journal of Biochemistry in 1913 (McCollum, 1913; Osbor ne, 1913). Both groups identified a fat-
5 soluble organic substance that was an essential nutritional component for rats. It was termed fat-soluble A and was the first di etary constituent to be recognized as a Â“vitamine.Â”(Rosenfeld, 1997) As scientist began to distill the va rious fat soluble and water-soluble requirements, the number of essential nutriti onal components isolated and characterized grew. It soon became clear that these diet ary factors were not in fact amines and subsequently the Â“eÂ” was dropped. Thus the early nomenclature, fat soluble A and water soluble B and C, would eventually be shortene d and referred to as vitamins A, B and C, respectively (Drummond, 1920). While the nutritional significance of vitamins is widely understood today, in the early 20th century the mere suggesti on that accessory dietary f actors were important was revolutionary and controversial. Though neit her Eijkman or Hopkins actually isolated the Â“anti-beriberi factorÂ” or any Â“accessory food factors,Â” it was their observations and research that would ultim ately led to the discovery of vitamins, thereby transforming nutritional research. The characterization of these trace nutrients not only revolutionized nutritional science, but contri buted greatly to ot her scientific disciplines. In 1929, Eijkman and Hopkins would share the Noble pr ize in Â“Physiology or MedicineÂ” for their work leading to the Â“discove ry of growth-stimulating vita mins.Â” By 1965, nine other scientists would be awarded the Nobel Prize for their work with vitamins and another seven Nobel Laureates would make significan t contributions to v itamin research.
6 Unraveling the Vitamin B complex Â“[T]he history of B6 is a further proof that success usually is preceded by trials, tribulations, and recurrent disappointment. The most helpful factor, apart from perseverance and timeliness of the line of research, is the deliberate recognition of a principle that is paramount in scientific research ; it is often almost beyond our control and touches closely on intuition. It is Walte r B. CannonsÂ’s Â‘serendipityÂ’(Gyorgy, 1971).Â” Paul Gyorgyy, M.D. Water-soluble vitamin B, the Â“anti-beriberiÂ” factor, was initially believed to be a single compound. The 1919 JBC review by Mitche ll was the earliest suggestion that vitamin B likely contained multiple compounds of nutritional significance (Mitchell, 1919; Carpenter, 2003). It was not until 1926 that Dutch scientists in Java isolated the first vitamin B component, thiamine, Grijn sÂ’ Â“anti-beriberiÂ” factor (Jansen, 1926; Carpenter, 2003). Thiamine, also called Vitamin B1, would be the first B vitamin identified and the first vitamin to be recognized as an enzyme cofactor. Seven additional chemically distinct co mponents of the vitamin B complex would eventually be isolated and identified (Table 1.1). Ultimately, all eight would be identified either as an enzyme cofactor or an essen tial cofactor component. The discovery that vitamins functioned as enzyme cofactors demons trated that vitamins play essential roles in various aspects of metabolism and began to bridge the gap betw een nutritional science and the then seemingly remote fields of cellular metabolism and enzymology. Among the vitamin B components that woul d eventually illicit the attention of nutritionists, biochemists and enzymologists, was pyridoxine. First identified as a
7 substance present in a crude supplement that cured or prevented florid dermatitis in rats, pyridoxine was initially given the name vitamin B6 by Paul Gyorgy (Gyorgy, 1934). This new vitamin was eventually isolated, synt hesized, and ultimate ly given the name pyridoxine (Gyorgy, 1971). An amine and an aldehyde form of vitamin B6, pyridoxamine and pyridoxal, respectively, was eventually identified by Esmond Snell, a biochemist studying microbial metabolism (Snell, 1942). Snell also identified the respective 5-phosphate esters of pyridoxal, pyridoxal 5-phosphate (PLP) as the major biological active form of the vitamin B6 (Christen and Metzler, 1985). Characterization of Vitamin B6 as a cofactor and coenzyme Â“Pyridoxal phosphate holds an exceptional place among the coenzymes with regard to both the unparalleled diversity of its catalytic function and to their para mount significance in biochemical transformations of amino acids an d in integral nitrogen metabolism (Boyer, Lardy et al., 1960).Â” A. E. Braunstein Scientists studying vitamin B6 began unraveling the signi ficance of its role as a cofactor in cellular metabolism. Snell eventually demonstrated that heating pyridoxal with various amino acids would produce an oxo-acid and pyridoxamine in a reversible fashion, prompting the proposal that vitamin B6 derivatives might be involved in enzymatic transaminations (Christen and Metzler, 1985). Almost simultaneously, tyrosine decarboxylase activity and enzyma tic transamination between glutamate and oxaloacetate in Streptococcus faecalis cells were found to be dependent on the availability of pyridoxal by another gr oup (Christen and Metzler, 1985).
8 However, it was the non-enzymatic catalytic activity of pyridoxal that would give the most insight into the magnitude of its role as a coenzyme. Pyridoxal demonstrated an ability to catalyze amino acid transaminations, racemization, and eliminations (Metzler and Snell, 1952; Olivard, Metzle r et al., 1952; Christen and Metzler, 1985; Miles and Metzler, 2004). The electron-withdrawing prope rty of the heterocyclic nitrogen was ultimately recognized as a fundamental element to all pyridoxal-catalyzed reactions (Christen and Metzle r, 1985). This led to a proposed mechanism based on the utilization of the electron w ithdrawing properties of the PLPcofactor in the labilization of an -carbon bond of the amino acid substrate, which was eventually found to be common for both non-enzymatic and PLP-assist ed enzymatic reactions (Christen and Metzler, 1985). The catalytic diversity of PLP-dependent enzymes ar ises from modulation and enhancement of the intrinsic chemical proper ties of PLP by the active site environment. In PLP-dependent enzymes, the PL P forms an imine bond with the -amino group of an active site lysine; this Schiff base linkage is termed the Â“internal aldimineÂ” and is outlined in Scheme 1.1. The reactions begin with a transaldimination in which the amino group of the amino acid substrate replaces the -amino of the lysine. This PLP-substrate complex is common to all reactions and is term ed the Â“external aldimineÂ”. Subsequently, the cleavage of an amino acid -carbon bond leads to a resona nce stabilized quinonoid intermediate in which the ex tended pi-system of the coenzy me acts as an electron sink, storing electrons from the cleaved bond thr ough the conjugated system of the Schiff base and pyridinium ring. Ultimately, the el ectrons are dispensed back to the -carbon
9 allowing the formation of new linkages (Chris ten and Metzler, 1985; Christen and Mehta, 2001). The fact that this essential and ubiquitous coenzyme is associated with at least 145 different enzymatic activities attests to the versatility and demonstrates the biological importance of this cofactor (Percudani an d Peracchi, 2003). PLP-dependent enzymes include amino acid racemases, transaminases, decarboxylases, synthases and aldolases. These enzymes play a key role in the synt hesis, interconverti on, and degradation of amino acids and are also involved in the synt hesis of nucleic acids and protein cofactors such as NAD+, biotin and heme (Mehta and Christen, 2000). Medical significance of Vitamin B6 Vitamin B6 would eventually be recognized by the medical field for being involved in significantly more bodily functions than any other nutrient, with over 100 essential biochemical reaction catalyzed by PLP-dependent enzymes in human metabolism (Tambasco-Studart, Titiz et al., 2005). A number of disorders have been associated with PLP-dependent enzymes, so me of which are overviewed in table 1.2. Because of the wide availability of vitamin B6 in both plantand animal-derived food, B6 deficiencies are uncommon. However, se veral different mechanisms can lead to an increased requirement for vitamin B6, including depletion by various drugs, renal dialysis, malabsorption, a nd hereditary errors in B6 metabolism (Holman, 1995). With the wide range of reactions that are catal yzed by PLP-dependent enzymes, it is not surprising that vitamin B6 deficiency has been associated with an array of symptoms including weakness, reduced resistance to infe ctions, weight loss, depression, irritability,
10 sleeplessness, peripheral neur opathy, seborrhoea-like dermat osis regions, pellagra-like dermatitis, glossitis/stomatitis, however, weakness, sleeplessness, and depression are the most prominent effects (Holman, 1995).
11 Table 1: Vitamin B components Vitamin Name Deficiency effects CofactorFunctionStructure B1 Thiamin Beri-beriThiamine Pyrophosphate Activated aldehyde transfer B2 Riboflavin Ariboflavinosis FMN and FADRedox reactions; Electron carrier B3 Niacin Pellagra NAD and NADPRedox reactions; Electron Carrier B5 Pantothnic Acid Paresthesis Coenzyme AAcyl group transfer B6 Pyridoxine Various symptoms Pyridoxal PhosphateAmino Acid Metabolism: Transaminations Deaminations Decarboxylations Condensations B7 Biotin InfantsImpair growth; neurological disorders BiotinCarrier of activted CO2 N N NH2N+S OH N N N N H O O OHOH OH OH H H H H H N O NH2 O O NH OH OH O H O H2PO3N+C H OH HO S N H NH O HO H H O
12 Table 1 (Con't) Vitamin Name Deficiency effects CofactorFunctionStructure B9 Folic Acid Macrocytic anemia; Birth defects TetrahydrofolateOne carbon transfer B12 Cobalamin Pernicious anemia Adenosylcobalamin; Methycobalamin Intramolecular rearrangements; Methylation; Ribonucleotide reduction N N H N N NH O O O N H2NH OOH OH N Co+N N N N N O P N O O-O O NH O O O O O H O O O NH2N H2N H2NH2N H2NH2O H H H H H
13 Lysine Amino Acid SubstrateN+C H O H2PO3C H N R C H OOO N+N+H OO H2PO3H H External Aldimine Internal Aldimine Quinonoid Intermediate Quinonoid IntermediateC O2 f r o m C H+ f r o m C R f r o m C N C H O H2PO3C N+R H OH Quinonoid Intermediate Aminotransferases -Eliminases Racemases Decarboxylases -Synthases Serine hydroxymethyl transferase -Synthases -Synthases Quinonoid Intermediate SCHEME 1.1 Reactions of PLP-dependent enzymes
14PLP dependent enzymes Function Assoicated Diseases Alanine aminotransferaseInterconversion of alanine and pyruvate Hyperoxaluria type IAminolevulinate synthase Heme biosynthesis X-linked sideroblastic anaemiaCystathionine b-SynthaseClearance of intracellular homocysteineHereditary homocystinuria and increase risk for atherosclerotic, cardio-, cerebroand peripheral vascular diseases, and deep vein thrombosis and thromboembolis m Cystathionine -lyase Trans-Sulfuration L-cystathionine to L-cysteine, aketobut y rate and ammonia Cystathioninuria, cystinosis and homocystinuriaDOPA decarboxylaseformation of dopamine decarboxylation of Laromatic amino acids into corresponding amines Parkinsons disease and hypertensionGABA transaminase Breakdown of GABA, regeneration of glutamateGABA inhibition/glutamate excitationGlutamate decarboxylase Conversion of glutamate to GABALow GABA levels implicated in the symptoms associated with epilepsy, Parkinsons disease, Huntingtons chorea, Alzheimers disease and tardive d y skinesiaHistidine decarboxylase Synthesis of histamine Inflammatory diseases, some neurological and neuroendocrine disorders, osteoporosis and several t yp es of neo p lasiasOrnithine d-aminotransferaseInitial and rate-limiting step in the biosynthesis of p ol y amines Ornithine anemia with gyrate atrophySerine hydroxymentyl transferaseOne-carbon metabolism (methionine and lipids b ios y nthesis ) Suspected increased risk of lung cancerSerine palmitoyltransferaseSphingomylin synthesis Hereditary sensory neuropathy type I
15 Vitamin B6-dependent enzymes Reactions catalyzed by PLP-dependent enzy mes play key roles linking different metabolic pathways and are thought to have been essential in the last universal ancestor of contemporary cells, in which the major metabolic pathways were likely established (Ouzounis and Kyrpides, 1996; Mehta and Chri sten, 2000). Thus, it has been proposed that this class of enzymes were already pr esent in a universal pr ogenitor cell 1000 to 1500 millions of years ago (Mehta and Christ en, 1998). The sequence similarities among this highly divergent class of enzymes have generally been too low to establish a common ancestry. However, after consider able progress was made in the threedimensional structure determination of a representative number of PLP-dependent enzymes, Christen and Mehta were finally able to use the high struct ural conservation of PLP-dependent enzymes to verify an evolu tionary relationship (C hristen and Mehta, 2001). Initially, PLP-dependent enzymes were cla ssified according to reaction specificity relative to the C but as the database of PLP-enzyme structures grew it became possible to classify according to fold-types derived fr om three-dimensional structures (Alexander, Sandmeier et al., 1994; Christen and Mehta, 2001; Eliot and Kirsch, 20 04). There are 4 fold-types of PLP-dependent enzymes a nd each corresponds to an independent evolutionary lineage. The aspartate aminot ransferase superfamily corresponds to the fold-type I, family is the largest and most divers e lineage of PLP-dependent enzymes, of which the members outnumber the combined total of the other families (Mehta and Christen, 1998; Mehta and Christen, 2000; Paiardini, Bossa et al., 2004).
16 The high structural conservation displayed by the -family, despite a low degree of sequence identity (1.1), is not unusual among proteins shari ng a common ancestor (Paiardini, Bossa et al., 2004). The dimer is the minimal functional unit for this family and each monomer consists of three domains (1.2), a short N-terminal domain (~50 residues), a central catalytic core (~250 residues), and a C-terminal domain (~100 residues) (Alexeev, Alexeeva et al., 1998; As tner, Schulze et al., 2005; Yard, Carter et al., 2007). Each dimer contains two identical active sites which lie at the interface between the two subunits. Both active sites interact with the PLP cofactor through a Schiff base linkage with an active site lysine (Christen and Mehta, 2001). In all known structures of the -family, the pyridine ring of the PLP co factor is nearly superimposible (Kack, Sandmark et al., 1999). The pyridoxal moiety interacts with the enzyme in a common motif, which also includes a salt-br idge between the pyridinium ring nitrogen and an aspartate, and a hydrogen bond with the phenolic oxygen which occurs through a variety of amino acids in addi tion to Schiff base linkage with an active site lysine (Kack, Sandmark et al., 1999). Interestingly, the activ e site lysine and aspartate are the only two perfectly conserved amino acids in the -family (Mehta and Christen, 1998). The -oxoamine synthases are a small group of enzymes within the -family that catalyze reactions between amino acids and Co A thioesters. 5-Aminolevulinate synthase (ALAS) and 8-amino-7-oxononanoate s ynthase (AONS) belong to this -oxoamine synthase subfamily, along with serine pa lmitoyltransferase (SPT) and 2-amino-3oxobutyrate CoA ligase (KBL) (Alexeev, Alexeev a et al., 1998; Schneider, Kack et al., 2000; Schmidt, Sivaraman et al., 2001; Astner Schulze et al., 2005; Yard, Carter et al.,
17 2007). These four members of the -oxoamine family share about 12% identity, with the sequence identity around 30% between any pair (Alexeev, Alex eeva et al., 1998). ALAS, AONS and SPT catalyze condensatio ns between amino acids and acylCoA thioesters with the concomitant decar boxylation of the amino acid, while KBL only catalyzes the condensation between the acyl-CoA thioester and the amino acid (Schneider, Kack et al., 2000). ALAS a nd AONS function in the biosynthesis of prosthetic groups essential for proteins that play fundamental roles in many biochemical processes. ALAS catalyzes the first co mmitted and rate limiting step of heme biosynthesis in non-plant eukaryotes and the al pha subdivision of purple bacteria (Jordan, 1991) and AONS catalyzes the first committed step in biotin biosynthesis (Alban, Job et al., 2000).
18 FIGURE 1.1 Ribbon diagrams of representative enzymes of Fold Types I IV. Each structure depicts a homodimer with the individual monomers distinguished by color. A. Fold-type I (aspartate aminotransferase family), 2-amino-3oxobutyrate CoA ligase, 1FC4, B. Fo ld-type II (tryptophan synthase family), tryptophan synthase, 1EX5, C. Fold-type III (alanine racemase family), Alanine racemase, 1FST, D. Fold-type IV (D-amino acid aminotransferase family), branch ch ain aminotransferase, 1KTA, (In A,B and E the PLP cofactor is shown in red, while in C the active site lysine is shown in red.) Images were generate d with Deep View (Guex and Peitsch, 1997; Kaplan and Littlejohn, 2001). A. B. C. D.
19 FIGURE 1.2 PLP-dependent enzymes of the -family are homodimers. The ALAS homodimer from R. capsulatus (PDB code: 2BWN) in ribbon representation with one subunit shown in yellow a nd the central catalytic core, N-terminal domain, and C-te rminal domain of the second subunit rendered in dark, medium and light blue respectively. De picted in ball-andstick are the PLP cofactor ( red ), the active site lysine ( green ) involved in the Schiff base linkage with PLP a nd the active site residue F342 ( purple) which is contributed by the adjacent ch ain. Image was generated with Deep View (Guex and Peitsch, 1997; Ka plan and Littlejohn, 2001).
20 Biotin SCHEME 1.2-Biotin Biotin Biosynthesis Biotin (Vitamin H or B7) is a cofactor for a small number of carboxylases and decarboxylases and is essential for life in all organisms (Scheme 1.2) (Alban, Job et al., 2000). Biotin biosynthesis occurs in most bacteria and archea as well as some fungi and plants. Animals and most fungi are biotin auxotrouphs (Alban, Job et al., 2000; Streit, W. R. and Entcheva, P., 2003). The first common precursor in bi otin biosynthesis is pimeloyl-CoA (Alban, Job et al., 2000; St reit, W. R. and Entcheva, P., 2003). The precursors for pimeloyl-CoA are not uniform and are not known for all microbes that synthesize biotin. In plants and gram-positive bacteria, pimeloyl CoA is synthesized from pimelic acid by pimeloyl -CoA synthetase. The four steps that convert pimeloylCoA to biotin are conserved among all organi sms known to produce biotin (Alban, Job et al., 2000; Streit, W. R. and Entcheva, P., 2003). The first enzyme of the common pa thway is AONS, which catalyzes the decarboxylative condensation of alanine and pime loyl-CoA in the first committed step of biotin biosynthesis, Scheme 1.3 (Schneider, G. and Lindqvist, Y., 2001). The next step is the conversion of 8-amino-7-oxopelargonic acid to 7, 8-diaminopelargonic acid (DAPA) catalyzed by DAPA aminotransferase. Both AONS and DAPA aminotransferase belong S N H NH O HO H H O
21 to the -family of PLP-dependent enzymes. DAP A aminotransferase is the only known aminotransferase that uses s-ad enosylmethionine (SAM) as the NH2 donor (Kack, Sandmark et al., 1999; Schneider, G. and Lindqvis t, Y., 2001). In th e following reaction, which is catalyzed by dethiobiotin synt hetase, an uriedo ring is formed via a carboxylation reaction a nd requires ATP, Mg2+ and CO2 (Schneider, G. and Lindqvist, Y., 2001). The final step of bi otin biosynthesis involves th e insertion of a sulfur atom between the non-reactive methyl and methylene carbon atoms adjacent to the ureido ring of dethiobiotin (Schneider, G. and Lindqvi st, Y., 2001) and is catalyzed by biotin synthase which functions as a homodimer. Biotin synthase is a s-adenosylmethionine dependent-enzyme with one 2Fe-2S cluster per monomer when isolated anerobically (Schneider, G. and Lindqvist, Y., 2001; Loti erzo, Tse Sum Bui et al., 2005). There is disagreement surrounding the identity of th e sulfur donor, while free cysteine has been the generally accepted donor, recent studies indica te that the Fe-S cluster may be the true sulfur source (Schneider, G. and Lindqvist, Y., 2001). Medical significance of biotin Biotin (Vitamin H or B7), another component of the vitamin B complex isolated by the physician Paul Gyorgy, is a cofactor for a small number of carboxylases and decarboxylases (Alban, Job et al., 2000). Biotin-dependent carboxylases add carbon dioxide to substrates and re quire ATP hydrolysis. The general mechanism for biotin dependent carboxylases is shown in Figur e 1.3. Biotin-dependent carboxylases play essential roles in the metabolism of choles terol, amino acids, and leucine degradation (Pacheco-Alvarez, Solorza no-Vargas et al., 2002).
22 In humans, there are five bi otin-dependent carboxylases: propionyl-CoA-, methylcrotonyl-CoA-, two forms of acetyl-CoA and pyruvate carboxylase (PachecoAlvarez, Solorzano-Vargas et al., 2002). The ab solute requirement for biotin rests in the central role that these four enzymes play in cell metabolism, which is summarized in Figure 1.4. The two most notable examples of biotin dependent-enzymes are acetyl-CoA carboxylase and pyruvate carboxylase. Acetyl -CoA carboxylase cata lyzes the production of malonyl-CoA from acetyl-C oA, the first and rate-limiting step in fatty acid synthesis (Pacheco-Alvarez, Solorzano-Vargas et al., 2002). Pyruvate carboxylase is an anaplerotic enzyme involve d in the formation of oxa loacetate from pyruvate, and oxaloacetate is a key intermediate in both gluconeogenesis and the TCA cycle (PachecoAlvarez, Solorzano-Vargas et al., 2002). Humans, like other animals, are unable to synthesize biotin and depend on diet to fulfill their need for this vitamin. The prim ary source of dietary biotin is protein-bound and biotin must be cleaved. Pancreatic bi otinidase cleaves the pr otein-bound biotin from the -amino group of the lysine residue to which it is attached (Suzuki, Aoki et al., 1994). Higher organisms have evolve d an efficient biotin cycle to survive with the low concentrations of biotin found in natura l food sources(Pacheco-Alvarez, SolorzanoVargas et al., 2002). Duri ng carboxylase turnover, biotin is cleaved by cytosolic plasma and biotinidase (Gravel and Narang, 2005). Once the biotin is inside the cell, holocarboxylase synthetase covalently att aches biotin to car boxylases (Gravel and Narang, 2005). Both biotinidase and holocarboxylase synthase deficiencies result in Multiple Carboxylase Deficiency, MCD. The symptoms for MCD include; alopeica,
23 developmental delay, organic aciduria, seiz ures, skin rashes, mild hyperammoneima, and breathing problems. Because biotinidase defi cient patients only lack the ability to generate free biotin, it is easily treated by biotin supplementation (5Â–20 mg per day) (Leon-Del-Rio and Gravel, 1994; Suzuki, Aoki et al., 1994). In holocarboxylase synthetase deficient patient s, carboxylase biotinylation is compromised globally, resulting in lethal conse quences if not diagnosed and treated rapidly. Since holocarboxylase synthase deficiency leading to MCD is due to mu tations resulting in decreased biotin affinity, the addition of pharmacologic doses of biotin (10Â–100 mg per day) to the diet can revers e the symptoms if treated promptly (Pacheco-Alvarez, Solorzano-Vargas et al., 2002).
24 CH3NH2N H2O H O CH3NH N H O H O O NH N H O H O O S CH3NH2O O H O CoAS O O H O KAPA Synthase DAPA Aminotransferase Dethiobiotin Synthetase Biotin Synthetase Alanine S-Adenosylmethionine (Sulfur Donor) ATP, CO2 S-Adenos y lmethionine SCHEME 1.3 Biotin Biosynthesis
25 FIGURE 1.3 Biotin-dependent carboxylases. A similar catalytic mechanism is shared by biotindependent carboxylases. Most biotin-dep endent enzymes have three functional domains, the biotin carboxylase (or decarboxylase) domain, the carboxyltransferase domain and the biotin carboxyl carrier domain. The reaction occurs in two steps in two separate subsite s. The first step occurs at the first subsite on the carboxylase domain and i nvolves the partial fixation of CO2 to biotin. The carboxybiotin then swi ngs to the second subsite on the carboxyltransferase domain where a carboxylat ed compound is formed with the carboxylases or the generation of free CO2 for the decarboxylases (Jitrapakdee and Wallace, 2003). S N H N N H O H H O Lysin e R step a step b Carboxylase Transcarboxylase Domain Domain ATP RH HCO3 Biotinyl Domain R =CO2
26 FIGURE 1.4 : Biotin-dependent car boxylases in metabolism OXALOACETATE GLUCOSE Pyruvate Carboxylase PYRUVATE PYRUVATE LEUCINE 3-METHYLCROTONYL-CoA FATTY ACIDS Methylcrotonyl-CoA Carboxylase 3-METHYLGLUTACONYLCoA MALONYL-CoA Acetyl-CoA Carboxylase ACETYL-CoA ACETYL-CoA VALINE ISOLEUCINE THREONINE ODD CHAIN FATTY ACIDS PROPIONYL-CoA PROPIONYL-CoA Propionyl-CoA Carboxylase METHYLMALONYL-CoA SUCC INYLCo A C ITRATE OXALOACETATE TCA CYCLE
27 Heme SCHEME 1.4 Heme Biosynthesis Tetrapyrroles are indispensable to various biological proces ses, including oxygen transport (heme), photosynthe sis (chlorophyll), electron transport (cytochromes), methionine and methylmalonyl CoA synthe sis (colabalamin) and nitrate reduction (siroheme) (Frankenberg, Moser et al., 2003). The universal precursor of heme (Scheme 1.4) and other tetrapyrroles is 5-aminolevu linic acid, ALA. There are two distinct pathways leading to ALA formation and both require PLP. The C5 pathway, found in plants, algae, and most bacteria, utilizes th e 5C -skeleton of glutamate and involves the action of two enzymes, the NADPH-dependent glutamyl-tRNA reductase (GluTR) and the PLP-dependent glutamate-1-semialdehyde-2 ,1-aminomutase (GSAM) (Frankenberg, Moser et al., 2003). The Shemin, or C4 path way, identified in non-plant eukaryotes and the -subdivision of purple bacteria requi res a single PLP-dependent enzyme, 5-HOOC NN N N Fe HOOC
28 aminolevulinate synthase (ALAS) (Shemin, Russell et al., 1955). The Shemin pathway for heme biosynthesis begins with the c ondensation of succinyl-CoA and glycine to produce carbon dioxide, CoA and ALA (Shemin, Russell et al., 1955; Ryter and Tyrrell, 2000). In humans and other mammals, heme bios ynthesis has been well characterized and begins in the mitochondria w ith the biosynthesis of ALA (Thunell, S. 2000). The prime location of heme biosynthesis occurs in the bone marrow and liver, however, the mechanisms controlling synthesis differ. Because heme demands vary significantly according to the unique requirements of different tissues, it is not unexpected that there is not a ubiquitous regulatory pathway. While th e major control of heme biosynthesis in all cells is the production of the initial precurs or ALA by ALAS, regula tion of ALAS varies in various cell types (Thunell, S. 2000). Two cytosolic ALAS isoenzymes are en coded by separate genes; the general housekeeping form, ALAS-1, and the erythroid specific form ALAS-2 and each are regulated by different mechanisms. The housekeeping isoenzyme, ALAS-1, has a short half-life and can be rapidly turned over in re sponse to the drain of the free cellular heme pool by present metabolic needs (Thunell, S. 2000). The erythroid isoenzyme, ALAS-2, regulation machinery is designed for unint errupted production and induced only during the period of active hemoglobin synthesis. ALAS-2 is regulated transcriptionally and post-transcriptionally by erythropoietin action and the amount of free iron present (Thunell, S. 2000; Zoller, Decristoforo et al., 2002). The regulation of ALAS-2 by intracellular free iron involves a mRNA stem loop structure in the 5Â’ untranslated region which contains an iron responsive elemen t (IRE). The binding of iron regulatory
29 proteins, IRP-1 and IRP-2, to the IRE inhi bits ALAS-2 mRNA tr anslation under ironpoor conditions. High intracellular iron ava ilability induces the post-translational modification of IRP-1 and degradation of IRP-2 to permit ALAS-2 mRNA translation ALA then moves into the cytosol, wh ere ALA dehydratase (or porphobilinogen synthase) catalyzes the condens ation of two ALA molecules to form the pyrrole ring of porphobilinogen, Scheme 1.5 (Thunell, S., 2000). Four porphobilinogen molecules are polymerized by porphobilinogen deaminase to form the linear tetrapyrrole of hydroxymethylbilane, which is then cy clized to uroporphyrinogen III by uroporphyrinogen III synthase (Frankenberg, Mo ser et al., 2003). The decarboxylation of 4 acetate groups of uroporphyrinogen III by uroporphyrinogen III decarboxylase yields coproporphyrinogen III (Ferreira, 2004). Heme biosynthesis then moves back into the mitochondria where two propionate side ch ains are oxidized by coproporphyrinogen III oxidase to yield protoporphyr inogen IX (Ferreira, 2004). Protoporphrinogen IX oxidase then catalyzes a six electron oxidation completing the conjugated system of the tetrapyrrole microcycle yieldi ng protoporphyrin IX (Ferreira, 2004). In the final step of heme biosynthesis, iron is inserted into protoporphyrin IX by ferrechelatase to produce heme (Ferreira, 2004). While the biosynthesis of the different te trapyrroles can requ ire between 7 and 30 reactions, the three reactions following AL A biosynthesis leading to uroporphyrinogen III found in heme biosynthesis are common for all tetrapyrroles (Figure 1.5). The steps leading from uroporphyrinogen III to protoporphyrin IX ar e common to both heme and chlorophyll biosyntheses. In plants, the major site of tetrapyrrole bi osynthesis is in the plastids, although there is some debate regarding biosynthesis of heme in the
30 mitochondria. Specifically, the co-localizati on of the three terminal enzymes of heme biosynthesis may occur in both the plastid and the mitochondria, however there is not consensus among researchers (Tanaka and Tanaka, 2007).
31HOOC NN N N Fe HOOC HOOC NHHN HN NH HOOC HOOC NHN HN N HOOC HOOC NHHN HN NH COOH COOH HOOC HOOC HOOC COOH COOH HN COOH HOOC N H2 NH2H O OH O O OH O CoAS O O H NH2 Aminolevulinate Succinyl CoA GlycineALA Dehydratase Porphobilinogen Deaminase Uroporphyrinogen III Synthase Uroporphyrinogen III DecarboxylaseHOOC HOOC NHHN HN NH COOH COOH HOOC HOOC COOH COOH O H HOOC NHHN HN NH HOOC COOH COOH Coproporphyrinogen III Oxidase Protoporphrinogen IX Oxidase Ferrochelatase Porphobilinogen Uroporphyrinogen III Coproporphyrinogen III Protoporphrinogen IX Protoporphrin IX Hydroxymethylbilane Heme 2CO2 H2O2 O2 3H2O2 3O2 4H 4CO2 H2O H2O 4NH2 Aminolevulinate SynthaseCO2 CoASH M I T O C H O N D R I A C Y T O S O LSCHEME 1.5
32 FIGURE 1.5: Overview of tetrapyrroles biosynthesis Glycine and Succinyl-CoA Glutamate Aminolevulinate Uroporphyrinogen III Siroheme Cobalamin Protoporphyrin IX Chlorophyll Heme
33 Medical significance of hemoprot eins and heme biosynthesis In humans, the role of heme as a pr osthetic group for the erythrocyte oxygen transporter protein hemoglobin is, perhaps, the most commonly known role. In fact, 85% of the bodyÂ’s heme is synt hesized in the erythroblast. While hemoglobin is found exclusively in erythrocytes, other specialized globins also function as oxygen carriers in fetal erythrocytes, neurons, cardiac muscle, and skeletal mu scle. Once molecular oxygen is captured by the heme prosthetic group in hemoglobin, it can be transported via the bloodstream to the other oxygen chaperones, su ch as myoglobin in muscle cells, where ultimately it can be used as an electron acceptor. Impaired globin synthesis is associated with thalassemia syndromes and sickle cel l disease is caused by mutations in the -globin gene (Cecil, Goldman et al., 2008). In addition to the ability of heme to interact with oxygen, heme can readily convert between oxidation states, Fe2+ and Fe3+ allowing heme to f unction as an electron carrier that may alternatively be reduced or oxidized. This characteristic makes heme an essential cofactor for a number of enzymes that catalyze redox and single electron carrier reactions during cellula r processes such as cellular re spiration, steroid metabolism and oxidative metabolism of foreign compounds such as drugs. The role of hemoproteins in various biological processes is summarized an d examples are given in Table 1.3. Heme can also function as a regulatory molecule in some transcription factors and proteins involved in various biochemical pathways by controlling the activ ity by binding Â“heme regulatory motifsÂ” (Rodgers, 1999). Cytochrome P450s (CYPs) are a large cl ass of hemoproteins and at least 27 functioning human CYPs have been identified (Nelson, Zeldin et al., 2004). CYPs
34 utilize the redox and electron car rier potential of heme and are involved in a variety of reactions. CYPs play key ro les in cholesterol homeostasis and in the metabolism of drugs and environmental chemicals (Agundez, 2004). CYP gene polymorphisms have been associated with cancer (Agundez, 2004) and altered re gulation of CYPs occurs in diseases such as obesity, diabetes, and nonalcoholic steatohepatitis (Luoma, 2008). The interruption of heme biosynthesis is also associated with a variety of diseases which include X-sideroblastic anemia and porphyrias. Si deroblastic anemias are a group of hematological disorders which is char acterized by microcytic hypochromic anemia with ringed sideroblasts in bone marrow (Bottomley, May et al., 1995). X-linked sideroblastic anemia, an inherited form of si deroblastic anemia, is due to any number of mutations in the erythroid-specific ALAS2 ge ne located on the X-chromosome (May and Bishop, 1998). In addition to the symptoms asso ciated with anemia, patients also exhibit iron overload resulting from increased intestinal absorption of dietary iron and result in the major complications of this disorder (Bottomley, 1982; Propper and Nathan, 1982). The increased iron delivery to the tissues ultimately leads iron overload and secondary hemochromatosis which can result in diabetes liver and heart failure without treatment (Bottomley, 1991; May and Bishop, 1998). Porphyrias are a group of diseases associ ated with the accumulation of porphyrin intermediates of heme biosynthesis. Porphyria s are classified as he patic or erythropoietic and neurovisceral or photocutaneous, according to the site of expression and the associated symptoms, although some overlap, Table 1.4. A wide array of neurovisceral and/or photocutaneous symptoms is associated with porphyrias. In addition to porphyrin neurotoxicity, the neurovisceral symptoms are also thought to be associated with the
35 build up of ALA or PBG, which also have a role neurotoxicity (Ryter and Tyrrell, 2000; Dombeck, T. A. and Satonik, R. C. 2005). Furthermore, ALA may act as a GABA analog and interact with neural GABA recepto rs (Dombeck, T. A. and Satonik, R. C. 2005). Photocutaneous symptoms are caused by porphyrin activation with long wave UV light and the subsequent gene ration of oxygen radicals that damages the skin (Dombeck, T. A. and Satonik, R. C. 2005). There are 8 types of porphyrias which are associated with the 8 enzymes involved in heme biosynthesis. The recently characterized X-linked-dominant protoporphyria results from gain-of-function mutations in the rate-limiting ALAS2 (Whatley, Ducamp et al., 2008). The increased AL A production is propagated down the pathway resulting in an overproduction of protoporphyrin, which ulti mately results in a disparity between the protoporphyrin and heme production. This porphyria results in photosensitivity indistinguishable from erythr opoietic protoporphyria which is associated with impaired ferrochelatase activity, with almost one-fif th of the cases exhibiting liver disease (Whatley, Ducamp et al., 2008). ALA dehydratase porphyria (AD-P) is the rarest, with only four confirmed cases (Sassa and Kappas, 2000). As the name suggests, this por phyria is attributed to a deficiency in ALA dehydratase. AD-P is au tosomal recessive and is classified as an acute hepatic porphyria (Sassa and Kappas, 2000). A major ch aracteristic found in patients is a marked increase in the urinar y excretion of ALA (Sassa and Kappas, 2000). Acute intermittent porphyria (AIP) is an autosomal dominant hepatic porphyria that is the most significant wi th respect to its incidence and clinical severity (Sassa and Kappas, 2000; Thunell, S. 2000). AIP resu lts from a deficiency in porphobilinogen
36 deaminase in which the porphobilinogen deamin ase activity is reduced by approximately 50% (Sassa and Kappas, 2000). The majority of those with AIP are c linically latent and only around one-half will experience an attack during their life span. The acute attacks occur in about 1-2 individuals per 100,000 in the US and in certain populations, such as Lapland, Sweeden, the rate may be as high as 1/1000 (Dombeck, T. A. and Satonik, R. C. 2005). Acute attacks may be life threaten ing and are often precipitated by various factors such as drugs, hormones, starvation, and infection (Dombeck, T. A. and Satonik, R. C., 2005). These factors induce heme synthe sis by either depleting heme stores or via more direct mechanisms of AL AS induction. This results in an increase in porphyrin intermediates in the heme pathway. Patients exhibit a variety of visceral and peripheral symptoms from the involvement of the aut onomic and central nervous system during an attack (Thunell, S. 2000; Dombeck, T. A. and Satonik, R. C., 2005). Extreme abdominal pain, tachycardia, nausea and va rious other neuropsychiatric symptoms are common indicator, with abdominal pain and tachycardia occurring in 90% and 80% of attacks, respectively (Dombeck, T. A. and Satonik, R. C., 2005). Congenital erythroblastic porphyria (CEP) is an auto somal recessive photosensitive porphyria charact erized by an uroporphyrinogen III synthase deficiency. Only about 100-200 cases of the relatively rare disorder has been reported (Sassa and Kappas, 2000). The porphyrin intermediates th at accumulate in CE P lead to cutaneous photosensitivity, hemolytic anemia and fr agile bones (Sassa and Kappas, 2000) Porphyria cutanea tarda is the most co mmon porphyria and ca n be inherited or acquired. Congenital PCT results from d ecrease uroporphyrinogen III decarboxylase activity and occurs in 1-2 individuals per 100,000. Acqui red PCT represents 80-90% of
37 all cases and among these 80-90% is hepatic wi th normal erythrocytes Iron overload is central to the pathogenesis of both acquired and c ongenital versions. The symptoms can be both neuoroviseral and photocutaneous an d include other neurops ychiatric conditions, abdominal pain, and chronic blistering lesions of sunlight-exposed skin. Hereditary coproporphyria is an autosomal dominant acute hepatic porphyria disorder caused by a deficiency in c oproporphyrinogen oxidase (Sassa and Kappas, 2000). Like AIP, factors that are associat ed the induction of heme biosynthesis can induce an attack in which ne urological symptoms predomin ate, although symptoms are usually milder than those displayed with AIP (Dombeck, T. A. and Satonik, R. C., 2005). Mild photosensitivity also occu rs in 30% of patient s (Dombeck, T. A. a nd Satonik, R. C., 2005). Variegate porphyria is autosomal dom inant hepatic porphyria and due to a deficiency in protoporphyrinogen oxidase ac tivity (Sassa and Kappas, 2000). It is relatively uncommon except in South Africa wher e the disorder is prevalent in the white population (Sassa and Kappas, 2000). The clin ical symptoms and precipitating factors also resemble those seen with AIP, with th e addition of skin photosensitivity symptoms including blisters and superf icial ulcers (Dombeck, T. A. and Satonik, R. C., 2005). Erythropoietic protoporphyria (EPP) is due to a partial deficiency of ferrochelatase activity and is an autosomal dominant disorder (Dombeck, T. A. and Satonik, R. C., 2005). EPP is usually charact erized by moderate skin photosensitivity and high concentrations of protoporphyrin IX in the erythrocytes, plasma, and bone marrow (Sassa and Kappas, 2000). Sun exposu re is the major contributing factor and symptoms usually occur within minutes of e xposure. The phototoxic reactions results in
38 burns causing intense pain, edema and vesicl e formation with repeated insults often leading to Â“velvet knuckles,Â” attributed to thick hyperkerat onic skin with deep skin markings over the dorsum the hands (Dombeck, T. A. and Satonik, R. C., 2005).
39General Roles Examples Oxygen carrierErythrocyte hemoglobin (HbA, HbF, HBA2, HbA1) Cardiac and skeletal muscle myoglobin Central and peripherial nervous system neuroglobin Brain cytoglobinElectron transportMitachondrial electron transport chain Cytochorme a, a3, b, cElectron carrier and Oxidation/Reduction Fatty Acid Metabolism (Fatty acids, Prostaglandins, Eicosanoids)(cytochrome P450 mediated)5-lipoxygenase Metabolism of steroids and related compounds (ie. vitamin D, bile acids) cholesterol 24-hydroxylase vitamin D3 25-hydroxylase retinoic acid hydroxylase 7-a-hydroxylase Drug and xenobiotic oxidation cyrochrome p450 mono-oxogenaseAntioxidantReduction of hydrogen perixide catalase peroxidase glutathione peroxidaseTryptophan degredationtryptophan pyrrolaseSensors (NO, CO,O2 and CO2) NO sensor nitric oxide sensitive guanylate cyclase Table 1.3: The role of some hemopr oteins in biological processes
40 Table 1.4: Classification of porphyrias Classification Deficient enzyme Manifestations Inheritance Erythropoietic Congenital erythropoietic porphyria Uroporphyrinogen III synthase Photosensitivity Autosomal recessive Erythropoietic protoporphyria Ferrochelatase Photosensitivity Autosomal dominant/recessive X-linked dominant protoporphyria Erythroid aminolevulinate synthase Acute Photosensitivity (hepatic in 20% of patients) Autosomal dominant Hepatic ALA dehydratase deficiency porphyria ALA dehydratase Chronic neurological symptoms Autosomal recessive Acute intermittent porphyria Porphobilinogen deaminase Acute neurological symptoms often severe Autosomal dominant Hereditary coproporphyria Coproporphyrinogen oxidase Acute neurological symptoms Autosomal dominant Variegate porphyria Protoporphyrinogen oxidase Acute photosensitivity and/or neurological symptoms Autosomal dominant Porphyria cutanea tarda Uroporphyrinogen decarboxylase Acute photosensitivity and neurological symptoms Autosomal dominant and sporadic
41 The medical significance of other -oxoamine synthases The -oxoamine subfamily of the fold-type I, -family of PLP-dependent enzymes has 4 members, two of which are invol ved in the synthesis of protein prosthetic groups which are essential for optimal healt h, biotin and heme. The two other members of this small family, 2-amino-3-ke tobutyrate ligase (KBL) and serine palmitoyltransferase (SPT), are involved in threonine degradation and the synthesis of sphingosine the precursor for spingolipids. Wh ile no disease has been associated with KBL, missense mutations in SPT have been associated with hereditary sensory neuropathy I (Bejaoui, Uc hida et al., 2002). SPT catalyzes the decarboxylative condens ation reaction of L-serine with palmitoyl-CoA to generate 3-ketodihydrosphing osine which is the first and rate-limiting step in sphingosine biosynthesis. SP T is expressed in many species ranging from bacteria to man. The bacterial enzyme, like most fold-type 1 -family members, is a water soluble homodimer. However, in mammals SPT is a heterodimer between the SPTLC1 subunit and either SPTLC2 or SPTLC3 and the heterodimer is anchored to the endoplasmic reticulum. Homozygous SPTLC 1 and SPTLC2 mice ar e embryonic lethal (Hojjati, Li et al., 2005). My riocin, a potent selective SPT inhibitor (Chen, Lane et al., 1999), is an immunosuppressant between 10-1 00 fold more potent than cyclosporin A (Fujita, Inoue et al., 1994), indicating that sphi ngolipids have a key role or roles in immunity. Sphingolipids are a class of lipids deri ved from the aliphatic amino alcohol sphingosine and include glycosphingolipids and sphingomyelins. These compounds are known to play important roles in signal transduction and cell recognition (Me rrill, 2002).
42 De novo synthesis and turnover of sphingolipids are involved in cell regulation including sphingolipid mediated cell death (Merrill 2002). Spingolipids are exploited during infection by some bacteria and viruses and have key roles in the toxic action of some exotoxins and endotoxins from a number of microorganisms (Heung, Luberto et al., 2006). Disruption of sphingolipid turnover results in sphingolip idoses, a group of lysosomal storage diseases in which the accummulation of spingolipids in the central nervous system results in neural degenera tion (Jeyakumar, Butters et al., 2002). The roles of this group of lipids are varied and scientists still have much work to do to understand all the mechanisms by which sphi ngolipds are used to mediate cellular processes. Research Aims PLP is the active form of vitamin B6 and is an essential cofactor for enzymes involved a wide array of reactions that are essential for life and the maintenance of health. The small fold-type I subfamily of -oxoamine synthases, composed of just 4 enzymes, exemplifies this in that they play roles in the biosynthesis of biotin, heme and sphingolipids, biomolecules that are involve d in a wide array of cellular processes essential for the maintenance of health. The aim of the research described in the following chapters is to better understand the relationships among this group of enzymes. Understanding the similarities and differences between these enzymes will provide insight into features that provide them with a shared catalytic mechanism, as well as at mechanisms by which these enzymes enforce selectivity.
43 The studies in the following chapter exam ine the phylogenic relationships and the remarkable structural plasticity of the a-oxoa mine synthases, as well as the significance of a conserved active site histidine. Ch apter 2 explains how sequence and phylogenetic analysis of the -oxoamine synthases were used to identify amino acids that may be significant in regulating the distinctive substr ates specificity of each enzyme. Chapter 3 describes how single chain ALAS dimers and chimeras between ALAS and AONS generated by engineering a single polypept ide chain linking two ALAS polypeptides or the ALAS and AONS polypeptides are used to explore the remarkable structural plasticity exhibited by this group of enzymes. Finally, in chapter 4 the role of an active site histidine that is conserved among the -oxoamine synthases is explored with the characterization of an ALAS variant in which the histidine is repla ced with an alanine.
44 References Agundez, J. A. (2004). "Cytochrome P450 gene polymorphism and cancer." Curr. Drug Metab. 5 (3): 211-224. Alban, C., D. Job, et al. (2000). "Biotin Metabolism in Plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 51 : 17-47. Alexander, F. W., E. Sand meier, et al. (1994). "Evo lutionary relationships among pyridoxal-5'-phosphate-dependent enzyme s. Regio-specific alpha, beta and gamma families." Eur. J. Biochem. 219 (3): 953-960. Alexeev, D., M. Alexeeva, et al. (1998). "The crystal structure of 8-amino-7oxononanoate synthase: a bacterial PL P-dependent, acyl-CoA-condensing enzyme." J. Mol. Biol. 284 (2): 401-419. Astner, I., J. O. Schulze, et al. (2005). "Crystal structure of 5-aminolevulinate synthase, the first enzyme of heme biosynthesi s, and its link to XLSA in humans." Embo. J. 24 (18): 3166-3177. Bejaoui, K., Y. Uchida, et al. (2002). "Her editary sensory neuropa thy type 1 mutations confer dominant negative effects on seri ne palmitoyltransferase, critical for sphingolipid synthesis." J Clin Invest 110 (9): 1301-1308. Bottomley, S. S. (1982). "Sideroblastic anaemia." Clin Haematol 11 (2): 389-409. Bottomley, S. S. (1991). "Sideroblastic anemia: death from iron overload." Hosp Pract (Off Ed) 26 Suppl 3 : 55-56. Bottomley, S. S., B. K. May, et al. ( 1995). "Molecular defects of erythroid 5aminolevulinate synthase in Xlinked sideroblastic anemia." J. Bioenerg. Biomembr. 27 (2): 161-168. Boyer, P. D., H. A. Lardy, et al. (1960) "Pyridoxal Phosphate. In: The Enzymes." 2 : 114. Carpenter, K. J. (2003). "A short history of nutritional science: part 2 (1885-1912)." J Nutr. 133 (4): 975-984. Carpenter, K. J. (2003). "A short history of nutritional science: part 3 (1912-1944)." J. Nutr. 133 (10): 3023-3032. Cecil, R. L., L. Goldman, et al. (2008). Cecil medicine. Philadelphia, Saunders Elsevier. Chen, J. K., W. S. Lane, et al. (1999). "The identification of myri ocin-binding proteins." Chem Biol 6 (4): 221-235.
45 Christen, P. and P. K. Mehta (2001). "From cofactor to enzymes. The molecular evolution of pyridoxal-5'-phos phate-dependent enzymes." Chem. Rec. 1 (6): 436447. Christen, P. and D. E. Metzler (1985). Transaminases. New York, Wiley. Dombeck, T. A. and R. C. Satonik (2005). "The porphyrias." Emerg Med Clin North Am 23 (3): 885-899, x. Dombeck, T. A. and R. C. Sa tonik (2005). "The porphyrias." Emerg. Med. Clin. North Am. 23 (3): 885-899, x. Drummond, J. C. (1920). "The nomenclature of the so-called acc essory food factors (vitamins)." Biochem J. (14): 660. Eliot, A. C. and J. F. Kirsch (2004). "Pyridoxal phosphate enzy mes: mechanistic, structural, and evolutio nary considerations." Annu. Rev. Biochem. 73 : 383-415. Ferreira, G. C. (2004). Heme Synthesis. Encyclopedia of biological chemistry. W. J. Lennarz and M. D. Lane. Amsterdam ; Boston, Elsevier. 2: 362-366. Frankenberg, N., J. Moser, et al. (2003 ). "Bacterial heme biosynthesis and its biotechnological application." Appl Microbiol Biotechnol 63 (2): 115-127. Fujita, T., K. Inoue, et al. (1994). "Fungal metabolites. Part 11. A potent immunosuppressive activity found in Isaria sinclairii metabolite." J Antibiot (Tokyo) 47 (2): 208-215. Funk, C. (1912). "The preparation from yeast and certain foodstuffs of the substance the deficiency of which in diet oc casions polyneuritis in birds." J. Physiol. 45 (1-2): 75-81. Gravel, R. A. and M. A. Narang (2005). "Mol ecular genetics of biotin metabolism: old vitamin, new science." J. Nutr. Biochem. 16 (7): 428-431. Guex, N. and M. C. Peitsch (1997). "S WISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling." Electrophoresis 18 (15): 27142723. Gyorgy, P. (1934). "Vitamin B2 and pellagra-like dermatitis in rats." Nature( 133): 498. Gyorgy, P. (1971). "Developments leading to the metabolic role of vitamin B 6." Am. J. Clin. Nutr. 24 (10): 1250-1256. Heung, L. J., C. Luberto, et al. (2006). "Rol e of sphingolipids in microbial pathogenesis." Infect Immun 74 (1): 28-39.
46 Hojjati, M. R., Z. Li, et al. (2005). "Serine palmitoyl-CoA transferase (SPT) deficiency and sphingolipid levels in mice." Biochim Biophys Acta 1737 (1): 44-51. Holman, P. (1995). "Pyridoxine-Vitamin B-6." Jour. of Aus. Col. of Nutrl. Environ. Med. 14 (1): 5-16. Hopkins, F. G. (1912). "Feeding experiment s illustrating the importance of accessory factors in normal dietaries." J. Physiol. 44 (5-6): 425-460. Jansen, B. C. P., and Donath, W.F. (1926 ). "On the isolation of the anti-beriberi vitamin." Kon. Akad. Wettensch, Amsterdam, Proc. (29): 1390-1400. Jeyakumar, M., T. D. Butters, et al. ( 2002). "Glycosphingolipid lysosomal storage diseases: therapy and pathogenesis." Neuropathol Appl Neurobiol 28 (5): 343-357. Jordan, P. M. (1991). Biosynthesis of tetrapyrroles Amsterdam ; New York, Elsevier. Kack, H., J. Sandmark, et al. (1999). "Cry stal structure of diaminopelargonic acid synthase: evolutionary relationships between pyridoxal-5'-phosphate-dependent enzymes." J. Mol. Biol. 291 (4): 857-876. Kaplan, W. and T. G. Littlejohn ( 2001). "Swiss-PDB Viewer (Deep View)." Brief Bioinform 2 (2): 195-197. Koesling, D., M. Russwurm, et al. (2004). "N itric oxide-sensitive guanylyl cyclase: structure and regulation." Neurochem Int 45 (6): 813-819. Leon-Del-Rio, A. and R. A. Gravel (1994). "S equence requirements for the biotinylation of carboxyl-terminal fragments of human propionyl-CoA carboxylase alpha subunit expressed in Escherichia coli." J Biol Chem 269 (37): 22964-22968. Lotierzo, M., B. Tse Sum Bui, et al. (2005). "Biotin synthase mechanism: an overview." Biochem Soc Trans 33 (Pt 4): 820-823. Luoma, P. V. (2008). "Cytochrome P450 an d gene activation--from pharmacology to cholesterol elimination and re gression of athe rosclerosis." Eur J Clin Pharmacol 64 (9): 841-850. May, A. and D. F. Bishop (1998). "The mol ecular biology and pyri doxine responsiveness of X-linked sideroblastic anaemia." Haematologica 83 (1): 56-70. McCollum, E. V., and Davis, M. (1913). "The necessity of certain lipins in the diet during growth." J. Biol. Chem. (15): 167-175. Mehta, P. K. and P. Christen (1998). "The Molecular Evolution of Pyridoxal 5-Phosphate Dependent Enzymes." Adv. Enzymol. Relat. Areas Mol. Biol. 74 : 129-185.
47 Mehta, P. K. and P. Christen (2000). "T he molecular evolution of pyridoxal-5'phosphate-dependent enzymes." Adv. Enzymol Relat. Areas Mol. Biol. 74 : 129184. Merrill, A. H., Jr. (2002). "De novo sphi ngolipid biosynthesis: a necessary, but dangerous, pathway." J Biol Chem 277 (29): 25843-25846. Metzler, D. E. and E. E. Snell (1952). "Deamin ation of serine. I. Catalytic deamination of serine and cysteine by pyridoxal and metal salts." J. Biol. Chem. 198 (1): 353-361. Miles, E. W. and D. E. Metzler (2 004). "Esmond Emerson Snell (1914-2003)." J. Nutr. 134 (11): 2907-2910. Mitchell, H. H. (1919). "On the identity of the water-soluble growth-promoting vitamine and the antineruic vitamine." J. Bio. Chem. (40): 399-413. Nelson, D. R., D. C. Zeldin, et al. (2004). "Comparison of cytochrome P450 (CYP) genes from the mouse and human genomes, in cluding nomenclature recommendations for genes, pseudogenes and al ternative-splice variants." Pharmacogenetics 14 (1): 1-18. Olivard, J., D. E. Metzler, et al. (1952). "Catalytic racemization of amino acids by pyridoxal and me tal salts." J. Biol. Chem. 199 (2): 669-674. Osborne, T. B., and Mendel, L.B. (1913). "T he relation of growth to the chemical constituents of the diet." J. Biol.Chem. (15): 311-326. Ouzounis, C. and N. Kyrpides (1996). "The emergence of major cellular processes in evolution." FEBS Lett 390 (2): 119-123. Pacheco-Alvarez, D., R. S. Solorzano-Vargas, et al. (2002). "Biotin in metabolism and its relationship to human disease." Arch Med Res 33 (5): 439-447. Pacheco-Alvarez, D., R. S. Solorzano-Vargas, et al. (2002). "Biotin in metabolism and its relationship to human disease." Arch. Med. Res. 33 (5): 439-447. Paiardini, A., F. Bossa, et al. (2004). "Evol utionarily conserved regions and hydrophobic contacts at the superfamily level: The case of the fold-type I, pyridoxal-5'phosphate-dependent enzymes." Protein Sci. 13 (11): 2992-3005. Panel, N. S.-o.-t.-S. (2006). "National Institu tes of Health State-of-the-science conference statement: multivitamin/mineral supplements and chronic disease prevention." Ann. Intern. Med. 145 (5): 364-371. Percudani, R. and A. Peracchi (2003). "A genomic overview of pyridoxal-phosphatedependent enzymes." EMBO Rep. 4 (9): 850-854.
48 Propper, R. and D. Nathan (1982). "Clinical removal of iron." Annu Rev Med 33 : 509519. Rodgers, K. R. (1999). "Heme-based sensors in biological systems." Curr. Opin. Chem. Biol. 3 (2): 158-167. Rosenfeld, L. (1997). "Vitamine--vitamin. The early years of discovery." Clin. Chem. 43 (4): 680-685. Ryter, S. W. and R. M. Tyrre ll (2000). "The heme synthesis and degradation pathways: role in oxidant sensitivity. Heme oxyge nase has both proand antioxidant properties." Free Radic Biol Med 28 (2): 289-309. Sassa, S. and A. Kappas (2000). "Molecula r aspects of the i nherited porphyrias." J. Intern. Med. 247 (2): 169-178. Schmidt, A., J. Sivaraman, et al. (2001). "Three-dimensional structure of 2-amino-3ketobutyrate CoA ligase from Escherichi a coli complexed with a PLP-substrate intermediate: inferred reaction mechanism." Biochemistry 40 (17): 5151-5160. Schneider, G., H. Kack, et al. (2000). "The manifold of vi tamin B6 dependent enzymes." Structure 8 (1): R1-6. Schneider, G. and Y. Lindqvist (2001). "Struc tural enzymology of biotin biosynthesis." FEBS Lett 495 (1-2): 7-11. Shemin, D., C. S. Russell, et al. (1955). "The succinate-glycine cycle. I. The mechanism of pyrrole synthesis." J Biol Chem 215 (2): 613-626. Snell, E. E., Guirard, B.M. (1942). "Occurrence in natural products of a physiologically active metabolite of pyridoxine." J. Bio. Chem. (143): 519. Streit, W. R. and P. Entche va (2003). "Biotin in microbes, the genes involved in its biosynthesis, its biochemical role and perspectives for biotechnological production." Appl. Microbiol. Biotechnol. 61 (1): 21-31. Streit, W. R. and P. Entche va (2003). "Biotin in microbes, the genes involved in its biosynthesis, its biochemical role and perspectives for biotechnological production." Appl Microbiol Biotechnol 61 (1): 21-31. Suzuki, Y., Y. Aoki, et al. (1994). "Isolati on and characterization of mutations in the human holocarboxylase synthetase cDNA." Nat Genet 8 (2): 122-128. Tambasco-Studart, M., O. Titiz, et al. (2005). "Vitamin B6 biosynthesis in higher plants." Proc Natl Acad Sci U S A 102 (38): 13687-13692.
49 Tanaka, R. and A. Tanaka (2007). "Tetra pyrrole biosynthesis in higher plants." Annu Rev Plant Biol 58 : 321-346. Thunell, S. (2000). "Porphyrins, porphyrin metabolism and porphyrias. I. Update." Scand J Clin Lab Invest 60 (7): 509-540. Thunell, S. (2000). "Porphyrin s, porphyrin metabolism and porphyrias. I. Update." Scand. J. Clin. Lab. Invest. 60 (7): 509-540. Whatley, S. D., S. Ducamp, et al. (2008). "C-t erminal deletions in the ALAS2 gene lead to gain of function and cause X-linked dominant protoporphyria without anemia or iron overload." Am J Hum Genet 83 (3): 408-414. Yard, B. A., L. G. Carter, et al. (2007). "The structure of serine palmitoyltransferase; gateway to sphingolipid biosynthesis." J. Mol. Biol 370 (5): 870-886. Zoller, H., C. Decristoforo, et al. (2002) "Erythroid 5-aminolevulinate synthase, ferrochelatase and DMT1 expression in erythroid progenitors: differential pathways for erythropoietin and iron-dependent regulation." Br J Haematol 118 (2): 619-626.
50 Chapter Two Sequence and phylogenetic an alysis of PLP-dependent -oxoamine synthases and their role in identifying residues regulating enzyme specificity Introduction Pyridoxal 5Â’-phosphate (PLP ) is a versatile cofact or, which exhibits a nonenzymatic ability to catalyze a variety of reactions invo lving amino acids, including transaminations, racemizations, and eliminations (Snell and Guirard, 1943; Metzler and Snell, 1952; Olivard, Metzler et al., 1952; Longenecker, Ikawa et al., 1957). PLPdependent enzymes exploit the diverse catalytic repertoire of the PLP-cofactor, as is evident from the fact that this group of enzymes belong to five of six total Enzyme Commission enzyme reaction classes (Meh ta and Christen, 2000). PLP-catalyzed reactions utilize the electron withdrawing prope rty of the positively charged heterocyclic nitrogen acting as an Â“electron sinkÂ” which a llows for the stabilization of a variety of carbanion intermediates and subsequent labilization of an amino acid -carbon bond (Christen and Metzler, 1985). Scrutiny of evolutionary pedigrees in a number of enzyme families has provided significant insight helping to deduce how functional specialization developed. Examination of the evolutiona ry relationships between the -family of PLP-dependent enzymes revealed that specia lization of the catalytic apparatus for reaction specificity generally requires more extensive adaptations than specialization for a specific substrate
51 (Alexander, Sandmeier et al., 1994; Mehta an d Christen, 2000; Schneider, Kack et al., 2000). This clear pattern found in the -family temporal sequence leading to functional specialization prompted the propo sal that primitive regio-specific enzymes first diverged into reaction specific protoenzymes, which later acquired substrate specificity and ultimately, phylogenetic diversity (Mehta and Christen, 2000; Christen and Mehta, 2001). PLP-dependent enzymes are represented by fo ur evolutionary di stinct structural groups, with the -family being the largest and most functionally diverse (Jansonius, 1998; Schneider, Kack et al., 2000; Christen and Me hta, 2001; Eliot an d Kirsch, 2004). The CoA-dependent aclytransferases, or -oxoamine synthases, constitute a small but widespread reaction specific subfamily within the -family of PLP-dependent enzymes. The -oxoamine synthase subfamily is comprised of 8-amino-7-oxononanoate synthase (AONS), 5-aminolevulinate synthase (ALAS) serine palmitoyltransferase (SPT), and 2amino-3-ketobutyrate-CoA ligase (KBL) (Alexeev, Alexee va et al., 1998; Schneider, Kack et al., 2000; Schmidt, Sivaraman et al., 2001; Astner, Schulze et al., 2005; Yard, Carter et al., 2007), which catalyze Claise n condensations betw een amino acids and carboxylic acid CoA thioesters (Ferreira, 1999; Bhor, Dev et al., 2006). At the amino acid level, these enzymes share about 12% id entity, with the identity between any pair around 30% (Alexeev, Alexeeva et al., 1998). Sequence alignments among members of the -oxoamine synthase subfamily have been instrumental in the identification of conserved amino acids, while crystallographi c and kinetic studies have increased our understanding of the structural and functiona l significance of these residues. However, the low sequence identities have made it ch allenging to disti nguish residues that differentiate substrate specificity among this group of enzymes.
52 The recent surge of sequence database s ubmissions resulting from advances in sequencing speed and reliability has generated a reservoir of data th at can be used to provide a better understanding of the phylogenetic relationships between the members of the -oxoamine synthases. A more thor ough understanding of the evolutionary progression of -oxoamine synthases could add insi ght into the mechanism by which functional specialization deve loped and provide valuable knowledge for re-engineering substrate specificity. Because of the diverse catalytic potential of the PLP-cofactor, selectivity of reaction type and substrate specificity also requires PLP-dependent enzymes to significantly diminish or abolish superfluous side reactions and utilization of undesired substrates. Hence, PLP-dependent enzymes commonly demonstrate activity to more than one substrate or reaction (Percudani and Peracchi, 2003). While enzyme promiscuity may seem incompatible with the evolutiona ry specialization of enzymes, it is often regarded as the origin for the divergent evol ution of new specificiti es and functions (Hult and Berglund, 2007). Engineering substrate pr omiscuity and/or new substrate specificity in the -oxoamine synthase family would faci litate a better unde rstanding of the mechanisms by which these enzymes enforce selectivity. The identification of key residues that distinguish between substrates is a challenge for the protein engineer that also may be overcome using directed evolution. Directed evolution, also known as evolutionary engineering, has emerged as a practical approach because it does not re ly on extensive stru ctural information or an understanding of the protein structure, f unction, or catalytic mechanism (Moore, Jin et al., 1997; Zhao and Arnold, 1997). Directed e volution involves repeated cycl es of random mutagenesis
53 (and/or gene recombination), functional expr ession, and finally, selection of variants acquiring new features (or exhi biting improvements in the targeted property) (Kuchner and Arnold, 1997). The number of amino aci d changes between th e parent and the laboratory evolved progeny is significantly smaller than that observed in naturally evolved enzymes. Thus, sequential changes in the in vitro evolved variants may provide insight into residues or pos itions regulating substrate pr eference. Incorporating the phylogenetic relationships of both natural and laboratory evolved enzymes with structural data may help identify amino acids key for enzyme selectivity. AONS and ALAS catalyze decarboxylativ e condensations leading to the formation of a 2-aminoketone product (Sch eme 2.1) (Alexeev, Alexeeva et al., 1998). The sequence, kinetic, and structural similar ities between these two enzymes have led to the proposal of a shared kinetic mechanis m (Alexeev, Alexeeva et al., 1998; Webster, Campopiano et al., 1998). The relatively short evolutionary space required for developing substrate specificity in the -family foldÂ–type I PLP-dependent enzymes suggests that alteration of substr ate specificity is a practical target for directed evolution. While directed evolution has been used su ccessfully to enhance specificity for minor substrates and relieve product inhi bition in other members of the -family of PLPdependent enzymes (Yano, Oue et al., 1998; Chow, McElroy et al., 2004; Rothman, Voorhies et al., 2004), there ar e no records of the use of either approach to redesign oxoamine synthases. Reengineering AONS to utilize ALAS substrates would enhance our knowledge of the mechanism by which thes e enzymes enforce subs trate selectivity by identifying residues with key roles in determining substrate specificity.
54 SCHEME 2.1 An enhanced understanding of the phylogenetic relationships among the oxoamine synthases is desirable to help iden tify amino acids that may be relevant for substrate specificity in evolve d variants. Therefore, an ex amination of the relationships among -oxoamine synthase sequences repres enting a broad taxonomic and phylogenetic spectrum was performed to help identify residu es that may regulate substrate specificity in ALAS, AONS and other -oxoamine synthases. The reengineering of ALAS to acquire AONS function using dire cted evolution requires an e fficient selection system to identify enzymes exhibiting AONS activity. T hus, one goal of this study is to develop a selection system to identify AONS function in which the growth of biotin-auxotrophic Escherichia coli may be rescued by evolved ALAS vari ants acquiring activity for AONS substrates. Experimental Details Sequence search, alignments and phylogenetic analysis A series of alignments using AONS, ALAS, KBL, and/or SPT amino acid sequences were performed with MAFFT using mafftE, which incorporates a Ruby script to align input sequences with up to 100 additional homologues from NCBI-BLAST or SwissProt to improve accuracy in the alignment (Katoh, Kuma et al., 2005). Sequences
55 from the additional homologues aligned using mafftE were selected and added to input sequence sets to obtain a broad phylogenetic distribution of the -oxoamine synthase family members. To obtain the final sequen ce set, additional sequences were added to reflect the taxonomic distri butions of AONS, ALAS, KBL, and SPT and the final alignments were performed with ClustalW using blssm (blocks amino acid substitution matrices), open gap penalty 10, and gap ex tension penalty 0.05 (Thompson, Higgins et al., 1994). The ClustalW alignment scores were generated using the SLOW/ACCURATE alignment parameters. The number of amino acid substitutions per site determined from averaging overall sequence pairs is base d on the pairwise sequences. Analyses were conducted using the Dayhoff matrix based method in MEGA4 (Schwartz and Dayhoff, 1979; Ta mura, Dudley et al., 2007). The evolutionary history was inferred using the neighbor-joining (NJ), the maximum parsimony (MP), and the minimum evolution (ME) methods (Eck and Dayhoff, 1966; Saitou and Nei, 1987; Rzhetsky and Nei, 1992). Phylogenetic analyses were conducted in MEGA4 (Tamura, Dudley et al., 2007). All positio ns containing gaps and missing data were eliminated from the data set. There were a to tal of 257 positions in the final dataset. The sum of the length branch in the NJ tree is =16.425 and. The sum of the branch length in the ME tr ee is 16.330 and the. The ME tree was searched using the close-neighbor-interchange algorithm at a search level of 3 (Tamura, Dudley et al., 2007). The evolutionary distances for the NJ and ME trees were computed using the Poisson correction method. The MP tree was obtaine d using the close-neighbor-interchange
56 algorithm with search level 3 in which the initial trees were obtained with the random addition of sequences (100 replicates) (Felsenstein, 1985 ; Nei and Kumar, 2000). Construction of ALAS random library Random mutagenesis of mature AL AS cDNAs by error-prone PCR was performed under conditions sim ilar to those previously desc ribed (Shafikhani, Siegel et al., 1997; Zhao, Giver et al., 1998). In error-prone PCR, the MgCl2 concentration is increased, MnCl2 is added and the dNTP concentra tion is increased to decrease the fidelity of the Taq polymerase during DNA sy nthesis (Cadwell and Joyce, 1992). The pGF23 plasmid containing the ALAS cDNA sequence flanked by a 5Â’ SalI and a 3Â’ Bam HI restrictions site was us ed as the template (Ferreir a and Dailey, 1993). The 5Â’ primer (TTCG TCT TCA AGT CTT CTC ATG TTT G) was desi gned to anneal upstream of the Sal I restriction site and the 3Â’ prim er (TAC AGA TGT ACA AAA GTT CAG ATA CTG GCG ATC ATC CGC CAC) was designe d to anneal with the sequence of the vector backbone, downstream of Bam HI site. The PCR was performed in 100 l containing 50 mM KCl, 10 mM Tris (pH 8.3), 7 mM MgCl2, 0.15 mM MnCl2, 0.2 mM dGTP and dATP, 1 mM dCTP and dTTP, 0.1 M primer and 5 to 10 ng template and 5 units Taq polymerase (Roche, Indian apolis, IN) using a MJ Res earch MiniCycler. The times and temperatures for annealing, extension, and denatura tion were 1 min at 94oC, 1 minute at 50oC and 1 minute at 72oC, respectively. This se quence was repeated for 30 cycles. Subsequently, the PCR product was purified using QIAquick spin columns (QIAGEN, Valencia, CA). The Â“megaprimer PCR of whole plasmidÂ”, MEGAWHOP, cloning method was used to subclone the mutant library using the error-prone PCR products as megaprimers
57 (Miyazaki and Takenouchi, 2002). PCR was performed in 100 l reactions containing 50 mM KCl, 10 mM Tris (pH 8.3), 1 mM MgCl2 and 2.5 units of Taq polymerase, 0.2 mM each dNTP, 0.5 to 1.0 g megaprimers and 50-100 ng of pGF23 using an MJ Research MiniCycler. The times and temperatures for annealing, extension, and denaturation were 30 seconds at 94oC, 1 minute at 50oC and 7 minutes at 72oC, respectively. This sequence was repeated for 30 cycles. After the PCR, 10 units of Dpn I were added directly to the reaction mixture and incubated at 37oC for 30 minutes. The Dpn I digestion was terminated by incubating the reaction at 80oC for 10 minutes. The amplified and Dpn Itreated DNA was purified using QiaQuick sp in columns (QIAGEN, Valencia, CA). Competent Escherichia coli DH5 cells were transformed with the Dpn I-treated MEGAWHOP library and grown on Luria-Bertani medium (Bertani, 1951) containing 50 mg/ml ampicillin for 16 hours at 37oC. After incubation, the ce lls were scraped from the plate and resuspended in sterile buffer P1 (QIAGEN, Valencia, CA). Plasmid DNA from the library of ALAS variants was purifie d using Q-10 mini columns according to the manufacturersÂ’ protocol (Q IAGEN, Valencia, CA). Selection and bioassay for AONS activity E. coli strain R872 (Del Campillo-Campbell, Kayajanian et al., 1967), which can only grow on a medium containing AON, biotin or when harboring a plasmid expressing functional AONS, were used to scre en for AONS function. Competent R872 E. coli cells were transformed with the variant ALAS plasmid library and plated on M-9 minimal medium supplemented with 0.1 % vitamin-free casamino acids and containing 1.5% agar, 0.1 mM CaCl2, 50 mg/ml ampicillin and 5 to 10 pg/ml biotin. The M9 medium contains
58 43 mM Na2HPO4-7H2O, 22 mM KH2PO4, 8.5 mM NaCl, and 18 mM NH4Cl, 2 mM MgSO4, and 0.4 % glucose (Smith and Levine, 19 64). The plates we re incubated at 37oC for 20 hours, followed by 16 hours at room temperature. The plates were examined for apparent colonies. R872 E. coli colonies expressing ALAS variants were selected and used to inoculate 0.3 ml of LB medium and then grown at 37oC while shaking at 115 rpm. Subs equently, 5 ml of M9 minimal medium supplemented with 0.1 % vitamin-free casamino acids, 0.1 mM CaCl2, 50 mg/ml ampicillin and 100 pg/ml biotin was inoculated with 20 l of the ALAS library expressing-R872 culture. The bacterial cells were incubated for 20 hours at 37o C while shaking at 200 rpm. The cell culture was centrifuged at 4000 x g for 15 minutes and the supernatant was discarded. Th e cells were washed twice wi th M-9 minimal medium. To wash the cells, 1.5 ml of M-9 minimal medi um was added, the cells were centrifuged at 4000 x g for 15 minutes and the supernatan t removed. The washed pellet was resuspended in 150 l of M-9 minimal medium and 20 l of the bacterial cell suspension was used to inoculate 1.5 ml of M-9 mini mal medium supplemented with 0.1 % vitaminfree casamino acid and various biotin concentr ations in a 24-well plate. Plates were incubated for 24 hours at 37oC while shaking at 115 rpm and the OD at 600 nm was read at various time intervals between 20 and 50 hours. Results Sequence selection and alignment A total of 44 amino acid sequences were selected to equitably represent the phylogenetic and taxonomic distri bution of each member of the -oxoamine synthase
59 subfamily of PLP-dependent enzymes. Hence, 13 AONS, 10 ALAS, 11 KBL, and 10 SPT sequences were selected. The taxon and species of these sequences are given in Table 2.1. Both AONS and SPT sequences we re found in four of the 5 kingdoms, with AONS being the most broadly distributed member of the -oxoamine synthase family. While all known AONSs, KBLs, ALASs and prokaryotic SPTs are homodimers, eukaryotic SPTs function as heterodimers composed of dissimilar subunits, LCB1 and LCB2, and possess a single PLP-binding moiet y, which is attributed to LCB2 (Han, Gable et al., 2006). In this study, SPT1 a nd SPT2 refer to LCB1 and LCB2 sequences, respectively, and SPT refers to the pr okaryotic sequences. Mammals and other vertebrates express genetically distinct hous ekeeping and erythroid ALAS genes for two distinct isoforms, ALAS1 and ALAS2, resp ectively (Bishop, 1990; Ferreira and Gong, 1995). After the removal of the non-homologous term inal residues, the ClustalW aligned sequences ranged from 361 to 409 amino acids. A selected subset of ALAS, KBL and AONS aligned sequences is shown in Figure 2 .1 and pairwise alignment scores of all sequences, given in percent id entity, are given in Table 2.2. The average evolutionary divergence, deduced by determining the number of amino acid substitutions per site, was 1.5 among all 44 sequences and the AONS seque nce set, 0.7 for both the ALAS sequence set and the KBL sequence set, and 0.8 for the SPT2 sequence set. The sequence alignment of ALAS, AONS, KBL, SPT2 and SPT indicated that nine positions are occupied with completely conserved residues, Gly88, H142, Ser189, Asp214, H217, Lys248, Gly254, Arg356, and Pro364 (Fi gure 2.1). As inferred from the ALAS and AONS crystal structures (Alexeev, Alexeeva et al., 1998; Webster, Alexeev et
60 al., 2000; Astner, Schulze et al., 2005), six of the nine conserved residues, Ser189, Asp214, Lys248 and Gly254, correspond to residues interacting with the PLP cofactor. Gly88 and Pro364 appear to bind succinyl-CoA while Arg356 appears to bind the amino acid carboxylate group, as deduced from the crystal structures of R. capsulatus ALAS bound to either glycine or succinyl-CoA s ubstrates (Astner, Schulze et al., 2005).
61 TABLE 2.1 Aligned Sequences Taxon Species (Abbreviation) Protein Accession No. Monera Actinobacteria Mycobacterium tuberculosis (MyTu) AONS NP_336073 Actinobacteria Streptomyces avermitilis (StAv) KBL NP_822803.1 Alpha Rhizobium etli (RhEt) ALAS YP_473106.1 Alpha Rhizobium etli KBL YP_470446 Alpha Rickettsia bellii (RiBe) ALAS YP_53807 Alpha Rhodobacter capsulatus (RhCa) ALAS P18079 Alpha Sphingomonas paucumobilis (SpPa) SPT 2JGT_B Alpha Sphingomonas sp SKA38 (SPSk) AONS ZP_01304166 Alpha Sphingomonas sp SKA38 SPT ZP_01303842 Alpha Sphingomonas sp SKA38 ALAS ZP_01304166 Aquifex Aquifex aeolicus (AqAe) AONS NP_213435 Bacteroidetes Gramella forsetii (GrFo) KBL YP_863382.1 Delta/Epsilon Helicopacter pylori J99 (HePy) AONS NP_223263 Firmicutes Bacillus cereus (BaCe) AONS NP_833835 Firmicutes Bacillus cereus KBL NP_830437 Firmicutes Bacillus subtilis (BaSu) AONS NP_390900 Firmicutes Bacillus subtilis KBL NP_389582 Firmicutes Staphphylococcus aureus (StAu) AONS NP_372948 Firmicutes Staphphylococcus aureus KBL YP_040004 1 Gamma Escherichia coli (EsCo) AONS 1DJ9A Gamma Escherichia coli KBL ZP_00736622 Gamma Haemophilus infllenzae (HaIn) AONS NP_439702 Euryarchaeota Thermococcus kodakarensis (ThKo) AONS YP_184630.1 Animalia Arthopoda Drosphilia melanogaster (DeMe) ALAS AAL89936 Ar thopoda Drosphilia melanogaster KBL NP_648509.1 Chordata Mus musculus (MuMu) SPT1 NP_033295 Chordata Mus musculus SPT2 NP_035609 Chordata Mus musculus ALAS1 NP_65584 Chordata Mus musculus ALAS2 NP_033783 Chordata Mus musculus KBL O88986 Fungi Ascomycota Aspergillus fumigatus (AsFu) SPT1 XP_7511171 Ascomycota Aspergillus fumigatus SPT2 XP_752555.1 Ascomycota Aspergillus fumigatus ALAS XP_754006.1 Ascomycota Saccharomyces cerevisiae (SaCe) SPT1 NP_014025 Ascomycota Saccharomyces cerevisiae SPT2 NP_010347 Ascomycota Saccharomyces cerevisiae ALAS NP010518 Ascomycota Aspergillus niger (AsNi) AONS Q58FL7 Plantae Magnoliophyta Arabidopsis thalia (ArTh) AONS NP 974731 Magnoliophyta Arabidopsis thalia SPT1 NP_568005 Magnoliophyta Arabidopsis thalia SPT2 NP_197756 Protist Alveolates Plasmodium falciparum (PiFa) ALAS XP_001350846 Alveolates Tetrahymena thermophilia (TeTh) AONS Q224308 Euglenozoa Trypanosoma brucei (TrBr) KBL XP_847436.1
62 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 20 7 ALAS_AsFu D S LNHAS M I Q GIR H S G A K K M V F K HND LVD LE Q K L ASLPL -----HVP K I IA FES V Y S M C G S IAP IEA ICDLA DK Y ALAS_DrMe D AG NHAS M I M GIR N S GVP K HI F R HND VDH L H Q L L K Q TDK-----SVP K IV A FET V H S M T G A I C PL EELL D V A HEH ALAS_PlFa D E M NHASII N GIR E S RCE K FI F K HND M ND LE RI L YNLRINK Q YENRKI M I V FES I Y S M S G H I SNIEY I V Q LA KK Y ALAS_RhCa D S LNHAS M I E GI KRNAGP K RI F R HND VAH L RELIAADDP-----AAP KLIA FES V Y S M DG DFG P IKE ICD I A DEF ALAS_RhEt D A LNHAS M I E GIR HA K CD K VI W K HND VAD LE AK L KAADP -----KAP KLIA FES V Y S M DG D IAP IKE ICDLA D Q Y ALAS_RiBe D E LNHASII A GI TG SKA E K HI Y R H L D VKH LE EL L Q SVDI-----NRP K I I V FES A Y S M DG FFS P VKD I IN LA KK Y ALAS_SaCe D E LNHAS M I V GI KHANVK K HI F K HND LNE LE Q L L Q SYPK-----SVP KLIA FES V Y S M A G SV A DIEK ICDLA DK Y ALAS_S p SK D E LNHAS M I A GIR H S GCE K RV F R HND VDH L REL L AAEDP-----DAP KLIA FES V Y S M DG D IAP IAA ICDLA DEF ALAS1_MuMu D SG NHAS M I Q GIR N S RVP K YI F R HND VNH L REL L Q RSDP-----SVP K IV A FET V H S M D A-VC PL EEL CD V A HEF ALAS2_MuMu D AG NHAS M I Q GIR N S G A A K FV F R HND PGH L KKL L EKSDP-----KTP K IV A FET V H S M DG A I C PL EEL CD V A H Q Y AONS_A q Ae D E LNHASIIDG V RLSKA Q K RV F K H K D YEE LE EF L KKNRKKFR----RV LI I TD T VFS M DG DV A D L KRLT Q ICEE Y AONS_BaCe D K LNHASI V DGI I LS G A EHKR Y R HND LDH LE KL L K M ASPEKR----KLI V TD T VFS M DG DT A Y L RDLV Q L KEK Y AONS_BaSu D Q LNHAS M IDG C RLSKA DTVV Y R H I D M ND LE NK L NET Q RY Q R-----RF I V TDGVFS M DG T IAPL D Q I IS LA KR Y AONS_ThKo EE LNHASIIDG M RLS G A P K VI Y K H I D M ED L KKR L EENKDK-----KK K I I VS DGVFS M DG DL APL PE M AE LA E Q Y KBl_BaCe D E LNHASIIDG S RLSKA KIIV Y K H S D M ED L R Q KAIAAKESGLY--NKL M VI TDGVFS M DG DV A K L PE I VEI A EEL KBl_BaSu D E LNHASIIDGIRL T KA D K KV Y Q H VN M SD LE RV L R---KS M NY--R M R LI V TDGVFS M DG N IAPL PD I VE LA EK Y KBL_DrMe D E LNHASIIDGIRL C KA K K Q R Y R H R D LGD LE E Q L KASDAR------L KLIATDGVFS M DG N IAPL AR I VE LA RK Y KBl_EsCo D A LNHASIIDG V RL C KA KRYR Y AN ND M Q E LE AR L KEAREAGA---RHV LIATDGVFS M DG V IA N L KGV CDLA DK Y KBL_GrFo D S LNHASIIDG V RL C KA ARYR Y ENGN M ED LE K Q L IDANEKGA---RF KLI V TDGVFS M DG LV APL DK ICDLA DK Y KBl_MuMu D E LNHASIIDGIRL C KA H K YR Y R H L D M AD LE AK L KEA Q KHR-----LR L V ATDG A FS M DG D IAPL Q D IC R LA A Q Y KBL_RhEt D A LNHASIIDG V RLSKA KRFR Y AN ND M AA LE EE L KKAEGS -----RF KLIATDGVFS M DG I IA N L GGV CDLA EK Y KBl_RhLe D A LNHASIIDG V RLSKA KRFR Y AN ND M AA LE EE L KKAEGS -----RF KL V ATDGVFS M DG I IA N L GGV CDLA EK Y KBl_StAu D E LNHASIIDG C RLSKA KIIRVN H S D M DD L RAKAKEAVESG Q Y--NKV M YI TDGVFS M DG DV A K L PE I VEI A EEF KBL_StAv D A LNHASIIDGIRLSKA RRFR Y ANR D M AD LE R Q L KEASGAR-----RR LI V TDGVFS M DG YV APL RE ICDLA DR Y KBL_TrBr D A LNHASIIDG V RL C KA ERHR Y A H L D M KE LE TA L Q KT Q HN-----RIR LI V TDGVFS M DG DV APL DK I V Q LA EK Y 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62ALAS_AsFu FD Y NAFYLGE L EKKHKDKS Y RY F NN I NRLAKEFPRAHT----------------AS M EER V T V W CSN D YLG M GRN ALAS_DrMe FP Y ERFFNE Q I M KK K RDHS Y RV F KKVNRLAGDGLFPHALEYS-----------ERTEKPIT V W CSN D YLG M SA H ALAS_PlFa SF Y EFY Q KYGYKPCIGNIR Y Q L S ASFEDNNKNICSFSHKNKENYLFNF W NLHIDNVSNEKTV V W CSN D YL C L SNN ALAS_RhCa M D Y NLA L DKAI Q KLHDE G R Y RT F ID I EREKGAFPKA Q W NRP------------DGGK Q DIT V W C G N D YLG M G Q H ALAS_RhEt M D F EAFFKNE L DGLHAE G R Y RV F ADLER Q RGHFPRATRHTA------------D-GEKD V T V W CSN D YLG M G Q N ALAS_RiBe SY Y DTIFSDHIDKI K SE G R Y RE F KALKR Q ADNFPFA M C----------------D--DK Q IV M W C I N D YLG M SK H ALAS_SaCe FD Y EGLIDSE L Q KKRLDKS Y RY F NN I NRLAKEFPLAHR Q REA---------------DK V T V W CSN D YL A L SK H ALAS_S p SK M N Y KHIFS Q AIDRLHSE G R Y RV F ID I LRNKGAFPNARCFHG------------HNGPKPIT V W CSN D YL A M G Q H ALAS1_MuMu F Q Y DHFFEKKIDEK K NDHT Y RV F KTVNRRA Q IFP M ADDYTDS------------LITKK Q V S V W CSN D YLG M SR H ALAS2_MuMu FG Y D Q FFRDKI M EK K Q DHT Y RV F KTVNR W ANAYPFA Q HFSEA------------S M ASKD V S V W CSN D YLG ISR H AONS_A q Ae --M R W IEEE L KRI K EANL Y RERILLEG------------------------------V KDF CSN D YLGL RK H AONS_BaCe Q T W RAH L Q CK L Q Q LHE Q G Q Y RDLHVTEKAEET W LIRDKK--------------------R M LNLA SNNYLGL AGD AONS_BaSu M KIDS W L NER L DR M K EA G VHRNLRS M DGAPVPERNIDGE--------------------N Q TV W S SNNYLGL ASD AONS_ThKo M GKLD W IREE L Q EL K DK G L Y VTIRKLESA Q GP W VVVDGK--------------------K V L N M CSNNYLGL AA H KBl_BaCe KTLAKF L EEN L EDL K SK G L Y NVIDPLESS N GPIITIGGK--------------------EYI N LS SNNYLGL ATD KBl_BaSu TKEFEF L KAE L NS M K ENHT W Q DIK Q LES M Q GPSVTVNH Q --------------------K V I Q LS SNNYLG FTS H KBL_DrMe A Q LREI L GT Q L AGI K DA G TFKAERI I TSS Q ST Q ITV Q GSDK------------------KIL N F C A NNYLGL ANN KBl_EsCo GEFY Q Q L TND L ETARAE G LFKEERI I TSA Q Q ADITVADG-------------------SH V I N F C A NNYLGL AN H KBL_GrFo GKIKEH L EKEIEEI K DD G L Y KRERI I TGP Q DAVIKIASG------------------Q E V I N F C A NNYLGL SS H KBl_MuMu A Q LRCI L DSE L EGIRGA G T W KSERV I TSR Q GPSIRVDGISG------------------GIL N F C A NNYLGL SS H KBL_RhEt SPFLSH L RAEISALRDA G L Y KSERV I SSK Q AGEIAISTG-------------------ER V L N F C A NNYLGL ADN KBl_RhLe S Q FLSH L SNEISAL K DA G L Y KSERV I SSK Q AGEIAISTG-------------------ER V L N F C A NNYLGL ADN KBl_StAu Q SLHEF L EENINYL K EN G L Y NEIDT I EGA N GPEIKINGK--------------------SYI N LS SNNYLGL ATN KBL_StAv DSVRDD L RTT L DEIRTA G LHKPERV I GTP Q SATVSVTAGGRP----------------GE V L N F C A NNYLGL AD H KBL_TrBr S M LREAAAA Q L AAI K EA G T Y KVERV I TSK Q SSTINVSTAAT------------------P V L N F C A NNYLGL AD H 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137ALAS_AsFu P E V LAT M HET L DTY G A G A G GT R N I S G HNKHAVA LE NT LA N LH G K E A AL V F S SC Y V AN DATLA TL GSK M P DCV I L S ALAS_DrMe P G V KR A V Q D AL NRH G S G A G GT R N I S G NSLH H ER LE SK LA E LH Q K E A ALLF T SCF V AN DSTLF TL AKLL P GCE I F S ALAS_PlFa EKIIEVGIET L KKI G NSS G GT R N I S G SLLN H TH LE YII A K W Y N K E SS LLF T S GY I AN V G ALE TL GKLL N -LIYI S ALAS_RhCa P V V LA A M HE AL EAV G A G S G GT R N I S GT TAY H RR LE AEI A D LH G K E A AL V F S S AY I AN DATLS TL RLLF P GLI I Y S ALAS_RhEt P K V IE A M KA A IDHC G A G A G GT R N I S GT NHY H VL LE RE LA D LH G K E A AL I F T S GY V S N W ASLG TL GGKI P GLI I F S ALAS_RiBe P K V V Q A SID A VLKY G V G S G GT R N I G G NNVAILE LE Q E LA S LH N K E AS L V F T S G F V AN DTTLA TL AKI M P NIVFF S ALAS_SaCe P E V LD A M HKTIDKY G C G A G GT R N I A G HNIPTLN LE AE LA T LH K K E G AL V F S SC Y V AN DAVLSL L G Q K M K DLV I F S ALAS_S p SK P K V VA A M EE AL HDV G A G S G GT R N I G G NTHY H VD LE GE LA D LH G K E A ALLF T S GY V S N EATLS TL AKLL P GCI I F S ALAS1_MuMu P R V CG A V M ETVK Q H G A G A G GT R N I S GT SKF H VE LE Q A LA D LH G K DA ALLF S SCF V AN DSTLF TL AK M M P GCE I Y S ALAS2_MuMu P R V L Q A IEET L KNH G A G A G GT R N I S GT SKF H VE LE Q E LA E LH Q K DS ALLF S SCF V AN DSTLF TL AKLL P GCE I Y S AONS_A q Ae P E V VEESIRV L KEA G L G S G AS Q LVS G YTKH H RE LE EK LA E F K G T E SCV LF G S G F L AN V G TIPA L VEEG D --LVL S AONS_BaCe ERLKE A AIACTKRY G T G ATAS R LVV G NHLLYEEV E RSICD W K G T E R AL IVN S GYT AN I G AISS L ASRH D --IVF S AONS_BaSu RRLID A A Q T AL Q Q F G T G SSGS R LTT G NSV W H EK LE KKI A S F KLT E A ALLF S S GYL AN V G VLSS L PEKE D --V I L S AONS_ThKo P EIRY A AIR A ILDY G V G A G AV R T I A GT M EL H VE LE EK LA K F KKR E A A I LF Q S GYN AN L G AISA L LKKGEDGVFI S KBl_BaCe SRL Q E A AIG A IHKY G V G A G AV R T I N GT LDL H IK LE ETI A K F KHT E A A IAY Q S G F NC N M AAISAV M DK N D A -I L S KBl_BaSu P RLIN A A Q E A V Q Q Y G A G T G SV R T I A GT FT M H Q E LE KK LA A F KKT E A AL V F Q S G FT T N Q G VLSSI L SK E D I --VI S KBL_DrMe P EIVEHS Q KL L E Q Y G A G LS SV R F I C GT Q DI H K Q LE KKI A Q F H G R E DTI L YA SC FD AN A G IFEAI L TP E D A --VF S KBl_EsCo P DLIA A AKAG M DSH G F G M A SV R F I C GT Q DS H KE LE Q K LA A F L G M E D A I L YS SC FD AN G G LFE TL L GA E D A -I I S KBL_GrFo P E V I Q A AKDT M DTH G F G M S SV R F I C GT Q DI H KE LE Q KISD F Y G T E DTI L YAA C FD AN G G IFEP L L TK E D A -I I S KBl_MuMu P A V I Q A GL Q T L EEF G A G LS ST R F I C GT Q SI H KN LE AKI A H F H Q R E D A I L YP SC FD AN A G LFEA L L TP E D A --VL S KBL_RhEt EELAG A GK Q AL DRY G Y G M A SV R F I C GT Q EE H K Q LE ARISA F L G M E DTI L YS SC FD AN G G LFE TL L SE E D A -I I S KBl_RhLe EELAE A GK Q AL DRY G Y G M A SV R F I C GT Q EE H K Q LE ARISS F L G M E DTI L YS SC FD AN G G LFE TL L SE E D A -I I S KBl_StAu EDLKS A AKA A IDTH G V G A G AV R T I N GT LDL H DE LE ET LA K F K G T E A A IAY Q S G FN C N M AAISAV M NK N D A -I L S KBL_StAv P E V IA A AHE AL DR W G Y G M A SV R F I C GT Q EV H KE LE RR L SA F L G Q E DTI L YS SC FD AN G G VFE TL L GA E D A --VI S KBL_TrBr P E V I Q A AKD AL DSH G Y G LA SV R F I C GT TDI H KK LE Q T M TE F L G M E DTI L YP SC FD AN A G VFEA L L TS E D A -I I S FIGURE 2.1 Alignment of representative sequences of ALAS and KBL with closest AONS sequences. The abbreviations are explained in Table 2.1. Amino acid shading in black and grey represent residues with homology greater than 50% identity or similarity, respectively. Residues in red or blue indicate positions that are conserved (or nonconserved) in ALAS or KBL sequences, respec tively. (Numbering us ed in figure one is based on R. capsulatus ALAS crystal structure (Astner, Schulze et al., 2005).) (ConÂ’t on next page.)
63 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259ALAS_AsFu G A I T FL DE V HAVG M Y G PH G A G VA E HLDYEIYAS Q DTPNPRSTKG----T V M DRV DII T GTLGKA Y G -CV GGY I A G ALAS_DrMe G A I T FI DE V HAVG L Y G DH G A G VG E RD G ---------------------V LHK M DII S GTLGKA F G -NI GGY I A G ALAS_PlFa N A L T Y VDE V HAVG L Y G NK G S G YL E ELH----------------------LCNHI DII N GTL S KA I G -SL GG FICA ALAS_RhCa G A L T YI DE V HAVG M Y G PR G A G VA E RD G ----------------------L M HRI DI FN GTL A KA Y G -VF GGY I A A ALAS_RhEt G A M T YL DE V HAVG M Y G PR G G G IA E RE G ----------------------L M DRLTV I E GTLGKA F G -V M GGY I A A ALAS_RiBe N A L T FI DE V H T VG L Y G KT G A G IA E LLD----------------------CSDEI DII Q GTL A KA Y G -TI GGY ITA ALAS_SaCe G A L T FL DE V HAVG L Y G PH G A G VA E HCDFESHRASGIATPKTNDKGGAKT V M DRV D M I T GTLGK SF G -SV GGY V A A ALAS_S p SK N A L T YL DE V HAVG M Y G AR G G G IS E RDE---------------------V ADRVT II E GTLGKA F G -V M GGY I A A ALAS1_MuMu G A I T F VDE V HAVG L Y G AR G G G IGDRD G ---------------------V M PK M DII S GTLGKA F G -CV GGY I A S ALAS2_MuMu G A L T F VDE V HAVG L Y G AR G A G IG E RD G ----------------------I M HKL DII S GTLGKA F G -CV GGY I A S AONS_A q Ae DC M LYI DE A H TT G TI G ---K G GLDYF G I---------------------EHKEYI I V M GTL S KA L G -SY G AFVCG AONS_BaCe G A III VDE A HA S G IY G IG G A G LSHIEKN---------------------LS Q KI DI H M GT FS KA L G -CY G A Y LTG AONS_BaSu H A FVV VD D A HA T G VL G DS G Q G TS E YF G ---------------------V CP-DI VI GTL S KA V G -AE GG FA A G AONS_ThKo D A ILYI D D A H GE G VL G DS G R G IVDHFK----------------------LHDKV D FE M GTL S KA F G -VI GGY V A G KBl_BaCe DL M TY VD D A H GS G VL G -K G A G TVKHF G ----------------------LSDKV D F Q I GTL S KA I G -VI GGY V A G KBl_BaSu D A FV M VD D A HA S G VL G EN G R G TVNHF G ----------------------LDGRVH I Q V GTL S KA I G -VL GGY A A G KBL_DrMe N A LVF VDE C HA T G FF G AT G R G TE E YDN---------------------V M GEV DII NS TLGKA L G GAS GGY TTG KBl_EsCo D A LV M VD DS HAVG FV G EN G R G SH E YCD---------------------V M GRV DII T GTLGKA L G GAS GGY T A A KBL_GrFo D A M V M I DE C HA T G FI G EK G I G TP E EK G ---------------------V LDRV DII T GTLGKA L G GA M GGY TTA KBl_MuMu G A LVF VDE C HA T G FL G PT G R G TD E LL G ---------------------V M D Q VT II NS TLGKA L G GAS GGY TTG KBL_RhEt G A M V M VD DS HAVG FV G RN G R G SA E HC G ---------------------V EGRI DII T GTLGKA L G GAS GGY TSA KBl_RhLe G A M V M VD DS HAVG FV G KN G R G SP E YC G ---------------------V EGRI DII T GTLGKA L G GAS GGY TSA KBl_StAu GLLTY VD DA H GS G V M G -K G A G TVKHF G ----------------------L Q DKI D F Q I GTL S KA I G -VV GGY V A G KBL_StAv D A M V M VD DS HAVG FV G PG G R G TP E LH G ---------------------V M DRV DII T GTLGKA L G GAS GGY V A A KBL_TrBr N A NV M VD DS HA S G F M G PG G R G TPALF G ---------------------V IDKV DI LNT TLGKA L G GAS GG LSSG 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334ALAS_AsFu SAA M V D T I R SLA PG FIF T TSLPP AT M A GANTAILY Q ARHKGD-R VL Q Q LHTRAVKKA L KELDIPVIP-NPS H I V P ALAS_DrMe THNLV D M I R SYA AG FIF T TSLPP TVLCGALEAVNI L ASEEGR Q R HL H Q R N VSYLKSL L KRE G FPVEE-TPS H I I P ALAS_PlFa NKYY ID V I R SYS SH FIF T TSL T P VNINTSAEAIHII Q ND M SL-R KK T Q -VVNKTK Q K L Q ER G I Q VLH-NNS H I VV ALAS_RhCa SAK M V D A V R SYA PG FIF S TSLPP AIA A GA Q ASIAF L KTAEG Q KL R DA Q Q M HAKVLK M R L KAL G M PIID-HGS H I V P ALAS_RhEt SAALC D F I R SFA SG FIF T T A LPP ALA A GAVASI Q H L KVS Q FERR AR H Q DRVRKLRA M L D Q R G IPH M H-NPS H I V P ALAS_RiBe NHSL ID A I R LSA SG FIF T TSLPP VISTAATHSIRH L KESN Q ERR KT H Q Q VVSKLKSSFDRFNIPYLK-NES H I V P ALAS_SaCe SRKL ID W F R SFA PG FIF T T T LPP SV M A GATAAIRY Q RCHIDLRR TS Q Q KHT M YVKKAFHEL G IPVIP-NPS H I V P ALAS_S p SK D Q M IV D V I R SYA PG FIF T TSL S P VLV A GVLASVRH L KGSSEE-R EG Q Q ASAALLK Q L M RDA G LPV M N-SVT H I V P ALAS1_MuMu TSLL ID T V R SYA AG FIF T TSLPP M LL A GALESVRI L KSSEGRAR R Q H Q R N VKLLR Q M L M DA G LPVIH-CPS H I I P ALAS2_MuMu TRDLV D M V R SYA AG FIF T TSLPP M VLSGALESVRL L KGEEG Q AR RA H Q R N VKH M R Q L L M DR G FPVIP-CPS H I I P AONS_A q Ae TKLL ID Y L VNKA R SL IFSTSLPP SVC A GAKKAIEIIEENPKL--IEF L RKKEKEILEI L E Q FSLDYKY-YSTP I I P AONS_BaCe DEIY I EY L Q N M M R S FIF T T A LPP STLGAV Q KAIEIVKEDNER-R EN L IA N GEYFRTK L RDA G FDIGN-SSTH I V P AONS_BaSu SAVF ID F L LNHA R T FIF Q T AI PP ASC A AAHEAFNIIEASREK-R Q L L FSYIS M IRTS L KN M G YVVKG-DHTP I I P AONS_ThKo PEEA I EY LR Q RA R P F L FS SAPN P PDV A AAIAAVEI L Q RSDEL--VRK L W D N TNFL Q KG L RDL G YDLGN-TKHP I T P KBl_BaCe K Q NL ID W L KV R S R P F L F S T A L T P ADA A AC M RSIEI L M ESTEL--HDR L W E N GRYLK Q G L KEL G FNIGE-SETP I T P KBl_BaSu SKVL ID Y LR H K G R P F L F S TS H PP AVT A AC M EAIDV L LEEPEH-M ER L W E N TAYFKA M L VK M G LTLTK-SETP I L P KBL_DrMe PAEL I SF LR Q K S R P YL F S NT LPP AVV A VGLKV M D M L L Q SSEL--T Q RV Q S N T Q RFR Q A M TKA G FTIAG-ENHP I C P KBl_EsCo RKEVVE W LR Q R S R P YL F S N SL A P AIV A ASIKVLE M VEAGSEL-R DR L W A N AR Q FRE Q M SAA G FTLAG-ADHA I I P KBL_GrFo KKEI I EI LR Q R S R P YL F S N SL A P SIVGASIKVFD M L KNDDSL-R KK L KE N TAYFKKEIKEA G FEIID-GEAA I V P KBl_MuMu PEPLVSL LR Q R S R P YL F S N SLPP AVVGCASKALDL L M ESNAI--I Q S M AAKTRRFRSK M EAA G FTVSG-ADHP I C P KBL_RhEt KAEVVE W LR Q R S R P YL F S NT L A P VIA A ASLKVFDLIENGDAV-R KS L SD N ADLFRTE M TKL G FTLAG-EGHP I I P KBl_RhLe KAEVVE W LR Q R S R P YL F S NT L A P VIA A ASLKVFDLIENGDAL-R KR L SD N ADLFRTE M TKL G FKLAG-EGHP I I P KBl_StAu TKEL ID W L KA Q S R P F L F S TSL A P GDTKAITEAVKK L M ASTEL--HDK L W D N A Q YLKNG L SKL G YDTGE-SETP I T P KBL_StAv RAEIVAL LR Q R S R P YL F S NT L A P VIA A ASLKVLDL L ESADDL-R VR L AE N TALFRSR M TEE G FDILP-GDHA I A P KBL_TrBr CKEIV D L Q R Q K G R P YL F S NTIA P AAVGGTLKV M EL L Q TTSSA-R K Q L Q D N THLFRTE M KKA G FTLSGHEECP I A P 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388ALAS_AsFu LLV GDA ELAKKASDK LL EE H G IYV Q A I N YPTVP R G E E R L R IT P T P G H VKEHRDHL-------------ALAS_DrMe IKI GD PLKSS Q ISNV L IE Q F GH Y L Q S I N YPTV AR G Q E KL R LA P T P F HT FE M M NAL-------------ALAS_PlFa L M INS A EKCK Q ICDD LL KE Y N IY I Q P I N YPTVP M G M E R I R IT P S P F HT DE Q IFKL--------------ALAS_RhCa V VI GD PVHTKAVSD M LL SD Y GV YV Q P I N F PTVP R G T E R L R FT P S P V H DLK Q IDGL--------------ALAS_RhEt V M V GDA AKCK W ISDL LL DN C GV YV Q P I N YPTVP KKT E R L R IT P T P L H SDADIAHL--------------ALAS_RiBe III GD PIKASKASN M LL NE Y S IYV Q H I N F PTVP R G T E R L R II P T P A HT DE M INDL--------------ALAS_SaCe V LI G N A DLAK Q ASDI L INK H Q IYV Q A I N F PTV AR G T E R L R IT P T P G HT NDLSDIL--------------ALAS_S p SK L M V GD PVKAKRISDI LL AE Y GA YV Q P I N YPTV SR G T E R L R FT P G P A HT EE M M RDLVGAL----------ALAS1_MuMu V RVA DA AKNTEICDE L M TR H N IYV Q A I N YPTVP R G E E LL R IA P T P H HT P Q M M NFF--------------ALAS2_MuMu IRV G N A ALNSKICDL LL SK H S IYV Q A I N YPTVP R G E E LL R LA P S P H H SP Q M M ENF--------------AONS_A q Ae I M VY D EKETVRIKEE LL KE G -VFI Q A I R YPTVP K G KA R L R LTASLNY T RKDLEFL--------------AONS_BaCe IVV G SNEHALRFSKR L Q EA G I AAIA I RP PTVP VHSS R I R FAVTS Q HT IADLK W A--------------AONS_BaSu V VI GDA HKTVLFAEK L Q GK G IY APA I RP PTV AP G ES R I R ITITSD H S M GDIDHL--------------AONS_ThKo V M LY D EKLA Q EFSRR L YDEYN I FA Q A I V YPTVP L G TA R I R LE P SAA H SKEDL Q YVIDAFEDLGKKTGFLK KBl_BaCe CII GD EV L T Q EFSKR L NEE G -V Y AKS I VF PTV AK G T G R V R N M P T A A HT KE M LDEA--------------KBl_BaSu ILI GD EG V AK Q FSD Q LL SR G -VFA Q S I VF PTV AK G K A R I R TIIT A E HT KDELD Q A--------------KBL_DrMe V M L GDA R L AS Q FADE M L TR G IYV IGFS YP V VP Q G K A R I R V Q IS A A HT EAEIDRAINAFIEVGRSLKVIK KBl_EsCo V M L GDA V V A Q KFARE L Q KE G IYV TGFF YP V VP K G Q A R I R T Q M S A A HT PE Q ITRA--------------KBL_GrFo V M LH DA K L S Q D M ADK LL EE G IYV IGFF YP V VP K G K A R I R V Q LS A A H NKEHLDKAIAAFKKVGKELKVIKBl_MuMu V M L GDA R L SS Q M ADD M L KK G I F V IGFS YP V VP K G K A R I R V Q IS A V H SEEDIDRC --------------KBL_RhEt V M L GDA K L ---------------------------------------------------------KBl_RhLe V M L GDA K L A Q D M ASL M L RK G IYV IGFSF P V VP K G Q A R I R T Q M S A A H SRADVERA--------------KBl_StAu V II G EEK T T Q EFSKR L KDE G -V YV KS I VF PTVP R G T G R V R N M P T A A HT KD M LDEA--------------KBL_StAv V M I GDA A V AGRLAEL LL ER G -V YV IGFS YP V VP Q G Q A R I R V Q LS A A H STDDVNRAVDAFVSARAELEA-KBL_TrBr V M LYE A R I AAEFAAK M M AE G IYV TAFS YP V VP K G Q A R I R V Q LS A A HT TEDVKLAVDAFTKIKKELNV-FIGURE 2.1 (ConÂ’t)
64 AONS ThKo AONS EsCo AONS MyTu AONS ArTh AONS HePy AONS AsNi AONS TeTh AONS HaIn AONS BaCe AONS AqAe AONS BaSu AONS StAu AONS SpSk KBL GrFo KBL DrMe KBL TrBr KBL RhEt KBL StAv KBl StAu KBl BaCe KBl BaSu KBl EsCo KBl RhLe KBl MuMu ALAS2 MuMu ALAS1 MuMu ALAS DrMe ALAS AsFu ALAS SpSK ALAS RhCa ALAS SaCe AONS RhEt ALAS RiBe ALAS PlFa SPT2 ArTh SPT2 MuMu SPT2 SaCe Spt2 AsFu SPT1 AsFu SPT1 ArTh SPT1 MuMu SPT1 SaCe SPT SpSk SPT SpPaAmino Acid #395 376 370 399 359 397 389 376 378 361 375 371 374 397 394 404 335 398 379 379 377 381 379 380 390 389 390 402 391 389 406 386 382 404 408 409 408 409 405 385 390 407 402 382AONS ThKo32303024212832394240303041394040415153474040393131303234313035322931323033232824213135 AONS EsCo32323227302533303234243327282728303030323229302830292731302730272326222525201919182826 AONS MyTu30323125272229293133233424262722282926272526282725262628282628262226222123171716193030 AONS ArTh30323127242331333433262928272929282728332929302724272929313029312523222425151515172424 AONS HeP y 24272527281831273026242719181919162324212017192218232526222025252319211917151817122322 AONS AsNi21302724281930242928163021232021212523262123252625272427262426212221222019131515152022 AONS TeTh28252223181919252523182123202222252323232223242119221919212023212123232223141816172024 AONS HaIn32332931313019373435303229312928283031302928282722272632252531302923252223182117192727 AONS BaCe39302933272425373939303132312932303738383032292930302933302831312927292729192119202929 AONS A q Ae42323134302925343939303434323229323636373231343331323131323031323327312627212323222528 AONS BaS u 40343333262823353939273436363637363941453637383133333632323334323227282729242422212929 AONS StAu30242326241618303030272425272426262425262325232624252826242428262719212425181815151921 AONS S p S k 30333429273021323134342426262525282526272828282426282630282532282322272328182119192827 KBL GrFo41272428192123293234362526565760614141406260552930273230312930313025292428212320173035 KBL DrMe39282627182320313132362726565655553637455758683130293133302930322823302728202120212932 KBL TrBr40272729192022292932362425575659593635415759562826262627252426282424282627192220192631 KBL RhEt40282229192122283229372625605559673935376895533229303231313032273025282528232622212929 KBL StA v 41302828162125283032362628615559673537416467563129283432342934292724252427192320193036 KBl StA u 51302927232523303736392425413636393576493937342830302932313031313127322731222525213031 KBl BaCe53302628242323313836412526413735353776513736352727323131323035312926332931222424202929 KBl BaSu47322733212623303837452627404541374149514136423231323332323335323133333131232523183032 KBl EsCo40322529202122293032362328625757686439374167563232293332333034282923302831212024193037 KBl RhLe40292629172323283231372528605859956737363667543329303128302934293027282530232219183139 KBl MuM u 39302830192524282934382328556856535634354256543530313532323135322823312830222023202934 ALAS2 MuMu31282727222621272933312624293128323128273232333576645251504749473723262326151818172729 ALAS1 MuMu31302524182519223031332426303026292930273132293076655153504750483622242229161719172628 ALAS DrMe30292627232722273032332528272926302830323229303164655152484847503622252225161820162325 ALAS AsFu32272629252419262931362826323126323429313333313552515153546556473926272422162122162524 ALAS SpS K 34312829262719323331322630303327313232313232283251535253614762534023262525171917152627 ALAS RhCa31302831222621253032322428313025313431323233303250504854615159493723252522181817132629 ALAS SaCe30272630202420252830332425292924302930303330293147474865475149483822222323171918172326 AONS RhEt35302829252623313131342832303026323431353534343549504756625949503824252024182118152630 ALAS RiBe32272631252121303132322628313228272931313228293247485047534948504324242525172119182729 ALAS PlFa29232225232221292933322723302824302731293129302837363639403738384325262424162322192123 SPT2 ArTh31262623192123232727271922252324252427263323272323222226232322242425534749212123152727 SPT2 MuMu32222222212223252931282127293028282532333330283126242527262522252426535154242524192626 SPT2 SaCe30252124192022222726272423242726252427293128252823222224252523202524475162222220162324 S p t2 AsFu33252325171923232927292528282827282731313131303026292522252223242524495462182526232723 SPT1 AsFu23201715151314181921241818212019231922222321232215161616171817181716212422183738362017 SPT1 ArTh28191715181518212123241821232122262325242520222018171821191819212123212522253745322324 SPT1 MuMu24191615171516171923221519202020222025242324192318192022171718181922232420263845382423 SPT1 SaCe21181917121517192022211519172119211921201819182017171616151317151819151916233632381719 SPT S p Sk31283024232020272925291928302926293030293030312927262325262623262721272623272023241774 SPT SpPa35263024222224272928292127353231293631293237393429282524272926302923272624231724231974 Table 2.2
65 Phylogenetic analysis of the -oxoamine synthase family of PLP-dependent enzymes The neighbor-joining (NJ), maximum pa rsimony (MP) and minimum evolution (ME) methods were used to infer the evolutionary history of the -oxoamine synthase family of PLP-dependent enzymes (Eck and Dayhoff, 1966; Sa itou and Nei, 1987; Rzhetsky and Nei, 1992; Tamura, Dudley et al., 2007). Analyses were conducted using the MEGA4 software package (Figure 2.2) (Tamura, Dudley et al., 2007). The three methods differ in the way the trees are constructed. Both the NJ and ME are distance matrix methods utilizing pairwise distan ces for phylogenetic r econstruction. The ME criterion incorporates a tree-search protocol and accepts the tree with the shortest sum of branch lengths, and thus minimizes the tota l amount of evolution assumed (Rzhetsky and Nei, 1992). The NJ method analyzes only a sm all number of trees and employs a Â“greedy searchÂ” based on locally optim ized choices (Saitou and Ne i, 1987). In the MP method, the preferred phylogenetic tree is the tree that requires the least number of evolutionary changes. The MP criterion incorporates a tree search protocol in which trees are scored based on how many evolutionary transitions ar e required to explain the distribution of each character(Eck and Dayhoff, 1966). The bootstrap method of Felsenstein was used to evaluate the level of c onfidence associated with the phylogenetic trees; bootstrap values between 70% and 100% indicate signi ficant support for a branch (Felsenstein, 1985 ; Soltis and Soltis, 2003). The overall topologies of the three trees ha ve significant congruencies. Each tree contained two main branches, one formed by SPT1s and SPT2s and the second formed by AONSs, ALASs, SPTs and KBLs (Figure 2.2) In all three trees, ALASs were isolated within a single cluster and all KBLs a ppeared together in a single cluster, which
66 also included B. cereus and T. kodakarensis AONSs. The distribution of SPTs and AONSs deviated between the three trees. The bo otstrap probability values were generally > 50% among the branches containing ALAS s, KBLs, SPT1s and SPT2s, whereas the branches containing AONSs generally exhibited < 50% scores. This suggests that larger evolutionary distances separate represen tative AONS sequences, relative to the evolutionary distance between eith er ALAS or KBL sequences.
67 A. B. C. I I I II II II III III III KBL RhEt KBl RhLe KBl EsCo KBL StAv KBL GrFo KBL TrBr KBL DrMe KBl MuMu KBl BaSu AONS ThKo KBl StAu KBl BaCe AONS BaSu AONS ArTh ALAS PlFa ALAS RiBe ALAS2 MuMu ALAS1 MuMu ALAS DrMe ALAS AsFu ALAS SaCe ALAS RhEt ALAS SpSK ALAS RhCa AONS AqAe AONS StAu AONS HaIn AONS BaCe AONS MyTu AONS SpSk AONS EsCo AONS TeTh AONS HePy AONS AsNi SPT SpSk SPT SpPa SPT2 ArTh SPT2 MuMu SPT2 SaCe Spt2 AsFu SPT1 AsFu SPT1 SaCe SPT1 ArTh SPT1 MuMu 99 99 93 99 99 99 99 99 99 99 99 99 90 97 99 97 99 98 88 99 99 93 94 99 91 88 84 99 94 76 KBl RhLe KBL RhEt KBl EsCo KBL StAv KBL GrFo KBL TrBr KBl MuMu KBL DrMe KBl BaSu AONS BaSu AONS ThKo KBl BaCe KBl StAu ALAS RhCa ALAS SpSK ALAS RhEt ALAS SaCe ALAS AsFu ALAS DrMe ALAS1 MuMu ALAS2 MuMu ALAS PlFa ALAS RiBe AONS BaCe AONS HaIn AONS StAu AONS ArTh AONS AqAe AONS SpSk AONS MyTu AONS AsNi AONS HePy AONS TeTh AONS EsCo SPT SpSk SPT SpPa SPT2 ArTh SPT2 MuMu SPT2 SaCe SPT2 AsFu SPT1 AsFu SPT1 SaCe SPT1 ArTh SPT1 MuMu 100 100 100 100 99 93 100 99 100 100 98 56 86 80 100 100 100 76 92 83 59 50 98 89 68 68 66 61 57 51 KBL RhEt KBl RhLe KBL StAv KBl EsCo KBL GrFo KBL TrBr KBL DrMe KBl MuMu AONS ThKo KBl StAu KBl BaCe KBl BaSu AONS BaSu AONS AqAe AONS StAu AONS HaIn AONS BaCe AONS ArTh ALAS PFa ALAS RiBe ALAS2 MuMu ALAS1 MuMu ALAS DrMe ALAS AsFu ALAS SaCe ALAS RhEt ALAS SpSK ALAS RhCa SPT SpSk SPT SpPa AONS MyTu AONS SpSk AONS HePy AONS EsCo AONS AsNi SPT2 ArTh SPT2 MuMu SPT2 SaCe Spt2 AsFu SPT1 ArTh SPT1 MuMu SPT1 AsFu SPT1 SaCe AONS TeTh 99 78 52 99 99 99 83 58 77 99 90 80 99 92 99 99 63 59 54 53 50 97 53 FIGURE 2 Phylogenetic trees for the -oxoamine synthase family of PLP-dependent en zymes. The phylogenetic trees were derived using the (A) minimum evolution method, (B) maximum pars imony method and (C) neighbo r-joining method. Numbers on branches indicate local bootstrap probability with values > 50% (Felsenstein, 1985 ). (The branch numbers in the minimum evolution tree represent the percen tage of replicate trees in which the asso ciated taxa clustered together in the bootstrap test. The branch numbers in th e maximum parsimony tree represent the pe rcentage of replicate trees in which the associated taxa clustered together, as inferred from 500 replicates. The branch numbers in the neighbor-joining tree represent the confidence probability that the interior branch length is greater than 0 as estimated using the bootstrap test and multiplied by 100.)
68 Strategy for selection of evolved ALAS variants with AONS activity The isolation of ALAS va riants evolved to acquire AONS activity relied on the development of a biological selection system. The bioF mutant E. coli R872 cells have no AONS function and thus are biotin-auxotro phic. When grown on a minimal plate, R872 cells cannot grow without biotin supplementation (> 100 pg/ml). Thus the biological system involved th e reversion of the biotin auxotrophy by transformation of the E. coli R872 cells with ALAS variants exhibi ting AONS activity. It should be noted that when transformed with a plasmid expressing E. coli AONS, the growth of R872 cells in a medium without biotin was rescued, wh ile transformation of R872 cells with an ALAS expression plasmid did not rescue their growth in a medium without biotin (data not shown). To increase the probability of selecting ALAS variants which would confer R872 cells the ability to synthesize biotin, the initial round of screeni ng was performed in a medium containing 5 pg/ml biotin, while biotin was omitted in the subsequent rounds of screening to the medium. Thus, a simple and efficient bioassay was developed to further confirm and compare AONS function in ALAS va riants selected in the initial screen. The bioassay also took advantage of the bioF mutant E. coli R872 strain. The assay was based on previous studies by Hwang et. al. who described that the growth of an E. coli bioF deletion strain in minimal medium was depe ndent on the addition of biotin, in the 0 10 ng/ml concentration range (Hwang, Su et al., 1999). The bioassay was modified to follow growth of R872 cells in 24-well plates (2. 3). Using the modified assay, AONS function in multiple ALAS variants can be efficiently screened by following the growth of R872 cells.
69 FIGURE 2.3 Growth of R872 cells in minimal media. Grow th of bacterial R872 cells as function of biotin concentration. Cell growth was monito ring by determining th e absorbance at 600 nm (OD600 nm) following 24 hours of incubation in mi nimal medium or minimal medium supplemented with biotin. ( Inset : Linear plot following R872 cell growth with increasing concentrations of biotin.) Biotin Concentration (pg/ml) 051001000100000 OD 600nm 0.00 0.25 0.50 0.75 Biotin Concentration (pg/ml) 05e+41e+5 OD 600nm 0.0 0.3 0.6
70 Directed Evolution of ALAS and screening for AONS function Error-prone PCR was used to create an ALAS cDNA library with a low mutation rate and thus reduce the number of simultaneous, hindering amino acid mutations and simplify the analysis of ALAS variants exhibiting AONS function (Zhao, Giver et al., 1998). The library wa s initially transformed into DH5 E. coli because high efficiency transformations of the ALAS libraries could not be attained with E. coli R872 Plasmids isolated from approximately 364,000 DH5 transformants were subsequently transformed into R872 cells. Approximately 1.5 x106 R872 transformants were plated on minimal plates supplemented with 5 pg/ml bio tin to screen for variants exhibiting AONS function. A total of 21 varian ts were selected for scr eening using the bioassay. Of the 21 selected variants screened using the AONS bioassay, one variant (B2) exhibited significantly higher cel l density after 24 hours in me dium supplemented with 5 and 10 pg/ml of biotin and two (E5 and E6) exhibited increased cell density at a single biotin concentration when compared to R872 cells transformed with an Â“emptyÂ” vector ( i.e. containing neither the AONSnor the ALAS-encoding DNA) (Figure 2.4). To ensure that the increased growth resulted from the transformation with evolved ALAS variants and not reversion of the R872 st rain cells, plasmids expressing the ALAS variants were obtained from the cells exhibi ting increased growth rates and transformed into new batches of freshly made competent R872 cells. (Plasmid encoding the E6 variant could not be obtained.) The bioassay was repeated using both the progeny of the screened R872 cells harboring plasmids encoding ALAS variants and R872 cells freshly transformed with the
71 isolated plasmids. When cell growth was monitored between 20 and 50 hours, growth observed with the freshly transformed cells wa s comparable to that of bacterial cells transformed with the Â“emptyÂ” vector controls in medium supplemented with 1, 5 or 10 pg/ml of biotin, while the proge ny of cells isolated during sc reening continued to exhibit growth over negative controls at all biotin concentrations (Figure 2.5). This indicated that cells were not rescued by the expressi on of the ALAS variants selected upon screening for AONS function.
72 0 0.1 0.2 0.3Cass3 Cass3 A1 A2 A3 A4 A5 A6 A7 A8 B1 B2 B3 B4 E3 E4 E5 E6 F13 F2 F3 F4 F5VariantsOD600 FIGURE 2.4 Growth of R872 cells expre ssing ALAS variants. Comparison of cell growth as measured by OD at 600 nm for R872 cells ex pression ALAS variants and the Â“emptyÂ” vector (Cass3) at 24 hours at 5 (whi te) and 10 (grey) pg/ml of biotin.
73 Time (hours) 01020304050 OD600nm 0.0 0.3 0.6 0.9 1.2 Time (hours) 01020304050 OD600nm 0.0 0.3 0.6 0.9 1.2 Time (hours) 01020304050 OD600nm 0.0 0.3 0.6 0.9 1.2 A B C FIGURE 2.5 Comparison of cell growth of freshly tran sformed R872 cells. R 872 cells expressing selected ALAS variants in minimal medium w ith (A) 10, (B) 5 or (C) 1 pg/ml of biotin. Cell growth was monito red by measuring the OD600 nm of the cell cultures. Red lines indicate growth for R872 cells expressing AONS, blue lines indicate growth for cells transformed with the Â“emptyÂ” vector, solid black lines indicate freshly transformed cells and dashed black indicates progeny from initial transformation of selected variants. Symbols indicates ALAS varian ts expressed in R872 cells, =B2 variant, =E5 variant and =E3 variant.
74 The bioassay was repeated using both the progeny of the screened R872 cells harboring plasmids encoding ALAS variants and R872 cells freshly transformed with the isolated plasmids. When cell growth was monitored between 20 and 50 hours, growth observed with the freshly transformed cells wa s comparable to that of bacterial cells transformed with the Â“emptyÂ” vector contro ls in medium supplemented with 1, 5, or 10 pg/ml of biotin, while the proge ny of cells isolated during sc reening continued to exhibit growth over negative controls at all biotin concentrations (Figure 2.5). This indicated that cells were not rescued by the expressi on of the ALAS variants selected upon screening for AONS function. Identification of residues that potentially regulate -oxoamine synthase specificity Given that no evolved ALAS variants were generated to identify residues with an ability to influence substrate specificity in the -oxoamine synthase family, we proceeded to scrutinize the primary structure and search for residue s potentially involved in the regulation of specificity in ALAS and KBL. Residues were identified that may have a role in cont rolling specificity by taking advantage of an improved understanding of the evolutionary relationships among this group of enzymes. The evolutionary space between AONS and ALAS or KBL sequences varied widely, 10-36% or 17-53% identity betw een AONS and ALAS or AONS and KBL sequences, respectively. Becau se of this large variation, we reasoned that AONS sequences closest to those of ALAS or KBL we re likely to be most similar to a shared ancestor sequence. Therefore, the AONS sequences closest to ALAS and KBL were used to help us suggest residue variations that may have influenced changes in enzyme
75 specificity. The phylogenetic tr ees and % identity scores we re used to select AONS sequences most similar to ALAS and KBL sequences. AqAe, BaCe, BaSu, and ThKo AONS sequences were most similar to both ALAS and KBL sequence sets. (Table 2.2). (Species abbreviations defined in Table 2. 1.) In order to identify residues that may play a role in determining enzyme specificity, the ClustalW-aligned ALAS and KBL sequences were compared to the 4 selected AONS sequences (Figure 2.1). The 37 residues identified and their structural position relative to the ALAS crystal structure are summarized in Table 2.2. Figures 2.6 and 2.7 illustrate the residues positioned near the SCoAand glycine-binding pockets, with the majority of the identified residues being located at or adjacent to the active site. Discussion Phylogenetic Analysis The evolutionary distances among all 44 -oxoamine synthase sequences and the 13 AONS sequences, as determined by the nu mber of amino acid substitutions, is equivalent and twice that observed among ALAS, KBL or SPT2 sequences. Moreover, the clustering of ALASs, KBLs and SPTs in si ngle branches and the bootstrap probability scores below 50% contrast with the relative ly broad distribution of AONSs in all three evolutionary trees. Together these data are consistent with the proposition of the appearance of AONS function early in the evolutionary time line of the -oxoamine synthases and the subsequent development of ALAS, KBL, SPT and SPT1/2 function. The -family of PLP-dependent enzymes cons ists of enzymes essential for amino acid metabolism as well as synthesis of the v ital protein cofactors heme and biotin (Ploux
76 and Marquet, 1992; Ferreira and Gong, 1995) AONS and 7,8 diaminopelargonic acid aminotransferase catalyze the first two common reactions in the biosynthesis of biotin (Schneider, G. and Lindqvist, Y. 2001) and are two of only six members of the -family of PLP dependent enzymes fo r which sequences have been found in all biological kingdoms (Mehta and Christen, 1998), suggesti ng that the biotin biosynthesis pathway was likely developed in the univers al ancestor cell. Consider ing that biotin is essential for all organisms and there are no naturally occurring alternative pathways known for biotin biosynthesis, the assertion of an early evolutionary appearance of AONS is plausible. Strategy for selection of evolved ALAS variants The selection system developed is based on the auxotrophy of the bioF mutant E. coli R872 strain, which does not exhibit AO NS function. The R872 strain of E. coli was obtained for this project on the premise that BioF gene, encoding AONS, was knocked out. Late into the first round screening fo r AONS function, it was discovered that this auxotrophic strain was a mutant and not a knockout strain (Del Campillo-Campbell, Kayajanian et al., 1967). Utilizing a mutant BioF strain instead of a BioF knockout strain, could permit BioF reversions that restore AONS function and eliminate biotin dependency. Therefore, ALAS variants-encoding plasmids, isolated during the initial screening for AONS function, were retransformed into new batches of R872 cells in orde r to verify whether the growth observed during se lection was not attributed to the reversion of the BioF mutant. Unfortunately, the freshly transf ormed cells could not grow in a medium supplemented with biotin concentrations used during selection. This almost certainly
77 indicates that the growth obs erved during the selection of the ALAS variants resulted from the reversion of the R872 BioF gene and is not attributed to the ALAS variants. This conclusion is further supported by the DNA sequencing data of the expression plasmid for the variant B2, which indicated th e absence of amino acid substitutions in the B2 variant-encoding cDNA, albeit the apparent ability of R872 cells harboring the B2 variant to grow in mi nimal medium or minimal medium su pplement with a limited biotin concentration (Figure 2.4). For directed evolution of AONS function to be feasible, an efficient selection is required. While a biotin auxotrouph encoding a mutant BioF gene can effectively select for AONS function, the efficiency is compromised with the screening of large libraries when reversions result in the selection of false positives. The flawed selection strategy prevented the continuation of this project. Identification of residu es that potentially regulate specificity in -oxoamine synthase While no ALAS variants with altered substrate specificity were identified using directed evolution, the enhanced understand ing of the evolutiona ry relationships among the -oxoamine synthases allowed for the iden tification of residues which potentially influence specificity. The large evolutio nary distance among AONS sequences resulting from an early evolutionary appearance of AONS presents the possibili ty that significant diversity was present in the -oxoamine synthases prior to the appearance of ALAS and KBL function. Thus, the ALAS and KBL precurs or sequences were likely most similar to AONS sequences with the closest identity to the ALAS and KBL sequence sets. By comparing the ALAS and KBL primary structures with the closest AONS relatives we identified the sequential position of residues that may function to regulate
78 enzyme specificity (Table 2.3). The ALAS crystal structure from R. capsulatus provided insight into the role of various positions identified. The majority of the 37 sequential positions were clustered near the active site (Figure 2.6 and 2.7); in fact, only 4 positions were distal to the active site. Interestingly, nine of the identified active site positions were occupied with residues that interact with either a substrate or th e PLP cofactor in the ALAS crystal structure. In the R. capsulatus ALAS crystal structure seven of the identified residues are involved in succinyl-CoA r ecognition. Gly82, Thr83, Asn85 and Ile86 are located in a glycine-rich stretch directly involved in positioning th e carboxylate group of succinylCoA (Astner, Schulze et al., 2005). The carboxylate is positioned through a van der Waals interaction with Ile 86 and hydrogen bonds betwee n the two carboxylate O and Thr83 and Asn85. Lys156 and Ile149 also have a role in succinyl-CoA recognition through a hydrogen bond between the O3Â’ of the succinyl-CoA ribose moiety and Lys156-N and a hydrophobic interaction between the adenine moiety and Ile149 (Astner, Schulze et al., 2005). Asn362 in the R. capsulatus ALAS is equivalent to Arg349 in the E. coli AONS crystal structures. Both ALAS Asn362 and AONS Arg349 are hydrogen-bonded to an N-terminal argini ne in the holoenzyme (ALAS or AONS) and respective SCoAor AON-bound structures. Th is arginine forms a hydrogen bond with the SCoA or AON carboxylate in the ALAS or AONS structure, respectively, and likely has a role coordinating the C-terminal tr ansition upon SCoA or AON binding (Webster, Alexeev et al., 2000; Astner Schulze et al., 2005). The crystal structures of -oxoamine synthases also reveal an active site loop, which upon the transition of the enzyme from the open to closed conformation, closes
79 over the active site (Webster, Alexeev et al., 2000; Astner, Schulze et al., 2005). In ALAS, the closure of this loop, comprised of residues 358-374, over the active site is important for the proper orientation of succinyl -CoA for catalysis (Ast ner, Schulze et al., 2005). Three residues identified, Gl n359, and Glu379 along with the previously mentioned Asn362, are at positions lo cated on this active site loop in R. capsulatus ALAS (Astner, Schulze et al., 2005). The three above identified residues appear to be involved in positioning the PLP cofactor in the R. capsulatus ALAS crystal structure. Tyr1 16 and Ser277 form two of six hydrogen bonds with the phosphate group (Ast ner, Schulze et al., 2005). At position 216 of the amino acid alignment, a Val residue is conserved among all ALAS sequences, while Ala is conserved among all AONS, SPT and SPT2 sequences (Figure 2.1). The corresponding residue in the ALAS and AONS crystal structures (Val216 and Ala206, respectively) forms hydrophobic contacts with PLP and functions to stabilize the pyridinium ring from underneath (Alexeev, Al exeeva et al., 1998; Astner, Schulze et al., 2005). Equivalent residues in aspartate am inotransferase and 1-aminocyclopropane-1carboxylate were mutated to isoleucine, i.e. Ala224Ile and Ile 232Ala, respectively, resulting in a 4-10 fold decrease in kcat/ Km and altered p Ka values for the internal aldimine in both enzymes (Eliot and Kirsch, 2002). Hereditary sideroblastic anemia or XLSA is attributed to point mutations in ALAS2 (Bottomley, May et al., 1995) Fi ve of the residues identified in the R. capsulatus ALAS structures are associated with mutations observed in XLAS patients (Astner, Schulze et al., 2005). Mutations at four of these resi dues, including those previously cited, Lys156 as well as Phy23, Gly273 and Leu133, disrupt the binding of the
80 succinyl-CoA substrate (Astner, Schulze et al., 2005). XLSA mutations impacting PLP affinity also occur at positions that were identified; these residues are equivalent to Lys30 and the abovementioned Gly273 (Astner, Schulze et al., 2005). While the significance of the residues occupying the positions identified as determining in substrate selectivity and/or discrimination can not be completely understood with the crystal structures alone, th e location of the majority of these residues at or near the active si te does not contradict the propositi on that they may be important in regulating substrate specific ity in ALAS, KBL and AONS. In fact, one third of the positions identified using the alignment are equi valent to ALAS residues that are either involved in direct interactio ns with succinyl-CoA or PLP, associated with XLSA, or located on an active site loop that functions to position succinyl CoA. This indicates that many of the residues located at the identified positions do have a role in enzyme function and determining function specificity. Furthe rmore, enzyme-dependent residue variation at each of the positions also suggests that the significance of each position for ALAS, AONS and KBL function is different. In summary, sequences were selected to represent the broad taxonomic and phylogenetic distribution of each member of the -oxoamine synthase subfamily of PLPdependent enzymes and to better comprehe nd their evolutionary relationships. The enhanced understanding of these relationships may have helped to provide insight into the significance of the primary structure di fferences in laboratory evolved variants. While no evolved variants we re identified, the knowledge gained by the phylogenetic analysis did aid in the identification of se quential positions at which amino acid residues varied according to enzyme function. Furthe rmore, one third of the identified amino
81 acids occupied relevant active site positions in the ALAS crystal structure and/or were previously recognized to be mutated in XLSA.
82# ALAS AONSKBL Role in ALAS Crystal Structure 3YXXRemote23FR,L,II,E Adjacent to active site (ALAS mutation disrupts SCoA b inding and Nand C-terminal interactionsa)30A,K (N) E,P,QQ (N) Adjacent to active site (ALAS mutation disrupts PLP b inding and enzyme stabilit y a) 50V (M) D,N (L,V) N(Q)Adjacent to residue involved in glycine binding55DN,DN Adjacent to active site N54 which forms hydrogen bond with the carboxylate of glycine82GA,GS Glycine rich stretch that positions terminal carboxylate of SCoA 83TSV Forms hydrogen bond with SCoA carboxylate O (In AONS, S has key role in governing enzyme specifityb) 85NL (T, H) F,T Glycine rich stretch that positions terminal carboxylate of SCoA86IV, L (T, A) I Glycine rich stretch that positions terminal carboxylate of SCoA103LF,W (L,R) FRemote104HXXRemote106KXXRemote113S (T) XXAdjacent116Y,FF,YFForms hydrogen bond with phosphate O 117I,VXD (T) Adjacent127A,G (R) XL,MAdjacent to active site and surface loop 130-134 130132M,L (F)P (N,K) -X XD,E(S,C)-X E (N) D-A (I) Surface loop adjacent to active site133134C,L (I) E,I,V DeletedDeleted Surface loop adjacent to active site (ALAS 133 mutation disrupts SCoA adenine binding pocket reducing SCoA affinitya) 147XDD Adjacent to active site149IV,C,MV,C,MBottom of hydrophobic pocket for SCoA adenine moity156KXK, R Forms hydrogen bond with SCoA ribose moity ALAS (ALAS mutation reduces SCOA affinit y a) 159F (Y,W) F,Y (T,A,V) Y (V) Adjacent to active site184F T (S) TAdjacent to active site185EDD Forms hydrogen bond with H142 which positions PLP from above 216VAA,C,S ALAS and AONS su pp orts the PLP rin g from belowa,b270Y,F (L,S) XK,R270-272 LOOP between 9 and 10 PLP binding271A (S) A (M) S,G270-272 LOOP between 9 and 10 PLP binding272P,A,SR (K,T) R, I270-272 LOOP between 9 and 10 PLP binding273G (H) XP ALAS mutation disrupts residues involved in both SCoA and PLP binding 277T (S) S, T, QSShares hydrogn bond with Phophate O359Q Q,I,PX Loop of C terminal loop that moves to widen the active site channel in open and closed conformation362NR (M,E,V) V, S, F Loop of C terminal loop the moves to narrow the active site channel in open and closed conformation371EX A (G) Loop of C terminal loop the moves to narrow the active site channel in open and closed conformation379PA, S, LA Remote-Possible role in intersubunit interaction TABLE 2.3 X = none conserved Italic = Occurs in single sequence a(Astner, Schulze et al., 2005)b(Alexeev, Alexeeva et al., 1998)
83 FIGURE 2.6 Succinyl-CoA binding pocket of R. capsulatus ALAS. Ribbon structure of the ALAS homodimer (PDB code:2BWO) active site in which one subunit is depicted in blue with the active site loop in green and the adjacent subunit is in grey (A) and (B) Cartoon representation of active site residue s with the succinyl -CoA substrate ( red ) depicted in ball and stick representation (carbon in green oxygen in red nitrogen in blue ). Variable residues according to enzy me function within the -oxoamine synthase family, which interact with or lie adjace nt to succinyl-CoA, are in yellow and residues that lie adjacent to active site loop are green. Images were generated with Deep View (Guex and Peitsch, 1997; Kaplan and Littlejohn, 2001). G273 P272 A271 Y270 G82 T83 F23 N85 V50 Q359 E371 G273 P272 A271 Y270 G82 T83 F23 N85 V50 Q359 E371 K156 R127 I149 I158 F159 P272 P272 K156 R127 I149 I158 F159 P272 P272 A B
84 FIGURE 2.7 Active site interactions with PLP cofactor and glycine substrate in R. capsulatus ALAS (A) Ribbon structure of the ALAS homodimer active site (PDB code: 2BWP) in which one subunit is depicted in blue and the adjacent subunit is in grey. Cartoon representation of active site residues with the PLP-bound glycine substrate ( red ) depicted in ball and stick representation (carbon in green oxygen in red nitrogen in blue ). Variable residues according to enzyme function within the a-oxoami ne synthase family, which interact with or lie adjacent to PLP or glycine, are in green (B) Ball and stick representation with ribbon structure removed for cl arity. (Guex and Peitsch, 1997; Kaplan and Littlejohn, 2001) Y270 P272 N117 A271 Y116 G273 S277 V216 D55 Y270 P272 N117 A271 Y116 G273 S277 V216 D55 A B
85 References Alexander, F. W., E. Sand meier, et al. (1994). "Evo lutionary relationships among pyridoxal-5'-phosphate-dependent enzyme s. Regio-specific alpha, beta and gamma families." Eur. J. Biochem 219 (3): 953-960. Alexeev, D., M. Alexeeva, et al. (1998). "The crystal structure of 8-amino-7oxononanoate synthase: a bacterial PL P-dependent, acyl-CoA-condensing enzyme." J. Mol. Biol. 284 (2): 401-419. Astner, I., J. O. Schulze, et al. (2005). "Crystal structure of 5-aminolevulinate synthase, the first enzyme of heme biosynthesi s, and its link to XLSA in humans." Embo. J. 24 (18): 3166-3177. Bertani, G. (1951). "Studies on lysogenesis. I. The mode of phage liberation by lysogenic Escherichia coli ." J Bacteriol 62 (3): 293-300. Bhor, V. M., S. Dev, et al. (2006). "Bro ad substrate stereospecificity of the Mycobacterium tuberculosis 7-keto-8-aminopelargonic acid synthase: Spectroscopic and kinetic studies." J. Biol. Chem. 281 (35): 25076-25088. Bishop, D. F. (1990). "Two different genes en code delta-aminolevulinate synthase in humans: nucleotide sequences of cDNAs for the housekeeping and erythroid genes." Nucleic Acids Res. 18 (23): 7187-7188. Bottomley, S. S., B. K. May, et al. ( 1995). "Molecular defects of erythroid 5aminolevulinate synthase in Xlinked sideroblastic anemia." J. Bioenerg. Biomembr. 27 (2): 161-168. Cadwell, R. C. and G. F. Joyce (1992). "R andomization of genes by PCR mutagenesis." PCR Methods Appl 2 (1): 28-33. Chow, M. A., K. E. McElroy, et al. (2004). "Narrowing substrat e specificity in a directly evolved enzyme: the A293D mutant of aspartate aminotransferase." Biochemistry 43 (40): 12780-12787. Christen, P. and P. K. Mehta (2001). "From cofactor to enzymes. The molecular evolution of pyridoxal-5'-phos phate-dependent enzymes." Chem. Rec. 1 (6): 436447. Christen, P. and D. E. Metzler (1985). Transaminases. New York, Wiley. Del Campillo-Campbell, A., G. Kayajanian, et al. (1967). "Biotin-requiring mutants of Escherichia coli K-12." J. Bacteriol. 94 (6): 2065-2066.
86 Eck, R. and M. Dayhoff (1966). Atlas of Protein Sequence and Structure. Silver Springs, Maryland, National Biomedical Research Foundation. Eliot, A. C. and J. F. Kirsch (2002). "Modul ation of the internal aldimine pK(a)'s of 1aminocyclopropane-1-carboxylate synthase and aspartate aminotransferase by specific active site residues." Biochemistry 41 (11): 3836-3842. Eliot, A. C. and J. F. Kirsch (2004). "Pyridoxal phosphate enzy mes: mechanistic, structural, and evolutio nary considerations." Annu. Rev. Biochem. 73 : 383-415. Felsenstein, J. (1985 ). "Confidence limits on phylogenies: An approach using the bootstrap." Evolution 39 : 783-791. Ferreira, G. C. (1999). Iron Metabolism. Inorganic Biochemistry and Regulatory Weinheim, Germany, Wiley-VCH. Ferreira, G. C. and H. A. Dailey (1993). "Expression of mammalian 5-aminolevulinate synthase in Escherichia coli. Overproduc tion, purification, and characterization." J. Biol. Chem. 268 (1): 584-590. Gong, J. and G. C. Ferreira (1995). "Aminol evulinate synthase: functionally important residues at a glycine loop, a putative pyr idoxal phosphate cof actor-binding site." Biochemistry 34 (5): 1678-1685. Guex, N. and M. C. Peitsch (1997). "S WISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling." Electrophoresis 18 (15): 27142723. Han, G., K. Gable, et al. ( 2006). "Expression of a novel marine viral single-chain serine palmitoyltransferase and construction of yeast and mammalian single-chain chimera." J. Biol. Chem. 281 (52): 39935-39942. Hult, K. and P. Berglund (2007). "Enzyme pr omiscuity: mechanism and applications." Trends Biotechnol. 25 (5): 231-238. Hwang, S. Y., V. Su, et al. ( 1999). "Bioassay of biotin concentration with a Escherichia coli bio deletion mutant." J Biochem Biophys Methods 39 (1-2): 111-114. Jansonius, J. N. (1998). "Structure, evolut ion and action of vitamin B6-dependent enzymes." Curr. Opin. Struct. Biol. 8 (6): 759-769. Kaplan, W. and T. G. Littlejohn ( 2001). "Swiss-PDB Viewer (Deep View)." Brief Bioinform 2 (2): 195-197.
87 Katoh, K., K. Kuma, et al. (2005). "MAFFT version 5: improvement in accuracy of multiple sequence alignment." Nucleic Acids Res. 33 (2): 511-518. Kuchner, O. and F. H. Arnold (1997). "D irected evolution of enzyme catalysts." Trends Biotechnol 15 (12): 523-530. Longenecker, J. B., M. Ikawa, et al. (1957) "The cleavage of alpha-methylserine and alpha-methylolserine by pyr idoxal and metal ions." J. Biol. Chem. 226 (2): 663666. Mehta, P. K. and P. Christen (1998). "The Molecular Evolution of Pyridoxal 5-Phosphate Dependent Enzymes." Adv. Enzymol. Relat. Areas Mol. Biol. 74 : 129-185. Mehta, P. K. and P. Christen (2000). "T he molecular evolution of pyridoxal-5'phosphate-dependent enzymes." Adv. Enzymol Relat. Areas Mol. Biol. 74 : 129184. Metzler, D. E. and E. E. Snell (1952). "Deamin ation of serine. I. Catalytic deamination of serine and cysteine by py ridoxal and metal salts." J. Biol. Chem. 198 (1): 353-361. Miyazaki, K. and M. Takenouchi (2002). "Cre ating random mutagenesis libraries using megaprimer PCR of whole plasmid." Biotechniques 33 (5): 1033-1034, 10361038. Moore, J. C., H. M. Jin, et al. (1997). "Strategies for the in vitro evolution of protein function: enzyme evolution by random r ecombination of improved sequences." J Mol Biol 272 (3): 336-347. Nei, M. and S. Kumar (2000). Molecular Evolution and Phylogenetics. New York, Oxford University Press. Olivard, J., D. E. Metzler, et al. (1952). "Catalytic racemization of amino acids by pyridoxal and me tal salts." J. Biol. Chem. 199 (2): 669-674. Percudani, R. and A. Peracchi (2003). "A genomic overview of pyridoxal-phosphatedependent enzymes." EMBO Rep. 4 (9): 850-854. Ploux, O. and A. Marquet (1992). "The 8-amino-7-oxopelargonate synthase from Bacillus sphaericus. Purification and preliminary characterization of the cloned enzyme overproduced in Escherichia coli." Biochem J. 283 ( Pt 2) : 327-331. Rothman, S. C., M. Voorhies, et al. (2004 ). "Directed evolut ion relieves product inhibition and confers in vivo functi on to a rationally designed tyrosine aminotransferase." Protein Sci. 13 (3): 763-772.
88 Rzhetsky, A. and M. Nei (1992). "A simple method for estimating and testing minimum evolution trees." Molecular Biology and Evolution 9 : 945-967. Rzhetsky, A. and M. Nei (1992). "A simple method for estimating and testing minimum evolution trees." Molecular Biology and Evolution 9 : 945-967. Saitou, N. and M. Nei (1987). "The ne ighbor-joining method: a new method for reconstructing phylogenetic trees." Mol. Biol. Evol. 4 (4): 406-425. Schmidt, A., J. Sivaraman, et al. (2001). "Three-dimensional structure of 2-amino-3ketobutyrate CoA ligase from Escherichi a coli complexed with a PLP-substrate intermediate: inferred reaction mechanism." Biochemistry 40 (17): 5151-5160. Schneider, G., H. Kack, et al. (2000). "The manifold of vi tamin B6 dependent enzymes." Structure 8 (1): R1-R6. Schneider, G., H. Kack, et al. (2000). "The manifold of vi tamin B6 dependent enzymes." Structure 8 (1): R1-6. Schneider, G. and Y. Lindqvist (2001). "Struc tural enzymology of biotin biosynthesis." FEBS Lett. 495 (1-2): 7-11. Schwartz, R. M. and M. O. Dayhoff (1979). "Protein and Nucleic Acid Sequence Data and Phylogeny." Science 205 (4410): 1038-1039. Shafikhani, S., R. A. Siegel et al. (1997). "Generation of large libraries of random mutants in Bacillus subtilis by PCR-based plasmid multimerization." Biotechniques 23 (2): 304-310. Smith, H. O. and M. Levine (1964). "Two Sequential Repressions of DNA Synthesis in the Establishment of Lysogeny by Phage P22 and Its Mutants." Proc. Natl. Acad. Sci. U S A 52 : 356-363. Snell, E. E. and B. M. Guirard (1943). "S ome Interrelationships of Pyridoxine, Alanine and Glycine in Their Effect on Certain Lactic Acid Bacteria." Proc. Natl. Acad. Sci. U S A 29 (2): 66-73. Soltis, D. E. and P. S. Soltis (2003). "The ro le of phylogenetics in comparative genetics." Plant Physiol 132 (4): 1790-1800. Tamura, K., J. Dudley, et al. (2007). "MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) Software Version 4.0." Mol. Biol. Evol. 24 (8): 1596-1599. Thompson, J. D., D. G. Higgins, et al. (1994 ). "CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting,
89 position-specific gap penalties and weight matrix choice." Nucleic Acids Res. 22 (22): 4673-4680. Webster, S. P., D. Alexeev, et al. (2000). "Mechanism of 8-amino-7-oxononanoate synthase: spectroscopic, kinetic, and crystallographic studies." Biochemistry 39 (3): 516-528. Webster, S. P., D. J. Campopiano, et al (1998). "Characteris ation of 8-amino-7oxononanoate synthase: a bacterial PL P-dependent, acyl CoA condensing enzyme." Biochem. Soc. Trans. 26 (3): S268. Yano, T., S. Oue, et al. (1998). "Directed evolution of an aspa rtate aminotransferase with new substrate specificities." Proc. Natl. Acad. Sci. U.S.A. 95 (10): 5511-5515. Yard, B. A., L. G. Carter, et al. (2007). "The structure of serine palmitoyltransferase; gateway to sphingolipid biosynthesis." J. Mol. Biol. 370 (5): 870-886. Zhao, H. and F. H. Arnold (1997). "Combinatori al protein design: stra tegies for screening protein libraries." Curr Opin Struct Biol 7 (4): 480-485. Zhao, H., L. Giver, et al. (1998). "Molecula r evolution by staggered extension process (StEP) in vitro recombination." Nat. Biotechnol. 16 (3): 258-261.
90 Chapter Three Functional asymmetry for active sites of single chain homoand chimeric dimers of 5-aminolevulinate synthase and 8-amino-7-oxononanoate synthase Abstract 5-Aminolevulinate synthase (ALAS) and 8-amino-7-oxononanoate synthase (AONS) are homodimeric members of the -oxoamine synthase family of pyridoxal 5Â’phosphate (PLP)-dependent enzymes. Previo usly, linking two ALAS subunits into a single polypeptide chain dimer re sulted in an enzyme (ALAS/ALAS) with a significantly greater turnover number than th at of wild-type ALAS. To examine the contribution of each active site to the enzymatic activity of ALAS/ALAS, the conserved lysine, which covalently binds the PLP cofactor, was substituted with alanine yielding single chain dimeric ALAS variants with one of the tw o active sites having no measurable enzymatic activity. Albeit the amount of ALA produced during the firs t turnover was identical in both active sites of ALAS/ALAS, the kcat values of the variants varied significantly (4.4 0.2 min-1 vs. 21.6 0.7 min-1) depending on which active site harbored the mutation. Chimeric ALAS/AONS proteins exhibited a di stinct oligomeric state from that of ALAS/ALAS, with two sites having AL AS activity and two AONS activity. Remarkably, while the steady-state rates fo r the ALAS and AONS activities were lower than in the parent enzymes, the reactivity of the ALAS sites in ALAS/AONS was similar to that of wild-type ALAS. We propose that the different contributi on of each active site
91 to the steady-state activity of ALAS/ALAS an d the reduced steady-state activities of the ALAS/AONS chimera, compared to the parent enzymes, relate to different extents of protein conformational changes associated w ith product release due to the strain caused with the linking the two ALAS (or ALAS and AONS) subunits. Introduction Pyridoxal 5Â’-phosphate (PLP) is a necessary cofactor for a cata lytically versatile class of enzymes (Ferreira and Gong, 1995). PLP-dependent enzymes that catalyze reactions involving amino acids share the common mechanism of using the electron withdrawing properties of th e PLP cofactor to labili ze bonds to the substrate -carbon (Christen and Mehta, 2001). PLP-dependent enzymes have been classi fied according to fold types inferred from three-dimensional structures and reac tion specificity (Chris ten and Mehta, 2001; Eliot and Kirsch, 2004). The largest an d most diverse group is the aspartate aminotransferase superfamily or fold-type I, family (Paiardini, Bossa et al., 2004). The -oxoamine synthases constitute a sm all but widespread subfamily within fold-type I PLP-dependent enzymes, comprising 8-amino-7-oxononanoate synthase (AONS), 5aminolevulinate synthase (ALAS), serine palmitoyltransferase (SPT), and 2-amino-3oxobutyrate CoA ligase (Alexeev, Alexeeva et al., 1998; Schneider, Kack et al., 2000; Schmidt, Sivaraman et al., 2001; Astner, Schul ze et al., 2005; Yard, Carter et al., 2007). These four members of the -oxoamine family share about 12% identity at the amino acid level, while the amino acid sequence identity between any pair is around 30% (Alexeev, Alexeeva et al., 1998). The thr ee-dimensional structures for all four -
92 oxoamine synthases are highly cons erved, as demonstrated by the C root-mean-square deviation values for SPT of 1.4-1.6 re lative to the other three members (Yard, Carter et al., 2007). AONS and ALAS catalyze Claisen condensat ions between amino acids and acylCoA thioesters with concomitant decarboxylation of the amino acid leading to the formation of a 2-aminoketone product, CoA and carbon dioxide (Ferreira and Gong, 1995; Alexeev, Alexeeva et al., 1998). AONS catalyzes the r eaction between alanine and pimeloyl-CoA to give 8-amino-7-oxononanoa te (AON), whereas ALAS catalyzes the reaction between glycine and succinyl-CoA to give 5-aminolevulinic acid (ALA) (Scheme 3.1). Both AON and ALA are esse ntial metabolic compounds: the first is an intermediate in biotin synthesis (Ploux and Marquet, 1992) and the second is an intermediate in the tetrapyrrole biosynthe tic pathway (Ferreira and Gong, 1995; Tan and Ferreira, 1996). SCHEME 3.1 Reaction catalyzed by ALAS and AONS ALAS and AONS function as homodimers (Tan and Ferreira, 1996; Alexeev, Alexeeva et al., 1998; Astner, Sc hulze et al., 2005). The activ e site is located at the
93 subunit interface, where the PLP cofactor is covalently bound to an active site lysine through a Schiff base linkage (Ferreira, Neam e et al., 1993; Alexee v, Alexeeva et al., 1998; Astner, Schulze et al., 2005) Each mono mer consists of three domains, a short Nterminal domain (~50 residues), a central catal ytic core (~250 residues), and a C-terminal domain (~100 residues). Although all three do mains participate in dimerization, the catalytic domain contributes the most to th e dimeric interface (Alexeev, Alexeeva et al., 1998; Astner, Schulze et al., 2005; Yard, Carter et al., 2007). Despite the highly conserved tertiary structure of the -oxoamine synthases, the plasticity of the PLP-binding and active site is quite remarkable, as demonstrated with circularly permuted murine erythroid ALAS variants (Che ltsov, Barber et al., 2001; Cheltsov, Guida et al., 2003). Circular perm utation of ALAS, which changed the primary sequence without altering the amino acid co mposition, did not prevent folding of the polypeptide chain into a structure compatib le with binding of th e PLP cofactor and assembly of the two subunits into a func tional enzyme (Cheltsov, Barber et al., 2001; Cheltsov, Guida et al., 2003). In fact, the circ ularly permuted ALAS variants were able to form tertiary structures similar to w ild-type (WT) ALAS that retained enzymatic function in spite of the rea rrangement of the s econdary structural elements (Cheltsov, Barber et al., 2001). Cheltsov et al. (Ferreira and Che ltsov, 2002) argued that a PLP fold and enzymatic activity are achievable as long as the polypeptide chain of ALAS (and presumably of other -oxoamine synthases) folds to a llow binding of the cofactor and correct positioning of the catalytic residues. Structural plasticity in the -oxoamine synthase family has also been substantiated with the creation of a Â“single chain dimericÂ” AL AS (Cheltsov, Barber et al.,
94 2001; Cheltsov, Guida et al., 2003; Zhang, Che ltsov et al., 2005), the discovery of a marine viral single-chain SPT and construction of yeast and mammalian single-chain SPT chimeras (Han, Gable et al., 2006) The Â“single chain, dimericÂ” ALAS ( i.e., ALAS/ALAS), created by linking two AL AS polypeptide chains, exhibited distinct spectroscopic properties and substantially gr eater enzymatic activity than WT ALAS (Zhang, Cheltsov et al., 2005). Si ngle-chain chimeras formed from either the two yeast or mammalian SPT subunits, LCB1 and L CB2, displayed novel s ubstrate specificity (Han, Gable et al., 2006). As a matter of fact similarly to the mari ne viral single-chain SPT and in contrast to the naturally occurr ing eukaryotic, heterodimeric SPTs, the yeast and mammalian single-chain SPT chimeras pref er myristoyl-CoA rather than palmitoylCoA as substrate (Han, Gable et al., 2006). Essentially, these findings indicate that significant flexibility is present in the arch itecture and formation of the PLP-binding and active sites of -oxoamine synthases. To address whether 1) each of the two active sites in the Â“single chain dimericÂ” ALAS contributes equally to the overall enzyma tic activity and 2) the structural plasticity and flexibility in active site formation can be extended to chimeras between different members of the -oxoamine synthase family, we char acterized Â“single chain dimericÂ” ALAS variants, in which one of the two active sites had no m easurable enzymatic activity, and single-chain ch imeras between WT or mutated ALAS and AONS. We report that each of the two active sites in ALAS/ALAS contributes differently to the steady-state activity of the enzyme, even though the amount of ALA produced during the first turnover is identical in both active sites. Further, while the chimeric ALAS/AONS proteins tolerate a high degree of structural plasticity, the oligomeric state and the active
95 site arrangement of the chimeric proteins di ffer from those of proteins generated from the fusion of identical polypeptide chains ( e.g., ALAS/ALAS).
96 Material and Methods Materials The following reagents were purchased fr om Sigma-Aldrich Chemical Company: DEAE-Sephacel, Ultrogel AcA-44, -mercaptoethanol, PLP, succinyl-CoA, ALAhydrochloride, -ketoglutaric acid, -ketoglutarate dehydrogenase, pyruvate dehydrogenase, HEPES-free acid, MOPS, tricine, thiamin pyrophosphate, NAD+, pyruvic acid, the gel filtration molecular weight markers kit (cytochrome c carbonic anhydrase, bovine serum albumin, al cohol dehydrogenase, -amylase), and the bicinchoninic acid protein determination kit. Glycerol, alan ine, glycine, disodium ethylenediamine tetraacetic acid dihydrate, ammonium sulf ate, magnesium chloride hexahydrate, perchloric acid, and potassium hydroxide were acquired from Fisher Scientific. Sodium dodecyl sulfate polyacrylamide ge l electrophoresis reagents were acquired from Bio-Rad. All resctriction enzymes, Vent DNA polymer ase, and T4 DNA ligase were from New England Biolabs. Superdex 200 gel filtrati on resin was from Amersham Biosciences-GE Healthcare and DNA oligonucleotides were from Integrated DNA Technologies. Escherichia coli R872 strain and pET6HAONS were generous gifts from Dr. Dominic Campopiano (University of Edinburg). E. coli HU227 strain was a generous gift from Dr. Charlotte S. Russell (City University of New York). Construction of ALAS/ALAS and ALAS/A ONS chimeric expression plasmids Chimeras between murine, mature erythroid ALAS and E. coli AONS or between mutated forms of either ALAS or AONS a nd the WT enzymes were engineered using pGF23 as the expression vect or (Figure 3.1, Table 3.1). The pGF23 plasmid contains the
97 sequence for murine, mature erythroid ALAS under the control of the alkaline phosphatase promoter (Ferreira and Dailey, 1993). pTDT1, an expression plasmid for 6x-histidine-tagged E. coli AONS, was constructed by PCR-amplification of the E. coli bioF gene, which encodes the AONS protein, and subcloning of the PCR product into pGF23, such that the bioF gene replaced the ALAS-encoding fragment (see Supporting Information for Experimental Details). The pTDT5 and pTDT4 expression plasmids (Table 3.1) were constructed to yield chimeric proteins between ALAS and AONS (F igure 3.1). Using the 5Â’ to 3Â’ convention for the Â“chimeric geneÂ” under the control of the alkaline phosphatase promoter, pTDT5 contains the cDNA coding for AL AS linked to the following bioF gene for AONS through a Mfe I site, whereas in pTDT4, the bioF gene precedes the ALAS cDNA (Figure 3.1). These constructs were based on the pAC1 plasmid (Cheltsov, Barber et al., 2001; Cheltsov, Guida et al., 2003; Zhang, Cheltsov et al., 2005) (Figure 3.1), which contains two tandem ALAS cDNA sequences separated by an Mfe I cloning site (see Supporting Information for Experimental Details). The pTDT12 and pTDT17 expression plasmi ds (Table 3.1) were constructed using the pGF27 expression plasmid for the ALASK313A variant (Ferreira, Vajapey et al., 1995) as the starting material fo r a DNA piece coding for the ALASK313A mutation. The pGF27 plasmid was digested with Kpn I and Xba I and the ALASK313A-encoding fragment was ligated into pTDT4 and pTDT 5 digested with the same enzymes. The pTDT8 plasmid (Table 3.1) was designe d to encode a full-length ALAS with phenylalanine-341 mutated to alanine. To introduce the F341A encoding mutation into
98 ALAS cDNA, the method prev iously described by Gong et al. was followed (Gong, Hunter et al., 1998) (see Supporting Information for Experimental Details). The pTDT14 and pTDT15 expression plasmi ds (Table 3.1) were constructed using pTDT8 as the source plasmid for the ALASF341A-encoding fragment. The pTDT8 plasmid was digested with Kpn I and Xba I and the fragment containing the ALASF341A mutation was isolated and subcloned into pTDT5 and pTDT4 to generate pTDT14 and pTDT15, respectively. The pTDT7 expression plasmid encodes the ALAS/AONS chimera in which the AONS active-site lysine involved in the Schi ff base linkage with the PLP cofactor, K236, is mutated to an alanine (Table 3.1; Figure 3.1). The method described by Gong et al. (11) was used to introduce th e K236A-encoding mutation in the bioF gene harbored in pTDT5 (see Supporting Information for Experimental Details). The pCA1 and pMG1 expression plasmids, encoding ALASK313A/ALAS and ALAS/ALASK313A, respectively, were cons tructed using the pAC1 and pGF27 plasmids (Table 3.1; Figure 3.1) (see Supporting Inform ation for Experimental Details).
99 Table 3.1: Nomenclature defining the plas mids and enzymes described in this report Plasmid Proteina Description Massb Reference pGF23 ALAS ALAS homodimer 56 kD (Ferreira and Dailey, 1993) pTDT1 AONS AONS homodimer 40 kD This study pGF27 ALASK313A ALAS homodimer harboring the K313A mutation 56 kD (Ferreira, Vajapey et al., 1995; Hunter and Ferreira, 1999) pTDT8 ALASF341A ALAS homodimer harboring the F341A mutation 56 kD This study pAC1 ALAS/ALAS Single polypeptide ALAS homodi mer 112 kD (Zhang, Cheltsov et al., 2005) pCA1 ALASK313A /ALAS Chimera of ALASK313A and ALAS 112 kD This study pMG1 ALAS/ALASK313A Chimera of ALAS and ALASK313A 112 kD This study pTDT5 ALAS/AONS Chimera of AL AS and AONS 96 kD This study pTDT4 AONS/ALAS Chimera of AONS and ALAS 96 kD This study pTDT12 ALASK313A /AONS Chimera of ALASK313A and AONS 96 kD This study pTDT17 AONS/ALASK313A Chimera of AONS and ALASK313A 96 kD This study pTDT14 ALASF341A /AONS Chimera of ALASF341A and AONS 96 kD This study pTDT15 AONS/ALASF341A Chimera of AONS and ALASF341A 96 kD This study pTDT7 ALAS /AONSK236A Chimera of ALAS and AONSK236A 96 kD This study aIn the linked proteins the first abbrevia tion refers to the N-terminal enzyme. b Monomeric molecular
100 FIGURE 3.1 Ribbon representation ALAS and AONS homodi mers and schematic representation of expression plasmids (A-B) ALAS and AONS homodimers in ribbon representation with one subunit shown in yellow and the central catalytic core, N-terminal domain, and Cterminal domain of the s econd subunit rendered in dar k, medium and light blue, respectively. (A) ALAS homodimer from R. capsulatus (PDB code: 2BWN). The PLP cofactor ( red ), the active site lysine ( green ) involved in the Schiff base linkage with PLP (K248 in R. capsulatus ALAS and K313 in murine erythroid ALAS) and F276 ( purple ; F341 in murine erythroid ALAS) are depi cted in ball-and-stick representation. (B) AONS homodimer from E. coli (PDB code: 1BS0). The PLP cofactor ( red ) and the active site lysine ( green) involved in the Schiff ba se linkage with PLP (K236) are depicted in balland-stick representation. (C) Schematic representation of the expression plasmids for mutated ALAS, AONS, ALAS/ALAS variants and single-chain chimeras between, ALAS (WT or mutated) and AONS (WT or muta ted). Each of the expression plasmids contains a DNA fragment encoding either a single protei n or a chimeric protein, under the control of the phoA promoter (Ferreira and Dailey, 1993) and possesses an ampicillin resistance selectable marker. (See Â“E xperimental Procedur esÂ” for details). Ampr, ampicillin resistance gene; Bam HI, Mfe I and Sal I, cloning sites.
101 Biological screening for ALAS and AONS function. HemAE. coli strain HU227 can only grow in a medium containing ALA, hemin or when harboring an expr ession plasmid for ALAS or a functional ALAS variant (Sasarman, Surdeanu et al., 1968; Li Brathwaite et al., 1989). BioFE. coli R872 strain can only grow in a medium containing AON, bi otin or when harboring an expression plasmid for AONS or a functional AONS variant (Del Campillo-Campbell, Kayajanian et al., 1967) Competent E. coli HU227 and R872 cells were tr ansformed by electroporation with expression plasmids containing the AL AS and AONS chimeric constructs. To screen for ALAS function, transformed HU 227 cells were plated on Luria-Bertani medium (0.5% yeast extract, 1% tryptone 1.0% NaCl and 1.5% agar) containing 50 mg/ml ampicillin. To screen for AONS func tion, transformed R872 cells were plated on M9 minimal medium containing 50 mg/ml ampici llin. M9 medium contains 1X M9-salts (12.8 g Na2HPO4-7H2O, 3.0 g KH2PO4, 0.5g NaCl, and 1.0g NH4Cl per 1 L ), 2 mM MgSO4, 0.1 mM CaCl2, 0.4 % glucose, 0.1 % vitamin-free casamino acids and 1.5% agar. Purification of ALAS, AONS, ALAS/ALAS, ALASK313A/ALAS, ALAS/ALASK313A and ALAS/AONS The purification of ALAS from E. coli DH5 cells harboring pGF23 was as described by Ferreira and Dailey (Ferreira and Dailey, 1993) Recombinant E. coli AONS and chimeric ALAS/AONS were purified from E. coli DH5 cells harboring pTDT1 and pTDT5, respectively. Tandem ALAS variants ( i.e., ALAS/ALAS, ALASK313A/ALAS and ALAS/ALASK313A) were purified from E. coli strain BL21(DE3) overproducing cells harboring pAC1, pCA1 or pMG1. E. coli cells harboring any of the six expression
102 plasmids were grown in MOPS medium c ontaining ampicillin and harvested as previously described (F erreira and Dailey, 1993) The cell pellet was resuspended in buffer A (20 mM potassium phosphate buffer, pH 7.5, containing 10% glycerol, 1 mM EDTA, 20 mM PLP and 5 mM -mercaptoethanol) with pr otease inhibitors (1 g/ml aprotinin, 1 g/ml leupeptine, 1 g/ml pepstatin and 1 mg/m l PMSF) as previously described (Ferreira and Dailey, 1993). The st eps following cell lysis and centrifugation were essentially as described in (Ferreira and Dailey, 1993) with s light modifications. Specifically, for both AONS and ALAS/AON S, the initial ammonium sulfate fractionation step was 20%. Af ter stirring for 20 min at 4 0C, the solution was centrifuged at 27,000xg for 30 min at 4 oC, and the supernatant was further fractionated with ammonium sulfate to a final concentration of 40%. For the purification of AONS, the protein pellet was resuspended in buffer A and loaded onto an Ultrogel ACA-44 gel filtration column equilibrated with buffer A. The protein solution was adjusted to 20% (w/v ) ammonium sulfate and the subsequent chromatographic steps using Phenyl-Sepha rose and Q-Sepharose anion exchange columns were as previously describe d (Webster, Alexeev et al., 2000). The purification of ALAS/ALAS from bacterial cells harboring the pAC1 expression plasmid (Table 3.1) was accordi ng to a previously published method (Cheltsov, Barber et al., 2001; Cheltsov, Guida et al., 2003; Zhang, Cheltsov et al., 2005). For the tandem ALAS variants, ALASK313A/ALAS and ALASK313A/ALAS encoded by pCA1 and pMG1, respectively (Table 3.1), the protein pellet was resuspended in buffer A and loaded onto a Superdex 200 column equilibrated with buffer A. The fractions containing protein were pooled and loaded ont o a DEAE-sephacel resin equilibrated with
103 buffer A. The resin was washed with buffer A, and the protein was eluted with buffer A containing 70 mM KCl. For the purificati on of ALAS/AONS, the protein pellet obtained after 40% ammonium sulfate fractionation was resuspended in buffer A, pH 7.9, and loaded onto an Ultrogel ACA-44 column equilibra ted with the same buffer. The fractions containing protein were pooled and loaded ont o a Q-Sepharose column equilibrated with buffer A, pH 7.9. The Q-Sepharose resin wa s washed with buffer A, pH 7.9, containing 25 mM KCl, until Abs280 of the Â“washed proteinsÂ” was lower than 0.1; ALAS/AONS was eluted from the Q-Sepharose resin with a 100 mM to 150 mM KCl gradient in buffer A, pH 7.5. The protein purity was assessed us ing SDS-PAGE. When the final product did not meet our purity criteria ( i.e., over 95% homogeneity as judged by SDS-PAGE), size exclusion chromatography with Sephadex 200 resin was used to eliminate protein contaminants. Protein-containing fractions we re pooled and concentrated in an Amicon 8050 stirred cell with an YM30 membrane. Th e purified and concentrated enzyme (WT ALAS, AONS, ALAS/ALAS, ALASK313A/ALAS, ALAS/ALASK313A or ALAS/AONS) was stored in liquid nitrogen. Protein concentrations were determined with the bicinchoninic acid method using bovine serum albumin as the standard (Smith, Krohn et al., 1985). ALAS/AONS concentrations are reported on the basis of a subunit molecular mass of 96 kD, while ALAS/ALASK313A and ALASK313A/ALAS concentrations are re ported on the basis of a subunit molecular mass of 112 kD. (ALAS and AONS have monomeric molecular masses of 56 kD and 40 kD, respectively (F erreira and Dailey, 1993; Alexeev, Alexeeva et al., 1998)).
104 Molecular mass determination by gel filtration chromatography. The native molecular mass of ALAS/AON S was determined using gel filtration chromatography as previously described by Cheltsov et al. (Cheltsov, Barber et al., 2001). Pimeloyl-CoA synthesis. Pimeloyl-CoA was synthesi zed as previously descri bed by Ploux and Marquet (Ploux and Marquet, 1992). Purity was assessed by reverse-phase HPLC and concentration was determined by measuri ng the absorbance at 260 nm and using an 260nm = 16800 M-1cm-1. Spectroscopic measurements. Absorption spectra were acquired at am bient temperature using a Shimadzu UV 2100 dual beam spectrophotometer, with a refe rence containing all components except the purified enzyme. Fluorescence spectra were collected on a Shimadzu RE-5301 PC spectrofluorophotometer using pr otein concentrations of 2-4 M. Fluorescence blank spectra were collected from samples containing all components except enzyme immediately prior to the measurement of sa mples containing enzyme. The blank spectra were subtracted from the spectra of samples containing enzyme. Steady-state kinetic characte rization of ALAS/AONS, ALASK313A/ALAS, and ALAS/ALASK313A. ALAS steady-state activity of the ALAS/AONS chimera, ALASK313A/ALAS and ALAS/ALASK313A was determined at 30C using a continuous spectrophotometric assay as described previously for WT ALAS (H unter and Ferreira, 1995). AONS steady-state activity of the of the ALAS/AON S chimera was determined at 30oC using a coupled
105 enzymatic spectrophotometric assay for AONS (Webster, Campopiano et al., 1998), which is similar to that developed for determination of ALAS activity (Hunter and Ferreira, 1995). Briefly, in the latter assay, -ketoglutarate dehydrogenase was replaced by pyruvate dehydrogenase as the coupling enzy me, and the reactions contained 20 mM HEPES, pH 7.5, 3 mM MgCl2, 1 mM pyruvic acid, 1 mM NAD+ and 0.25 to 1 M enzyme. Data were acquired using a Shim adzu UV 2100 dual-beam spectrophotometer. Enzymatic activity data were plotted vs. substrate concentration in which one of the substrate concentrations varied, while the s econd was kept constant. The steady-state kinetic parameters ( i.e.,Gly mK, SCoA mK, and kcat of the ALAS/AONS chimera, ALASK313A/ALAS and ALAS/ALASK313A andAlanine mK, PCoA mK, and kcat of the ALAS/AONS chimera) were determined by fitting the data to the Michaelis-Menten equation using non-linear regre ssion analysis software. Rapid chemical quenched-flow exp eriments and data analysis. Rapid chemical quenched-flow experime nts were performed using a SFM-400/Q mode quenched-flow apparatus (BioLogic Scie nce Instruments, France), equipped with a circulating water bath to control the temperat ure of the reactants as described in Zhang and Ferreira (Zhang and Ferreira, 2002). ALA concentration in the quenched samples was also determined as previously desc ribed (Zhang and Ferreira, 2002). ALA produced at different reaction times were plotted against time and fitted to equation 1 (Yard, Carter et al., 2007), using the nonlinear least-square s regression analysis program SigmaPlot, where Pt represents the product concentration at an aging time t, A is the amplitude of the
106 burst phase, kb is the burst rate constant, kss is the steady-state rate constant, and E0 is the total enzyme concentration (Zhang and Ferreira, 2002). Pt A(1 e kbt) kssE0t Equation 1 Results In vivo activity screen of ALASK313A/ALAS and ALAS/ALASK313A. To rapidly verify if ALASK313A/ALAS and ALAS/ALASK313A, which harbor the K313A mutation in either of the two linked ALAS subunits, exhibited ALAS activity, positive genetic complementation of hemAE. coli HU227 cells was employed. Since HU227 cells can only grow in a medium supplem ented with ALA (Li, Brathwaite et al., 1989) or when transformed with functi onal ALAS expression plasmids (Gong and Ferreira, 1995; Tan and Fe rreira, 1996) the rescue of HU227 cells in a non-ALA supplemented medium indicates the production of a variant with ALAS activity. Indeed, while HU227 cells harboring the ALAS ho modimer or the single chain dimeric ALAS/ALAS can grow in a medium without ALA (Gong and Ferreira, 1995; Tan and Ferreira, 1996), HU227 cells overproduci ng the K313A homodimer cannot (Tan and Ferreira, 1996) (Table 3.2). Both ALASK313A/ALAS and ALAS/ALASK313A variants retained function as indicated by the ability of transformed HU227 cells to grow when harboring either va riant (Table 3.2).
107Spectroscopic characterization of ALASK313A/ALAS and ALAS/ALASK313A. Although, as previously reported (Zhang, Cheltsov et al., 2005), the UV/visible absorbance spectra for both ALAS and the single chain dimeric ALAS/ALAS exhibited the characteristic maxima for PLP-dependent enzymes at ~330 and ~420 nm, the prominence of these absorbance bands varied between the spectra of the two proteins (Figure 3.2). The absorbance band at ~420 nm was more prominent in the spectrum of ALAS/ALAS than that of ALAS, whereas the absorbance band at 330 nm was less distinct in ALAS/ALAS. Indeed, at pH 7.5, the ratio of the 420 nm to 330 nm absorbance changed from 0.44 for ALAS to 0.67 for ALAS/ALAS (Figure 3.2). Similarly, the UV/visible absorbance spectra of the ALASK313A/ALAS and ALAS/ALASK313A variants retained th e characteristic maxima at ~330 and ~420 nm while the prominence of the two absorbance bands varied, resulting in a 420 to 330 nm absorption ratio of 0.35 and 0.58 at pH 7.5 for ALASK313A/ALAS and ALAS/ALASK313A, respectively. The ~330 and ~420 nm maxima were previously a ttributed to the substituted aldimine and ketoenamine forms of the in ternal aldimine between PLP and the -amino group of K313, respectively (Zhang, Cheltsov et al., 2005). Clearly, the mutation of the lysine residue in either of the two active sites of ALAS/ALAS has unique affects on the environment of the PLP cofactor of each site. These results indi cate that each active site makes distinctive contributions to the UV/visible spectrum of the single chain dimeric ALAS/ALAS and that the linking of th e two subunits of WT ALAS produces asymmetrical cofactor-binding sites in the single chain dimeric ALAS/ALAS.
108 Table 3.2: Growth of transformed E. coli strains on selective media E. coli HU227 E. coli R872 Plasmid Protein LB/Amp/ALA LB/Amp LB/Amp Biotinminimal pGF23 ALAS + + + pTDT1 AONS + + + pGF27 ALASK313A + + pTDT8 ALASF341A + + pAC1 ALAS/ALAS + + + pCA1 ALASK313A /ALAS + + + pMG1 ALAS/ALASK313A + + + pTDT5 ALAS/AONS + + + + pTDT4 AONS/ALAS + + + + pTDT12 ALASK313A /AONS + + + pTDT17 AONS/ALASK313A + + + pTDT14 ALASF341A /AONS + + + pTDT15 AONS/ALASF341A + + + pTDT7 ALAS /AONSK236A + + +
109 Wavelength (nm) 300350400450500 Absorbance 0.0 0.3 0.6 0.9 Wavelength (nm) 300350400450500 Absorbance 0.0 0.1 0.2 FIGURE 3.2 Absorption spectra of ALAS, ALAS/ALAS, ALASK313A/ALAS and ALAS/ALASK313A. The inset includes th e region from 300Â–500 nm. Protein concentrations were adjusted to 30 M for ALAS/ALAS, ALASK313A/ALAS and ALAS/ALASK313A or 20 M for ALAS in 20 mM Hepes, pH 7.5. ALAS/ALASK313A ( Â—) ALASK313A/ALAS ( ), ALAS/ALAS( -), and ALAS (Â– Â–).
110Kinetic characterization of the ALASK313A/ALAS and ALAS/ALASK313A variants and determination of the dissociation constants for the binding of ALA. The steady-state kinetics of the ALASK313A/ALAS and the ALAS/ALASK313A reactions were examined, and the results are presented in Table 3.3. Previously, it was reported that linking the two ALAS subunits into the single chai n dimeric ALAS/ALAS resulted in over 5-fold and 28-fold increases in the kcat and SCoA catm/ kK values, respectively (Zhang, Cheltsov et al., 2005). In addition, mu tagenesis of K313, the active site residue involved in the Schiff base linkage with the PLP cofactor (Ferreira, Neame et al., 1993) and in catalysis (Hunter and Ferreira, 1999), to alanine rendered the Â“naturalÂ” ALAS homodimer into a variant of the enzyme (K 313A) with no measurable ALAS activity (Ferreira, Vajapey et al., 1995; Hunter and Ferreira, 1999). To determine the individual contribution of the two active sites to the overall steady-state activity of ALAS/ALAS, the steady-state kinetic parameters of the ALASK313A/ALAS and ALAS/ALASK313A variants, in which the K313A mutation was independently introduced in each of the two active sites, were determined (Table 3.3). While the K313A mutation in ALASK313A/ALAS decreased the kcat 2.5-fold, the same mutation in ALAS/ALASK313A resulted in a 12.6-fold decrease of the kcat value. Similarly, the catalytic efficiencies of ALASK313A/ALAS for glycine and succinyl-CoA were decreased approximately 2.9and 1.8-fold, respectively, whereas those of ALAS/ALASK313A were reduced approximately 9and 51-fold for glycine and succinyl-CoA, re spectively. If both active sites were to contribute equally to the steady-state activity of ALAS/ALAS, then a 50% reduction of the turnover number would be expected for the single chain dimer in which the enzymatic activity of one of the two active sites was impaired. This seems to be the case for
111 ALASK313A/ALAS but not for the ALAS/ALASK313A variant. These findings suggest that linking the N-terminus of one subunit to th e Cterminus of the other subunit in ALAS/ALAS may have created strain that hi ndered the steady-state enzymatic activity of one of the active sites, and, consequently, produced a single chain dimer with unequal participation of the two active sites. Table 3.3: Summary of steady-state kinetic parameters for ALAS, ALAS/ALAS, ALAS/ALASK313A and ALASK313A/ALAS catk (min-1) Gly mK (mM) Gly catm/ kK (min-1mM-1)SCoA mK ( M) SCoA catm/ kK (min-1 M-1) ALA dK ( M) ALAS 10 1a 23 1a0.43 0.06a2.3 0.1a4.3 0.2a 25 3a ALAS/ALAS 55.4 16.7 0.2b3.32 0.04b0.45 0.03b123 8b 1.46 0.01 ALAS/ALASK313A 4.4 0.2 11.8 2.80.37 0.091.8 0.3 2.4 0.4 6.8 1.1 ALASK313A/ALAS 21.6 0.7 14.8 11.71.46 0.170.32 0.05 67.5 10.8 2.3 0.8 a Data from ref (Gong, Hunter et al., 1998). b Data from ref (Zhang, Cheltsov et al., 2005).
112 Pre-steady-state burst experiments for ALASK313A/ALAS and ALAS/ALASK313A. Similarly to the ALAS-catalyzed reaction, the ra te-limiting step for the single chain dimeric ALAS occurs after catalysis and is proposed to be associated with a protein conformational change associated with ALA re lease (Zhang, Cheltsov et al., 2005). To determine if the rate-limiting step of the reactions catalyzed by the K313A single chain dimeric variants occurs afte r the chemical step and to compare the amount of ALA produced in the first turnover of the ALASK313A/ALAS and ALAS/ALASK313A reactions with that generated in the ALAS/ALAS first turnover, pre-steady-st ate burst experiments using chemical quenched-flow were perf ormed (Figure 3.3). The reaction of ALASK313A/ALAS with saturating glycine and succ inyl-CoA occurred with a burst in ALA production at a rate of 30.6 4.2 s-1, whereas the first turnover of the ALAS/ALASK313A reaction was at a rate of 45.1 6.1 s-1. These burst rates are similar to the burst rate previously reported fo r the ALAS/ALAS reaction, 48.6 6.1 s-1 (Zhang, Cheltsov et al., 2005). Anothe r important piece of inform ation provided by the burst experiments shown in Figure 3.3 relates to the burst amplitudes, representing the concentration of ALAS active sites for th e reactions. The burst amplitudes were 0.25/active site and 0.21/active site for ALASK313A/ALAS and ALAS/ALASK313A, respectively, nearly 50% less than the bur st amplitude previous ly determined for ALAS/ALAS, 0.49/active site (Z hang, Cheltsov et al., 2005). These results are consistent with ALASK313A/ALAS and ALAS/ALASK313A possessing half-sites of the ALAS/ALAS reactivity. Further, given that the number of the active sites of either ALASK313A/ALAS or ALAS/ALASK313A is roughly 50% of that of ALAS /ALAS, the results also indicate
113 that the different contributions of each of the two active sites of ALAS/ALAS to its steady-state activity emanate from a step occurring after chemistry.
114 FIGURE 3.3 Kinetics of a pre-steady-state burst of ALA product in the ALAS/ALASK313A and ALASK313A/ALAS reactions. ALAS/ALASK313A (square) or ALASK313A/ALAS (triangle) (15 M) preincubated with glyc ine (200 mM) was quickly reacted with succinyl-CoA (150 M) at 20C. The concentrations shown in parentheses are final concentrations after mixing. The reactions were quenched w ith 0.14 M perchloric ac id at various aging times, and the ALA concentration was determined. The inset illustrates the first 0.3 seconds of the the reaction time course. Th e curves represents the best fits to equation 1 with a burst amplitude of 6.27 0.36 M and a burst rate of 45.1 6.1 s 1 for ALAS/ALASK313A and a burst amplitude of 7.50 0.31 M and a burst rate of 30.6 4.2 s 1 for ALASK313A/ALAS.
115 Biological screening for ALAS and AONS function. The determination of the crystallographic structures of ALAS (Astner, Schulze et al., 2005) and AONS (Alexeev, Alexeeva et al., 1998) confirmed the prediction drawn from the high degree of sequence similarity between ALAS and AONS (37%) that these two -oxoamine synthases have similar 3D fold and active site architecture. To extend our studies on the plasticity of the PLP-bi nding and active site of ALAS to other members of the -oxoamine synthase family, chim eras of ALAS and AONS and chimeras of singly-mutated ALAS variants and AONS were constructed (Table 3.1 and Figure 3.1). The two major objectives were 1) to determine whether the chimeras retain ALAS and AONS activities and 2) to examine the active site arrangement in the ALAS/AONS chimera. Both ALAS/AONS and AONS/ALAS exhibited ALAS and AONS activities as assessed by the positive genetic complementation of hemAHU227 and bioFR872 cells, which are ALA/heme (S asarman, Surdeanu et al., 1968; Li, Brathwaite et al., 1989) and biotin auxotrophes (Del Camp illo-Campbell, Kayajanian et al., 1967), respectively (Table 3.2). From these in vivo activity assays, ALAS/AONS and AONS/ALAS appear to be bifunctional enzymes ( i.e. ALAS and AONS activities). To start addressing the question related to the active site arrangement of ALAS/AONS chimera, directed-mutagenesis of critical residues in ALAS and AONS had to be established. K236 of E. coli AONS, the conserved active site lysine involved in PLP binding and catalysis (Webster Campopiano et al., 1998), corre sponds to K313 of murine ALAS (Ferreira, Vajapey et al ., 1995; Hunter an d Ferreira, 1999). T hus, when the K236A mutation was introduced in the AONS homodimer, bioFR872 cells did not grow in a medium without biotin (dat a not shown), similar to th e absence of growth of hemA-
116 HU227 cells overproducing ALASK313A when plated onto a medi um lacking ALA (Table 3.2). The crystal structure of the R. capsulatus ALAS homodimer revealed that amino acids of the two polypep tide chains contribute to the same active site, which is located at the dimer interface (Astner, Schulze et al., 2005). For example, F276 ( R. capsulatus ALAS numbering or F341 in the equivalent murine er ythroid ALAS numbering), a phenylalanine crucial for interaction with the pantetheine moiety of CoA, and the lysine involved in the Schiff base linka ge participate in the architecture of the same active site but reside in different polypeptide chains. Of significance to this study, F341 is critical to ALAS function, as the F341A mutation abolished the pro duction of ALA necessary to sustain growth of HU227 cells on a selective medium (Table 3.2). Thus, while K313 and F341 are critical for function of the muri ne erythroid ALAS homodimer, of the corresponding K236 and Y264 (Alexeev, Alex eeva et al., 1998), only K236 is essential for function in the E. coli AONS homodimer. The three possible arrangements for the act ive sites of ALAS/A ONS chimera are 1) formation of two active sites with the cont ribution of amino acids from one ALAS and one AONS polypeptide chains to each active site (Figure 3.4B) in an analogous arrangement to that of single chain dimeric ALAS/ALAS (Figure 3.4A ); 2) formation of four active sites with the contribution of amino acids from one ALAS and one AONS polypeptide chains to each active site (Figure 3.4C); 3) forma tion of four active sites with two of the active sites being formed with only ALAS amino acids and the other two of the active sites with only AONS amino acids (F igure 3.4D), such that they represent WT ALASand AONS-like active sites. While the first active site arrangement would result from one single, chimeric polypeptide chai n (Figure 3.4B), the latter two active site
117 arrangements would arise from two chimeric polypeptide chains (Figures 3.4C and 3.4D). The determined molecular mass of ~182 KDa for the chimeric ALAS/AONS (see below and Figure 3.5) ruled against a single chai n, dimeric ALAS/AONS and the active site arrangement depicted in Figure 3.4B. To di stinguish between the other two possibilities for the active site arrangement of the ALAS/ AONS chimera (Figures 3.4C and 3.4D), an experimental approach involving the use of specific amino acid mutations targeted to abolish either ALAS or AONS functi on and biological selection systems ( E. coli HU227 and E. coli R872) was followed. If the active si te arrangement in Figure 3.4C were correct the K313A mutation w ould eliminate ALAS function in two of the four active sites, and the AONS residues in one or two of the remaining sites would have to complement the ALAS residues, yielding a chimeric protein with ALAS and AONS activities. In contrast, with the active site arrangement presented in Figure 4D, the ALAS and AONS activities arise from WT ALASand AONS-like active sites, and thus a deleterious mutation of a critical active site residue in ALAS or AONS would produce a chimeric enzyme with impair ed ALAS or AONS activity. HU227 cells transformed with either pTDT12 or pTDT17 [ i.e., expression plasmids for the ALAS/AONS chimera harboring the K313A mutation in ALAS ( i.e., Table 3.1)] did not support the growth of these cells in an ALA-depleted medium (T able 3.2). However, R872 cells harboring either of these two plasmids could grow in a medium without biotin. A similar situation was observed with the chimeras ALASF341A/AONS and AONS/ALASF341A, in which the phenylalanine at position 341 of murine erythr oid ALAS was substituted with an alanine (Table 3.2). When the ALAS sequence was maintained intact and a mutation of the Schiff base linkage-lysine residue was introduced into AONS ( i.e., K236A), the
118 generated chimeric proteins, AONSK236A/ALAS or ALAS/AONSK236A, could rescue the growth of HU227 cells in a me dium without ALA but not of R872 cells in a non-biotin supplemented medium (Table 3.2). Taken toge ther, these findings are consistent with the active site arrangement for ALAS/AONS depicted in Figure 3.4D, in which the active sites responsible for ALAS activity are built with ALAS residues, while those with AONS activity are made of AONS residues. Oligomeric state of the ALAS/AONS chimera. Although ALAS/AONS and AONS/ALAS coul d be overproduced as active chimeric enzymes in E. coli DH5 cells (data not shown), only ALAS/AONS could be purified. Poor solubility and low stabili ty were among the factors preventing the purification of AONS/ALAS. The molecular mass of each subunit of the WT ALAS and WT AONS homodimers is ~56 and ~40 kD, respectively. Thus a single chim eric subunit, derived from the fusion of an ALAS and AONS polypeptide, is expected to be ~96 kD. This is in agreement with the molecular mass estimated by SDS-PAGE (Figure 3.5, inset). The molecular mass of the native ALAS/AONS chimera was determined to be ~182 kD (Figure 3.5), consistent of a Â“homodimerÂ” of ~96 kD subunits. Further, coupled enzyme assays confirmed that the 182-kD protein exhibits bot h ALAS and AONS activities, indicating that the ALAS/AONS chimera is bi functional as a homodimer of two ~96 kD chimeric polypeptide subunits.
119 FIGURE 3.4 Schematic representation illustrating the act ive site active site arrangement for ALAS/ALAS and possible active site arrangements for ALAS/AONS. Red bars represent ALAS polypeptide chains, with th e essential active si te residues K313 and F341 from one single polypeptid e chain contributing to separate active sites. Green bars represent AONS polypeptide chains, with the essential active s ite residues K236 and Y264 from one single polypeptide chain contributing to separate active sites. In (A) and (B), there are two active sites per single chain dimer, whereas in (C) and (D), there are four active sites per chimeric ALAS/AONS dimer. (A) Active site arrangement of the ALAS/ALAS Â“single chain dimerÂ”. (B) (D) Possible active site arrangements for ALAS/AONS. (B) ALAS/AONS single chain dimer. (C) ALAS/AONS chimeric dimer with Â“hybridÂ” active sites comprised of both ALAS and AONS residues. (D) ALAS/AONS chimeric dimer with two of th e active sites containing only ALAS residues and the other two active sites containing only AONS residues.
120 Spectroscopic characterizati on of AONS and ALAS/AONS. At pH 7.5, absorbance maxima at ~ 330 and ~420 nm were also observed in the ALAS/AONS chimera (Figure 6A). Because the UV-visible spectra of both ALAS and AONS exhibited maxima at 330 nm and 4 20 nm, their fluorescence spectra were examined for distinctive features among ALAS, AONS and AL AS/AONS. With excitation at 420 nm, the PLP cofactor of eith er ALAS or AONS e xhibited a fluorescence emission maximum at 510 nm for ALAS and AONS (Figure 6B), albeit the magnitude of the 510 nm fluorescence emission maximum was more than three times greater for the AONS cofactor than that of ALAS. Sim ilar to AONS and ALAS, the PLP cofactor of the ALAS/AONS chimera exhibited a fluorescence emission maximum at 510 nm upon excitation at 420 nm, and the magnitude of this fluorescence emission maximum fell between the values observed for the PLP cofa ctor of ALAS and AONS (Figure 6B). Upon excitation at 330 nm, the fluorescen ce emission spectra of ALAS and AONS exhibited maxima at 385 nm and 430 nm, re spectively, while the fluorescence emission spectrum of the ALAS/AONS chimera displa yed a broad emission band between 385 nm and 430 nm (Figure 6C).
121 FIGURE 3.5 Determination of the molecular mass of the ALAS/AONS chimera by gel filtration chromatography. ALAS/AONS (5 mg) wa s applied to a Pharmacia Sephadex 200 filtration column and eluted with 20 mM potassium phosphate buffer containing 10% glycerol at 4 0C and pH 7.5 (flow rate 1.0 mL/min). The molecular mass calibration curve for the Superdex 200 column using cytochrome c (12.4 kDa), carbonic anhydrase (29.0 kDa), ovalbumin (45 kDa), bovine serum albumin (66.0 kDa), alcohol dehydrogenase (150.0 kDa), and -amylase (200.0 kDa) as pr otein standards (indicated by circles). The ALAS/AONS chimer a is indicated by a triangle. ( Inset ) 12.5% SDSPAGE of the purified ALAS, AONS and ALAS/AONS chimera, which as were detected using Coomassie Brilliant Blue staining. Approximately 5 g of each protein sample was loaded per lane.
122 Wavelength (nm) 450475500525550 Relative Fluorescence 0 25 50 75 100 125 Wavelength (nm) 350400450500550 Relative Fluorescence 0 300 600 900 1200 Wavelengnth (nm) 300325350 Relative Fluorescence 0 200 400 Wavelength (nm) 375425475 Relative Fluorescence 0 50 100 Wavelength (nm) 250300350400450500 Absorbance 000 025 050 075 Wavelength (nm) 300 400 500 Absorbance 000 005 010 015 FIGURE 3.6 Absorption and fluorescence spectra of ALAS, AONS and ALAS/AONS. (A) UVvisible absorption spectra. The inse t includes the region from 300 nm. Fluorescence emission spectra upon excitation at (B) 420 nm and (C) 330 nm and fluorescence excitation spectra upon emission at 510 nm ( inset B ) and 385 nm ( inset C ). For absorption spectra, protein concentrations were adjusted to 15 M (AONS and ALAS) or 7.5 M (ALAS/AONS) and for fluorescence spectra, protein concentrations were adjusted to 4 M (AONS and ALAS) or 2 M (ALAS/AONS) in 20 mM Hepes, pH 7.5 containing 10% glyc erol. For (A) (C ), AONS (--), ALAS (), and ALAS/AONS ( ).
123 Steady-state and transient kinetics of ALAS/AONS. To examine the ALAS and AONS activit ies of the ALAS/AONS chimera, the steady-state kinetic parameters of the chimer ic protein associated with both activities were determined using substrates for ALAS and AONS (Tables 3.4 and 3.5). Regarding the ALAS activity, the kcat decreased almost 40%, the catal ytic efficiency for glycine increased ~1.4-fold, and the catalytic effici ency for succinyl-CoA remained virtually the same relative to ALAS (Table 3.4). With respect to the AONS activity, while the value for kcat decreased approximately 50%, the cataly tic efficiencies towards alanine and pimeloyl-CoA of the ALAS/AONS chimera were similar to those of AONS (Table 3.5). Using chemical quenched-flow, pre-st eady-state experiments of the ALAS reaction were performed to examine the extent of ALA production by ALAS/AONS and ascertain the reactivity of the ALAS activ e sites in the ALAS/AONS chimera (Figure 3.7). The time course associated with AL A formation was biphasic and is described by a burst phase with a ra te of 13.2 2.6 s-1 and a steady-state rate of 0.015 s-1. While the values for the burst and the steady-state rate s were approximately 70% and 50% of those previously determined for the ALAS-cataly zed reaction (Zhang, Ch eltsov et al., 2005), the burst amplitude of 0.10/active site was similar to that formerly observed with ALAS (Zhang, Cheltsov et al., 2005). Significantly, the ~50% decrease in the value of the steady-state rate agrees with that inferred from the experiments performed under steadystate conditions (above). T hus, the reactivity of the ALAS sites in ALAS/AONS is similar to that of ALAS, and the diminished steady-state ALAS activity of the ALAS/AONS chimera in relation to that of AL AS must arise from a step occurring after
124 the reaction chemistry, presumably a strained protein conformational change associated with product release.
125 Table 3.4: ALAS activity: Summary of st eady-state kinetic parameters for ALAS and ALAS/AONS chimera catk (min-1) Gly mK (mM) Gly catm/ kK (min-1mM-1) SCoA mK ( M) SCoA catm/ kK (min-1 M-1) ALASa 10 1 23 10.43 0.022.3 0.1 4.3 0.2 ALAS/AONS 6.2 0.8 5.5 0.81.1 0.21.5 0.2 4.1 0.7a Data from ref (Gong, Hunter et al., 1998) Table 3.5: AONS activity: Summary of stea dy-state kinetic parameters for AONS and ALAS/AONS chimera catk (min-1) Alanine mK (mM) Alanine catm/ kK (min-1mM-1) PCoA mK ( M) PCoA catm/ kK (min-1 M-1) AONSa 3.6 0.6 0.5 0.4 7.2 1.3 25 2 0.14 0.03 ALAS/AONS 1.7 0.3 0.25 0.05 6.8 1.8 10 1 0.16 0.03 a Data from ref (Webster, Alexeev et al., 2000)
126 Figure 3.7 Kinetics of a pre-steady-st ate burst of ALA product in the ALAS/AONS reaction. ALAS/AONS (50 M) preincubated with glycine ( 140 mM) was quickl y reacted with succinyl-CoA (150 M) at 20C. The concentrations shown in parentheses are final concentrations after mixing. The reactions were quenched with perc hloric acid (0.14 M) at various aging times, and the ALA concentration was determined. The inset illustrates the reaction time course, while the first 0.7 s econds of the reaction ar e shown in a larger plot. The first 0.7 s of the time course is expanded while in the inset the time course is extended to 3.0 s. The curve represents the best fit to equation 1 with a burst amplitude of 4.7 0.4 M and a burst rate of 13.2 2.6 s 1.
127Discussion ALAS and other fold-type I PLP-depende nt enzymes function as homodimers with two active sites pe r dimer; each active site is created at the interface between the two monomeric subunits (Christen and Mehta, 2001; Eliot and Kirsch, 2004). The crystal structure of the R. capsulatus ALAS holoenzyme revealed that the enzyme symmetrically binds two PLP molecules, one at each activ e site (Astner, Schulze et al., 2005). Previously, we demonstrated that linking the two subunits of ALAS into a single polypeptide chain yielded a more -active enzyme that functioned as a single-chain dimer (Zhang, Cheltsov et al., 2005). Although linking the N-terminus of one subunit and the C-terminus of the adjacent subunit without the introduction of a polypeptide linker did not affect the global conformation, changes in the environment of the PLP cofactor altered the predominant tautomeric form of the internal aldimine, which contributed to the greater activity of the si ngle-chain dimer (Zhang, Chelts ov et al., 2005). However, it was not clear if these changes affected the enzymatic activity of each active site to a similar extent. To determine whether the two active sites in ALAS/ALAS contribute equally to enzymatic activity, we characteri zed variants in which one of the two active sites had no measurable en zymatic activity due to a mu tation of the conserved K313 residue that binds to the cofactor. Spectral characterizatio n of and ALAS/ALASK313A and ALASK313A/ALAS revealed asymmetric cofactor environments in the two active sites, which was reflected in the disproportionate kinetic behavior of the two sites. In contrast to ALAS/ALAS and ALASK313A/ALAS, in which the steady-state rates increased 5-fold and 2-fold, respectively, relative to the values for WT ALAS, the steady-state rate decreased
128 approximately 50% for ALAS/ALASK313A. The concentration of catalytically active sites in ALAS/ALASK313A and ALASK313A/ALAS was half that of the ALAS/ALAS but, unlike the variations observed in the steady-state rates, the chemical rates were similar in all three forms. As with ALAS/ALAS and WT ALAS (Zhang and Ferreira, 2002; Zhang, Cheltsov et al., 2005), the pre-st eady-state burst results for the two K313A va riants were consistent with the rate-limiting step occurring after the reaction chemistry. In ALAS, the rate-limiting step has been ascribed to the opening of an active site loop that allows ALA release; the rate of this assigned loop ope ning closely corresponds to the steady-state rate (Zhang and Ferreira, 2002; Hunter, Zhang et al., 2007). Thus, the variations observed in the steady-state kinetic parameters are likely to be due to alterations in the energy barrier for this c onformational change required for ALA release. The ALAS crystal structure i ndicates that the N-terminus of one subunit and the Cterminus of the adjacent s ubunit are located near the su rface on opposite faces of the holoenzyme (Astner, Schulze et al., 2005). The strain resulting from linking the remote Nand C-termini of two ALAS subunits in ALAS/ALAS appears to increase the barrier for product release at one site while decreasing the barrier at the other; that is, the steadystate enzymatic activity is enhanced at one active site and hindered at the other. Consequently, the active sites contribute asymmetrically to enzyme function. Generally, the active sites in fold type I PLP-dependent enzymes are equivalent and independent (Eliot and Kirsch, 2004), alt hough some examples of kinetic asymmetry have been documented. In glutam ate-1-semialdehyde aminomutase (GSAM), allosteric interactions between active s ites lead to inactivation of the site in one subunit by the activation of the site in the other subunit (Stetefeld Jenny et al., 2006). In aspartate
129aminotransferase (AAT), dissimilar lattice co ntact in the crystalline enzyme contributes to kinetic asymmetry in the active site s, although the active sites display kinetic equivalence in solution (Kirsten, Gehring et al., 1983). In addition, heterodimeric variants created for a number of compleme ntation studies have contained asymmetric active sites (Onuffer and Kirs ch, 1994; Tan and Ferreira, 1996; Tan, Harrison et al., 1998; Tarun and Theologis, 1998), but the crea tion of ALAS/ALAS is the first example of engineered active site asym metry without the in troduction of active site mutations in fold-type I PLP-dependent enzymes. Despite the unequal steadystate rates in the functioning active sites of ALAS/ALASK313A and ALASK313A/ALAS, we would expect the total activity of the two variants to be similar to ALAS/ALAS, in which both sites contribute to enzyme function. Clearly, this was not the case. The tota l steady-state activities for the two K313A variants was only about 50% of the activity observed in the non-mutated single-chain dimer, suggesting that the enzymatic activity of at least one active site is impaired by the elimination of the Schiff base linkage betw een K313 and the PLP cofactor at the adjacent site. Conversely, studies in the unlinked ALASK313A and ALASR149A heterodimer, containing one nonactive site with both K313A and R149A mutations and one WT catalytically active site, show that the varian t heterodimer retained approximately 50% of the activity observed in the AL AS homodimer (Tan and Ferr eira, 1996); that is, the WT active site was not significantly affected by mu tations at the adjacent site. Thus, linking the termini of the two ALAS subunits likely altered the intermolecular dynamics such that movement at one site can be transmitted to the adjacent site.
130 Because the single-chain ALAS dimer s howed structural pl asticity and had increased activity, we wondered whether the st ructural plasticity would extend to singlechain chimeras constructed from two members of the -oxoamine synthase family, ALAS and AONS. We also hypothe sized that it might be possi ble to generate an enzyme with novel activity by creating hybr id ALAS/AONS active sites. In vivo assays indicated that the ALAS/AONS and AONS/ALAS chimeras possessed both ALAS and AONS activities. Thus, the chimeric protein had su fficient structural plasticity to achieve the conformations necessary to produce both enzyma tic activities. The linking of two ALAS subunits did not significantly affect the dimeri c interface or the foldi ng of the core in the individual subunits, and the dissociation of the dimeric interf ace of ALAS and the ALAS/ALAS single-chain dimer exhibited similar free energi es and resulted in stable intermediates that retained a substan tial amount of their secondary and tertiary structure (Cheltsov, Barber et al., 2001; Cheltsov, Guida et al., 2003). Similarly, we would not expect the linking of ALAS and AONS to stro ngly affect the folding core or subunit interface of the two doma ins, and it is not surprising th at the two domains retained their overall structural character given that th e entire ALAS and AONS polypeptides were used in the creation of the chimeras. Despite our initial hypothesis that the chimeric protein would create chimeric active sites with potentially novel enzy matic activities, both ALAS/AONS and AONS/ALAS appeared to function as ch imeric homodimers with functionally independent ALAS a nd AONS modules. In these modul es, two active sites exhibiting ALAS activity were built exclusively with ALAS residues, and two active sites with AONS activity were built with AONS residues (Figure 3.4D). Nonetheless, the
131 dimerization of two chimeric polypeptides into a bifunctiona l homodimer with functionally independent ALAS and AONS modules suggests that the structural plasticity observed in ALAS can be extend ed to other members of the -oxoamine synthase family. It is possible that creating a chimera from individual domains of each protein (for example, the N-terminal domain of ALAS an d the catalytic and C-terminal domains of AONS) would produce a functional hybrid, but this remains to be tested. Although we succeeded in expressing both the ALAS/AON S and AONS/ALAS chimeras as active enzymes, we were only ab le to purify the ALAS/AONS chimera. The molecular mass of the native ALAS/AONS ch imera was consistent with that of a homodimer containing two chimeric polype ptide subunits. The fluorescence spectra exhibited by the ALAS/AONS chimera were distinct from those of ALAS and AONS and were consistent with an enzyme exhibiting a mixture of ALAS and AONS spectroscopic characteristics. The ALAS a nd AONS steady-state ki netic activities were diminished by roughly one-half in the chimera, and the catalytic efficiencies were not impaired. The pre-steady-state kinetic analys is for the ALAS reaction demonstrated that the reactivity of the ALAS sites in ALAS/AON S was similar to that of ALAS, with the rate-limiting step occurring after catalysis. The linking of two proteins or f unctional domains in natural and de novo fusion proteins generally involves a peptide linker (Gokhale and Khos la, 2000; Arai, Ueda et al., 2001; Wriggers, Chakravarty et al., 2005). A primary goal of linker engineering is to effectively separate the two functional domains to avoid intermolecula r strain and prevent unwanted interactions between the two modules (Carlsson, Ljung et al., 1996; Seo, Koo et al., 2000; Arai, Ueda et al ., 2001; Arai, Wriggers et al., 20 04; Wriggers, Chakravarty et
132 al., 2005). Because our objective was to faci litate interaction be tween the subunits of ALAS/ALAS and the AONS and ALAS chimer as, only two amino acids (Glu-Leu), which were introduced with the construction of a restriction site between the cDNAs, link the two subunits. Like both ALAS single-ch ain dimeric variants, the linking of the ALAS and AONS subunits appeared to change the energy ba rrier associated with the structural rearrangement that occurs upon ALA fo rmation to allow product release. It is likely that the use of the s hort dipeptide to link the ALAS C-terminus with the Nterminus of either ALAS or AONS introduced intermolecular strain which altered conformational flexibility. Remarkably, the strain introduced with the short linker di d not significantly impair the catalytic efficiencies in the any of the fusion proteins. In fact, the catalytic efficiency for succinyl-CoA increas ed roughly 15and 30-fold in ALASK313A/ALAS and ALAS/ALAS, respectively (Table 3.3). The short linker also appears to have facilitated interdomain communication in the single-chain dimeric variants, allowing changes at one site to be propagated to the adjacent site Generally, efforts are made to avoid the introduction of strain when engineering hybrid proteins; however, with the increased use of high-throughput protein engineering, the in troduction of intermolecu lar strain through linker domains may be a reasonable approach to creating new proteins with enhanced or novel functions. In summary, we have shown that the tw o active sites in the ALAS/ALAS singlechain dimer make asymmetrical steady-state co ntributions to the activity of the enzyme. This and the reduced steady-state activity of the single-chain chimeras of ALAS and AONS relative to the parent enzymes are likely to be caused by differences in
133 conformational changes at product release, which in turn are due to the strain introduced by joining the two subunits wi thout a linker region. Alt hough the chimeric ALAS/AONS and AONS/ALAS proteins did not form hybrid active sites, they we re able to form dimers with separate regions of ALAS and A ONS activity. Thus, the extensive structural plasticity seen in ALAS exte nds to another member of the -oxoamine family, AONS.
134References Alexeev, D., M. Alexeeva, et al. (1998). "The crystal structure of 8-amino-7oxononanoate synthase: a bacterial PL P-dependent, acyl -CoA-condensing enzyme." J. Mol. Biol. 284(2): 401-419. Arai, R., H. Ueda, et al. ( 2001). "Design of the linkers which effectively separate domains of a bifunctional fusion protein." Protein Eng. 14(8): 529-532. Arai, R., W. Wriggers, et al. (2004). "Conformations of variab ly linked chimeric proteins evaluated by synchrotron X -ray small-angle scattering." Proteins 57(4): 829-838. Astner, I., J. O. Schulze, et al. (2005). "Crystal structure of 5-aminolevulinate synthase, the first enzyme of heme biosynthesi s, and its link to XLSA in humans." Embo. J. 24(18): 3166-3177. Carlsson, H., S. Ljung, et al. (1996). "Physical and kinetic effects on induction of various linker regions in beta-galactosidase/g alactose dehydrogenase fusion enzymes." Biochim Biophys Acta 1293(1): 154-160. Cheltsov, A. V., M. J. Barber et al. (2001). "Circular perm utation of 5-aminolevulinate synthase. Mapping the polypept ide chain to its function." J. Biol. Chem. 276(22): 19141-19149. Cheltsov, A. V., W. C. Guida, et al. (2003). "Circular perm utation of 5-aminolevulinate synthase: effect on folding, confor mational stability, and structure." J Biol. Chem. 278(30): 27945-27955. Christen, P. and P. K. Mehta (2001). "From cofactor to enzymes. The molecular evolution of pyridoxal-5'-phos phate-dependent enzymes." Chem. Rec. 1(6): 436447. Del Campillo-Campbell, A., G. Kayajanian, et al. (1967). "Biotin-requiring mutants of Escherichia coli K-12." J. Bacteriol. 94(6): 2065-2066. Eliot, A. C. and J. F. Kirsch (2004). "Pyridoxal phosphate enzy mes: mechanistic, structural, and evolutio nary considerations." Annu. Rev. Biochem. 73: 383-415. Ferreira, G. C. and A. V. Cheltsov (2002). "Circular perm utation of 5-aminolevulinate synthase as a tool to evaluate folding, structure and function." Cell Mol. Biol. (Noisy-le-grand) 48(1): 11-16. Ferreira, G. C. and H. A. Dailey (1993). "Expression of mammalian 5-aminolevulinate synthase in Escherichia coli. Overproducti on, purification, and characterization." J. Biol. Chem. 268(1): 584-590.
135 Ferreira, G. C. and J. Gong (1995). "5-Amino levulinate synthase and the first step of heme biosynthesis." J. Bioenerg. Biomembr. 27(2): 151-159. Ferreira, G. C., P. J. Neame, et al. (1993) "Heme biosynthesis in mammalian systems: evidence of a Schiff base linkage between the pyridoxal 5'-phosphate cofactor and a lysine residue in 5-am inolevulinate synthase." Protein Sci. 2(11): 1959-1965. Ferreira, G. C., U. Vajapey, et al. (1995). "Aminolevulinat e synthase: lysine 313 is not essential for binding the pyridoxal phospha te cofactor but is essential for catalysis." Protein Sci. 4(5): 1001-1006. Gokhale, R. S. and C. Khosla (2000). "Role of linkers in communi cation between protein modules." Curr. Opin. Chem. Biol. 4(1): 22-27. Gong, J. and G. C. Ferreira (1995). "Aminol evulinate synthase: functionally important residues at a glycine loop, a putative pyr idoxal phosphate cof actor-binding site." Biochemistry 34(5): 1678-1685. Gong, J., G. A. Hunter, et al. (1998). "Aspar tate-279 in aminolevulin ate synthase affects enzyme catalysis through enhancing the function of the pyridoxal 5'-phosphate cofactor." Biochemistry 37(10): 3509-3517. Han, G., K. Gable, et al. ( 2006). "Expression of a novel marine viral single-chain serine palmitoyltransferase and construction of yeast and mammalian single-chain chimera." J. Biol. Chem. 281(52): 39935-39942. Hunter, G. A. and G. C. Fe rreira (1995). "A continuous sp ectrophotometric assay for 5aminolevulinate synthase that utilizes substrate cycling." Anal. Biochem. 226(2): 221-224. Hunter, G. A. and G. C. Fe rreira (1999). "Lysine-313 of 5-Aminolevulinate Synthase Acts as a General Base during Formation of the Quinonoid Reaction Intermediates." Biochemistry 38(38): 12526-12531. Hunter, G. A., J. Zhang, et al. (2007). "Tra nsient kinetic studies support refinements to the chemical and kinetic mechanisms of aminolevulinate synthase." J Biol Chem 282(32): 23025-23035. Kirsten, H., H. Gehring, et al. (1983). "Crystalline aspa rtate aminotransferase: latticeinduced functional asymmetr y of the two subunits." Proc Natl Acad Sci U S A 80(7): 1807-1810. Li, J. M., O. Brathwaite, et al. (1989) "5-Aminolevulinic acid synthesis in Escherichia coli ." J. Bacteriol. 171(5): 2547-2552.
136 Onuffer, J. J. and J. F. Kirsch (1994). "C haracterization of the apparent negative cooperativity induced in Escherichia coli aspartate aminotransferase by the replacement of Asp222 with alanine. Evidence for an extremely slow conformational change." Protein Eng. 7(3): 413-424. Paiardini, A., F. Bossa, et al. (2004). "Evol utionarily conserved regions and hydrophobic contacts at the superfamily level: The case of the fold-t ype I, pyridoxal-5'phosphate-dependent enzymes." Protein Sci. 13(11): 2992-3005. Ploux, O. and A. Marquet (1992). "The 8-amino-7-oxopelargonat e synthase from Bacillus sphaericus. Purification and preliminary characterization of the cloned enzyme overproduced in Escherichia coli." Biochem J. 283 ( Pt 2): 327-331. Sasarman, A., M. Surdeanu, et al. (1968). "L ocus determining the synthesis of deltaaminolevulinic acid in Escherichia coli K-12." J. Bacteriol. 96(5): 1882-1884. Schmidt, A., J. Sivaraman, et al. (2001). "Three-dimensional structure of 2-amino-3ketobutyrate CoA ligase from Escherichia coli complexe d with a PLP-substrate intermediate: inferred reaction mechanism." Biochemistry 40(17): 5151-5160. Schneider, G., H. Kack, et al. (2000). "The manifold of vi tamin B6 dependent enzymes." Structure 8(1): R1-6. Seo, H. S., Y. J. Koo, et al (2000). "Characterization of a bifunctional enzyme fusion of trehalose-6-phosphate synthetase and trehalose-6-phosphate phosphatase of Escherichia coli ." Appl. Environ. Microbiol. 66(6): 2484-2490. Smith, P. K., R. I. Krohn, et al. (1985). "M easurement of protein using bicinchoninic acid." Anal. Biochem. 150(1): 76-85. Stetefeld, J., M. Jenny, et al. (2006). "Intersubunit signaling in glutamate-1semialdehyde-aminomutase." Proc. Natl. Acad. Sci. U S A 103(37): 13688-13693. Tan, D. and G. C. Ferreira (1996). "Active si te of 5-aminolevulinat e synthase resides at the subunit interface. Evidence from in vivo heterodimer formation." Biochemistry 35(27): 8934-8941. Tan, D., T. Harrison, et al. (1998). "Role of arginine 439 in su bstrate binding of 5aminolevulinate synthase." Biochemistry 37(6): 1478-1484. Tarun, A. S. and A. Theo logis (1998). "Complementation analysis of mutants of 1aminocyclopropane1-carboxylate synthase reveals the enzyme is a dimer with shared active sites." J. Biol. Chem. 273(20): 12509-12514.
137 Webster, S. P., D. Alexeev, et al. (2000). "Mechanism of 8-amino-7-oxononanoate synthase: spectroscopic, kinetic, and crystallographic studies." Biochemistry 39(3): 516-528. Webster, S. P., D. J. Campopiano, et al. (1998). "Characteris ation of 8-amino-7oxononanoate synthase: a bacterial PL P-dependent, acyl CoA condensing enzyme." Biochem. Soc. Trans. 26(3): S268. Wriggers, W., S. Chakravarty, et al. (2005). "Control of pr otein functional dynamics by peptide linkers." Biopolymers 80(6): 736-746. Yard, B. A., L. G. Carter, et al. (2007). "The structure of serine palmitoyltransferase; gateway to sphingolipid biosynthesis." J. Mol. Biol. 370(5): 870-886. Zhang, J., A. V. Cheltsov, et al. (2005). "C onversion of 5-aminolevul inate synthase into a more active enzyme by linking the two subunits: spectroscopic and kinetic properties." Protein Sci. 14(5): 1190-1200. Zhang, J. and G. C. Ferreira (2002). "Tra nsient state kinetic investigation of 5aminolevulinate synthase reaction mechanism." J. Biol.Chem. 277(47): 4466044669.
138 Chapter Four Histidine-282 in 5-Aminolevulinate Synt hase Affects Substrate Binding and Catalysis Abstract 5-Aminolevulinate synthase (ALAS), the first enzyme of the heme biosynthetic pathway in mammalian cells, is member of the -oxoamine synthase family of pyridoxal 5Â’-phosphate (PLP)-dependent enzymes. In all structur es of the enzymes of the oxoamine synthase family, a conserved hi stidine hydrogen bonds with the phenolic oxygen of the PLP cofactor and may be significant for substrate-binding, PLPpositioning, and maintaining the p Ka of the imine nitrogen. In ALAS, replacing the equivalent histidine, H282, with alanine re duces the catalytic effi ciency for glycine 450fold. The slow phase rate for glycine binding is decreased 60% while the overall Kd increased 4.5 fold. The distribution of th e absorbing 420 and 330 nm species was altered with an increased A420/A330 ratio from 0.45 to 1.05. This shift in species distribution was mirrored in the cofactor fluorescence and 300 to 500 nm circular dichroic spectra and likely reflects variation in th e tautomer distribution of th e holoenzyme. The 300 to 500 nm circular dichroic spectra of ALAS and H282A diverged in the presence of either glycine or aminolevulinate indicating that the reorientation of the PLP cofactor upon external aldimine formation is impeded in H 282A. Alterations were also observed in the Gly dK value and spectroscopic and ki netic properties, while the PLP dK increased 9-fold.
139 Altogether, the results imply that H282 coordinates the mo vement of the pyridine ring with the reorganization of the active-site hydrogen bond network and acts as a hydrogen bond donor to the phenolic oxygen to maintain the protonated Schiff base and enhance the electron sink function of the PLP cofactor. Introduction Heme is an essential tetrapyrrole in nearly all living cells, and all tetrapyrroles are biosynthesized from the same precursor, 5-am inolevulinic acid (ALA). In mammals, 5aminolevulinate synthase (ALA S, EC 126.96.36.199) catalyzes th e condensation of glycine and succinyl-CoA to form ALA, CoA, and carbon di oxide, in the first an d regulatory step of heme biosynthesis. Mammals express geneti cally distinct eryt hroid and housekeeping ALAS isoforms, and mutations in the erythroi d specific ALAS have been implicated in X-linked sideroblastic anemia, a disease char acterized by inadequate formation of heme and the accumulation of iron in the erythrob last mitochondria (May and Bishop, 1998). ALAS belongs to a catalyti cally versatile cla ss of enzymes that require pyridoxal 5Â’-phosphate (PLP) as a cofactor (Ferreir a and Gong, 1995). PLP-dependent enzymes that catalyze reactions involving amino acids share common mechanis tic characteristics based on utilization of the electron withdrawin g properties of the cofactor to labilize bonds to the substrate -carbon (Christen and Mehta, 2001). Specifically, the PLP cofactor covalently binds to the -amino group of an active site lysine via a Schiff base linkage to form the Â“internal aldimine.Â” The incoming amino acid s ubstrate replaces the lysine amino group to form an Â“external aldimineÂ” via a gem -diamine intermediate, in a reaction often referred to as transaldiminati on. Subsequently, the cleavage of one of the
140 substrate -carbon bonds leads to a resonance-stabili zed quinonoid intermediate in which the coenzyme acts an electron sink, stori ng electrons from the cleaved bond through the conjugated system of the Schiff base and pyrid inium ring. Ultimately, the electrons are dispensed back for the forma tion of new linkages to the C(Christen and Mehta, 2001). PLP-dependent enzymes have been classi fied according to reaction specificity relative to the C (Mehta and Christen, 1994) and fold-types derived from threedimensional structures (Alexander, Sandmei er et al., 1994; Elio t and Kirsch, 2004). ALAS is classified within the -oxoamine synthase sub-family of the within class II of fold type I of PLP-dependent enzyme supe rfamilies (Schneider, Kack et al., 2000). In all known structures of fold-t ype I, the pyridine ring of th e PLP cofactor superimposes very well (Kack, Sandmark et al., 1999). The pyridoxal moiety intera cts with the enzyme in a common motif, which includes the previously mentioned Schiff base linkage with an active site lysine, a salt-bridge between the pyridinium ring nitrogen a nd an aspartate, and a hydrogen bond with the phenolic oxygen which occurs through a variety of amino acids (Kack, Sandmark et al., 1999). In ALAS and other -oxoamine synthase enzymes, the hydrogen bond of the phenolic oxygen involves a conserved histidin e (Alexeev, Alexeeva et al., 1998; Schmidt, Sivaraman et al., 2001; Astner, Schulze et al ., 2005) which corresponds to H282 in murine erythroid ALAS (4.1). No studies have examined the role of this residue in any -oxoamine synthase family member, although ba sed on structural data alone it has been suggested that it may function as an acid cataly st during transald imination (Webster, Alexeev et al., 2000; Zhang, Cheltsov et al., 2005), play a key role in positioning the PLP
141 aromatic ring (Schmidt, Sivaraman et al., 2001), or influence the p Ka of the imine nitrogen (Webster, Alexeev et al., 2000). Studies in other -family enzymes indicate that the significance of interaction between the protein and the phenolic oxyge n of PLP may vary according to the requirements of the enzyme. In aspartate aminotransferase and 1-aminocyclopropane-1carboxylate synthase, the phenolic oxygen interacts with a tyro sine residue (Goldberg, Swanson et al., 1991; White, Vasquez et al ., 1994). The deletion of the hydrogen bond through the replacement of the active site tyrosine with phe nylalanine reveals a different function in the kinetic properties of each en zyme. In aspartate aminotransferase, the tyrosine stabilizes the reactiv e form of the internal aldi mine at physiological pH and increases the kcat value (Goldberg, Swanson et al ., 1991; White, Vasquez et al., 1994). Similar studies of 1-aminocyc lopropane-1-carboxylat e synthase reveal that the tyrosine decreases the Km, but has no affect on kcat (White, Vasquez et al., 1994). In murine erythroid ALAS, H282 is te thered between PLP and Y121 through hydrogen bonds between the imidazole N 2 and N 1, respectively (Figure 4.1). Previous studies have demonstrated that the Hbond between the Y121 hydroxyl group and H282 N 1 is important for efficient cofactor and s ubstrate binding (Tan, Barber et al., 1998), providing evidence for a probable role for H282 in these interactions. The ordered ALAS catalytic pathway is comprise d of the following steps (Scheme 4.1): the association of glycine with the enzyme forming the Mich aelis complex (I); the transaldimination reaction between glycine and th e active site lysine (K313) to generate the external aldimine (II); the removal of the pro -R proton to generate a transient quinonoid intermediate (III); the condensation of succi nyl-CoA (IV); the removal of Co-A and the
142 formation -amino--ketoadipate (V); decarboxylation of the -amino-ketoadipate(VI); the protonation of the second quinonoid intermediate (VII) and finally the releases of ALA (VII) (Zhang and Ferreir a, 2002),(Hunter and Ferreira, 1999). In order to characterize the role of the cons erved histidine in mu rine erythroid ALAS function, a series of H282 vari ants were constructed. The results provide evidence that H282 impacts a variety of ALAS functions including s ubstrate and PLP binding and catalysis. Materials and Methods Materials The following reagents were purchased from Sigma-Aldrich Chemical Company: DEAESephacel, Ultrogel AcA-44, -mercaptoethanol, PLP, bovine serum albumin, succinylCoA, ALA-hydrochloride, -ketoglutaric acid, -ketoglutarate dehydrog enase, Bis-Tris, HEPES-free acid, AMPSO-free acid, MO PS, tricine, thiamin pyrophosphate, NAD+, and the bicinchoninic acid protein determina tion kit. Glycerol, glycine, disodium ethylenediamine tetraacetic acid dihydrate, ammonium sulf ate, magnesium chloride hexahydrate, perchloric acid, and potassi um hydroxide were ac quired from Fisher Scientific. Sodium dodecyl sulfate polyacr ylamide gel electrophoresis reagents were acquired from Bio-Rad. Phenylhydrazine was from by Eastman Kodak. PD-10 columns were from Amersham Biosciences. Chamel eon mutagenesis kit was from Stratagene. Xho I and Xba I restriction enzyme s were from New England Biolabs. Methods Mutagenesis.
143 The pGF23 plasmid encoded the full-le ngth sequence for the murine, mature erythroid ALAS. Site-directed mutagene sis for the H282Y and H282F mouse ALAS mutant was performed on the single-stranded pGF23 using the chameleon mutagenesis kit from Stratagene. The mutagenic oli gonucleotides for H282Y and H282F were GAT GAA GTC TAT GCT TAT GCT GTA GGA CTG TAT GGA and GAT GAA GTC TTT GCT TAT GCT GTA GGA CTG TA T GGA, respectively, w ith the intr oduced codon substitutions underlined. The H282A mutant was generated using the method previously described by Gong (1998). Brie fly, two rounds of PCR were performed to obtained DNA fragments with the desired mu tation flanked by unique restric tion sites. The mutagenic primers used to generate the H282A mutation were 5Â’-GTA GAT GAA GTC GCT GCT GTA GGA CTG or 5Â’-GAG TCC TAC AGC AGC GAC TTC ATC TAC with the introduced codon substitution underlined. The two fragme nts containing the mutation were used as megaprimers and amplified by a third round of PCR. The product was then digested with Xba I and Xho I and subclone d into pGF23 vector. Clones obtained after mutagenesis procedures were confirmed by sequencing. Protein purification, SDS-PAGE, protein de termination and steady-state analysis Recombinant murine erythroid ALAS and the H282A variant were purified from DH5 Escherichia coli bacterial cells containing the ove rexpressed protein as previously described (Hunter and Ferre ira 1995). Sufficient expr ession of H282Y and H282F variants could not be obtain ed. Purity was determined by SDS-PAGE (Laemmli, 1970) and protein concentration dete rmined by the bicinchoninic acid method using BSA as the standard (Smith, Krohn et al., 1985). All prot ein concentrations are reported on the basis of a subunit molecular weight of 56,000 kD Enzymatic activity was determined by a
144 pH pK pH pK pK pHa a aY Y Y Y 10 1 log log 10 10 1 log logmax max2 1continuous spectrometric assay at 30oC (Hunter and Ferreira, 1995) To evaluate the pH dependence of the kinetic para meters, assays were perfor med in 20 mM MOPS for pH 6.7, HEPES for pH 7-8 or AMPSO for pH 8.2-9.5. The pH dependences of log kcat and log kcat/ Km were fit to equation 1 while the pH dependence of log 1/ Km was fit to equation 2 Equation 1 Equation 2 Spectroscopic measurements. Prior to spectroscopic measurement, enzy me was dialyzed in 20 mM HEPES, pH 7.5 with 10% glycerol to remove free PLP. Absorption spectra were acquired at ambient temperature using a Shimadzu UV 2100 dual beam spectrophotometer, with a reference containing all components except the purified en zyme. Circular dichroism (CD) spectra were obtained using an AVIV CD spectrometer calibrated for both wavelength maxima and signal intensity with an aqueous solutio n of D-10 camphorsulfonic acid (Chen, 1977). Protein concentrations were 10-11 M and 100 M for the near and far CD spectra, respectively, in 20 mM Bis-Tris, pH 7.5 containing 10% gly cerol. Spectra were recorded in triplicate and averaged, using a 0.1 cm path length cuvette with a total volume of 300 l. Fluorescence spectra were coll ected on a Shimadzu RE-5301 PC spectrofluorophotometer using pr otein concentrations of 2-4 M. The pH was adjusted with 20 mM MOPS (pH range 6.7-7.0), 20 mM HEPES (pH range 78.2), or 20 mM
145 min ) ( min max10 1 Y Y Y YpK pH AMPSO, (pH range 8.3-9.5). 10% glycerol was also included in the buffers. CD and fluorescence blank spectra were collected from samples containing all components except protein immediately prior to the measurement of samples. The blank spectra were subtracted from spectra of sample contai ning enzyme. The pH dependence of the 510 nm-fluorescence emission upon 420 nm-e xcitation was fit to equation 3. Equation 3 Stopped-flow spectroscopy. Rapid scanning stopped-flow measuremen ts were conducted using a model RSM100 stopped-flow spectrophotometer (OLIS Inc. This instrument has a dead-time of approximately 2-ms and an observation cham ber path length of 4 mm. Scan spectra covering a wavelength range of 300-510 nm were collected at a rate of 1000 scans/s and then averaged to 62 scans/s to reduce data f iles to a manageable level. The temperature of the syringes and the stopped-flow cell compartment was maintained at 30oC by an external water bath. The concentration of glycine was always at least 10-fold greater than the enzyme concentration to ensure pseudo-first order kinetic were observed. For each experimental condition, three rep licate experiments were performed. The Absorbance 420 nm were globally fit using the simulation software Dynafit to the binding models described in scheme 4.2 and 4.3 (Kuzmic, 1996).
146 ] [ ] ][ [ Enz Gly Enz Gly Kd k1 E E-G k -1 Scheme 4.2 Scheme 4.3 Determination of dissociation c onstants of glycine and ALA. Dissociation constants were determined spectroscopically by monitoring spectral changes upon the binding of glycine and ALA (Gong, Hunter et al., 1998). The Kd values for glycine from pH 6.7-9.5 were determined at 30 oC for ALAS and the H282A variant by monitoring the increase in cofactor ab sorbance at 420 nm upon glycine binding. The pH was adjusted with 20 mM MOPS (pH range 6.7-7.0), 20 mM HEPES (pH range 78.2), or 20 mM AMPSO, (pH range 8.3-9.5). 10% glycerol was al so included in the buffers. Glycine was prepared as 2 M st ocks adjusted to the same pH as the corresponding buffers. Kd is defined by equation 4 where [Gly] and [Enz] are the concentrations of free glycine and free enzy me, respectively, and [Gly-Enz] represents the concentration of glycine-bound ALAS. Equation 4 The changes in absorbance at 420 nm we re plotted as a function of glycine concentration and the data were fit to equation 5 to determine Kd, where Abs is the absorbance increase at 420, Absmax is the maximum increase in absorbance, and [Gly] is k1 k2 E E G* E-G k -1 k -2
147 2[E] ] ][ [ 4 ]) [ ] [ ( ] [ ] [ ) (2 E L E L K E L K A A A Ad d i f i ] [ ] [maxGly K Gly Abs Absd pH pK pK pHa aY Y 2 110 10 1maxthe total glycine concentrat ion. The pH dependence of Kd for ALAS was fit to equation 3 and for H282A equation 6 Equation 5 Equation 6 The ALA Kd for the H282A variant was determ ined by monitoring the decrease in absorbance at 420 nm at 30oC in 20 mM HEPES, pH 7.5 and 10% glycerol. Enzyme (2530 M) solution was titrated with small aliquots of concentrated ALA solution, and the change in absorbance measured. Data were analyzed by non-linear regression fitting to equation 7 where A is the observed absorbance, and Ai and Af are the fitted values of the initial and final absorbance, respectively. [L ] is the ligand concentration, and [E] is the enzyme concentration. Determinations were made in duplicate and the reported values represent the mean and standard error of measurement. Equation 7
148 min ) ( min max10 1 Y Y Y YpH pK Preparation of apoenzyme and determination of the PLP dissociation constant. To obtain H282A apoenzyme, 1 mg/ml enzyme in 20 mM HEPES pH 7.5, containing 20% glycerol was treated with 150 mM phenylhy drazine for 1.5 hours at 4oC, following which phenylhydrazine was remove d by running the solution through a PD-10 column. The phenylhydrazine treatment was then repeated to ensure all PLP was removed. The PLP Kd for the H282A variant was determined at 25oC by monitoring the PLP-dependent increase in 510 nm fluorescen ce emission upon excitation at 420 nm, in a buffer composed of 20 mM HEPES, pH 7.5 and 10% glycerol. To determine Kd, data were analyzed by non-linear regression fitting to equation 7 where A is the observed fluorescence, and Ai and Af are the fitted values of the initial and final fluorescence, respectively. pH titration of quinonoid intermedia te formation for H282A variant. The pH dependence of quinonoid intermedia te formation was investigated with ALA saturated enzymes as described prev iously (Gong, Hunt er et al., 1998). Equation 8 was used to fit the quinonoid intermediate titration curves where Y is the observed absorbance at 510 nm, Ymax and Ymin are the theoretical maximal and minimal absorbance values at 510 nm, and p Ka is the equivalence point for quinonoid intermediate formation. Equation 8
149Results Spectroscopic properties of the H282A variant. At pH 7.4, three absorbance maxima at approximately 278, 330 and 420 nm are observed in both ALAS and the H282A variant (Figure 4.2A). The absorbance at 278 nm is primarily due to the protein, while th e 330 nm and 420 nm maxima are common in PLP-dependent enzymes and are typically at tributed to deprotona ted and protonated aldimine species, respectively (Metzler and Metzler, 1987). A similar assignment for ALAS is ambiguous because the spectrum is unchanged in the pH range 6.5-9.51. The mutation had no discernable eff ect on the protein absorptio n band centered at 278 nm, but the cofactor absorption peaks were significantl y altered. The ratio of the 420 nm to 330 nm absorbance was increased from 0.45 in the w ild-type enzyme to 1.05 in the variant. The changes in the absorption spectra we re reflected in the fluorescence spectra (Figure 4.2B and C). Upon excitation at 330 nm ALAS exhibits only one maximum at 385 nm, while in H282A the 385 nm fluores cence emission maximum is shifted to 410 nm with a 6-fold decrease and a second maximum is observed at 510 nm. With excitation at 420 nm, the cofactor exhibits fluorescence emission maximum at 510 nm for both enzymes; however the magnitude of the 510 nm emission was ~7 times greater in the H282A. The pH titration of the species emitting at 510 nm upon excitation at 420 nm demonstrated that this species diminished as a result of loss of a single proton for both enzymes (Figure 4.3). A fit of the data to equation 3 yield a p Ka of 8.05 0.043 and 9.02 0.07 for ALAS and H282A, respectively. 1 G.C. Ferreira and G.A. Hunter, unpublished results
150 Kinetic characterization of the H282A variant. The steady-state kinetic parameters of the H282A variant were determined and the results are summarized in Table 4.1. The mutation resulted in a kcat of 1.4% of the wild-type ALAS value. The Km for glycine was increased 5-fold relative to ALAS, while the Km for succinyl-CoA was not significantly affected. The overall catalytic efficiency for glycine and succinyl-CoA decr eased 450-fold and 87-fold respectively as compared to ALAS values. If H282 acts as a hydrogen bond donor to th e phenolic oxygen of the cofactor, then the H282A mutation may lower the p Ka for the imine nitrogen. To investigate this possibility the pH dependence of the steady-st ate kinetic parameters was studied, with the results summarized in Table 4.2 and Figure 4.4. The log kcat vs pH profile for H282A decreased on both the acidic and basic si des, and the best fit of the data to equation 1 generated a p Ka of 7.2 0.1 for a residue in the enzyme-substrate complex that must be protonated for optimal catalysis. A second p Ka of 8.6 0.1 for a residue that is deprotonated during catalysis was also observed for H282A which shifted from the previously reported p Ka of 9.1 0.03 in ALAS (Zhang, Cheltsov et al., 2005). The possibility of an acidic limb p Ka, below 7.0, in the wild-type enzyme could not be investigated due to instability at pH values below 6.5-7.0 (Zhang, Cheltsov et al., 2005), but the available data do sugge st that the H282A mutation resu lts in a substantial increase to the p Ka of an important enzyme-substrate comp lex ionization. This ionization might be assigned directly to H282, or it could be assigned to the imine nitrogen that presumably shares a proton with the phenolic oxygen atom.
151 The log kcat/ Km pH profile for the mutant was si milar to that of the wild-type enzyme, decreasing on both the acidic and basi c limbs. Nonlinear regression of the data using equation 1 yielded a p Ka for the acidic and basic limb of 8.00 0.14 and 8.50 0.14 for H282A, which reflects a shift in the acidic limb from the at 8.60 0.11 p Ka value previously reported in ALAS (Zhang, Cheltsov et al., 2005). The pH variation of the log 1/ Km Gly decreased with increasing pH for bot h enzymes. The data were fit to equation 2 to generate a p Ka of 8.36 0.1 for ALAS and 7.76 0.16 for the H282A variant. The log kcat and logGly m cat/ K k profiles limiting slopes of approximately 1 or -1 indicate the ionization of a sing le group for acidic and basic li mbs. Given that glycine is not a sticky substrate an d does not ionize over the pH range studied, the p Ka observed for 1/ Km Gly and the acidic limb of the log Gly m cat/ K k likely represents gr oup(s) in the free enzyme.
152aDatafrom (Gong, 1998)14 () 40 (4) 0.05 (0.002) 2.75 (.07) 49 (5) 9.5 x10-4(0.06x10-4) 144 (.7) 0.137 (0.003) H282A 1.6 () 25 (3) 4.35 (.62) 2.3 (.1) 22 (2) 0.4 (.06) 23 () 10 () ALASa( M) ( M) (min-1 M-1) ( M) (mM) (min-1mM-1) (mM) (min-1) Protein Table Summary of steady-state kinetic parameters and dissociation constants aDatafrom (Gong, 1998)14 () 40 (4) 0.05 (0.002) 2.75 (.07) 49 (5) 9.5 x10-4(0.06x10-4) 144 (.7) 0.137 (0.003) H282A 1.6 () 25 (3) 4.35 (.62) 2.3 (.1) 22 (2) 0.4 (.06) 23 () 10 () ALASa( M) ( M) (min-1 M-1) ( M) (mM) (min-1mM-1) (mM) (min-1) Protein Table Summary of steady-state kinetic parameters and dissociation constants Gly mK Gly mcatKkGly dKSCoA mKALA dKPLP dKcatk SCoA mcatKk Table 4.2: Summary of p K values obtained from the pH dependence of kinetic parameters ALAS H282A p Ka p Ka 1 2 1 2 Log catk nda (<6.7)9.1.03 7.2.09 8.6 .09 Log Gly mK 1 8.4.10 7.8.16 Log Gl y m catK k 8.6.11a 8.75.13b 8.0.14 8.5 .14 Log Gly dK nd (<7) 7.4.2 8.1.2 aData from ( 12) b Data from (21) nd, not determined
153 Reaction of glycine w ith H282A variant. The reaction of 60 M H282A variant with glycine resulted in an increased absorbance at 420 nm (Figure 4.5a). The data best fit to the two step model described by scheme 4.2 (Figure 4.5b-c). A fit of the data yielded values for k1 of 0.001654 3.8 x 105 s-1, k-1 of 0.14 0.0064 s-1, k2 of 0.022 0.0025 s-1, and k-2 of 0.0455 0.0016. Dissociation constants for the bi nding of glycine and ALA. To elucidate a potential role of H282 in substrate binding, the enzymes were titrated with glycine and ALA to determine th e dissociation constants for formations of the external aldimine with the subs trate and product. At pH 7.5 the Kd for ALA and glycine increase 8.5-fold and 5-fold, respectiv ely, relative to ALAS (Table 4.1). To establish if the ionization of groups reflected in the kcat profiles are involved in substrate binding or catalysis, the pH dependence of Kd for glycine was determined. The loss of the PLP-O3-H282 interaction also had a ma rked effect on the pH profile for the Gly dK values. For ALAS, the Gly dK decreases with increasing pH and, when fitted to equation 1, yielded a p Ka value at the boundary of the pH range tested, therefore a p Ka < 7 was assumed. In contrast, the Kd Gly for the H282A variant f it to a bell curve with p Ka values at 7.4 0.2 and 8.1 0.2 (Figure 4.6). Th e data indicate that th e H282 mutation results in a substantial modification to the p Ka of an enzyme-glycine complex ionization. pH titration of quinonoid intermedia te formation for H282A variant. When ALAS is saturated with ALA, the external aldimine is converted to a quinonoid intermediate in a pH-dependent manne r; the extent of th is reaction can be monitored by following the absorbance of the quinonoid intermediate at 510 nm. Formation of the ALA-bound quinonoid intermed iate in ALAS has been reported to
154 occur with an apparent p Ka of 8.1 0.1 (Gong, Hunter et al., 1998), and involves participation of the active site K313, which acts as a general base catalyst for the reaction by abstracting a proton from the ALA-aldi mine to form the quinonoid intermediate (Hunter and Ferreira, 1999). The ALA-bound quinonoid intermediate was observed to increase with pH for H282A as was observed previously in ALAS (Hunter and Ferreira, 1999), although the amplitude of the abso rption of the quinonoid intermediate was markedly diminished by the mutation at all pH values tested (Figure 4.7A). pH titration of the H282A quinonoid intermediate absorbance demonstrated that the intermediate was formed as a result of loss of a single proton w ith an equivalence point at 8.8 0.1 (4.7B). The higher p Ka value in the variant i ndicates that one function of H282 is to lower the apparent p Ka for quinonoid intermediate formation su ch that the PLP co factor functions more effectively as an electron sink at physio logical pH. The observation that disruption of a hydrogen bond to the phenolic oxygen of th e cofactor has a significant effect on quinonoid intermediate formation indicates th at the equivalence point of 8.1 0.1 observed with ALAS is a complex function of th e electronic interactio n of the active site lysine with the ALA-PLP aldimine and its active site environment, and not simply reflective of an ionization constant for the active site lysine. CD spectroscopy. The disruption of the H-bond be tween the phenolic oxygen and the enzyme could potentially a lter the time-averaged orientat ion of the PLP cofactor in the active site. The circular dichroism in the UV-visible re gion reflects the PLP microenvironment by monitoring the asymmetry of the bound cof actor. Formation of an external aldimine results in the reorientati on of the PLP cofactor which can be followed with CD spectroscopy (Moore, Dominici et al ., 1995). Spectra of the holoand ligand-
155 bound enzymes were collected (Figure 4.8). Spectra of the free enzyme exhibited positive dichroic bands at ~330 and 420 nm with an increase in the 420 nm band with an associated decrease in the ~330 nm band observe d in the variant. Th e addition of glycine to ALAS or H282A resulted in comparable d ecrease in the ~330 nm dichroic band, while the ~420 nm band decreased 75% in the varian t and disappeared in the ALAS spectra. The addition of a saturating concentration of ALA to ALAS or H282A had strikingly different effects on the relative chiral environment of the ex ternal aldimine in the two enzymes. Specifically, the ~330 nm dichro ic band was decreased and the ~420 nm band disappeared in the ALAS spectra, while th e addition of ALA to H282A resulted in a moderate increase in the 330 nm and little ch ange to the ~420nm band. The CD spectra for ALAS and H282A between the 200-300 nm were similar, indicating that no significant changes occurred in the overall c onformation as a result of the mutation (data not shown). Dissociation constants fo r the binding of PLP. To address the role of H282 in cofactor bi nding, the effect of the H282 to alanine mutation on Kd of PLP was studied. The titration of the H282A apoenzyme with PLP leads to the reconstitution of the holoen zyme, which can be monitored by following changes in the intensity of fluorescence emission at 510 nm upon excitation at 420 nm (Figure 4.6). Theoretical saturation curves were generated from which the dissociation constant of PLP from H282A was determined. When compared to the wild-type enzyme, the Kd for PLP was increased ~9 fold by the H 282A mutation, as re ported in Table 4.1.
156Discussion The crystal structure of Rhodobacter capsulatus ALAS reveals the existence of a hydrogen bond between H282 and the phenolic oxyge n atom of the PLP cofactor (Astner, Schulze et al., 2005). A clusta l sequence alignment demonstr ated that th is histidine residue was perfectly conserved in over 70 known ALAS sequences from bacteria to mammals (data not shown). The existe nce of one, and often two, hydrogen bonds between the enzyme and the PLP phenolic oxygen is common in fold type I PLPdependent enzymes, and is likely multifunctiona l. The ALAS crystal structures suggest possible roles for H282 in binding and orientation of the cofactor within the active site, as well as control of the electronic status of the cofactor during catalysis(Alexeev, Alexeeva et al., 1998). These possibiliti es led us to postulate that mutation of H282 should have multiple effects on substrate and cofactor bi nding, as well as catalysis. In this communication, we constructed ALAS va riants harboring the H282A, H282Y, and H282F mutations, of which only the H282A vari ant was recoverable as a soluble enzyme. The effects of the H282A mutation on the sp ectroscopic and kineti c properties of the enzyme were characterized in order to bette r understand the functional roles of H282 in the ALAS-catalyzed formation of ALA. The absorption spectra (Figure 4.2) indica te the mutation has a substantial effect on the electronics of the PLP cofactor. A d ecrease in the absorbance of the 330 nm peak is accompanied by an increase in the absorbance of the 420 nm peak. These changes are reflected in the cofactor fluorescence spect ra. In some transaminases, including aspartate aminotransferase and tyrosine am inotransferase, the co rresponding absorbance peaks titrate as a function of pH with the long wavelength peak favored at low pH and the
157 short wavelength peak favored at high pH (Goldberg, Swanson et al., 1991; Chow, McElroy et al., 2004) These are generally at tributed to the ketoenamine and enolamine tautomers, respectively, which differ in th e position of the proton shared between the phenolic oxygen and the Schiff base nitrogen atoms. The changes in the absorbance spectra for H282A suggest that the mutati on significantly alters the equilibrium of cofactor tautomeric structures to favor the ketoenamine, but this assignment is ambiguous because, unlike aspartate and tyrosine aminot ransferases, the absorbance spectrum of ALAS is largely pH-independent and the H 282A mutation did not al ter this property (data not shown). In contrast to the absorption spectra, fluorescence spectra of ALAS upon excitation at 330 or 420 nm are pH-depende nt (Zhang, Cheltsov et al., 2005). Upon excitation at 420 nm, the ALAS 510 nm fluor escence emission titrates with a single p Ka of 8.05 0.043, while in H282A a p Ka of 9.02 0.07 is observed under similar conditions (Figure 4.3). The 385 nm fluorescence emi ssion signal resulting from excitation at 330 nm, which occurs with a p Ka of 8.4 0.1 in the wild-type enzyme, is greatly diminished in the mutant, and an e quivalent titration coul d not be performed. The two p Kas observed in ALAS fluorescence sp ectra are presumably indicative of more complex chemistry than simple titration of the Schiff base nitrogen atom. This is not unprecedented, as in dialky lglycine decarboxylase, three p Kas are observed during absorbance spectra titrations with both ketoenamine and enolamine species present in each ionization state (Zhou and Toney, 1999). For both dialkylglycine decarboxylase and glutamate decarboxylase it has been proposed that the multiple ionizations observed reflect active site residues that regulate the distribution of ketoenamine and enolamine
158 tautomers through electrostatic effects (Chu a nd Metzler, 1994; Zhou and Toney, 1999). In ALAS, the ionizations observed in the fluorescence spectra, but not the absorption spectra, are also likely to be attributable to active site residues and not the Schiff base nitrogen. Alterations observed in the H282A spectra may be due to changes in both tautomeric equilibria and the electrostatic interactions between the phenolic oxygen and other active site residues. The steady-state kinetic parameters of the variant indicate loss of H282 interaction with the phenolic oxygen impairs both glyc ine binding and catalysis. The Gly mK increased 5-fold and the kcat decreased by two orders of magnitude. Rapid-scanning stopped-flow analysis experime nts were performed to furthe r characterize the effect of the mutation. A pre-steady-state burst of the quinonoid intermediate for the reaction of H282A-glycine and succinyl-CoA was not ob served, presumably due to diminished absorption of the quinonoid intermediate that is typically observed with the wild-type enzyme (data not shown). The transimination reaction expected to occur during glycin e binding involves nucleophilic attack of the prot onated Schiff base internal al dimine by the deprotonated amine of glycine, to form a transient g em -diamine intermediate. If the hydrogen bond donated by H282 to the phenolic oxygen of the cofactor is important in maintaining a protonated Schiff base, then th e loss of this hydrogen bond in H282A might be expected to slow the rate at which glycine binds to the enzyme. In the absence of succinyl-CoA, glycine binding to H282A is a two-step pr ocess (Figure 4.5). Previous studies demonstrated that glycine also binds with ALAS in two steps; however, the rates associated with the fast phase were not slow enough to be re solved (Hunter and Ferreira,
159 1999). The slow phase rate for H282A decreas ed 85% relative to ALAS. The slower binding of glycine observed in H282A may be at tributed to alteratio ns in the electronic status of the Schiff base, but other interpre tations are also possibl e. One interesting possibility is that H282 is dir ectly or indirectly involved in proton transfers that convert the internal aldimine and glycine to the reacti ve ionic states necessary to formation of the glycine external aldimine (Scheme 4.1, I). In any case, these data, along with the data in Figure 4.6, indicate an important ro le for H282 in glycine binding. The pH-dependence of log kcat, log Gly m cat/ K k and log 1/Gly mK were all diminished in H282A-catalyzed reaction, indicating that the mutation had severe catalytic consequences (Figure 4. 4), but only the log kcat profile contained an ionization that was obviously changed by the mutation. The appearance of a new p Ka of ~7.3 for H282A in the acidic limb of both the log kcat and logGly dK suggests that the mutation results in a substantial change to a p Ka for the enzymeÂ–glycine complex. In ALAS kcat is known to be determined by release of ALA, or a c onformational change associated with ALA release (Hunter and Ferreira 1999 ). The appearance of an ac id limb ionization in the log kcat vs. pH profile for H282A shifted the pH op timum from less than 6.5 to slightly over 8.0, and indicates a change in the nature of the rate-determini ng step for catalysis, at least at lower pH values. The furt her observation that a similar p Ka is apparent in the log Gly dK pH profile suggests that in H 282A the rate-determining step at pH values less than 8.0 may be associated with binding of gl ycine. The ALAS spectroscopic p Ka of 8.4 0.1 observed upon excitation at 330 nm is mirrored in the log 1/ Km and the acidic limb of log kcat/ Km pH profile of ALAS (Zhang, Cheltsov et al., 2005). Although it was not possible
160 to titrate the equivalent speci es in the H282A spectra, th e pH dependence of the log 1/ Km and acidic log kcat/ Km was shifted to ~7.9. This ionization controls the reactive free enzyme species, and the disappearance or si gnificant reduction of the equivalent species in the H282A spectrum suggests H282 stabili zes the reactive form of the internal aldimine. In ALAS, the binding of ALA results in the appearance of an ALA-quinonoid intermediate with a 510 nm absorbance (Gong, Hunter et al., 1998). In H282A the addition of ALA results in a decrease in absorbance at 420 nm w ith an associated increase at 330 nm in addition to the app earance of a 510 nm absorbance, though the amplitude of the 510 nm absorption associ ated with the ALA-quinonoid species is markedly diminished. While Kd for ALA is minimally affected, the p Ka of the ALAquinonoid intermediate is increased from 8.1 0.1 in ALAS to 8.8 0.1 in H282A. The data suggest that proton abstra ction from ALA is impaired in the variant. One possible explanation is that the lo ss of the hydrogen bond between the phenolic oxygen and H282 is likely to cause a net flow of electrons into the conjugated -bond system, thereby disrupting the electron sink cap acity of the cofactor. Additionally, it has been suggested that the process of ALA binding and quinonoid intermediate formation may involve some structural reorganization of the active site (Hunter and Ferreir a, 1999). In both the glyc ine and succinyl-CoA soaked R. capsulatus ALAS crystals, a 15o rotation of the pyridine ring around the C5-C5A bonds occurs such that the O3 and C4A atoms move away from the catalytic lysine (Astner, Schulze et al., 2005). Upon the binding of pr oduct in AONS, a similar rotation of the pyridine ring occurs along w ith subtle rearrangement of the active site hydrogen bond
161 system (Webster, Alexeev et al., 2000) In ALAS, the movement of the O3 is tracked by the residue equivalent to H282 (Astner, Sc hulze et al., 2005), indicat ing H282 is probably involved in coordinating the movement of the pyridine ring with the reorganization of the hydrogen bond system occurring upon substrate binding. H282 is te thered between the phenolic O3 of PLP and Y121 by way of hydrogen bonds between the imidazole N 2 and N 1, respectively. The loss of H282 hydroge n bonds with the cofactor and Y121 would likely affect the PLP movement and orientation within the activ e site that are presumably a crucial aspect of the catalytic process. The possibility that the PLP microenvironm ent is affected in solution by the H282A mutation was examined using CD sp ectroscopy, performed in the absence and presence of ALA or glycine. The spectra for the ALAS and H282A holoenzymes are relatively similar and exhibit two positive dichroic bands around 330 and 420 nm (Figure 4.8) which mirror the shift from the 330 to 420 nm species observed in the absorbance spectra (Figure 4.2). In these holoenzyme spect ra the cofactor is covalently anchored to the enzyme via the internal aldimine linkage with the active site lysine, and this attachment would be expected to maintain th e orientation of the cofactor in the active site. However, upon formation of an exte rnal aldimine with glycine or ALA, the attachment of the cofactor to the active site lysine is lost, and the resulting CD spectra diverge in the two enzymes. Upon the addi tion of glycine, the 420 nm dichroic band disappears in ALAS, while the band continue s to be observed in the H282A spectra (4. 8). In ALAS, the binding of ALA results in the loss of the 330 nm band and a significantly diminished ~420 nm band. In contrast, the bindi ng of ALA to H282A results in a spectrum remarkably similar to the holoenzyme. The divergence observed in
162 the ligand bound CD spectra of ALAS and H282A suggest that the reorientation of the PLP cofactor, observed with ALAS upon extern al aldimine formati on, is blocked or diminished by the H282A mutation. This would influence both the cofactor position and interaction with key catalytic residues and could help explain the multiple effects caused by the mutation in the variant. In summary, H282 is involved in a hydr ogen bond with the phenolic oxygen of the PLP cofactor. The deletion of this in teraction in the H282A variant has multiple effects on the spectral, binding, and kinetic properties of th e enzyme that support the conclusion that H282 plays multiple roles in th e enzymology of ALAS. It may also be further concluded that the impaired function of the variant results from a combination of direct and indirect effects, including alterations in the prot onation of the phenolic oxygen and changes to the stereoelec tronic relationships between the cofactor and active site residues, through the disruption in the proces sional PLP positioning that normally occurs during catalysis.
163 Scheme 4.1
164 FIGURE 4.1 Spatial position of active site residues in the R. capsulatus ALAS holoenzyme crystal structure. This view highlights the interactio n of the pyridinium ring of the cofactor with active site residues and the H282 imidazole N 2 and N 1 hydrogen bonds between the cofactor phenolic oxygen and Y121. The im age was constructed using Pymol (DeLano 2002) and PDB file 2BWN. Residue numbering is relative to murine erythroid ALAS.
165 FIGURE 4. 2 Absorption and fluorescence spectra of ALAS and H282A variant. (A) UV-visible absorption spectra. The inset in cludes the region from 250Â–300 nm Protein concentrations were adjusted to 13 M in 20 mM Hepes, pH 7.5. (B) Fluorescence emission spectra of 5 M ALAS and H282A in 20 mM He pes, pH 7.5 containing 10% glycerol upon excitation at (B) 330 nm a nd (C) 420 nm. For (A) Â– (C), ALAS ( ) and H282A ( Â— ).
166 FIGURE 4.3 The pH dependence of fluorescence emission. The fluorescence emission at 510 nm upon excitation at 420 nm for 2.0 M ALAS ( ) and 3.5 M H282A ( ) at varying pH. Each of the line represents the nonlinear regression fit to equation 3
167 FIGURE 4.4 pH dependence of (A) log kcat (B) log kcat/ Km Gly and (C) log 1/ Km Gly for ALAS ( ) and H282A ( Â— ). The lines represent the nonlinear re gression fits to equation 1 or 2 as described in Materials and Methods. The pr ofiles for the pH dependence of the steadystate kinetic parameters for ALAS ( ) are from (Zhang, Cheltsov et al., 2005).
168 FIGURE 4.5 Reaction of 60 M H282A variant with glycine. (A) Spectra changes observed during the reaction of 300 mM glycine with H282A. Spectra were collected at 1, 5, 11, 23, 41 and 52 seconds and are shown sequentially with th e lowest to the highest absorbance at 420 nm. The A420 were globally fit to a two-step model using the simulation software Dynafit. (B) The A420 data for the time course re action of 300 mM glycine is represented by circles, with the line repres enting the fitted data. (C) The fit of the A420 data for glycine binding at 100, 125, 150, 200, 300, 400, 500 and 600 mM.
169 FIGURE 4.6 pH dependence of the Kd for glycine for ALAS ( ) and H282A ( ). The data were fit to equation 5 (ALAS) or equation 6 (H282A) using non-linear re gression analysis.
170 FIGURE 4.7 UV-visible absorption spectra of H282A in the presence of ALA and pH-dependence of ALA-quinonoid intermediate formation. (A) Absorption spectra of H282A ( Â— ) and ALAS ( ) in the presence of 500 M ALA. ( Inset ) Absorption spectra of H282A ( Â— ) and ALAS ( ) in the presence of 300 M ALA. Spectra were acquired at 30 C and pH 7.5. (B) pH-dependence of quinonoid interm ediate absorption upon addition of 20 mM ALA to either ALAS ( ) or H282A ( Â— ). The lines represent th eoretical curv es based on the best fit of the data to equation 8
171 FIGURE 4.8 Circular dichroism spectra of ALASand H2 82A-ligand complexes. Spectra of ALAS and H282A (A) Holoenzymes; (B) in the pr esence of 200 mM gl ycine; (C) in the presence of 300 M ALA. Spectra were recorded in 20 mM Bis-Tris with 10% glycerol, pH 7.5, at an enzyme concentration of 100 M.
172 Supporting Information Histidine-282 in 5-Aminole vulinate Synthase Affects S ubstrate Binding and Catalysis FIGURE 4.9 The A420 were globally fit to a two-step model using the simulation software Dynafit. The A420 data for the time course reaction of va rying glycine concentrations represented by circles, with the lines representing the fitted data.
173References Alexander, F. W., E. Sand meier, et al. (1994). "Evo lutionary relationships among pyridoxal-5'-phosphate-dependent enzyme s. Regio-specific alpha, beta and gamma families." Eur. J. Biochem. 219(3): 953-960. Alexeev, D., M. Alexeeva, et al. (1998). "The crystal structure of 8-amino-7oxononanoate synthase: a bacterial PL P-dependent, acyl -CoA-condensing enzyme." J. Mol. Biol. 284(2): 401-419. Astner, I., J. O. Schulze, et al. (2005). "Crystal structure of 5-aminolevulinate synthase, the first enzyme of heme biosynthesi s, and its link to XLSA in humans." Embo. J. 24(18): 3166-3177. Chen, G. C., Yang, J.T. (1977). "Two-point calibration of circular dichrometer with D10-camphorsulfonic acid." Analytical letters 10: 1195Â–1207. Chow, M. A., K. E. McElroy, et al. (2004). "N arrowing substrate specificity in a directly evolved enzyme: the A293D mutant of aspartate aminotransferase." Biochemistry 43(40): 12780-7. Christen, P. and P. K. Mehta (2001). "From cofactor to enzymes. The molecular evolution of pyridoxal-5'-phos phate-dependent enzymes." Chem. Rec. 1(6): 436447. Chu, W. C. and D. E. Metzler (1994). "E nzymatically active truncated cat brain glutamate decarboxylase: expression, purification, and absorption spectrum." Arch. Biochem. Biophys. 313(2): 287-295. DeLano, W. (2002). The PyMOL Molecular Graphics System. San Carlos, CA, De Lano Scientific. Eliot, A. C. and J. F. Kirsch (2004). "Pyridoxal phosphate enzy mes: mechanistic, structural, and evolutio nary considerations." Annu. Rev. Biochem. 73: 383-415.
174 Ferreira, G. C. and J. Gong (1995). "5-Amino levulinate synthase and the first step of heme biosynthesis." J. Bioenerg. Biomembr. 27(2): 151-159. Goldberg, J. M., R. V. Swanson, et al. (1991). "The tyrosine-225 to phenylalanine mutation of Escherichia coli aspartate aminotransferase results in an alkaline transition in the spectrophotom etric and kinetic pKa values and reduced values of both kcat and Km." Biochemistry 30(1): 305-312. Gong, J., G. A. Hunter, et al. (1998). "Aspar tate-279 in aminolevulin ate synthase affects enzyme catalysis through enhancing the function of the pyridoxal 5'-phosphate cofactor." Biochemistry 37(10): 3509-3517. Hunter, G. A. and G. C. Ferreira (1995). "A continuous spectrophotometric assay for 5aminolevulinate synthase that utilizes substrate cycling." Anal. Biochem. 226(2): 221-224. Hunter, G. A. and G. C. Fe rreira (1999). "Lysine-313 of 5-Aminolevulinate Synthase Acts as a General Base during Formation of the Quinonoid Reaction Intermediates." Biochemistry 38(38): 12526-12531. Hunter, G. A. and G. C. Ferreira (1999). "Pre -steady-state reaction of 5-aminolevulinate synthase. Evidence for a rate-determining product release." J. Biol. Chem. 274(18): 12222-12228. Kack, H., J. Sandmark, et al. (1999). "Cry stal structure of diaminopelargonic acid synthase: evolutionary relationships between pyridoxal-5'-phosphate-dependent enzymes." J. Mol. Biol. 291(4): 857-876. Kuzmic, P. (1996). "Program DYNAFIT for the analysis of enzyme kinetic data: application to HIV proteinase." Analytical Biochemistry 237(2): 260-273. Laemmli, U. K. (1970). "Cleavage of structural proteins during the assembly of the head of bacteriophage T4." Nature 227(5259): 680-685. May, A. and D. F. Bishop (1998). "The mo lecular biology and pyri doxine responsiveness of X-linked sideroblastic anaemia." Haematologica 83(1): 56-70.
175 Mehta, P. K. and P. Christen (1994). "H omology of 1-aminocyclopropane-1-carboxylate synthase, 8-amino-7-oxononanoate synthase 2-amino-6-caprolactam racemase, 2,2-dialkylglycine decar boxylase, glutamate-1-semialdehyde 2,1-aminomutase and isopenicillin-N-epimerase with aminotransferases." Biochem. Biophys. Res. Commun. 198(1): 138-43. Metzler, C. M. and D. E. Metzler (1987). "Q uantitative description of absorption spectra of a pyridoxal phosphate-dependent enzyme using lognormal distribution curves." Anal. Biochem. 166(2): 313-327. Moore, P. S., P. Dominici, et al. (1995 ). "Transaldiminati on induces coenzyme reorientation in pig kidney dopa decarboxylase." Biochimie 77(9): 724-728. Schmidt, A., J. Sivaraman, et al. (2001). "Three-dimensional structure of 2-amino-3ketobutyrate CoA ligase from Escherichia coli complexe d with a PLP-substrate intermediate: inferred reaction mechanism." Biochemistry 40(17): 5151-5160. Schneider, G., H. Kack, et al. (2000). "The manifold of vi tamin B6 dependent enzymes." Structure 8(1): R1-6. Smith, P. K., R. I. Krohn, et al. (1985). "M easurement of protein using bicinchoninic acid." Anal. Biochem. 150(1): 76-85. Tan, D., M. J. Barber, et al. (1998). "The role of tyrosine 121 in cofactor binding of 5aminolevulinate synthase." Protein Sci 7(5): 1208-13. Webster, S. P., D. Alexeev, et al. (2000). "Mechanism of 8-amino-7-oxononanoate synthase: spectroscopic, kinetic, and crystallographic studies." Biochemistry 39(3): 516-528. White, M. F., J. Vasquez, et al. (1994). "Expression of apple 1aminocyclopropane-1carboxylate synthase in Escher ichia coli: kinetic characte rization of wild-type and active-site mutant forms." Proc. Natl. Acad. Sci. U. S. A. 91(26): 12428-12432.
176 Zhang, J., A. V. Cheltsov, et al. (2005). "C onversion of 5-aminolevul inate synthase into a more active enzyme by linking the two subunits: spectroscopic and kinetic properties." Protein Sci. 14(5): 1190-1200. Zhang, J. and G. C. Ferreira (2002). "Tra nsient state kinetic investigation of 5aminolevulinate synthase reaction mechanism." J. Biol.Chem. 277(47): 4466044669. Zhou, X. and M. D. Toney (1999). "pH studies on the mechanism of the pyridoxal phosphate-dependent dialkylglycine decarboxylase." Biochemistry 38(1): 311320.
177 Chapter Five Summary and Conclusions Sequence and phylogenetic analysis of PLP-dependent -oxoamine synthases and their role in identifying residues regulating enzyme specificity The CoA-dependent acyltransferases or -oxoamine synthases constitute a small but widespread reaction-specific subfamily within the -family of PLP-dependent enzymes; it is comprised of ALAS, AONS SPT and KBL (Alexeev, Alexeeva et al., 1998; Schneider, Kack et al., 2000; Schmidt, Sivaraman et al., 2001; Astner, Schulze et al., 2005; Yard, Carter et al., 2007). Here we perform phylogeneti c analysis of the oxoamine synthase subfamily of PLP-depende nt enzymes to understand the evolutionary progression of functional sp ecialization and facilitate a better understanding of the mechanisms by which these enzymes enforce selectivity. Sequences were selected to represent the phylogenetic and taxonomic distribution of each member of the -oxoamine synthase subfamily of PLP-dependent enzymes. Using the sequence alignment of the selected sequences, the evolu tionary history was inferred and phylogenetic trees were construc ted using three methods. The data are consistent with the appearan ce of AONS function early in the evolutionary time line of the -oxoamine synthases with the subsequent development of ALAS, KBL, SPT and SPT1/2 function. This is not surprisi ng given that AONS was likely the only -oxoamine
178 synthase present in the unive rsal ancestor cell and that AONS is one of only five PLPdependent enzymes found in all biological kingdoms (Mehta and Christen, 1998). Although the phylogenetic analys is was initially performed as a tool to help characterize the significance of amino acid changes in labor atory-evolved ALAS variants that were not realized, the alignments were ut ilized to identify residues in positions that may be significant for the regulation substrat e specificity. Of the 37 residues identified, all but four were located at or adjacent to th e active site. Interestingly, one third of the residues identified by our analysis were located at key active site positions in the ALAS crystal structure from R. capsulatus and/or were previously r ecognized to be mutated in ALAS-2 in patients with x-linked sideroblastic anemia. Functional Asymmetry for Active Sites of Single Chain Homoand Chimeric Dimers of 5-Aminolevulinate Synthase and 8-Amino-7-Oxononanoate Synthase To determine whether the two active sites in ALAS/ALAS contribute equally to enzymatic activity, we characterized variants in which one of the two active sites had no measurable enzymatic activity due to a mu tation of the conserved K313 residue that binds to the cofactor. Spectral characterization of ALAS/ALASK313A and ALASK313A/ALAS revealed asymmetric cofactor e nvironments in the two active sites, which was also reflected in the disproportiona te kinetic behavior of the two sites. The pre-steady-state burst results for th e two K313A-containing variants indicate that the chemical rates were similar to ALAS/ALAS and WT ALAS and, like ALAS/ALAS and WT ALAS, were consistent with the rate-l imiting step occurring after the reaction chemistry (Zhang and Ferreira, 2002; Zhang, Cheltsov et al., 2005). In
179 ALAS, the rate-limiting step has been ascribed to a conformation change that occurs prior to the release of the ALA product (Zhang and Ferreira, 2002; Hunter, Zhang et al., 2007). The strain resulting from linking the remote Nand C-termini of two ALAS subunits appears to increase the energy barrier for produc t release at one site while decreasing the barrier at the other; that is the steady-state enzymatic activity is enhanced at one active site and hindered at the other. Consequently, the active sites contribute asymmetrically to enzyme function. Because the single-chain ALAS dimer s howed structural plasticity and had increased activity, we wondered whether the st ructural plasticity would extend to singlechain chimeras constructed from two members of the -oxoamine synthase family, ALAS and AONS. Both ALAS/AONS and AONS/ALAS chimeras had sufficient structural plasticity to achieve the conf ormations necessary to produce both enzymatic activities. Despite our initial hypothesis that the chimeric protein would create chimeric active sites with potentially novel enzy matic activities, both ALAS/AONS and AONS/ALAS appeared to function as ch imeric homodimers with functionally independent ALAS and AONS modules. None theless, the dimerization of two chimeric polypeptides into a bifuncti onal homodimer with functionally independent active sites suggests that the structural plasticity obs erved in ALAS can be extended to other members of the -oxoamine synthase family. The ALAS/AONS chimera was purified for further analysis. The fluorescence spectra exhibited by the ALAS/AONS chimer a were consistent with an enzyme exhibiting a mixture of ALAS and AONS spectroscopic charact eristics. The ALAS and
180 AONS steady-state kinetic activities were dimi nished by roughly one-half in the chimera, and the catalytic efficiencies were not impair ed. The pre-steady-state kinetic analysis for the ALAS reaction demonstrated that the r eactivity of the ALAS sites in ALAS/AONS was similar to that of ALAS, with the rate -limiting step occurri ng after catalysis. Like ALAS/ALAS, the linking of the ALAS and AONS subunits appeared to change the energy barrier associated with th e structural rearrangeme nt that occurs upon ALA formation to allow product release. It is likely that the use of the short dipeptide to link the ALAS C-terminus with the N-terminus of either ALAS or AONS introduced intermolecular strain which altered conformati onal flexibility. Our studies involving the chimeras between ALAS and AONS demonstrate that the extensive structural plasticity seen in ALAS extends to another member of the -oxoamine family, AONS. Histidine-282 in 5-Aminolevulinate Synthase Affects Substrate Binding and Catalysis A clustal sequence alignment demonstrated that a histidine re sidue was perfectly conserved in over 70 known ALAS sequences from bacteria to mammals and this histidine was also conserve among the -oxoamine synthases. In murine ALAS-2, this conserved histidine corresponds to H282. The crystal structure of Rhodobacter capsulatus ALAS reveals the existence of a hydrogen bond between the equivalent histidine and the phenolic oxygen at om of the PLP cofactor (Ast ner, Schulze et al., 2005). A series of H282 murine ALAS-2 variants were constructed to characterize the role of this conserved residue, however H282A was the only variant recoverable as a soluble enzyme. Though this residue was predicted to have multiple roles, including functioning as an acid catalyst during transaldiminati on (Webster, Alexeev et al., 2000; Zhang,
181 Cheltsov et al., 2005), positioning the PLP arom atic ring (Schmidt, Sivaraman et al., 2001), and regulating the p Ka of the imine nitrogen (Webster, Alexeev et al., 2000), we were surprised by the range of effects on the spectral, binding and kinetic properties that resulted from the replacement of H282 with alanine. The absorption and fluorescent spectra indicated that the mutation had a substantial effect on the elec tronics of the PLP cofactor and suggest that the mutation significantly alters the equilibrium of cofactor tautomeric structures to favor the ketoenamine. The steady-state kinetic parameters of the variant revealed that the loss of the H282 interaction with the phenolic oxygen impairs both glycine binding and catalysis, reducing the catalytic efficiency for glycine 4505-fold. The slow phase rate for glycine binding in H282A decreased 60% rela tive to ALAS, while the overall kinetic Kd increased 4.5 fold. The rate-determining step is also altered in the H282A and is likely associated with glycine binding (at pH below 8. 0) and not ALA release, as is observed in ALAS. The pH dependence of the log 1/ Km and acidic log kcat/ Km provides evidence that H282 stabilizes the reactive form of the inte rnal aldimine and is consistent with the change in tautomeric structures observed in the absorbance spectra. In H282A the ALAquinonoid species is markedly diminished, while Kd for ALA is minimally affected. In the R. capsulatus crystal structure, H282 hydrogen bonds with both the phenolic oxygen of the PLP cofactor and Y121. We suspected that the loss of the H282 hydrogen bonds with the cofactor and Y121 likely affect the PLP movement and orientation within the active site impacting multiple aspects of the catalytic process. Using CD spectroscopy, we determined that while the PLP microenvironment in the holoenzyme was similar in ALAS and H282A the microenvironment diverged upon the
182 binding of the 5-aminolevulinate product and gl ycine substrate. This alteration in the cofactor microenvironment would impact both co factor position and interactions with key catalytic residues. Therefore, we conclude that the multiple effects of the loss resulting from the loss of H282 results from a combinati on of direct and indire ct effects, including alterations in the protonation of the phenolic oxygen and changes to the stereoelectronic relationships between the cof actor and active site residues, through the disruption in the processional PLP positioning that normally occurs during catalysis.
183References Alexeev, D., M. Alexeeva, et al. (1998). "The crystal structure of 8-amino-7oxononanoate synthase: a bacterial PL P-dependent, acyl -CoA-condensing enzyme." J. Mol. Biol. 284(2): 401-419. Astner, I., J. O. Schulze, et al. (2005). "Crystal structure of 5-aminolevulinate synthase, the first enzyme of heme biosynthesi s, and its link to XLSA in humans." Embo. J. 24(18): 3166-3177. Hunter, G. A., J. Zhang, et al. (2007). "Tra nsient kinetic studies support refinements to the chemical and kinetic mechanisms of aminolevulinate synthase." J Biol Chem 282(32): 23025-23035. Mehta, P. K. and P. Christen (1998). "The Molecular Evolution of Pyridoxal 5-Phosphate Dependent Enzymes." Adv. Enzymol. Relat. Areas Mol. Biol. 74: 129-185. Schmidt, A., J. Sivaraman, et al. (2001). "Three-dimensional structure of 2-amino-3ketobutyrate CoA ligase from Escherichia coli complexe d with a PLP-substrate intermediate: inferred reaction mechanism." Biochemistry 40(17): 5151-5160. Schneider, G., H. Kack, et al. (2000). "The manifold of vi tamin B6 dependent enzymes." Structure 8(1): R1-6. Webster, S. P., D. Alexeev, et al. (2000). "Mechanism of 8-amino-7-oxononanoate synthase: spectroscopic, kinetic, and crystallographic studies." Biochemistry 39(3): 516-528. Yard, B. A., L. G. Carter, et al. (2007). "The structure of serine palmitoyltransferase; gateway to sphingolipid biosynthesis." J. Mol. Biol. 370(5): 870-886. Zhang, J., A. V. Cheltsov, et al. (2005). "C onversion of 5-aminolevul inate synthase into a more active enzyme by linking the two subunits: spectroscopic and kinetic properties." Protein Sci. 14(5): 1190-1200. Zhang, J. and G. C. Ferreira (2002). "Tra nsient state kinetic investigation of 5aminolevulinate synthase reaction mechanism." J. Biol.Chem. 277(47): 4466044669.
About the Author Tracy D. Turbeville was born in 1970 in Lakeland, Florida. She graduated cum laude from the University of South Florida in 1997, earning a Bachelor Â’s degree Secondary Science Education with a speci alization in Biology. As an undergraduate, Tracy was a recipient of the Woods Undergraduate Research Fellowship sponsored by the Institute for Biomolecular Science. Tracy entered the Inte rdisciplinary Ph.D. Program in Cellular and Molecular Biology, IP2CMB, and joined Dr. Gloria FerreiraÂ’s laboratory in the Department of Molecular Medicine in the Colle ge of Medicine at the University of South Florida. While a graduate student, she received a Burroughs Wellcome Tuition Bursary to attend a Canadian Bioinformatics Work shop, earning a Certificate in Protein Informatics. Tracy served as the Associat ion of Medical Scien ces Graduate Students (AMSGS) Vice President and Secretary, as we ll as acting as the AMSGS representative to the Graduate Student and Professional Student Council.