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 2200373Ka 4500
controlfield tag 001 001956485
008 080910s2006 flua sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001450
b S555h 2006
Heme biosynthesis: structure-activity studies of murine ferrochelatase /
by Zhen Shi.
[Tampa, Fla] :
University of South Florida,
x, 206 leaves ;
ill. (col.) :
ABSTRACT: Ferrochelatase catalyzes the terminal step of heme biosynthesis by inserting ferrous iron into protoporphyrin IX. The current study is aimed at understanding the structural basis of porphyrin binding and distortion in ferrochelatase-catalyzed reaction by functional analysis of a highly conserved active site loop motif. The loop was shown to contact bound porphyrin based on crystallographic and molecular modelling observations, and its role in murine ferrochelatase was assessed by random mutagenesis and steady-state kinetic analysis. To overcome the limitations of conventional kinetic assay methods for ferrochelatase, a continuous assay was developed by monitoring porphyrin fluorescence decrease using natural substrates ferrous iron and protoporphyrin IX under anaerobic conditions. For wild-type murine ferrochelatase, the assay yielded KmPPIX of 1.4 uM, KmFe2+ of 1.9 uM and kcat of 4.0 min-1 at 30 C.The results of random mutagenesis indicated that all the loop residues spanning Q248-L257 tolerated functional substitutions. While Q248, S249, G252, W256 and L257 possessed high informational content, the other five positions contained low informational content. Selected active loop variants exhibited kcat comparable to or higher than that of wild-type enzyme, while KmPPIX was increased in most variants. The kcat/KmPPIX remained largely unchanged, with the exception of a 10-fold reduction in variant K250M/V251L/W256Y. Molecular modeling of the active loop variants suggested that loop mutations resulted in alterations of the active site architecture. Distortion of porphyrin substrate, a crucial step in ferrochelatase-catalyzed metal chelation, was examined using resonance Raman spectroscopy. The results revealed that both wild-type enzyme and loop variants induced saddling of substrate protoporphyrin.Further, loop mutations generally interfered with porphyrin saddling, with the least deformation observed in variant K250M/V251L/W256Y.N-alkyl porphyrins are potent competitive inhibitors of mammalian ferrochelatase. The present study showed that while N-methyl protoporphyrin strongly inhibited the wild-type enzyme with an inhibition constant in the nanomolar range, it was less effective in inhibiting variants P255R and P255G. These results suggest that inhibitor binding may be associated with a protein conformational change mediated by P255. Wild-type ferrochelatase is a homodimeric [2Fe-2S] cluster-containing protein. Variants S249A/K250Q/V251C and S249A/K250R/G252W were found to retain enzymatic activity in the absence of FeS cluster and form active, higher order oligomers. These observations raise the possibility that FeS cluster and homodimeric organization are not essential to ferrochelatase catalysis.
Dissertation (Ph.D.)--University of South Florida, 2006.
Includes bibliographical references.
Also available online.
Advisor: Gloria Ferreira, Ph.D.
Biochemistry and Molecular Biology
t USF Electronic Theses and Dissertations.
Heme Biosynthesis: StructureActivity Studies of Murine Ferrochelatase by Zhen Shi A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Biochemist ry and Molecular Biology College of Medicine University of South Florida Major Professor: Gloria Ferreira, Ph.D. Michael Barber, Ph.D. Huntington Potter, Ph.D. George Blanck, Ph.D. Peter Medveczky, Ph.D. Kenton Rodgers, Ph.D. Date of Approval: February 10, 2006 Keywords: porphyrin, iron, resonance Rama n, continuous assay, random mutagenesis Copyright 2006, Zhen Shi
ACKNOWLEDGEMENTS I wish to express my deep gratitude to the members of my committee, Dr. Michael Barber, Dr. Huntington Potter, Dr. George Blanck, Dr. Peter Medveczky, and most of all, to Dr. Gloria Ferreria, for their cons istent guidance, understanding and support throughout the course of my graduate work. I am thankful to all the professors and colleagues in the Department of Biochemistr y, the members of Dr. Ferreira lab, as well as the lab group of Dr. John Shelnutt at the Univ ersity of New Mexico for their help and advice during the studies. I would like to acknowledge the financial support from the American Heart Association for a pre-doc toral fellowship from 2000 to 2002, and from the Institute of Biomolecular Sciences for a graduate fellowship from 1998 to 2000. I wish to express my appreciation to Ms. Ka thy Zahn and Susan Chapman at the Office of Research and Graduate Affairs for their continuous administrative assistance. I am forever grateful to my family for thei r enduring understanding and encouragement.
NOTE TO THE READER The original of this document contains color that is necessary for understanding the data. The original dissertation is on file with the USF library in Tampa, Florida.
i TABLE OF CONTENTS LIST OF TABLES iv LIST OF FIGURES v LIST OF ABBREVIATIONS vii ABSTRACT ix CHAPTER ONE INTRODUCTION 1 The importance of heme in biological systems 1 Enzymes in the heme biosynthetic pathway 7 Aminolevulinic acid synthase 7 Porphobilinogen synthase 13 Porphobilinogen deaminase 18 Uroporphyrinogen III synthase 22 Uroporphyrinogen III decarboyxlase 27 Coproporphyrinogen III oxidase 32 Protoporphyrinogen IX oxidase 37 Ferrochelatase 42 CHAPTER TWO MATERIALS AND METHODS 55 Materials 55 Experimental Methods 57 Media preparation for bacterial cultures 57 Competent cell preparation a nd bacterial transformation 58 Glycerol stock prepara tion for bacterial cells 59 Plasmid DNA purification 60 Sodium dodecyl sulfate-polyacylamide gel electrophoresis and protein concentration determination 60 Construction of a random library an d genetic selection of functional ferrochelatase loop variants 61 Large-scale purification of the wild-t ype ferrochelatase and loop variants 65 UV-visible absorbance spectra of purified ferrochelatase 67 Metal content analysis of purified ferrochelatase 68 Pyridine-hemochromogen assay 68 Continuous assay of fe rrochelatase activity 69 Steady-state kinetic analysis of the loop variants 71
ii Homology modeling of mu rine ferrochelatase 72 Resonance Raman spectroscopy of por phyrin binding to the wild-type ferrochelatase and loop variants 72 Profiling the active variants by hi gh-throughput protein purification 74 Liposomal binding assays of ferrochelatase variants 75 Inhibition assay of ferrochelatase by N -methyl protoporphyrin on agar plates 76 Quantification of N -methyl protoporphyrin binding to ferrochelatase by fluorescence quenching measurements 77 Transient kinetic analysis of ferrochelatase activity 78 Ligand binding pocket size measurement 79 Enzymatic activity of ferrochelatase in the absence of FeS cluster synthesis 80 Molecular mass assessment of purified ferrochelatase 81 Electron paramagnetic resonance spect roscopy of purified ferrochelatase 82 CHAPTER THREE RESULTS 83 Purification of recombinant ferrochelatase 83 Large-scale purification of wild-type ferrochelat ase and loop variants 83 Small-scale purification of ferrochelatase loop variants 84 Developing a continuous assay for steady-st ate kinetic analysis of ferrochelatase 87 Characterization of the functiona l ferrochelatase loop variants 92 Biological selection of th e active loop variants 92 Distribution of the functiona l amino acid substitutions 96 Steady-state kinetic analysis of the active loop variants 96 Homology modeling of wild-type murine ferrochelatase and selected loop variants 98 Interaction of ferrochelatase with mitochondrial membrane lipids 103 Resonance Raman spectroscopic analysis of porphyrin binding in the loop variants 105 Binding of substrate protopor phyrin to the variants 105 Binding of hemin to the variants 108 Binding of nickel-protopor phyrin to the variants 108 Inhibition of ferrochelatase by N -methyl protoporphyrin 112 Equilibrium binding of inhibitor to ferrochelatase 114 Kinetic pathway of inhibition 114 Size measurement of the active site pocket 119 FeS cluster and oligomeric assembly in ferrochelatase variants 121 UV-visible absorbance spectra of the variants 121 Metal content analysis 123 Dependence of enzymatic activity on FeS cluster synthesis 123 Subunit assembly of purified ferrochelatase 125 Electron paramagnetic resonance spectra of ferrochelatase 128 CHAPTER FOUR DISCUSSION 131 Continuous assay for fe rrochelatase activity 131 Characterization of the active site loop variants 135 Resonance Raman spectroscopy analysis of ferrochelatase-induced porphyrin distortion 142
iii Inhibition of ferrochelatase by N -methyl protoporphyrin 151 FeS cluster assembly and oligomer ic organization in ferrochelatase 155 REFERENCES 161 APPENDICES 205 ABOUT THE AUTHOR END PAGE
iv LIST OF TABLES Table 1. Steady-state kine tic parameters of wild-type ferrochelatase and selected loop variants 99 Table 2. Comparison of the steady-state kinetic parameters of ferrochelatase determined by various assay methods 134 Table 3. Results of simulation for the low-frequency resonance Raman spectra of ferrochelatase-bound protoporphyrin 145
v LIST OF FIGURES Figure 1. The heme biosynthetic pathway in animal cells 3 Figure 2. The heme degradati on pathway in mammalian cells 6 Figure 3. The reaction catalyzed by ferrochelatase 43 Figure 4. Random mutagenesis of the ferrochelatase activ e site loop motif and biological selection of the functional variants 62 Figure 5. The UV-visible absorbance sp ectra of purified wild-type murine ferrochelatase and selected loop variants 85 Figure 6. SDS-polyacrylamide gel elec trophoresis of purified ferrochelatase 86 Figure 7. The fluorescence spectra of protoporphyrin 88 Figure 8. Time course for the di sappearance of prot oporphyrin in the ferrochelatase-catalyzed reaction 89 Figure 9. Dependence of the initial rate of protoporphyrin consumption on ferrochelatase concentration 90 Figure 10. Determination of the steadystate kinetic parameters of wild-type murine ferrochelatase 91 Figure 11. Sequence alignment of the loop motif in ferrochelatase 93 Figure 12. Activity assessment and distribution of the number of the functional loop variants 95 Figure 13. Spectrum and frequency of amino acid substitutions in the functional loop variants 97 Figure 14. Molecular modeling of w ild-type murine ferrochelatase and selected loop variants 101
vi Figure 15. Interaction between fe rrochelatase and the mitochondrial membrane lipids 104 Figure 16. The resonance Raman spect ra of protoporphyrin incubated with ferrochelatase at a porphyrin-to -protein molar ratio of 0.1 106 Figure 17. The resonance Raman spectra of hemin incubated with ferrochelatase at a hemin-to-pro tein molar ratio of 0.1 109 Figure 18. The resonance Raman spectra of nickel-protopo rphyrin incubated with ferrochelatase at a porphyrin-toprotein molar ratio of 0.1 111 Figure 19. The structural diagram of N -methyl protoporphyrin 113 Figure 20. The intrinsic fluor escence of ferrochelatase 115 Figure 21. Binding curves generated from protein fluorescence quenching measurements following titration with N -methyl protoporphyrin 116 Figure 22. Transient kinetic analysis of the ferrochela tase-catalyzed reaction 118 Figure 23. Dependence of the rate c onstants for ferrochelatase binding on N methyl protoporphyrin concentration 120 Figure 24. The UV-visible absorbance spectra of purified wild-type ferrochelatase and variants 122 Figure 25. Dependence of ferrochelatas e activity on FeS cluster synthesis 124 Figure 26. Molecular weight assessment of purified wild-type ferrochelatase and variants by gel filtration chromatography 126 Figure 27. Molecular size determinati on of purified wild-type ferrochelatase and variants by dynamic light scattering 127 Figure 28. EPR spectra of purified wild -type ferrochelatase and variants 129 Figure 29. Temperature-dependence of th e EPR signal intensity for a purified ferrochelatase variant 130 Figure 30. Structural fe atures of the murine fe rrochelatase models 141 Figure 31. Three-dimensional views of the porphyrin-binding cleft in ferrochelatase 149 Figure 32. Position of the FeS cluster relative to the active site in ferrochelatase 156
vii LIST OF ABBREVIATIONS AIP, acute intermittent porphyria ALA, -aminolevulinic acid ALAS, -aminolevulinic acid synthase AONS, 8-amino-7-oxononanoate synthase Biotin-DHPE, N -(biotinoyl)-1,2-dihexadecanoylsn-glycero-3-phosphoethanolamine triethylammonium salt CEP, congenital erythropoietic porphyria CPO, coproporphyrinogen III oxidase ECL, enhanced chemiluminescence EPP, erythropoietic protoporphyria EPR, electron paramagnetic resonance FAD, flavin adenine dinucleotide HCP, hereditary coproporphyria Hemin, iron(III) protoporphyrin IX HMB, hydroxymethylbilane IRE, iron-responsive element IRP, iron regulatory protein KBL, 2-amino-3-oxobutyrate CoA ligase
viii NiPP, Ni(II) protoporphyrin N -MeMP, N -methyl mesoporphyrin NMPP, N -methyl protoporphyrin PBG, porphobilinogen PBGS, porphobilinogen synthase PBGD, porphobilinogen deaminase PCT, porphyria cutanea tarda PLP, pyridoxal 5-phosphate PPIX, protoporphyrin IX PPO, protoporphyrinogen IX oxidase RMSD, root mean square deviation RR, resonance Raman SDS-PAGE, sodium dodecyl sulfatepolyacylamide gel electrophoresis Tween-80, polyethylene glycol sorbitan monooleate Tween-20, polyethylene glycol sorbitan monolaurate TBS, Tris-buffered saline UROD, uroporphyrinogen III decarboyxlase UROS, uroporphyrinogen III synthase VP, variegate porphyria XLSA, X-linked sideroblastic anemia
ix Heme Biosynthesis: StructureActivity Studies of Murine Ferrochelatase Zhen Shi ABSTRACT Ferrochelatase catalyzes the terminal st ep of heme biosynthesis by inserting ferrous iron into protoporphyr in IX. The current study is aimed at understanding the structural basis of porphyrin binding and distortion in ferr ochelatase-catalyzed reaction by functional analysis of a hi ghly conserved active site l oop motif. The loop was shown to contact bound porphyrin based on crys tallographic and molecular modelling observations, and its role in murine ferroch elatase was assessed by random mutagenesis and steady-state kinetic analysis. To ove rcome the limitations of conventional kinetic assay methods for ferrochelatase, a con tinuous assay was developed by monitoring porphyrin fluorescence decrease using natural substrates fe rrous iron and protoporphyrin IX under anaerobic conditions. For wild-type murine ferrochelatase, the assay yielded Km PPIX of 1.4 M, Km Fe2+ of 1.9 M and kcat of 4.0 min-1 at 30 oC. The results of random mutagenesis indicated that a ll the loop residues spanning Q248-L257 tolerated functional substitutions. While Q248, S249, G252, W256 and L257 possessed high informational content, the other five positions contained lo w informational content. Selected active loop variants exhibited kcat comparable to or higher than that of wild-type enzyme, while Km PPIX was increased in most variants. The kcat/ Km PPIX remained largely unchanged, with
x the exception of a 10-fold reduction in variant K250M/V251L/W256Y. Molecular modeling of the active loop varian ts suggested that loop mutati ons resulted in alterations of the active site architecture. Distortion of porphyrin substrate, a crucial step in ferrochelatase-catalyzed me tal chelation, was examined using resonance Raman spectroscopy. The results revealed that both wild-type enzyme and loop variants induced saddling of substrate protoporphyrin. Further, loop mutations generally interfered with porphyrin saddling, with the least deformation observed in variant K250M/V251L/W256Y. N -alkyl porphyrins are potent comp etitive inhibitors of mammalian ferrochelatase. The present study showed that while N -methyl protoporphyrin strongly inhibited the wild-type enzyme with an inhibi tion constant in the nanomolar range, it was less effective in inhibiting variants P255R and P255G. These results suggest that inhibitor binding may be associated with a protein conformational change mediated by P255. Wild-type ferrochelatase is a homodime ric [2Fe-2S] cluster-containing protein. Variants S249A/K250Q/V251C and S249A/K 250R/G252W were found to retain enzymatic activity in the absence of FeS cluster and form active, higher order oligomers. These observations raise the possibility th at FeS cluster and hom odimeric organization are not essential to ferrochelatase catalysis.
1 Chapter One Introduction The importance of heme in biological systems Heme, an iron-containing macrocyclic te trapyrrole compound, serves a wide range of important physiological func tions in many organisms including oxygen transport, electron transfer, detoxification, metabolite produ ction, biological sensing and signal transduction. In hemoglobin and myoglob in the heme prosthetic group acts as an oxygen carrier which is crucial to gas tran sport and exchange in blood and tissues (Hardison 1996; Wittenberg and Wittenberg 2003) Interestingly, hexacoordinated heme has been found in the newly identified vert ebrate globins includ ing neuronal specific neuroglobin and ubiquitously expressed cyto globin (Pesce et al. 2002). The primary function of neuroglobin has been thought to promote heme-mediated O2 diffusion into mitochondria, which confers protection agai nst hypoxic-ischemic attacks in neurons (Hankeln et al. 2005). In various forms of cytochromes associated with the electron transport chain, heme provides a redox-active metal center and thereby promotes electron transfer in aerobic respira tion and photosynthesis to incr ease energy production via ATP synthesis (Rich 2003; Gray and Winkler 2005). Enzymatic activities of a diverse set of heme monooxygenases including cytochrome P 450s and nitric oxide synthase depend
2 upon protein interaction with th e ubiquitous cofactor heme (Poulos 2005). By supplying catalytically active iron, heme plays an essent ial role in mediating electron transfer and O2 activation to facilitate s ubstrate oxidation (Poulos 2005). In an increasing number of proteins the heme moiety has been shown to pl ay a central role in biological sensing of small molecules including O2, carbon monoxide (CO) and nitr ic oxide (NO), which leads to modulation of protein activity (Rodgers 1999; Gilles-Gonzalez and Gonzalez 2005). For instance, for mammalian neuronal hete rodimeric transcription factor NPAS2BMAL1, CO binding to heme coordinated in the PAS domain of NPAS2 has been found to inhibit DNA binding of the transactivation domain requ ired for circadian rhythmn regulation (Dioum et al. 2002; Uchida et al. 2005). Recently, heme has been implicated in regulating the activity of large-conduc tance calcium-dependent potassium (BKCa) channels, which control cerebral arterial smoot h muscle excitability (Tang et al. 2003). While unliganded heme binds to the prot ein via a conserved heme-binding motif (CXXCH) and blocks channel activity, CO bindi ng to heme relieved this inhibition and thus allowing channel activation and possibly le ading to vasodilation (Tang et al. 2003). In all heme-synthesizing organisms, the heme biosynthetic pathway consists of a series of enzymes encoded by the nuclear genes (Dailey 1997). In eukaryotes, the enzymes and reaction intermediates of heme biosynthesis are distributed in both the mitochondria and cytosol (Figure 1). The first committed compound in the pathway is aminolevulinic acid (ALA) (Dailey 1997). In animals, fungi and some photosynthetic bacteria ( -proteobacteria), ALA is generated via a four-carbon pathway (Shemin pathway) from the condensation of gl ycine and succinyl-CoA catalyzed by aminolevulinic acid synthase (ALAS) (Shemi n and Russell 1953; Avissar et al. 1989). In
3 Figure 1. The heme biosynthetic pathway in animal cells. The enzymes and the subcellular distribution of the reaction intermediates in the pathway are shown in the diagram. Adapted from (Dailey 1997). Cytosol Mitochondrion Porphobilinogen N H COOH COOH H2NHN HN NH NH COOH HOOC HOOC HOOC HOOC COOH HOOC HOOC HO A B C DH y drox y meth y lbilaneHN HN NH NH COOH HOOC HOOC HOOC HOOC COOH HOOC COOH A B D CUroporphyrinogen III HN HN NH NH COOH HOOC HOOC COOH A B D CCoproporphyrinogen III HN HN NH NH HOOC COOH DC B AProtoporphyrinogen IX N HN NH N HOOC COOH A B D CProtoporphyrin IX N N N NHOOC COOHFeA D B CProtoheme Uroporphyrinogen III synthase Porphobilinogen deaminase Uroporphyrinogen III decarboxylase Coproporphyrinogen III oxidase Protoporphyrinogen IX oxidase Ferrochelatase Aminolevulinic acid synthase Porphobilinogen synthase HOOC H2C C H2 SCoA O Succin y l-Co A H2N C H2COOHGlycine H2N CH2C CH2H2C COOH O -Aminolevulinic acid
4 plants, algae, archaea and most eubacteria ALA is synthesized using a five-carbon pathway which starts from the C5-skeleton of glutamate and involves the action of three enzymes (Beale et al. 1975; Kannangara et al. 1988; Frankenberg et al. 2003). Once ALA is formed, the subsequent reactions in the pathway are common to all organisms (Frankenberg et al. 2003). The second st ep involves the condensation of two ALA molecules to form the monopyrrole porphobili nogen (PBG) catalyzed by PBG synthase. In the third step, a linear tetrapyrrole, hydr oxymethylbilane (HMB), is formed by headto-tail condensation of four PBG molecules with concomitant deamination using PBG deaminase. HMB is immediately cyclized by the 4th enzyme uroporphyrinogen III synthase to form uroporphyrinogen III, the first tetrapyrrole macrocycle in the pathway. In the 5th step, uroporphyrinogen decarboxylase cataly zes the stepwise decarboxylation of the four acetyl side chains in uroporphyrin ogen III to methyl groups to generate coproporphyrinogen III. Oxid ative decarboxylation of coproporphyrinogen III to protoporphyrinogen IX is catalyzed by the 6th enzyme coproporphyrinogen IX oxidase. Subsequently, protoporphyri nogen IX is oxidized to pr otoporphyrin IX by the 7th enzyme protoporphyrinogen IX oxidase. In the last step, ferrochelatas e catalyzes the insertion of iron into the protoporphyrin macrocycle to form protoheme. In mammals, heme biosynthesis takes place primarily in the liver and bone marrow erythroid cells (Ponka 1997). Genetic defects in the first enzyme ALAS cause aberrant iron metabolism and associated disorders (Alcindor and Bridges 2002). Enzymatic deficiency in any one of the steps after ALAS results in accumulation of porphyrin s or porphyrin precursors and lead to various forms of a disease st ate known as porphyria (Kauppinen 2005).
5 In animals, plants and some bacter ia, heme catabolism involves oxidative degradation catalyzed by heme oxygenase to yi eld biliverdin, fe rrous iron and CO (Figure 2). Mammalian heme oxygenase is as sociated with biliverdin reductase in a microsomal enzyme system, which further re duces biliverdin to bilirubin (Kutty and Maines 1981; Maines and Gibbs 2005). Th e reaction catalyzed by heme oxygenase uses molecular oxygen and requires reducing equi valents from NADPH coupled with a redox partner, which is served by cytochrome-P450 reductase in ma mmals (Colas and Ortiz de Montellano 2003) and ferredoxin in some bacteria and plan ts (Cornejo et al. 1998; Wegele et al. 2004). In mammals, heme degradation primarily o ccurs in the reticuloendothelial system and hepatic cells. Heme oxygenase plays a major role in the physiological breakdown of hemoglobin and other hemoproteins to bile pi gments (Tenhunen et al. 1969; Maines and Gibbs 2005). While the primary catabolite bi lirubin is water-insoluble and potentially toxic, conjugation of bilirubin with gl ucuronic acid improves water solubility and facilitates excretion in to bile (Tukey and Strassburg 2000). In patients with excessive red cell lysis and heptobiliary diseases, bilirubi n and its precursors accumulate in the serum leading to hyperbilirubinemia, while skin deposition of excess bilirubin gives the appearance of jaundice (Kapla n et al. 2003). In spite of its putative cytotoxicity, numerous studies have shown that bilirubin ca n elicit potent antioxida nt effects providing cellular protection against tissue injury, oxi dative stress, inflammation and apoptosis (Otterbein and Choi 2000; Se dlak and Snyder 2004; Miralem et al. 2005). The gaseous catabolic product CO has gained much atten tion as a versatile pa racrine and autocrine messenger and neurotransmitter mediating diverse physiological responses such as
6 N N N N HOO C C OOH Fe N N N N H O O C C O O H Fe OH N N N N O+ H O O C C O O H Fe N H N H N H COOH N H HOOC O O N H N H N H COOH N H HOOC O O H H Heme -meso-hydroxyheme Verdoeme Biliverdin Bilirubin Heme oxygenase Fe 2 +NADPH/O2 Heme oxygenase NADPH/O2 H2O Heme oxygenase O 2 CO Biliverdin reductase NADPH NADP+ Figure 2. The heme degradation pa thway in mammalian cells. The enzymes and the reaction intermediates in the pathway are shown in the diagram. Adapted from (Poulos 2005).
7 muscle relaxation, blood vessel dilation, pain perception and an ti-inflammation (Morse et al. 2002; Boehning and Snyder 2003). Heme o xygenase has also been found to play an important role in mammalian iron homeostasis (Poss and Tonegawa 1997). Iron released from heme is thought to exit cells via a microsomal Fe-ATPase pump to bind to transferrin and to ente r recycling route for instance towa rds hemoglobin synthesis (Ferris et al. 1999; Hentze et al. 2004) In pathogenic bacteria heme oxygenase allows the organism to obtain iron from acquired host heme to enable infection under low iron conditions and protects bacteria against heme toxicity (Schmitt 1997; Zhu et al. 2000). In the past decade, the structures of all the enzymes in the heme biosynthesis and degradation pathways have been determined. These high resolution crystal structures provide many opportunities to st udy the structural basis of the catalytic mechanism of each enzyme and genetic diseases resulted fr om enzymatic deficiency. In the following sections, recent discoveries are described in terms of structure, function, regulation and genetic disorder associated w ith each of the heme biosyntheti c enzymes with an emphasis on ferrochelatase, the focus of this dissertation. Enzymes in the heme biosynthetic pathway Aminolevulinic acid synthase -Aminolevulinic acid synthase (ALAS) (E C 188.8.131.52) is the first enzyme in the heme biosynthetic pathway in animal cells, fungi and -proteobacteria including Rhodobacter Agrobacterium Rhizobium and Rickettsia species (Shemin and Russell 1953; Avissar et al. 1989). ALAS catalyzes the condensation of L-glycine and succinyl-
8 CoA (SCoA) to form -aminolevulinic acid (ALA), CO2 and coenzyme A (Gibson et al. 1958; Shemin and Kikuchi 1958). The enzymatic reaction generally requ ires the cofactor pyridoxal 5-phosphate (PLP ) (Nandi 1978b; 1978a; Davies and Neuberger 1979). Bacterial ALAS is a cytosolic soluble pr otein consisting of ~400 residues with a monomeric molecular mass of ~45-60 kDa (Warnick and Burnham 1971; Tait 1973; Nandi and Shemin 1977; Davies and Neuberg er 1979; Hornberger et al. 1990). In eukaryotic cells, ALAS is nuclear-encoded, sy nthesized in the cytoplasm as a precursor and translocated to the mitochondrial matrix (Whiting and Elliott 1972; Ades and Harpe 1981; Urban-Grimal et al. 1986; Ades and Stevens 1988; Munakata et al. 1993). The transport usually involves an N-terminal l eader sequence of ~60 residues, which is cleaved proteolytically in the mitochondria (Bawden et al. 1987; Cox et al. 1991). Additionally, a second targeting signal invo lving an internal hydrophobic sequence has also been suggested (Volland and Urban-Grim al 1988). The mature form of animal and fungal ALAS is typically present as a PLP-containing homodimer with a subunit molecular mass of ~50-65 kDa (Whiting an d Granick 1976; Volland and Felix 1984; Ades and Friedland 1988; Ferreira and Dailey 1993). In mammals, there are two forms of ALAS encoded by separate genes (Bishop et al. 1981; Watanabe et al. 1983; Yamamoto et al. 1986; Riddle et al. 1989; Bishop 1990). The housekeeping form, ALAS-1, is present in all nonerythroid cells (Watanabe et al. 1983; Yamamoto et al. 1986; Srivastava et al. 1988; Riddle et al. 1989). ALAS-1 transcription is upregulated by various hor mones, porphyrinogenic drugs and xenobiotics (de Verneuil et al. 1983a; Sriv astava et al. 1988; Podvinec et al. 2004; Handschin et al. 2005), whereas product ALA inhibits ALAS-1 e xpression (Granick 1966; Marks et al.
9 1988; Srivastava et al. 1988). Notably, m itochondrial import and maturation of ALAS-1 are inhibited by very low levels of intracel lular heme (Kikuchi and Hayashi 1981; Ades 1983; Ades and Stevens 1988; Fujita et al. 1991). The inhibitory effect is mediated by heme binding to three heme regulatory motifs (HRM) characterized by the Cys-Pro dipeptide in the ALAS-1 N-terminal region (Zhang and Guarente 1995; Munakata et al. 2004; Dailey et al. 2005). The second isoform, ALAS-2, is only produc ed in developing erythrocytes and its expression is subject to transcriptiona l and translational regulation (Cox et al. 1991; Sadlon et al. 1999). Ir on has long been known to stimulate ALAS-2 translation via an iron-responsive element (IRE), a stem-loop structure found in the 5untranslated region of all the ALAS-2 transc ripts (Cox et al. 1991; Dandekar et al. 1991; Melefors et al. 1993). While binding of the iron regulatory pr oteins (IRP1 and IRP2) to the IRE typically block ALAS-2 translation, this repression is relieve d at high levels of intracellular iron, and IRP1-mediated transla tional regulation has been shown to be associated with FeS cluster assembly (Melefors et al. 1993; Sadlon et al. 1999; Cooperman et al. 2005; Wingert et al. 2005). In spite of major differences in their regulatory mechanisms, both ALAS-1 and ALAS-2 share a highly homologous catalytic co re, which consists of the C-terminal twothirds of the enzyme (Riddle et al. 1989; C ox et al. 1991; Munakata et al. 1993). In fact, the catalytic domain of all ALASs is remark ably conserved throughout evolution with an overall sequence identity of ~40 % from b acterial to human enzymes (Duncan et al. 1999; Astner et al. 2005). ALAS belongs to a family of PLP-dependent enzymes, which include 8-amino-7-oxononanoate synthase ( AONS), 2-amino-3-oxobutyrate CoA ligase (KBL) and serine palmitoyl tr ansferase (Alexander et al. 1994; Grishin et al. 1995).
10 These enzymes are members of the -oxoamine synthase family of PLP-dependent enzymes, which catalyze C-C bond formation or cleavage between an amino acid and an acyl-CoA substrate with release of coenzy me A and formation of an enzyme-bound 2amino-3-ketoacid intermediate (Alexander et al. 1994; Grishin et al. 1995; John 1995). These proteins share a large number of c onserved residues, exhibit homologous folding patterns and active site geometries, and empl oy similar catalytic mechanisms (Alexander et al. 1994; Grishin et al. 1995; Alexeev et al. 1998). For ALAS, the proposed catalytic path way involves the initial formation of a Schiff-base between the enzyme and cofactor PLP, followed by a transimination reaction between the enzyme-PLP complex and glycine, subsequent condensat ion of glycine with succinyl-CoA, and decarboxylation of the glyc inyl carboxyl group results in release of ALA from the enzyme (Scholnick et al. 1972; Nandi 1978c). Kinetic analysis indicated that the enzymatic reaction is initiated upon binding of the first substrate glycine to ALAS (Fanica-Gaignier and Clement-Metr al 1973; Nandi 1978c). Glycine forms a Schiff-base linkage with PLP to generate an external aldimine (Scholnick et al. 1972; Nandi 1978c; Ferreira et al. 1995; Hunter and Ferreira 1999a). Subsequently the glycineALAS external aldimine is deprotonated on glycine to yield a transient quinonoid intermediate (Nandi 1978c; Hunter and Ferreira 1999a). This complex condenses with succinyl-CoA to convert aldimine to -amino-ketoadipate (Hunter and Ferreira 1999b). In the subsequent decarboxylati on reaction, rearrangement of the -amino-ketoadipate results in the release of CO2 and a second quinonoid intermediate (Hunter and Ferreira 1999b). Protonation of the second quinonoid inte rmediate yields a third external ALAaldimine (Hunter and Ferreira 1999b). In th e last step, ALA is released from the
11 aldimine and the holoenzyme is regenerate d (Nandi 1978c; Hunter and Ferreira 1999b). The rate-determining step appears to be product rele ase, or a conformational change of the enzyme associated with ALA re lease (Hunter and Ferreira 1999b). This catalytic model is consistent with the recent X-ray crys tal structures of Rhodobacter capsulatus ALAS holoenzyme and substrate-enzyme complexes (ALASrc) (Astner et al. 2005). As expected, ALASrc bears close resemblance to the crystal structures of AONS and KBL (Alexeev et al. 1998; Schmidt et al. 2001) and to a R. spheroides ALAS model previously constructe d by homology modeling using AONS and KBL as templates (Shoolingin-Jordan et al 2003b). ALAS is a symmetrical homodimer with each monomer folded in to 3 domains, all of which contribute to the dimeric assembly. The largest domain is the central catalytic core consisti ng of a seven-stranded mostly parallel -sheet covered on both sides by nine -helices in alternating / motifs. The catalytic core provides the most extensive contribution to the dimer interface. In the vicinity of the dimer interface lies the C-terminal domain which comprises three-stranded antiparallel -sheet and three -helices. A small N-terminal domain containing a threestranded antiparallel -sheet and an -helix is located more distantly from the catalytic core. The PLP cofactor is bound to the catalytic domain near the dimer interface. The active site is proposed to be located in an interdomain cavity between the juxtaposed catalytic domains and the C-terminal domain of the PLP-containing monomer (Astner et al. 2005). Glycine binding occurs in this pocket, whereas succinyl-CoA binding occurs in a channel which connects the active site to the protein exterior (Astner et al. 2005). The active site pocket is enriched with ma ny conserved residues, some of which have been found previously to be necessary for catalysis (Astner et al. 2005).
12 As indicated earlier (Nandi 1978c; Ferre ira et al. 1993; Ferreira et al. 1995; Hunter and Ferreira 1999a), th e initial enzyme-cofactor complex involves a Schiff-base between PLP and an invariant, catalytically -essential lysine resi due. PLP-binding is stabilized by residues from both monomers (Astner et al. 2005). Consistent with a common observation among many PLP-dependent enzymes, the electron withdrawl capacity of PLP appears to be enhanced by hydrogen-bonding betw een its pyridinium ring and an adjacent conserved aspartate residu e; this hydrogen-bond facilitates PLP to be held in the active site while the initial PLP-ly sine Schiff-base linkage is replaced with a new Schiff-base between PLP and substrate glycine (Gong et al. 1998; Hunter and Ferreira 1999a). In the enzyme-glycine complex, initial binding and orientation of glycine is shown to involve a salt bridge between the glycine carboxylate group and a conserved arginine, which is consistent with a previous study (Tan et al. 1998). Succinyl-CoA binds to a hydrophobic pocket near the protein surface in an extended conformation, and a conserved glycine-rich interface loop appears to be crucial in positioning the succinate carboxylate group as proposed previously (Gong et al. 1996). Both succinyl-CoA and glycine are positioned in close proximity in the active site, which makes it feasible for the nucleophilic attack of the succinyl-CoA CS1 by the glycine C atom in the first quinonoid intermediate to generate a new C C bond (Astner et al. 2005). In mammals, enzymatic deficiency in the erythroid isozyme ALAS-2 leads to Xlinked sideroblastic anemia (X LSA), an X chromosome-linked hypochromic and microcytic anemia characterized by the presen ce of ring sideroblasts in the bone marrow (Cotter et al. 1992; Bottomley et al. 1995). Ring sideroblasts are er ythroblasts containing iron-laden mitochondria which are generated in response to disrupted heme synthesis
13 (Alcindor and Bridges 2002). In some patients progressive toxic accumulation of iron in many tissues can have fatal consequences (Cot ter et al. 1999). To date more than 40 mutations in human ALAS-2 have been know n to cause XLSA (Astner et al. 2005). They include frameshift mutations, nonsense mutations and a large number of missense mutations involving nearly 40 highly conserve d residues (Cotter et al. 1999; ShoolinginJordan et al. 2003b; Astner et al. 2005) Mi ssense mutations are primarily located in the catalytic domain and appear to directly in terfere with PLP and s ubstrate binding (Cotter et al. 1999; Shoolingin-Jordan et al. 2003b; Astner et al. 2005). Some substitutions are found in the hydrophobic core or on the enzyme su rface and seem likely to destabilize the protein fold (Shoolingin-Jordan et al. 2003b; Astner et al. 2005). Pyridoxine (vitamin B6) supplementation has been shown to be an overall effectiv e therapy for XLSA particularly in patients with ALAS-2 mutati ons near the catalytic center (Cotter et al. 1999; Alcindor and Bridges 2002; Shoolingin-Jordan et al. 2003b) The treatment can be combined with phlebotomy or iron chelatio n with desferrioxamine to correct ironoverloading and improve pyridoxine responsiv eness (Cotter et al. 1999; Alcindor and Bridges 2002). Porphobilinogen synthase The first step common to all tetrapyrrole biosynthetic pathways is catalyzed by porphobilinogen synthase (PBGS), also known as ALA dehydratase (ALAD) (EC 184.108.40.206). PBGS catalyzes the condensation of two molecules of -aminolevulinic acid (ALA) to form the monopyrrole porphobilinogen (PBG) (Nandi et al. 1968). PBG is an essential metabolic building block which is incorporated into various tetrapyrrole
14 compounds including heme, chlorophyll and cobalamin (Warren et al. 1998). From bacteria to mammals, the amino acid sequenc e of PBGS is hi ghly conserved and comprised of ~330 residues (Jaffe 2003). PB GS is a cytosolic enzyme and typically exists in a large homo-octame ric assembly with a subunit molecular mass of ~35-40 kDa (Wu et al. 1974; Anderson and Desnick 1979; Gibbs et al. 1985; Sp encer and ShoolinginJordan 1993). Metal ions are often found in the active site of PBGS and are commonly required for enzymatic activity (Jaffe 2003), although metal-independ ent PBGS has also been identified in some bacterial species (Bollivar et al. 2004). The condensation reaction catal yzed by PBGS is initiate d with an ordered binding of two ALA molecules in the ac tive site (Jaffe 2004b). The first ALA molecule binds to the P-site in PBGS, which involves formati on of Schiff-base linkage of ALA with an active site lysine (Gibbs and ShoolinginJordan 1986; Jaffe 2004a). The first ALA contributes the propionate side chain and the pyrrole nitrogen of PBG (ShoolinginJordan and Gibbs 1985; Jaffe 2004a). The seco nd ALA binds to the A-site, giving rise to the acetate side chain and the amino nitr ogen of PBG (Shoolingin-Jordan and Gibbs 1985; Jaffe 2004a). The rate-limiting step of the overall reaction appears to be product release or a conformational change of the enzyme that is coupled to product release (Jaffe 2004a). Interestingly, product formation has been found to occur with a half-of-the-sites reactivity (Kundrat et al. 2003; Jaffe 2004a). Metal ions typically serve as cofactors in the PBGS-catalyzed reaction, and metal binding specificity is largely phylogenetically dependent (Jaffe 2004a). In animals, fungi, archaea and some bacteria PBGS contains catalytic Zn2+ whose main function is to enhance A-site binding of the second AL A substrate molecule possibly by acting as a
15 Lewis acid or electrophile to react with the bound first AL A (Jaffe and Hanes 1986; Jaffe 1995; Spencer and Shoolingin-Jordan 1995; Ersk ine et al. 2001b; Erskine et al. 2003). The catalytic Zn2+ is coordinated by three cysteine ligands in a conserved C-X-C-X7-C motif (Jaffe 2003). In contrast, Mg2+ has been found in PBGS from photosynthetic plants, algae and some bacteria (Boese et al. 1991; Jaffe 2003). At least some of the bound Mg2+ appear to be allosteric rather than catalytic (Boese et al. 1991; Jaffe 2003). Additionally, both Zn2+ and Mg2+ have been found to be associated with PBGS from many plants and bacteria (Jaffe 2003). The structure of PBGS has been determin ed for the enzyme from a variety of organisms (Jaffe 2004b). To date the crystal structures are availabl e for the enzyme from human (Breinig et al. 2003), yeast S. cerevisiae (Erskine et al. 1997; Erskine et al. 2001b; Erskine et al. 2003; Erskine et al. 2005), and a number of bacteria including E. coli (Erskine et al. 1999; Kervinen et al. 2001), Pseudomonas aeruginosa (Frankenberg et al. 1999; Frere et al. 2002), and Chlorobium vibrioforme (Coates et al. 2004). PBGS, either in an apo form or with bound inhibitor or reaction intermediate, exhibits homologous overall fold and active site structure (J affe 2004b). The holoenzyme is typically assembled into a homo-octamer (Erskine et al 1997; Erskine et al. 1999; Frankenberg et al. 1999; Kervinen et al. 2001; Coates et al. 2004). The eight monome rs are organized in 422 or D4 point group symmetry with the central cavity in each subunit oriented towards the solvent. The holoenzyme is tightly packed into four pairs of hugging dimers, in which one monomer extends an N-terminal arm to wrap around the adjacent monomer (Erskine et al. 1997; Erskine et al. 1999; Fran kenberg et al. 1999; Kervinen et al. 2001; Coates et al. 2004). Ea ch monomer adopts a ( / )8 barrel fold characteristic of the triose
16 phosphate isomerase (TIM) barrel proteins in the adolase su perfamily. The active site is located in a deep cavity in the center of th e TIM barrel, while an extended loop at the exposed end of TIM barrel forms a mobile li d to gate solvent access. Many conserved residues are clustered in the active site cavity and have been shown to mediate substrate binding, catalysis and active site conformationa l change (Erskine et al. 1997; Erskine et al. 1999; Frankenberg et al. 1999; Kervinen et al. 2001; Coates et al. 2004). Crystal structural analysis has provided further details in the PBGS reaction mechanism (Jaffe 2004b). The in itial step of catal ysis involves Schiff-base formation of the P-site ALA substrate with an invariant lysine located at the bottom of active site cavity (Gibbs and Shoolingin-Jordan 1986; Er skine et al. 1997; Er skine et al. 1999; Erskine et al. 2001b; Frere et al. 2002; Coates et al. 2004). This first step is conserved for all of the PBGS investigated to date and does not require metal ions (Jaffe 2004b). The next step involves the second ALA binding to A-site, which c onsists of a solvent-filled hydrophilic pocket (Erskine et al 1999; Erskine et al. 2001b). It is possible that this ALA forms a second Schiff-base linkage with another invariant lysine in the active site (Erskine et al. 2001a; Kervinen et al. 2001). In Zn-dependent PBGS, this step is facilitated by the presence of solvent-coordinated catalytic Zn2+ (Erskine et al. 1999; Erskine et al. 2001a; Erskine et al. 2001b; Kervinen et al. 2001) It has been proposed that active site closure is promoted by th e second ALA binding (Frankenberg et al. 1999; Kervinen et al. 2001). The final steps have been shown to proceed via a five-membered ring intermediate held at the active site, and PBG formation and release is facilitated by added ALA substrate (Erskine et al. 2003). Alternatively, a reciprocating motion model suggests that a covalently bound almost-product intermedia te is transferred to the
17 adjacent active site, which opens the lid and re leases a PBG molecule (Jaffe 2004b). This latter hypothesis is consistent with the obs ervation of half-site reactivity in PBGS (Kundrat et al. 2003). Important features of metal-binding in PBGS have also been revealed by structural analysis (Jaffe 2003). While Zn2+-dependent PBGS re quires active site Zn2+ in catalysis, Mg-containing PBGS uses Mg2+ for allosteric regulation of enzymatic activity (Jaffe 2003). The binding of Mg2+ has been shown to occur at a remote location near the active site flap and is likely to play a ro le in stabilizing the closed lid conformation (Frankenberg et al. 1999; Coat es et al. 2004). Interestin gly, an analogous allosteric Mg2+-binding site is occupied by a second Zn2+ in E. coli PBGS, which provides a structural explanation fo r the observation that Mg2+ stimulates E. coli PBGS activity (Mitchell and Jaffe 1993; Spencer and Shoo lingin-Jordan 1993; Er skine et al. 1999; Kervinen et al. 2001). Mamm alian PBGS has long been known as a target for lead poisoning (Warren et al. 1998). Human PB GS activity is inhibited by very low concentrations of Pb2+ (Simons 1995) (Jaffe 2001). Struct ural analysis suggests that this inhibition is caused by Pb2+ replacement of the catalytic ally important, tri-cysteinecoordinated Zn2+, which leads to alteration of the ac tive site environment and thereby interferes with substrate binding a nd catalysis (Jaffe et al. 2001). Hereditary PBGS deficiency, or Doss por phyria, is the most uncommon form of all the porphyrias (Doss et al. 1979). The dis ease is autosomal recessive and fewer than 10 cases have been reported to date (Maruno et al. 2001; Do ss et al. 2004). Doss porphyria belongs to the acute he patic porphyrias with character istic clinical features of acute neurovisceral attacks and neuropa thy (Nordmann and Puy 2002). Patients
18 generally show elevated ur inary excretion of ALA and coproporphyrin, which can be used for diagnosis (Nordmann and Puy 2002; Doss et al. 2004). Only ~10 mutations in PBGS have been associated with Doss porphyria including predominantly missense mutations and a few frameshift mutations (Maruno et al. 2 001; Doss et al. 2004). Most patients are compound heterozygous for the gene tic defects (Maruno et al. 2001; Doss et al. 2004). Missense mutations primarily occur near the catalytic zinc-binding site or substrate-binding site and resu lt in markedly diminished enzymatic activity (Maruno et al. 2001). A few substitutions are located di stant from the active site and appear to disrupt protein structure and stability (Mar uno et al. 2001). Intravenous heme arginate infusion has been shown to be an effective th erapy for the disease l eading to reduction of urinary ALA and porphyrins and correction of polyneuropathy (Doss et al. 2004). Porphobilinogen deaminase Porphobilinogen deaminase (PBGD), also known as hydroxymethylbilane synthase (HMBS) (E.C.220.127.116.11) catalyzes the st epwise head-to-tail condensation of four porphobilinogens to form the linear tetrap yrrole, hydroxymethylbilane (HMB) or preuroporphyrinogen (Bogorad 1958b). PBGD contai ns ~310-370 residues and exhibits an exceedingly high degree of sequence conservati on with ~40% sequence identity between the E. coli and human sequences (Brownlie et al. 1994; Louie et al. 1996). The protein is a monomer of ~34-39 kDa (Anderson and Desn ick 1980; Raich et al. 1986; Grandchamp et al. 1987; Louie et al. 1996). In plants PB GD is localized in th e plastid chloroplast (Bogorad 1958a; Witty et al. 1996). Human PBGD is encoded by a single gene, which contains two distinct promoters to gene rate housekeeping and erythroid-specific
19 transcripts by alternative splic ing (Grandchamp et al. 1987). This gives rise to two isoforms of PBGD, i.e. the erythroid form a nd the slightly larger non-erythropoietic form (Grandchamp et al. 1987). An interesting feature of the PBGD catalytic mechanism is that the reaction is primed using a dipyrromethane cofactor that the enzyme assembles from two molecules of porphobilinogen (Hart et al. 1987; Shooli ngin-Jordan and Warren 1987; Warren and Shoolingin-Jordan 1988). The cofactor is covale ntly linked to a cysteine residue (Hart et al. 1988; Miller et al. 1988). Polymerizati on of four PBGs is initiated by linking one pyrrole ring to the cofactor, followed by st epwise addition of th ree pyrroles (Anderson and Desnick 1980; Shoolingin-Jordan and Warren 1987; Warren and Shoolingin-Jordan 1988). This results in an enzyme-bound hexapy rrole, which is hydrolyzed to release the tetrapyrrole product, thus regenerating th e holoenzyme (Shoolingin-Jordan and Warren 1987; Warren and Shoolingin-Jordan 1988). The three-dimensional structure of PBGD has been determined for the E. coli enzyme (Louie et al. 1992; Louie et al. 1996; Helliwell et al. 2003). The crystal structures reveal a flexible multidomain polymerase with a single active site (Louie et al. 1992; Louie et al. 1996; Helliwell et al. 2003). The protein is folded into three domains (Louie et al. 1992; Louie et al. 1996; Helliwell et al. 2003). The active site is located in a deep interdomain cleft and contains a dipy rromethane cofactor (Shoolingin-Jordan and Warren 1987; Warren and Shoolingin-Jordan 1988; Louie et al. 1992; Lo uie et al. 1996). Domains 1 and 2 exhibit similar fold s which resemble transferrins and periplasmic binding proteins (L ouie et al. 1992; Louie et al 1996). Domain 1 contains the substrate-binding site and is enriched with catalytically important residues (Louie et
20 al. 1992). Domain 2 has most of the cofact or-binding residues prov iding extensive saltbridges and hydrogen-bonds to hold the cofactor within the cleft between domain 1 and 2 (Louie et al. 1992; Louie et al 1996). Domain 3 supplies an invariant cysteine residue which forms a covalent link with the dipyrrome thane cofactor (Louie et al. 1992; Louie et al. 1996). It also appears to shield domain 1 and 2 from so lvent exposure (Louie et al. 1996). The multi-domain organization of PBGD thus provides intrinsic flexibility to the active site to facilitate substrate entry, cof actor attachment, chain elongation and product release (Louie et al. 1996). Many conser ved residues are found in the hydrophobic core and are important for protein stability (B rownlie et al. 1994). Conserved residues clustered in the active site clef t have been shown to be cataly tically essential (Brownlie et al. 1994; Louie et al. 1996). A group of conserved arginine residues have been found to be necessary for binding and positioning of the monoand poly-pyrrole substrates, cofactor binding, chain elonga tion and product release (Lande r et al. 1991; ShoolinginJordan and Woodcock 1991; Louie et al. 1992; L ouie et al. 1996; Shool ingin-Jordan et al. 2003a; Solis et al. 2004). Significantly, an invari ant active site aspartate residue has been hypothesized to play a key role in facilitating s ubstrate interaction with the cofactor (Woodcock and Jordan 1994). The carboxylate side chain of this aspartate is found at a position suitable for hydrogen-bonding with the pyrrole nitrogen groups of both the cofactor and substrate and in th is manner, allows stabilization of the reaction intermediate (Louie et al. 1992; Woodcock and Jo rdan 1994; Louie et al. 1996). Partial deficiency of PBGD causes one of the most common of the porphyrias, acute intermittent porphyria (A IP) (Meyer et al. 1972; Tschudy et al. 1975). AIP is an autosomal dominant disease with very low penetrance (Nordmann et al. 1997; Nordmann
21 and Puy 2002). Patients typically have acute episodes of neuropathic symptoms, most frequently with abdominal pains, and at tim es also with psychi atric manifestations (Anderson et al. 2005). The ne urovisceral attacks are often precipitated and sometimes exacerbated by alcohol, low caloric intake sex hormones and porphyrinogenic drugs such as barbiturates and sulfonamides (Jover et al. 2000; Anderson et al. 2005; Fraunberg et al. 2005). AIP is usually diagnosed by el evated concentrations of urinary porphyrin precursors ALA and PBG duri ng a symptomatic relapse (Tschudy et al. 1975; Anderson et al. 2005). Neurologic impairment is often at tributed to toxicity of the accumulated porphyrin precursors including AL A, although the mechanism is still not very well understood (Brennan et al. 1980; Lindberg et al. 1999; Solis et al. 2004). Currently more than 240 genetic lesions in PBGD have been described for AIP (Kauppinen 2005). They include splice site mutations, fram eshifts, small insertions, nonsense mutations, and nearly half of all, missense mutations (Brownlie et al. 1994; Floderus et al. 2002; Nordmann and Puy 2002; Fraunberg et al. 2005). While the molecular defects for AIP are highly heter ogeneous, a few mutations have been found with high prevalence in Sweden, Holland, Canada, Switzerland and France suggesting a founder effect (Lee and Anvr et 1991; Puy et al. 1997; Sc hneider-Yin et al. 2000a; Cappellini et al. 2002; Nordmann and Puy 2002; Schneider-Yin et al. 2002). PBGD mutations generally lead to low residual en zymatic activity, poor protein expression or instability or retention of the enzyme-int ermediate complexes (Desnick et al. 1985; Brownlie et al. 1994; Solis et al. 2004; Fra unberg et al. 2005; Pischik et al. 2005). Many single amino acid substitutions occur in cons erved residues located in the interdomain active site cleft and conserved hydrophobic core (Brownlie et al. 1994; Puy et al. 1997;
22 Floderus et al. 2002; Shoolingin-Jordan et al. 2003a). They have been shown to disrupt protein fold and stability, cof actor assembly, substrate bindi ng and polypyrrole formation and release (Brownlie et al. 1994; Puy et al. 1997; Floderus et al. 2002; ShoolinginJordan et al. 2003a). Two conventional therapeutic methods have been found to benefit some patients which involve high carbohydrate intake an d intravenous hematin administration (Tschudy et al. 1975; Anderson et al. 2005). None theless, recurrent severe AIP attacks are difficult to treat and can lead to high mo rtality rate. Liver tr ansplantation has been shown to be curative for some severe cases (Soonawalla et al. 2004). The first gene therapy trial based on recombinant adenovi ral transduction of functional PBGD has successfully increased enzymatic expression in hepatic tissues and restored metabolism of porphyrin precursors ALA and PBG in an AIP mouse model (Johansson et al. 2004). This approach points to th e potential for effective gene therapy in human AIP. Uroporphyrinogen III synthase The fourth enzyme in the heme biosynthetic pathway is uroporphyrinogen III synthase (UROS) (EC 18.104.22.168). UROS cataly zes cyclization of hydroxymethylbilane with concurrent D ring flipping to fo rm uroporphyrinogen III (Bogorad 1958b; 1958c; Shoolingin-Jordan et al. 1979). Uroporphyrinogen III is the fi rst tetrapyrrole macrocycle in the heme biosynthetic pathway and a ubiqu itous precursor of all other tetrapyrrole compounds including chlorophyll, siroheme, F430 and vitamin B12 (Schubert et al. 2002). While non-enzymatic cyclization of HMB produces uroporphyrinogen I isomer, D ring inversion is exclusive to th e enzymatic cyclization of lin ear bilane (Battersby 1978).
23 Asymmetry of the D ring thus serves as a hallmark of all biosynthetic porphyrins (Battersby 1978). From bacteria to human, UROS has been found as a small cytosolic protein comprising ~250-300 residues (Tsai et al. 1988; Alwan et al. 1989; Mathews et al. 2001). The protein sequence is highly diverged th roughout evolution and retains only a small number of conserved residues scattered in th e central portion of th e sequence. (Mathews et al. 2001). UROS has been purified from bacteria, plant and mammals, and it is typically present as a monomer of ~30 kD a (Higuchi and Bogorad 1975; Clement et al. 1982; Hart and Battersby 1985; Tsai et al. 1987; ShoolinginJordan et al. 1988; Smythe and Williams 1988; Tsai et al. 1988; Alwan et al. 1989; Omata et al. 2004). Interestingly, both prokaryotic and eukaryotic UROS have been shown to be extremely thermolabile (Tsai et al. 1987; Alwan et al. 1989). For pur ified human erythrocyte UROS, the half life at 37, 45, and 60 C was around 30, 4 and 1 min, respectively (Tsai et al. 1987). E. coli UROS was completely inactivated at 60 C in 1 min (Alwan et al. 1989). Another interesting feature of human UR OS is that although there is only a single enzyme in all tissues, distinct housekeeping and erythroi d-specific transcripts are generated by alternative promoter elements (Aizencang et al. 2000a; Aizencang et al. 2000b). While expression of the housekeeping form is driv en by a TATA-less prom oter, the erythroid promoter contains binding site s for erythroid-specific tran scription factors GATA-1 and NF-E2 (Aizencang et al. 2000a ; Aizencang et al. 2000b). According to a widely accepted catalytic model, the reaction pathway of UROS involves a spiro mechanism to allow D ring r earrangement (Battersby and Leeper 1990; Leeper 1994). In this scheme, enzymatic r eaction is initia ted with rearranging the A ring
24 of HMB to remove the C20 hydroxyl group, followed by el ectrophilic attack of C16 of the D ring by C20 carbon-cation to generate a spirocyclic pyrrolenine (Battersby and Leeper 1990; Leeper 1994). This intermediate is reso lved in the opposite di rection yielding an azafluvene, which cyclizes to form uropor phyrinogen III concomitant with D ring inversion (Battersby and Leeper 1990; Leeper 1994). This model is in agreement with the observation that UROS is i nhibited by spirolactam with a Ki of 1-2 M (Stark et al. 1986; Cassidy et al. 1991). To date the structure of UROS has only been determined for the human enzyme (Mathews et al. 2001). The crystal structure shows that human UROS is an elongated bilobed monomer (Mathews et al. 2001). The pr otein is folded into two separate domains with similar topology and interconnect ed via a two-stra nded anti-parallel -ladder (Mathews et al. 2001). Each domain comprises a parallel -sheet surrounded by helices (Mathews et al. 2001). Domain 1 cont ains the Nand C-terminal sequences and belongs to a flavodoxin-like fold family, wh ile domain 2 adopts a DNA glycosylase-like fold (Mathews et al. 2001). The active site is proposed to be located in a large open cleft between the two domains on the basis of structural modeling of a putative product-UROS complex and clustering of conserved residu es along the lining of the interdomain cleft (Mathews et al. 2001). While mutational anal ysis indicated that the conserved residues are not absolutely required for activity, subs trate binding is thought to involve a large number of weaker interactions in the activ e site and enzymatic r eaction may not require acid/base catalysis (Mathews et al. 2001). Opening and closi ng of the cleft appears to be mediated, at least in part, by movement of the bridging -ladder residues (Mathews et al.
25 2001). This intrinsic flexibility of the interdomain cleft is likely to play a key role in substrate recognition and product rele ase (Mathews et al. 2001). The first porphyria ever described, congeni tal erythropoietic porphyria (CEP), or Gunthers disease, is caused by enzymatic defi ciency of UROS (Rom eo and Levin 1969). CEP is a rare autosomal recessive disease with variable clinical severity (Desnick and Astrin 2002). Characteristic symptoms incl ude severe cutaneous photosensitivity, red urine and red teeth (Desnick and Astrin 2002). Hemolytic anemia and splenomegaly are observed in some patients (Desnick and Astrin 2002). Cutaneous le sions often begin in early childhood, and the diseas e can be fatal without trea tment (Desnick and Astrin 2002). CEP is usually diagnosed by elevated concentrations of urinary uroporphyrin I and coproporphyrin I (Desnick and Astrin 2002 ). The disease symptoms develop as a result of marked decrease of UROS activity usually to <1% normal level (Desnick and Astrin 2002). UROS deficiency leads to accumulation of hydroxymethylbilane, which undergoes non-enzymatic conversion to yi eld isomer I porphyrinogens, including the non-enzymatically cyclized uroporphyrinogen I and its decarboxylation product coproporphyrinogen I (Desnick and Astrin 2002). Because isomer I porphyrinogens cannot be further metabolized towards heme s ynthesis, they undergo autoxidation to yield uroporphyrin I and coproporphyrin I (Desnick and Astrin 2002). Excess type I porphyrins are cytotoxic to the developing er ythrocytes in bone marrow, cause skin photosensitivity and give the br ight red color in urine and deposited tissues (Desnick and Astrin 2002). To date nearly 40 mutations in UROS have been attributed to CEP pathogenesis (Desnick and Astrin 2002). They include more than 20 missense mutations, and a small
26 number of deletions, insertions, splicing defects and erythroid-specific promoter mutations (Desnick and Astrin 2002). Most of the missense mutations occur at conserved residues in UROS (Desnick an d Astrin 2002). Only a few residues affected are mapped close to the active site cleft, most substitutions are scattered throughout the tertiary structure making it possible that the primary mu tational effect results from disruption of protein fold (Mathews et al. 2001). The major ity of mutant enzymes are inactive or show very low residual activity at less than 2% normal level (Xu et al. 1995; Fontanellas et al. 1996; Shady et al. 2002). Only a few muta nts have been found to retain up to 1/3rd normal level activity, although they also become less thermostable (Xu et al. 1995; Shady et al. 2002). Further, these less severe muta tions also seem to produce milder clinical expression (Xu et al. 1995; Sh ady et al. 2002). The invarian t residue C73 has been found to be a mutational hotspot (Xu et al. 1995) C73R is the most commonly found CEP allele with an observed fre quency of ~30% (Desnick and Astrin 2002). C73R mutant enzyme retains less than 1% residual act ivity (Boulechfar et al. 1992). Among all CEP alleles, C73R homozygotes manifest the most severe clinical symptoms (Desnick and Astrin 2002). Structural mapping indicates that C73 residue is buried in the second domain hydrophobic core and does not appear to be catalytically essent ial (Mathews et al. 2001). Instead, loss of activity in C73R is most likely due to destabilization of the enzyme structure (Omata et al. 2004). CEP patients generally need to avoid sun and require skin protection (MathewsRoth 1998). More severe cases may require treatment which reduces bone marrow erythropoiesis and thereby decreases porphyr in production, and these therapies include frequent blood transfusions, hydroxyurea treatment a nd splenectomy (Desnick and Astrin
27 2002). Bone marrow transplant has proved cu rative for severely affected young patients resulting in correction of enzymatic defici ency in the bone marrow and marked reduction of porphyrin levels and cutaneous lesions (D upuis-Girod et al. 2005). Current efforts are focused on developing hematopoietic stem cell-mediated gene therapy towards an effective cure (Desnick a nd Astrin 2002). This method is becoming feasible with progress in viral transduction of hematopoietic stem cells with functional UROS and murine model studies (de Verneuil et al. 2003 ; Gronimi et al. 2003; Bishop et al. 2005). Uroporphyrinogen III decarboyxlase The fifth enzyme, uroporphyrinogen III decarboyxlase (UROD) (EC 22.214.171.124), catalyzes four successive decarboxylation reactions on acetate groups of uroporphyrinogen III to yiel d the four corresponding -pyrrole methyl groups of coproporphyrinogen III (Mauzerall and Gr anick 1958; Romeo and Levin 1971). Although uroporphyrinogen III is the normal physiological substrate, both naturally occurring uroporphyrinogen isomers (I and III) may serve as substrates (Mauzerall and Granick 1958; de Verneuil et al. 1983b; Elde r et al. 1983; Straka and Kushner 1983). The decarboxylation product of isomer I, c oproporphyrinogen I, ca nnot be processed by the next enzyme coproporphyrinogen oxi dase, but undergoes autoxidation to uroporphyrin I (Bogorad 1958a). UROD has been identified in many speci es. The proteins typically contain 300400 residues and the sequence is considerab ly conserved throughout evolution with the highest degree of identity found in the N-terminal region (Romeo et al. 1986; Jones and Jordan 1993; Nishimura et al. 1993; Mock et al. 1995; Wyckoff et al 1996; Whitby et al.
28 1998) In most species, UROD has an a pparent monomeric mo lecular weight of 40-50 kDa (Jones and Jordan 1993; Wyck off et al. 1996). In bacteria l, yeast and animal cells, UROD is a cytosolic enzyme (Romeo a nd Levin 1971; Elder et al. 1983; Felix and Brouillet 1990). In plants, UROD is synthesi zed in the cytosol as a precursor form and imported into the chloroplast using a N-termin al sequence of ~40 residues, and the transit peptide is cleaved in the mature protein (Mock et al. 1995). Wh ile purified UROD is commonly monomeric (de Verneu il et al. 1983b; Elder et al. 1983; Straka and Kushner 1983; Romeo et al. 1986; Juknat et al. 1989; Fe lix and Brouillet 1990; Jones and Jordan 1993), dimeric UROD has also been shown in solution and in crystal structure (Kawanishi et al. 1983; Phillip s et al. 1997; Whitby et al. 1998; Phillips et al. 2003). The human UROD gene shows enhanced transc ription and abundant expression in erythropoietic cells, whereas non-erythroid expr ession is constitutive but occurs at a low level (Romeo et al. 1986). In contrast to most decarboxylases, UROD does not require cofactors for catalysis, and its enzymatic activity is inhibite d by divalent metal ions (de Verneuil et al. 1983b; Elder et al. 1983; Kawanishi et al. 1983; Straka and Kushner 1983; Felix and Brouillet 1990). It is known that the UR OD-catalyzed decarboxylation produces three intermediates, the hepta, hexaand penta-carboxyporph yrinogens (Mauzerall and Granick 1958). This reaction sequence appears to proceed via a preferred route that starts at the acetate side chain on the D ring py rrole and proceeds around the macrocycle in alphabetical order (Jackson et al. 1976). The rate-limiting step in the conversion of uroporphyrinogen III to coproporphyrinogen III is the decarboxylation of the first reaction intermediate hept acarboxylate porphyri nogen III (Straka and Kushner 1983;
29 Felix and Brouillet 1990). While kinetic analysis of mammalian UROD suggested the possibility of multiple catalytic sites. (de Verneuil et al. 1980; Straka and Kushner 1983; Chaufan et al. 2005), an altern ative model proposed that ther e is only a single active site which contains conserved residues for substr ate binding (Chelstowska et al. 1992; Garey et al. 1992). Decarboxylation may in volve the protonated pyrrole ring of uroporphyrinogen substrate to function as an electron sink to promot e electron withdrawl in a manner similar to the pyrid ine ring of PLP, e.g. in ALAS -catalyzed reaction (Barnard and Akhtar 1979). This latter model is consiste nt with the crystal st ructures (Whitby et al. 1998; Martins et al. 2001; Phillips et al. 2003). The X-ray crystal structure of UROD has been solved for the human and tobacco enzymes, revealing a high degree of conserva tion in the overall topology (Whitby et al. 1998; Martins et al. 2001; Phillips et al. 2003). The protein crystallized as a homodimer with a globular tiertia ry structure (Whitby et al. 1998; Ma rtins et al. 2001; Phillips et al. 2003). Each monomer contains one domain consisting of a distorted ( )8 TIM (triose phosphate isomerase) barrel (Whitby et al. 1998; Martins et al. 2001; Phillips et al. 2003). A single active site is pr oposed to be located in a de ep cleft which extends from the hydrophobic -barrel core to the dimer interface (Whitby et al. 1998; Martins et al. 2001; Phillips et al. 2003). Hi ghly conserved residues are cl ustered around the active site cleft (Whitby et al. 1998; Martins et al. 2001; Phillips et al. 2003). Most recently, the structure was dete rmined for the complex between human UROD and product coproporphyrinogen (Phillip s et al. 2003). Both coproporphyrinogen I and III isomers were found at the same position in the active site cavity in each monomeric subunit and both active sites in the UROD dimer we re occupied (Phillips et
30 al. 2003). Specificity for the I and III isomer s appears to involve differential localization of the D-ring propionate side chain which is mediated by interaction with adjacent residues including conserved argi nines (Phillips et al. 2003). Most interestingly, the bound coproporphyrinogen was found to adopt a distorted conformation, i.e. a domed macrocycle, and one face of the macrocycle was aligned against a hydrophobic surface of the active site formed by a ring of cons erved residues centrally located in the -barrel core (Phillips et al. 2003). An invariant aspartate (human D86), which provides the only acidic side chain in the active site, was shown to approach the center of the bound macrocycle and its carboxylate side chain was situated in an optimal position for hydrogen-bonding with all four pyr role nitrogens (Phillips et al. 2003). The active site geometry is consistent with a previous pr oposal suggesting that a hypothetical protonated intermediate serves as an electron sink to accept the electrons retain ed when the acetyl side chain is released as CO2 (Barnard and Akhtar 1979; Akhtar 1994). Further, the invariant active site aspartate (D86) seems ve ry likely to function in substrate binding by coordinating the four pyrrole nitrogens in a domed conformation, and it might also directly promote catalysis by electrostatic stabilization of a protonated intermediate (Phillips et al. 2003). Kinetic and structural analysis of the site-directed D86 mutants provided additional support for the proposed catalytic model (Phillips et al. 2003). The most common form of porphyria in humans, porphyria cutanea tarda (PCT), is associated with subnormal UROD activity (Kushner et al. 1976). Clinically PCT is characterized by cutaneous hyperpigmentatio n (Fritsch et al. 1998). The patients typically exhibit severe photosensitivity associated with skin lesions as a result of the deposition of uroporphyrin and partially decarb oxylated porphyrins (Fritsch et al. 1998).
31 Excessive uroporphyrins are generated and accumul ated in the liver, circulate in plasma and excreted in the urine (Fritsch et al. 1998). The disease is classifi ed as either familial or sporadic (Kushner et al. 1976). Familial PCT (F-PCT) is transmitted as an autosomal dominant trait (Kushner et al 1976). F-PCT patients are he terozygous for mutant UROD alleles and typically exhibit half-normal UROD activity in all tissues (Phillips et al. 2001). Most carriers of mutant UROD alleles do not ha ve clinical expression unless additional factors are present that further redu ce UROD activity in the liver (Bulaj et al. 2000). In sporadic PCT (S-PCT) patients, defi ciency of UROD activity is restricted to the liver, although no mutation in UROD has b een identified (de Ve rneuil et al. 1978; Elder et al. 1978b; Garey et al. 1993). To da te more than 30 mutations in UROD have been associated with F-PCT, and a majority of them are single amino acid substitutions scattered throughout the protein scaffold (Wh itby et al. 1998; Phillips et al. 2001). None of the residues affected is directly involved in catalysis in the active site (Whitby et al. 1998; Phillips et al. 2001). Most mutations are located near the dimer interface or regions remote from the active site (Whitby et al. 1998; Philli ps et al. 2001). They are likely to disrupt protein fold or stability resu lting in activity decreas e or rapid degradation of the enzyme (de Verneuil et al. 1986; Whitby et al. 1998; Phillips et al. 2001). An interesting aspect of PCT is its well-known association with hemochromatosis, an autosomal recessive di sease characterized by hepatic iron overload (Berlin and Brante 1962). Hereditary hemo chromotosis is most commonly caused by mutations in the HFE locus (Feder et al. 1996) In many clinical studies, mutant HFE alleles, primarily with the single substitu tion C282Y, have been shown to occur at increased frequencies in overt PCT patients co mpared to the general population (Roberts
32 et al. 1997; Bygum et al. 2003; Harper et al. 2004; Mehrany et al. 2004). It has been widely accepted that inheritance of the hemoch romatosis gene is a primary risk factor for PCT (Elder and Worwood 1998; Bulaj et al. 20 00). Other susceptibility factors include hepatitis C viral infection, alcohol consump tion, heavy metal exposure and estrogen use, although the mechanism by which PCT pathogenesis is promoted by these risk factors is not yet clear (Bulaj et al. 2000; Egger et al. 2002; Sams et al. 2004). PCT can usually be treated with sunscreen for skin protection and phlebotomy therapy for iron removal (Thunell and Harper 2000; Sarkany 2001; Nordmann and Puy 2002). Repeated phlebotomies deplete hepatic iron stores and thereby lead to clinical and biochemical remission (Lundvall and Weinfeld 1968). Coproporphyrinogen III oxidase The sixth enzyme in the pathway, c oproporphyrinogen III oxidase (CPO) (EC 126.96.36.199), converts coproporphyrinogen III to pr otoporphyrinogen IX. CPO catalyzes the oxidative decarboxylation of tw o propionate side chains on pyrrole rings A and B in coproporphyrinogen III to yield two vinyl groups in prot oporphyrinogen IX (Sano and Granick 1961; Sano 1966). There are two distinct CPOs, oxygen-dependent and oxygenindependent. These enzymes share no sequence or structural similar ity (Dailey 2002). In eukaryotes and some aerobic prokaryot es, there is a single, oxygen-dependent CPO (odCPO) (Dailey 2002). The odCPO s use molecular oxygen to convert coproporphyrinogen and releases CO2 (Dailey 2002). In c ontrast to many other decarboxylases, odCPO-catalyzed reaction does not require a ny cofactor or metal ions (Yoshinaga and Sano 1980; Camadro et al 1986a; Medlock and Dailey 1996; Dailey
33 2002). odCPO is highly conserved throughout e volution and contains a large number of invariant residues (Martasek et al. 1994; Medlock and Da iley 1996; Lamoril et al. 2001; Phillips et al. 2004). From bacteria to hu man, the protein typically occurs as a homodimer of 70-80 kDa (Yoshinaga and Sano 1980; Camadro et al. 1986a; Bogard et al. 1989; Kohno et al. 1993; Kruse et al 1995; Martasek et al. 1997; Macieira et al. 2003). While odCPOs from yeast S. cerevisiae and E. coli are cytosolic proteins (Camadro et al. 1986b; Zagorec et al. 1988; Macieira et al 2003), odCPO is present in the intermembrane space betw een the inner and outer mito chondrial membranes in higher animal cells (Elder and Evans 1978a; Grandc hamp et al. 1978). This results from translocation of a cytosolic precursor into mitochondria, mediated by a N-terminal leader sequence which is then proteolytically rem oved (Kohno et al. 1993; Delfau-Larue et al. 1994; Martasek et al. 1994; Su sa et al. 2003). An unus ually long bipartite signal sequence consisting of ~120 N-terminal residu es is found to be essential for mammalian CPO targeting (Susa et al. 2003; Dailey et al. 2005). Kinetic studies of odCPO suggest that the rate-limiting step for the overall reaction is the first decarboxylation with transient fo rmation of the tricarboxylic intermediate, harderoporphyrinogen (Elder and Evans 1978b; Elder et al. 1978a). Although deprotonation of the pyrrole ring and pe roxide formation have been implicated, the catalytic mechanism is not well understood (Seehra et al. 1983; Colloc'h et al. 2002). Recently, new insights have begun to develop wi th the structural determination of odCPO (Phillips et al. 2004; Lee et al. 2005). Crystal structures have been solved first for the yeast S. cerevisiae enzyme ( hem13 p), and most recently, for the human protein revealing a homologous folding
34 pattern (Phillips et al. 2004; Lee et al. 2005 ). The yeast odCPO crystallized as a homodimer of ~70 kDa. Each monomer c ontains a single domain consisting of a relatively flat 7-stranded anti-parallel -sheet flanked on both sides by -helices (Phillips et al. 2004). There is one active site in each subunit, which is located in a deep cavity sandwiched between one face of the -sheet and a C-terminal he lical surface (Phillips et al. 2004). The crystal structures revealed two conformations, an open form and a more closed form. Active site closure appears to be mediated by movement of two short helices positioned above the opening. Possi bly, substrate binding induces the closed conformation, which allows sequestration of the active site from the bulk solvent and leaves a cavity with a size and shape comparable to the substrate molecule (Phillips et al. 2004). As indicated by a structural mode l of coproporphyrinogen I II docking into the CPO open conformer, the substrate fits tightly into the active site and is surrounded by many conserved residues possibly involved in coproporphyrinogen bindi ng (Phillips et al. 2004). Interestingly, an invariant aspa rtate (D274) which belongs to a short, conformationally mobile helix covering the pocket is found di rectly above the center of the macrocycle. This led to the suggestion that D274 plays a crucial role in substrate binding by interacting with the four pyrrole nitrogen groups via its carboxylate side chain. An updated catalytic model has been proposed based on the structural features of odCPO (Lash 2005). It emphasizes the importan ce of substrate straining in the oxidative reaction, e.g. by bending c oproporphyrinogen macrocycle into a domed conformation (Lash 2005). Distortion of the substrate facil itates interaction with an adjacent acidic residue, which deprotonates the pyrrole nitrogens (Lash 2005). The deprotonated
35 pyrrolic ring favors reaction with molecula r oxygen to generate an anionic peroxide intermediate (Lash 2005). This intermediate is resolved to yield a vinyl pyrrole derivative concomitant w ith the release of CO2 and peroxide (Lash 2005). Indeed, peroxide formation has been observed in the reaction catalyzed by E. coli odCPO (Breckau et al. 2003). Oxidative decarboxylation of coproporphyrinogen in anaerobic organisms is catalyzed by an oxygen-inde pendent CPO encoded by the hemN gene (Dailey 2002). Although both hemF and hemN genes are present in facultative bacteria such as E. coli and Salmonella typhimurium the hemN product is selectively upr egulated during growth under low oxygen tension or an aerobic conditions, while the hemF product (odCPO) is synthesized only aerobically (Xu et al. 1992; Troup et al 1995). Unlike odCPO, the reaction catalyzed by the hemN -encoded CPO requires Mg2+, ATP, NAD(P)+, methionine and an FeS cluster (Coomber et al. 1992; Layer et al 2002). Also, the en zyme is entirely unrelated to odCPO on a structural le vel. The crystal structure of the E. coli hemN product reveals features common to the radi cal SAM enzymes (Layer et al. 2003). The protein is a monomer of ~53 kDa (Layer et al. 2002; Layer et al. 2003). It is folded into two domains featuring a curved parallel -sheet and contains a [4Fe-4S] cluster and Sadenosylmethionine (SAM) which are necessa ry for enzymatic activity (Layer et al. 2002; Layer et al. 2003). Partial enzymatic deficiency of CPO cau ses two distinct fo rms of porphyria hereditary coproporphyria a nd harderoporphyria (Lamoril et al. 2001; Nordmann and Puy 2002). Hereditary coproporphyria (HCP) is a rare acute hepatic porphyria (Delfau-Larue et al. 1994). The disease is transmitted as an autosomal dominant trait with low
36 penetrance (Martasek 1998; Nordmann and Puy 2002). Clinical symptoms are primarily acute neurovisceral attacks commonly associat ed with abdominal pain, hypertension and peripheral neuropathy (Martasek 1998; K uhnel et al. 2000; Nordmann and Puy 2002). The symptoms are often provoked by drugs, al cohol consumption or fasting (Nordmann and Puy 2002; Anderson et al. 2005; Kauppine n 2005). Urinary and fecal excretion of coproporphyrin III is elevated (Kuhnel et al. 2000; Gross et al. 2002a). Sometimes patients also exhibit photosens itive skin lesions (Kuhnel et al. 2000). Enzymatic activity of CPO is reduced, typically to half-normal in heterozygous patients (Gross et al. 2002a; Nordmann and Puy 2002). Harderoporphyria is a rare erythropoietic variation of HCP (Nordmann et al. 1983). While the patients do not have ac ute attacks of neurological dysfunction, their clinical manifestations ar e characterized by fecal excretion of a large amount of harderoporphyrin and hemolytic an emia associated with iron overload (Nordmann et al. 1983; Lamoril et al. 1995; La moril et al. 1998; Schmitt et al. 2005). Currently nearly 40 mutations in huma n CPO have been described including small insertions, deletions, splice site mutations and missense mutations (Rosipal et al. 1999; Lamoril et al. 2001; Wiman et al. 2003a; Schmitt et al. 2005). More than half of the defects are single amino acid substitutions which occur mostly in the conserved residues spread throughout the protein, and th ey are likely to disrupt protein fold or dimerization (Lamoril et al. 2001; Phillips et al 2004; Lee et al. 2005) While active site mutations are very rare, CPO defects identified in harderoporphyria patients are restricted to K404E and R401W substitutions (Lamoril et al. 1995; Lamoril et al. 1998; Schmitt et al. 2005). Both residues ar e located in a short mobile -helix (D400-K404), which covers the active site cavity (Phillips et al. 2004). The mu tant enzymes are thermolabile,
37 and exhibit markedly decreased binding affi nities for substrate coproporphyrinogen and the first reaction intermediate harderoporphyr inogen (Nordmann et al. 1983; Lamoril et al. 1995; Schmitt et al. 2005). Structural an alysis suggests that the mutations may interfere with active site cl osure, and thus promote diffusion of the harderoporphyrinogen intermediate out of the substrate-binding poc ket (Phillips et al. 2004). Intravenous hemin arginate infusion has been shown to effec tively alleviate acute porphyric attacks and remains a preferable therapy for severely affected HCP pa tients (Kuhnel et al. 2000; Gross et al. 2002b; Nordmann and Puy 2002). Protoporphyrinogen IX oxidase Protoporphyrinogen IX oxidase (PPO) (EC 188.8.131.52), the penultimate enzyme in heme biosynthesis, catalyzes the six-electron oxidation of protoporphyrinogen IX to form fully conjugated macrocyclic protoporphyrin IX (Porra and Falk 1964). PPO utilizes three molecules of molecular oxygen, re ducing them to three molecules of H2O2 and generating the protoporphyrin product. The oxidation reac tion requires a noncovalently bound cofactor flavin adenine dinucleotide (F AD) (Shoolingin-Jordan 1991). PPO has been widely identified in bacteria, yeast, plants and higher animals (Dailey and Karr 1983; Dailey et al. 1994b; Nish imura et al. 1995b; Camadro and Labbe 1996; Dailey and Dailey 1996a; Dailey and Dailey 1996b; Narita et al. 1996; Wang et al. 2001). Although its sequence became significantly diversified du ring evolution, the protein generally has a monomeric molecular mass of ~50-60 kDa and contains a highly conserved N-terminal dinucleotide-binding motif G-X-G-X2-G which mediates the coordination of a FAD cofactor (Dailey and Karr 1983; Proulx a nd Dailey 1992; Dailey et al. 1995; Dailey and
38 Dailey 1996b; Wang et al. 2001). Plant PPO ha s two isozymes, with PPO1 found in the plastid chloroplast and PPO2 in the mitoc hondrion (Narita et al. 1996; Lermontova et al. 1997). Eukaryotic PPO is tig htly bound to the inner mitochondrial membrane (Dailey and Karr 1983) (Deybach et al. 1985; Nishimur a et al. 1995b). It is typically found as a homodimer of ~100 kDa (Dailey and Da iley 1996b). Interestingly, the monomeric subunit exhibits a highly protease-resistan t, compact domain structure (Arnould and Camadro 1998; Arnould et al. 1999). A block of ~60 conserved residues near the Nterminal region has been implicated to serve as a membrane anchor for the enzyme (Arnould and Camadro 1998; Koch et al. 2004; Morgan et al. 2004), although a short conserved N-terminal sequence appears suffi cient for mitochondrial import (Fraunberg et al. 2003; Dailey et al. 2005). It is possible that the two regi ons interact to optimize the efficiency of mitoc hondrial targeting. In prokaryotes, as with CPO, PPO has two forms, but only a single form is present in a given cell (Dailey 2002). One is th e oxygen-dependent form encoded by the hemY gene; the other is the oxygen-i ndependent form encoded by the hemG gene (Dailey 2002). In most bacteria, oxygendependent PPO is associated with the plasma membrane as a monomer or homodimer (Dailey and Da iley 1996a; Wang et al. 2001). However, B. subtilis PPO ( hemY p) is water soluble a nd exists in a monome ric form of ~51 kDa (Hansson and Hederstedt 1992; Dailey et al. 1994b). In addition to protoporphyrinogen, this enzyme can efficiently oxidize copr oporphyrinogen to copr oporphyrin (Hansson et al. 1997). An interesting feature of B. subtilis PPO is that it is only very weakly inhibited by diphenyl ether herbicides (Dailey et al. 1994b ). This family of herbicides are known as potent inhibitors of eukaryotic PPOs and cause light-d ependent phytotoxicity in plants
39 as a result of substrate prot oporphyrinogen accumulation in the inappropriate cellular compartment (Matringe et al. 1989; Camadr o and Labbe 1996; Narita et al. 1996). Overexpression of B. subtilis PPO in the plastids of transgenic plants has been achieved and conferred resistance to the herbicid e oxyfluorfen (Li and Nicholl 2005). In anaerobic and facultative bacteria, protoporph yrinogen oxidation is catalyzed by the hemG -encoded PPO, which belongs to a multimeric protein complex in the cellular respiratory chain (Klemm a nd Barton 1987; Dailey 2002). E. coli PPO ( hemG p) product is a small, cytosolic soluble, monomeri c protein of ~21 kDa containing 181 residues (Sasarman et al. 1993; Nish imura et al. 1995a). The structure of PPO was recently dete rmined for the mitochondrial isozyme (PPO2) from common tobacco Nicotiana tabacum (Koch et al. 2004). The X-ray crystal structure of PPO2 was solved at 2.9 reso lution with the inhi bitor 4-bromo-3-(5'carboxy-4'-chloro-2'-fluoro-phenyl)-1-methyl -5-trifluoromethyl-pyrazol bound in the active site (Koch et al. 2004). The protein is a homodimer of ~110 kDa and contains one non-covalently attached FAD cofactor in each subunit. Each monomer consists of three domains, which are proposed for binding to FAD, substrate a nd the mitochondrial membrane, respectively. Detergent Trit on X-100 used for solubilization and crystallization was found in th e interdomain cleft. While PPO has long been known to possess a FAD-binding motif commonly found in a superfamily of flavoproteins including monoamine oxidases and some phytoene dehydrogenases (Dailey and Dailey 1996b), the FAD-binding domain of PPO2 also e xhibits significant ove rall structural resemblance to this family of flavin-depe ndent enzymes. The substrate-binding domain, suggested by the orientation of th e bound inhibitor, contains mostly -strands (Koch et al.
40 2004). The active site is thought to lie in a narrow cavity between the FAD and substrate-binding domains of each monomer (Koch et al. 2004). The docking model of protoporphyrinogen binding to PPO2 indicated th at the vinyl groups of pyrrole rings A and B are buried deeply in th e interior of the active si te cavity, while the propionyl groups of pyrrole rings C and D are directed outwards to th e solvent-exposed end (Koch et al. 2004). Since the reactive group of FAD lies adjacent to pyrrole rings A and D, the methylene bridge between ring s A and D is proposed to be the site of initial oxidation (Koch et al. 2004). This arrangement app ears to be required for protoporphyrinogen binding and catalysis, as macrocycle rotation is unlikely due to steric constraint within the active site. The putative membrane-binding domain protrudes from the globular core and consists primar ily of a group of -helices formed by a consecutive N-terminal sequence of ~100 residues (Koch et al. 2004). As was found with monotopic membrane proteins (Binda et al. 2004), a conserve d helical region forms a hydrophobic surface at the base of PPO and probably mediates memb rane insertion in a monotopical manner. The domain structure of PPO2 lends support to a previous model which suggested that the active site of PPO faces the cytoso lic side of the inner mitochondria membrane and substrate channeling might occur betw een the terminal heme synthetic enzymes (Ferreira et al. 1988). Spa tially, it seems possible that the product protoporphyrin can be translocated to the next enzyme ferrochel atase via a U-shaped channel which connects the active site cavity to the membrane (Koch et al. 2004). Enzymatic deficiency in PPO causes th e disease variegate porphyria (VP) (Brenner and Bloomer 1980; Deybach et al. 1 981; Meissner et al. 1986). VP is an autosomal dominant disease with low pene trance (Nordmann and Puy 2002). Patients
41 typically exhibit skin photosensitivity, or less co mmonly have acute episodes of neuropathy, or sometimes show both symptoms (Bales et al. 1980; Whatley et al. 1999; Fraunberg et al. 2002; Wiman et al. 2003b; Hift et al. 2004). Neurovisceral attacks are often precipitated by porphyri nogenic drugs (Hift and Meissn er 2005). PPO activity is approximately half-normal in all tissues of the clinically overt patients as well as asymptomatic carriers (Brenner and Bloomer 1980; Deybach et al. 1981; Da Silva et al. 1995). Currently about 130 mutatio ns of PPO have been associated with VP (Kauppinen 2005). They include splice site mutations, exon deleti ons, frameshifts, non-sense mutations, and half of all are missense mutati ons (Deybach et al. 1996; Whatley et al. 1999; Frank et al. 2001; Fraunberg et al. 2001; Morgan et al 2002; Wiman et al. 2003b). Single mutations are spread throughout th e protein, and many occur at conserved residues. Most VP mutations cause total or near complete (<10%) loss of enzymatic activity, thus exhibiting complete haplodefici ency (Deybach et al. 1996; Meissner et al. 1996; Fraunberg et al. 2001; Morgan et al. 2002). Recently, the first case report has been described for a patient recovery from VP after liver transplantation (S tojeba et al. 2004). Interestingly, one of the best known PPO mutations R59W gives a prime example of the founder gene effect. VP is highly pr evalent in the Afrikane r population of South Africa, which can be traced back to an an cestral PPO allele with R59W substitution (Meissner et al. 1996; Hift et al. 2004). Only minimal activity is retained in R59W mutant, probably because the R59 residue is requ ired for catalysis and active site integrity (Manelia et al. 2003). Modeling based on th e tobacco PPO2 crystal structure indicates that the equivalent residue (N67 in tobacco enzyme) occupies an active site position between FAD and substrate-binding site, wh ich implies that a single mutation can
42 interfere with FAD-substrate interaction and di sturb the active site geometry (Koch et al. 2004). Ferrochelatase The last enzyme in heme biosynthesis pathway is ferrochelatase (FECH) (E.C. 184.108.40.206). Ferrochelatase catalyzes the insertion of ferrous iron into protoporphyrin IX to form protoheme (Figure 3). Ferrochelatase activity was first described by Goldbergs group in 1956 using crude preparations of av ian erythrocytes (Goldberg et al. 1956). Subsequently, its enzymatic role in bi ological porphyrin metallation was further examined in a variety of cell types and organisms (Porra and Jones 1963a; 1963b; Neuberger and Tait 1964; Porra and Ro ss 1965; Jones 1969; 1970). With the identification of ferrochelatase-deficient, heme auxotrophic mutants in the bacterium Spirillum itersonii and yeast S. cerevisiae ferrochelatase was recognized as the genetic basis for iron chelation in heme biosynthesi s (Dailey and Lascelles 1974; Gollub et al. 1977). Ferrochelatase was firs t successfully cloned from S. cerevisiae in 1990 (LabbeBois 1990), followed with the cloning of hu man and murine genes (Nakahashi et al. 1990; Taketani et al. 1990; Br enner and Frasier 1991). As hi gh yield bacterial expression systems for recombinant mammalian ferrochelat ases became available, purified enzymes were given extensive biophysical and bioche mical examination (F erreira 1994; Sellers and Dailey 1997; Dailey et al. 2000). Signi ficantly, the X-ray crystal structure was
43 N HN NH NHOOC COOHN N N NHOOC COOHFe+ Fe2++ 2H+A D C B A D C B Figure 3. The reaction catalyzed by ferrochelatase. Adapted from (Al-Ka radaghi et al. 1997).
44 determined first for B. subtilis ferrochelatase in 1997 (AlKaradaghi et al. 1997), followed by the human (Wu et al. 2001) and yeas t proteins (Karlberg et al. 2002) In the past decade, the ease of cloning, expression and purification of ferrochelatase and the extent to which they have been characterized have grown rapidly and expanded to diverse organisms, thus a more complete understand ing of ferrochelatase evolution has emerged (Dailey and Dailey 2003). In most organisms, ferrochelatase is found as a membrane protein (Dailey and Dailey 2003). In animal cells and yeast, ferro chelatase is synthesized in the cytoplasm as a nuclear-encoded precursor protein and tran slocated into the mitochondria using a Nterminal leader sequence consisting of ~ 50 residues (Camadro and Labbe 1988; Karr and Dailey 1988; Taketani et al 1990; Dailey et al. 2005). The mature form, typically consisting of 400-500 residues, is associated with the ma trix side of the inner mitochondrial membrane, and the enzymatic r eaction it catalyzes takes place in the mitochondria (Jones and Jones 1968; Jones 19 69; Harbin and Dailey 1985; Prasad and Dailey 1995). Prokaryotic ferrochelatases ar e smaller, and while most of them are bound to the plasma membrane, some are soluble (Hansson and Hederstedt 1994; Wang et al. 2001; Dailey and Dailey 2002). Biochemical characterization of ferrochelat ase from different species has revealed that the reaction mechanism is conserved (D ailey and Dailey 2003). The catalytic model was initially proposed nearly two decades a go, and has been advanced to incorporate a multitude of details (Dailey and Dailey 2003). During catalysis the enzyme transiently binds the iron (II) and porphyrin substrates. Metal chelation occurs only when the porphyrin ring is distorted from planarity, concomitant with proton release from the
45 pyrrole nitrogen atoms. The product heme is flat and exits the active site once formed. An important feature of ferrochelatase-cataly zed metallation is the distortion of porphyrin macrocycle. The original proposal drew on the observation that porphyrin analog, N methyl-protoporphyrin, acts as a tight-binding competitive inhibitor of ferrochelatase with a Ki in the nanomolar range (Dailey et al. 1989). N alkyl porphyrins have an alkylated pyrrole ring bent 20-30o out of the otherwise plan ar porphyrin core, and this nonplanar conformation has been thought to yield high affinity binding in the active site (Dailey and Fleming 1983; Dailey et al 1989). Based on quantum mechanical calculation of porphyrin binding to B. subtilis ferrochelatase, tilting of a pyrrole ring induced by the enzyme significantly lowers the energy required for metal insertion, suggesting that macrocycle distortion may be a thermodynamically favorable process (Sigfridsson and Ryde 2003). Actually, antibodies made against N methylmesoporphyrin were found to catalyze metal ch elation into porphyrin s and thus possess ferrochelatase-like activity (Cochran and Schultz 1990). Resonance Raman spectroscopic studies revealed that murine ferroc helatase is able to induce formation of a saddled porphyrin even in th e absence of metal substrate (Franco et al. 2000), whereas yeast ferrochelatase binds and distorts porphyrin into a dome d or ruffled conformation in the presence of the inhibitor Hg2+ (Blackwood et al. 1997; Bl ackwood et al. 1998). A direct observation comes from the crystal structure of B. subtilis ferrochelatase with bound N -methyl-mesoporphyrin, which clearly show ed that a distorted macrocycle is held tightly in the active site (Lecerof et al. 2000). In comparison, deformation of porphyrin in the binding sites of catalytic antibodies has also been demonstrated using resonance Raman spectroscopy and crystal st ructure analysis, which lends additional
46 support that macrocycle distortion is a crucial step in enzymatic metallation (Yin et al. 2003; Venkateshrao et al. 2004). Although the importance of por phyrin deformation was widely recognized, little was known about the porphyrin substrate bind ing site and residues responsible for porphyrin distortion in the enzyme (Dailey and Dailey 2003). With the structure determination of ferrochelatase, it became possible to address this important question so as to understand the structur al basis of porphyrin-enzyme in teraction. The X-ray crystal structure first solved was that of the sma ll, cytosolic soluble, monomeric enzyme from the Gram-positive bacterium B. subtilis (Al-Karadaghi et al. 1997). Secondly, the structure of human ferrochelatase was dete rmined (Wu et al. 2001). This structure revealed a homodimer with one [2Fe-2S] cluster associated with each subunit, which is characteristic of animal ferrochelatases (Wu et al. 2001). More recently, the structure of yeast S. cerevisiae ferrochelatase was solved, showi ng a homodimeric protein without the FeS cluster (Karlberg et al. 2002). Although the overall sequence identity is only ~20% for these proteins, they exhibit remarkably similar folds and active site geometries (AlKaradaghi et al. 1997; Wu et al. 2001; Karl berg et al. 2002). A monomeric unit of the enzyme comprises two struct urally-related domains, each of which consists of a Rossmann-type fold, which is character ized by a four-stranded, parallel -sheet flanked by -helices on both sides. The active site is situated in a deep cavity between the two domains. The cavity is enriched with many c onserved residues, some of which have been shown to play important roles in catalysis (Al-Karadaghi et al. 1997; Lecerof et al. 2000; Wu et al. 2001; Karlbe rg et al. 2002).
47 The opening of the active site cleft consists of two sides, with one side composed of residues from the first two N-terminal -helices, and an intervening loop. The opposite side is formed by a short loop connecting a -strand and an -helix in the second domain (Al-Karadaghi et al. 1997; Lecer of et al. 2000; Wu et al. 2001; Karlberg et al. 2002). The ferrochelatase loop shows a high degree of sequence identity across a wide taxonomic range (Shi and Ferreira 2004). It was located in close proximity to the porphyrin macrocycle in the cr ystal structure of inhibited B. subtilis ferrochelatase (Lecerof et al. 2000). Consistent with this observation, the loop was predicted to contact the porphyrin based on molecular dynamics-b ased docking of nickel-protoporphyrin in B. subtilis ferrochelatase (Franco et al. 2000). Thereby, it was hypothesized that the conserved loop residues are i nvolved in substrate porphyrin binding and distortion during enzymatic reaction. In agreem ent with this proposal, structural features of human and yeast ferrochelatases indicate that the activ e site opening defines the membrane-facing side of ferrochelatase and provides an entr y site for substrate po rphyrin (Gora et al. 1996a; Wu et al. 2001; Karlberg et al. 2002). The experiment s described in the following chapters were aimed at detailed characterizati on of the loop motif in order to gain further understanding of the structural basis of the ferrochelatase-catalyzed reaction. Metal substrate binding to ferrochelatase ha s been studied in detail over the past decade. Current models suggest that a cha nneling mechanism is involved in the transfer of metal ions from the protein exterior into the internal reaction center near the distorted porphyrin (Sellers et al. 2001; Lecerof et al 2003). This notion received support from studies of bacterial and yeast ferrochelatases (Karlberg et al 2002; Lecerof et al. 2003). In the active site, an invariant ca talytic histidine residue (H183 in B. subtilis enyzme) has
48 been found necessary for metal chelation (K ohno et al. 1994; Gora et al. 1996b; Lecerof et al. 2000; Lecerof et al. 2003; Shipovskov et al. 2005), while an invariant glutamate and a conserved serine can also serve as metal lig ands (Karlberg et al. 2002; Lecerof et al. 2003). These residues constitute an interior si te for metal-coordination; and possibly, this site may interact with a second, surface-exposed metal-binding site, which is likely to play a regulatory role (Karlber g et al. 2002; Lecerof et al. 2 003). Transport of metal ions between the two sites is mediated by a group of conserved acidic re sidues aligned along a -helix on an exposed surface inside the active site cavity (Karlberg et al. 2002; Lecerof et al. 2003). Further, recent studies have pr ovided evidence that accelerating the rate of ligand exchange for metal ions is an important element of the catalytic mechanism (Shipovskov et al. 2005). Exam ination of human ferrochelatas e has led to the proposal that iron binding involves the conserved argini ne and tyrosine residues in the interior active site cavity, whereas a conduit formed by a group of conserved, mostly aromatic residue shuttles the iron substrate from an external location into the catalytic center (Sellers et al. 2001). In eukaryotic cells, the iron substrate is thought to be supplied by the mitochondrial membrane (Taketani et al. 1986; Lange et al. 1999; Napier et al. 2005). Recently, the iron chaperone frataxin and additi onal carrier proteins have been implicated as iron donors for ferrochelatase (Lesuisse et al. 2003; He et al. 2004; Yoon and Cowan 2004; Zhang et al. 2005). Frat axin, a small, evolutionari ly conserved, mitochondrial protein, is known for its role in iron ho meostasis by sequestering iron and making iron available for FeS cluster biogene sis and heme synthesis inside the mitochondria (Lesuisse et al. 2003; Park et al 2003; Yoon and Cowan 2003). In yeast and human
49 ferrochelatases, frataxin is shown to recrui t ferrous iron and medi ate its delivery to ferrochelatase by binding to the enzyme surfac e (Lesuisse et al. 2003 ; He et al. 2004; Yoon and Cowan 2004). Additionally, other carrier proteins including the yeast mitochondrial membrane proteins Mrs3p and Mrs4p have been suggested to cooperate with frataxin in a rapid transport system to move iron into mitochondria and direct its utilization in heme synthe sis (Zhang et al. 2005). One of the least understood aspects of ferro chelatase concerns the role of the FeS cluster. Although it is not present in ferrochelatases fr om some species including B. subtilis (Hansson and Hederstedt 1994), E. coli (Miyamoto et al. 1991) and S. cerevisiae (Labbe-Bois 1990; Gora et al. 1996b), cluste r-containing ferrochel atases are commonly found in evolution and a [2Fe-2S] cluster is associated with the enzyme in animals (Dailey et al. 1994a; Ferreira et al 1994; Sellers et al. 1998b), yeast Schizosaccharomyces pombe (Medlock and Dailey 2000) and bacteria Caulobacter crescentus and Mycobacterium tuberculosis (Dailey and Dailey 2002). In mammalian ferrochelatase, the [2Fe-2S] cluster is bound to a 30-50 residue long C-terminal tail, which constitutes a domain distinct from the st ructurally conserved catalytic core (Dailey et al. 1994a). The cluster is coordinated by four cysteine residues, with three ligands provided by a conserved C-X2-C-X4-C motif in the C-terminal tail and one much further removed in the N-terminal region (Crouse et al. 1996; Sellers et al. 1998b; Wu et al. 2001). In human ferrochelatase the cluster is labile, an d is rapidly destroyed upon exposure to nitric oxide concomitant with lo ss of enzymatic activity (Furukawa et al. 1995; Sellers et al. 1996). It is known that the cluster iron does not se rve as substrate iron and its redox state does not affect enzymatic ac tivity (Dailey et al. 1994a; Ferreira et al.
50 1994). Although mutations in the cysteine liga nds can abolish catalyt ic activity (Sellers et al. 1998b; Schneider-Yin et al. 2000b), the cl uster does not appear to be indispensable for enzymatic function (Medlock and Da iley 2000; Najahi-Missaoui and Dailey 2005; Ohgari et al. 2005). Instead, it may help to enhance prot ein stability in mammalian ferrochelatases (Medlock and Dailey 2000; Najahi-Missaoui and Da iley 2005; Ohgari et al. 2005). Further, FeS cluster assembly has been implicated in regulating overall heme production by modulating ferrochelatase activit y (Taketani et al. 2003; Lange et al. 2004). In human ferrochelatase, the C-terminal cluster-containing regi on is involved in binding ABC7, a member of the mitoc hondrial half-type ATP-binding cassette (ABC) transporters (Taketani et al. 2003 ). This interacti on is thought to medi ate an increase in heme synthesis during erythroid differen tiation, which contributes to elevated hemoglobin production in erythropoiesi s (Taketani et al. 2003). In mammals, naturally occurring genetic de fects cause decrease in ferrochelatase activity leading to the diseas e erythropoietic protoporphyria (EPP) (Magnus et al. 1961; Todd 1994). Typical symptoms of EPP incl ude skin photosensitivity beginning in childhood which results from accumulation of fr ee protoporphyrin primarily generated in the erythropoietic tissues (Todd 1994). Alt hough EPP is generally not life threatening, retention of excessive porphyrin s in the liver can cause hepat obiliary diseases and entails liver transplantation in a small fraction of patients (Bloomer 1988). In humans, EPP is primarily transmitted as an autosomal dominant trait with incomplete penetrance and variable clinical expression (D eLeo et al. 1976). A puzzling feature of clinical EPP is that the patients exhibit ferrochelatase activity at only ~10-30% of the normal level, lower than the expected, 50% residual activity (Bonkowsky et al. 1975; Norris et al.
51 1990). One explanation based on the multiallel ic inheritance model is that clinical manifestations of EPP require the inheritan ce of a low expression wild-type allele in combination with a defective ferrochelatase al lele (Gouya et al. 1996; Gouya et al. 1999). In support of this model, an intronic pol ymorphism mapped to the intron/exon splice acceptor site, i.e. an IVS3-48C/T allele, was shown to cause rapid degradation of aberrantly spliced transcripts and significan tly reduce the steady-state mRNA levels, thus resulting in decreased expres sion from an otherwise functional ferrochelatase allele (Gouya et al. 2002). Indeed, the low expression IVS3-48C allele is pr evalent in families with overt EPP from diverse human populations (Risheg et al. 2003; Wiman et al. 2003a; Whatley et al. 2004; Bloomer et al. 2005). Additional factors including individual genetic background have also been implicated in modulating ferrochelatase expression level and EPP pathogenesis (Chen et al. 2002; Gouya et al. 2004; Abitbol et al. 2005; Cooperman et al. 2005; Navarro et al. 2005; Zhou et al. 2005). While the size (45 kb) of the human FECH gene locus (chromosome 18q21.31) precludes the use of a comprehensive analysis of gene defects as routine examination for EPP most patients exhibit mutations in the prot ein coding region (Rufen acht et al. 1998). To date ~80 deleterious mutations have been identified in human ferrochelatase (Whatley et al. 2004). Exon skipping, frameshifts and nonsense mutations are common and result in inactive enzymes (Gouya et al. 1996; Rufe nacht et al. 1998; Sellers et al. 1998a; Schneider-Yin et al. 2000c). Over 30 misse nse EPP mutations have been identified, and they typically occur in conserved residues and result in large reduction of enzymatic activity (Rufenacht et al. 1998; Schneider-Yin et al. 2000b; Schneider-Yin et al. 2000c; Rufenacht et al. 2001; Whatley et al. 2004). Structural mappi ng indicates that relatively
52 few mutations are located in the active site, and the majority are scattered throughout the protein scaffold (Najahi-Missaoui and Daile y 2005). Interestingly, mutations near the dimer interface or FeS cluster are highly detrimental to residual enzymatic activity (Schneider-Yin et al. 2000b; Najahi-Missa oui and Dailey 2005; O hgari et al. 2005). A common treatment for EPP consists of preventing or lessening photosensitivity by -carotene ingestion, but this method is refractory to some patients and does not protect against liver damage (Mathews-Rot h 1993; Todd 1994). Long-term intake of Lcysteine has also been shown to ameliorate photosensitivity (Mathews-Roth and Rosner 2002). Bone marrow transplantation and hemat opoietic stem cell gene therapy offer the potential of more effective cu res (Pawliuk et al. 1999; Fontan ellas et al. 2000; Richard et al. 2004). In an EPP mouse model, it has b een feasible to correct photosensitization by transplanting bone marrows containing hematopo ietic stem cells transduced with human ferrochelatase (Pawliuk et al. 1999; Fontanel las et al. 2001; Richard et al. 2004). Recently, localized skin graft ha s been demonstrated to effectively protect against dermal photosensitivity (Pawliuk et al. 2005). Howeve r, liver damage remains difficult to cure (Bloomer et al. 1996; Bloomer et al. 2005) Bone marrow transplantation in young patients has been suggested as an option to prevent hepatobiliary complications (Fontanellas et al. 2000). In summary, with the availaibility of crystal structures for all the heme biosynthetic enzymes, it has become increasingly possible to carry out extensive structure-activity characterization for each prot ein. Altogether, these studies shall lead to a better comprehension of the driving for ces underlying the biologi cal process of heme
53 synthesis. More importantly, extending molecu lar analysis into clinical settings offers promise toward improving the diagnosis and treatment of porphyria and anemia. The experimental investigation presente d in this dissertation focuses on the terminal enzyme, ferrochelatase. As de scribed above, although enzymatic studies of ferrochelatase have been carried out for decades and encompass a wide range of biochemical and biophysical charact erization, only in the past few years have the crystal structures been solved for ferrochelatase is olated from different organisms (Dailey and Dailey 2003). This opens up the possibility of conducting structure-based functional assays to probe ligand-protein interac tions and to infer the enzymatic reaction mechanism. The primary purpose of this stu dy is to explore the structural basis of substrate porphyrin binding and distortion with regard to ferrochelatase catalysis. This investigation was initiated based on crystallographic observation and molecular modeling that a highly conserved loop motif was located in the active site and appeared to interact with bound porphyrin. The loop wa s subjected to random muta genesis and the properties of the active variants were ch aracterized. This study revealed a number of important results regarding the molecular details of ferrochelatase functi on. In support of the initial proposal, the loop residues are shown to play a role in s ubstrate porphyrin binding. The loop also regulates active site conformation resulting in mo dulation of reaction rate and substrate interaction. The loop residues ar e directly involved in distortion of the porphyrin macrocycle into a non-planar confor mation, which contribute s to the catalytic efficiency of ferrochelatase. Structural cha nges in the loop are also found to play a role in determining the sensitivity to the inhibitor N -methyl-protoporphyrin. One unexpected finding is that loop mutations allow the enzy me to retain activity even without FeS
54 cluster assembly and adopt an al ternative oligomeric states. Overall, the results described here provide new insights into the structural basis of porphyrin metallation catalyzed by ferrochelatase. Ferrochelatase is present at the converg ence of the intrace llular tetrapyrrole biosynthetic and iron supply pathways. Because accumulati on of unmetallated porphyrins and ferrous iron are cytotoxic, inte gration of both pathways by ferrochelatase is crucial for cellular metabolism, especially in the process of erythroid cell development (Ponka 1999). While enzymatic deficiency of ferrochelatase leads to the disease erythropioetic protoporphyria (Gouya et al. 2002), growi ng evidence indicates that ferrochelatase function is also intimately related to ir on metabolism (Taketani 2005). Therefore, it is hoped that ferro chelatase shall cont ribute to a better understanding of its physiological importance, for instance, towards the pathogenesis mechanism and therapy of erythropoietic porphyria as we ll as diseases associated with aberrant iron metabolism.
55 Chapter Two Materials Plasmid pGF42 contains the mature form of wild-type muri ne ferrochelatase subcloned into the pCASS3 vector (Ferreira 1994). Plasmid pGF47 contains the mature form of wild-type murine ferrochelatase with a N-terminal 5X-His-tag (Franco et al. 2000). Plasmid pGF23 contains the murine erythroid 5-am inolevulinate synthase cDNA cloned into the pCASS3 vector (Ferreira and Dailey 1993). E. coli strain vis was a kind gift of Dr. H. Inokuchi at Kyoto Un iversity (Nakahigashi et al. 1991). E. coli strains PK4331 ( isc S-) and RZ4500 were from Dr. P. Kiley at the University of Wisconsin (Schwartz et al. 2000). Oligonucleotides were synthesized by Cybersyn, IDT Integrated Technologies and Invitrogen. Protoporphyrin IX, Ni -protoporphyrin, zinc-p rotoporphyrin and N methyl protoporphyrin were obtained from Fr ontier Scientific and were used without further purification. Hemin, tris(hydr oxymethyl)amino-methane carbonate (Trizma base), MOPS, tricine, thiamine, cholic acid, sodium citrate, polyeth ylene glycol sorbitan monooleate (Tween-80), polyethyl ene glycol sorbitan monolaurate (Tween-20), agarose, bovine serum albumin standard solution ampules, the bicinchonic acid protein determination reagents, gel filtration molecular weight markers, CelLytic B II bacterial
56 cell lysis extraction reagent, ExtrAvidin-perox idase, palmitic acid, stearic acid, oleic acid, linoleic acid, arachidonic acid and N -(biotinoyl)-1,2-dihexadecanoyl-sn-glycero-3phosphoethanolamine triethyl-ammonium salt (biotin-DHPE) were obtained from Sigma Chemicals. Bacto agar, Bact o tryptone, yeast extract, vitamin assay casamino acids, dextrose and ferrous ammonium sulfate were from Fisher Scientific. Restriction enzymes, DNA ligase and DNA polymerases we re from New England Biolabs. Plasmid purification kit, QiaQuick gel extraction ki t and Ni-NTA agarose resins were from Qiagen. Talon metal chelate affinity resins and Talon HT 96-well plate were from BD Biosciences. Blue-sepharose CL-6B, PD -10 columns and Microspin S200-HR columns were from Pharmacia. Amicon stirred ce lls, centricon, microcon, Amicon Ultra-4 and Ultra-15 centrifugal filter uni ts were from Millipore. Chelex-100 resins, sodium dodecyl sulfate-polyacylamide gel electrophoresis reag ents and silver staini ng kit were from BioRad Laboratories. Superdex-200 gel filtra tion matrix, T7 sequenase version 2.0 DNA sequencing kit, ECL chemiluminescent We stern blotting detection reagents and Hyperfilm ECL were from GE Healthecare. Superblock blocking buffer and Gelcode blue stain were from Pier ce. Protran BA85 nitr ocellulose membrane (0.45 m) was from Schleicher and Schuell. P hosphatidylcholine (bovine liver), phosphatidylethanolamine (bovine liver) phosphatidylinositol (bovine liver), phosphatidylserine (porcine brain), lys ophosphatidylcholine (soy), lysophosphatidylethanolamine (chicken egg), car diolipin (bovine heart), chol esterol, diacylglycerol and sphingomyelin (porcine brain) were from Avan ti Polar Lipids. Microtiter and gas-tight syringes were from Hamilton. Screw-cap fl uorescence cuvettes and magnetic stir bars
57 were from Starna Cells. Se ptum caps were from Supelco. All reagents were of the highest purity available. Experimental Methods Media preparation for bacterial cultures All the media for bacteria growth were prepared and stored aseptically. LB medium (for r outine growth of E. coli cells) contains 1% Bacto tryptone, 0.5% yeast extract, 1% NaCl. LB/agar plate contains: 1% Bacto trypton e, 0.5% yeast extract, 1% NaCl., and 1.5% agar. Terrific broth (enriche d medium for growing E. coli cells) contains: 1.2% Bacto tryptone, 2.4% yeast extr act, 0.4% glycerol, 0.17 M KH2PO4, and 0.72 M K2HPO4. SOB media (for recovery of transformed E. coli DH5 and BL21 cells) contains: 2% Bacto tryptone, 0.5% Bacto yeast extrac t, 10 mM NaCl, 2.5 mM KCl, 10 mM MgCl2 and 10 mM MgSO4. SOC media (for recovery of transformed vis cells) contains: 2% Bacto tryptone, 0.5% Bacto yeast extract, 10 mM NaCl, 2.5 mM KCl, 10 mM MgCl2, 10 mM MgSO4 and 20 mM glucose. MOPS media (for induced protein expression in E. coli cells) consists of the following components: 1 liter complete medium is prepared by mixing 200 mL M solution (described below), 2 mL O solu tion (described below), 0.1 mL P solution
58 (described below), 1 mL S solution (descr ibed below), 0.5 mL 0.2% thiamine, 40 mL 3.75% vitamin assay casamino acids, 20 mL 20% glucose, and supplemented with sterile H2O to a final volume of 1 liter. 200 mL M solution is composed of 4.2 % (w/v) MOPS, 0.4% (w /v) tricine, 0.8% (w/v) KOH, 1.46% (w/v) NaCl, 0.26% (w/v) NH4Cl and 0.42% (w/v) NaHCO3. M solution is sterilized by au toclaving and kept at 4 oC. 50 mL O solution is prepar ed by dissolving 0.1g FeCl2H2O in 10 mL HCl and 10 mL H2O, mixing with 1 mL T solution (d escribed below), adding 2.68 g MgCl2H2O, and filled with H2O to a final volume of 50 mL. O solution is filter-sterilized and kept at 4 oC. 100 mL P solution is prepared by dissolving 13.6 g KH2PO4 in 100 mL H2O. P solution is sterilized by au toclaving and kept at 23 oC. 100 mL S solution is prepar ed by dissolving 4.81 g K2SO4 in 100 mL H2O. S solution is sterilized by au toclaving and kept at 23 oC. 100 mL T solution is prepared by dissolving 184 mg CaCl2H2O, 64 mg H3BO3, 40 mg MnCl2H2O, 18 mg CoCl2H2O, 4 mg CuCl2H2O, 340 mg ZnCl2 and 605 mg Na2MoO4H2O in 8 mL concentrated HCl, and filled with H2O to a final volume of 100 mL. T solution is filter-sterilized and kept at 23 oC. Competent cell preparation a nd bacterial transformation The procedures described here were used to prepare competent cells for electroporation for the bacter ial strain BL21(DE3) and DH5 All reagents used were prepared sterilely. LB me dium was inoculated with E. coli cells stored as a frozen
59 glycerol stock and the inoculum wa s grown on a LB/agar plate at 37 oC for 16 hr. Bacterial cells from a single colony were used as an inoculum to grow a 20 mL culture in LB medium at 37 oC overnight with shaking at 220 RPM on a platform shaker. The saturated bacterial culture was diluted 50-fo ld into 1 liter LB medium. The freshly diluted culture was incubated at 37 oC with shaking at 220 RPM, until OD600 reached ~0.6 when measured using a 1 cm path-length cuvette, which usually took ~1-2 hr. The cells were pelleted by centrifuging at 6000 RP M in a Beckman JA-10 rotor for 10 min at 4 oC. Cell pellets were washed twice in ice-cold H2O, followed by two washes in icecold 10% glycerol. After the washes, cells were resuspended in 10% gl ycerol to a total of ~6 mL and aliquoted in ~200 L to pre-chilled microfuge t ubes. The aliquots were quickly frozen in liquid N2 and stored at -80 oC. For electro-transformation, typically, 40 L competent cells were incubated on ice with 1-2 L solution containing a total of ~1 ng to 1 g DNAs. The mixture was transferred to a pre-chilled 0.1 cm electroporation cuvette and electroporated usi ng a Bio-Rad Gene Pulser under the setting of 1.8 KV and 25 F. The mixture was grown in 1 mL SOC medium for 1-3 hr at 37 oC in a platform shaker at 220 RPM. 100-200 l of the transformation was sp read on an agar plate and incubated at 37 oC overnight. Transformants were ob tained from colonies developed on the plate. Glycerol stock preparatio n for bacterial cells An individual colony developed on an ag ar plate was inoculated into ~5 mL growth media and the culture was incubated at 37 oC overnight with shaking at 220 RPM. The next day, the culture was collected and mi xed with glycerol to make a stock solution
60 containing 10% glycerol. Ali quots of ~1 mL were transfe rred to microfuge tubes and stored at oC. Plasmid DNA purification Plasmids from bacterial cultures were is olated using Qiagen Tip-20 or Tip-100 according to the manufacturers protocol. For small-scale preparation, colonies of BL21 or DH5 cells were inoculated into ~10 mL gr owth media. The culture was grown overnight at 37 oC, and the plasmids were extracted usually at a yield of ~1 g/mL culture. Concentration of double-stranded DNA was determined by measuring OD260 in a 1 cm path-length cuvette using an extinction coefficient of 1OD = 50 g/mL DNA. An aliquot of the prepar ation, usually 0.2-0.5 g, was run on agarose gel, typically 0.8%, to assess purity of the plasmid. Sodium dodecyl sulfate-polyacylamide gel electrophoresis and protein concentration determination SDS-polyacrylamide gel electrophoresi s was run as described by LaemmLi (Laemmli 1970). Typically, 1-10 g of denatured protein samples were run on 15% acrylamide and 0.75-mm thick mini-gels. The denatured protein sample was prepared by mixing proteins with loading buffer cont aining 50 mM TrisCl pH 6.8, 2% SDS, 0.1% bromophenol blue and 10% glycerol, along with 0.1 M dithiothreitol dissolved in 1 mM NaAcetate pH 5.2. The mixture was boiled at 95 oC for 2-3 min, and an aliquot of ~1015 L was loaded onto a gel. Usually, gels were run at 100-120 V for 2-3 hr at 23 oC. Protein bands were visualized using Coom assie Brilliant blue (R-250) staining by
61 incubating gels in 0.25% Coomassie blue solu tion in 45% methanol and 10% acetic acid for a few hours, followed by destaining in 30% methanol and 10% acetic acid for overnight. Alternatively, gels were stained using Gelcode blue stain (Pierce). Gels were rinsed in ddH2O several times, followed by overnight incubation in Gelcode solution and destained in ddH2O for a few hours. Both methods allow ~1 g or more proteins to be shown. For improved sensitivity, gels were silv er stained, which allo ws visualization of ~ 10 ng or more proteins. Protein concentrations were determined by the bicinchoninic acid assay. In a 2 mL reaction, bicinchoninic acid and 4% (w/v) CuSO4 solution were mixed at a 50:1 ratio, and the mixture was incubated with 100 l protein sample. The reactions were allowed to proceed for 30 min at 37 oC, or 2 hr at 23 oC. At the end, the absorbance of each sample at 562 nm was measur ed using 1 cm path-length cuvette on a UVPC-2100U dualbeam spectrophotometer. A standard curve was generated using dilutions of 1 mg/mL bovine serum albumin (BSA) stoc k solution in the reaction. Construction of a random library and gene tic selection of functional ferrochelatase loop variants Electro-competent cells of the E. coli strain vis were prepared following a standard procedure. To complement defici ency in the endogenous ferrochelatase activity, LB medium was supplemented with 10 g/mL hemin and 0.4 % (w/v) glucose to allow growth of vis cells. To eliminate background from wild-type ferrochel atase in library construction, a mock vector was generated by replacing the segment in the murine ferrochelatase expression plasmid pGF47 wh ich spans the ten codons for the loop motif
62 x x x x x x x x x x x x x x x x x x x x x BstEII PspOMI A nnealing & extensiono f mutagenic primers PCR amplification o f mutagenic templates Restriction digest XPspOMI BstEIIRandomized ferrochelatase insert 60 bp Figure 4. Random mutagenesis of the ferro chelatase active site loop motif and biological selection of th e functional variants.
63 with a stuffer sequence derived from the murine ALAS expression plasmid pGF23 via two unique restrictions sites BstEII and PspOMI (Figure 4). The ferrochelatase random library was constructed based on previously described protocols (M unir et al. 1992; Gong and Ferreira 1995; Landis and Loeb 1998) (Fig ure 4). Two single-stranded DNA oligos with 15 complementary base pairs at their 3 ends were annealed. Oligo 1 (5-ATG GAA AAG CTG GGT TAC C CC AAC CCC TAC CGA CTG GTT TGG-3) is a 42mer corresponding to the sense strand nucleotides and contains a BstEII site ( italics ) for cloning. Oligo 2 (5-AGC GTC ATC TGT CTG A GG GCC C AA CCA GGG TAC TGG ACC AAC CTT GGA CTG CCA AAC CAG TCG GTA-3) is a 66mer spanning the antisense strand nucleotides with a PspOMI restriction site ( italics ) for cloning. Oligo 2 contains degenerate nucleotides (underl ined) corresponding to codons 248-257 of the mature murine ferrochelatase loop residues. Each of the ten loop codons was randomized using 85% wild-type nucleotides and 15% mixture of th e other three nucleotides. A 60 L annealing reaction, containing 0.5 nmole of each oligo, 200 mM Tris-HCl, pH 7.5, 100 mM MgCl2 and 250 mM NaCl, was incubated at 80 C for 5 min, followed by 55 C for 15 min, 37 C for 15 min and finally at 23 C for 30 min. The hybrid was extended using the Klenow fragment of E. coli DNA polymerase I (5 units) in a 40 L mixture containing 50 pmole of annealed oligos, 62.5 M of each of the four dNTPs and EcoPol buffer (10 mM Tris-HCl, pH 7.5, 5 mM MgCl2 and 7.5 mM dithiothreit ol). The extension reaction was carried ou t at 37 C for 2 hr. The ge nerated double-stranded mutagenic DNA was then amplified by the polymerase chain reaction (PCR) using two primers corresponding to the 5 termini of oligos 1 and 2. The two pr imers used were: FC-66, 5ATG GAA AAG CTG GGT TAC CCC AA-3, which covers the first 23 nucleotides of
64 the 5 terminus of oligo 1 and FC-R67, 5-AGC GTC ATC TGT CTG AGG-3, which corresponds to the 18 nucleotides of the 5 terminus of oligo 2. The PCR reactions were performed in 100 L reaction volumes consisting of 20 pmole of primers FC-66 and FCR67, 3-50 pmole of the extended double-stra nded DNA as template, 50 M each of the four dNTPs, 1 unit of VentR DNA polymerase, and ThermoPol reaction buffer (20 mM Tris-HCl, pH 8.8, 10 mM KCl, 10 mM (NH4)2SO4, 2 mM MgSO4 and 0.1% Triton X100). The reaction mixture was subjected to temperature cycling in a MJ Research programmable thermal controller by running 1 cycle of 95 C for 2 min, followed by 30 cycles of 95 C for 1 min, 45 C for 2 min, and 72 C for 1 min, and a final extension at 72 C for 10 min. The PCR product of 93 bp was purified using Microspin S200-HR columns, and subsequently digested with BstEII and PspOMI. The ~60 bp digested fragment was isolated from an agarose gel using GLASSFOG matrix and buffers in the MERmaid kit and subcloned into the mock v ector previously digested with the same enzymes. The 20 L reaction ligation contained the randomLy mutagenized insert and vector at a 5:1 molar ratio, 1 mM ATP, 7 un its of T4 DNA ligase, ligase buffer (50 mM Tris-HCl, pH 7.5, 10 mM MgCl2, 10 mM dithiothreitol, 1 mM ATP, and 25 g/mL bovine serum albumin), 100 mM NaCl and 10% PEG 6000. After 16-20 hr incubation at 16 C, the ligation reaction was purified by passing through Microspin S200-HR columns (Pharmacia), and used to transform competent vis cells with an Electroporator II (Invitrogen). Aliquots of 1-2 L of the ligated products we re electroporated into 40 L vis cells, using 0.1 cm cuvettes (Invitrogen) at 1.5 KV, 50 F and 150 After the pulse, transformants were resuspended in 1 mL SOC medium (2% Bacto tryptone, 0.5%
65 Bacto yeast extract, 10 mM NaCl, 2.5 mM KCl, 10 mM MgCl2, 10 mM MgSO4 and 20 mM glucose) supplemented with 10 g/mL hemin, and incubated at 37 0C for 4 hr with continuous shaking. To assess transforma tion efficiency, a small portion of the transformed cells was spread on to LB plates containing 50 g/mL ampicilin, 0.4 % (w/v) glucose and 10 g/mL hemin. The remaining transf ormed cells were plated onto LB/ampicilin plates to allow selection of the active variants. Only vis cells expressing functional ferrochel atase can form colonies on heminfree media. Aliquots of vis cells transformed with the ra ndom library ligation products were plated on LB/agar medium containing 50 g/mL ampicilin and 0.4 % (w/v) glucose, and incubated at 37 C for 24-30 hr. Plasmids were isolated from all of the surviving colonies, and the randomized regions were sequenced to determine the permissible amino acid substitutions. To prepare doublestranded DNA templates, ~20 mL of vis culture for each clone was grown overnight in terrific br oth containing 50 g/mL ampicilin and 0.4 % (w/v) glucose, and plasmids were pu rified using Qiagen Tip-20 (Qiagen). Sequences were determined for the sense st rand using the primer MFC-73: 5-AGA GGG GAC CCC TAT CCC CAA GAG GT A-3, which is 63 bp 5 to the randomized loop region. 166 clones were sequenced by the dideoxynucleotide chain termination method using a T7 sequenase 2. 0 DNA sequencing kit and 35S-dATP. The remaining 48 clones were sequenced by the ABI automated DNA sequ encers at the University of Florida. Large-scale purification of the wild-ty pe ferrochelatase and loop variants The wild-type murine ferrochelatase and the functional loop variants were expressed under the E. coli alkaline phosphatase promoter phoA in BL21(DE3) cells or
66 DH5 cells as described previously (Ferreira 1994). Bacterial colonies were used to inoculate ~100 mL LB/amp media and th e culture was grown overnight at 37 oC with shaking at 220 RPM. Approximately 10 mL of the overnight culture was used to inoculate 1 liter of MOPS medium contai ning 50 mg/liter ampicilin. The MOPS culture was grown at 37 oC for 4 hr with shaking at 220 RPM in a Beckman platform shaker, and the growth was continued at 25-30 oC for 16-20 hr. Typically, 8 liters of MOPS culture were grown. Bacterial cells were pe lleted at 6,000 RPM for 10 min at 4 oC in a Beckman JA-10 rotor. Cells were resuspended in ~ 30 mL solution containing 20 mM TrisCl pH 8 and 10% glycerol. Cells were homogenized, an d ruptured using a Fr ench press with 3-4 passages at 14,000 psi. To reduce proteolysis, 0.0017% (w/v) phenylmethylsulfonyl fluoride was included in the lysate. Lysate was resuspended in a so lution containing 0.6 M NaCl, 0.5% cholate, 20 mM TrisCl pH 8 a nd 10% glycerol under constant stirring at 4 oC for 30 min. The solubilized lysate was centrifuged at 44,000 RPM using a 50.2Ti rotor in a Beckman ultracentrifuge for 1 hr at 4 oC. The supernatant fraction was saved and used for loading onto the affinity columns. For protein purification using blue se pharose resins, the supernatant from ultracentrifugation was mixed with 35% (NH4)2SO4 and stirred at 4 oC for 20 min. The mixture was centrifuged at 11,000 RPM for 30 min. The pellet fraction was homogenized in ~20 mL equilibration buffe r containing 20 mM TrisCl pH 8, 10% glycerol, 0.5 M NaCl and 0.5% NaCholate. The homogenate was lo aded onto a 15 x 2.5 cm gravity column packed with ~20 mL blue sepharose resins. The column was washed in ~150 mL equilibration buffer until A280 < 0.1. Subsequently, the column was washed with ~100 mL wash buffer containing 20 mM TrisCl pH 8, 10% glycerol, 1 M NaCl and
67 0.5% NaCholate until A280 < 0.1. Proteins were eluted in elution buffer containing 20 mM TrisCl pH 8, 10% glycerol, 1.5 M NaCl and 1% NaCholate. Elution fractions, ~50 mL total, were pooled and concentrated in a 50 mL Amicon stirred cell to ~1-5 mL. Aliquots of ~200 L of purified concentrated protei ns were stored in liquid N2 until use. For purification of ferrochelatase with N-terminal His-tag using metal chelate affinity resins, the supernatant fraction obtai ned from ultracentrifugation was loaded onto a 17 x 1.5 cm gravity column packed with ~10 mL Talon resins equilibrated in buffer containing 20 mM TrisCl pH 8, 10% glycerol 150 mM NaCl and 0.5% NaCholate. The column was first washed with ~200 mL e quilibration buffer until A280 < 0.1. It was further washed in ~100 mL equilibration buffer supplemented with 10 mM, 20 mM and 30 mM imidazole. Proteins were eluted us ing equilibration buffe r containing 100 mM imidazole. The protein in the elution frac tions in a total of ~100 mL was pooled and concentrated in an Amicon stirred cell. To remove imidazole, purified samples were loaded onto a 100 x 2 cm HPLC column pack ed with ~150 mL Superdex-200 resins in equilibration buffer. The proteins were elut ed at a flow rate of 0.4 mL/min and elution fractions were collected and concentrated. UV-visible absorbance spectra of purified ferrochelatase UV-visible absorption spectra of purifie d ferrochelatase and variants were recorded using a UVPC-2100U dual-beam sp ectrophotometer (Shimadzu) with 1 cm path-length cuvette at 23 oC. Typically, a prot ein solution at ~20 M was added to the cuvette, and the absorption spectrum was co llected across the wavelength range of 250700 nm.
68 Metal content analysis of purified ferrochelatase The metal ion content of purified ferroch elatase was determined using plasma emission spectroscopy. His-tagged wild-type ferrochelatase was purified using bluesepharose affinity column chromatogr aphy and variant S 249A/K250Q/V251C was purified using Talon metal chelate affin ity column chromatography followed by gel filtration chromatography. The proteins we re concentrated in a centricon (MWCO 10,000 Da). The sample for plasma emission spectroscopy was prepared in 1 mL assay buffer containing 20 mM TrisCl pH 8, 10% glycerol and 150 mM NaCl. Metal ion concentrations in 60 M solution of His-tagged w ild-type ferrochelatase, 40 M and 23 M solutions of variant S249A/K250Q/V25 1C and the assay buffer alone, were determined using 20-element analysis at the Chemical Analysis Laboratory at the University of Georgia. Pyridine-hemochromogen assay A conventional method to measure ferroch elatase activity is by quantifying heme production using the pyridine-hemochromogen assay. A 2 mM protoporphyrin IX stock solution was prepared by dissolving protoporphyrin in 100 mM NH4OH and 0.5% (v/v) Tween-80. A 4 mM ferrous iron stock solution was prepared anaerobically by dissolving 15.7 mg Fe(NH4)2(SO4)2H2O and 27 mg Na3Citrate in 10 mL deaerated H2O. A mixture containing the enzyme incuba ted with a solution containing 100 M protoporphyrin and 100 mM TrisCl pH 8 was deareated under Argon flow. Reaction was initiated by adding the ferrous iron solu tion to a final concentration of 100 M. The
69 entire reaction was incubated at 23 oC for 20 min. At the end of incubation, 0.5 mL of 1 N NaOH and 0.5 mL pyridine were added to th e mixture. Half of the sample was reduced with dithionite, and the other half was used as reference. The difference spectrum of the reduced sample versus th e oxidized sample was collected at 23 oC using a UVPC-2100U dual-beam spectrophotometer in 1 cm path-length cuvette. Wavelength scan was run between 500 nm and 600 nm. The absorbance differential between the maximum at 557 nm and minimum at 541 nm was recorded and used to calculate the amount of pyridine-hemochromogen form ed using the extinction coefficient A557A541) of 20.7 mM-1 cm-1 (Porra and Jones 1963a). Continuous assay of ferr ochelatase activity An initial rate assay was developed for ferrochelatase by monitoring the decrease of porphyrin fluorescence using protoporphyrin IX and ferrous iron as substrates. Caution was taken in preparing the assay reagents in order to minimize trace metal ion contamination. All of the buffers and solutions including 1 M Tris-acetate pH 8.1 and 10% (v/v) Tween-80 were made in HPLC-grade H2O and further purifi ed using Chelex100 resins. Protoporphyrin IX was prepared as 200 M stock solution by dissolving porphyrin solid in 100 mM NH4OH and 0.5% (v/v) Tween-80. The solution was adjusted to pH 8.1 with acetic acid, and kept in 0.1 M Tris-acetate pH 8.1. Ferrous iron was prepared under anaerobic conditions as a 2 mM ferrous ammoni um citrate stock by dissolving equimolar amounts of ferrous ammonium sulfate and sodium citrate, i.e. 7.8 mg Fe(NH4)2(SO4)2H2O and 13.6 mg Na3Citrate in 10 mL H2O (Taketani and
70 Tokunaga 1981). All glassware used in the assays was so aked in HCl overnight to remove traces of iron. Protoporphyrin fluorescence was mon itored on a Shimadzu RF-5301PC fluorimeter equipped with a red-sensitive phot omultiplier tube. The thermostatically controlled cell holder was maintained at 30 oC. A 2 mL solution containing protoporphyrin IX in 0.1 M Tris-acetate pH 8.1 and 0.5 % (v/v) Tween-80 was deaerated in a fluorescence cuvette sealed with a se ptum cap and contai ning a micro-magnetic stirring bar. Deaeration continued for at le ast 30 minutes by applying repeated cycles of vacuum and purging with ultra-high purity argon gas. The cuvette was then placed in the fluorimeter cell holder. Purified ferrochel atase was added to the mixture by injection using a fixed-needle syringe, and the mi xture was equilibrated for 5 min at 30 oC under constant stirring. The reaction was initiated by injecting ferrous ir on, and the progress of the reaction was monitored continuously by recording the emitted light through an emission monochromator set at 635 nm (5-n m slit width) upon excitation at 505 nm (3nm slit width). The rate of change in fluorescence intensity (arbitrary unit min-1) was converted to velocity (n mol of protoporphyrin min-1) using a standard curve of fluorescence intensity versus protoporphyrin concentration. The time course for each reaction was collected using the Shimadzu so ftware RF-5301PC and the initial rates were calculated from the linear portion of the progress curves. The steady-state kinetic parameters Km Fe2+, Km PPIX and kcat of wild-type ferrochelatase were determined at 30 oC using the continuous fluorimetric assay as described above. The data were analyzed in matrices of five pr otoporphyrin and five Fe2+ concentrations by fitting the in itial velocities to Equation 1:
71 ][PPIX] [Fe [PPIX] ] [Fe [PPIX] ] Fe ][ PPIX [2 Fe2 m 2 PPIX m Fe2 i 2 max K K K V v (Eq.1) where Km PPIX and Km Fe2+ are the Michaelis consta nts for protoporphyrin and Fe2+, Ki Fe2+ is the dissociation constant for Fe2+, v is the initial velocity, and Vmax is the maximal velocity of the reaction. Equa tion 1 describes the steady-stat e velocity equation for a bireactant system in the abse nce of products (Segel 1975). The kinetic constants and standard deviations were calculated from regr ession analysis using th e statistical program DATAFIT (Oakdale Engineering). Steady-state kinetic analysis of the loop variants Ferrochelatase activity was determined by monitoring the consumption of substrate protoporphyrin using ferrous iron as the metal substrate in a continuous spectrofluorimetric assay conducted under strictly anaerobic conditions as described previously (Shi and Ferreira 2003). Typi cally, a mixture contai ning 100 mM Tris-acetate pH 8.1, 0.5% (v/v) Tween-80 and protoporphyrin IX was incubated with ferrochelatase for ~5 min at 30 C, and the enzymatic r eaction was initiated by ferrous ammonium citrate injection. Activities are expressed as nmole protoporphyrin consumed per minute per mg of enzyme. The steady-state kinetic pa rameters were determined from matrices of five protoporphyrin and five Fe2+ concentrations. The reported values and the standard deviations were obtained from the nonlinear leas t squares fit of the data to the MichaelisMenten equation for bi-reactant systems using the software DataFit.
72 Homology modeling of murine ferrochelatase Comparative protein modeling of the thre e-dimensional structures of wild-type murine ferrochelatase and selected loop va riants was performed using the amino acid sequence and coordinates for human ferrochelat ase (PDB code 1hrk) as the template. Sequence alignments, molecular modeling, energy minimization and PROCHECK model analyses were performed on the Geno3D server s provided by the Institute of Biology and Chemistry of Proteins (IBCP) in Lyon, Fran ce (Combet et al. 2002). The monomeric models for wild-type ferrochelatase and the loop variants were vi sualized using the program VMD version 1.7 (Humphrey et al. 1996). The secondary structures in the protein models were identified by the STRIDE algorithm included in VMD. Resonance Raman spectroscopy of porphyrin binding to the wild-type ferrochelatase and loop variants Stock solutions of protoporphyrin, hemin and Ni-protoporphyrin were prepared by dissolving porphyrins in 100 mM NH4OH and 0.5% (v/v) Tween-80. Typically, porphyrin stocks were made as 2 mM solutions, and further diluted in equilibration buffer containing 20 mM TrisCl pH 8, 150 mM NaCl and 10% glycerol prior to incubation with proteins. Wild-type ferrochelatase and the variants were purified by Talon metal chelate affinity column without adding detergent so dium cholate during th e entire pur ification procedure. Aliquots of ~200 L of purified proteins (typically at 100-150 M) in equilibration buffer were stored in liquid N2 until use. To prepare samples for resonance Raman spectroscopy, porphyrins (typically 6 M) were incubated with ferrochelatase (typically 60 M) in equilibration buffer on ice and the final concentration of detergent
73 Tween-80 was adjusted to 12 M. Each sample, in a final volume of ~100 L, was transferred to a stoppered 3 x 3 mm cross-section optical ce ll (NSG Precision Cells) for spectra collection. The spectra we re routinely recorded at 26 oC for 2-10 min using 20 mW laser power. The spectra were record ed on a Raman spectrometer by applying the 406.7-nm line of an INNOVA 304 Kr+ laser (Coherent) as previ ously described (Lu et al. 2002). Briefly, the spectrometer consists of a 0.75m monochrom ator with a 2048channel liquid N2-cooled CCD detector (Instruments, SA ). Position mode was used for CCD detection, with each section covering about 500 cm1 of the Raman spectrum without moving the grating. The resonance Raman spectra were exporte d as even-X ASCII files for plotting with SigmaPlot (SPSS). Lorentzian decomposition of the spectra was performed using the program PeakFit (SYSTAT). Curve-fitti ng was used to obtain the frequencies for the centers of the structure-sensitive bands. For the region near the saddling-symmetry (B2u) mode 15, the spectral fitting procedure was adjust ed to take into account the varying signal-to-noise characteristics of the lowfrequency Raman spectra for protoporphyrin binding to the proteins. First, the sp ectra of protoporphyrin bound to wild-type ferrochelatase and the quadruple mutant Q248P/S249G/K250P/G252W, which exhibited the highest signal-to-noise level, were fit with all parameters free, including the center, width and amplitude of each Lorentzian band, as well as the slope and intercept of a linear baseline. Eight lines were used to fit the frequency region from 580 to 780 cm-1, and the average half-width at half maximum of 15 was determined to be 2.814 cm-1. This value for the width was then fixed and the data were refit. These parameters were used for the constrained fits to yield the simulation results.
74 Profiling the active variants by hi gh-throughput protein purification Plasmids encoding all the act ive loop variants were transformed into BL21 cells and the variant proteins were purified using a TALON HT 96-well plate. BL21 transformants were inoculated into ~ 10 mL LB/ampicilin media and grown at 37 oC overnight. Typically, ~5 mL culture was in oculated into 150 mL MOPS media, and the culture was grown at 37 oC for 4 hr and then at 25 oC overnight. Bacterial cells were pelleted at 4,000 RPM in an Eppendorf centr ifuge, resuspended in a 10 mL solution containing 20 mM TrisCl pH 8 and 10% glycerol. The cel ls were lysed by passing through the French press twice. The lysate was solubili zed in 20 mM TrisCl pH 8 containing 0.6 M NaCl, 0.5% NaCholate and 10% glycerol with stirri ng for 1 hr at 4 oC. The mixture was centrifuged at 15,000 RPM for 30 min and the supernatant fraction was saved. In a TALON 96-well plate, each well contained 200 mg BD Talon Superflow resin and the resin was equili brated with a buffer of 20 mM TrisCl pH 8 containing 10% glycerol, 150 mM NaCl and 0.5% NaCholate. The supernatant was loaded into each well. The wells were then washed with ~15 mL equilibration buffer until A280 < 0.1. Further washes were in ~5-10 mL equilibra tion buffer supplemented with 10 mM and 20 mM imidazole. Ferrochelatase was eluted using 2 mL equilibration buffer supplemented with 100 mM imidazole. The eluate was collected in a 96-well plate. Eluate for each sample was concentrated to ~100 L in an Amicon Ultra-4 ce ntrifugal filter unit (MWCO 10,000). Imidazole in the sample was removed by exchanging the proteins into a storage buffer of 20 mM TrisCl pH 8 containing 10% glycerol and 150 mM NaCl. UV-visible absorbance scan was taken for each purified variant.
75 Liposomal binding assays of ferrochelatase variants Liposomes were formulated with a lipid composition similar to that found in the inner mitochondrial membranes of mouse liver cells (Ardail et al. 1990). Biotinylated liposomes were prepared following a previous ly described procedure (Zhu et al. 2001). Each lipid was weighed and suspended in chlo roform to make a 1 mg/mL stock solution. The lipid stocks were mixed in a glass tube at ratios (w/w) of 1% biotin-DHPE, 35% phosphatidylcholine, 26.5% phosphatidylet hanolamine, 18% cardiolipin, 5% phosphatidylinositol, 2.3% cholesterol, 2.8% diacylglycerol, 0.7% sphingomyelin, 0.5% phosphatidylserine, 0.5% lysophosphatidylchol ine, 0.3% lysophosphatidylethanolamine, 2.2% palmitic acid, 1.2% stearic acid, 1.3% oleic acid, 1.8% li noleic acid and 0.9% arachidonic acid. The mixture was dried under N2, resuspended in 25 mM TrisCl pH 7.6 and 150 mM NaCl (TBS). The resuspension was sonicated to obtain a 100 g/mL solution for use as a 1000X liposome stock fo r protein-binding assay. Liposome size distribution was determined by dyn amic light scattering at 26 oC using a submicron particle size analyzer (Beckman Coulter) at the University of New Mexico. Purified ferrochelatase was diluted to 1 g/ L in equilibrati on buffer containing 20 mM TrisCl pH 8, 10% gly cerol, 150 mM NaCl and 0.5% NaC holate. A serial dilution by 70% was made for each protein stock. 10 L of the diluted samples were spotted onto nitrocellulose membrane (~10 x 15 cm) and ai r-dried for 30 min. The blot was then wetted in H2O followed by TBS. For binding assay, th e blot was first blocked with 50 mL Superblock buffer (Pierce) for 2 hr at 23 oC, followed by washing in TBS with 0.05% (v/v) Tween-20 (TBST) 3 times for 10 min eac h. The blot was then transferred to a
76 hybrization bottle (Amersham), and incubated with 50 L liposome stock solution diluted in 50 mL TBS. The hybridization bottle wa s placed on a rotor in a hybridization oven (Fisher Scientific) at 30 oC for 1 hr. Subsequently, the bl ot was washed in TBST three times for 15 min each at 23 oC. For biotin detection, th e blot was incubated with 1 L peroxidase-conjugated extravid in (Sigma) at 2 mg/mL diluted in 50 mL TBS for 30 min, at 23 oC. Final washes were in TBST twice for 15 min each, and in TBS three times for 15 min each. In the last ste p, the blot was treated with chemiluminescence substrate solutions (Amersham Biosciences) according to the manufacturers instructions and exposed to a Hyperfilm (Amersham Bioscien ces). The bound liposomes were quantified using the densitometry program supplied with ChemiImager 4400 (Alpha Innotech). The relative binding affinity was calculated from the intensity of chemiluminescent signals plotted against the amount of protein spotte d onto the nitrocellulose membrane. Wildtype ferrochelatase wa s run as the standar d, and a variant (G252D/V 254I) was used as an internal control; essentially G 252D/V254I was included in every single blot to ensure that the ratio between the chemiluminescent signal intensity for wild-type protein and the variant remained constant from blot to blot. Inhibition assay of ferrochelatase by N -methyl protoporphyrin on agar plates A 2 mM stock solution of N -methyl protoporphyrin (NMPP) was made by dissolving NMPP in 20 mM NH4OH containing 2% (v/v) Tween-20. The solution was filter-sterilized and kept at 4 oC. The stock solution was diluted 100-, 1000and 10,000fold to prepare plates containing 20 M, 2 M and 0.2 M NMPP. All plates were kept at 4 oC in the dark. vis cells transformed with wild-type ferrochelatase and functional
77 variants were grown in LB medium containing 50 g/mL ampicilin and 0.4 % (w/v) glucose at 37 oC overnight with shaking at 220 RPM. Each of the overnight cultures was diluted 20-fold into fresh media, and grown at 37 oC for ~1 hr until OD600 reached ~0.40.5. Subsequently, these cultures we re diluted with LB medium by 106-fold and 100 L aliquots of the diluted stock were spread on plates containing LB/ampicilin only or with added NMPP at 0.2 M, 2 M and 20 M, respectively. The plates were incubated at 37 oC for 16 hr, and the number of colonies developed on each plate was counted. Quantification of N -methyl protoporphyrin bindi ng to ferrochelatase by fluorescence quenching measurements Purified wild-type ferrochelatase and vari ants P255R and P255G were diluted into a 3 mL solution containing 10 mM Tris-aceta te and 0.05% (v/v) Tween-80 at pH 8. Protein fluorescence was excited at 283 nm and monitored at 331 nm at 23 oC using a Shimadzu RF-5301PC fluorimeter. Equilibrium binding of N -methyl protoporphyrin to the protein was monitored by quenching the intrinsic protein fluorescence. A stock solution of NMPP at 200 M was made by dissolvi ng NMPP in 100 mM NH4OH containing 0.5% (v/v) Tween-80. Aliquots were added to th e protein solution. After incubating on ice for 1 hr, fluorescence of the mixture was measured as described above. The fluorescence intensity was plotted agai nst NMPP concentration and the dissociation constant of NMPP for ferrochelatase was determ ined using the least-squares fit of each data set to Equation 2 by nonlinea r regression (Anderson et al. 1988): t t d t d tE L E K L E K L E F F F 2 42 0 (Eq.2)
78 where Kd is the dissociation constant, F is the measured fluorescence, F0 is the fluorescence in the absence of NMPP, F is the quenched fluorescence, Et is the total protein concentration, and L is the total NMPP concentration. Transient kinetic analysis of ferrochelatase activity The kinetic steps involved in NMPP binding to ferrochelatase were assessed using stopped-flow absorption spectroscopic analys is by monitoring the enzymatic reaction in the presence of various concen trations of the inhibitor NMPP. The reactions were run on an OLIS model RSM-1000 stopped-flow spect rophotometer equipped with a stoppedflow mixer and observation chamber of 4 mm op tical path-length. The dead time of the instrument is ~2 ms. Wavelength scans spanning 347 nm to 574 nm were collected at 1000 scan/sec. The syringes containing the reaction components and the stopped-flow cell were maintained at 30 C using an external water-bath. The concentrations of reagents loaded into each syringe were twi ce the final concentrations in the observation chamber upon mixing, which were reported as the actual values. Enzymatic activity was monitored by following the increase in abso rbance at 420 nm for zinc-protoporphyrin production using Zn2+ and protoporphyrin as substrates. The enzyme and substrate stock solutions were diluted in a buffer of 100 mM Tris-acetate pH 8 containing 0.5% (v/v) Tween-80. A 3 mM zinc-acetate stock solution was made by dissolving zinc-acetate in ddH2O. A 200 M protoporphyrin stock solution was prepared by dissolving protoporphyrin in 50 mM NH4OH and 0.5% (v/v) Tween-80. A 2 mM N -methyl protoporphyrin stock solution wa s prepared by dissolving NMPP in 100 mM NH4OH containing 0.5% (v/v) Tween-80. Typically, a 2 mL solution
79 containing 1 M ferrochelatase, 10 M protoporphyrin and various amounts of NMPP was pre-incubated on ice for 30 min. The mixture was subseque ntly transferred to one of the stopped-flow syringes. The other stopp ed-flow syringe was loaded with ~2 mL reaction buffer containing 30 M zinc-acetate. The enzymatic reaction was initiated by mixing the contents from both syringes. Reaction progression was monitored by following the increase in absorbance at 420 nm. The data were analyzed by fitting the absorbance values to Equation 3: kt s se k v v t v A A 10 0 (Eq.3) where A is the observed absorbance, A0 is the initial absorbance, s is the steady-state velocity, 0 is the initial velocity, and k is the pseudo-first order ra te constant for inhibitor binding. Equation 3 describes the progress of enzymatic reactions in the presence of tight-binding competitive inhibitors (Williams and Morrison 1979). Ligand binding pocket size measurement Structural models of w ild-type ferrochelatase and variants P255R and P255G were generated using the amino acid sequen ce and coordinates for human ferrochelatase (PDB code 1hrk) as a template. Sequen ce alignments, molecular modeling, energy minimization and PROCHECK model analysis were performed on the Geno3D servers at the Institute of Biology and Chem istry of Proteins (IBCP) in France (Combet et al. 2002). Dimensions of the porphyrin binding pocket in the monomeric models of wild-type protein and the variants were calculated on th e CASTp server (Liang et al. 1998). Values for the area and volume of Connollys surface of the active site cavity were determined using the CAST program (Liang et al. 1998).
80 Enzymatic activity of ferrochelatase in the absence of FeS cluster synthesis RZ4500 cells ( iscS+) developed large colonies on LB/agar plates, whereas small colonies of PK4331 cells ( iscS-) were formed on LB/agar plates containing 30 g/mL kanamycin after overnight incubation 37 oC upon inoculating the cells stored as frozen glycerol stock (Schwartz et al. 2000). Compet ent cells of both strains were prepared for electroporation following previously descri bed procedures. RZ4500 cells from a single colony were inoculated into 10 mL LB medium and grown overnight at 37 oC with shaking at 220 RPM. The saturated culture was diluted 100-fold into 1 liter LB medium, and grown at 37 oC with shaking at 220 RPM until OD600 reached ~0.6, which usually took ~2 hr. Cells were washed twice in H2O, then twice in 10% glycerol. Cells were resuspended in 5 mL of 10% aqueous glycer ol. Aliquots of 0.5 mL suspension were rapidly frozen a nd stored at -80 oC. PK4331 cells from a singl e colony were inoculated into 40 mL LB medium containing 30 g/mL kanamycin and grown overnight at 37 oC. The saturated culture was diluted 100-fold into 4 liter LB/kanamycin medium, and grown at 37 oC with shaking at 220 RPM until OD600 reached ~0.6, which took ~6 hr. Cells were washed twice in H2O, then twice in 10% glycerol. Cells were resuspended in 20 mL 10% glycerol. Aliquots of 0.5 mL suspen sion were rapidly frozen and stored at -80 oC. Aliquots of 40 L electro-competent RZ4500 cells and 100 L electro-competent PK4331 cells were mixed separately with pl asmids encoding wild-type ferrochelatase, single mutant C341A, triple mutants S249A/K250Q/V251C and S249A/K250R/G252W. The mixtures were transferred to pre-ch illed 0.1 cm electroporation cuvettes and
81 electroporated at 1.8 KV, 25 F using a Bio-Rad Gene Pulser. Transformation of RZ4500 cells was grown in SOB medium at 37 oC for 1 hr and plated onto LB/ampicilin plates. Transformation of PK4331 cells was grown in SOB medium at 37 oC for 5 hr and plated onto LB/ampicilin/kanamycin plates Subsequently, 50 mL MOPS culture was grown for each transformant. The culture was pelleted, and the pelle t was resuspended in 20 mM TrisCl pH 8 and 10% glycerol. Th e resuspension was passed through French press twice, and the lysate was solubilized by incubating with 0.6 M NaCl and 0.5% NaCholate with stirring at 4 oC for 30 min. 10 mL total lysate was obtained for each transformant. Enzymatic activity was measured using a zinc-chelatase assay as described previously (Shi and Ferreira 2004). A 2 mL reaction mixt ure was set up by adding 200 L cell lysate, 3 M ZnCl2 and 2 M protoporphyrin in a buffer of 100 mM TrisCl pH 7.6, and the mixture was incubated for 1 hr at 37 oC. The amount of ZnPP formed was measured by monitoring fluorescence at 592 nm upon excitation at 421 nm. Molecular mass assessment of purified ferrochelatase The molecular mass of purified ferrochel atase was assessed using size exclusion column chromatography. A 100 x 2 cm HPLC column packed with ~150 mL Superdex200 resin was equilibrated with 20 mM Tris Cl pH 8 containing 150 mM NaCl and 5% glycerol. The flow rate was 0.4 mL/min. Pu rified wild-type ferrochelatase and variants S249A/K250R/G252W and S 249A/K250Q/V251C were injected in 100-200 L aliquots at a protein concentration of 240-800 M. Elution fractions were collected at 1 mL intervals, and absorbance at 280 nm of each fraction was measured using a Shimadzu UVPC-2100U dual-beam spectr ophotometer with 1 cm path-length cuvette at 23 oC. The
82 column was calibrated using marker proteins of known molecular masses which included amylase (200 kDa), alcohol dehydrogenase (150 kDa), albumin (66 kDa), carbonic anhydrase (29 kDa) and cytochrome c (12.4 kD a). Enzymatic activity of the eluted protein was determined using a zinc-chelatase assay as described previously (Shi and Ferreira 2004). Dynamic light scattering was used as an alternative method for molecular size measurement. Approximately 1 mg each of the purified wild-type ferrochelatase and variants S249A/K250Q/V251C and S249A/K250R/G252W was diluted into a 4 mL buffer of 20 mM TrisCl pH 8 containi ng 150 mM NaCl, 0.5% NaCholate and 10% glycerol. The molecular mass distributio n in each sample was analyzed at 23 oC on a particle size analyzer at the Particle Engin eering Research Center at the University of Florida. Electron paramagnetic resonance spectro scopy of purified ferrochelatase EPR spectra were collected for wild -type ferrochelatase and variants S249A/K250R/G252W and S249 A/K250Q/V251C. Wild-type protein was purified using blue-sepharose affinity resin (Ferreira 1994; Shi a nd Ferreira 2003) or Talon metal chelate affinity resin (Shi and Ferreira 2004) while the variants were purified using Talon resin as described prev iously (Shi and Ferreira 2004) X-band EPR spectra were collected in collaboration with Dr. David Ti erney at the University of New Mexico. Purified proteins were reduced by titrating with dithionite at an equal molar ratio. Samples in the concentration range of 190-400 M were analyzed by spectroscopy.
83 Chapter Three Results Purification of recombinant ferrochelatase Large-scale purification of wild-type ferrochelatase and loop variants In order to obtain large qua ntities of purified ferrochelatase for characterization, recombinant wild-type protein and variants we re isolated from an overexpression system by growing the BL21 cells or DH5 transformants in complete MOPS media. Wild-type protein and selected variants were typically purified by affinity column chromatography using blue-sepharose resins or metal chelate re sins. Interestingly, while certain variants, including triple mutant K250M/V251L/W256Y, were difficult to purify by either bluesepharose or Ni-NTA agarose resins followi ng the standard procedures, the problem was circumvented by applying cobalt-based metal chelate affinity resins (Talon resins). Subsequently, Talon resins were used routin ely to purify His-tagged ferrochelatase. On average, 1-2 mg purified protei ns could be obtained from 1 liter culture. The yield was lower for some variants in cluding K250M/V251L/W256Y and Q248P/S249G/K250P/G252W by 2-3 fold compar ed to wild-type ferrochelatase. Typically, purified proteins were concentrated to 1-10 mg/mL (20-200 M) and stored in liquid N2.
84 The UV-visible absorption spectra of pur ified ferrochelatase exhibit distinct features derived from the FeS cluster (Fig ure 5). In wild-typ e ferrochelatase, one monomeric unit is associated with one [2Fe-2S] cluster, giving rise to the characteristic absorbance band at 330 nm and shoulders at 440 nm and 525 nm. Typically, for wildtype protein, the absorbance ratio A330/ A278 is 3 and A440/ A278 is 6. Most variants exhibited UV-visible spectra similar to wild-type ferrochelatase. Variants S249A/K250R/G252W and S249A/K250Q/V251C showed absorption features with some alterations and were further analyzed. Purity of the proteins was checked by SDSPAGE. Similar to wild-type ferrochelatase, purified variants migr ated on the SDS-PAGE gel with an estimated molecular mass of 42 kDa (Figure 6). Small-scale purification of fe rrochelatase loop variants All of the 33 active variants generated in the random library were over-expressed and purified from BL21 cells using a Talon-HT 96-well plate. The yield of purified proteins obtained by the plate method was si milar to that obtained from large-scale cultures. Typically, 1-2 g ferrochelatase could be isolated from 1 mL MOPS culture. Purified proteins, 20-200 g in total, were concentrated to 200 L for use in the concentration range of 0.1-1 mg/mL (2-20 M). The triple mutant S249A/K250R/G252W was produced at a lower yield of ~0.3 g/mL culture. The lowest yield was observed in the double mutant V 254I/L257K, which could only be obtained at ~0.01 g/mL culture.
85 Wavelength (nm) 250340430520610700 Absorbance 0.0 0.2 0.4 0.6 0.8 1.0 b d a c e Figure 5. The UV-visible absorbance sp ectra of purified wild-type murine ferrochelatase and selected loop variants. The recombinant wild-type ferrochelatase and variants were over-produced in E. coli DH5 or BL21(DE3) cells and purified using bl ue-sepharose or metal chelate affinity column chromatography. The absorption spect ra are shown for: (a) wild-type protein (8.8 M), (b) single mutant V251L (15.2 M), (c) triple mutant S249A/K250Q/V251C (18.8 M), (d) triple mutant K250M/V251L/W256Y (10.5 M), (e) quadruple mutant Q248P/S249G/K250P/G252W (19.4 M). Similar to wild-type ferrochelatase, most variants e xhibited the electronic absorption features characteristic of [2Fe-2S] cluster-containi ng proteins, including absorption bands at 330 nm, and shoulders at 440 nm and 525 nm. Variant S249A/K250Q/V251C lacked the absorption shoulder at 440 nm.
86 1 2 3 29 36 50 93 113 kDa Figure 6. SDS-polyacrylamide gel electro phoresis of purified ferrochelatase. Migration of purified murine ferrochelatase with N-terminal 5X -His-tag on 15% SDSpolyacrylamide gel is shown. The estimat ed apparent molecular mass for wild-type protein (10 g) was 42 kDa (lane 2). Va riant Q248P/S249G/K250P/G252W (5 g) exhibited a similar size (lane 3).
87 Developing a continuous assay for steady-s tate kinetic analysis of ferrochelatase A continuous assay was developed to de termine ferrochelatase activity based on the fluorescence properties of porphyrin. Th e substrate protoporphyrin IX is fluorescent, whereas the reaction product protoheme does not fluoresce. Thereby, it was possible to monitor the progress of the ferrochelatasecatalyzed reaction with ferrous iron and protoporphyrin IX as subs trates by following the d ecrease of protoporphyrin fluorescence. To minimize inner filter effect associated with excitation in the Soret band at 406 nm, porphyrin fluorescence was monitored using ex= 505 nm and em = 635 nm (Figure 7). Strictly anaerobic conditions we re imposed during the assay in order to maintain iron substrate in the reduced state. Reactions were conducted at the optimal pH of 8, and 0.5% (v/v) Tween-80 was included in the medium to optimize solubilization of protoporphyrin. Under these conditions, the in itial rate of the r eaction was obtained from the slope of the progress curve by monitori ng the disappearance of substrate porphyrin fluorescence (Figure 8). The rate of porphyr in disappearance was found to be linearly dependent on the concentration of wild-typ e enzyme in the range of 25-150 nM (Figure 9). Thus, the assay allows enzymatic activity to be determined for ferrochelatase in the nM concentration range, in contrast to the conventional pyridine-hemochromogen method, which requires a relatively large am ount of protein (>100 nM). Steady-state kinetic analysis of wild-type ferrochelatase indicates that the enzymatic activity exhibited hyperbolic dependence on prot oporphyrin concentration at various ferrous iron concentrations (Figure 10). The calculated Km values for protoporphyrin IX and ferrous iron are 1.4 0.2 M and 1.9 0.3 M, respectively, while kcat is 4.0 0.3 min-1.
88 A B Figure 7. The fluorescence spectra of protoporphyrin. The spectra were shown for protoporphyrin IX (1 M) in assay buffer containing 100 mM Tris-acetate pH 8 and 0.5% (v/v) Tw een-80. (A) The excitation spectrum displayed four distinct bands at 406 nm, 505 nm, 541 nm and 576 nm ( em = 635 nm). (B) The emission spectrum exhibited a maximum at 635 nm ( ex = 505 nm).
89 Fe2+ injection Figure 8. Time course for the disappearance of protoporphyrin in the ferrochelatase-catalyzed reaction. Wild-type murine ferrochelatase (176 nM) was pre-incubated with protoporphyrin (2.7 M) at 30 oC under strictly anaerobic conditions. Reaction was initiated by addition of ferrous iron (5 M). The decrease in fluor escence intensity at 635 nm ( ex = 505 nm) was continuously monitored over the course of the reaction. The initial rate was determined from the slope of the tangent to the initial, linear part of the progress curve.
90 Figure 9. Dependence of the initial ra te of protoporphyrin consumption on ferrochelatase concentration. Initial rates were determined for enzyma tic reactions by the continuous assay using protoporphyrin (2 M), Fe2+ (4 M) and purified wild-type ferrochelatase at various concentrations.
91 Figure 10. Determination of the steady-s tate kinetic parameters of wild-type murine ferrochelatase. Enzymatic activity was measured using a continuous assay conducted at 30 oC under strictly anaerobic conditions. The reactions contained purified ferrochelatase (70 nM), protoporphyrin and ferrous iron. The hyperbolic curves represent the best fits to the Michaelis-Menten equation by plotting in itial velocity against protoporphyrin concentration at constant Fe2+ concentrations of 0.5 M ( ), 1 M ( ), 2 M ( ), 3 M ( ) and 4 M ( ). The steady-state kinetic parameters of the wild-type ferrochelatase were determined to be Km PPIX = 1.4 0.2 M, Km Fe2+ = 1.9 0.3 M, and Vmax = 97.4 7.9 nmol min-1 mg-1 enzyme.
92 Characterization of the function al ferrochelatase loop variants Crystal structural analysis of ferrochel atase reveals that the enzymes exhibit similar folding patterns and ac tive site structures throughout evolution. Each monomeric unit consists of two domains, with each dom ain folded into a Rossmann-type fold. The active site is located in a deep cleft between the two domains. One side of the cleft is formed by the first two N-terminal -helices, and a loop between them, and the other side consists of a short loop sequence (Al-Karadaghi et al. 1997; Lecerof et al. 2000; Wu et al. 2001; Karlberg et al. 2002). The loop motif exhibits a high degree of sequence identity among all known ferrochel atases (Figure 11). The loop has been proposed to be directly i nvolved in porphyrin interactio n based on molecular dynamics calculations of B. subtilis ferrochelatase with bound ni ckel-protoporphyrin (NiPP) (Franco et al. 2000) and crystal structural analysis of ferroch elatase with the heme analog N -methyl mesoporphyrin bound in the active site (Lecerof et al. 2000). To test this hypothesis, random mutations were introdu ced to the ten consecutive loop residues Q248-L257 in murine ferrochelatase and functio nal variants were further characterized. Biological selection of th e active loop variants By incorporating a mixture of four nucleoti des at each of the 30 bases in the target sequence, random substitutions were generated simultaneously in all the residues. Using a combination of 85% of the wild-type nucleot ides and 15% of the ot her three nucleotides at each position, the bias was set towards th e wild-type sequence, thus increasing the probability of recovering active ferrochelatase variants. Under these conditions, a full
93 Figure 11. Sequence alignment of the loop motif in ferrochelatase. A representative subset illustrates high degree of identity in the loop motif among all known sequences of ferrochelatase. The ami no acid sequences were obtained from NCBI using BLAST search and aligned usin g CLUSTAL W (Thompson et al. 1994).
94 spectrum of permissible substitutions at each position was drawn. To eliminate background activity from wild-t ype enzyme, a mock vector was constructed by replacing a segment of wild-type ferrochelatase plasmi d with a nonfunctional stuffer sequence, and it was used as the cloning vector for the rando m nucleotide sequences. In order to select the functional variants, E. coli strain vis was chosen as the host for the plasmid library containing the randomized loop sequences. Because vis cells need hemin in the growth media for survival, transformants of vis cells carrying the mock vector with inactivated ferrochelatase could not grow in hemin-free media, whilevis cells transformed with a functional ferrochelatase expression plasmi d could override the requirement for hemin and form colonies on hemin-free plates Screening of the unselec ted library by genetic complementation showed that 90% of the random mutants were inactive. This is consistent with the hypothesis that the loop motif is important for enzymatic function, and also demonstrates that the biological selection procedure was efficient. From a total of 2,210 vis transformants harboring randomized l oop mutations, 214 clones were able to grow in hemin-free medium, and the plas mids were sequenced to identify nucleotide changes in the loop region. Most of them were single, double and trip le mutants, with an average of 2.2 amino acid changes (Figure 12A). For a relatively quick assessment of the active library, all of the functional vis transformants were assayed for enzymatic activity by monitoring zinc-protoporphyrin production usi ng crude cell extracts. Most variants showed zinc-chelatase activity reduced to less than 60% of the wild -type level, whereas a subset of triple mutants displayed activities comparable to, or even higher than, wild-type ferrochelatase (Figure 12B).
95 Figure 12. Activity assessmen t and distribution of the nu mber of the functional loop variants. (A) Distribution of the number of the functiona l variants relative to the number of amino acid changes. Random substitutions were intr oduced into the ten re sidues encoding the murine ferrochelatase loop motif. Ac tive variants were recovered by genetic complementation in the E. coli strain vis, and mutations were identified by DNA sequencing. (B) Zinc-chelatase activity of the variants relative to the number of amino acid substitutions. Zinc-chelatase ac tivity was assayed by monitoring zincp rotoporphyrin formation using cell extracts prepared from vis cells transformed with the active variants. Enzymatic activity of each variant was normalized to the activity level of wild-type enzyme. B A
96 Distribution of the functional amino acid substitutions The spectrum and frequency of substitutions observed in the active variants are shown in Figure 13. Functional mutations we re detected at ever y residue in the loop motif. However, single mutants were only recovered at five positions, i.e., K250, V251, P253, V254 and P255, and they are grouped as low informational content residues. Changes were mostly conservative at K250 and V251, while P253, V254 and P255 tolerated a diverse set of alterations. Muta tions in any of the remaining five residues were rare and only occurred in conjunction with substitutions at other positions, and thereby they contain high informational cont ent. Permissible replacements at Q248, S249, G252, W256 and L257 were very restri cted, with the exception of G252, which could be converted to residue s of various sizes and charge s in the context of double, triple, quadruple and penta mutants. Steady-state kinetic analysis of the active loop variants Five active variants were purified to homogeneity and characterized by steadystate kinetic analysis. The single mutant V251L was select ed because it was the only point mutation allowed at the low informa tion content residue V251 and was also coselected in a number of double, triple and quadruple mutants. The single mutant P255R was selected because, although P255 has low information content, arginine replacement was a drastic amino acid alteration and wa s observed in single, double and quadruple mutants (Figure 13). Due to the lack of single substitutions at the high information content positions, multiply-substi tuted variants were chosen for further analysis (Figure 13). By characterizing the triple mu tants S249A/K250Q/V251C K250M/V251L/W256Y
97 Single mutants K 5G 1 N 2L 15T 4N 22R 12 Q248S249K250V251G252P253V254P255W256L257Double mutants N 4A 4A 11 Q 5I 12I 46 T 1 L 1D 41H 1L 5R 1K 6 Q248S249K250V251G252P253V254P255W256L257Triple mutants G 2 M 3 N 2A 7A 2 Q 28C 10G 2D 1 R 1D 16K 6H 16 S 1I 11R 18L 8G 1S 2 T 4A 17T 6L 3W 1Q 8M 16T 24Y 3Q 6 Q248S249K250V251G252P253V254P255W256L257Quadruple mutants M 1 N 1A 5 A 1P 1F 5S 1 P 1G 1T 5L 1W 1A 5L 1R 1G 1 Q248S249K250V251G252P253V254P255W256L257Penta mutant P 1G 1R 1V 1G 1 Q248S249K250V251G252P253V254P255W256L257 Figure 13. Spectrum and frequency of amino acid substitutions in functional loop variants. Mutations at each residue are listed along with the number of times each substitution was observed. The wild-type loop sequence is s hown on the bottom of each mutant category.
98 and the quadruple mutant Q248P/S249G/K250P/G252 W, attempts were made to examine the mutational effects of Q248, S249, G252 and W256 in combination with changes at other positions. Steady-state analysis indicated that, with exception of the triple mutant K250M/V251L/W256Y, all of the variants had kcat values comparable to or higher than that of wild-type ferrochelatase (Table 1). There was a 2-fold increase in kcat for the single mutants V251L and P255R, 4.5-fo ld increase for the triple mutant S249A/K250Q/V251C, and 3.5-fold in crease for the quadruple mutant Q248P/S249G/K250P/G252W. Relative to wild-type enzyme, the Km values for protoporphyrin ( Km PPIX) were elevated ranging from a 2-fold increase for the single mutant P255R to 8-fold for the triple mutant K250M/V251L/W256Y, suggesting that the interaction with the porphyrin substrate was disrupted due to the loop mutations (Table 1). The Km values for ferrous iron ( Km Fe2+) in the single mutants V251L and P255R and triple mutant K250M/V251L/W256Y were of th e same order of magnitude as that in wild-type enzyme. The Km Fe2+ value was 2.5-fold higher for the triple mutant S249A/K250Q/V251C, while a 4fold decrease was observed for the quadruple mutant Q248P/S249G/K250P/G252W (Table 1). Homology modeling of wild-type murine ferr ochelatase and selected loop variants To assess how loop mutations could po ssibly affect the three-dimensional structure of ferrochelatase and the active site architecture, structural models of the
99 Table 1. Steady-state kinetic parameters of wild-type ferrochelatase and selected loop variants. Ferrochelatase activity was determined by a continuous assay using protoporphyrin and Fe2+ as substrates (Shi and Ferreira 2003) The assays were conducted at 30 o C under strictly anaerobic conditions. Steady-state kinetic parameters were determined using matrices of concentrations for both substrate protoporphyrin and Fe2+ (Shi and Ferreira 2004). Proteins kcat (min-1) Km PPIX ( M) kcat / Km PPIX (min-1 M-1) Km Fe2+ ( M) kcat / Km Fe2+ (min1 M-1) Wild-type ferrochelatase 4.1 0.31.40 0.20 2.93 1.90 0.30 2.16 V251L 8.3 1.21.82 0.35 4.56 1.13 0.53 7.35 P255R 7.8 0.82.65 0.44 2.94 1.46 0.38 5.34 S249A/K250Q/V251C 18.0 2.56.84 1.74 2.63 5.15 1.44 3.50 K250M/V251L/W256Y 3.0 0.411.25 2.450.27 1.08 0.48 2.78 Q248P/S249G/K250P/G252W14.1 0.910.34 1.141.36 0.48 0.21 29.40
100 wild-type protein and lo op variants were generated using the crystal structure of human ferrochelatase as a template. The sequences used in modeling wild-type and mutant murine ferrochelatase were at least 91% similar to the human ferrochelatase template. The average free energy of the structural m odels was approximately 16 kcal/mol. The Ramachandran plots showed a normal distribu tion of points with at least 99% residues occupying the allowed regions. Cchirality, amide torsion ( ) and side chain torsions ( 1 and 2) showed no major deviation from the corresponding allowed values. Superimposition of the monomeric model of w ild-type murine ferrochelatase with the Xray crystal structure of hu man ferrochelatase gave a RMSD of 1.23 for the C -atoms of the 357 aligned residues. Structural models of the loop variants (Figure 14) exhibited slightly more resemblance to human ferrochelatase, as indicated by a small reduction in RMSD values (~0.8-0.9 ). In the wild-type model, the N-terminal loop residues Q248 and S249 were buried deeply inside the active site cavity and in very close proximity (about 4 ) to H209 and E289, which were previously proposed to be essential for catal ysis (Kohno et al. 1994; Gora et al. 1996b; Franco et al. 2001; Sellers et al. 2001). K250 was solvent-exposed with the side chain projecting to the protein exterior and in close contact with the Q260 residue. V251 oriented its side chain towards the interior of the ac tive site pocket, while G252 occupied the tip of the loop, and together with P253, V254 and P255, formed a solvent-exposed patch. The indole ring of W2 56 was positioned nearly perpendicular to the H209 and E289 side chains, a d L257 was e xposed to the protein surface with the side chain extended away from the active site poc ket. The spatial arrangement of the loop residues was not significantly altered by the functional substitutions in the selected
101 Figure 14. Molecular modeling of wild-t y pe murine ferrochelata se and selected loop variants. The structural models (monomeric) were generated by homology modeling on Geno3D servers (Combet et al. 2002) using the X-ray crystal structure of human ferrochelatase (PDB code 1hrk) as a template. S ubstituted residues in the loop motif ( red ) are marked ( green ). The color scheme for th e secondary structures is: -helix in blue -strand in y ellow 310-helix in lime -helix in purple turns and coils in cyan The N-terminal 2 helix and its extension in th e loop variants are indicate d by open arrows; the second domain -helix and its variations are indicate d by filled arrows. When not occluded by structural elements, the Nand C-termini ar e indicated. The three-dimensional models are shown for: (A) wild-type murine ferrochelatase, (B) single mutant V251L, (C)single mutant P255R, (D) triple mutant S249A/K250Q/V251C, (E) triple mutant K250M/V251L/W256Y, (F)quadruple mutant Q248P/S249G/K250P/G252W. Images were generated using VMD 1.7 (Humphrey et al. 1996). E D A F B C-term N-term C-term C
102 variants (Figure 14). For all loop variants, changes in the sec ondary structure were observed in the Nterminal domain forming one side of the active site cavity, and in the second domain helix region located adjacent to the active si te cleft (Figure 14). At the N-terminus, immediately following the second -helix ( 2), the variants showed an increase of helical content in place of the relatively unstruc tured turns. While in the structural model of wild-type protein the 2 helix consists of 10 residues Q52-K61 (Figure 14A), in the loop variants, the 2 helix was extended by a short -helix or 310-helix to include 5-7 additional residues (Figure 14B-F). These a dditional residues along with the meandering loop that follows appear to c ontact residues in the second doma in of the enzyme close to the active site cavity. The structural model of wild-type mu rine ferrochelatase revealed a short -helix, consisting of five residues 290TLYEL294 in the second domain near the bottom of the active site pocket (Figure 14A), in an analogous fashion to the -helix observed in the crystal structures of human (349ELDIE353) and yeast (318EIDLG322) ferrochelatases (Fodje and Al-Karadaghi 2002). This -helix was replaced by an -helix in the triple mutant S249A/K250Q/V251C (Figure 14D). In contrast, the quadruple mutant Q248P/S249G/K250P/G252W had a long -helix comprising the 289ETLYELDIEY298 sequence, which is enriched with acidic re sidues (Figure 14F). In the single mutants V251L and P255R and the triple mutant K250M/V251L/W256Y, the -helix shifted positions slightly to include two glutamate residues (Figure 14B, 14C, 14E). In these structures, the -helix was formed by groups of five residues including 293ELDIE297 in V251L and P255R and 289ETLYE293 in K250M/V251L/W256Y.
103 Interaction of ferrochelatase with mitochondrial membrane lipids Polarity distribution among the functional substitutions in the loop variants indicated that mutations in the three N-terminal loop residues, Q248, S249 and K250, reduced the hydrophilic environment of th e N-terminal looplet (Figure 15A). Conversely, the seven non-polar C-termin al residues V251-L 257 were replaced by mostly polar, acidic or basic amino acids (F igure 15A). Furthermore, basic amino acid substitutions were enriched in the C-terminal loop region of the active variants. In order to examine potential interaction of the vari ants with the negatively-charged membrane lipids, liposomes were formulated to mimi c lipid composition of the mouse liver inner mitochondrial membrane, with whic h ferrochelatase is associated in vivo The average diameter of liposomes was determined to be approximately 170 nm by dynamic light scattering. Liposomal binding assay results indicated that affinities for the membrane lipids can be modified by loop mutations (Figur e 15B). In general, the variants carrying basic residues in the C-terminal looplet ex hibited stronger interaction with liposomes than wild-type protein, implying that the localized positive charges favor lipid interaction (Figure 15C).
104 B 1 2 3 4 Ferrochelatase Figure 15. Interaction between ferrochel atase and the mitochondrial membrane lipids. (A) Polarity distribution at each loop positi on in the functional variants. Basic amino acid substitutions accumulated in the C-te rminal looplet positions G252, P253, V254, P255 and L257. Hydrophobic residues include A, F, G, I, L, M, P, V and W; pola r residues include C, N, Q, S, T and Y; aci dic residues include D and E; basic residues include H, K and R. (B) Protein-lipid in teractions among wild-type ferrochelatase and active loop variants. Prot eins were purified in a 96-well format, spotted onto nitrocellulose membrane in a dilution seri es, and probed with bi otinylated liposomes. Bound liposomes were detected using peroxi dase-conjugated extravidin and visualize d using the enhanced chemiluminescence (ECL) system. Examples are shown fo r enhanced lipid-binding to variant P253Q/V 254K/P255T (row 1) and decreased binding to variant Q248P/S249G/K250P/G252W (row 2) relati ve to the wild-type protein (row 3). The variant G252D/V254I (row 4) was included as an internal control. (C) Analysis o f liposomal binding affinities of the active variants. The right panel illustrates the liposome-binding affinity of each variant no rmalized against that of the wild-type ferrochelatase. The left panel indicates the polarity of indivi dual loop residues in each variant using a color code corresponding to: blue for hydrophobic residues, yellow for p olar residues, black for acidic residues, and red for basic residues. Mutated positions i n each variant are marked by an X. A 0 10 20 30 40 50 60 70 80 90 100Q248S249K250V251G252P253V254P255W256L257Wild-type loop residuesPercentage of active variants Hydrophobic Polar Acidic BasicC QSKVGPVPWL Loop substitutions x x x xxx x X x X x X xx x xxx x X x xx X xx X xx X x xx xxx x X xxx xx xxx x X x x x xxx xxx xxx x x X xx x xxxx xxx xxx xxxx xx
105 Resonance Raman spectroscopic analysis of porphyrin binding in the loop variants An important step in the enzymatic reaction catalyzed by ferrochelatase involves distortion of the porphyrin macrocycle to expose the lone-pair or bitals of pyrrole nitrogens to incoming metal ion, thereby facili tating metal chelation. In order to assess the role of the loop residue s in porphyrin deformation, the vibrational modes of porphyrins bound to wild-type enzy me and loop variants were examined by resonance Raman spectroscopy. Binding of substrate protoporphyrin to the variants Substrate ( i.e. free-base protoporphyrin) bi nding in the active site of ferrochelatase loop variants was examined at a 0.1 porphyrin / prot ein molar ratio using RR spectroscopy in a similar fashion to that previously devised for wild-type murine ferrochelatase (Lu et al. 2002). The RR spect ra of the variant-bound porphyrin exhibited clear differences from those of unbound porphyrin in the detergent miscelle environment and porphyrin bound to wild-type enzyme (Figure 16). In th e low-frequency region, as was found previously for wild-type ferrochel atase [(Lu et al. 2002) and Figure 16A, spectrum b], the 7 line upshifted from 666 cm-1 for unbound porphyrin in the equilibration buffer to 673-675 cm-1 in four variants (Figure 16A spectra d-g); the largest shift (to 677 cm-1) was detected in porphyrin bound to the triple mutant K250M/V251L/W256Y (Figure 16A, spectrum c). The 16 line was at 738 cm-1 for unbound porphyrin (Figure 16A, spectrum a) a nd its intensity and frequency were reduced upon binding to wild-type protein and four variants (F igure 16A, spectra b, d-g)
106 Figure 16. The resonance Raman spectra of protoporph y rin incubated with ferrochelatase at a porphyrin-to-protein molar ratio of 0.1. The spectra are aligned in the (A) low-frequency (300-800 cm-1) and (B) high-frequency (1300-1700 cm-1) regions. The spectra are shown for (a) unbound porphyrin in equilibration buffer; (b) porphyrin bound to wild-type ferrochelatase; (c)-(g) porphyri n bound to loop variants includ ing (c) triple mutant K250M/V251L/W256Y, (d) single mutant P255R, (e) single mutant P255G, (f ) triple mutant S249A/K250Q/V251C, (g) quadruple mutant Q248P/S249G/K250P /G252W. Asterisks denote b uffer lines. Vibrational bands of the major stru cture-sensitive modes are labeled. B A
107 in the triple mutant K250M/V251L/W256Y, the 16 band was shifted to higher frequency albeit with diminished in tensity (Figure 16A, spectrum c). More importantly, the 15 mode was activated upon protei n binding (Figure 16A). The 15 mode, a B2u symmetry saddling-like mode, was previously observed and attributed to saddling deformation of the porphyrin upon binding to wild-type enzyme (Hu et al. 1996; Lu et al. 2002). The intensity of 15 mode for porphyrin bound to the variants exhibited differences from that for wild-type protein. In the high-frequency region, the 2 line was considerably diminished in intensity for porphyrin bound to both wild-type enzyme a nd the variants (Figure 16, spectra b-g) when compared to unbound porphyrin (Figure 16B, spectrum a). In particular, the 2 line upshifted by an additional 5 cm-1 in porphyrin bound to the triple mutant K250M/V251L/W256Y relative to w ild-type protein (Figure 16B, spectra b, c), and this shift was the highest observed when all varian ts were considered and indicated important changes in the vinyl interaction with the pr otein (Figure 16B, spect ra b-g). Moreover, protein binding led to sharpening of the vi nyl stretching mode Ca=Cb of free-base porphyrin (Figure 16B). While the frequency of the Ca=Cb mode remained unchanged for porphyrin bound to wild-type protein and four variants (Figure 16B, spectra b, d-g), the Ca=Cb band was broadened and downshi fted upon binding to the variant K250M/V251L/W256Y (Figure 16B, spectrum c), again implicating disruption of interaction between the porphyrin vinyl groups and the active site matrix.
108 Binding of hemin to the variants Interaction of the loop variants with hemi n, an inhibitor of ferrochelatase (Dailey and Fleming 1983), was examined by compar ing the RR spectra of protein-bound hemin with free hemin in the equilibration buffe r (Figure 17). Hemin binding to both P255 variants resulted in a 5 cm-1 downshift of the 2 band for P255R (Figure17B, spectrum d) and 4.2 cm-1 for P255G (Figure 17B, spectrum e), whereas all other variants produced smaller (typically 1.5 cm-1) shifts to higher frequencies (Figure 17B, spectra c, f, g). The 38 line, associated with the C-C and C-C1 stretches, increased intensity in the variants P255R and P255G as compared to wild-type pr otein, although it did not appreciably shift (Figure 17B, spectra b, d, e). Both the 2 and 38 lines derived from vinyl-associated modes, so the fact that they altered in tande m indicates significant structural changes in the vinyl groups of the variant-bound hemi n. In the low-frequency region, the 7 line was upshifted by 1.8 cm-1 in the variant K250M/V251L/W256Y (Figure 17A, spectrum c) and 1.5 cm-1 in the variant S249A/K250Q/V251C (Figure 17A, spectrum f) when compared to binding to wild-type ferrochelat ase (Figure 17A, spectrum b). Binding of nickel-protopo rphyrin to the variants NiPP is a strongly distortio n-sensitive metalloporphyrin and has been shown to undergo deformation upon binding to wild-type ferrochelatase in previous RR studies (Lu et al. 2002). Compared to unbound NiPP in the equilibration buffer, incubation with wild-type protein and th e variants resulted in sharpening the RR lines of NiPP in the high-frequency region of the spectra (Figure 18 B). Protein binding al so led to downshift
109 B A Fi g ure 17. The resonance Raman spectra of hemin incubated with ferrochelatase at a hemin-to-protein molar ratio of 0.1. The spectra are aligned in th e (A) low-frequency (300-800 cm-1) and (B) high-frequeny (1300-1700 cm-1) regions. The spectra are shown for (a) unbound hemin in equilibratio n buffer; (b) hemin bound to wild-type ferrochelatase; (c)-(g) hemin bound to the loop variants including (c) triple mutant K250M/V251L/W256Y, (d ) single mutant P255R, (e) single mutant P255G, (f) triple mutant S249A/K250Q/V251C, (g) quadruple mutan t Q248P/S249G/K250P/G252W. Asterisks denote bu ffer lines. Vibrational bands of the major structure-sensitive modes are labeled.
110 of the 2 and 10 bands of NiPP, while the vinyl stretching band Ca=Cb, was sharpened and enhanced in intensity (Figure 18B). Co mpared with NiPP bound to wild-type protein (Figure 18A, spectrum b), the low-frequency re gions of the RR spectra for the variants K250M/V251L/W256Y (Figure 18A, spectru m c) and S249A/K250Q/V251C (Figure 18A, spectrum f) exhibited broader 7 lines and propionate lin es with less pronounced frequency shifts. These alterations make th e spectra for the vari ant-bound NiPP resemble more closely the spectrum of unbound NiPP (Figur e 18A, spectrum a) rath er than that of NiPP bound to wild-type enzyme (Figure 18A, sp ectrum b). This suggests that, relative to wild-type ferrochelatase, both triple mutants had lower affinity towards NiPP, or the bound NiPP was less conformationally constr ained in the triple mutants.
111 Figure 18. The resonance Raman sp ectra of nickel-protoporph y rin incubated with ferrochelatase at a porphyrin-to-protein molar ratio of 0.1. The spectra are aligned in the (A) low-frequency (300-800 cm-1) and (B) high-frequency (1300-1700 cm-1) regions. The spectra are shown for (a) unbound NiPP in equilibration buffer; (b) NiPP bound to wild-type ferrochelatase; (c)-(g) NiPP bound to the loop variants including (c) triple mutant K250M/V251L/W256Y, (d ) single mutant P255R, (e) single mutant P255G, (f) triple mutant S249A/K250Q/V251C, (g) quadruple mutan t Q248P/S249G/K250P/G252W. Asterisks denote bu ffer lines. Vibrational bands of the major structure-sensitive modes are labeled. A B
112 Inhibition of ferrochelatase by N -methyl protoporphyrin N -alkylated porphyrins have been known to competitively inhibit substrate porphyrin binding to mammalian fe rrochelatase (Figure 19). N -methyl protoporphyrin exhibits a very high binding affinity for ferrochelatase (Dailey and Fleming 1983; Cole and Marks 1984; Nunn et al. 1988) The X -ray crystal structural analysis of B. subtilis enzyme with bound N -methyl mesoporphyrin indicated that the loop residues are in close contact with the porphyr in macrocycle (Lecerof et al. 2000) This raises the possibility that mutations in the lo op may modulate binding of N -alkylated porphyrins. In order to test this hypothesis, loop variants were examined with respect to interaction with N methyl protoporphyrin. To find out whether loop muta tions could possibly alter i nhibitor binding, a subset of the functional loop variants including K250N, V251L, P255R, P255G, K250M/V251L/W256Y and Q248P/S249G/K250P /G252W were transformed into vis cells, and the cells were grown on agar plates with increasing concentrations of NMPP. vis cells transformed with vari ants P255R and P255G could not grow on plates in the presence of 20 M NMPP, whereas vis cells transformed with wild-type protein and other variants could. Based on this observati on, P255 variants were selected for further characterization. It is recognized that the current assay was not intended to be a comprehensive screen to cover the full spectrum of mutations that can affect interaction with the inhibitor. However, it revealed the variants which behaved differently from wild-type enzyme and therefor e provided support to the or iginal hypothesis that mutation in the loop could modulate inhibitor binding. This may serve as an initial step towards understanding the structural basis of ferrochelatase inhibition by N -alkylated porphyrins.
113 N HN N NH O O C C O O H DC B ACH3 Figure 19. The structural diagram of N -methyl protoporphyrin. An isomeric form containing methylated A ri ng pyrrole nitrogen atom is shown in the diagram.
114 Equilibrium binding of inhibitor to ferrochelatase Binding affinity of NMPP to the enzyme was determined from fluorescence quenching measurements. Intrin sic fluorescence derives from tryptophan residues in the protein and is sensitive to protein conforma tion (Eftink and Ghiron 1976). In wild-type ferrochelatase, five tryptopha n residues give rise to the intrinsic fluorescence which exhibits an excitation maximum near 280 nm and an emission maximum near 330 nm (Figure 20A). The fluorescence is linearly dependent on protein c oncentration (Figure 20B). Interestingly, in a number of variants, the intens ity of protein fluorescence appeared to be lower than that of wild-type ferrochelatase (Figure 20B). The decrease in intrinsic fluorescence suggest s that loop mutations lead to modification of protein conformation consistent with observations obtained from st ructural mode ls (Shi and Ferreira 2004). Because incubation of ferroch elatase with NMPP resulted in reduction of the intrinsic fluorescence, the dissociation c onstant of NMPP for the enzyme could be measured by quantifying the quenching effect. While the inhibitor exhibited high affinity to wild-type protein with a Kd of 9 5 nM, binding affinity to the variants was decreased by one order of magnitude with a Kd of 160 7 nM for P255R and Kd of 300 100 nM for P255G (Figure 21). Kinetic pathway of inhibition The kinetic steps involved in NMPP binding to ferrochelatase were investigated by monitoring enzymatic reaction with Zn2+ and protoporphyrin as substrates using stopped-flow absorbance spectroscopy (Fig ure 22A). Kinetic traces for zincprotoporphyrin formation in the presence of va rious concentrations of NMPP were fitted
115 Figure 20. The intrinsic fluores cence of ferrochelatase. (A) The excitation and emission spectra of wild-type ferrochelatase. 0.9 M purifie d p rotein was diluted in 10 mM Tris-acetate pH 8 containing 0.05% (v/v) Tween-80. The emission spectrum exhibited a maximum at 331 nm ( ex= 283 nm), while the excitation spectrum showed a maximum at 283 nm ( em = 331 nm). (B) The intrinsic fluorescence is proportional to protein concentration, a nd fluorescence intensity is diminished in the loop variants. Protein fluorescence was measured at 331 nm across a range o f concentrations for wild-type ferrochelatase ( ), single mutant V251L ( ), single mutant P255R ( ), quadruple mutant Q248P/S249G/K250P/G252W ( ), and triple mutan t K250M/V251L/W256Y ( ). 0 20 40 60 80 00.10.20.30.40.50.6Protein concentration ( M)Fluorescence (arbitrary unit ) B Wavelength (nm) 250300350400450 Fluorescence (arbitrary units) 0 50 100 150 200 Excitation spectrum Emission spectrum A
116 Fi g ure 21. Bindin g curves g enerated from protein fluorescence quenchin g measurements following titration with N -methyl protoporphyrin. Aliquots of NMPP solution were mixed with pu rified ferrochelatase in quenching buffer containing 10 mM Tris-acetate pH 8 and 0.05% (v/v) Tween-80. The mixture was incubated on ice for 1 hr. Fluorescence inte nsity of the solution was measured at 23 oC at 331 nm ( ex = 283 nm). Kdfor NMPP was determined by fitting data to Equation 2 t t d t d tE L E K L E K L E F F F 2 42 0 Proteins used in each set of assays are: (A) wild-type ferrochelatase (40 nM); (B) vari ant P255R (250 nM);(C ) variant P255G (120 nM). B C A
117 to Equation 3 (Williams and Morrison 1979) to obtain the pseudo first-order rate constants for inhibitor binding (Figure 22C): kt s se k v v t v A A 10 0 (Eq.3) For wild-type enzyme and variant P255R, the rate constants exhibited a hyperbolic dependence on NMPP concentrat ion (Figure 23A, 23B ) suggesting that inhibitor binding occurred in two kinetic steps, i.e. k1k-1 k 2FC + NMPP [FC NMPP]1[FC NMPP]2k-2 The first step was rapid and involved the initia l binding of the inhibitor to ferrochelatase. The second step was slow, and possibly it was associated wi th a conformational change of the inhibitor-protein complex. The hype rbola can be described by the equation: m iK S K I I k k k0 0 0 2 21 (Eq.4) where k is the pseudo first-order rate constant for the approach to the steady-state phase I0 is the inhibitor concentration, S0 is the initial substrate concentration, m iK S K01 is the apparent inhibition constant Ki app, k2 and k-2 are the second step forward and reverse rate constants. For wild-type protein, k2 was 6.4 0.4 s-1, k-2 was 24.5 0.3 s-1, and Ki app was 2.8 0.6 nM. For variant P255R, k2 was 8.3 1.1 s-1, k-2 was 19.3 0.7 s-1, and Ki app was 1.0 0.5 M. In contrast, for variant P255G, the pseudo-first order rate constant k was linearly dependent upon inhi bitor concentration (Figure 23C), consistent with a single kinetic step for NMPP binding, i.e.
118 Figure 22. Transient kinetic analysis of the ferrochelatase-ca talyzed reaction. Purified variant P255R (2 M) was pre-incubated with protoporphyrin (10 M) in 100 mM Tris-acetate pH 8 containing 0.5% (v/v ) Tween-80. The solution was mixed with zinc-acetate (15 M) to initiate enzymatic reaction. (A) Wavelength scans of the reactio n mixture at different reaction times. Spect ra were collected at 96 ms, 144 ms, 192 ms, 240 ms and 288 ms since the inception of reac tion. Isosbestic point at 412 nm indicate d conversion of molecular species in the course of enzymatic reaction. (B) The absorbance spectrum of zinc-protoporphyrin. The spectrum was shown for ZnPP (5 M) in assay b uffer with an absorbance maximum at 420 nm (C) The time course for ZnPP formation catalyzed by ferrochelatase. The increas e of absorbance at 420 nm was continuously monitored over the course of reaction. Progress curve was fitted to Equation 3 kt s se k v v t v A A 10 0 to obtain the rate constant k for NMPP binding. Wavelength (nm) 350360370380390400410420430440450 Absorbance 0.0 0.1 0.2 0.3 0.4 0.5 B Time (sec) 0.00.51.01.52.02.53.0 Absorbance at 420 nm 0.17 0.18 0.19 0.20 0.21 C Wavelength (nm) 350360370380390400410420430440450 Absorbance 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 0.22 96 ms 144 ms 192 ms 240 ms 288 ms A
119 k 1k-1FC + NMPP [FC NMPP] By fitting the line to the equation mK S I k k k0 0 1 11 (Eq.5) the forward rate constant mK S k0 11 was determined to be 8.2 0.7 s-1 M-1, the reverse rate constant k-1 was 18.7 2.3 s-1. Ki app was calculated from mK S k k0 1 11 and was found to be 2.3 M. Size measurement of the active site pocket In order to compare the size of the active si te in the variants with that of wild-type protein, the volume and Conno llys surface area of the porphyrin-binding pocket in the structural models were estimated using the program CAST (Liang et al. 1998). The active site of wild-type ferrochelatase wa s found to occupy a surface area of 2138 2 and a volume of 3015 3, while variant P255R had a smaller surface area of 1340 2 and a reduced volume of 1932 3, and variant P255G had the smallest surface area of only 1048 2 and a volume of 1604 3. Overall, the binding poc ket size was significantly reduced as a result of P255 mutations. Vari ant P255R retained ~ 70% of the volume and surface area relative to wild -type protein, whereas P255G only kept ~50% the size of wild-type porphyrin -binding pocket.
120 Figure 23. Dependence of the rate cons tants for ferrochelatase binding on N -meth y l protoporphyrin concentration. Ferrochelatase (0.5 M) was reacted with protoporphyrin (5 M) and zinc-acetate (15 M) in the presence of various concentratio ns of NMPP. The ps eudo first-order rate constant k for inhibitor binding was obtained from by analyzing progress curve as described previously. The values of k were plotted against inhib itor concentrations. The p roteins used in each set are: (A) wild-t ype ferrochelatase; (B ) variant P255R; (C) variant P255G. NMPP concentration (nM) 05101520253035 k (sec -1 ) 24 26 28 30 32 A NMPP concentration ( M) 01234567 k (sec -1 ) 16 18 20 22 24 26 28 30 B 0 10 20 30 40 50 60 70 80 01234567 NMPP concentration ( M)k (sec-1) C
121 FeS cluster and oligomeric assembly in ferrochelatase variants During profiling of the UVvis absorbance spectra of th e functional loop variants, a number of variants exhibited absorbance feat ures with some differences from wild-type protein. This raises the que stion of whether the variants have alterations in the FeS cluster. Because the FeS cluster has been t hought to be located near the dimer interface in wild-type ferrochelatase, it was suspected th at dimeric interaction may be disrupted in the variants. To address this issue, the vari ants were further examined with regard to their oligomerization state and th e properties of FeS cluster. UV-visible absorbance spectra of the variants The UVvis absorbance spectra of selected variants including triple mutants S249A/K250Q/V251C and S249A/K 250R/G252W exhibited featur es somewhat different from wild-type protein (Figure 24). In wild-type ferrochelatase, the absorbance maximum at 330 nm and the shoulder regions around 450 nm and 510 nm are consistent with the presence of a [2Fe-2S] cluster. In the variants, the region around 450 nm appears blue-shifted to around 410 nm, while the absorbance maxima at 330 nm and 510 nm remain. These absorbance features bear some resemblance to those for biotin synthase, which contains mixtures of [2Fe-2 S] and [4Fe-4S] clusters (Ugulava et al. 2000; Ugulava et al. 2001).
122 Figure 24. The UV-visible absorb ance spectra of purified wild-t y pe ferrochelatase and variants. The absorbance spectra are shown fo r (A) wild-type ferrochelatase (9 M), (B) triple mutant S249A/K250Q/V251C (27 M), and (C) triple mu tant S249A/K250R/G252W (31 M). In wild-type protein, the absorb ance maximum at 330 nm and the shoulde r regions around 450 nm and 510 nm are indicative of a [2Fe-2S] cluster. In the variants, the absorbance shoulder region around 450 nm appears to blue-shi ft to around 400 nm, while the absorbance maximum at 330 nm and shoulder around 510 nm remain. A C B
123 Metal content analysis Determination of metal content of purif ied wild-type ferrochelatase and variant S249A/K250Q/V251C by plasma emission spectroscopy showed that the stoichiometry ratio of Fe/protein was 1.3 for each mono mer of wild-type prot ein and 2 for variant S249A/K250Q/V251C (Shi and Ferreira 2004). The ratio for wild-type protein is in agreement with a previous re port indicating a stoichiometry of one [2Fe-2S] cluster per monomer (Lloyd et al. 1996). The molar ratio of Fe/protein was approximately 50% higher in variant S249A/K 250Q/V251C relative to wild-type enzyme. Dependence of enzymatic activi ty on FeS cluster synthesis The requirement of FeS cluster for en zymatic activity was assessed using the bacteria strain iscS(PK4331), which is deficient in cluster assembly as a result of deletion of cysteine desulfurase, the sulfur donor for FeS cluster synthesis (Schwartz et al. 2000). Growth in PK4331 cells rendered wild-type fe rrochelatase catalytically inactive, whereas the enzyme was active in RZ4500 cells (Fi gure 25). These results are consistent with the suggestion that FeS cluster is important to maintaining wild-type ferrochelatase activity (Sellers et al. 1998b; Schneider-Yin et al. 2000b). Single mutant C341A was not functional in either RZ4500 or PK4331 cells and was used as a negative control (Figure 25). In terestingly, both varian ts S249A/K250Q/V251C and S249A/K250R/G252W retained 50-75 % residual activities in PK4331 cells suggesting that catalysis could take pl ace in the absence of FeS cluster assembly (Figure 25).
124 W i ld ty p e S 2 4 9 A /K 2 5 0 Q /V 2 5 1 C S 2 4 9 A /K 2 5 0 R /G 2 5 2 W C 3 4 1 A PK4331 RZ4500 0 50 100 150 200 250 300 350Zinc-protoporphyrin fluorescence (arbitrary units) Figure 25. Dependence of ferrochelatase ac tivity on FeS cluster synthesis. RZ4500 ( iscS+) and PK4331 ( iscS-)cells were transformed with expression plasmids encoding wild-type ferrochelatase, singl e mutant C341A, and triple mutants S249A/K250Q/V251C and S249A/K250R/G252 W. Enzymatic activities were determined by monitoring zinc-protoporphyrin formationusing cell lysates prepared fro m each transformant.
125 Subunit assembly of purified ferrochelatase Consistent with previous observations that mammalian ferrochelatase is a homodimer (Straka et al. 1991; Wu et al. 2001) gel filtration chromatography of purified recombinant wild-type murine ferrochelatase yielded a single dimeric species with an approximate apparent molecular weight of 88 kDa (Figure 26). Dynami c light scattering analysis showed that wild-type murine fe rrochelatase in solution had an estimated hydrodynamic radius of 4.6 nm (Figure 27). This is in agreemen t with the results obtained for purified human ferrochelatase, which showed a hydrodynamic radius of 4.4 nm corresponding to an estimated molecula r mass of 100 kDa fo r globular protein (Burden et al. 1999). In contrast, gel filtr ation analysis of va riants S249A/K250Q/V251C and S249A/K250R/G252W indicated the presen ce of higher order oligomers (Figure 26B, 26C). Interestingly, all of the molecular species eluted from the gel filtration column were enzymatically active and exhi bited comparable spec ific activity. Size distribution of the ferrochel atase oligomers was further analyzed using dynamic light scattering, and the results were comparable to those obtained from gel filtration chromatography. Variant S249A/K250Q/V251C exhibited an estimated hydrodynamic radius of 20 nm suggesting that it is pres ent as a homogeneous high molecular weight species in solution (Figure 27A ). Variant S249A/K250R/G252W also yielded oligomeric assembly larger than wild-type protein w ith a hydrodynamic diameter of 12.8 nm (Figure 27B).
126 Figure 26. Molecular weight assessment of purified wild-t y pe ferrochelatase and variants by gel filtration chromatography. All the proteins were loaded onto a Supe rdex-200 HPLCcolumn equilibrated with a buffer containing 20 mM TrisCl pH 8.0, 150 mM NaCl and 5% glycerol. Purified wildtype ferrochelatase (240 M) and variants S 249A/K250Q/V251C (600 M) and S249A/K250R/G252W (800 M) were injected separately. Elution was monitored b y absorption at 280 nm. (A) Ca libration of the column with protein markers of known molecular masses including amylase (200 kDa), alcohol dehydrogenase (150 kDa), albumin (66 kDa), carbonic anhydrase (29 kD a) and cytochrome c (12.4 kDa). The voi d volume (V0) was determined to be 38 mLbased on the elution of blue dextran. (B) Elution profiles of wild-type ferrochelatase ( ) and variant S249A/K250Q/V251C ( ). (C) Elution profiles of w ild-type ferrochelatase ( ) and variant S249A/K250R/G252W ( ). 10 100 1000 220.127.116.11.92.02.12.22.3Ve/V0Molecular Weight (kDa) Amylase Alcohol dehydrogenase Albumin Carbonic anhydrase Cytochrome cA 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 18.104.22.168.22.214.171.124.02.12.22.3 Ve/V0Absorbance at 280 nm 0.00 0.03 0.06 0.09 0.12B 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 126.96.36.199.71.81.92.02.1Ve/V0Absorbance at 280 nm 0.00 0.03 0.06 0.09 0.12C
127 0 5 10 15 20 25 30 35 110100 Diameter (nanometers)Differential Volume %0 5 10 15 20 25 30 35 40 45B 0 5 10 15 20 25 30 110100 Diameter (nanometers)Differential Volume %0 5 10 15 20 25 30 35 40 45A Figure 27. Molecular size determination of purified wild-t y pe ferrochelatase and variants by dynamic light scattering (DLS). Solutions of purified ferrochelatase (6 M) in solubilization bu ffer containing 20 mM TrisCl pH 8.0, 150 mM NaCl, 0.5% NaChol ate and 10% glycerol were examined on a p article size analyzer. (A) DLS results for wild-type protein ( ) and varian t S249A/K250Q/V251C ( ). (B) DLS results for wild-type protein ( ) and varian t S249A/K250R/G252W ( ).
128 Electron paramagnetic resonance spectra of ferrochelatase At high power level (25 mW ) and high temperature (70 K), the EPR spectra of wild-type ferrochelatase exhibited features of a [2Fe-2S] cluste r-containing protein (Figure 28A), consistent with the previous observations (Ferreira et al. 1994). Both variants S249A/K250Q /V251C and S249A/K250R/G252W showed similar characteristics indicating they also contained [2 Fe-2S] clusters (Figure 28B). In order to examine whether [4Fe-4S] cluster was also pr esent in the variants, temperature sweep of EPR signals was carried out. Only varian t S249A/K250Q/V251C was analyzed so far. Temperature-dependence of the signal intensit y at g value of 1.93 exhibited a maximum at low power levels (0.002-25 mW) and be came linear at higher power levels (25-200 mW) (Figure 29).
129 Wild-type A S249A/K250Q/V251C S249A/K250R/G252W B Figure 28. EPR spectra of purified w ild-type ferrochelatase and variants. Typically purified proteins were reduced by dithionite prior to collecting EPR spectra. Xb and EPR spectra were recorded at 70 K and 25 mW. (A) The spectra are shown fo r dithionite-reduced wild-type ferrochelatase with 5X His-tag at 400 M ( ), with 5X His-tag at 190 M ( ), and without His-tag at 200 M ( ). The spectra of as purifie d wild-type ferrochelatase with 5X His-tag at 400 M is also shown ( ). (B) EPR spectr a are shown for dithionite-reduced variants S249A/K250Q/V251C at 426 M ( ) an d S249A/K250R/G252W at 95 M ( ).
130 Figure 29. Temperature-dependence of the EPR signal intensit y for a purified ferrochelatase variant. Purified variant S249A/K250Q/V251C (426 M) was reduced with dithionite, and Xband EPR spectra were collected Signal intensity at g va lue of 1.93 was plotted as a function of temperature at diffe rent microwave power levels.
131 Chapter Four Discussion Continuous assay for ferrochelatase activity Over the past four decades, a number of assays have been developed to determine ferrochelatase activity (Dailey and Dailey 2003). One of the most commonly used methods is a discontinuous assay using ferrous iron and porphyrin as su bstrates. In this method, synthesized heme is spectrophotom etrically quantified following quantitative conversion to the pyridine-hemochromogen (Porra and Jones 1963a; 1963b; Porra et al. 1967). This assay facilitates quantification of enzymatic activities as low as 1 nmol protoheme min-1 (Taketani and Tokunaga 1981). To ma intain the iron substrate in the ferrous state, thiol compounds have been routinely included in the reaction mixture (Porra et al. 1967). However, hemes degrad e in the presence of thiol compounds (Porra et al. 1967); additionally, dith iothreitol competes with fe rrochelatase for binding to free ferrous iron (Franco et al. 1995). A second, dual-wavelength assay measures ferrochelatase activity by following the disa ppearance of the porphyrin substrate (Labbe et al. 1963; Porra et al. 1967; Jones 1969; Camadro and Labbe 1982). Under these assay conditions, the absorption spectra of porphyrin and heme exhibit clear isosbestic points indicating conversion of substrate into produc t, and the progress of the reaction can be followed using electronic absorption spectrosc opy (Porra et al. 1967) While this assay offers the prospect for development of a continuous assay, its inherently reduced
132 sensitivity (Porra et al. 1967), compared to that of the pyridine-hemochromogen method, has restricted its application. In addition, reducing reagen ts, such as glutathione, which appear to interfere with heme stability, were included in this pr ocedure (Porra et al. 1967). A radiochemical assay for ferrochelatas e activity quantifies th e incorporation of 59Fe into porphyrin (G oldberg et al. 1956). However, this method has had limited use because of requirements for safety precautions, expense and labor (Goldberg et al. 1956; Jones and Jones 1968; Dailey 1977). To overcome the technical constraints of maintaining the iron substrate in the reduced state and the requirement of strict an aerobicity, alternative assays take advantage of the fact that ferrochelatase can catalyze th e incorporation of othe r divalent metal ions (Johnson and Jones 1964; Neuberger and Tait 1964; Porra and Ross 1965; Jones 1969; Camadro and Labbe 1982; Taketani and T okunaga 1982). Indeed, a convenient method is to determine zinc-chelatase activity (N euberger and Tait 1964), since the fluorescence of the zinc-protoporphyrin pr oduct can be followed contin uously (Camadro and Labbe 1982; Abbas and Labbe-Bois 1993). Chelation of Co2+ has also been used to determine the enzymatic activity of ferrochelatase (Johnson and Jones 1964; Porra and Ross 1965; Jones 1969). Cobalt-chelatase activity can be measured by the pyridine-hemochromogen assay or by monitoring the ab sorbance decrease of the porph yrin substrate (Jones 1969; 1970; Canepa and Llambias 1988). Despite the convenience of these assays using alternative metal substrates, kinetic and Xray absorption studies (Camadro and Labbe 1982; Ferreira et al. 2002) have provided evidence that chelation of non-physiological, metal substrates by ferrochelatase might not be mechanistically identi cal to that of the Fe2+ substrate. While Fe2+ was a strong competitive inhibitor of zinc-chelatase activity,
133 Zn2+ competitively inhibited iron chelation with a Ki of only 1.5 M (Camadro and Labbe 1982). Moreover, the recent X-ray absorption spectroscopy studies of metal binding sites in murine and yeast ferrochelatases indicated that coordination of Zn2+ was different from that of Fe2+ or Co2+ (Ferreira et al. 2002). Thus, while there are convenient methods available to determine the enzymatic activity of ferrochelatase, each has limitations that reduce its usefulness, particularly in enzyme kinetic studies. In this study, a continuous, fluor imetric assay for ferrochelatase activity has been developed using the phys iological substrates ferrous iron and protoporphyrin, by measuring the rate of porphyr in consumption. Using this method, the kinetic constants determined for the wild-typ e murine ferrochelatase were shown to be comparable to values reported for the wild-type human ferrochelatase (Table 2). Overall, the continuous assay offers si gnificant advantages over the existing procedures. Because the time course of reaction can be monitored by following the decrease of porphyrin fluorescence, this assay allows the initial rate to be determined from the slope of progress curves at the inception of enzymatic reaction, and thereby eliminates the problem of non-lin earity in velocity calculati ons in end-point assays which results from prolonged incubation. By impos ing strictly anaerobic conditions, the assay permits the determination of ferrochelatase activity for the physiological substrate, ferrous iron, instead of metal ion substitutes, such as Zn2+ and Co2+. Additionally, reducing agents are eliminated in the reac tion, including the th iol-containing compounds, which are known to promote he me degradation (Porra et al 1967). The reaction medium
134 Ferrochelatase source Vmax (nmol mg-1 min-1) Km Porp ( M) Km Fe2+ ( M) Porphyrin substrate Reducing agent Assay Method Ref. Murine (recombinant) 97.4 1.4 1.9 Protoporphyrin None Continuous fluorimetric Assay (Shi and Ferreira 2003) Murine (recombinant) 136 95 112.5 Deuteroporphyrin DTT Hemochromogen assay (Ferreira 1994) Human (recombinant) 165 a 9.3 a 9.0 a Protoporphyrin Not specified in the text Porphyrin absorbance decrease (Sellers et al. 2001) Human (recombinant) --8.19 a 9.35 aProtoporphyrin -mercaptoethanol Hemochromogen assay (Sellers and Dailey 1997) Yeast (mitochondrial membrane extracts) --5.9 a 1.63 aProtoporphyrin None; anaerobiosis bPorphyrin absorbance decrease (Camadro and Labbe 1982) Bovine (liver) 105 a 80 a 11 a Protoporphyrin DTT Hemochromogen assay (Dailey and Fleming 1983) Bovine (liver) 88 a 54 a 46 a Protoporphyrin DTT Hemochromogen assay (Taketani and Tokunaga 1982) Rat (liver) 120 a 28.5 a 33.1 aProtoporphyrin DTT Hemochromogen assay (Taketani and Tokunaga 1981) Table 2. Comparison of the steady-state kinetic parameters of ferrochelatase determined by various assay methods. a Apparent steady-state kine tic parameter values. b Anaerobiosis was achieved by enzymatic oxygen uptake (Camadro and Labbe 1982).
135 is prepared at the opt imal pH of 8.0 (Tak etani and Tokunaga 1981; 1982; Franco et al. 1995), and the detergent, Tween-80 [0.5% (v/v)], is included to allow maximal solubilization of protoporphyrin (Porra and Jones 1963a; Po rra et al. 1967; Jones 1969; Camadro and Labbe 1982). As ferrochelatas e in the range of 10100 nM has been found sufficient for the assay, the procedure allows the enzyme to be used in concentrations approximately one to two orders of magnit ude lower than those routinely employed in non-fluorimetric assays. Moreover, the calcul ated limit of reliable detection is improved up to two orders of magnitude over that of the pyridine-h emochromogen assay (Taketani and Tokunaga 1981). Thus, the assay permits economical use of purified enzyme stocks, and also makes it possible to detect ferrochel atase in biological samples of low specific activity such as in porphyria patients. In summary, the continuous assay provides a novel method to measure ferrochelatase activity using the physiological fe rrous iron substrate. The sensitivity of this assay is enhanced up to two orders of magnitude over other assays currently in use, making it possible to determine the initial rate of enzymatic reaction at lower substrate concentrations than those previously reported. Characterization of the ac tive site loop variants The overall fold and active site structure of ferrochelatase are highly conserved throughout evolution. Cr ystal structures of B. subtilus yeast and human enzymes reveal two domains in each monomeric unit and each domain is comprised of a Rossmann-type fold (Al-Karadaghi et al. 1997; Lecerof et al. 2000; Wu et al. 2001; Karlberg et al. 2002). The active site is located in a deep cleft betw een the two domains. Th is cleft is enriched
136 with many conserved residues that have been found to be important in metal binding and catalysis (Al-Karadaghi et al 1997; Lecerof et al. 2000; Wu et al. 2001; Karlberg et al. 2002). One side of the cleft is formed by the first two N-terminal -helices, and a loop between them, and the other side cons ists of a short conserved loop sequence, which connects a -strand and an -helix in the second doma in (Al-Karadaghi et al. 1997; Lecerof et al. 2000; Wu et al. 2001; Karlberg et al. 2002). A lthough ferrochelatase only exhibits an overall sequence identity of no more than 20% in evolution, the loop motif is highly conserved (Figure 11). Based on molecular dynamics calculations of B. subtilis ferrochelatase with bound ni ckel-protoporphyrin (NiPP) (Franco et al. 2000) and crystal structural analysis of fe rrochelatase with bound porphyrin analog N -methyl mesoporphyrin (Lecerof et al. 2000 ), it has been proposed th at the conserved loop is directly involved in porphyrin interaction. To test this hypothesis, random mutagenesis and genetic complementation were applied to identify functional substitutions and to evaluate the information content of each of the 10 consecutive loop residues (Q248L257) in murine ferrochelatase. The results indicated that despite conser vation of the loop (Figure 11), every amino acid within this sequence tolerated f unctionally permissible substitutions (Figure 13). The degree of acceptable substitutions varied, however, from position to position. K250, V251, P253, V254 and P255 tolerate d diverse mutations including single substitutions and contained low informational c ontent. Interestingly, single amino acid substitutions identified at the low informationa l content positions were also co-selected in the multiply substituted variants (Figure 13), and a few of these amino acid substitutions are also present in nature (Figure 11). Th e remaining five loop positions Q248, S249,
137 G252, W256 and L257 had high informational content, because they only allowed very limited substiutions, and mutations were only po ssible when they were associated with changes in other loop residues whereas no single substitutions were observed (Figure 13). While G252 was the most amenable to change, even with drastic s ubstitution such as W or R (Figure 13), substitutions at Q248, S 249 and W256 were scar ce, occurring primarily in triple and occasionally in quadruple muta nts, suggesting that complementary changes in the loop are necessary to counterbalan ce the loss of wild-type residues at these positions. The attachment of ferrochelatase to the mitochondrial membrane has been proposed to play a role in f acilitating uptake of the water-i nsoluble porphyrin substrates from a hydrophobic environment and also provid ing a pathway for heme release (Gora et al. 1996a; Wu et al. 2001). As shown in the X-ray crystal structures of human and yeast ferrochelatases, the active si te entrance is delimited by two, oppositely located loops, which are enriched with hydr ophobic residues and possib ly involved in mediating membrane association (Gora et al. 1996a; Wu et al. 2001; Karlberg et al. 2002). Further, a number of positively charged residues in these loops have been implicated in interaction with the phosphate groups of the membrane phospholipids (Karlberg et al. 2002). One of these loops surr ounding the active site entrance corresponds to the murine ferrochelatase sequence Q248-L 257, and consistent with th e above hypothesis, positively charged amino acid substitutions were f ound to accumulate at the C-terminal hydrophobic loop residues G252, P253, V254 and P255 (Figure 15A). They conferred an overall increase in the binding affinities of the variants towards liposomes mimicking the composition of the inner mitochondrial membranes (Figure 15B, 15C).
138 An important step in ferrochelatase cat alysis is the binding and distortion of porphyrin macrocycle to faci litate metal chelation. An overall increase in the Km PPIX values of the selected loop va riants (Table 1) supports the proposal that loop residues contribute to porphyrin-ferrochelatase inter action. K250M/V251L/W256Y was the only variant with both a lower kcat and a higher Km PPIX than those of wild-type ferrochelatase (Table 1). Indeed, th is variant exhibited a catalytic e fficiency towards porphyrin 10-fold lower than wild-type enzyme (Table 1). A lthough three residues were mutated in variant K250M/V251L/W256Y, the primary effect was thought to derive from the W256Y mutation. Specifically, the V251L mutati on alone did not bring a decrease in the kcat value or a significant increase in the Km PPIX value (Table 1), and mutation of the K250 residue was not expected to pr oduce dramatic effect on the f unction of ferrochelatase, as it occupies a low information content position (Figure 13). In addi tion, mutation of the corresponding residue in yeast ferrochelatas e (W282) yielded a vari ant (W282L) showing a 10-fold decrease in Km PPIX and an unaltered Km for the metal substrate (Gora et al. 1996b). Actually, the homologous residue in B. subtilis ferrochelatase (W230) was shown to stack against the pyrrole ring, ther eby stabilizing the posi tion of the porphyrin ring in the crystal structure (Lecerof et al. 2000). The over 7-fold increase in the Km PPIX of the quadruple mutant Q248P/S249G/K250P/G252W (Table 1) most likel y stemmed from the G252W mutation. Replacement of G252 with bulky aromatic tryptophan possibly blocked active site opening, thus hindering entr y and binding of the porphyrin substrate. This hypothesis agrees with the observation, based on the crystal structure of B. subtilis ferrochelatase (Al-Karadaghi et al. 1997), that any side chain other than glycine would introduce steric
139 hindrance at this position. The triple mutant S249A/K250Q/V251C also exhibited a considerable increase (approximately 5-fold) in the Km PPIX. While S249 contains high informational content (Figure 13) and is burie d deeply in the active site cavity, the V251 side chain is directed towards the path of porphyrin entry. Possibly, combinatorial changes in the three positions led to misa lignment or destabilization of porphyrin binding. Unexpectedly, with the exception of the triple mutant K250M/V251L/W256Y, kcat, increased in all of the loop variants (Table 1). The increase, ranging from 2to 4.4fold, conceivably resulted from conformati onal flexibility of th e loop motif. Surface loops are frequently identifie d as mobile elements medi ating protein conformational changes (Pompliano et al. 1990; Osborne et al. 2001). They may control access to the active site by adopting an open conformati on to permit substrate entry and product release, and a closed confor mation to protect active site from the solvent and promote enzyme-substrate interactions required for catal ysis (Pompliano et al. 1990). It is likely that some variants prefer closed conforma tions which are more conducive to catalysis. For example, in variant Q248P/S249G/K 250P/G252W, the indole ring and the pyrrolidine rings at the 252 and 248/250 positions, resp ectively, might restrict conformational plasticity of the loop regi on and promote a closed conformation more favorable to catalysis. Alternatively, a higher kcat might result from stabilization of the transition state intermediates. Although the protein ligands for iron s ubstrate remain to be unequivocally identified, nitrogenous and/or oxygenous side ch ains appear to coordi nate metal substrate in ferrochelatase (Franco et al 1995; Gora et al. 1996b; Ferrei ra et al. 2002). The active site residues H209 and E289 (murine ferrochelat ase numbering) have been recognized as
140 metal ligands and/or have an essential role in catalysis (Gora et al. 1996b; Lecerof et al. 2000; Franco et al. 2001). Analysis of the cr ystal structure of yeas t ferrochelatase with bound metal ions suggested th at yeast H235, E314 and S 275 (equivalent to murine residues H209, E314 and S249, respectiv ely) are involved in coordinating Co2+ and Cd2+ (Karlberg et al. 2002). Thus, the increase in the Km Fe2+ of the triple mutant S249A/K250Q/V251C may result fr om the loss of stabilizat ion provided by S249 (Table 1). Curiously, the differences in the Km Fe2+ values of the loop variants correlate with the presence of a -helix in the second domain of ferroch elatase (Figure 14). In a number of proteins, -helix residues function to coordinate metal ions required for enzymatic activity (Weaver 2000; Fodje and Al-Karadaghi 2002; Shin et al. 2003). In the structural model of murine ferrochelat ase, residues arranged along the helical edge of the -helix form a channel that connects the interior of the active site to the protein exterior (Figure 30A). The quadruple mutant Q248P/S249G/K250P/G252W, whose Km Fe2+ value was the lowest among the analyzed variants (Table 1), exhibits a long ( i.e., ten residues) -helix (Figure 30C). Given that th e unit rise per residue of a -helix is the shortest among all helical structures (Al-Kara daghi et al. 1997; Fodje a nd Al-Karadaghi 2002), the -helical alignment of residues allows the side chai ns of E289, E293 and E297 to packed more closely and thereby might promote the efficien cy of metal substrate uptake. This is consistent with the observed decrease in the Km Fe2+ value in the quadruple mutant. In contrast, S249A/K250Q/V251C is th e only variant with a regular -helix in place of the -helix (Figure 30B); this should result in a longer distance between the glutamate residues and thus slower metal transf er, which might account for a higher Km Fe2+. The mode of metal substrate in teraction proposed here
141 Figure 30. Structural features of the murine ferrochelatase models. The opening of the active site cavity consists of two N-terminal helices ( lime ) and the conserved loop ( cyan ). Substituted residues in the loop are marked ( green ). Three glutamate residues ( purple ) are located in the back of the active site cavity. (A) In wildtype ferrochelatase, two of the gl utamate residues are arranged along a -helix ( red ). (B) In triple mutant S249A/K 250Q/V251C, the glutamate re sidues are located on the helix which is in place of the -helix. (C) In quadruple mutan t Q248P/S249G/K250P/G252W, thre e glutamate residues ar e aligned along a long helix. All images were generated us ing VMD 1.8.3 (Humphrey et al. 1996). A C B
142 is consistent with metal binding studies in yeast ferrochelatase (Karlberg et al. 2002; Lecerof et al. 2003). Ac cording to this model, metal bi nding would require the conserved H209 and occur on the same side of active site as that of porphyrin binding, whereas the distal residues, includi ng glutamates in the -helical region, play a regulatory role by promoting metal release from the primar y binding site (Lecer of et al. 2003). Overall, a mutational surv ey of the conserved loop motif (Q248-L257, murine ferrochelatase numbering) demonstrated that th e loop possessed a considerable degree of plasticity. While multiple functional substitutions were tolerated, the positions occupied by Q248, S249, G252, W256 and L257 exhibited th e highest stringency. Kinetic analysis of selected, active variants s uggested that loop mutations resu lt in a general disruption of interaction between the porphyrin substrate an d ferrochelatase. Further, the loop motif was also shown to play an im portant role in maintaining the active site architecture as well as modulate ferrochelatase interaction with the metal substrate. Resonance Raman spectroscopy analysis of ferrochelatase-induced porphyrin distortion The most widely accepted reaction mechan ism for ferrochelatase suggests that a critical step involves distortion of the por phyrin macrocycle to expose the lone-pair orbitals of pyrrole nitrogens to incoming metal ion (Da iley and Dailey 2003). This notion has received support from both experime ntal and theoretical studies (Cochran and Schultz 1990; Blackwood et al. 1997; Black wood et al. 1998; Lece rof et al. 2000). Antibodies raised against non-planar alkyl ated porphyrins were found to catalyze porphyrin metallation by Zn2+ and Cu2+ (Cochran and Schultz 1990). Resonance Raman
143 (RR) spectra of antibody-bound mesoporphyr in revealed a saddling distortion (Blackwood et al. 1998), and X-ray crystal structure of the Michaelis complex of catalytic antibody bound to substrate mesopor phyrin in the absence of metal ions demonstrated out-of-plane macrocycle dist ortion including saddling, ruffling and some doming deformations (Yin et al. 2003). These studies indi cated that an tibody-induced porphyrin distortion resembles deformation observed in N -alkyl porphyrins (Blackwood et al. 1998; Yin et al. 2003). Recent studies to wards quantification of distortion led to the proposal that activation of the Raman band assigned to 15, an out-of-plane vibrational mode for bound porphyrin, is directly related to the degree of affinity maturation of catalytic antibodies (Venkate shrao et al. 2004). Theoreti cal methods including quantum and molecular mechanics calculations have been used to calculate the energetics associated with porphyrin distor tion, especially within protein matrices (Sigfridsson and Ryde 2003). These model studies indicated that other than exposing pyrrole nitrogens to incoming metal ions, porphyrin de formation favors decreases in p Ka of the macrocycle, thereby promoting deprotonation of the pyrro le rings. Since deprotonation precedes metal insertion, catalytic reaction is allowed to progress. In add ition, porphyrin distortion may facilitate product release, as metalloporphyri ns are harder to distort than free-base porphyrin and thus have lower affinity for th e active site (Sigfridsson and Ryde 2003). Previous RR spectroscopic analysis of porphyrin binding to wild-type murine ferrochelatase showed that the enzyme coul d induce porphyrin distortion even in the absence of metal substrate (Franco et al. 2000; Lu et al. 2002). A high degree of specificity for porphyrin interaction with the protein was established by titrating ferrochelatase with a substoichiometric am ount of porphyrin, which allowed observation
144 of the vibrational modes of bound porphyrins in the RR spectra, and revealed frequency shifts in the structure-sensitive lines upon protein binding (Lu et al. 2002). In particular, 15, an out-of-plane saddling-like mode of B2u symmetry, was activat ed in the RR spectra of free-base protoporphyrin bound to wild-type ferrochelatas e (Lu et al. 2002). Saddling was also shown to be a major compone nt of the nonplanar distortion of N -methylmesoporphyrin when bound to B. subtilis ferrochelatase by X-ray crystal structural analysis (Lecerof et al. 2000). Significantly, saddling is an out-of-plane deformation that exposes both the protons and lone-pair orbitals of the nitrogen atoms of the macrocycle, thus consistent with a porphyr in set-up for metal insertion in a reaction facilitated by ferrochelatase. Previous studies have shown that a c onserved active site loop motif mediates interaction of porphyrin with ferrochelatase (Shi and Ferre ira 2004). Porphyrin binding to the loop variants was examined using RR sp ectroscopy. Analysis of the RR spectra for substrate protoporphyrin de monstrated that the 15 mode was activated upon porphyrin binding to wild-type protein a nd variants (Figure 16A). C onsistent with previous RR studies (Franco et al. 2000; Lu et al. 2002), this obse rvation confirmed porphyrin saddling is an important form of non-planarity induced by the enzyme. Further, the extent of porphyrin saddling could be quantif ied by normalizing the integrated intensity of 15 line to the intensity of 7 line for the same spectrum (Table 3). Interestingly, as the variants became deficient in producing saddl ed porphyrins, their cat alytic efficiencies were diminished accordingly (Table 3). Th e strongest reduction was observed in variant K250M/V251L/W256Y in which po rphyrin saddling was less th an 30% of that in the wild-type enzyme (Table 3). This dimini shed out-of-plane distortion was accompanied
145 Proteins 7 (cm-1) 7 intensity (area) 7 total intensity (area) 15 (cm-1) 15 total intensity (area) 15/ 7 relative intensity % Wildtype 15/ 7 relative intensitya kcat / Km PPIX (min-1 M-1) Wild-type ferrochelatase 673.3 678.0 1624.5 1408.8 3033.3 700.4 455.8 0.150 100.0 2.93 P255R 674.3 679.0 1059.6 745.6 1805.2 701.0 244.4 0.135 90.1 2.94 S249A/K250Q/ V251C 673.8 678.5 2166.5 1517.9 3684.4 700.8 481.8 0.131 87.0 2.63 P255G 675.3 682.7 1752.3 514.4 2266.7 700.6 126.0 0.056 37.0 2.35 Q248P/S249G/ K250P/G252W 674.9 679.8 6632.7 2349.2 8981.9 701.0 1088.9 0.121 80.7 1.36 K250M/V251L /W256Y 676.8 685.5 3124.6 531.9 3656.5 701.3 158.6 0.043 28.8 0.27 Table 3. Results of simulation for the low-frequency resonance Raman spectra of ferrochelatase-bound protoporphyrin. The Raman spectra were exported as even-X ASCII files, and spectral simulation for the frequency region from 580 to 780 cm-1 was based on least squares fitting with Lorentzian line shapes and a fixed half-width at half maximum (2.814 cm-1) for the 15 line using the program P eakFit (SYSTAT). a % wild-type 15/ 7 relative intensity was determined by normalizing the relative intensity of the 15 line versus the 7 line observed in porphyrins bound to each varian t against that obtained fr om porphyrins bound to wild-t ype ferrochelatase.
146 by a catalytic efficiency ( kcat/ Km PPIX) one order of magnitude lo wer than the wild-type enzyme (Table 3). Studies using synthetic nonplanar porphyrins dissolv ed in solvents of different polarity suggested th at saddling could affect the potential-energy surfaces to allow porphyrin to access multiple configurat ions with small energy barriers from the ground state (Sazanovich et al. 2001). Conceiva bly, a similar mechanism might be used for ferrochelatase to solvate porphyrin substr ate and to facilitate its conversion into catalytically competent forms. Similar observations have been made for porphyrin distor tion by catalytic antibodies (Venkateshrao et al. 2004). An out-of-plane vibrational mode, 15, assigned to the Raman band at 680 cm-1, was activated in mesoporphy rin upon binding to affinitymatured antibody, and its intensity was ~3-fold higher than that resulted from binding to the germLine precursor anti body (Venkateshrao et al. 2004). Somatic mutations which allowed reversion to the germLi ne residues yielded intermediate intensity values for the 15 band of bound porphyrins (Venka teshrao et al. 2004). Ind eed, an approximate linear correlation was shown between th e logarithm of catalytic efficiency and the intensity of 15 line normalized to 7. These results implied that the extent of 15 activation was mainly proportional to activation energy of th e reaction, and reiterated the idea that catalysis involves substrate straining, and binding energy is evolved to lower the activation energy required for reaction to proceed (Venkateshrao et al. 2004). The mechanisms of porphyrin metallation seem to proceed through a highly conserved transition state, regardless of whether the r eaction is catalyzed by a catalytic antibody or ferrochelatase. Out-of-plane porphyrin di stortion appears to be a common and crucial thread that connects all of the transition states.
147 It is likely that multiple factors can infl uence the competency of ferrochelatase to distort porphyrin from its na turally planar conformation. Maximal porphyrin distortion may require optimal active site geometry. Pr evious studies indicate d that loop mutations led to alteration of th e active site architecture (Shi a nd Ferreira 2004), and as a result, conformational changes of porphyr ins bound in the active sites of the loop variants might not achieve those of th e transition state for Fe2+ insertion. This is consistent with the correlation between the decrease in porphyrin saddling observed in the variants and the loss of catalytic efficiency ( kcat/ Km PPIX) (Table 3). Studies of catalytic antibodies suggested that specific residue s appeared to play a dominan t role in induc ing porphyrin distortion (Venkateshrao et al 2004). Although residues of the loops in both light and heavy chains contributed to mesoporphyrin de formation, mutational analysis indicated that heavy chain residues domin ated the activation of the 15 mode. The dominant effect of certain amino aci d residues on porphyrin deformation in ferrochelatase is best dem onstrated by variant K250M/V251L /W256Y. This variant was the least efficient in porphyrin distortion, as it induced only about 30% of the saddling seen in the wild-type enzyme. Previous studies showed that while K250 and V251 tolerated multiple replacements and were less likely to play a critical role in porphyrin interaction, functional substitutions at the W256 position were very limited suggesting the importance of W256 to enzymatic functi on (Shi and Ferreira 2004). Thus, while the possibility that K250, V251 and W 256 act synergistically in variant K250M/V251L/W256Y to promote deformation of the porphyrin substrate cannot be dismissed, it seems unlikely that K250 and V2 51 are themselves the major determinants in porphyrin distortion. In the structural m odel of murine ferrochelatase, K250 side chain
148 was oriented towards the protein exterior, thus making it very unlikely that K250 would contact the bound porphyrin directly (Figur e 31). Although the side chain of V251 pointed into active site pocke t (Figure 31), porphyrin bindi ng to variant V251L yielded RR spectra similar to that of wild-type pr otein (data not shown), indicating that the V251L mutation alone does not alter the de gree of porphyrin dist ortion from that observed for the wild-type enzyme. In ag reement with these observations, variants S249A/K250Q/V251C and Q248P /S249G/K250P/G252W, whic h also contained K250 and V251 substitutions, exhibited only a mild decrease in porphyrin saddling at 10-15% of the wild-type level (Table 3). The inva riant W256 oriented it s bulky aromatic side group towards the interior of the active site cleft. X-ray cr ystal structural analysis of B. subtilis ferrochelatase indicated that W230, equi valent to the murine W256 residue, was in close proximity to the bound N -methyl-mesoporphyrin (Lecerof et al. 2000). In particular, the minimum distance between the C6 of tryptophan indole group and the nitrogen of pyrrole ring C of porphyrin was only ~3 This made it possible for the tryptophanyl indole ring to be positioned to tilt down the pyrrole group in a manner required for generating the observed non-planar structure. Based on this model, it seems likely that W256Y substitution in murine ferr ochelatase would make it difficult for the side group to reach the proximal pyrrole ring and tilt it down, because tyrosine presents a shorter phenolic ring than the indole ring in tryptophan (Figure 31). Taken together, the loop residue W256 appears to play a dominant role in producing porphyrin deformation in the active site matrix of murine ferrochelatase. In addition to modulating saddling of th e porphyrin core, loop residues are found to mediate interaction of the porphyrin vinyl substituents with the act ive site. While the
149 A B C Figure 31. Three-dimensional views of the po rphyrin-binding cleft in ferrochelatase. (A) The active site of B. subtilis ferrochelatase with bound N -MeMP as shown in the Xray crystal structure (PDB code 1C1H). The side chain of the conserved W230 ( green ) i n the loop motif ( cyan ) lies in close proximity to py rrole ring C of bound porphyrin ( red ). (B) The active site (arrow) in the structural mo del of wild-type murine ferrochelatase. The side group of the conserved W256 ( green ) in the loop motif (cyan) is oriented towards the interior of porphyr in-binding pocket. (C)The active site (arrow) in the structural model o f variant K250M/V251L/W256Y. Among the mutated residues in the loop ( cyan ), the side chains of L251 ( blue ) and Y256 ( green ) are directed towards the interior of the porphyrinbinding cleft, while the side group of M250 ( orange ) points to the protein exterior.
150 vinyl groups adopted defined conformations upon porphyrin binding to wild-type protein, the interaction was generally weakened in loop variants, possibly due to global changes in the active site structure (Shi and Ferreira 2004). The RR spectra of protoporphyrin and NiPP revealed sharpening of the vinyl stretching band Ca=Cb upon protein binding (Figures 16B, 18B), suggesting that vinyl groups had a reduced freedom of movement and adopted a more restricted orie ntation within the protein matrix However, in variant K250M/V251L/W256Y, the vinyl gr oups of protoporphyrin appeared less defined than in wild-type protein and other varian ts (Figure 16B). Notably, the Ca=Cb band was downshifted suggesting rotation of the vinyl si de chains into the plane of the porphyrin core, whereas broadening of the Ca=Cb line indicated that vinyl groups might take alternative conformations (Kalsbeck 1995; Marzocchi and Smulevich 2003). The loop residues may also function to rest rict the configura tion of porphyrin ring imposed by the active site. Using 2 as a core-size marker (Parthasarathi 1987), hemin exhibited expansion of the macrocycle upon bi nding to wild-type ferrochelatase (Figure 17B). The ring was further en larged in the variants P255R and P255G, as indicated by the downshift of the 2 line for bound hemin (Figure 17B). However, in the active sites of the variants K250M/V251L /W256Y, S249A/K250Q /V251C and Q248P/S249G/K250P/G252W, the macrocycle was more contracted as suggested by the upshift of 2, making it resemble unbound hemin. For NiPP binding, the 2 and 10 lines were downshifted in wild-type protein and the variants (Fig ure 18B), indicating that the ring was similarly expanded in the prot ein matrix. While unbound NiPP adopted both planar and non-planar conformations in solu tion, as indicated by the deconvolution of the 10 band into a planar and non-planar compone nt, only the non-planar form was retained
151 in the protein environment (Figure 18B), re miniscent of the conf ormation previously observed for NiPP bound to wild-type enzyme (Franco et al. 2000). In conclusion, the conserve d active site loop motif in ferrochelatase plays an important role in porphyrin binding and di stortion. Most significantly, the induced porphyrin saddling is show n to modulate catalytic efficien cy of ferrochelatase towards its substrate, and the invariant tryptophan is a major protein determinant of the saddled conformation of the macrocycle. Additionally, the loop residues function in specifically orienting the porphyrin vinyl substituents and likely th e entire porphyrin ring. Inhibition of ferrochelatase by N -methyl protoporphyrin N -methyl protoporphyrin has long been known as a poten t inhibitor of mammalian ferrochelatase (Dailey and Flemi ng 1983; Cole and Marks 1984; Nunn et al. 1988). Since alkylation of a pyrrole nitrog en atom introduces nonplanarity into the porphyrin macrocycle (Caughey and Iber 1963), NMPP is thought to bear structural resemblance to the distorted porphyrin inte rmediate in the enzy matic reaction, and therefore is strongly inhibitory to ferrochelatase catalysis. N -alkylated porphyrins are generated in vivo from modification of the cytoch rome P450 heme moiety upon drug exposure (Marks et al. 1988; Lavigne et al. 2002). Xenobiotics taken up by hepatic cells have been shown to bind to the active s ite of P450 and inactiv ate the enzyme by N alkylation of a heme pyrrole ring (Marks et al. 1988). Recently, N -alkylporphyrin formation has been associated with speci fic isoforms of P450 undergoing mechanismbased inactivation in human liver mi crosomes (Gamble et al. 2003).
152 While wild-type ferrochelatase showed high affinity towards NMPP consistent with previous studie s (Dailey and Fleming 1983), the st ability of this complex was reduced due to mutations in loop residue P 255. Studies of enzymes including thrombin (Lewis et al. 1998) and pyruvate dehydrogenase (Nemeria et al. 2001) suggested that a slow conformational change of the proteininhibitor complex could play an important role in providing high affinity binding of a substrate analog to the protein, because it would ensure optimal alignment of the ligand in the active site environment. In order to find out whether this mechanism is applicable to ferrochelatase, the rate constants for NMPP binding to ferrochelatase were determined using stopped-flow analysis. For wildtype enzyme, the inhibitor bi nding process was found to cons ist of two kinetic steps. There was a fast step for initial binding of NMPP to the enzyme (Figure 23A). It was followed by a slow step, which might be rela ted to conformational rearrangement of the protein-inhibitor complex (Figure 23A). Consistent with this mechanism, crystallographic studies revealed that binding of porphyrin analog N -methyl mesoporphyrin to B. subtilis ferrochelatase triggered a si gnificant conformational change of the active site pocket involving the loop residues (Lecerof et al. 2000). Similar to wild-type enzyme, binding of NMPP to variant P255R also proceeded in two steps (Figure 23B). For the slow step i.e. the second step, the forward and reverse rate constants k2 and k-2, were comparable to those of wild-type protein. Therefore diminished inhibition of varian t P255R did not stem from ch anges in the rate of NMPP binding. In variant P255G, NMPP binding o ccurred in one step and did not involve conformational change of the protein-inhibitor complex (Fi gure 23C). Nevertheless, the rate constants k1 and k-1 for variant P255G were similar to those determined for the rate-
153 limiting step, i.e. k2 and k-2, in wild-type protein and vari ant P255R, suggesting that the rate of NMPP binding was unlikel y to contribute to its lessened inhibitory effect on P255G. It is possible that binding affinity fo r NMPP can be modulated by the steric features of the active site of ferrochelatase. Struct ural modeling suggested that loop mutations result in alterations in the active site architecture (Shi and Ferreira 2004). Examination of the dimension of the active si te cavity using the CAST program indicates that the size of porphyrin-bindi ng pocket is reduced by 30% in variant P255R and 50% in variant P255G relative to wild-type prot ein. Based on these observations, it seems plausible that the P255 resi due plays a role in supporting an open conformation of the active site in wild-type ferroch elatase, whereas P255 variants are deficient in opening the active site cleft, which can lead to restrict ed NMPP entry and decreas ed binding affinity. This possibility is consistent with the func tion of proline as a conformational switch, which has been demonstrated in a variety of proteins such as interleukin-2 tyrosine kinase and serine proteases (Mallis et al 2002; Bobofchak et al. 2005). In spite of marked differences with respect to NMPP binding, the variants exhibited kinetic properties similar to wild-t ype enzyme. While wild-type ferrochelatase had a kcat of 4.1 0.3 min-1, the catalytic activity was slightly increased in the variants, with a kcat of 7.8 0.8 min-1 for P255R and 5.9 1.4 min-1 for P255G (Table 1 and unpublished results). The Km PPIX for wild-type enzyme was 1.4 0.2 M, and it was slightly higher in the variants with 2.65 0.44 M for P255R and 2.51 0.80 M for P255G (Table 1 and unpublishe d results). Thus, the sp ecificity constant for protoporphyrin, i.e. kcat/ Km PPIX, was 2.92 for wild-type ferroc helatase, 2.94 for P255R and
154 2.35 for P255G. These results imply that specificity towards the endogenous substrate porphyrin is distinct from selectivity to wards NMPP, and the mechanism involving substrate porphyrin interaction in the catalytic pathway might be different from that for inhibitor binding. Overall, while selectivity towards NMPP is prone to perturbation as a result of active site mutations, it is much more difficult to alter specificity towards substrate porphyrin. It is c onceivable that the active site of ferrochelatase is optimized for interaction with substrate protoporphyrin and retains a certain degree of plasticity towards this end. Possibly, this facilitates induction of porphyrin deformation in the protein matrix, and in so doing, ensures catalyt ic reaction to proceed. In contrast, active site does not appear to be flexible towards interacti on with porphyrin analog such as NMPP. One potential application of the ferrochelatase variants with improved tolerance towards NMPP is to use them in the ce ll assay system for studying physiological responses resulted from heme deficiency. Recent reports have suggested that heme deficiency is associated with various disord ers such as Alzheimers disease, frataxindeficiency mediated diseases and oxidative damages resu lted from hyperoxia (Atamna and Frey 2004; Campian et al 2004; Schoenfeld et al. 2005) In these studies, NMPP treatment of cultured cells has been used as a convenient model system to induce heme deficiency (Atamna et al. 2001; Campian et al. 2004). Interestingly, NMPP-treated cells typically exhibited impaired enzymatic acti vity and assembly of mitochondrial electron transport complex IV, i.e. cytochrome c oxi dase (COX) (Tangeras 1986; Atamna et al. 2001; Campian et al. 2004). COX has been pr oposed to have a cyt oprotective role by removing reactive oxygen species, and COX defici ency renders cells mo re susceptible to
155 oxidative stress (Atamna et al. 2001; Campian et al. 2004). It shoul d be interesting to find whether the observed NMPP-induced cytot oxicity can be alleviated by expressing constitutively active ferrochelatase, for inst ance, the P255 variants which would confer resistance to NMPP inhibiti on and thereby keep heme synthesis uninterrupted. In summary, ferrochelatase variants P255R and P255G have been found to be much less inhibited by the porphyrin analog NMPP when compared to wild-type protein. This is likely to derive from structural features of the mu tant active sites and not due to changes in the rate constants of inhibitor bindi ng. This raises the possibility of lowering selectivity towards active site-directed inhibitor without compromising substrate specificity and catalysis, for instance, through modification of protein conformation. The variants can also be applied to disease studies relevant to heme deficiency, for example, to protect cells from cytotoxicity resulted from heme deficiency. FeS cluster assembly and oligomeric organization in ferrochelatase Wild-type mammalian ferrochelatase has be en shown to exis t as a homodimer (Straka et al. 1991; Wu et al. 2001), and ther e is one [2Fe-2S] clus ter associated with each monomeric subunit (Wu et al. 2001). FeS cluster is coordinated by four cysteine ligands C403, C406 and C411 in a conserved Cterminal motif and a distant C196 in the N-terminal portion [(Crouse et al. 1996; Wu et al. 2001) and Figure 32]. Mutations in the C-terminus including cysteine ligand substitutions resulted in deficiency in catalytic activity and cluster assembly, suggesting that the FeS cluster plays an important role in wild-type protein func tion (Brenner et al. 1992; Dailey et al. 1994a; Crouse et al. 1996). The current study indicates while FeS cluster as sembly is important to enzymatic activity
156 Figure 32. Position of the FeS cluster relative to the active site in ferrochelatase. Locations of the [2Fe-2S] cluster (sulfur atoms in pink and iron atoms in green ) and the four coordinating cysteine ligands ( red ) are shown in the X-ray crystal structure o f human ferrochelatase ho modimer (PDB code 1hrk ). The active site cavity in a monomeric subunit is indicated by an arrow. The loop motif is marked in cyan The homology model of murine ferrochelatase ( lime ) is superimposed onto the monomeric structure of human ferrochelatase ( yellow ). Image was generated using VMD 1.8.3 (Humphrey et al. 1996).
157 of wild-type murine ferrochel atase, it is dispensible in variants S249A/K250Q/V251C and S249A/K250R/G252W (Figure 25). This implies that the catalytic mechanism of ferrochelatase may be independent of FeS clus ter, which is in agreement with previous observations (Medlock and Dailey 2000). Usi ng chimeric constructs of human and yeast ferrochelatase, it has been shown that a fusi on protein containing the C-terminal tail from yeast enzyme and the remaining of human enzyme was active without having the FeS cluster from human ferrochelatase (Medlock and Dailey 2000). Also, truncated human ferrochelatase lacking the C-terminal cluste r-coordinating region wa s catalytically active (Ohgari et al. 2005). Because the catalytic core of ferroch elatase is highly conserved (Dailey and Dailey 2003), it is likely that the reaction pathway for mammalian enzyme is analogous to that for the clusterless enzyme from B. subtilis and S. cerevisiae and does not directly involve FeS cluster (Al-Kara daghi et al. 1997; Karl berg et al. 2002). Consistent with the notion that mammalian ferrochelatase exists as a homodimer (Straka et al. 1991; Wu et al 2001), size determination by gel filtration analysis (Figure 26) and dynamic light scattering (Figure 27) indicated that purified wild-type murine ferrochelatase is a homodimer in solution. Nevertheless, higher order oligomers were found in variants S249A/K250Q/V251C and S249A/K250R/G252W and they remained enzymatically active (Figure 26 and 27). Thes e observations suggest that ferrochelatase can adopt alternative quaterna ry structures other than hom odimer, and oligomerization state is not directly related to enzymatic activity. In line w ith this proposal, recent studies of human ferrochelatase showed that a w ild-type monomer was capable of forming a heterodimer with a mutant monomer encoded by alleles derived from EPP patients and the heterodimers were cataly tically active (Najahi-Missaoui and Dailey 2005; Ohgari et
158 al. 2005). Interestingly, the presence of hi gh molecular weight species in loop variants S249A/K250Q/V251C and S249A/K 250R/G252W suggest that th e quaternary structure of ferrochelatase can be modified as a result of protein struct ural changes, and therefore is not completely controlled by interaction of residues on the dimer interface as suggested in the crystal structural analysis of human ferrochelatase (Wu et al. 2001). This result is reminiscent of a conformational study of porphobilinogen synthase, which exhibits dynamic equilibrium of quaternary structur es (Jaffe 2005). Determining subunit arrangement in the variants, for instance, by X-ray crystallography, may help to further understand the structural basis of fe rrochelatase oligomerization. EPR spectroscopic analysis indicated that similar to wild-type protein, [2Fe-2S] is a predominant, if not the only, form of clus ter in the variants (Figure 28). Although attempts were made to examine the temperature-dependence of EPR spectra in order to differentiate the possibilities that there are mi xed-type clusters such as the [4Fe-4S] and [2Fe-2S] forms, further analyses will be re quired to decipher the sp ectral features (Figure 29). Because variant S 249A/K250Q/V251C and S249A/K250R /G252W can exist in large oligomeric assemblies, the FeS cluster does not appear to di rectly promote subunit interaction. This proposal is consistent with recent studies of human ferrochelatase showing that homodimers and heterodimers c ould form among the wild-type and mutant protein subunits in the abse nce of cluster (Najahi-Missa oui and Dailey 2005; Ohgari et al. 2005). Therefore, although [2 Fe-2S] cluster has been f ound near the dimer interface in human ferrochelatase crystal structure (Wu et al. 2001), th e notion that th e cluster acts to facilitate dimerization may be questionable.
159 Taken together, the current characterizati on of ferrochelatase variants suggests an interesting aspect of the catalytic mechanism. It seems plausible that catalytic reaction may take place on a scaffold provided by higher order oligomeric forms as revealed in the variants, and the reaction may proceed ev en without FeS cluster. This helps to explain the observation th at variants S249A/K250Q/V 251C and S249A/K250R/G252W retained significant amount of enzymatic activ ity in the absence of cluster assembly, i.e. in PK4331 cells, when compared to the contro l levels, i.e. in RZ4500 cells, whereas wild-type protein was inactivated by the loss of cluster assembly (Figure 25). An alternative hypothesis is that the FeS cluster plays a role in stabi lizing the active site conformation (Dailey 1997). Because the loop mo tif has been proposed to play a role in maintaining the active site structure (Shi and Ferreira 2004), it is lik ely that the variants can adopt alternative active site conformations which suppor t enzymatic reaction without the FeS cluster (Figure 32). Overall, the current results suggest that th e FeS cluster is not directly involved in catalysis, but rather, it is more likely to regu late enzymatic activity. One possibility is that the FeS cluster can sense the iron state in vivo and modulate enzymatic activity accordingly (Taketani et al. 2003; Lange et al. 2004). This would imply that ferrochelatase is active if an adequate amount of cluster is assembled, whereas under conditions such as iron depr ivation, FeS cluster assembly is diminished leading to inactivation of the enzyme and inhibition of the completion of heme synthesis. This possibility is particularly interesting cons idering that FeS cluster assembly has very recently been shown to modulat e erythroid heme s ynthesis by translational regulation of the first enzyme ALAS2 production (Wingert et al. 2005). Thus, by extending the
160 regulatory role of the cluster to the terminal enzyme, ferrochelatase, FeS cluster assembly machinery becomes tightly coupled to heme biosynthesis pathway, thereby providing a rapid response to iron status under physiological conditions. Investigation of the mechanism of ferroch elatase oligomerization and FeS cluster assembly are relevant to the understanding of EPP pathogenesis. Since EPP mutations found near the cluster and dimer interface are hi ghly deleterious to ferrochelatase activity (Schneider-Yin et al. 2000b; Najahi-Missaoui and Daile y 2005; Ohgari et al. 2005), regulation of enzymatic activity and protein stru cture and stability is likely to play an important part in disease development. One potential application of the variants described here is to study the regulation of ferrochelatase activity in mammalian cells. For instance, if ferrochelatase is regulated by FeS cluster assembly, expression of the variants may override this control and unc ouple protoporphyrin IX accumulation from inadequate cluster assembly. It is hoped that further stud ies of cluster assembly and oligomerization shall provide insights into the regulatory mechanism for ferrochelatase activity.
161 References Abbas, A., and Labbe-Bois, R. 1993. Structur e-function studies of yeast ferrochelatase. Identification and functional analysis of amino acid substitutions that increase Vmax and the Km for both substrates. J. Biol. Chem. 268: 8541-8546. Abitbol, M., Bernex, F., Puy, H., Jouault, H., Deybach, J.-C., Guenet, J.-L., and Montagutelli, X. 2005. A mouse mode l provides evidence that genetic background modulates anemia and liver in jury in erythropoie tic protoporphyria. Am. J. Physiol. Gastrointest. Liver Physiol. 288: G1208-1216. Ades, I.Z. 1983. Biogenesis of mitochondrial proteins regulation of maturation of deltaaminolevulinate synthase by hemin. Biochem. Biophys. Res. Commun. 110: 4247. Ades, I.Z., and Friedland, D.M. 1988. Pr operties of chicke n erythrocyte deltaaminolevulinate synthase. Int. J. Biochem. 20: 965-969. Ades, I.Z., and Harpe, K.G. 1981. Biogenesis of mitochondrial proteins. Identification of the mature and precursor forms of the subunit of delta-aminolevulinate synthase from embryonic chick liver. J. Biol. Chem. 256: 9329-9333. Ades, I.Z., and Stevens, T.M. 1988. Maturation of embryonic chick liver deltaaminolevulinate synthase: precursor pools and regulation by intra-cellularly produced heme. Int. J. Biochem. 20: 959-964. Aizencang, G., Solis, C., Bishop, D.F., Wa rner, C., and Desnick, R.J. 2000a. Human uroporphyrinogen-III synthase: genomic orga nization, alternative promoters, and erythroid-specific expression. Genomics 70: 223-231. Aizencang, G.I., Bishop, D.F., Forrest, D., Astrin, K.H., and Desnick, R.J. 2000b. Uroporphyrinogen III synthase. An altern ative promoter controls erythroidspecific expression in the murine gene. J. Biol. Chem. 275: 2295-2304. Akhtar, M. 1994. The modification of acetat e and propionate side chains during the biosynthesis of haem and chlorophylls: m echanistic and stereochemical studies. Ciba Found. Symp. 180: 131-151.
162 Al-Karadaghi, S., Hansson, M., Nikonov, S ., Jonsson, B., and Hederstedt, L. 1997. Crystal structure of ferrochelatase: the terminal enzyme in heme biosynthesis. Structure 5: 1501-1510. Alcindor, T., and Bridges, K.R. 2002. Sideroblastic anaemias. Br. J. Haematol. 116: 733743. Alexander, F.W., Sandmeier, E., Mehta, P.K., and Christen, P. 1994. Evolutionary relationships among pyridoxal-5'-phosphate -dependent enzymes. Regio-specific and families. Euro. J. Biochem. 219: 953-960. Alexeev, D., Alexeeva, M., Baxt er, R.L., Campopiano, D.J., Webster, S.P., and Sawyer, L. 1998. The crystal structure of 8-am ino-7-oxononanoate synthase: a bacterial PLP-dependent, acyl-CoA-condensing enzyme. J. Mol. Bio. 284: 401-419. Alwan, A.F., Mgbeje, B.I., and Jordan, P.M. 1989. Purification and properties of uroporphyrinogen III synthase (co-synthase) from an overproducing recombinant strain of Escherichia coli K-12. Biochem. J. 264: 397-402. Anderson, K.E., Bloomer, J.R., Bonkovsky, H.L ., Kushner, J.P., Pierach, C.A., Pimstone, N.R., and Desnick, R.J. 2005. Recommenda tions for the diagnosis and treatment of the acute porphyrias. Ann. Intern. Med. 142: 439-450. Anderson, K.S., Sikorski, J.A., and Johnson, K.A. 1988. Evaluation of 5enolpyruvoylshikimate-3-phosphate synthase substrate and inhibitor binding by stopped-flow and equilibrium fluorescence measurements. Biochemistry 27: 1604-1610. Anderson, P.M., and Desnick, R.J. 1979. Purification and properties of deltaaminolevulinate dehydratase from human erythrocytes. J. Biol. Chem. 254: 69246930. Anderson, P.M., and Desnick, R.J. 1980. Pu rification and properties of uroporphyrinogen I synthase from human erythrocytes. Id entification of stable enzyme-substrate intermediates. J. Biol. Chem. 255: 1993-1999. Ardail, D., Privat, J.P., Egret-Charlier, M., Le vrat, C., Lerme, F., and Louisot, P. 1990. Mitochondrial contact sites. Lipid composition and dynamics. J. Biol. Chem. 265: 18797-18802. Arnould, S., and Camadro, J.-M. 1998. Th e domain structure of protoporphyrinogen oxidase, the molecular target of diphenyl ether-type herbicides. Proc. Natl. Acad. Sci. 95: 10553-10558.
163 Arnould, S., Takahashi, M., and Camadro, J. -M. 1999. Acylation stabilizes a proteaseresistant conformation of protoporphyrinoge n oxidase, the molecular target of diphenyl ether-type herbicides. Proc. Natl. Acad. Sci. 96: 14825-14830. Astner, I., Schulze J. O., van den Heuvel, J ., Jahn, D., Schubert, W.D., and Heinz, D.W. 2005. Crystal structure of 5-aminolevulinat e synthase, the first enzyme of heme biosynthesis, and its lin k to XLSA in humans. EMBO J. 24: 3166-3177. Atamna, H., and Frey, W.H. 2004. A role for heme in Alzheimer's disease: Heme binds amyloidand has altered metabolism. Proc. Natl. Acad. Sci. 101: 11153-11158. Atamna, H., Liu, J.K., and Ames, B.N. 2001. Heme deficiency selectively interrupts assembly of mitochondrial complex IV in human fibroblasts. Relevance to aging. J. Biol. Chem. 276: 48410-48416. Avissar, Y.J., Ormerod, J.G., and Beale, S. I. 1989. Distribution of delta-aminolevulinic acid biosynthetic pathways am ong phototrophic bacterial groups. Arch. Microbiol. 151: 513-519. Bales, L., Day, R.S., and Blekkenhorst, G.H. 1980. The clinical and biochemical features of variegate porphyria: An analysis of 300 cases studied at Groote Schuur Hospital, Cape Town. Int. J. Biochem. 12: 837-853. Barnard, G.F., and Akhtar, M. 1979. Stereo chemical and mechanistic studies on the decarboxylation of uroporphyrinogen III in haem biosynthesis. J. Chem. Soc. Perkin Trans. I 10: 2354-2360. Battersby, A.R. 1978. The discovery of nature's biosynthetic pathways. Experientia 34: 1-13. Battersby, A.R., and Leeper, F.J. 1990. Biosynthe sis of the pigments of life: mechanistic studies on the conversion of por phobilinogen to uroporphyrinogen III. Chem. Rev. 90: 1261-1274. Bawden, M.J., Borthwick, I.A., Healy, H.M., Morris, C.P., May, B.K., and Elliott, W.H. 1987. Sequence of human 5-aminolevulinate synthase cDNA. Nucleic Acids Res. 15: 8563. Beale, S.I., Gough, S.P., and Granick, S. 1975. Biosynthesis of delta-aminolevulinic acid from the intact carbon skeleton of glutamic acid in greening barley. Proc. Natl. Acad. Sci. 72: 2719-2723. Berlin, S.O., and Brante, G. 1962. Iron meta bolism in porphyria and hemochromatosis. The Lancet 280: 729-729.
164 Binda, C., Hublek, F., Li, M., Edmondson, D.E., and Mattevi, A. 2004. Crystal structure of human monoamine oxidase B, a drug target enzyme monot opically inserted into the mitochondrial outer membrane. FEBS Lett. 564: 225-228. Bishop, D., Johansson, A., and Desnick, R. J. 2005. Congenital eryt hropoietic porphyria: generation and characterizatio n of knock-in mouse models. J. Inher. Metab. Disease 28: 258. Bishop, D.F. 1990. Two different genes encode -aminolevulinate synthase in humans: nucleotide sequences of cDNAs for th e housekeeping and erythroid genes. Nucleic Acids Res. 18: 7187. Bishop, D.F., Kitchen, H., and Wood, W.A. 1981. Evidence for erythroid and nonerythroid forms of delta-a minolevulinate synthetase. Arch. Biochem. Biophys. 206: 380-391. Blackwood, M.E., Rush, T.S., Medlock, A. Dailey, H.A., and Spiro, T.G. 1997. Resonance Raman spectra of ferrochela tase reveal porphyrin distortion upon metal binding. J. Am. Chem. Soc. 119: 12170-12174. Blackwood, M.E., Rush, T.S., Rosenberg, F. Schultz, P.G., and Spiro, T.G. 1998. Alternative modes of substrate distor tion in enzyme and antibody catalyzed ferrochelation reactions. Biochemistry 37: 779-782. Bloomer, J., Wang Y, Singhal A, and H., R. 2005. Molecular studies of liver disease in erythropoietic protoporphyria. J. Clin. Gastroenterol. 39: S167-175. Bloomer, J.R. 1988. The liver in protoporphyria. Hepatology 8: 402-407. Bloomer, J.R., Rank, J.M., Payne, W.D., Snov er, D.C., Sharp, H.L., Zwiener, R.J., and Carithers, R.L. 1996. Follow-up after liver transplantation fo r protoporphyric liver disease. Liver Transpl. Surg. 2: 269-275. Bobofchak, K.M., Pineda, A.O., Mathews, F. S., and Di Cera, E. 2005. Energetic and structural consequences of perturbing Gly-193 in the oxyanion hole of serine proteases. J. Biol. Chem. 280: 25644-25650. Boehning, D., and Snyder, S.H. 2003. Novel neural modulators. Ann. Rev. Neurosci. 26: 105-131. Boese, Q.F., Spano, A.J., Li, J.M., and Timko, M.P. 1991. Aminolevulinic acid dehydratase in pea ( Pisum sativum L .). Identification of an unusual metal-binding domain in the plant enzyme. J. Biol. Chem. 266: 17060-17066.
165 Bogard, M., Camadro, J.M., Nordmann, Y., and Labbe, P. 1989. Purification and properties of mouse liver coproporphyrinogen oxidase. Eur. J. Biochem. 181: 417-421. Bogorad, L. 1958a. The enzymatic synthesi s of porphyrins from porphobilinogen. I. Uroporphyrin I. J Biol Chem 233: 501-509. Bogorad, L. 1958b. The enzymatic synthesis of porphyrins from porphobilinogen. II. Uroporphyrin III. J Biol Chem 233: 510-515. Bogorad, L. 1958c. The enzymatic synthesis of porphyrins from porphobilinogen. III. Uroporphyrinogens as intermediates. J Biol Chem 233: 516-519. Bollivar, D., Clauson, C., Lighthall, R., Forb es, S., Kokona, B., Fair man, R., Kundrat, L., and Jaffe, E. 2004. Rhodobacter capsulatus porphobilinogen synthase, a high activity metal ion independent hexamer. BMC Biochemistry 5: 17. Bonkowsky, H.L., Bloomer, J.R., Ebert, P.S ., and Mahoney, M.J. 1975. Heme synthetase deficiency in human protoporphyria. Demons tration of the defect in liver and cultured skin fibroblasts. J. Clin. Invest. 56: 1139-1148. Bottomley, S.S., May, B.K., Cox, T.C., Cotte r, P.D., and Bishop, D.F. 1995. Molecular defects of erythroid 5-aminolevulinate sy nthase in X-linked sideroblastic anemia. J. Bioenerg. Biomembr. 27: 161-168. Boulechfar, S., Da Silva, V., Deybach, J. C., Nordmann, Y., Grandchamp, B., and de Verneuil, H. 1992. Heterogeneity of mutations in the uroporphyrinogen III synthase gene in congenita l erythropoietic porphyria. Hum. Genet. 88: 320-324. Breckau, D., Mahlitz, E., Sauerwald, A., Layer, G., and Jahn, D. 2003. Oxygendependent coproporphyri nogen III oxidase ( HemF) from Escherichia coli Is stimulated by manganese. J. Biol. Chem. 278: 46625-46631. Breinig, S., Kervinen, J., Stith, L., Wasson, A.S., Fairman, R., Wlodawer, A., Zdanov, A., and Jaffe, E.K. 2003. Control of tetrapyrrole biosynthesis by alternate quaternary forms of porphobilinogen synthase. Nat. Struct. Mol. Biol. 10: 757763. Brennan, M.J., Cantrill, R.C ., and Kramer, S. 1980. Effect of delta-aminolaevulinic acid on GABA receptor binding in synaptic plasma membranes. Int. J. Biochem. 12: 833-835. Brenner, D.A., and Bloomer, J.R. 1980. The enzymic defect in variegate porphyria. Studies with human cultured skin fibroblasts. New Eng. J. Med. 302: 765-769.
166 Brenner, D.A., Didier, J.M., Fraiser, F., Chri stensen, S.R., Evans, G.A., and Dailey, H.A. 1992. A molecular defect in human protoporphyria. Am. J. Hum. Genet. 50: 12031210. Brenner, D.A., and Frasier, F. 1991. Cloning of murine ferrochelatase. Proc. Natl. Acad. Sci. 88: 849-853. Brownlie, P.D., Lambert, R., Louie, G.V., Jordan, P.M., Blundell, T.L., Warren, M.J., Cooper, J.B., and Wood, S.P. 1994. The th ree-dimensional structures of mutants of porphobilinogen deaminase: Toward an und erstanding of the structural basis of acute intermittent porphyria. Protein Sci. 3: 1644-1650. Bulaj, Z.J., Phillips, J.D., Ajioka, R.S., Franklin, M.R., Griffen, L.M., Guinee, D.J., Edwards, C.Q., and Kushner, J.P. 2000. Hemochromatosis genes and other factors contributing to the pathogenesis of porphyria cutanea tarda. Blood 95: 1565-1571. Burden, A.E., Wu, C., Dailey, T.A., Busch, J. L., Dhawan, I.K., Rose, J.P., Wang, B., and Dailey, H.A. 1999. Human ferrochelatase: cr ystallization, char acterization of the [2Fe-2S] cluster and determination that the enzyme is a homodimer. Biochim. Biophys. Acta 1435: 191-197. Bygum, A., Christiansen, L., Petersen, N.E., Horder, M., Thomsen, K., and Brandrup, F. 2003. Familial and sporadic porphyria cutanea tarda: clinical, biochemical and genetic features with emphasis on iron status. Acta Derm. Venereol. 83: 115-120. Camadro, J.-M., and Labbe, P. 1996. Cloni ng and characterization of the yeast Hem14 gene coding for protoporphyrinogen oxida se, the molecular target of diphenyl ether-type herbicides. J. Biol. Chem. 271: 9120-9128. Camadro, J.M., Chambon, H., Jolles, J., a nd Labbe, P. 1986a. Purification and properties of coproporphyrinogen oxidase from the yeast Saccharomyces cerevisiae Eur. J. Biochem. 156: 579-587. Camadro, J.M., Chambon, M., Jolles, J., a nd Labbe, P. 1986b. Purification and properties of coproporphyrinogen oxidase from the yeast Saccharomyces cerevisiae Eur. J. Biochem. 156: 579-587. Camadro, J.M., and Labbe, P. 1982. Kinetic studies of ferrochelatase in yeast. Biochim. Biophys. Acta. 707: 280-288. Camadro, J.M., and Labbe, P. 1988. Purificati on and properties of ferrochelatase from the yeast Saccharomyces cerevisiae Evidence for a precursor form of the protein. J. Biol. Chem. 263: 11675-11682. Campian, J.L., Qian, M., Gao, X., and Eat on, J.W. 2004. Oxygen tolerance and coupling of mitochondrial electron transport. J. Biol. Chem. 279: 46580-46587.
167 Canepa, E.T., and Llambias, E.B. 1988. Puri fication and characterization of ferroand cobalto-chelatases. Biochem. Cell Biol. 66: 32-39. Cappellini, M.D., di Montemuros, F.M., Di Pierro, E., and Fiorelli, G. 2002. Hematologically important mutations: Acute intermittent porphyria. Blood Cells, Molecules and Diseases 28: 5-12. Cassidy, M.A., Crockett, N., Leeper, F.J., a nd Battersby, A.R. 1991. Synthetic studies on the proposed spiro intermediate for bios ynthesis of the natu ral porphyrins: the stereochemical probe. J. Chem. Soc., Chem. Comm. 6: 384-386. Caughey, W.S., and Iber, P.K. 1963. Ring no nplanarity and aromatic ity in porphyrins. nuclear magnetic resonance spectra of etioporphyrin II and its N-alkyl compounds. J. Org. Chem. 28: 269-270. Chaufan, G., Corvi, M.M., San Martin de Vi ale, L.C., Cardenas, M.L., and Carmen Rios de Molina, M. 2005. Abnormal kine tic behavior of uroporphyrinogen decarboxylase obtained from rats with hexachlorobenzene-induced porphyria. J. Biochem. Mol. Toxicol. 19: 19-24. Chelstowska, A., Zoladek, T., Garey, J., Kus hner, J., Rytka, J., a nd Labbe-Bois, R. 1992. Identification of amino acid change s affecting yeast uroporphyrinogen decarboxylase activity by sequence analysis of hem12 mutant alleles. Biochem. J. 288: 753-757. Chen, F.P., Risheg, H., Liu, Y., and Bloomer J. 2002. Ferrochelatase gene mutations in erythropoietic protoporphyria : focus on liver disease. Cell Mol. Biol. (Noisy-legrand) 48: 83-89. Clement, R.P., Kohashi, M., and Piper, W.N. 1982. Rat hepatic uroporphyrinogen III cosynthase: purificati on, properties, and inhibition by metal ions. Arch. Biochem. Biophys. 214: 657-667. Coates, L., Beaven, G., Erskine, P.T., Beale, S.I., Avissar, Y.J., Gill, R., Mohammed, F., Wood, S.P., Shoolingin-Jordan, P., and C ooper, J.B. 2004. The X-ray structure of the plant like 5-aminolaevulin ic acid dehydratase from Chlorobium vibrioforme complexed with the inhibitor laevul inic acid at 2.6 A resolution. J. Mol. Biol. 342: 563-570. Cochran, A.G., and Schultz, P.G. 1990. Antibody-catalyzed porphy rin metallation. Science 249: 781-783. Colas, C., and Ortiz de Montellano, P.R. 2003. Autocatalytic ra dical reactions in physiological prosthetic heme modification. Chem. Rev. 103: 2305-2332.
168 Cole, S.P., and Marks, G.S. 1984. Ferrochelatase and N -alkylated porphyrins. Mol. Cell Biochem. 64: 127-137. Colloc'h, N., Mornon, J.-P., and Camadro, J.-M. 2002. Towards a ne w T-fold protein?: The coproporphyrinogen III oxidase sequen ce matches many structural features from urate oxidase. FEBS Letters 526: 5-10. Combet, C., Jambon, M., Deleage, G., and Geourjon, C. 2002. Geno3D: automatic comparative molecular modelling of protein. Bioinformatics 18: 213-214. Coomber, S.A., Jones, R.M., Jordan, P.M., and Hunter, C.N. 1992. A putative anaerobic coproporphyrinogen III oxidase in Rhodobacter sphaeroides I. Molecular cloning, transposon muta genesis and sequence an alysis of the gene. Mol. Microbiol. 6: 3159-3169. Cooperman, S.S., Meyron-Holtz, E.G., Olivie rre-Wilson, H., Ghosh, M.C., McConnell, J.P., and Rouault, T.A. 2005. Microcytic anemia, erythropoietic protoporphyria, and neurodegeneration in mice with targeted deletion of iron-regulatory protein 2. Blood 106: 1084-1091. Cornejo, J., Willows, R.D., and Beale, S.I. 1998. Phytobilin biosynthesis: cloning and expression of a gene encoding soluble ferredoxin-dependent heme oxygenase from Synechocystis sp. PCC 6803. Plant J. 15: 99-107. Cotter, P.D., Baumann, M., and Bishop, D.F. 1992. Enzymatic defect in X-linked sideroblastic anemia: Molecular eviden ce for erythroid delta-aminolevulinate synthase deficiency. Proc. Natl. Acad. Sci. 89: 4028-4032. Cotter, P.D., May, A., Li, L., Al-Sabah, A.I ., Fitzsimons, E.J., Cazzola, M., and Bishop, D.F. 1999. Four new mutations in the erythroid-specific 5-aminolevulinate synthase (ALAS2) gene causing X-linked sideroblastic anemia: Increased pyridoxine responsiveness after removal of iron ove rload by phlebotomy and coinheritance of hereditary hemochromatosis. Blood 93: 1757-1769. Cox, T.C., Bawden, M.J., Martin, A., and May, B.K. 1991. Human erythroid 5aminolevulinate synthase: promoter an alysis and identification of an ironresponsive element in the mRNA. EMBO J. 10: 1891-1902. Crouse, B.R., Sellers, V.M., Finnegan, M. G., Dailey, H.A., and Johnson, M.K. 1996. Site-directed mutagenesis and spect roscopic characterization of human ferrochelatase: identification of residue s coordinating the [2 Fe-2S] cluster. Biochemistry 35: 16222-16229. Da Silva, V., Simonin, S., Deybach, J.C., Puy, H., and Nordmann, Y. 1995. Variegate porphyria: diagnostic value of fluoromet ric scanning of plasma porphyrins. Clinica Chimica Acta 238: 163-168.
169 Dailey, H.A. 1977. Purification and ch aracterization of the membrane-bound ferrochelatase from Spirillum itersonii. J. Bacteriol. 132: 302-307. Dailey, H.A. 1997. Enzymes of heme biosynthesis. J. Biol. Inorg. Chem. 2: 411-417. Dailey, H.A. 2002. Terminal st eps of haem biosynthesis. Biochem. Soc. Trans. 30: 590595. Dailey, H.A., and Dailey, T.A. 1996a. Protoporphyrinogen oxidase of Myxococcus xanthus J. Biol. Chem. 271: 8714-8718. Dailey, H.A., and Dailey, T.A. 2003. Ferrochelatase. In The Porphyrin Handbook (eds. K.M. Kadish, K.M. Smith, and S.R. Gu ilard), pp. 93-121. Elsevier Science. Dailey, H.A., Dailey, T.A., Wu, C.K., Medloc k, A.E., Wang, K.F., Rose, J.P., and Wang, B.C. 2000. Ferrochelatase at the millenniu m: Structures, mechanisms and [2Fe2S] clusters. Cell. Mol. Life. Sci. 57: 1909-1926. Dailey, H.A., Finnegan, M.G., and Johnson, M.K. 1994a. Human fe rrochelatase is an iron-sulfur protein. Biochemistry 33: 403-407. Dailey, H.A., and Fleming, J.E. 1983. Bovi ne ferrochelatase. Kinetic analysis of inhibition by N -methylprotoporphyrin, manganese, and heme. J. Biol. Chem. 258: 11453-11459. Dailey, H.A., Jones, C.S., and Karr, S.W. 1989. Interaction of free porphyrins and metalloporphyrins with mouse ferrochelat ase. A model for the active site of ferrochelatase Biochim. Biophys. Acta 999: 7-11. Dailey, H.A., and Karr, S.W. 1983. Purifi cation and characterization of murine protoporphyrinogen oxidase. Biochemistry 26: 2697-2701. Dailey, H.A., and Lascelles, J. 1974. Ferrochelatase activity in wild-type and mutant strains of Spirillum itersonii. Solubilization with chaotropic reagents. Arch. Biochem. Biophys. 160: 523-529. Dailey, T.A., and Dailey, H.A. 1996b. Hu man protoporphyrinogen oxidase: Expression, purification, and characteri zation of the cloned enzyme. Protein Sci. 5: 98-105. Dailey, T.A., and Dailey, H.A. 2002. Identific ation of [2Fe-2S] clusters in microbial ferrochelatases. J. Bacteriol. 184: 2460-2464. Dailey, T.A., Dailey, H.A., Meissner, P., and Prasad, A.R.K. 1995. Cloning, sequence and expression of mouse pr otoporphyrinogen oxidase. Arch. Biochem. Biophys. 324: 379-384.
170 Dailey, T.A., Meissner, P., and Dailey, H.A. 1994b. Expression of a cloned protoporphyrinogen oxidase. J. Biol. Chem. 269: 813-815. Dailey, T.A., Woodruff, J.H., and Dailey, H.A. 2005. Examination of mitochondrial protein targeting of haem synthetic en zymes: in vivo identification of three functional haem-responsive motifs in 5-aminolaevulinate synthase Biochem. J. 386: 381-386. Dandekar, T., Stripecke, R., Gray, N.K., Goos sen, B., Constable, A., Johansson, H.E., and Hentze, M.W. 1991. Identification of a novel iron-responsive element in murine and human erythr oid delta-aminolevulinic acid synthase mRNA. EMBO J. 10: 1903-1909. Davies, R.C., and Neuberger, A. 1979. Contro l of 5-aminolaevulinat e synthetase activity in Rhodopseudomonas spheroides Binding of pyridoxal phosphate to 5aminolaevulinate synthetase. Biochem. J. 177: 661-671. de Verneuil, H., Aitken, G., and Nordmann, Y. 1978. Familial and sporadic porphyria cutanea: two different diseases. Hum. Genet. 44: 145-151. de Verneuil, H., Deybach, J.C., Phung, N., Da Silva, V., and Nordmann, Y. 1983a. Study of anaesthetic agents for their ability to elicit porphyrin biosynthesis in chick embryo liver. Biochem. Pharmacol. 32: 1011-1018. de Verneuil, H., Geronimi, F., Lamrissi-Garcia, I., Richard, E., Moreau-Gaudry, F., and Morey, M. 2003. Congenital erythropoiet ic porphyria and gene therapy in erythropoietic porphyrias. J. Eur. Acad. Derm. Vener. 17: 61. de Verneuil, H., Grandchamp, B., Beaumont C., Picat, C., and Nordmann, Y. 1986. Uroporphyrinogen decarboxylase structural mu tant (Gly281----Glu) in a case of porphyria. Science 234: 732-734. de Verneuil, H., Grandchamp, B., and Nord mann, Y. 1980. Some kinetic properties of human red cell uroporphyrinogen decarboxylase. Biochim. Biophys. Acta 611: 174-186. de Verneuil, H., Sassa, S., and Kappas, A. 1983b. Purification and properties of uroporphyrinogen decarboxylase from huma n erythrocytes. A single enzyme catalyzing the four sequential decarboxyl ations of uroporphyrinogens I and III. J. Biol. Chem. 258: 2454-2460. DeLeo, V.A., Poh-Fitzpatrick, M., Ma thews-Roth, M., and Harber, L.C. 1976. Erythropoietic protoporphyr ia. 10 years experience. Am. J. Med. 60: 8-22.
171 Delfau-Larue, M.H., Martasek, P., and Grandchamp, B. 1994. Coproporphyrinogen oxidase: gene organization and descrip tion of a mutation leading to exon 6 skipping. Hum. Mol. Genet. 3: 1325-1330. Desnick, R.J., and Astrin, K.H. 2002. Conge nital erythropoietic por phyria: Advances in pathogenesis and treatment. Br. J. Haemat. 117: 779-795. Desnick, R.J., Ostasiewicz, L.T., Tishle r, P.A., and Mustaj oki, P. 1985. Acute intermittent porphyria: characterization of a novel mutation in the structural gene for porphobilinogen deaminase. Demons tration of noncatalytic enzyme intermediates stabilized by bound substrate. J. Clin. Invest. 76: 865-874. Deybach, J.C., da Silva, V., Grandchamp, B., and Nordmann, Y. 1985. The mitochondrial location of protoporphyrinogen oxidase. Eur. J. Biochem. 149: 431-435. Deybach, J.C., de Verneuil, H., and Nordmann, Y. 1981. The inherited enzymatic defect in porphyria variegata. Hum. Genet. 58: 425-428. Deybach, J.C., Puy, H., Robreau, A.M., Lamoril, J., Da Silva, V., Grandchamp, B., and Nordmann, Y. 1996. Mutations in the protoporphyrinogen oxidase gene in patients with variegate porphyria. Hum. Mol. Genet. 5: 407-410. Dioum, E.M., Rutter, J., Tuckerman, J.R., Gonzalez, G., Gilles-Gonzalez, M.-A., and McKnight, S.L. 2002. NPAS2: A gas -responsive transcription factor. Science 298: 2385-2387. Doss, M., Von Tiepermann, R., Schneider, J., and Schmid, H. 1979. New type of hepatic porphyria with porphobilinogen synthase de fect and intermittent acute clinical manifestation. Klin. Wochenschr. 57: 1123-1127. Doss, M.O., Stauch, T., Gross, U., Renz, M ., Akagi, R., Doss-Frank, M., Seelig, H.P., and Sassa, S. 2004. The third case of Do ss porphyria (delta-aminolevulinic acid dehydratase deficiency) in Germany. J. Inher. Metab. Disease 27: 529-536. Duncan, R., Faggart, M.A., Roger, A.J., and Cornell, N.W. 1999. Phylogenetic analysis of the 5-aminolevulinate synthase gene Mol. Biol. Evol. 16: 383-396. Dupuis-Girod, S., Akkari, V., Ged, C., Galambrun, C., Ke bali, K., Deybach, J.-C., Claudy, A., Geburher, L., Philippe, N., de Verneuil, H., et al. 2005. Successful match-unrelated donor bone marrow transpla ntation for congen ital erythropoietic porphyria (Gnther disease). Eur. J. Pediat. 164: 104-107. Eftink, M.R., and Ghiron, C.A. 1976. Expos ure of tryptophanyl residues in proteins. Quantitative determination by fluorescence quenching studies. Biochemistry 15: 672-680.
172 Egger, N.G., Goeger, D.E., Payne, D.A ., Miskovsky, E.P., Weinman, S.A., and Anderson, K.E. 2002. Porphyria cutanea tarda: multiplicity of risk factors including HFE mutations, hepatitis C, and inherited uroporphyrinogen decarboxylase deficiency. Digest. Dis. Sci. 47: 419-426. Elder, G.H., and Evans, J.O. 1978a. Eviden ce that coproporphyrinoge n oxidase activity of rat liver is situat ed in the inte rmembrane space of mitochondria. Biochem. J. 172: 345-347. Elder, G.H., and Evans, J.O. 1978b. A radi ochemical method for the measurement of coproporphyrinogen oxidase and the utiliz ation of substrates other than coproporphyrinogen III by the enzyme from rat liver. Biochem. J. 169: 205-214. Elder, G.H., Evans, J.O., Jackson, J.R., and Jackson, A.H. 1978a. Factors determining the sequence of oxidative decarboxylation of the 2and 4-propionate substituents of coproporphyrinogen III by coproporphyr inogen oxidase in rat liver. Biochem. J. 169: 215-223. Elder, G.H., Lee, G.B., and Tovey, J.A. 1978b. Decreased ac tivity of hepatic uroporphyrinogen decarboxylase in spor adic porphyria cu tanea tarda. N. Engl. J. Med. 299: 274-278. Elder, G.H., Tovey, J.A., and Sheppard, D.M. 1983. Purification of uroporphyrinogen decarboxylase from human erythrocytes. Immunochemical evidence for a single protein with decarboxylase activity in human erythrocytes and liver. Biochem. J. 215: 45-55. Elder, G.H., and Worwood, M. 1998. Mutations in the hemochromatosis gene, porphyria cutanea tarda, and iron overload. Hepatology 27: 289-291. Erskine, P.T., Coates, L., Butler, D., Youell, J.H., Brindley, A.A ., Wood, S.P., Warren, M.J., Shoolingin-Jordan, P.M., and Coope r, J.B. 2003. X-ray structure of a putative reaction intermediate of 5aminolaevulinic acid dehydratase. Biochem. J. 373: 733-738. Erskine, P.T., Coatesa, L., Newboldb, R., Brindleyb, A.A., Stau fferc, F., Wooda, S.P., Warrenb, M.J., Coopera, J.B., ShoolinginJordana, P.M., and Neierc, R. 2001a. The X-ray structure of yeast 5-aminolae vulinic acid dehydratase complexed with two diacid inhibitors. FEBS Lett. 503: 196-200. Erskine, P.T., L. Coates, R. Newbold, A. A. Brindley, F. Stauffer, G. D. E. Beaven, R. Gill, A. Coker, S. P. Wood, M. J. Warren, et al. 2005. Structure of yeast 5aminolaevulinic acid dehydratase complexed with the inhibitor 5hydroxylaevulinic acid. Acta Cryst. 61: 1222-1226.
173 Erskine, P.T., Newbold, R., Brindley, A. A., Wood, S.P., Shoolingin-Jordan, P.M., Warren, M.J., and Cooper, J.B. 2001b. The X-ray structure of yeast 5aminolaevulinic acid dehydratase complexed with substrate and three inhibitors. J. Mol. Biol. 312: 133-141. Erskine, P.T., Norton, E., Cooper, J.B., Lamb ert, R., Coker, A., Lewis, G., Spencer, P., Sarwar, M., Wood, S.P., Warren, M.J ., et al. 1999. X-ray structure of 5aminolevulinic acid dehydratase from Escherichia coli complexed with the inhibitor levulinic acid at 2.0 A resolution. Biochemistry 38: 4266-4276. Erskine, P.T., Senior, N., Awan, S., Lambert, R., Lewis, G., Tickle, I.J., Sarwar, M., Spencer, P., Thomas, P., Warren, M.J ., et al. 1997. X-ray structure of 5aminolaevulinate dehydratase, a hybrid aldolase. Nat. Struct. Biol. 4: 1025-1031. Fanica-Gaignier, M., and Clemen t-Metral, J. 1973. 5-Aminole vulinic-acid synthetase of Rhodopseudomonas spheroides Y Kinetic mechanism and inhibition by ATP. Eur. J. Biochem. 40: 19-24. Feder, J.N., Gnirke, A., Thomas, W., Ts uchihashi, Z., Ruddy, D.A., Basava, A., Dormishian, F., Domingo, R., Ellis, M. C., Fullan, A., et al. 1996. A novel MHC class I-like gene is mutated in patien ts with hereditary haemochromatosis. Nat. Genet. 13: 399-408. Felix, F., and Brouillet, N. 1990. Purificat ion and properties of uroporphyrinogen decarboxylase from Saccharomyces cerevisiae Yeast uroporphyrinogen decarboxylase. Eur. J. Biochem. 188: 393-403. Ferreira, G.C. 1994. Mammalian ferro chelatase. Overexpression in Escherichia coli as a soluble protein, purificat ion and characterization. J. Biol. Chem. 269: 4396-4400. Ferreira, G.C., Andrew, T.L., Karr, S.W., and Dailey, H.A. 1988. Organization of the terminal two enzymes of the heme bi osynthetic pathway. Orientation of protoporphyrinogen oxidase and evidence for a membrane complex. J. Biol. Chem. 263: 3835-3839. Ferreira, G.C., and Dailey, H.A. 1993. Expression of mammalian 5-aminolevulinate synthase in Escherichia coli Overproduction, purificati on, and characterization. J. Biol. Chem. 268: 584-590. Ferreira, G.C., Franco, R., Lloyd, S.G., Pereir a, A.S., Moura, I., Moura, J.J., and Huynh, B.H. 1994. Mammalian ferrochelatase, a new addition to the metalloenzyme family. J. Biol. Chem. 269: 7062-7065. Ferreira, G.C., Franco, R., Mangravita, A ., and George, G.N. 2002. Unraveling the substrate-metal binding site of ferrochelat ase: An X-ray absorption spectroscopic study. Biochemistry 41: 4809-4818.
174 Ferreira, G.C., Neame, P.J., and Dailey, H.A. 1993. Heme biosynthesis in mammalian systems: Evidence of a Schiff base li nkage between the pyridoxal 5'-phosphate cofactor and a lysine residue in 5-aminolevulinate synthase. Protein Sci. 2: 19591965. Ferreira, G.C., Vajapey, U., Hafez, O ., Hunter, G.A., and Barber, M.J. 1995. Aminolevulinate synthase: Lysine 313 is not essential for binding the pyridoxal phosphate cofactor but is essential for catalysis. Protein Sci 4: 1001-1006. Ferris, C.D., Jaffrey, S.R., Sawa, A., Takahash i, M., Brady, S.D., Barrow, R.K., Tysoe, S.A., Wolosker, H., Baranano, D.E., Do re, S., et al. 1999. Haem oxygenase-1 prevents cell death by re gulating cellular iron. Nat. Cell Biol. 1: 152-157. Floderus, Y., Shoolingin-Jordan, P.M., and Ha rper, P. 2002. Acute intermittent porphyria in Sweden. Molecular, functional and clinical consequences of some new mutations found in the porphobilinogen deaminase gene. Clin. Genet. 62: 288297. Fodje, M.N., and Al-Karadaghi, S. 2002. O ccurrence, conformational features and amino acid propensities for the -helix. Protein Eng. 15: 353-358. Fontanellas, A., Bensidhoum, M., Enriquez de Salamanca, R., Moruno Tirado, A., de Verneuil, H., and Ged, C. 1996. A systema tic analysis of the mutations of the uroporphyrinogen III synthase gene in congenital erythropoietic porphyria. Eur. J. Hum. Genet. 4: 274-282. Fontanellas, A., Frdric Mazurier, Marc Landry, Laurence Taine, Carine Morel, Monique Larou, Jean-Yves Daniel, Xavi er Montagutelli, Rafael Enriquez de Salamanca, and de Verneuil, H. 2000. Re version of hepatob iliary alterations by bone marrow transplantation in a murine model of erythropoietic protoporphyria. Hepatology 32: 73-81. Fontanellas, A., Mendez M, Mazurier F, Ca rio-Andre M, Navarro S, Ged C, Taine L, Geronimi F, Richard E, Moreau-Gaudry F, et al. 2001. Successful therapeutic effect in a mouse model of erythropoi etic protoporphyria by partial genetic correction and fluorescence-based se lection of hematopoietic cells. Gene Ther. 8: 618-626. Franco, R., Ma, J.G., Lu, Y., Ferreira, G.C., and Shelnutt, J.A. 2000. Porphyrin interactions with wild-type a nd mutant mouse ferrochelatase. Biochemistry 39: 2517-2529. Franco, R., Moura, J.J.G., Moura, I., Ll oyd, S.G., Huynh, B.H., Forbes, W.S., and Ferreira, G.C. 1995. Characterization of the iron-binding site in mammalian ferrochelatase by kinetic and Mssbauer methods. J. Biol. Chem. 270: 2635226357.
175 Franco, R., Pereira, A.S., Tavares, P., Mangravita, A., Barber, M.J., Moura, I., and Ferreira, G.C. 2001. Substitution of murine ferrochelatase glutamate-287 with glutamine or alanine leads to por phyrin substrate-bound variants. Biochem. J. 356: 217-222. Frank, J., Jugert, F.K., Merk, H.F., Kalka, K., Goerz, G., Anderson, K., Bickers, D.R., Poh-Fitzpatrick, M.B., and Christiano, A.M. 2001. A spectrum of novel mutations in the protoporphyrinogen oxidase gene in 13 families with variegate porphyria. J. Invest. Derm. 116: 821-823. Frankenberg, N., Erskine, P.T., Cooper, J. B., Shoolingin-Jordan, P.M., Jahn, D., and Heinz, D.W. 1999. High resoluti on crystal structure of a Mg2+-dependent porphobilinogen synthase. J. Mol. Biol. 289: 591-602. Frankenberg, N., Moser, J., and Jahn, D. 2003. Bacterial heme biosynthesis and its biotechnological application. Appl. Microbiol. Biotechnol. 63: 115-127. Fraunberg, M., Nyronen, T., and Kauppinen, R. 2003. Mitochondrial targeting of normal and mutant protoporphyrinogen oxidase. J. Biol. Chem. 278: 13376-13381. Fraunberg, M., Pischik, E., Udd, L., and Ka uppinen, R. 2005. Clin ical and biochemical characteristics and genotype-phenotype correlation in 143 Finnish and Russian patients with acute intermittent porphyria. Medicine (Baltimore) 84: 35-47. Fraunberg, M., Tenhunen, R., and Kauppinen, R. 2001. Expression and characterization of six mutations in the protoporphyrinogen oxidase gene among Finnish variegate porphyria patients. Mol. Med. 7: 320-328. Fraunberg, M., Timonen, K., Mustajoki, P ., and Kauppinen, R. 2002. Clinical and biochemical characteristics and genot ype-phenotype correlation in Finnish variegate porphyria patients Eur. J. Hum. Genet. 10: 649-657. Frere, F., Schubert, W.-D., Stauffer, F., Franke nberg, N., Neier, R., Jahn, D., and Heinz, D.W. 2002. Structure of por phobilinogen synthase from Pseudomonas aeruginosa in complex with 5-fluorolevulinic Acid suggests a double Schiff-base mechanism. J. Mol. Biol. 320: 237-247. Fritsch, C., Lang, K., von Schmiedeberg, S ., Bolsen, K., Merk, H., Lehmann, P., and Ruzicka, T. 1998. Porphyria cutanea tarda. Skin Pharmacol. Appl. Skin Physiol. 11: 321-335. Fujita, H., Yamamoto, M., Yamagami, T ., Hayashi, N., and Sassa, S. 1991. Erythroleukemia differentiation. Distinc tive responses of the erythroid-specific and the nonspecific -aminolevulinate synthase mRNA. J. Biol. Chem. 266: 17494-17502.
176 Furukawa, T., Kohno, H., Tokunaga, R., and Taketani, S. 1995. Nitric oxide-mediated inactivation of mammalian ferrochelatase in vivo and in vitro : possible involvement of the iron-sul phur cluster of the enzyme. Biochem. J. 310: 533-538. Gamble, J.T., Nakatsu, K., and Marks, G. S. 2003. Comparison of the formation of N alkylprotoporphyrin IX afte r iteraction of porphyrinogeni c xenobiotics with single cDNA-expressed human P450 enzymes in microsomes prepared from baculovirus-infected in sect cells and human lymphoblastoid cell lines. Drug Metab. Dispos. 31: 202-205. Garey, J.R., Franklin, K.F., Brown, D.A., Ha rrison, L.M., Metcalf, K.M., and Kushner, J.P. 1993. Analysis of uroporphyrinogen decarboxylase complementary DNAs in sporadic porphyria cutanea tarda. Gastroenterology 105: 165-169. Garey, J.R., Labbe-Bois, R., Chelstowska, A ., Rytka, J., Harrison, L., Kushner, J., and Labbe, P. 1992. Uroporphyrinogen decarboxylase in Saccharomyces cerevisiae HEM12 gene sequence and evidence for two conserved glycines essential for enzymatic activity. Eur. J. Biochem. 205: 1011-1016. Gronimi, F., Richard, E., Lamrissi-Garcia, I., Lalanne, M., Ged, C., Redonnet-Vernhet, I., Moreau-Gaudry, F., and de Verneu il, H. 2003. Lentivirus-mediated gene transfer of uroporphyrinogen III synthase fully corrects the porphyric phenotype in human cells. J. Mol. Med. 81: 310-320. Gibbs, P.N., Chaudhry, A.G., and Shoolingin-Jordan, P.M. 1985. Purification and properties of 5-aminolaevulinate de hydratase from human erythrocytes. Biochem. J. 230: 25-34. Gibbs, P.N., and Shoolingin-Jordan, P.M. 1986. Id entification of lysine at the active site of human 5-aminolaevulinate dehydratase. Biochem. J. 236: 447-451. Gibson, K.D., Laver, W.G., and Neuberger, A. 1958. Initial stages in the biosynthesis of porphyrins. 2. The formation of delta-a minolaevulic acid from glycine and succinyl-coenzyme A by particle s from chicken erythrocytes. Biochem. J. 70: 7181. Gilles-Gonzalez, M.-A., and Gonzalez, G. 2005. Heme-based sensors: defining characteristics, recent developm ents, and regulatory hypotheses. J. Inorg. Biochem. 99: 1-22. Goldberg, A., Ashenbrucker, H., Cartwright, G.E., and Wintrobe, M. M. 1956. Studies on the biosynthesis of heme in vitro by avian erythrocytes. Blood 11: 821-833. Gollub, E.G., Liu, K.P., Dayan, J., Adle rsberg, M., and Sprinson, D.B. 1977. Yeast mutants deficient in heme biosynthesis and a heme mutant additionally blocked in cyclization of 2,3-oxidosqualene. J. Biol. Chem. 252: 2846-2854.
177 Gong, J., and Ferreira, G.C. 1995. Aminolev ulinate synthase: f unctionally important residues at a glycine loop, a putative pyridoxal phospha te cofactor-b inding site. Biochemistry 34: 1678-1685. Gong, J., Hunter, G.A., and Ferreira, G. C. 1998. Aspartate-279 in aminolevulinate synthase affects enzyme catalysis thr ough enhancing the function of the pyridoxal 5'-phosphate cofactor. Biochemistry 37: 3509 -3517. Gong, J., Kay, C.J., Barber, M.J., and Ferreira, G.C. 1996. Mutations at a glycine loop in aminolevulinate synthase affect py ridoxal phosphate cofactor binding and catalysis. Biochemistry 35: 14109-14117. Gora, M., Chacinska, A., Rytka, J., and Labbe -Bois, R. 1996a. Isolation and functional characterization of muta nt ferrochelatases in Saccharomyces cerevisiae Biochimie 78: 144-152. Gora, M., Grzybowska, E., Rytka, J., and La bbe-Bois, R. 1996b. Probing the active-site residues in Saccharomyces cerevisiae ferrochelatase by directed mutagenesis. J. Biol. Chem. 271: 11810-11816. Gouya, L., Deybach, J.C., Lamoril, J., Da S ilva, V., Beaumont, C., Grandchamp, B., and Nordmann, Y. 1996. Modulation of the phenotype in dominant erythropoietic protoporphyria by a low expr ession of the normal ferrochelatase allele. Am. J. Hum. Genet. 58: 292-299. Gouya, L., Puy, H., Lamoril, J., Da Silv a, V., Grandchamp, B ., Nordmann, Y., and Deybach, J.-C. 1999. Inheritance in er ythropoietic protoporphyria: A common wild-type ferrochelatase allelic variant w ith low expression accounts for clinical manifestation. Blood 93: 2105-2110. Gouya, L., Puy, H., Robreau, A.-M., Lyoumi, S ., Lamoril, J., Silva, V., Grandchamp, B., and Deybach, J.-C. 2004. Modulation of pe netrance by the wild -type allele in dominantly inherited erythr opoietic protoporphyria and acute hepatic porphyrias. Hum. Genet. 114: 256-262. Gouya, L., Puy, H., Robreau, A.M., Bourge ois, M., Lamoril, J., Da Silva, V., Grandchamp, B., and Deybach, J.C. 2002. The penetrance of dominant erythropoietic protoporphyria is modulated by expressi on of wild-type FECH. Nat. Genet. 30: 27-28. Grandchamp, B., De Verneuil, H., Beaum ont, C., Chretien, S., Walter, O., and Nordmann, Y. 1987. Tissue-specific expr ession of porphobilinogen deaminase. Two isoenzymes from a single gene. Eur. J. Biochem. 162: 105-110. Grandchamp, B., Phung, N., and Nordmann, Y. 1978. The mitochondrial localization of coproporphyrinogen III oxidase. Biochem. J. 176: 97-102.
178 Granick, S. 1966. The induction in vitro of the synthesis of delta-aminolevulinic acid synthetase in chemical porphyria: A res ponse to certain drugs, sex hormones and foreign chemicals. J. Biol. Chem. 241: 1359-1375. Gray, H.B., and Winkler, J.R. 2005. Long-range electron transfer. Proc. Natl. Acad. Sci. 102: 3534-3539. Grishin, N.V., Phillips, M.A., and Goldsmith, E.J. 1995. Modeling of the spatial structure of eukaryotic ornithine decarboxylases. Protein Sci. 4: 1291-1304. Gross, U., Puy, H., Kuhnel, A., Meissauer, U., Deybach, J.C., Jacob, K., Martasek, P., Nordmann, Y., and Doss, M.O. 2002a. Mo lecular, immunological, enzymatic and biochemical studies of c oproporphyrinogen oxidase deficiency in a family with hereditary coproporphyria. Cell Mol. Biol. (Noisy-le-grand) 48: 49-55. Gross, U., Puy, H., Meissauer, U., Lamoril, J., Deybach, J.C., Doss, M., Nordmann, Y., and Doss, M.O. 2002b. A molecular, enzy matic and clinical study in a family with hereditary coproporphyria. J. Inher. Metab. Disease 25: 279-286. Handschin, C., Lin, J., Rhee, J., Peyer, A. K., Chin, S., Wu, P.H., Meyer, U.A., and Spiegelman, B.M. 2005. Nutritional regula tion of hepatic heme biosynthesis and porphyria through PGC-1alpha. Cell 122: 505-515. Hankeln, T., Ebner, B., Fuchs, C., Gerlach, F ., Haberkamp, M., Laufs, T.L., Roesner, A., Schmidt, M., Weich, B., and Wystub, S. 2005. Neuroglobin and cytoglobin in search of their role in th e vertebrate globin family. J. Inorg. Biochem. 99: 110119. Hansson, M., Gustafsson, M.C., Kannangara, C.G., and Hederstedt, L. 1997. Isolated Bacillus subtilis HemY has coproporphyrinogen III to coproporphyrin III oxidase activity. Biochim. Biophys. Acta. 1340: 97-104. Hansson, M., and Hederstedt, L. 1992. Cloning and characterization of the Bacillus subtilis HemEHY gene cluster, which encode s protoheme IX biosynthetic enzymes. J. Bacteriol. 174: 8081-8093. Hansson, M., and Hederstedt, L. 1994. Puri fication and characterisation of a watersoluble ferrochelatase from Bacillus subtilis Eur. J. Biochem. 220: 201-208. Harbin, B.M., and Dailey, H.A. 1985. Orient ation of ferrochelatase in bovine liver mitochondria. Biochemistry 24: 366-370. Hardison, R.C. 1996. A brief history of hemoglob ins: Plant, animal, pr otist, and bacteria. Proc. Natl. Acad. Sci. 93: 5675-5679.
179 Harper, P., Floderus, Y., Holmstrom, P., Eggertsen, G., and Gafvels, M. 2004. Enrichment of HFE mutations in Swedish patients with familial and sporadic form of porphyria cutanea tarda. J. Intern. Med. 255: 684-688. Hart, G.J., and Battersby, A.R. 1985. Purifica tion and properties of uroporphyrinogen III synthase (co-synthetase) from Euglena gracilis Biochem. J. 232: 151-160. Hart, G.J., Leeper, F.J., and Battersby, A.R. 1987. J. Chem. Soc. Chem. Commun.: 17621765. Hart, G.J., Miller, A.D., and Battersby, A.R. 1988. Evidence that the pyrromethane cofactor of hydroxymethylbilane synthase (porphobilinogen deaminase) is bound through the sulphur atom of a cysteine residue. Biochem. J. 252: 909-912. He, Y., Alam, S.L., Proteasa, S.V., Zhang, Y., Lesuisse, E., Dancis, A., and Stemmler, T.L. 2004. Yeast frataxin solution struct ure, iron binding, and ferrochelatase interaction. Biochemistry 43: 16254 -16262. Helliwell, J.R., Nieh, Y.P., Habash, J., Faulder, P.F., Raftery, J., Cianci, M., Wulff, M., and Hadener, A. 2003. Time-resolved and st atic-ensemble structural chemistry of hydroxymethylbilane synthase. Faraday Discuss. 122: 131-144; discuss. 171-190. Hentze, M.W., Muckenthaler, M.U., and Andrew s, N.C. 2004. Balancing acts: molecular control of mammalian iron metabolism. Cell 117: 285-297. Hift, R.J., Meissner, D., and Meissner, P.N. 2004. A systematic study of the clinical and biochemical expression of variegate porphyria in a la rge South African family. Br. J. Dermatol. 151: 465. Hift, R.J., and Meissner, P.N. 2005. An anal ysis of 112 acute porphyric attacks in Cape Town, South Africa: Evidence that acute intermittent porphyria and variegate porphyria differ in susceptibility and severity. Medicine (Baltimore) 84: 48-60. Higuchi, M., and Bogorad, L. 1975. The purif ication and propertie s of uroporphyrinogen I synthases and uroporphyrinogen III cosynthase. Interactions between the enzymes. Ann. N.Y. Acad. Sci. 244: 401-418. Hornberger, U., Liebetanz, R ., Tichy, H.V., and Drews, G. 1990. Cloning and sequencing of the hemA gene of Rhodobacter capsulatus and isolation of a deltaaminolevulinic acid-dependent mutant strain. Mol. Gen. Genet. 221: 371-378. Hu, S., Smith, K.M., and Spiro, T.G. 1996. Assignment of protoheme resonance Raman spectrum by heme labeling in myoglobin. J. Am. Chem. Soc. 118: 12638 -12646. Humphrey, W., Dalke, A., and Schulten, K. 1996. VMD Visual Molecular Dynamics. J. Mol. Graphics 14: 33-38.
180 Hunter, G.A., and Ferreira, G.C. 1999a. Lysine-3 13 of 5-aminolevulinate synthase acts as a general base during formation of the quinonoid reaction intermediates. Biochemistry 38: 3711-3718. Hunter, G.A., and Ferreira, G.C. 1999b. Presteady-state reaction of 5-aminolevulinate synthase. Evidence for a rate -determining product release. J. Biol. Chem. 274: 12222-12228. Jackson, A.H., Ferramola, A.M., Sancovich, H.A., Evans, N., Matlin, S.A., Ryder, D.J., and Smith, S.G. 1976. Heptaand hexa -carboxylic porphyrinogen intermediates in haem biosynthesis. Ann. Clin. Res. 8: 64-69. Jaffe, E.K. 1995. Porphobilinogen synthase, the first source of heme's asymmetry. J. Bioenerg. Biomembr. 27: 169-179. Jaffe, E.K. 2003. An unusual phylogenetic vari ation in the metal ion binding sites of porphobilinogen synthase. Chem. Biol. 10: 25-34. Jaffe, E.K. 2004a. The porphobilinogen synt hase catalyzed reaction mechanism. Bioorganic Chemistry 32: 316-325. Jaffe, E.K. 2004b. The porphobilinogen synt hase catalyzed reaction mechanism. Bioorg. Chem. 32: 316-325. Jaffe, E.K. 2005. Morpheeins a new structur al paradigm for allosteric regulation. Trends Biochem. Sci. 30: 490-497. Jaffe, E.K., and Hanes, D. 1986. Dissection of the early steps in the porphobilinogen synthase catalyzed reaction. Requireme nts for Schiff's base formation. J. Biol. Chem. 261: 9348-9353. Jaffe, E.K., Martins, J., Li, J., Kervinen, J., and Dunbrack, R.L., Jr. 2001. The molecular mechanism of lead inhibition of human porphobilinogen synthase. J. Biol. Chem. 276: 1531-1537. Johansson, A., Nowak, G., Mller, C., Blom berg, P., and Harper, P. 2004. Adenoviralmediated expression of porphobilinogen deam inase in liver rest ores the metabolic defect in a mouse model of acute intermittent porphyria. Mol. Ther. 10: 337-343. John, R.A. 1995. Pyridoxal phosphate-dependent enzymes. Biochim. Biophys. Acta 1248: 81-96. Johnson, A., and Jones, O.T.G. 1964. Enzymatic formation of haems and other metalloporphyrins. Biochim. Biophys. Acta 93: 171-173.
181 Jones, M.S., and Jones, O.T.G. 1969. The struct ural organization of ha em synthesis in rat liver mitochondria. Biochem. J. 113: 507-514. Jones, M.S., and Jones, O. T.G. 1970. Ferrochelatase of Rhodopseudomonas spheroides Biochem. J. 119: 453-462. Jones, M.S., and Jones, O.T. 1968. Evidence for the location of ferrochelatase on the inner membrane of rat liver mitochondria. Biochem. Biophys. Res. Commun. 31: 977-982. Jones, R.M., and Jordan, P.M. 1993. Purificat ion and properties of the uroporphyrinogen decarboxylase from Rhodobacter sphaeroides Biochem. J. 293: 703-712. Jover, R., Hoffmann, F., Scheffler-Koch, V., and Lindberg, R.L.P. 2000. Limited heme synthesis in porphobilinogen deaminase-de ficient mice impairs transcriptional activation of specific cytochrome P450 genes by phenobarbital. Eur. J. Biochem. 267: 7128-7137. Juknat, A.A., Seubert, A., Seubert, S., and Ip pen, H. 1989. Studies on uroporphyrinogen decarboxylase of etiolated Euglena gracilis Z Eur. J. Biochem. 179: 423-428. Kalsbeck, W.A., Ghosh, A. Pandey, K.P., Sm ith, K.M., Bocian, D.F. 1995. Determinants of the vinyl stretching frequency in prot oporphyrins. Implications for cofactorprotein interactions in heme proteins. J. Am. Chem. Soc 117: 10959-10968. Kannangara, C.G., Gough, S.P., Bruyant, P., Hoober, J.K., Kahn, A., and von Wettstein, D. 1988. tRNAGlu as a cofactor in -aminolevulinate biosynthesis: steps that regulate chlorophyll synthesis. Trends Biochem. Sci. 13: 139-143. Kaplan, M., Hammerman, C ., and Maisels, M.J. 2003. Bilirubin genetics for the nongeneticist: Hereditary defects of neonatal bilirubin conjugation. Pediatrics 111: 886-893. Karlberg, T., Lecerof, D., Gora, M., Silvegren, G., Labbe-Bois, R., Hansson, M., and AlKaradaghi, S. 2002. Metal binding to Saccaromyces cerevisiae ferrochelatase. Biochemistry 41: 13499 -13506. Karr, S.R., and Dailey, H.A. 1988. The synthesis of murine ferrochelatase in vitro and in vivo Biochem. J. 254: 799-803. Kauppinen, R. 2005. Porphyrias. The Lancet 365: 241-252. Kawanishi, S., Seki, Y., and Sano, S. 1983. Uroporphyrinogen decarboxylase. Purification, properties, and inhibition by polychlorin ated biphenyl isomers. J. Biol. Chem. 258: 4285-4292.
182 Kervinen, J., Jaffe, E.K., Stauffer, F., Neie r, R., Wlodawer, A., and Zdanov, A. 2001. Mechanistic basis for suic ide inactivation of porphobi linogen synthase by 4,7dioxosebacic acid, an inhibitor that shows dramatic species selectivity. Biochemistry 40: 8227-8236. Kikuchi, G., and Hayashi, N. 1981. Regulatio n by heme of synthesi s and intracellular translocation of delta-aminolevu linate synthase in the liver. Mol. Cell. Biochem. 37: 27-41. Klemm, D.J., and Barton, L.L. 1987. Puri fication and properties of protoporphyrinogen oxidase from an anaerobic bacterium, Desulfovibrio gigas J. Bacteriol. 169: 5209-5215. Koch, M., Breithaupt, C., Kiefersauer, R., Fr eigang, J., Huber, R., and Messerschmidt, A. 2004. Crystal structure of protoporphyrinogen IX oxidase: a key enzyme in haem and chlorophyll biosynthesis. EMBO J. 23: 1720-1728. Kohno, H., Furukawa, T., Yoshinaga, T ., Tokunaga, R., and Taketani, S. 1993. Coproporphyrinogen oxidase. Pu rification, molecular cloning, and induction of mRNA during erythroi d differentiation. J. Biol. Chem. 268: 21359-21363. Kohno, H., Okuda, M., Furukawa, T., Tokunaga, R., and Taketani, S. 1994. Site-directed mutagenesis of human ferrochelatase: Id entification of his tidine-263 as a binding site for metal ions. Biochim. Biophys. Acta 1209: 95-100. Kruse, E., Mock, H.P., and Grimm, B. 1995. Coproporphyrinogen III oxidase from barley and tobacco sequence analysis and initial expression studies. Planta 196: 796803. Kuhnel, A., Gross, U., and Doss, M.O. 2000. Hereditary coproporphyria in Germany: clinical-biochemical studies in 53 patients. Clin. Biochem. 33: 465-473. Kundrat, L., Martins, J., Stith, L., Dunbrack, R. L., Jr., and Jaffe, E.K. 2003. A structural basis for half-of-the-sites metal binding revealed in Drosophila melanogaster porphobilinogen synthase. J. Biol. Chem. 278: 31325-31330. Kushner, J.P., Barbuto, A.J., and Lee, G. R. 1976. An inherited enzymatic defect in porphyria cutanea tarda: decreased uroporphyrinogen decarboxylase activity. J. Clin. Invest. 58: 1089-1097. Kutty, R.K., and Maines, M.D. 1981. Purifi cation and characterization of biliverdin reductase from rat liver. J. Biol. Chem. 256: 3956-3962. Labbe-Bois, R. 1990. Th e ferrochelatase from Saccharomyces cerevisiae Sequence, disruption, and expression of its structural gene Hem15 J. Biol. Chem. 265: 72787283.
183 Labbe, R.F., Hubbard, N., and Caughey, W.S. 1963. Porphyrin specificity of ferro:protoporphyrin chelatase from rat liver. Biochemistry 2: 372-374. Laemmli, U.K. 1970. Cleavage of structural prot eins during the assemb ly of the head of bacteriophage T4. Nature 227: 680-685. Lamoril, J., Martasek, P., Deybach, J.C., Da Silva, V., Grandchamp, B., and Nordmann, Y. 1995. A molecular defect in c oproporphyrinogen oxida se gene causing harderoporphyria, a variant form of hereditary coproporphyria. Hum. Mol. Genet. 4: 275-278. Lamoril, J., Puy, H., Gouya, L., Rosipal, R ., Da Silva, V., Grandchamp, B., Foint, T., Bader-Meunier, B., Dommergues, J.P., Deybach, J.C., et al. 1998. Neonatal Hemolytic Anemia Due to Inherited Hard eroporphyria: Clinical Characteristics and Molecular Basis. Blood 91: 1453-1457. Lamoril, J., Puy, H., Whatley, S.D., Martin, C., Woolf, J.R. Da Silva, V., Deybach, J.C., and Elder, G.H. 2001. Charact erization of mutations in the CPO gene in British patients demonstrates absence of genot ype-phenotype correlation and identifies relationship between hereditary coproporphyria and harderoporphyria. Am. J. Hum. Genet. 68: 1130-1138. Lander, M., Pitt, A.R., Alefounder, P.R., Bardy, D., Abell, C., and Battersby, A.R. 1991. Studies on the mechanism of hydroxymethyl bilane synthase c oncerning the role of arginine residues in substrate binding. Biochem. J. 275: 447-452. Landis, D.M., and Loeb, L.A. 1998. Random sequence mutagenesis and resistance to 5fluorouridine in human thymidylate synthases. J. Biol. Chem. 273: 25809-25817. Lange, H., Kispal, G., and Lill, R. 1999. Mechan ism of iron transport to the site of heme synthesis inside yeast mitochondria. J. Biol. Chem. 274: 18989-18996. Lange, H., Muhlenhoff, U., Denzel, M., Kispal G., and Lill, R. 2004. The heme synthesis defect of mutants impaired in mitochondr ial iron-sulfur prot ein biogenesis is caused by reversible inhibi tion of ferrochelatase. J. Biol. Chem. 279: 2910129108. Lash, T.D. 2005. The enigma of coproporphyrinogen oxidase: How does this unusual enzyme carry out oxidative decarboxylations to afford vinyl groups? Bioorg. Med. Chem. Lett. 15: 4506-4509. Lavigne, J.A., Nakatsu, K., and Marks, G. S. 2002. Identification of human hepatic cytochrome P450 sources of N -alkylprotoporphyrin IX after interaction with porphyrinogenic xenobiotics, implications for detection of xenobiotic-induced porphyria in humans. Drug Metab. Dispos. 30: 788-794.
184 Layer, G., Moser, J., Heinz, D.W., Jahn, D., and Schubert, W.-D. 2003. Crystal structure of coproporphyrinogen III oxida se reveals cofactor ge ometry of radical SAM enzymes. EMBO J. 22: 6214-6224. Layer, G., Verfurth, K., Mahlitz, E., and Jahn, D. 2002. Oxygen-independent coproporphyrinogen-III oxidase HemN from Escherichia coli J. Biol. Chem. 277: 34136-34142. Lecerof, D., Fodje, M., Hansson, A., Hansson, M., and Al-Karadaghi S. 2000. Structural and mechanistic basis of porphyrin metallation by ferrochelatase. J. Mol. Biol. 297: 221-232. Lecerof, D., Fodje, M.N., Alvarez, L.R., Olss on, U., Hansson, A., Sigfridsson, E., Ryde, U., Hansson M., and Al-Karadag hi, S. 2003. Metal binding to Bacillus subtilis ferrochelatase and interaction between metal sites. J. Biol. Inorg. Chem. 8: 452458. Lee, D.-S., Flachsova, E., Bodnarova, M., Deme ler, B., Martasek, P., and Raman, C.S. 2005. Structural basis of hereditary coproporphyria. Proc. Natl. Acad. Sci. 102: 14232-14237. Lee, J., and Anvret, M. 1991. Identificati on of the most common mutation within the porphobilinogen deaminase gene in Swedish patients with acute intermittent porphyria. Proc. Natl. Acad. Sci. 88: 10912-10915. Leeper, F.J. 1994. The evidence for a spiroc yclic intermediate in the formation of uroporphyrinogen III by cosynthase. In Ciba Found. Symp. pp. 111-123. Lermontova, I., Kruse, E., Mock, H.-P ., and Grimm, B. 1997. Cloning and characterization of a plastidal and a mitochondrial isoform of tobacco protoporphyrinogen IX oxidase. Proc. Natl. Acad. Sci. 94: 8895-8900. Lesuisse, E., Santos, R., Matzanke, B.F., Kni ght, S.A.B., Camadro, J.-M., and Dancis, A. 2003. Iron use for haeme synthesis is und er control of th e yeast frataxin homologue (Yfh1). Hum. Mol. Genet. 12: 879-889. Lewis, S.D., Lucas, B.J., Brady, S.F., Si sko, J.T., Cutrona, K.J., Sanderson, P.E., Freidinger, R.M., Mao, S.S., Gardell, S.J ., and Shafer, J.A. 1998. Characterization of the two-step pathway for inhibition of thrombin by -ketoamide transition state analogs. J. Biol. Chem. 273: 4843-4854. Li, X., and Nicholl, D. 2005. Development of PPO inhibitor-resist ant cultures and crops. Pest Manag. Sci. 61: 277-285.
185 Liang, J., Edelsbrunner, H., and Woodward, C. 1998. Anatomy of protein pockets and cavities: Measurement of binding site geometry and implications for ligand design. Protein Sci. 7: 1884-1897. Lindberg, R.L.P., Martini, R., Baumgartner, M ., Erne, B., Borg, J., Zielasek, J., Ricker, K., Steck, A., Toyka, K.V., and Meye r, U.A. 1999. Motor neuropathy in porphobilinogen deaminasedeficient mice imitates the peripheral neuropathy of human acute porphyria. J. Clin. Invest. 103: 1127-1134. Lloyd, S.G., Franco, R., Moura, J.J.G., Mour a, I., Ferreira, G.C., and Huynh, B.H. 1996. Functional necessity and physicochemical ch aracteristics of the [2Fe-2S] cluster in mammalian ferrochelatase. J. Am. Chem. Soc. 118: 9892 -9900. Louie, G.V., Brownlie, P.D., Lambert, R ., Cooper, J.B., Blundell, T.L., Wood, S.P., Malashkevich, V.N., Hdener, A., Warre n, M.J., and Shoolingin-Jordan, P.M. 1996. The three-dimensional structure of Escherichia coli porphobilinogen deaminase at 1.76resolution. Proteins Struct. Funct. Genet. 25: 48-78. Louie, G.V., Brownlie, P.D., Lambert, R ., Cooper, J.B., Blundell, T.L., Wood, S.P., Warren, M.J., Woodcock, S.C., and Jordan, P.M. 1992. Structure of porphobilinogen deaminase reveals a flexible multidomain polymerase with a single catalytic site. Nature 359: 33-39. Lu, Y., Sousa, A., Franco, R., Mangravita, A ., Ferreira, G.C., Moura, I., and Shelnutt, J.A. 2002. Binding of protoporphyrin IX and me tal derivatives to th e active site of wild-type mouse ferrochelatase at low porphyrin-to-protein ratios. Biochemistry 41: 8253-8262. Lundvall, O., and Weinfeld, A. 1968. Studies of the clinical and metabolic effects of phlebotomy treatment in porphyria cutanea tarda. Acta Med. Scand. 184: 191-199. Macieira, S., Martins, B.M., and Huber, R. 2003. Oxygen-dependent coproporphyrinogen-III oxidase from Escherichia coli : one-step purification and biochemical characterisation. FEMS Microbiol. Lett. 226: 31-37. Magnus, I., Jarrett, A., Prankerd, T., and Rimington, C. 1961. Erythropoietic protoporphyria: a new porphyria syndrom e with solar urticaria due to protoporphyrinaemia. Lancet 2: 448-451. Maines, M.D., and Gibbs, P.E.M. 2005. 30 some years of heme oxygenase: From a "molecular wrecking ball" to a "mesme rizing" trigger of cellular events. Biochem. Biophys. Res. Comm. 338: 568-577. Mallis, R.J., Brazin, K.N., Fulton, D.B ., and Andreotti, A.H. 2002. Structural characterization of a proline-driven c onformational switch within the Itk SH2 domain. Nat. Struct. Biol. 9: 900-905.
186 Manelia, M.H., Corrigalla, A.V., Klumpb, H. H., Davidsa, L.M., Kirscha, R.E., and Meissner, P.N. 2003. Kinetic and physical characterisation of recombinant wildtype and mutant human protoporphyrinogen oxidases. Biochim. Biophys. Acta 1650: 10-21. Marks, G.S., McCluskey, S.A., Mackie, J. E., Riddick, D.S., and James, C.A. 1988. Disruption of hepatic heme biosynthesis after interaction of xenobiotics with cytochrome P-450. FASEB J. 2: 2774-2783. Martasek, P. 1998. Hereditary coproporphyria. Semin. Liver Dis. 18: 25-32. Martasek, P., Camadro, J.M., Delfau-Larue M.-H., Dumas, J., Montagne, J.J., de Verneuil, H., Labbe, P., and Grandc hamp, B. 1994. Molecular cloning, sequencing, and functional expre ssion of a cDNA encoding human coproporphyrinogen oxidase. Proc. Natl. Acad. Sci. 91: 3024-3028. Martasek, P., Camadro, J.M., Raman, C.S., Le comte, M.C., Le Caer, J.P., Demeler, B., Grandchamp, B., and Labbe, P. 1997. Human coproporphyrinogen oxidase. Biochemical characterization of r ecombinant normal and R231W mutated enzymes expressed in E. coli as soluble, catalytically active homodimers. Cell Mol. Biol. (Noisy-le-grand) 43: 47-58. Martins, B.M., Grimm, B., Mock, H.-P., Hube r, R., and Messerschmidt, A. 2001. Crystal structure and substrate binding modeling of the uroporphyrinogen-III decarboxylase from Nicotiana tabacum Implications for the catalytic mechanism. J. Biol. Chem. 276: 44108-44116. Maruno, M., Furuyama, K., Akagi, R., Ho rie, Y., Meguro, K., Garbaczewski, L., Chiorazzi, N., Doss, M.O., Hassoun, A ., Mercelis, R., et al. 2001. Highly heterogeneous nature of delta-aminole vulinate dehydratase (A LAD) deficiencies in ALAD porphyria. Blood 97: 2972-2978. Marzocchi, M.P., and Smulevich, G. 2003. Relationship between heme vinyl conformation and the protein matrix in peroxidases. J. Raman Spectrosc. 34: 725 736. Mathews-Roth, M.M. 1993. Carotenoids in erythropoietic protopor phyria and other photosensitivity diseases. Ann. N.Y. Acad. Sci. 691: 127-138. Mathews-Roth, M.M. 1998. Treatment of the cutaneous porphyrias. Clin. Dermatol. 16: 295-298. Mathews-Roth, M.M., and Rosner, B. 2002. Long-term treatment of erythropoietic protoporphyria with cysteine. Photodermatol. Photoimmunol. Photomed. 18: 307309.
187 Mathews, M.A.A., Schubert, H.L., Whitby, F.G., Alexander, K.J., Schadick, K., Bergonia, H.A., Phillips, J.D., and Hill, C.P. 2001. Crystal structure of human uroporphyrinogen III synthase. EMBO J. 20: 5832-5839. Matringe, M., Camadro J.M., Labbe, P., a nd Scalla, R. 1989. Pr otoporphyrinogen oxidase as a molecular target for diphenyl ether herbicides. Biochem. J. 260: 231-235. Mauzerall, D., and Granick, S. 1958. Por phyrin biosynthesis in erythrocytes. III. Uroporphyrinogen and its decarboxylase. J. Biol. Chem. 232: 1141-1162. Medlock, A.E., and Dailey, H.A. 1996. Hu man coproporphyrinogen oxidase is not a metalloprotein. J. Biol. Chem. 271: 32507-32510. Medlock, A.E., and Dailey, H.A. 2000. Examin ation of the activity of carboxyl-terminal chimeric constructs of human and yeast ferrochelatases. Biochemistry 39: 74617467. Mehrany, K., Drage, L.A., Brandhagen, D.J ., and Pittelkow, M.R. 2004. Association of porphyria cutanea tarda with hereditary hemochromatosis. J. Am. Acad. Dermatol. 51: 205-211. Meissner, P.N., Dailey, T.A., Hift, R.J., Zi man, M., Corrigall, A.V., Roberts, A.G., Meissner, D.M., Kirsch, R.E., and Da iley, H.A. 1996. A R59W mutation in human protoporphyrinogen oxidase results in decreased enzyme activity and is prevalent in South Africans with variegate porphyria. Nat. Genet. 13: 95-97. Meissner, P.N., Day, R.S., Moore, M. R., Disler, P.B., and Harley, E. 1986. Protoporphyrinogen oxidase and porphob ilinogen deaminase in variegate porphyria. Eur. J. Clin. Invest. 16: 257-261. Melefors, O., Goossen, B., Johansson, H.E., Stripecke, R., Gray, N.K., and Hentze, M.W. 1993. Translational control of 5-ami nolevulinate synthase mRNA by ironresponsive elements in erythroid cells. J. Biol. Chem. 268: 5974-5978. Meyer, U.A., Strand, L.J., Doss, M., Rees, A.C., and Marver, H.S. 1972. Intermittent acute porphyria--demonstration of a genetic defect in porphobilinogen metabolism. N. Engl. J. Med. 286: 1277-1282. Miller, A.D., Hart, G.J., Packman, L.C., and Battersby, A.R. 1988. Evidence that the pyrromethane cofactor of hydroxymethyl bilane synthase (porphobilinogen deaminase) is bound to the protein thr ough the sulphur atom of cysteine-242. Biochem. J. 254: 915-918.
188 Miralem, T., Hu, Z., Torno, M.D., Lelli, K.M., and Maines, M.D. 2005. Small interference RNA-mediated gene silencing of human biliverdin reductase, but not that of heme oxygenase-1, attenuates arsenite-mediated induction of the oxygenase and increases apopt osis in 293A kidney cells. J. Biol. Chem. 280: 17084-17092. Mitchell, L.W., and Jaffe, E.K. 1993. Porphobilinogen synthase from Escherichia coli is a Zn(II) metalloenzyme stimulated by Mg(II). Arch. Biochem. Biophys. 300: 169177. Miyamoto, K., Nakahigashi, K., Nishimura, K., and Inokuchi, H. 1991. Isolation and characterization of visible light-sensitive mutants of Escherichia coli K12. J. Mol. Biol. 219: 393-398. Mock, H.P., Trainotti, L., Kruse, E., and Gr imm, B. 1995. Isolation, sequencing and expression of cDNA sequences enc oding uroporphyrinogen decarboxylase from tobacco and barley. Plant Mol. Biol. 28: 245-256. Morgan, R.R., Da Silva, V., Puy, H., Deyba ch, J.C., and Elder, G.H. 2002. Functional studies of mutations in the human prot oporphyrinogen oxidase ge ne in variegate porphyria. Cell Mol. Biol. (Noisy-le-grand) 48: 79-82. Morgan, R.R., Errington, R., and Elder, G.H. 2004. Identification of sequences required for the import of human protoporphy rinogen oxidase to mitochondria. Biochem. J. 377: 281-287. Morse, D., Sethi, J., and Choi, A.M. 2002. Carbon monoxide-d ependent signaling. Crit. Care Med. 30: S12-17. Munakata, H., Sun, J.Y., Yoshida, K., Naka tani, T., Honda, E., Hayakawa, S., Furuyama, K., and Hayashi, N. 2004. Role of th e heme regulatory motif in the hememediated inhibition of mitochondrial im port of 5-aminolevulinate synthase. J. Biochem. (Tokyo) 136: 233-238. Munakata, H., Yamagami, T., Nagai, T., Yamamoto, M., and Hayashi, N. 1993. Purification and structure of rat erythroid-specific -aminolevulinate synthase. J. Biochem. (Tokyo) 114: 103-111. Munir, K.M., French, D.C., Dube, D.K., a nd Loeb, L.A. 1992. Permissible amino acid substitutions within the putative nucleosid e binding site of herpes simplex virus type 1 encoded thymidine kinase establ ished by random sequence mutagenesis. J. Biol. Chem. 267: 6584-6589. Najahi-Missaoui, W., and Dailey, H.A. 2005. Production and characterization of erythropoietic protoporphyric he terodimeric ferrochelatases. Blood 106: 10981104.
189 Nakahashi, Y., Taketani, S., Okuda, M., I noue, K., and Tokunaga, R. 1990. Molecular cloning and sequence analysis of cDNA encoding human ferrochelatase. Biochem. Biophys. Res. Commun. 173: 748-755. Nakahigashi, K., Nishimura, K., Miyamoto K., and Inokuchi, H. 1991. Photosensitivity of a protoporphyrin-accumulati ng, light-sensitive mutant ( visA ) of Escherichia coli K-12. Proc. Natl. Acad. Sci. 88: 10520-10524. Nandi, D.L. 1978a. Delta-aminole vulinic acid synthase of Rhodopseudomonas spheroides Binding of pyridoxal phos phate to the enzyme. Arch. Biochem. Biophys. 188: 266-271. Nandi, D.L. 1978b. Mode of binding of pyridoxal phosphate to 5-aminolevulinate synthase. Z. Naturforsch. 33: 1003-1005. Nandi, D.L. 1978c. Studies on -aminolevulinic acid synthase of Rhodopseudomonas spheroides Reversibility of the reaction, ki netic, spectral, and other studies related to the mechanism of action. J. Biol. Chem. 253: 8872-8877. Nandi, D.L., Baker-Cohen, K.F., and Shem in, D. 1968. Delta-Aminolevulinic Acid Dehydratase of Rhodopseudomonas spheroides I. Isolation and properties. J. Biol. Chem. 243: 1224-1230. Nandi, D.L., and Shemin, D. 1977. Quaternary structure of delta -aminolevulinic acid synthase from Rhodopseudomonas spheroides J. Biol. Chem. 252: 2278-2280. Napier, I., Ponka, P., and Richardson, D.R. 2005. Iron trafficking in the mitochondrion: novel pathways revealed by disease. Blood 105: 1867-1874. Narita, S., Tanaka, R., Ito, T., Okada, K., Taketani, S., and Inokuchi, H. 1996. Molecular cloning and characterization of a cDNA that encodes protoporphyrinogen oxidase of Arabidopsis thaliana Gene 182: 169-175. Navarro, S., del Hoyo, P., Campos, Y., Abitbol, M., Moran-Jimenez, M.-J., GarciaBravo, M., Ochoa, P., Grau, M., Montagutelli, X., Frank, J., et al. 2005. Increased mitochondrial respiratory chain enzyme activ ities correlate with minor extent of liver damage in mice suffering fr om erythropoietic protoporphyria. Exp. Dermatol. 14: 26-33. Nemeria, N., Yan, Y., Zhang, Z., Brown, A.M ., Arjunan, P., Furey, W., Guest, J.R., and Jordan, F. 2001. Inhibition of the Escherichia coli pyruvate dehydrogenase complex E1 subunit and its tyrosine 177 variants by thiamin 2-thiazolone and thiamin 2-thiothiazolone diphosphates. Evidence for reversible tight-binding inhibition. J. Biol. Chem. 276: 45969-45978.
190 Neuberger, A., and Tait, G.H. 1964. Studi es on the biosynthesi s of porphyrin and bacteriochlorophyll by Rhodopseudomonas spheroides. 5. Zinc-protoporphyrin chelatase. Biochem. J. 90: 607-616. Nishimura, K., Nakayashiki, T., and Inokuc hi, H. 1993. Cloning a nd sequencing of the hemE gene encoding uroporphyrinogen III decarboxylase (UPD) from Escherichia coli K-12. Gene 133: 109-113. Nishimura, K., Nakayashiki, T., and Inokuc hi, H. 1995a. Cloning and identification of the hemG gene encoding protoporp hyrinogen oxidase (PPO) of Escherichia coli K-12. DNA Res. 2: 1-8. Nishimura, K., Taketani, S., and Inokuchi H. 1995b. Cloning of a human cDNA for protoporphyrinogen oxidase by complementation in vivo of a hemG mutant of Escherichia coli J. Biol. Chem. 270: 8076-8080. Nordmann, Y., Grandchamp, B., de Verneuil, H., Phung, L., Cartigny, B., and Fontaine, G. 1983. Harderoporphyria: a vari ant hereditary coproporphyria. J. Clin. Invest. 72: 1139-1149. Nordmann, Y., and Puy, H. 2002. Hu man hereditary hepatic porphyrias. Clinica Chimica Acta 325: 17-37. Nordmann, Y., Puy, H., Da Silva, V., Simoni n, S., Robreau, A.M., Bonaiti, C., Phung, L.N., and Deybach, J.C. 1997. Acute in termittent porphyri a: Prevalence of mutations in the porphobili nogen deaminase gene in blood donors in France. J. Intern. Med. 242: 213-217. Norris, P.G., Nunn, A.V., Hawk, J.L., and Cox, T.M. 1990. Genetic heterogeneity in erythropoietic protoporphyria: a study of the enzymatic de fect in nine affected families. J. Invest. Dermatol. 95: 260-263. Nunn, A.V., Norris, P., Hawk, J.L., and C ox, T.M. 1988. Zinc chelatase in human lymphocytes: detection of th e enzymatic defect in eryt hropoietic protoporphyria. Anal. Biochem. 174: 146-150. Ohgari, Y., Sawamoto, M., Yamamoto, M., Kohno, H., and Taketani, S. 2005. Ferrochelatase consisting of wild-type a nd mutated subunits from patients with a dominant-inherited disease, erythropoi etic protoporphyria, is an active but unstable dimer. Hum. Mol. Genet. 14: 327-334. Omata, Y., Sakamoto, H., Higashimoto, Y., Hayashi, S., and Noguchi, M. 2004. Purification and characterization of human uroporphyrinogen III synthase expressed in Escherichia coli J. Biochem. (Tokyo) 136: 211-220.
191 Osborne, M.J., Schnell, J., Benkovic, S.J., Dy son, H.J., and Wright, P.E. 2001. Backbone dynamics in dihydrofolate reductase comple xes: role of loop flexibility in the catalytic mechanism. Biochemistry 40: 9846-9859. Otterbein, L.E., and Choi, A.M.K. 2000. He me oxygenase: colors of defense against cellular stress. Am. J. Physiol. Lung Cell. Mol. Physiol. 279: L1029-1037. Park, S., Gakh, O., O'Neill, H.A., Mangravita, A., Nichol, H., Ferreira, G.C., and Isaya, G. 2003. Yeast frataxin sequentially ch aperones and stores iron by coupling protein assembly with iron oxidation. J. Biol. Chem. 278: 31340-31351. Parthasarathi, N., Hansen, C., Yamaguchi, S., Spiro, T.G. 1987. Metalloporphyrin core size resonance Raman marker bands revisite d: implications for the interpretation of hemoglobin photoproduct Raman frequencies. J. Am. Chem. Soc. 109: 38653871. Pawliuk, R., Bachelot, T., Wise, R.J., Mathews-Roth, M.M., and Leboulch, P. 1999. Long-term cure of the phot osensitivity of murine eryt hropoietic protoporphyria by preselective gene therapy. Nat. Med. 5: 768-773. Pawliuk, R., Tighe, R., Wise, R.J., Math ews-Roth, M.M., and Leboulch, P. 2005. Prevention of murine erythropoietic protoporphy ria-associated skin photosensitivity and liver disease by dermal and hepatic ferrochelatase. J. Invest. Dermatol. 124: 256-262. Pesce, A., Bolognesi, M., Bocedi, A., Ascenzi P., Dewilde, S., Moens, L., Hankeln, T., and Burmester, T. 2002. Neuroglobin and cytoglobin. Fresh blood for the vertebrate globin family. EMBO Rep. 3: 1146-1151. Phillips, J.D., Parker, T.L., Schubert, H.L., Whitby, F.G., Hill, C.P., and Kushner, J.P. 2001. Functional consequences of naturally occurring mutations in human uroporphyrinogen decarboxylase. Blood 98: 3179-3185. Phillips, J.D., Whitby, F.G., Kushner, J.P., and Hill, C.P. 1997. Characterization and crystallization of human uroporphyrinogen decarboxylase. Protein Sci. 6: 13431346. Phillips, J.D., Whitby, F.G., Kushner, J.P ., and Hill, C.P. 2003. Structural basis for tetrapyrrole coordination by ur oporphyrinogen d ecarboxylase. EMBO J. 22: 6225-6233. Phillips, J.D., Whitby, F.G., Warby, C.A., Labbe, P., Yang, C., Pflugrath, J.W., Ferrara, J.D., Robinson, H., Kushner, J.P., and H ill, C.P. 2004. Crystal structure of the oxygen-dependant coproporphyrinogen oxidase ( hem13p ) of Saccharomyces cerevisiae J. Biol. Chem. 279: 38960-38968.
192 Pischik, E., Mehtl, S., and Kauppinen, R. 2005. Nine mutations including three novel mutations among Russian patients with acute intermittent porphyria. Hum. Mut. 26: 496. Podvinec, M., Handschin, C., Looser, R., a nd Meyer, U.A. 2004. Identification of the xenosensors regulating human 5-aminolevulinate synthase. Proc. Natl. Acad. Sci. 101: 9127-9132. Pompliano, D.L., Peyman, A., and Knowles, J.R. 1990. Stabilization of a reaction intermediate as a catalytic device: definition of the functional role of the flexible loop in triosephosphate isomerase. Biochemistry 29: 3186-3194. Ponka, P. 1997. Tissue-specific regulation of iron metabolism and heme synthesis: distinct control mechanisms in erythroid cells. Blood 89: 1-25. Ponka, P. 1999. Cell biology of heme. Am. J. Med. Sci. 318: 241. Porra, R.J., and Falk, J.E. 1964. The enzymi c conversion of coproporphyrinogen III into protoporphyrin IX. Biochem J 90: 69-75. Porra, R.J., and Jones, O.T.G. 1963a. Studies on ferrochelatase. 1. Assay and properties of ferrochelatase from a pig liver mitochondrial extract. Biochem. J. 87: 181-185. Porra, R.J., and Jones, O.T.G. 1963b. Studies on ferrochelatase. 2. An investigation of the role of ferrochelatase in the biosynth esis of various haem prosthetic groups. Biochem. J. 87: 186-192. Porra, R.J., and Ross, B.D. 1965. Haem synt hase and cobalt porphyrin synthase in various micro-organisms. Biochem. J. 94: 557-562. Porra, R.J., Vitols, K.S., Labbe, R.F., and Newt on, N.A. 1967. Studies on ferrochelatase. The effects of thiols and other factors on the determination of activity. Biochem. J. 104: 321-327. Poss, K.D., and Tonegawa, S. 1997. Heme oxygenase 1 is required for mammalian iron reutilization. PNAS 94: 10919-10924. Poulos, T.L. 2005. Structural biology of heme monooxygenases. Biochem. Biophys. Res. Comm. 338: 337-345. Prasad, A.R.K., and Dailey, H.A. 1995. Eff ect of cellular location on the function of ferrochelatase. J. Biol. Chem. 270: 18198-18200. Proulx, K.L., and Dailey, H.A. 1992. Char acteristics of murine protoporphyrinogen oxidase. Protein Sci. 1: 801-809.
193 Puy, H., Deybach, J.C., Lamoril, J., R obreau, A.M., Da Silva, V., Gouya, L., Grandchamp, B., and Nordmann, Y. 1997. Molecular epidemiology and diagnosis of PBG deaminase gene defects in acute intermittent porphyria. Am. J. Hum. Genet. 60: 1373-1383. Raich, N., Romeo, P.H., Duba rt, A., Beaupain, D., Cohen-So lal, M., and Goossens, M. 1986. Molecular cloning and complete pr imary sequence of human erythrocyte porphobilinogen deaminase. Nucleic Acids Res. 14: 5955-5968. Rich, P.R. 2003. The molecular machinery of Keilin's respiratory chain. Biochem. Soc. Trans. 31: 1095-1105. Richard, E., Robert, E., Cario-Andre, M., Ged, C., Geronimi, F., Gerson, S.L., de Verneuil, H., and Moreau-Gaudry, F. 2004. Hematopoietic stem cell gene therapy of murine protoporphyria by methylguani ne-DNA-methyltransferase-mediated in vivo drug selection. Gene Ther. 11: 1638-1647. Riddle, R.D., Yamamoto, M., and Engel, J.D. 1989. Expression of -aminolevulinate synthase in avian cells: separate genes encode erythroid-specific and nonspecific isozymes. Proc. Natl. Acad. Sci. 86: 792-796. Risheg, H., Chen, F.-P., and Bloomer, J.R. 2003. Genotypic determinants of phenotype in North American patients with erythropoietic protoporphyria. Mol. Genet. Metab. 80: 196-206. Roberts, A.G., Whatley, S.D., Nicklin, M., Wo rwood, J.J., Pointon, C.S., and Elder, G.H. 1997. The frequency of hemochromatosis-associ ated alleles is increased in British patients with sporadic porphyria cutanea tarda. Hepatology 25: 159-161. Rodgers, K.R. 1999. Heme-based se nsors in biological systems. Curr. Opion.Chem. Biol. 3: 158-167. Romeo, G., and Levin, E.Y. 1969. Uropor phyrinogen III cosynt hetase in human congenital erythropoietic porphyria. Proc. Natl. Acad. Sci. 63: 856-863. Romeo, G., and Levin, E. Y. 1971. Uroporphyrinogen decarboxyl ase from mouse spleen. Biochim. Biophys. Acta 230: 330-341. Romeo, P.H., Raich, N., Dubart, A., Beaupain D., Pryor, M., Kushner, J., Cohen-Solal, M., and Goossens, M. 1986. Molecular cloning and nucleo tide sequence of a complete human uroporphyrinogen decarboxylase cDNA. J. Biol. Chem. 261: 9825-9831.
194 Rosipal, R., Lamoril, J., Puy, H., Da Silva, V., Gouya, L., De Rooij, F.W., Te Velde, K., Nordmann, Y., Martsek, P., and Deybach J.-C. 1999. Systematic analysis of coproporphyrinogen oxidase gene defect s in hereditary coproporphyria and mutation update. Hum. Mut. 13: 44-53. Rufenacht, U.B., Gouya, L., Schneider-Yin, X., Puy, H., Schafer, B.W., Aquaron, R., Nordmann, Y., Minder, E.I., and Deybach J.C. 1998. Systematic analysis of molecular defects in the ferrochelatase gene from patients with erythropoietic protoporphyria. Am. J. Hum. Genet. 62: 1341-1352. Rufenacht, U.B., Gregor, A., Gouya, L., Ta rczynska-Nosal, S., Schneider-Yin, X., and Deybach, J.-C. 2001. New missense mutation in the human ferroch elatase gene in a family with erythropoietic protoporphyria : Functional studies and correlation of genotype and phenotype. Clin. Chem. 47: 1112-1113. Sadlon, T.J., Dell, rsquo, Oso, T., Surinya K.H., and May, B.K. 1999. Regulation of erythroid 5-aminolevulinate synthase expression during erythropoiesis. Int. J. Biochem. Cell Biol. 31: 1153-1167. Sams, H., Kiripolsky, M.G., Bhat, L., and Stri cklin, G.P. 2004. Porphyria cutanea tarda, hepatitis C, alcoholism, and hemochroma tosis: a case report and review of the literature. Cutis 73: 188-190. Sano, S. 1966. 2,4-bis-beta -hydroxypropionic acid. Deut eroporphyrinogen IX, a possible intermediate between coproporphyri nogen III and protoporphyrin IX. J. Biol. Chem. 241: 5276-5283. Sano, S., and Granick, S. 1961. Mitoc hondrial coproporphyri nogen oxidase and protoporphyrin formation. J. Biol. Chem. 236: 1173-1180. Sarkany, R.P.E. 2001. The manageme nt of porphyria cutanea tarda. Clin. Exp. Dermatol. 26: 225-232. Sasarman, A., Letowski, J., Czaika, G., Ra mirez, V., Nead, M.A., Jacobs, J.M., and Morais, R. 1993. Nucleotide sequence of the hemG gene involved in the protoporphyrinogen oxid ase activity of Escherichia coli K12. Can. J. Microbiol. 39: 1155-1161. Sazanovich, I.V., Galievsky, V.A., van Hoek, A., Schaafsma, T.J., Malinovskii, V.L., Holten, D., and Chirvony, V.S. 2001. Phot ophysical and structur al properties of saddle-shaped free base porphyrins: Eviden ce for an "orthogonal" dipole moment. J. Phys. Chem. B. 105: 7818-7829.
195 Schmidt, A., Sivaraman, J., Li, Y., Larocque, R., Barbosa, J., Smith, C., Matte, A., Schrag, J.D., and Cygler, M. 2001. Thr ee-dimensional structure of 2-amino-3ketobutyrate CoA ligase from Escherichia coli complexed with a PLP-substrate intermediate: inferred reaction mechanism. Biochemistry 40: 5151-5160. Schmitt, C., Gouya, L., Malonova, E., Lamoril, J., Camadro, J.-M., Flamme, M., Rose, C., Lyoumi, S., Da Silva, V., Boileau, C., et al. 2005. Mutations in human CPO gene predict clinical expression of eith er hepatic hereditary coproporphyria or erythropoietic ha rderoporphyria. Hum. Mol. Genet. 14: 3089-3098. Schmitt, M.P. 1997. Utilization of host iron sources by Corynebacterium diphtheriae : identification of a gene whose product is homologous to eukaryotic heme oxygenases and is required for acquisition of iron from heme and hemoglobin. J. Bacteriol. 179: 838-845. Schneider-Yin, X., Bogard, C., Rfenacht, U. B., Puy, H., Nordmann, Y., Minder, E.I., and Deybach, J.-C. 2000a. Identificati on of a prevalent nonsense mutation (W283X) and two novel mutations in the porphobilinogen deaminase gene of Swiss patients with acute intermittent porphyria. Hum. Hered. 50: 247-250. Schneider-Yin, X., Gouya, L., Dorsey, M., Ru fenacht, U., Deybach, J.-C., and Ferreira, G.C. 2000b. Mutations in the iron-su lfur cluster ligands of the human ferrochelatase lead to er ythropoietic protoporphyria. Blood 96: 1545-1549. Schneider-Yin, X., Gouya, L., Meier-Weinand, A., Deybach, J.-C., and Minder, E.I. 2000c. New insights into the pathogenesis of erythropoietic protoporphyria and their impact on patient care. Eur. J. Pediatr. 159: 719-725. Schneider-Yin, X., Hergersberg, M., Goldgar, D.E., Rufenacht, U.B., Schuurmans, M.M., Puy, H., Deybach, J.C., and Minder, E. I. 2002. Ancestral founder of mutation W283X in the porphobilinogen deaminase gene among acute intermittent porphyria patients. Hum. Hered. 54: 69-81. Schoenfeld, R.A., Napoli, E., Wong, A., Zhan, S., Morin, D., Buckpitt, A.R., Taroni, F., Lonnerdal, B., Ristow, M., Pucci o, H., et al. 2005. Frataxin deficiency alters heme pathway transcripts and decreases mito chondrial heme metabolites in mammalian cells. Hum. Mol. Genet.: Epub ahead of print. Scholnick, P.L., Hammaker, L. E., and Marver, H.S. 1972. So luble delta-aminolevulinic acid synthetase of rat liver. II. Studies related to the mechanism of enzyme action and hemin inhibition. J. Biol. Chem. 247: 4132-4137. Schubert, H.L., Raux, E., Matthews, M.A., Ph illips, J.D., Wilson, K. S., Hill, C.P., and Warren, M.J. 2002. Structural diversity in metal ion chel ation and the structure of uroporphyrinogen III synthase. Biochem. Soc. Trans. 30: 595-600.
196 Schwartz, C.J., Djaman, O., Imlay, J.A., and K iley, P.J. 2000. The cysteine desulfurase, IscS has a major role in in vivo Fe-S cluster formation in Escherichia coli Proc. Natl. Acad. Sci. 97: 9009-9014. Sedlak, T.W., and Snyder, S.H. 2004. B ilirubin benefits: Cellular protection by a biliverdin reductase antioxidant cycle. Pediatrics 113: 1776-1782. Seehra, J.S., Jordan, P.M., and Akhtar, M. 1983. Anaerobic and aerobic coproporphyrinogen III oxidases of Rhodopseudomonas spheroides Mechanism and stereochemistry of vinyl group formation. Biochem. J. 209: 709-718. Segel, I.H. 1975. In Enzyme kinetics: behavior and analys is of rapid equilibrium and steady state enzyme systems pp. 564-565. John Wiley, New York. Sellers, V.M., and Dailey, H.A. 1997. Expressi on, purification, and characterization of recombinant mammalian ferrochelatase. Methods Enzymol. 281: 378-387. Sellers, V.M., Dailey, T.A., and Dailey, H.A. 1998a. Examination of ferrochelatase mutations that cause eryt hropoietic protoporphyria. Blood 91: 3980-3985. Sellers, V.M., Johnson, M.K., and Dailey, H.A. 1996. Function of the [2Fe-2S] cluster in mammalian ferrochelatase: A possible role as a nitric oxide sensor. Biochemistry 35: 2699-2704. Sellers, V.M., Wang, K.-F., Johnson, M.K., and Dailey, H.A. 1998b. Evidence that the fourth ligand to the [2Fe-2S] cluster in animal ferrochelatase is a cysteine. Characterization of the enzyme from Drosophila melanogaster J. Biol. Chem. 273: 22311-22316. Sellers, V.M., Wu, C.K., Dailey, T.A., and Dailey, H.A. 2001. Human ferrochelatase: characterization of substrate-iron bi nding and proton-abstracting residues. Biochemistry 40: 9821-9827. Shady, A.A., Colby, B.R., Cunha, L.F., Astr in, K.H., Bishop, D.F., and Desnick, R.J. 2002. Congenital erythropoietic porphyria: id entification and expr ession of eight novel mutations in the uroporphyrinogen III synthase gene. Br. J. Haematol. 117: 980-987. Shemin, D., and Kikuchi, G. 1958. Enzymatic synthesis of sigma-aminolevulinic acid. Ann. NY Acad. Sci. 75. Shemin, D., and Russell, C.S. 1953. Delta-a mminolevulinic acid, its role in the biosynthesis of porphyrins and purines. J. Am. Chem. Soc. 75: 4873-4874.
197 Shi, Z., and Ferreira, G.C. 2003. A c ontinuous anaerobic fluorimetric assay for ferrochelatase by monitoring porphyrin disappearance. Anal. Biochem. 318: 1824. Shi, Z., and Ferreira, G.C. 2004. Probing the active site loop motif of murine ferrochelatase by random mutagenesis. J. Biol. Chem. 279: 19977-19986. Shin, D.H., Roberts, A., Jancarik, J., Yokota, H., Kim, R., Wemmer, D.E., and Kim, S.H. 2003. Crystal structure of a phosphata se with a unique substrate binding domain from Thermotoga maritima Protein Sci. 12: 1464-1472. Shipovskov, S., Karlberg, T., Fodje, M., Hans son, M.D., Ferreira, G.C., Hansson, M., Reimann, C.T., and Al-Karadaghi, S. 2005. Metallation of the transition-state inhibitor N -methyl mesoporphyrin by ferrochelatase: Implications for the catalytic reaction mechanism. J. Mol. Biol. 352: 1081-1090. Shoolingin-Jordan, P.M. 1991. Biosynthesis of Tetrapyrroles Elsevier, New York. Shoolingin-Jordan, P.M., Al-Abass, A., McNeill, L.A., Sarwar, M., and Butler, D. 2003a. Human porphobilinogen deaminase mutations in the investigation of the mechanism of dipyrromethane cofactor as sembly and tetrapyrrole formation. Biochem. Soc. Trans. 31: 731-735. Shoolingin-Jordan, P.M., Al-Daihan, S., Alexeev, D., Baxter, R.L., Bottomley, S.S., Kahari, I.D., Roy, I., Sarwar, M., Sawyer, L., and Wang, S.F. 2003b. 5Aminolevulinic acid synthase: mech anism, mutations and medicine. Biochim. Biophys. Acta 1647: 361-366. Shoolingin-Jordan, P.M., Burton, G., Nordlv, H., Schneider, M.M., Pryde, L., and Scott, A.I. 1979. Pre-uroporphyrinogen: a subs trate for uroporphyrinogen III cosynthase. J. Chem. Soc. Chem. Comm.: 204-205. Shoolingin-Jordan, P.M., and Gibbs, P.N. 1985. Mechanism of action of 5aminolaevulinate dehydratase from human erythrocytes. Biochem. J. 227: 10151020. Shoolingin-Jordan, P.M., Mgbeje, B.I., Thom as, S.D., and Alwan, A.F. 1988. Nucleotide sequence for the hemD gene of Escherichia coli encoding uroporphyrinogen III synthase and initial evidence for a hem operon. Biochem. J. 249: 613-616. Shoolingin-Jordan, P.M., and Warren, M.J. 1987. Evidence for a dipyrromethane cofactor at the catalytic site of E. coli porphobilinogen deaminase. FEBS Lett. 225: 87-92.
198 Shoolingin-Jordan, P.M., and Woodcock, S.C. 1991. Mutagenesis of ar ginine residues in the catalytic cleft of Escherichia coli porphobilinogen deaminase that affects dipyrromethane cofactor assembly a nd tetrapyrrole chain initiation and elongation. Biochem. J. 280: 445-449. Sigfridsson, E., and Ryde, U. 2003. The importance of porphyrin distortions for the ferrochelatase reaction. J. Biol. Inorg. Chem. 8: 272-282. Simons, T.J.B. 1995. The affin ity of human erythrocyte porp hobilinogen synthase for Zn and Pb. Eur. J. Biochem. 234: 178-183. Smythe, E., and Williams, D.C. 1988. Rat liver uroporphyrinogen III synthase has similar properties to the enzyme from Euglena gracilis including absence of a requirement for a reversibly bound cofactor for activity. Biochem. J. 253: 275279. Solis, C., Martinez-Bermejo, A., Naidich, T. P., Kaufmann, W.E., Astrin, K.H., Bishop, D.F., and Desnick, R.J. 2004. Acute intermittent porphyria: Studies of the severe homozygous dominant disease provides insi ghts into the neur ologic attacks in acute porphyrias. Arch. Neurol. 61: 1764-1770. Soonawalla, Z.F., Orug, T., Badminton, M.N ., Elder, P.G.H., Rhodes, J.M., Bramhall, S.R., and Elias, E. 2004. Liver transplantation as a cure for acute intermittent porphyria. Lancet 363: 705-706. Spencer, P., and Shoolingin-Jordan, P.M. 1993. Purification and ch aracterization of 5aminolaevulinic acid dehydratase from Escherichia coli and a study of the reactive thiols at the metal-binding domain. Biochem. J. 290: 279-287. Spencer, P., and Shoolingin-Jordan, P.M. 1995. Characterization of the two 5aminolaevulinic acid binding sites, the Aand P-sites, of 5-am inolaevulinic acid dehydratase from Escherichia coli Biochem. J. 305: 151-158. Srivastava, G., Borthwick, I.A., Maguire, D.J ., Elferink, C.J., Bawden M.J., Mercer, J.F., and May, B.K. 1988. Regulation of 5aminolevulinate synthase mRNA in different rat tissues. J. Biol. Chem. 263: 5202-5209. Stark, W.M., Hart, G.J., and Battersby, A. R. 1986. Synthetic studies on the proposed spiro intermediate for biosynthesis of the natural porphyrins: Inhibition of cosynthetase. J. Chem. Soc. Chem. Comm. 6: 465-467. Stojeba, N., Carole, M., Jeanpi erre, C., Perrot, F., Hirth, C., Pottecher, T., and Deybach, J.-C. 2004. Recovery from a variegate po rphyria by a liver transplantation. Liver Transplant. 10: 935-938.
199 Straka, J.G., Bloomer, J.R., and Kempne r, E.S. 1991. The functional size of ferrochelatase determined in situ by radiation inactivation. J. Biol. Chem. 266: 24637-24641. Straka, J.G., and Kushner, J.P. 1983. Purifi cation and characteriza tion of bovine hepatic uroporphyrinogen decarboxylase. Biochemistry 22: 4664-4672. Susa, S., Daimon, M., Ono, H., Li, S., Yoshid a, T., and Kato, T. 2003. The long, but not the short, presequence of human copropor phyrinogen oxidase is essential for its import and sorting to mitochondria. Tohoku J. Exp. Med. 200: 39-45. Tait, G.H. 1973. Aminolae vulinate synthetase of Micrococcus denitrificans Purification and properties of the enzyme, and the eff ect of growth conditions on the enzyme activity in cells. Biochem. J. 131: 389-403. Taketani, S. 2005. Acquisition, mobilization and utilization of ce llular iron and heme: endless findings and growing evidence of tight regulation. Tohoku J. Exp. Med. 205: 297-318. Taketani, S., Kakimoto, K., Ueta, H., Masaki R., and Furukawa, T. 2003. Involvement of ABC7 in the biosynthesis of heme in er ythroid cells: interac tion of ABC7 with ferrochelatase. Blood 101: 3274-3280. Taketani, S., Nakahashi, Y., Osumi, T ., and Tokunaga, R. 1990. Molecular cloning, sequencing, and expression of mouse ferrochelatase. J. Biol. Chem. 265: 1937719380. Taketani, S., Tanaka-Yoshioka A., Masaki, R., Tashiro, Y., and Tokunaga, R. 1986. Association of ferrochelatase with Co mplex I in bovine heart mitochondria. Biochim. Biophys. Acta 883: 277-283. Taketani, S., and Tokunaga, R. 1981. Rat live r ferrochelatase. Purification, properties, and stimulation by fatty acids. J. Biol. Chem. 256: 12748-12753. Taketani, S., and Tokunaga, R. 1982. Purifica tion and substrate specificity of bovine liver-ferrochelatase. Eur. J. Biochem 127: 443-447. Tan, D., Harrison, T., Hunter, G.A., and Fe rreira, G.C. 1998. Role of arginine 439 in substrate binding of 5-am inolevulinate synthase. Biochemistry 37: 1478 -1484. Tang, X.D., Xu, R., Reynolds, M.F., Garcia, M. L., Heinemann, S.H., and Hoshi, T. 2003. Haem can bind to and inhibit mammalia n calcium-dependent Slo1 BK channels. Nature 425: 531-535.
200 Tangeras, A. 1986. Effect of decreased fe rrochelatase activity on iron and porphyrin content in mitochondria of mice with porphyria i nduced by griseofulvin. Biochim. Biophys. Acta 882: 77-85. Tenhunen, R., Marver, H.S., and Schmid, R. 1969. Microsomal heme oxygenase. Characterization of the enzyme. J. Biol. Chem. 244: 6388-6394. Thompson, J.D., Higgins, D.G., and Gibs on, T.J. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22: 4673-4680. Thunell, S., and Harper, P. 2000. Porphyrin s, porphyrin metabolis m, porphyrias. III. Diagnosis, care and monitoring in porphyr ia cutanea tarda --suggestions for a handling programme. Scand. J. Clin. Lab. Invest. 60: 561-579. Todd, D.J. 1994. Erythro poietic protoporphyria. Br. J. Dermatol. 131: 751-766. Troup, B., Hungerer, C., and Jahn, D. 199 5. Cloning and characterization of the Escherichia coli hemN gene encoding the oxygen-independent coproporphyrinogen III oxidase. J. Bacteriol. 177: 3326-3331. Tsai, S.F., Bishop, D., and Desnick, R. J. 1987. Purification and properties of uroporphyrinogen III synthase from human er ythrocytes. J. Biol. Chem. 262: 1268-1273. Tsai, S.F., Bishop, D.F., and Desnick, R. J. 1988. Human uroporphyrinogen III synthase: molecular cloning, nucleotide sequence, and expression of a full-length cDNA. Proc. Natl. Acad. Sci. 85: 7049-7053. Tschudy, D.P., Valsamis, M., and Magnussen, C.R. 1975. Acute intermittent porphyria: clinical and selected research aspects. Ann. Intern. Med. 83: 851-864. Tukey, R.H., and Strassburg, C.P. 2000. Human UDP-Glucuronosyltransferases: Metabolism, Expression, and Disease. Ann. Rev. Pharm. Toxicol. 40: 581-616. Uchida, T., Sato, E., Sato, A., Sagami, I., Shimizu, T., and Kitagawa, T. 2005. COdependent activity-controlling mechanism of heme-containing CO-sensor protein, neuronal PAS domain protein 2. J. Biol. Chem. 280: 21358-21368. Ugulava, N.B., Gibney, B.R., and Jarrett, J.T. 2000. Iron-sulfur cl uster interconversions in biotin synthase: dissociation and r eassociation of iron during conversion of [2Fe-2S] to [4Fe-4S] clusters. Biochemistry 39: 5206-5214.
201 Ugulava, N.B., Sacanell, C.J., and Jarrett, J.T. 2001. Spectroscopic changes during a single turnover of biotin synthase: dest ruction of a [2Fe-2S] cluster accompanies sulfur insertion. Biochemistry 40: 8352-8358. Urban-Grimal, D., Volland, C., Garnier, T., Dehoux, P., and Labbe-Bois, R. 1986. The nucleotide sequence of the HEM1 gene and evidence for a precursor form of the mitochondrial 5-aminolevulinate synthase in Saccharomyces cerevisiae Eur. J. Biochem. 156: 511-519. Venkateshrao, V., Yin, J., Jarzcki, A.A., Sc hultz, P.G., and Spiro, T.G. 2004. Porphyrin distortion during affinity maturation of a ferrochelatase antibody, monitored by resonance Raman spectroscopy. J. Am. Chem. Soc. 126: 16361-16367. Volland, C., and Felix, F. 1984. Isolation and properties of 5-aminolevulinate synthase from the yeast Saccharomyces cerevisiae Eur. J. Biochem. 142: 551-557. Volland, C., and Urban-Grimal, D. 1988. The presequence of yeast 5-aminolevulinate synthase is not required fo r targeting to mitochondria. J. Biol. Chem. 263: 82948299. Wang, K.-F., Dailey, T.A., and Dailey, H.A. 2001. Expression and ch aracterization of the terminal heme synthetic enzymes from the hyperthermophile Aquifex aeolicus FEMS Microbiol. Lett. 202: 115-119. Warnick, G.R., and Burnham, B.F. 1971. Regulation of porphyrin biosynthesis. Purification and characterization of delta-aminolevulinic acid synthase. J. Biol. Chem. 246: 6880-6885. Warren, M.J., Cooper, J.B., Wood, S.P., and Shoolingin-Jordan, P.M. 1998. Lead poisoning, haem synthesis and 5-am inolaevulinic acid dehydratase. Trends Biochem. Sci. 23: 217-221. Warren, M.J., and Shoolingin-Jordan, P.M. 1988. Investigation into the nature of substrate binding to the di pyrromethane cofactor of Escherichia coli porphobilinogen deaminase. Biochemistry 27: 9020-9030. Watanabe, N., Hayashi, N., and Kikuchi, G. 1983. DeltaAminolevulinate synthase isozymes in the liver and erythroid cells of chicken. Biochem. Biophys. Res. Commun. 113: 377-383. Weaver, T.M. 2000. The helix translates structure into function. Protein Sci. 9: 201206. Wegele, R., Tasler, R., Zeng, Y., Rivera, M., and Frankenberg-D inkel, N. 2004. The heme oxygenase(s)-phytochrome system of Pseudomonas aeruginosa J. Biol. Chem. 279: 45791-45802.
202 Whatley, S.D., Mason, N.G., Khan, M., Zami ri, M., Badminton, M. N., Missaoui, W.N., Dailey, T.A., Dailey, H.A., Douglas, W.S., Wainwright, N.J., et al. 2004. Autosomal recessive erythropoietic pr otoporphyria in the United Kingdom: prevalence and relationshi p to liver disease. J. Med. Genet. 41: e105-. Whatley, S.D., Puy, H., Morgan, R.R., Robreau, A.M., Roberts, A.G., Nordmann, Y., Elder, G.H., and Deybach, J.C. 1999. Variegate porphyria in Western Europe: identification of PPOX gene mutations in 104 families, extent of allelic heterogeneity, and absence of correla tion between phenotype and type of mutation. Am. J. Hum. Genet. 65: 984-994. Whitby, F.G., Phillips, J.D., Kushner, J.P., and Hill1, C.P. 1998. Crystal structure of human uroporphyrinogen decarboxylase. EMBO J. 17: 2463. Whiting, M.J., and Elliott, W.H. 1972. Pu rification and propert ies of solubilized mitochondrial delta-aminolevulinic acid synthetase and comparison with the cytosol enzyme. J. Biol. Chem. 247: 6818-6826. Whiting, M.J., and Granick, S. 1976. Delta-A minolevulinic acid synthase from chick embryo liver mitochondria. I. Puri fication and some properties. J. Biol. Chem. 251: 1340-1346. Williams, J.W., and Morrison, J.F. 1979. Th e kinetics of reversible tight-binding inhibition. Methods Enzymol. 63: 437-467. Wiman, ., Floderus, Y., and Harper, P. 2003a Novel mutations and phenotypic effect of the splice site modulator IVS3-48C in ni ne Swedish families with erythropoietic protoporphyria. J. Hum. Genet. 48: 70-76. Wiman, ., Harper, P., and Floderus, Y. 2003b. Nine novel mutations in the protoporphyrinogen oxidase gene in Swedish families with variegate porphyria. Clin. Genet. 64: 122-130. Wingert, R.A., Galloway, J.L., Barut, B., Foott, H., Fraenkel, P., Axe, J.L., Weber, G.J., Dooley, K., Davidson, A.J., Schmidt, B., et al. 2005. Deficiency of glutaredoxin 5 reveals Fe-S clusters are required for vertebrate haem synthesis. Nature 436: 1035-1039. Wittenberg, J.B., and Wittenberg, B.A. 2003. Myoglobin function reassessed. J. Exp. Biol. 206: 2011-2020. Witty, M., Jones, R.M., Robb, M.S., Jordan, P.M., and Smith, A.G. 1996. Subcellular location of the tetrapyrro le synthesis enzyme porpho bilinogen deaminase in higher plants: an immunological investigation. Planta 199: 557-564.
203 Woodcock, S.C., and Jordan, P.M. 1994. Eviden ce for participation of aspartate-84 as a catalytic group at the active site of porphobilinogen deaminase obtained by sitedirected mutagenesis of the hemC gene from Escherichia coli Biochemistry 33: 2688-2695. Wu, C.K., Dailey, H.A., Rose, J.P., Burden, A., Sellers, V.M., and Wang, B.C. 2001. The 2.0 structure of human ferrochelatas e, the terminal enzyme of heme biosynthesis. Nat. Struct. Biol. 8: 156-160. Wu, W.H., Shemin, D., Richards, K.E., and Williams, R.C. 1974. The quaternary structure of delta-aminolevulinic acid dehydratase from bovine liver. Proc. Natl. Acad. Sci. 71: 1767-1770. Wyckoff, E.E., Phillips, J.D., Sowa, A.M., Franklin, M.R., and Kushner, J.P. 1996. Mutational analysis of huma n uroporphyrinogen decarboxylase. Biochim. Biophys. Acta 1298: 294-304. Xu, K., Delling, J., and Elliott, T. 1992. The genes required for heme synthesis in Salmonella typhimurium include those encoding alte rnative functions for aerobic and anaerobic coproporphyrinogen oxidation. J. Bacteriol. 174: 3953-3963. Xu, W., Warner, C.A., and Desnick, R.J. 1995. Congenital erythr opoietic porphyria: identification and expression of 10 mutati ons in the uroporphyrinogen III synthase gene. J. Clin. Invest. 95: 905-912. Yamamoto, M., Fujita, H., Watanabe, N., Hayashi, N., and Kikuchi, G. 1986. An immunochemical study of delta-aminol evulinate synthase and deltaaminolevulinate dehydratase in li ver and erythroid cells of rat. Arch. Biochem. Biophys. 245: 76-83. Yin, J., Andryski, S.E., Beuscher IV, A. E., Stevens, R.C., and Schultz, P.G. 2003. Structural evidence for substrate strain in antibody catalysis. Proc. Natl. Acad. Sci. 100: 856-861. Yoon, T., and Cowan, J.A. 2003. Iron-Sulfur cluster biosynthesis. Characterization of frataxin as an iron donor for assembly of [2Fe-2S] clusters in ISU-type proteins. J. Am. Chem. Soc. 125: 6078-6084. Yoon, T., and Cowan, J.A. 2004. Frataxin-m ediated Iron Delivery to Ferrochelatase in the Final Step of Heme Biosynthesis. J. Biol. Chem. 279: 25943-25946. Yoshinaga, T., and Sano, S. 1980. Copr oporphyrinogen oxidase. I. Purification, properties, and activ ation by phospholipids. J. Biol. Chem. 255: 4722-4726.
204 Zagorec, M., Buhler, J.M., Treich, I., Keng, T., Guarente, L., and Labbe-Bois, R. 1988. Isolation, sequence, and re gulation by oxygen of the yeast HEM13 gene coding for coproporphyrinogen oxidase. J. Biol. Chem. 263: 9718-9724. Zhang, L., and Guarente, L. 1995. Heme binds to a short sequence that serves a regulatory function in diverse proteins. EMBO J. 14: 313-320. Zhang, Y., Lyver, E.R., Knight, S.A.B., Lesu isse, E., and Dancis, A. 2005. Frataxin and mitochondrial carrier proteins, Mrs3p and Mrs4p, cooperate in providing iron for heme synthesis. J. Biol. Chem. 280: 19794-19807. Zhou, S., Zong, Y., Ney, P.A., Nair, G., Stewart, C.F., and Sorrentino, B.P. 2005. Increased expression of the Abcg2 trans porter during erythroi d maturation plays a role in decreasing cellular protoporphyrin IX levels. Blood 105: 2571-2576. Zhu, H., Bilgin, M., Bangham, R., Hall, D., Casamayor, A., Bertone, P., Lan, N., Jansen, R., Bidlingmaier, S., Houfek, T., et al. 2001. Global analysis of protein activities using proteome chips. Science 293: 2101-2105. Zhu, W., Wilks, A., and Sto jiljkovic, I. 2000. Degradation of heme in Gram-negative bacteria: the product of the hemO gene of Neisseriae is a heme oxygenase. J. Bacteriol. 182: 6783-6790.
206 Appendix 1: The primary sequence of th e mature form of wild-type murine ferrochelatase. The porphyrin-binding loop motif is underlined. TTKPQAQPERRKPKTGILMLNMGGPETLGEVQDFLQRLFLDRDLMTLPIQNKLAPFIAKR RTPKIQEQYRRIGGGSPIKMWTSKQGEGMVKLLDELSPATAPHKYYIGFRYVHPLTEEAI EEMERDGLERAIAFTQYPQYSCSTTGSSLNAIYRYYNEVGQKPTMKWSTIDRWPTHPLLI QCFADHILKELNHFPEEKRSEVVILFSAHSLPMSVVNRGDPYPQEVGATVHKVMEKLGYP NPYRLVWQSKVGPVPWL GPQTDEAIKGLCERGRKNILLVPIAFTSDHIETLYELDIEYSQ VLAQKCGAENIRRAESLNGNPLFSKALADLVHSHIQSNKLCSTQLSLNCPLCVNPVCRKT KSFFTSQQL
About the Author Zhen Shi is a native of P. R. China. She r eceived a Bachelor of Arts degree in Biology Magna Cum Laude from Bryn Mawr College in 1992. She had a Master of Arts degree in Molecular and Cell Biology from University of California at Berkeley in 1996. Since 1998, she has been a graduate student in the Ph.D. program in the Department of Biochemistry and Molecular Biology, College of Medicine, University of South Florida, Tampa, FL.