HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chu, G. C. Articles by Ikeda-Saito, M. Search for Related Content PubMed PubMed Citation Articles by Chu, G. C. Articles by Ikeda-Saito, M. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 30, 21319-21325, July 23, 1999

Heme Degradation as Catalyzed by a Recombinant Bacterial Heme Oxygenase (Hmu O) from Corynebacterium diphtheriae^*

Grace C. Chu, Koki Katakura, Xuhong Zhang, Tadashi Yoshida, and Masao Ikeda-Saito

From the Department of Physiology and Biophysics, Case Western Reserve University School of Medicine, Cleveland, Ohio 44106-4970 and the  Department of Biochemistry, Yamagata University School of Medicine, Yamagata 990-9585, Japan

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Hmu O, a heme degradation enzyme in the pathogen Corynebacterium diphtheriae, catalyzes the oxygen-dependent conversion of hemin to biliverdin, carbon monoxide, and free iron. A bacterial expression system using a synthetic gene coding for the 215-amino acid, full-length Hmu O has been constructed. Expressed at very high levels in Escherichia coli BL21, the enzyme binds hemin stoichiometrically to form a hexacoordinate high spin hemin-Hmu O complex. When ascorbic acid is used as the electron donor, Hmu O converts hemin to biliverdin with -hydroxyhemin and verdoheme as intermediates. The overall conversion rate to biliverdin is approximately 4-fold slower than that by rat heme oxygenase (HO) isoform 1. Reaction of the hemin-Hmu O complex with hydrogen peroxide yields a verdoheme species, the recovery of which is much less compared with rat HO-1. Reaction of the hemin complex with meta-chloroperbenzoic acid generates a ferryl oxo species. Thus, the catalytic intermediate species and the nature of the active form in the first oxygenation step of Hmu O appear to be similar to those of the mammalian HO. However, the considerably slow catalytic rate and low level of verdoheme recovery in the hydrogen peroxide reaction suggest that the active-site structure of Hmu O is different from that of its mammalian counterpart.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Heme oxygenase (HO),¹ first characterized in eukaryotes, is an enzyme that catalyzes the regiospecific oxidative degradation of iron protoporphyrin IX to biliverdin IX, iron, and CO in the presence of NADPH and NADPH-cytochrome P-450 reductase, which serves as an electron donor (1-5). Not a hemoprotein per se, HO binds heme at a 1:1 ratio and utilizes it as both a substrate and a prosthetic group (2-6).

Among the well characterized isoforms of the mammalian HO are HO-1, which is 33 kDa, inducible, and highly expressed in the spleen and liver, and HO-2, which is 36 kDa, constitutive, and found primarily in the brain and testes (2). Both isoforms function similarly in their catalytic cycle (5). HO first binds hemin stoichiometrically to form a hemin-HO complex. Then, an electron donated from NADPH-cytochrome P-450 reductase reduces the hemin iron to the ferrous state (7). This allows dioxygen to bind to the ferrous iron, forming a metastable oxy complex (8). Additional electron donation to this oxy complex initiates the conversion of the heme to biliverdin IX with -meso-hydroxyheme and verdoheme as intermediates (Scheme 1) (3-5). The CO released during HO catalysis has been reported to act as a messenger molecule, participating in neuronal transmission and vascular regulation through the activation of soluble guanylyl cyclase (9-11). The product biliverdin is subsequently converted to bilirubin by biliverdin reductase (2, 12-13). The significant outcome of this is the removal of excess heme via the excretion of bilirubin and the recycling of iron (14).

View larger version (26K): [in this window] [in a new window]   Scheme I.   Intermediates in the conversion of heme to biliverdin as catalyzed by the mammalian heme oxygenase.

Iron is required for the survival of most bacteria and is particularly essential for pathogens to cause diseases (15-17). To circumvent the low concentration of free extracellular iron, pathogenic bacteria have developed sophisticated systems to acquire iron from iron-containing proteins found in their hosts (17-20). One mechanism is via a bacterial heme degradation enzyme (18, 21). In contrast to the mammalian HO, whose primary purpose is to maintain iron homeostasis, that of the bacterial HO is to release iron from heme so that the iron may be utilized. Within the last few years, through genetic studies, the presence of heme-degrading enzymes has been identified in pathogenic bacteria (17, 21-22).

Hmu O, the first and only prokaryotic HO isolated to date, is expressed by the hmu O gene found in Corynebacterium diphtheriae, the causative agent of diphtheria. In comparison to the mammalian HO, Hmu O is not membrane-bound but soluble and has a smaller molecular mass of 24 kDa. It is 33% identical in sequence to the first 221 amino acids of human HO-1 (22). This enzyme has been proposed to be utilized by C. diphtheriae to release iron from the heme supplied by the pathogen's host. Using an Escherichia coli expression system, Wilks and Schmitt (23) have purified a 24-kDa Hmu O protein that stoichiometrically binds hemin and converts it to biliverdin IX and CO upon the aerobic addition of electron donors. However, the oxygen activation mechanisms and the catalytic pathway of Hmu O are not clear. Whether or not they are similar to the mammalian system remains to be established.

In this study, we have constructed a highly efficient bacterial expression system of Hmu O using a synthetic gene based on the hmu O gene sequence (22). We have examined the heme degradation reaction catalyzed by Hmu O using the hemin-, -hydroxyhemin-, and verdoheme-complexes of the purified recombinant Hmu O. Reactions of the hemin-Hmu O complex with H₂O₂ and mCPBA have also been studied. We have found that the nature of the active form in the first oxygenation step and the catalytic intermediates of Hmu O are similar to those of the mammalian HO. However, the distal heme pocket structure of Hmu O appears to be different from that of its mammalian counterpart.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Construction of the Bacterial Heme Oxygenase Hmu O Expression Plasmid, pMWHmuO-- Plasmid purification, subcloning, and bacterial transformations were performed according to Sambrook et al. (24). A T7 promoter-based expression vector, pMW172, was a gift from Dr. K. Nagai (MRC Laboratory of Molecular Biology, Cambridge, UK). A 48-base pair double-stranded synthetic oligonucleotide with unique restriction enzyme sites for XhoI, ClaI, and AvrII was incorporated between the NdeI and HindIII sites in the multiple cloning site of pMW172 to make pMWA. Eight oligonucleotides and their complements, 82-91 nucleotides in length, were synthesized to prepare a 648-base pair synthetic gene coding for the entire Hmu O from the initiator ATG to the TAA stop codon. Each oligonucleotide was phosphorylated with a T₄ polynucleotide kinase, then annealed with its complement to make a double-stranded DNA, e.g. Oligo I to Oligo VIII. Oligos I and II were designed so that the 5' end of Oligo I could be ligated to the NdeI site, whereas its 3' end was complementary to the 5' end of Oligo II. The 3' end of Oligo II could be ligated to the XhoI site. Similarly, the 5' ends of Oligos III, V, and VII were designed to ligate to the XhoI, ClaI, and AvrII sites, respectively, and their 3' ends had sequences complementary to the 5' ends of Oligos IV, VI, and VIII, respectively. The 3' ends of Oligos IV, VI, and VIII had sequences for ligation to the ClaI, AvrII, and HindIII sites. Given these Oligos I to VIII, the construction of pMWHmuO was as followed.

pMWA was excised with NdeI and XhoI, and the resulting larger fragment of the digests was gel-purified. Using T₄ ligase, both Oligos I and II were ligated simultaneously between the NdeI and XhoI sites of the NdeI/XhoI fragment. The resulting plasmid, pMWB, was cleaved with XhoI and ClaI, and the larger fragment was gel-purified. Both Oligos III and IV were ligated between the XhoI and ClaI sites of the larger XhoI/ClaI fragment, and pMWC was obtained. Excised by ClaI and AvrII, the larger ClaI/AvrII fragment from pMWC was gel-purified and ligated with Oligos V and VI simultaneously. The ligation of these fragments yielded pMWD, which then was cut with AvrII and HindIII. Oligos VII and VIII were ligated to the larger AvrII/HindIII fragment from pMWD. Ligation of Oligos VII and VIII resulted in the formation of pMWHmuO, the expression plasmid for the recombinant bacterial heme oxygenase Hmu O of C. diphtheriae. The nucleotide sequence was determined by an Applied Biosystems 373A DNA sequencer. Fig. 1 is a comparison of the coding sequence of the synthetic gene to that of the cloned DNA (22).

View larger version (55K): [in this window] [in a new window]   Fig. 1.   Nucleotide and amino acid sequences of the synthesized pMWHmuO as compared with hmu O. The nucleotide sequence of the cloned hmu O is shown below the amino acid sequence of the synthetic Hmu O enzyme. The arabic numerals above the sequence indicate the numbering of the nucleotide and amino acid sequences.

Protein Expression and Purification-- A 10-ml inoculum in LB-Amp media of A_600 nm 1.0 was prepared with a fresh colony of E. coli BL21 (DE3) cells transformed with the expression construct pMWHmuO. 500-ml cultures were inoculated with 1 ml of the prepared inocula and grown in LB-Amp media at 37 °C until A_{600 nm} reached 1.3-1.4. The cells were grown for an additional 16 h at 25 °C before being harvested by centrifugation and stored at 80 °C until use. Typically, 2 g of cells were obtained from a 500-ml culture.

Thawed on ice, the cell pellets were resuspended in 100 mM Tris buffer, pH 8, containing 100 mM NaCl and 1 mM EDTA. Lysozyme (2 mg/g cells) and phenylmethysulfonyl fluoride of final concentration 1 mM were added to the resuspension, which was stirred continuously at 4 °C for 2 h. The cells then were sonicated (Branson 450 Sonifier) until no longer viscous and centrifuged at 39,000 × g for 1 h. The resulting supernatant was recovered for further purification.

Solid ammonium sulfate was added to the supernatant to a concentration of 60% saturation, and the solution was stirred for 30 min at 4 °C. After centrifugation at 14,000 × g for 30 min, the supernatant then was increased to 85% ammonium sulfate saturation and centrifuged. The subsequent precipitates, which contained the Hmu O protein, were dissolved in 20 mM phosphate buffer, pH 7. The dissolved solution was gel-filtered through a Sephadex G75 column, which had been equilibrated with 20 mM phosphate buffer, pH 7. Fractions with A_{400 nm} greater than 0.35 were pooled and loaded onto a column of DEAE-cellulose equilibrated with 20 mM phosphate buffer, pH 7. Eluted in 20 mM phosphate buffer, pH 7, with a linear gradient of 0 to 0.4 M KCl, fractions with A_400 nm/A_{280 nm} values greater than 0.3 were pooled and concentrated by ultrafiltration. The enzyme was stored at 77 K until use.

Reconstitution of Hmu O with Hemin-- 250 µM hemin in 5-µl increments was added to 10 µM of Hmu O in 800 µl of 0.1 M phosphate buffer, pH 7, at 20 °C. After each addition of hemin, the absorption spectrum was recorded. By plotting the absorbance at 404 nm against the amount of hemin added, titration curves were constructed, and the amount of hemin required to form the hemin-Hmu O complex was determined. To ensure that all Hmu O were hemin-bound, hemin of concentration twice that of Hmu O was added to the purified protein solution. After incubation at 4 °C, the excess hemin was removed by a DEAE-cellulose column. Fractions with A_{404 nm}/A_{280 nm} greater than or equal to 3 were collected, concentrated by ultrafiltration, and stored at 77 K until use.

Formation of the -meso-Hydroxyhemin- and Verdoheme-Hmu O Complexes-- Verdoheme (protoverdoheme-IX) was synthesized as described by Saito and Itano (25). -Hydroxyhemin (-meso-hydroxyprotohemin) was made by hydrolyzing -meso-benzoyloxyprotohemin as described previously (26). The ferric -hydroxyheme- and ferrous verdoheme-Hmu O complexes were prepared as described by Mansfield Matera et al. (26) and Fujii (27), respectively.

Reaction of the Hemin-Hmu O Complex with Ascorbic Acid-- Heme degradation catalyzed by Hmu O was studied using ascorbic acid as the electron donor (12). Ascorbic acid of final concentration 17.5 mM was added to an optical cuvette containing 10 µM hemin-Hmu O in 2.3 ml of 0.1 M phosphate buffer, pH 7, at 20 °C. Spectral changes between 300-800 nm were recorded until the reaction was complete, as indicated by the maximum loss of the Soret band (A_404 nm) and the formation of biliverdin.

For experiments performed with CO, the hemin-Hmu O complex was prepared as described above but injected into a sealed cuvette filled with phosphate buffer that had been presaturated with 50% CO and 50% O₂. Ascorbic acid of final concentration 17.5 mM was added to initiate heme degradation. The formation of the verdoheme-CO complex was monitored spectrophotometrically. When there was no further change in the spectrum, pyridine (20% final concentration) was added to the mixture (28), and the formation of the pyridine-verdoheme complex was recorded.

In other experiments, after the reaction had been arrested at the verdoheme-CO stage, the verdoheme that had accumulated in the presence of CO was allowed to continue to the fully oxidized product by displacing the CO with 100% O₂.

Reaction of the Heme-Hmu O Complex with H₂O₂ and mCPBA-- Preparation for the assays of the hemin-Hmu O complex with H₂O₂ and mCPBA was similar to that described previously (29). Reagent grade 30% H₂O₂ was diluted in 0.1 M phosphate buffer, pH 7, to make working solutions, the concentrations of which were determined spectroscopically using _{240 nm} = 43.6 M¹·cm¹ (30). The appropriate amount of H₂O₂ was added to separate solutions containing the hemin-Hmu O complex in 0.1 M phosphate buffer, pH 7, to yield final H₂O₂:hemin-Hmu O ratios of 1:1, 2:1, 3:1, and 5:1. mCPBA:hemin-Hmu O of these final concentration ratios were also prepared. Reactions at these different concentration ratios were monitored spectroscopically at 20 °C.

Analytical Methods-- Protein expression and purity were analyzed by SDS-PAGE on 10-20% gradient gels. The isoelectric point of the purified Hmu O was determined by a Pharmacia FastSystem following the manufacturer's instruction. Optical absorption spectra were recorded on a Hewlett-Packard 8453 spectrophotometer at 20 °C in 0.1 M phosphate buffer, pH 7.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Expression and Purification of Hmu O-- Sligar and co-workers (31) have pioneered the use of synthetic genes for bacterial expression of hemoproteins. The advent of efficient and inexpensive synthesis has made it feasible to obtain longer (80-90-mer) oligonucleotides than those (~20 mer) used earlier (31). In our approach here, using longer oligonucleotides has enabled us to ligate the eight double-stranded oligonucleotides (Oligos I to VIII) to the expression vector in minimal ligation steps. The ligation products generated during the plasmid preparation were conveniently purified by standard agarose gel electrophoresis, and the final expression plasmid was readily obtained. Furthermore, the synthetic gene has allowed for the incorporation of unique restriction enzyme sites convenient for cassette mutagenesis.

Using pMWHmuO, we have expressed successfully a recombinant Hmu O protein, as depicted in the SDS-PAGE (Fig. 2, lane 2). The initial expression of Hmu O by culturing the transformed BL21 cells at 37 °C resulted in an accumulation of the expressed protein mostly in inclusion bodies. Solubilization of the inclusion bodies, following the methods used for mammalian HO mutants (32), successfully yielded an active Hmu O. However, our current method of culturing the E. coli at 37 °C then 25 °C, instead of 37 °C continuously, has increased significantly the yield of the protein in the soluble form. Both the purified Hmu O in the soluble form and from inclusion bodies have the same enzymatic properties and optical absorption spectra. Our expression of Hmu O mostly in the soluble form has facilitated the purification process and yields 100-120 mg of purified Hmu O/liter of bacterial culture, 5-fold greater than that reported by Wilks and Schmitt (23). As indicated by SDS-PAGE, the purified Hmu O we obtain is homogeneous and has a single band at 24 kDa (Fig. 2, lane 3), the size as expected from the deduced Hmu O amino acid sequence (24.1 kDa) (22). The pI value estimated by isoelectricphoresis is 5.4. 

View larger version (47K): [in this window] [in a new window]   Fig. 2.   SDS-PAGE analysis of the recombinant Hmu O protein. Lane 1, molecular mass markers; lane 2, Hmu O expressed in E. coli; lane 3, purified Hmu O.

We have observed that the transformed E. coli cultured at 37 °C are pale yellow, whereas those cultured at 37 °C then 25 °C are green. The former is because of the expressed protein being in mostly inclusion bodies, whereas the latter is in the soluble form. The green E. coli cells indicate that biliverdin has been formed from the hemin degradation catalyzed by the expressed Hmu O and a reductase system in the E. coli, a phenomenon observed also for the E. coli expression of mammalian HO isoforms (29, 32-33). During the purification of Hmu O, even after DEAE-cellulose column chromatography, the fractions with the heme oxygenase enzyme activity have a broad Soret band between 380 and 400 nm, indicating that biliverdin is bound to Hmu O. In the case of rat HO-1, most of the biliverdin is removed by the DEAE-cellulose. Thus, the biliverdin affinity of Hmu O appears to be higher than that of rat HO-1. However, despite biliverdin being attached to the enzyme, hemin still binds to Hmu O readily, as described below. This indicates that biliverdin bound to Hmu O can be easily displaced by hemin, contrary to that claimed by Wilks and Schmitt (23).

Properties of the Hemin-Hmu O Complex-- Fig. 3 shows the optical absorption spectra of the purified recombinant Hmu O (broken line) and its stoichiometric complex with hemin (solid line). The optical absorption spectrum of the hemin-Hmu O complex at pH 7 has a Soret maxima at 404 nm and peaks at 500 and 630 nm in the visible region (Fig. 3, solid line). The absorption spectrum is similar to those of the mammalian hemin-HO complexes and indicates that the hemin-Hmu O complex is ferric hexacoordinate high spin at neutral pH, which is the case for the mammalian HO isoforms (5, 33-34). EPR and resonance Raman spectra of the hemin-Hmu O complex have confirmed this.²

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Optical absorption spectra of Hmu O (broken line) and its hemin complex (solid line). The protein concentration was 10 µM. Inset, titration of Hmu O (10 µM) with hemin as monitored by the absorbance increase at 404 nm (closed circles). The absorbance increase because of the addition of hemin in the absence of Hmu O is indicated by open circles.

Because the Soret region of the optical absorption spectrum of the hemin-Hmu O complex is different from that of free hemin in pH 7 buffer, spectrophotometric titration of Hmu O with hemin was carried out utilizing this difference. Illustrated in the inset of Fig. 3, the titration curve of Hmu O with hemin has a well defined inflection point. From this, the molar stoichiometry of their binding has been established to be 1:1, which is in agreement with that reported by Wilks and Schmitt (23). By the pyridine hemochrome method (35), the extinction coefficient at 404 nm for the hemin-Hmu O complex is determined to be 150 mM¹·cm¹, larger than the 121 mM¹·cm¹ value reported by Wilks and Schmitt (23). Our value of 150 mM¹·cm¹ is within the range of Soret extinction coefficients for hexacoordinate high spin protohemin complexes including the hemin complex of rat HO-1 (165 mM¹·cm¹) (34).

Degradation of the Hemin Bound to Hmu O with Ascorbic Acid-- Ascorbic acid can serve as the electron donor in the oxidative degradation of hemin by mammalian HO (12). In this study, ascorbic acid is shown to support the Hmu O-catalyzed conversion of hemin to biliverdin. As depicted in Fig. 4, the addition of ascorbic acid to the hemin-Hmu O complex initiates heme degradation. In a period of 2 h, the Soret peak at 404 nm disappears, and broad absorption bands centered near 380 and 680 nm appear, indicating that hemin is converted to biliverdin (Fig. 4, spectrum b). Independently, by high performance liquid chromatography analysis, we have confirmed biliverdin-IX formation. We have noticed that heme degradation by Hmu O is 4-fold slower than that by rat HO-1 when ascorbic acid is used as an electron donor.

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Reaction of the hemin-Hmu O complex with ascorbic acid. The spectra were recorded immediately before the addition of ascorbic acid, 1, 3, 6, 10, 16, 22, 33 min and 2 h after the addition of ascorbic acid. Before the addition of ascorbic acid, the Soret absorbance at 404 nm is maximum. After the addition of ascorbic acid, the Soret decreases with time, and the absorbance at 680 nm increases. The shoulder near 395 nm in the 2 h spectrum is because of residual verdoheme. The visible region is expanded for the spectrum before (spectrum a) and 2 h after (spectrum b) the addition of ascorbic acid.

In previous HO-1 studies, Yoshida and Kikuchi (12) observed that when ascorbic acid was used as the electron donor, the heme degradation product was not biliverdin but an Fe³⁺-biliverdin complex. This differs from the Hmu O system, in which biliverdin, not the Fe³⁺-biliverdin complex, is the final product of the ascorbic acid-supported heme degradation. The implication of this is that iron is more readily extruded in the Hmu O system in contrast to the mammalian HO. This appears to be consistent with the physiological role of Hmu O, which is to release iron from heme for iron utilization by the bacteria (23).

Formation of the Verdoheme-CO Complex Using Ascorbic acid-- Heme catabolism catalyzed by mammalian HO yields biliverdin with verdoheme as a precursor (28). When the hemin-HO complex is incubated with reducing equivalents under an atomosphere of O₂ and CO, heme degradation can be arrested by CO at the verdoheme stage, resulting in an accumulation of the ferrous verdoheme-CO complex and thus inhibiting biliverdin formation (28-29). In our current study of Hmu O catalysis, we have conducted the reaction of the hemin-Hmu O complex with ascorbic acid in an atmosphere of approximately 50% CO and 50% O₂. The product formed exhibits an absorption spectrum with distinct peaks at 351, 404, 538, and 636 nm, as indicated in Fig. 5A, spectrum a. The spectrum is similar to that of the CO-bound ferrous verdoheme-rat HO-1 complex (36). With the addition of 20% pyridine, a new spectrum with peaks at 398, 417, 537, and 679 nm appears (data not shown). This spectrum resembles that of the pyridine complex of verdoheme-HO-1 (28).

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Formation of the CO bound verdoheme-Hmu O species with CO. Panel A (top), reaction of the hemin-Hmu O complex with ascorbic acid in the presence of CO. Under an atmosphere of 50% CO and 50% O2, Hmu O catalysis forms a species with spectrum a. Exposure of this species to 100% O2 yields a final product having spectrum b. Spectrum c is of the biliverdin complex formed in Hmu O catalysis in the presence of air only. Panel B (bottom), optical absorption spectra of the CO (spectrum a) and ferrous deoxy (spectrum b) forms of the verdoheme IX-Hmu O complex, prepared from chemically synthesized verdoheme IX and Hmu O.

Prolonged incubation of the bacterial verdoheme-CO complex in the 50% CO and 50% O₂ environment does not change its spectrum. When the CO and O₂ mixture is replaced with 100% O₂, the optical absorption spectrum of the reaction product (Fig. 5A, spectrum b) becomes similar to that obtained by the ascorbic acid reaction of the hemin-Hmu O complex in the absence of CO (Fig. 5A, spectrum c).³ The likeness of the two spectra indicates that the originally CO-bound verdoheme-Hmu O complex has converted to biliverdin and that verdoheme is a precursor to biliverdin in Hmu O catalysis.

The formation of the verdoheme intermediate during the Hmu O-catalyzed heme degradation reaction is further corroborated by the optical absorption spectrum of Hmu O complexed with a chemically synthesized verdoheme. The spectrum of the CO-bound ferrous verdoheme IX-Hmu O, shown in Fig. 5B, spectrum a, has peaks at 352, 405, 541, and 634 nm, similar to those aforementioned of the verdoheme-CO intermediate formed during Hmu O catalysis (Fig. 5A, spectrum a). The removal of the bound CO by evacuation results in the ferrous deoxy verdoheme IX-Hmu O complex, whose optical absorption spectrum (Fig. 5B, spectrum b) has a Soret peak at 400 nm and bands at 532 and 685 nm. The spectrum is similar to those of the mammalian HO verdoheme complexes (28, 36). Subsequent exposure of this ferrous verdoheme IX-Hmu O complex to O₂ and ascorbic acid leads to its conversion to biliverdin (data not shown). These results unequivocally demonstrate verdoheme as the precursor to biliverdin.

Reaction of the -meso-Hydroxyhemin-Hmu O Complex with O₂-- In HO catalysis, -meso-hydroxyheme is the precursor to verdoheme (3-5, 26). To determine whether or not it is an intermediate as well as a precursor to verdoheme in Hmu O-mediated heme catabolism, we have prepared -meso-hydroxyhemin-Hmu O using chemically synthesized -meso-hydroxyhemin and studied its reactivity with O₂. As depicted in Fig. 6, spectrum a, the optical absorption spectrum of the -meso-hydroxyhemin-Hmu O complex exhibits a broad Soret peak at 404 nm with a relatively featureless visible region, similar to that of the -meso-hydroxyhemin-HO complex (26). Upon addition of dithionite, the heme iron is reduced, thus forming a ferrous -hydroxyheme-Hmu O complex (Fig. 6, spectrum b) with a Soret maxima at 430 nm. The binding of exogenous CO to this complex yields the CO form of the ferrous -hydroxyheme-Hmu O complex, whose spectrum has a distinct Soret peak at 418 nm (Fig. 6, spectrum c). The optical absorption spectra of the ferric, ferrous, and ferrous CO forms of the -meso-hydroxyheme-Hmu O complex are similar to those of the respective forms of the -meso-hydroxyheme complex of HO-1 (26). Based on these results, we infer that the coordination structure and electronic states of the -hydroxyheme-Hmu O complexes are pentacoordinate high spin for the ferric and ferrous forms and hexacoordinate low spin for the ferrous CO form, concurring with those established for the -hydroxyheme complexes of HO-1 (26).

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Optical absorption spectra of the -hydroxyheme-Hmu O complex and its reaction products. The initially ferric complex (spectrum a) was reduced anaerobically to the ferrous form (spectrum b) by dithionite and then coordinated with CO, forming the ferrous-CO derivative (spectrum c). Reaction of the ferrous CO -hydroxyheme-Hmu O with O2 resulted in spectrum d.

Reaction of the ferrous CO-bound -hydroxyheme-Hmu O with O₂ causes an immediate shift of the Soret band to 404 nm, and a defined peak at 634 nm emerges (Fig. 6, spectrum d). This is similar to HO-1 (26). The optical absorption spectrum of the newly formed species is identical to that we have observed for CO-bound verdoheme-Hmu O.⁴ Prolonged exposure of the aforementioned species to O₂ in the presence of ascorbic acid produces an optical spectrum similar to that of a biliverdin complex (data not shown). The results described above undeniably demonstrate that -hydroxyheme is an intermediate of the Hmu O-mediated heme degradation, specifically as the precursor to verdoheme.

Reactions of the Hemin-Hmu O Complex with mCPBA and H₂O₂-- In a previous study, to identify the active intermediate in the first oxygenation step of HO catalysis, which is the conversion of hemin to -hydroxyhemin, the hemin-HO-1 complex was reacted with mCPBA (29). This resulted in the formation of the ferryl oxo species (Fe⁴⁺=O). In our current Hmu O study, reaction of the hemin-Hmu O complex with mCPBA causes a decrease and a shift in the Soret peak from 404 to 418 nm and the concomitant appearance of absorption bands at 527 and 549 nm (Fig. 7). The spectrum is characteristic of the ferryl oxo-heme species (29, 33). No further heme degradation occurred hereafter. Hence, ferryl oxo is not an active intermediate in Hmu O catalysis.

View larger version (15K): [in this window] [in a new window]   Fig. 7.   Reaction of the hemin-Hmu O complex with mCPBA. a, spectrum of the hemin-Hmu O complex; b, spectrum after the addition of 3 eq of mCPBA.

In the case of the mammalian HO, a ferric hydroperoxide (Fe³⁺-OOH) species has been proposed to be an active intermediate in the first oxygenation step (29). Unlike mCPBA, H₂O₂ hydroxylates the -meso position of hemin to form -meso-hydroxyhemin, which then is converted to verdohemin in the presence of O₂ (29). Here we have reacted the hemin-Hmu O complex with H₂O₂ and found that a verdohemin complex, as indicated by the absorption band at 688 nm (Fig. 8A), is formed. This verdohemin formation was further corroborated by the optical absorption spectrum of its pyridine complex (data not shown). Based on these results, a ferric hydroperoxide species must be an active intermediate in the first oxygenation step of Hmu O catalysis as it is in mammalian HO.

View larger version (28K): [in this window] [in a new window]   Fig. 8.   Reactions of the hemin-Hmu O and hemin-rat HO-1 complexes with H2O2. Panel A (top), spectra of the hemin-Hmu O complex () after the addition of 1 eq (- - - - - -), 2 eq (- · - · -), 3 eq (     ), and 5 eq (- · · - · ·) of H2O2. Panel B (bottom), spectra of the hemin-rat HO-1 complex () after the addition of 1 eq (- - - - - -), 2 eq (- · - · -), 3 eq (     ), and 5 eq (- · · - · ·) of H2O2.

We have also found that the extent of conversion to verdohemin is dependent on the H₂O₂ concentration, with 5 equivalents of H₂O₂ inducing the most change, and 1 equivalent, the least. In comparison to the H₂O₂ reactions with rat HO-1 (Fig. 8B), those with Hmu O are less efficient and do not result in the virtual loss of the Soret band. Even with 1 equivalent of H₂O₂, verdoheme formation by rat HO-1 is much greater than that by Hmu O.

In the H₂O₂ reaction with the hemin-HO complex, deprotonation of the peroxide by a distal group in the heme pocket facilitates its binding to the hemin iron (38). If the deprotonation is deficient, formation of the active intermediate ferric hydroperoxide is hindered, and consequently, verdoheme recovery is reduced. Based on this, one possible reason for the decreased verdohemin recovery in Hmu O is that the active-site structure, namely the nature of the distal group, of Hmu O differs from that of rat HO-1.

Structural differences in the distal heme pocket between Hmu O and rat HO-1 might also explain our observation of the slow ascorbic acid-supported heme degradation by Hmu O. In the first segment of the proposed Hmu O catalytic pathway, ascorbic acid reduces the hemin iron to the ferrous state, thus permitting O₂ to bind to the iron to form a metastable ferrous oxy Hmu O complex. The bound O₂ in HO-1 forms a hydrogen bond with a distal group (39). This hydrogen bond facilitates the formation of the ferric hydroperoxide active species by decreasing the reduction potential of the oxy form. The structure of the distal heme pocket of Hmu O might perturb the hydrogen bond interaction between the bound O₂ and a distal group so that the reduction potential of the ferrous oxy Hmu O complex is altered, thereby reducing the rate of ferric hydroperoxide formation. Consequently, the turnover of Hmu O is retarded when ascorbic acid is used as the source of the reducing equivalents.

Conclusion-- Utilizing a full-length synthetic hmu O gene, we have expressed in high yield a recombinant bacterial heme oxygenase that is catalytically identical to the cloned Hmu O from C. diphtheriae. We have found similarities between Hmu O and the mammalian HO in terms of enzymatic properties. As with the mammalian HO, under aerobic conditions and in the presence of an electron donor, Hmu O catalyzes heme degradation and yields biliverdin as the final product. Verdoheme is demonstrated to be the immediate precursor to the biliverdin complex in both the degradation of the hemin-Hmu O complex and the oxidation of a verdoheme IX-Hmu O complex prepared from a chemically synthesized verdoheme and Hmu O. When the -hydroxyheme-Hmu O complex we have prepared is oxidized in the presence of a reductant, verdoheme and subsequently a biliverdin complex, are formed. In addition, we have found that a ferric hydroperoxide species is an active intermediate of the first oxygenation step in Hmu O catalysis. These results and that of a previous study, which showed CO being concomitantly released during Hmu O catalysis (23), suggest that the Hmu O catalytic pathway is similar to that of mammalian HO.

However, despite these similarities, there are differences between Hmu O and mammalian HO. Among them are the higher affinity of Hmu O for biliverdin, the formation of biliverdin instead of an Fe³⁺-biliverdin complex in the reaction with ascorbic acid, and the different rate of heme catabolism by Hmu O as compared with rat HO-1. We propose that the distal pocket hydrogen bonding in Hmu O is different from that in mammalian HO. This might be one primary reason for the slower overall Hmu O catalytic rate in the ascorbic acid-supported heme degradation and the low verdoheme formation in the H₂O₂ reaction. Further investigation of the Hmu O active site structure and the mechanisms by which heme is catabolized in Hmu O catalysis is needed to explain these differences.

	ACKNOWLEDGEMENTS

We thank Drs. H. Fujii and C. T. Migita for the preparation of -meso-hydroxyhemin and verdoheme and Dr. S. Park for pI measurements.

	FOOTNOTES

^* This work was supported by National Institutes of Health Research Grants GM 57272 and Ministry of Education, Science, Sports, and Culture Grants-in-aid for Scientific Research 09480158, 10129101, and 10044233, Japan.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EMBL Data Bank with accession number(s) AB019621.

² G. C. Chu, T. Tomita, F. Sonnichsen, T. Yoshida, and M. Ikeda-Saito, submitted for publication.

³ These results demonstrate that exogenous CO can potentially inhibit the conversion of verdoheme to biliverdin by forming the CO complex of verdoheme-Hmu O. However, CO released during Hmu O catalysis (23) does not inhibit this conversion in an air-saturated buffer. One explanation for this is that verdoheme has extremely low reactivity toward CO because of its inherent ferric character (36-37). Hence, under physiological conditions, in which the concentration of CO is much less than that of O₂, the product CO does not inhibit the conversion of verdoheme to biliverdin.

⁴ CO does not inhibit the conversion of -hydroxyheme to verdoheme in Hmu O catalysis. As reported for HO-1, this conversion appears to proceed without the dissociation of CO from the -hydroxyheme iron. Direct oxygen attack of the porphyrin ring may be an alternative mechanism (26).

	ABBREVIATIONS

The abbreviations used are: HO, heme oxygenase; mCPBA, meta-chloroperbenzoic acid; PAGE, polyacrylamide gel electrophoresis; eq, equivalent(s).

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. D. L. Suits, N. Jaffer, and Z. Jia Structure of the Escherichia coli O157:H7 Heme Oxygenase ChuS in Complex with Heme and Enzymatic Inactivation by Mutation of the Heme Coordinating Residue His-193 J. Biol. Chem., December 1, 2006; 281(48): 36776 - 36782. [Abstract] [Full Text] [PDF]

				K. A. Ridley, J. D. Rock, Y. Li, and J. M. Ketley Heme Utilization in Campylobacter jejuni J. Bacteriol., November 15, 2006; 188(22): 7862 - 7875. [Abstract] [Full Text] [PDF]

				T. Ueno, N. Yokoi, M. Unno, T. Matsui, Y. Tokita, M. Yamada, M. Ikeda-Saito, H. Nakajima, and Y. Watanabe Design of metal cofactors activated by a protein-protein electron transfer system PNAS, June 20, 2006; 103(25): 9416 - 9421. [Abstract] [Full Text] [PDF]

				M. D. L. Suits, G. P. Pal, K. Nakatsu, A. Matte, M. Cygler, and Z. Jia Identification of an Escherichia coli O157:H7 heme oxygenase with tandem functional repeats PNAS, November 22, 2005; 102(47): 16955 - 16960. [Abstract] [Full Text] [PDF]

				J. Wang, L. Lad, T. L. Poulos, and P. R. O. de Montellano Regiospecificity Determinants of Human Heme Oxygenase: DIFFERENTIAL NADPH- AND ASCORBATE-DEPENDENT HEME CLEAVAGE BY THE R183E MUTANT J. Biol. Chem., January 28, 2005; 280(4): 2797 - 2806. [Abstract] [Full Text] [PDF]

				T. Matsui, M. Furukawa, M. Unno, T. Tomita, and M. Ikeda-Saito Roles of Distal Asp in Heme Oxygenase from Corynebacterium diphtheriae, HmuO: A WATER-DRIVEN OXYGEN ACTIVATION MECHANISM J. Biol. Chem., January 28, 2005; 280(4): 2981 - 2989. [Abstract] [Full Text] [PDF]

				M. Unno, T. Matsui, G. C. Chu, M. Couture, T. Yoshida, D. L. Rousseau, John. S. Olson, and M. Ikeda-Saito Crystal Structure of the Dioxygen-bound Heme Oxygenase from Corynebacterium diphtheriae: IMPLICATIONS FOR HEME OXYGENASE FUNCTION J. Biol. Chem., May 14, 2004; 279(20): 21055 - 21061. [Abstract] [Full Text] [PDF]

				S. Hirotsu, G. C. Chu, M. Unno, D.-S. Lee, T. Yoshida, S.-Y. Park, Y. Shiro, and M. Ikeda-Saito The Crystal Structures of the Ferric and Ferrous Forms of the Heme Complex of HmuO, a Heme Oxygenase of Corynebacterium diphtheriae J. Biol. Chem., March 19, 2004; 279(12): 11937 - 11947. [Abstract] [Full Text] [PDF]

				E. P. Skaar, A. H. Gaspar, and O. Schneewind IsdG and IsdI, Heme-degrading Enzymes in the Cytoplasm of Staphylococcus aureus J. Biol. Chem., January 2, 2004; 279(1): 436 - 443. [Abstract] [Full Text] [PDF]

				O. Protchenko and C. C. Philpott Regulation of Intracellular Heme Levels by HMX1, a Homologue of Heme Oxygenase, in Saccharomyces cerevisiae J. Biol. Chem., September 19, 2003; 278(38): 36582 - 36587. [Abstract] [Full Text] [PDF]

<