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

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted 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 Request Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Kostanjevecki, V. Articles by Van Beeumen, J. Search for Related Content PubMed PubMed Citation Articles by Kostanjevecki, V. Articles by Van Beeumen, J. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 50, 35614-35620, December 10, 1999

Structure and Characterization of Ectothiorhodospira vacuolata Cytochrome b₅₅₈, a Prokaryotic Homologue of Cytochrome b₅^*

Vesna Kostanjeveki, David Leys^§, Gonzalez Van Driessche, Terrance E. Meyer^¶, Michael A. Cusanovich^¶, Ulrich Fischer, Yves Guisez, and Jozef Van Beeumen^**

From the  Department of Biochemistry, Physiology, and Microbiology, Laboratory of Protein Biochemistry and Protein Engineering, University of Gent, B-9000 Gent, Belgium, the ^¶ Department of Biochemistry, University of Arizona, Tucson, Arizona 85721, and the  Department of Marine Microbiology, University of Bremen, D-28359 Bremen, Germany

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

A soluble cytochrome b₅₅₈ from the purple phototropic bacterium Ectothiorhodospira vacuolata was completely sequenced by a combination of automated Edman degradation and mass spectrometry. The protein, with a measured mass of 10,094.7 Da, contains 90 residues and binds a single protoheme. Unexpectedly, the sequence shows homology to eukaryotic cytochromes b₅. As no prokaryotic homologue had been reported so far, we developed a protocol for the expression, purification, and crystallization of recombinant cytochrome b₅₅₈. The structure was solved by molecular replacement to a resolution of 1.65 Å. It shows that cytochrome b₅₅₈ is indeed the first bacterial cytochrome b₅ to be characterized and differs from its eukaryotic counterparts by the presence of a disulfide bridge and a four-residue insertion in front of the sixth ligand (histidine). Eukaryotes contain a variety of b₅ homologues, including soluble and membrane-bound multifunctional proteins as well as multidomain enzymes such as sulfite oxidase, fatty-acid desaturase, nitrate reductase, and lactate dehydrogenase. A search of the Mycobacterium tuberculosis genome showed that a previously unidentified gene encodes a fatty-acid desaturase with an N-terminal b₅ domain. Thus, it may provide another example of a bacterial b₅ homologue.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Soluble cytochromes from purple phototropic bacteria are almost invariably found to be of the c-type with covalently bound heme (1, 2). One of the few previously recognized b-type soluble cytochromes in purple bacteria is a minor component and has a large molecular mass (3). It is now known to be a bacterioferritin and to be widespread in bacteria as a whole (4). The best characterized soluble cytochrome b is the b₅₆₂ from Escherichia coli, with a molecular mass of 12 kDa and a redox potential of +200 mV (5-7). The three-dimensional structure is an antiparallel four-helix bundle, similar to that of cytochrome c' (8).

Kusche and Trüper (9) found a small, soluble cytochrome b₅₅₈ in Ectothiorhodospira shaposhnikovii, in addition to cytochromes c₅₅₄ and c'. We found very similar proteins in the related species, Ectothiorhodospira vacuolata. High potential iron protein isozymes were also purified from these two species (10), and the sequences were found to be very similar (11).¹ In fact, they are so similar that the two bacteria may be considered to be strains of the same species. The cytochrome b₅₅₈ from E. shaposhnikovii was reported to have a mass of 15.8 kDa and a redox potential of 210 mV (9). The immediate goal of this study was to identify the cytochrome and to determine its primary and tertiary structures as a contribution to the determination of its functional role.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Protein Isolation and Modification-- Native cytochrome b₅₅₈ from E. vacuolata was prepared essentially as described by Kusche and Trüper (9). The amino acid sequence was determined with 1.9 mg of protein. Prior to sequencing, the cysteines of the protein were reduced with dithiothreitol and pyridylethylated as described by Friedman et al. (12). Heme and reagents were removed from the denatured protein by gel filtration on a Sephadex G-25F column (22 × 1.5 cm; Amersham Pharmacia Biotech, Uppsala, Sweden) equilibrated in and eluted with 5% formic acid.

Protein Sequence Analysis-- Digestion with Lys-C endoproteinase (Wako, Osaka, Japan) was performed on 18.6 nmol of pyridylethylated apoprotein in 100 mM ammonium bicarbonate buffer, pH 8.0, at an enzyme/substrate ratio of 1:30 for 2 h at 37 °C. The same conditions were used for Glu-C endoproteinase (Roche Molecular Biochemicals) digestion, but the digestion mixture was incubated for 4 h at 37 °C.

Peptides obtained after enzymatic digestions were separated by reversed-phase high performance liquid chromatography using a Pep-S C2/C18 column (250 × 4 mm; Amersham Pharmacia Biotech). The chromatographic equipment consisted of two Model 6000A chromatographic pumps (Waters, Milford, MA), an injector with a loop of 100-µl volume (Rheodyne, Inc., Cotati, CA), and a UV detector set at 220 nm (Pye Unicam, Cambridge, MA). The solvents used were 0.1% trifluoroacetic acid in water (solvent A) and 0.1% trifluoroacetic acid in acetonitrile (solvent B) at a flow rate of 1 ml/min. Automated N-terminal sequence and electrospray ionization mass analyses were performed as described (13).

PCR² Strategy-- The gene for the E. vacuolata cytochrome b₅₅₈ was obtained via PCR amplification (Vent polymerase, New England Biolabs Inc., Beverly, MA) using degenerate primers based on the amino acid sequence of the protein. The sequence for the amino-terminal primer (based on the first six residues of the protein sequence) was 5'-AAYGARACNGARGCNACN-3', and that of the carboxyl-terminal primer (based on sequence 84-90) was 5'-HTANCARAGRAGITCNCANGG-3'. The PCR fragment was cloned in the vector pGEM-T (Promega, Madison, WI), and the resulting construct (pGEMT-b₅₅₈) was verified by sequence analysis.

Construction of the Overexpression Plasmid-- To achieve secretion of cytochrome b₅₅₈ into the periplasmic space of E. coli, the gene was fused with the nucleotide sequence encoding the signal peptide (sOmpA) of the bacterial outer membrane protein OmpA. Thus, the b₅₅₈ gene was inserted after the sOmpA leader sequence using the NaeI and HindIII restriction sites of plasmid pT10sOmpArPDI (14-16). As verified by DNA sequence analysis, sOmpA was fused in-frame to the mature b₅₅₈. The hybrid gene encoding the fusion sOmpA/b₅₅₈ was transferred from pT10sOmpAb₅₅₈ to pLPPsOmpArPDI using the restriction sites XbaI and HindIII and also to pQE60 (QIAGEN Inc.) as a NcoI/HindIII fragment. The b₅₅₈ gene was also cloned without signal sequence in the pQE60 plasmid. The resulting constructs (pLPPsOmpAb₅₅₈, pQE60sOmpAb₅₅₈, and pQE60b₅₅₈) were verified by nucleotide sequencing.

Production of Recombinant Cytochrome b₅₅₈ in E. coli-- The bacterial strain MC1061 (17) harboring the expression plasmid pLPPsOmpAb₅₅₈ was grown in LB medium supplemented with carbenicillin (100 µg/ml). After 16 h of incubation, cells were harvested by centrifugation (4000 rpm, 10 min, 4 °C), and the periplasmic fraction was collected after treatment as described by Koshland and Botstein (18). After SDS-polyacrylamide gel electrophoresis and electroblotting, the N-terminal amino acid sequence was determined to check for the correct removal of the signal peptide from the secreted protein.

Protein Purification-- Purification of recombinant cytochrome b₅₅₈ was achieved by a two-step procedure utilizing ion-exchange and size-exclusion chromatographies. The periplasmic protein fraction was first loaded on a Q-Sepharose Fast Flow column (1-ml bed volume; Amersham Pharmacia Biotech) in 50 mM Tris-Cl, pH 7.4. The cytochrome b₅₅₈ fraction was completely retained on the column and was eluted with 0.35 M NaCl and 50 mM Tris-Cl, pH 7.4. The pooled protein was further purified on an HL-Superdex 75 column (1.5 × 60 cm) connected to an Äkta chromatographic system (Amersham Pharmacia Biotech). The cytochrome b₅₅₈-containing fractions were pooled and concentrated using Vivaspin filters (molecular mass cutoff of 5 kDa). Mass spectroscopy and N-terminal amino acid sequence analysis were used to judge the purity of the preparation.

Characterization of Recombinant Cytochrome b₅₅₈ from E. coli-- Protein samples were subjected to reducing SDS-polyacrylamide gel electrophoresis (19) and stained with Coomassie Blue or silver (20). Native electrophoresis was used to detect the presence of the heme group by peroxidase reaction (21). Total protein concentration was determined by the method of Bradford (22) using the Bio-Rad 500-0006 kit with bovine serum albumin as a standard curve.

Crystallization and X-ray Data Collection-- The protein solution was concentrated to 10 mg/ml and dialyzed against 10 mM Tris-Cl, pH 7.4, to set up crystallization trials using the hanging drop diffusion method. The protein crystallized in 4 M NaCl, pH 5.6, and 0.1 M MES. Crystals grew over a period of several days to a maximum size of 0.2 × 0.05 × 0.05 mm. The crystals belong to space group P3₂21 with unit cell parameters a and b = 46.406 Å and c = 91.66 Å. Assuming a molecular mass of 10,094.7 Da, as determined by electrospray ionization mass spectrometry, and with one molecule in the asymmetric unit, the V_m value is 3.42, corresponding to a solvent content of 63.8%. These values are within the observed range for protein crystals (23). Data collection was performed at the LURE Synchrotron (Orsay, France) on beam-line DW32 ( = 0.97 Å) using a Mar-Research Mar345 imaging plate detector. The x-ray diffraction data were indexed, processed, scaled, and merged using DENZO and SCALEPACK (24). A total of 92,730 measurements from 14,370 unique reflections were recorded, and R_merge was 5.9% for the data between 15 and 1.65 Å with a completeness of 99.6%.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Protein Sequence Determination-- The complete amino acid sequence of cytochrome b₅₅₈ (Fig. 1, sequence 1) was obtained by N-terminal sequence analysis of the native protein up to position 16 and of overlapping peptides obtained from Glu-C and Lys-C enzymatic digestions on pyridylethylated apoprotein. Peptides obtained from these two digestions covered the complete sequence. Fragment S12B contained an intramolecular disulfide bridge, which was found to connect Cys²⁴ and Cys⁵³ of the complete protein sequence. Mass data on the different peptides confirming the sequence data are summarized in Table I. The measured mass of native E. vacuolata cytochrome b₅₅₈ is 10,094.7 Da, which is in agreement with the calculated mass of 10,095.3 Da (Fig. 2).

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Structure-based alignment of the amino acid sequences of cytochrome b558 with several eukaryotic cytochromes b5. Sequence 1, b558 (complete); sequence 2, bovine microsomal b5 (residues 1-87); sequence 3, Nicotiana tabacum b5 (residues 1-87); sequence 4, S. cerevisiae flavocytochrome-b2:lactate dehydrogenase (residues 1-82); sequence 5, chicken sulfite oxidase (residues 1-85); sequence 6, M. tuberculosis (residues 1-91). Boldface letters indicate conserved residues and those determining the cytochrome b5 fold. It is remarkable that sulfite oxidase has a Gly-Glu-Pro-His tetrapeptide after the sixth ligand, whereas the same tetrapeptide sequence occurs four residues earlier in E. vacuolata b558 and provides the sixth ligand in that protein. The three-dimensional structures clearly show that the tetrapeptides do not align.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Electrospray ionization mass spectrum of native cytochrome b558. The number at the top of each peak represents the number of positive charges for the particular m/z peak. The inset is the original spectrum with indication of the m/z values for the series z = 7-10.

A BLAST search of GenBank^TM indicated that cytochrome b₅₅₈ is related to cytochrome b₅, yeast flavocytochrome b₂, plant nitrate reductases, animal sulfite oxidases, and the fatty-acid desaturases of plants and animals. We developed the strategy to determine the crystal structure of cytochrome b₅₅₈ to identify it more conclusively. Because we had insufficient native protein for crystallization trials, we prepared recombinant protein for these studies.

Expression of Recombinant Cytochrome b₅₅₈ in E. coli-- Information obtained from back-translation of the amino acid sequence was used to design degenerate oligonucleotide primers for PCR against the E. vacuolata genomic DNA template. A PCR fragment of the appropriate size and sequence was obtained, showing that it was the correct product derived from the b₅₅₈ gene. The periplasmic protein fractions from the pQE60sOmpAb₅₅₈, pT10sOmpAb₅₅₈, and pLPPsOmpAb₅₅₈ constructs were prepared from E. coli cells, grown aerobically at 37 °C until saturation (A₆₀₀ = 2.3). Recombinant cytochrome b₅₅₈ was expressed from plasmid pLPPsOmpAb₅₅₈ only when isopropyl--D-thiogalactopyranoside was omitted. When isopropyl--D-thiogalactopyranoside was added, expression was considerably suppressed. The other plasmids did not produce any detectable amounts of cytochrome. Therefore, pLPPsOmpAb₅₅₈ was used for expression of cytochrome b_558. SDS-polyacrylamide gel electrophoresis analysis (data not shown) showed cytochrome b₅₅₈ to be the major component in the periplasmic protein fraction. N-terminal amino acid sequence analysis revealed the expected sequence of mature cytochrome b₅₅₈, confirming the correct processing of the signal peptide by an E. coli signal peptidase.

To ascertain that recombinant cytochrome b₅₅₈ did incorporate a heme cofactor, the tetramethylbenzidine-hydrogen peroxide test was performed. On native polyacrylamide gels, a unique band was indeed visualized due to the peroxidase activity of the cytochrome. The dithionite-reduced and air-oxidized difference absorption spectra indicated that the oxidized Soret peak maximum was at 414 nm; and upon reduction, the -peak was at 557.5 nm, and the Soret peak was shifted to 424.5 nm and increased in magnitude (Fig. 3), confirming that it had native redox properties.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Dithionite-reduced and air-oxidized difference absorption spectra of recombinant cytochrome b558. The spectra of 4.6 µg of protein were measured in 10 mM sodium phosphate buffer, pH 7.0.

Purification of Cytochrome b₅₅₈ from E. coli Periplasm-- Cytochrome b₅₅₈ was successfully purified from the osmotic shock fluid by ion-exchange chromatography and gel filtration. Analysis using isoelectric focusing gels showed that cytochrome b₅₅₈ was retained at a pH of 4.5, which is in accordance with the calculated pI value of 4.75 based upon the sequence (data not shown). On a Q-Sepharose Fast Flow column, cytochrome b₅₅₈ bound very efficiently at pH 7.4. The eluted protein was judged to be 90% pure. To remove minor contaminating proteins, the cytochrome b₅₅₈ fraction was loaded on an HL-Superdex 75 column in 100 mM Tris-Cl, pH 7.4, and 100 mM NaCl. Analysis on silver-stained SDS-polyacrylamide gels and mass spectrometry showed the protein to be fully pure.

Three-dimensional Structure Analysis-- The crystal structure of cytochrome b₅₅₈ was solved by molecular replacement with bovine cytochrome b₅ (Protein Data Bank code 1CYO) as the search model. Molecular replacement calculations were performed using AmoRe (25). Refinement was carried out using the maximum likelihood-based program REFMAC (26). Ordered water molecules were included by selecting the peaks based on F_o  F_c difference Fourier maps contoured at 3.0  and the 2F_o  F_c density contoured at 1  (Fig. 4). Residues and side chains not visible in the electron density were omitted from the model. For data collection and refinement statistics, see Table II.

View larger version (62K): [in this window] [in a new window]   Fig. 4.   Detail of the weighted 2Fo  Fc electron density map surrounding the heme group and both heme ligands with the final model superposed.

The polypeptide chain of cytochrome b₅₅₈ is folded in five short -helices and five -strands involved in a twisted sheet structure (Fig. 5). This is the same overall three-dimensional structure as for eukaryotic cytochrome b₅ (31). The first five N-terminal and the last four C-terminal residues are not visible in the electron density. Most -helices are very short, having incomplete -helical geometry (31). In Fig. 1, we show the structure-based alignment of cytochrome b₅₅₈ with cytochrome b₅ sequences for which the three-dimensional structures are already known. All of them are soluble or membrane-bound redox proteins, although in some, the cytochrome b is a domain within a larger enzyme structure (such as, for example, sulfite oxidase and flavocytochrome-b₂:lactate dehydrogenase).

View larger version (45K): [in this window] [in a new window]   Fig. 5.   Ribbon fold representation of cytochrome b558. -Helices are red, whereas -strands are blue. Loops are yellow. This figure was generated using Molscript (29) and Raster3D (30).

Structure of Cytochrome b₅₅₈ and Comparison with Homologous Proteins-- When comparing cytochrome b₅₅₈ with the cytochrome b₅ domain of Saccharomyces cerevisiae flavocytochrome b₂ (32) and with bovine microsomal cytochrome b₅ (33), it is apparent that cytochrome b₅₅₈ resembles the latter more closely, having the same orientation for both -helix 5 and the heme group. The C- traces of b₅₅₈ and bovine microsomal b₅ are very similar (Fig. 6A), the only major differences being situated in the loops connecting the -helices involved in heme binding. A close comparison of the two molecules clearly indicates a different orientation of the heme-binding region relative to the twisted sheet structure in b₅₅₈. This rather small movement is mediated by changes in backbone torsion angles around Gln⁵⁵, Leu³⁵, and Tyr⁸¹.

View larger version (38K): [in this window] [in a new window]   Fig. 6.   Stereo representation of the C- trace of E. vacuolata cytochrome b558 (blue), with the heme group and the heme ligand, with bovine microsomal cytochrome b5 (Protein Data Bank code 1CYO) (A), yeast flavocytochrome b2 (Protein Data Bank code 1FCB) (B), and chicken sulfite oxidase (Protein Data Bank code 1SOX) (C) (all in red). Residues involved in reorienting the backbone conformation compared with bovine b5 have been labeled for b558. The disulfide bridge is green.

A structural characteristic novel to the b₅ fold in cytochrome b₅₅₈ is the disulfide bridge between two strands formed by Cys²⁴ and Cys⁵³. Although a single cysteine is observed at position 24 (b₅₅₈ numbering) in plant cytochromes b₅ (34) and yeast flavocytochrome b₂, most animal cytochromes b₅ lack cysteine altogether. The disulfide bridge appears to span a hydrophobic cleft on the surface of the molecule. This cleft is situated in between two hydrophobic sites of the protein as described by Mathews and co-workers (31). It functions as a sort of a channel making the heme more accessible to the solvent. Recent molecular dynamics simulations (35) as well as NMR and fluorescence studies (36) provide evidence of localized motion at the cleft area. The same authors (36, 37) studied a "de novo" incorporated disulfide bond in the b₅ protein, near the same position as in native cytochrome b₅₅₈. Their result suggests that the engineered disulfide bond inhibits the dynamics of the cleft formation. In the case of cytochrome b₅₅₈, the influence of the disulfide bridge on the native dynamics remains to be investigated, as does its influence on the interaction with redox partners.

A comparison of the structures of cytochrome b₅₅₈, bovine cytochrome b₅, yeast flavocytochrome b₂ (Fig. 6B), and chicken sulfite oxidase (38, 39) (Fig. 6C), a molybdopterin cytochrome b₅, clearly shows the locations of insertions and deletions. Cytochrome b₅₅₈ has a four-residue insertion one position ahead of the sixth heme ligand (His⁷⁰). This is illustrated in the sequence alignment of Fig. 1. It is curious that several of the insertions and deletions occur at the molecular surface around the heme where electron transfer is thought to take place. They may thus play a role in molecular recognition of reaction partners.

The heme group in cytochrome b₅₅₈ is buried in a hydrophobic cleft flanked by four short -helices (-helices 2-5) at the sides and by the twisted -sheet at the bottom (Fig. 5). In comparison with bovine b₅, the heme group is in the so-called B-conformation, rotated 180° around the a,y-meso-carbon atoms relative to the observed A-conformation in bovine b₅ (Fig. 6A). In b₅₅₈, the A-orientation would be less stable due to a close contact with Leu⁷⁸, a bulky residue that replaces the smaller Ser⁷⁸ of bovine b₅. His⁴² and His⁷⁰, extending from the side of the cavity, ligate the iron atom. Whereas in bovine cytochrome b₅, one propionic acid group is curved toward the interior of the molecule and the other projects outward in solution, cytochrome b₅₅₈ has both heme propionates bent back toward the interior of the molecule. Both propionates form salt bridges with Lys⁶⁴ (Fig. 7), a structural property not normally encountered in cytochromes b₅, but frequently observed in cytochromes c. Whereas propionate B is hydrogen-bonded to Tyr⁶⁶, propionate A is hydrogen-bonded to the hydroxyl group and the peptide nitrogen of Ser⁷¹, the latter contact being preserved in bovine b₅. Both Lys⁶⁴ and Tyr⁶⁶ are close to or part of the four-residue insertion in the b₅₅₈ sequence when compared with the b₅ family.

View larger version (34K): [in this window] [in a new window]   Fig. 7.   Stereo representation of the heme pocket. The heme group and selected residues involved in heme binding are represented by atom colored sticks, whereas the backbone is represented by ribbons. Color coding is the same as described in the legend to Fig. 5. This figure was generated using Molscript (29) and Raster3D (30).

The loop connecting -helices 2 and 3 at the other side of the heme plane is formed by the strictly conserved Pro⁴³-Gly⁴⁴-Gly⁴⁵ sequence in the cytochrome b₅ family. This sequence is positioned at a strategic point where chain reversal occurs, leaving no space to accommodate a side chain at positions 44 and 45. In cytochrome b₅₅₈, however, the second glycine is replaced by another proline residue, reorienting the local backbone conformation relative to bovine b₅. This relocation results in a change in the hydrogen bond partner for the second imidazole nitrogen of His⁴², which is normally the carbonyl oxygen of Gly⁴⁵, to a water molecule. The other heme ligand (His⁷⁰) is hydrogen-bonded to the carbonyl oxygen of Trp⁶¹, similar to the interaction with Phe⁶¹ in bovine b₅. The Phe⁶¹ aromatic ring in cytochrome b₅₅₈ is stacked parallel to the imidazole plane of the second histidine, positioning it firmly in the heme-binding site. This residue is replaced by Trp⁶¹ in cytochrome b₅₅₈, with the indole group having the same structural role.

In addition to the hydrophobic core that constitutes the heme-binding site, a second and smaller hydrophobic core is situated at the other side of the twisted -sheet. This core is even more compact in cytochrome b₅₅₈, where the replacement of the bulky side chains of Ile¹⁵ and Ile²⁷ by valine and alanine allows -helix 1 to pack more closely to the -sheet. Trp²⁵ is part of a conserved structural motif in the small hydrophobic core: its indole side chain is sandwiched between the imidazole ring of His¹⁸ and a methyl group of Ile⁸³. In bovine b₅, Ser²¹ is hydrogen-bonded to Arg⁵⁰, which is in electrostatic interaction with Glu⁴⁶. This hydrophilic interaction has a hydrophobic counterpart in b₅₅₈, where Pro²¹ is in van der Waals interaction with Leu⁵⁰, resulting in a reorientation of the local backbone conformation around Pro²¹. In addition to the Arg⁵⁰/Glu⁴⁶ contact, bovine b₅ has two more salt bridges between exposed side chains. These contacts do not exist in cytochrome b₅₅₈, which has no salt bridges except for the interaction of Lys⁶⁴ with the heme propionates (see above).

Two sets of polar interactions that contribute to anchoring the wall of the heme-binding site to the bottom are conserved in both proteins. On one side of the heme, the hydroxyl group of Thr³⁶ is hydrogen-bonded to both the peptide amide of Cys²⁴ and the carboxylate of Asp³⁴, whereas on the other side, Thr⁵⁸ interacts with Glu⁵⁶ and the backbone nitrogen of His²⁹.

Bovine b₅ has several negatively charged residues surrounding the exposed heme edge. These residues are implicated in electrostatic interactions with their physiological partners. Redox partners carry a complementary positively charged ring as described in the model of Salemme and co-workers (40) for the cytochrome b₅-cytochrome c complex. On the other hand, intramolecular interactions between the cytochrome b₂ domain and the FAD domain in flavocytochrome b₂ involve but one salt bridge interaction, all others being water-mediated hydrogen bonds or van der Waals interactions. Likewise, cytochrome b₅₅₈ has no predominant negative charge positioned around the exposed heme edge (Fig. 8). Besides Glu⁶⁸, only the two propionic acid groups carry a negative charge in this region, and they are partially neutralized by Lys⁶⁴. The interactions of cytochrome b₅₅₈ with hypothetical reaction partners could thus be quite different from those of cytochrome b₅.

View larger version (54K): [in this window] [in a new window]   Fig. 8.   Comparison of the electrostatic molecular surfaces around the heme pockets of cytochromes b558 (A) and b5 (B). The surface is colored according to the electrostatic potential (red, negative; blue, positive) and made transparent; the heme group is blue. This figure was generated by GRASP (41).

Possible Functional Role for Cytochrome b₅₅₈-- A biological function for E. vacuolata cytochrome b₅₅₈ is not known so far. However, we do want to propose two possibilities, which we deduced from considering the distribution of cytochrome b₅ homologues, including an analysis of the >20 bacterial and two eukaryotic genomes that are presently known.

Cytochrome b₅ had not been reported in any bacterial species until now, and b₅ genes have not been identified in any of the bacterial genomes. However, a BLAST search of the Mycobacterium tuberculosis genome (42) revealed the presence of an unidentified gene (Rv1371) with similarity to the fatty-acid desaturases of plants, animals, and fungi. That it is indeed a b₅ homologue is shown in the alignment of Fig. 1. It has the two His heme ligands located at the appropriate positions in the sequence, the Pro-Gly-Gly tripeptide after the fifth heme ligand, and Trp²² as well as other conserved residues. It has the flavocytochrome b₂ deletion ahead of the sixth heme ligand and a two-residue insertion after helix 5. It also appears to have one of the sulfite oxidase insertions after the sixth heme ligand. The M. tuberculosis cytochrome b₅ differs from other cytochromes b₅ in having a single-residue deletion three or four residues after Trp²². We think there are two likely functional roles for cytochrome b₅₅₈ in E. vacuolata. It could be a subunit of a fatty-acid desaturase as found in M. tuberculosis, yeast (43), and animals such as Caenorhabditis elegans (44). An equally likely alternative because of the utilization of reduced sulfur compounds by E. vacuolata is a sulfite oxidase subunit.

Evolutionary Considerations-- The present eukaryotic cytochromes b₅ can be assumed to be derived from an ancestral prokaryotic cytochrome b₅, for which the E. vacuolata cytochrome b₅₅₈ could stand as a model. The structure of the prokaryotic cytochrome b₅₅₈ shows that this protein provides an evolutionarily stable framework that permits a considerable exchange of residues, including the amino acids at the surface as well as amino acids outlining the heme environment, causing no fundamental structural change in the polypeptide backbone. Comparing ~40 deposited b₅ sequences, Lederer (6) located eight invariant residues in the b₅ fold, of which two now appear not to be conserved in the prokaryotic cytochrome b₅₅₈. The two residues that are replaced are involved in the interactions with the heme ligands (His⁴² and His⁷⁰). The proline that replaces Gly⁴⁵ in the His-Pro-Gly-Gly loop causes slight changes in the backbone conformation and creates space for an extra water molecule in the His⁴² environment. The invariant Phe⁶¹ is replaced by tryptophan, retaining the same role in the stabilization of His⁷⁰. Both changes do not seem to be typical of prokaryotic b₅, as the M. tuberculosis open reading frame has Gly at position 45 and lacks an aromatic side chain at position 61 altogether.

Although it is surprising how strict the fold is conserved throughout evolution, the function of cytochrome b₅ diverged in eukaryotic organisms, and the protein has become a modular domain of several redox systems. Residues conserved throughout the b₅ family are therefore to be associated with folding of the protein and binding of a heme group and not with interacting with the redox partners. With the exception of Gly⁴⁵, replaced by Pro⁴⁵, all these residues are conserved in cytochrome b₅₅₈. It is interesting to note that the E. vacuolata protein is not significantly more related to any particular functional class of eukaryotic cytochrome b₅, whether the microsomal b₅ domain of flavocytochrome b₂, sulfite oxidase, or fatty-acid desaturases. It is in fact almost equidistant to these four subclasses of the cytochrome b₅ family.

	ACKNOWLEDGEMENTS

We sincerely thank Drs. K. De Sutter and N. Mertens for providing E. coli expression plasmids.

	FOOTNOTES

^* The work was supported in part by Grant GM 21277 from the National Institutes of Health (to M. A. C.) and by Grant Fi 295/1 from the Deutsche Forschungsgemeinschaft (to U. F.).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) AF183259.

The atomic coordinates and structure factors (code 1CXY, 009616) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

^§ Research Assistant of the Fonds voor Wetenschappelijk Onderzoek-Vlaanderen.

^** Supported by Fonds voor Wetenschappelijk Onderzoek-Vlaanderen Research Projects G.0068.96 and G.0054.97. To whom correspondence and reprint requests should be addressed: Dept. of Biochemistry, Physiology, and Microbiology, Lab. of Protein Biochemistry and Protein Engineering, University of Gent, Ledeganckstraat 35, B-9000 Gent, Belgium. Tel.: 32-92645109; Fax: 32-92645338; E-mail: jozef.vanbeeumen@rug.ac.be.

¹ V. Kostanjeveki, D. Leys, G. Van Driessche, T. E. Meyer, M. A. Cusanovich, U. Fischer, Y. Guisez, and J. Van Beeumen, unpublished results.

	ABBREVIATIONS

The abbreviations used are: PCR, polymerase chain reaction; MES, 4-morpholineethanesulfonic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted 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 Request Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Kostanjevecki, V. Articles by Van Beeumen, J. Search for Related Content PubMed PubMed Citation Articles by Kostanjevecki, V. Articles by Van Beeumen, J. Social Bookmarking                      What's this?