Arabidopsis thaliana sequence analysis confirms the presenceof cyt b-561 in plants: Evidence for a novel protein family

Han Asarda*, Javier Terol-Alcaydeb, Valeria Pregerc, Jurgen Del Faverod, Wim Verelsta, Francesca Sparlac,Manuel Pérez-Alonsob, Paolo Trostc

a Laboratory of Plant Physiology, University of Antwerp (RUCA), Groenenborgerlaan 171, B-2020 Antwerp, Belgiumb Departamento de Genética, University of Valencia, Dr. Moliner 50, Burjasot 46100, Spainc Laboratory of Plant Physiology, Department of Biology, University of Bologna, Via Irnerio 42, 40126 Bologna, Italyd Department of Molecular Genetics, Flanders Interuniversity Institute for Biotechnology (VIB), University of Antwerp (UIA),Universiteitsplein 1, 2610 Wilrijk, Belgium

Received 22 May 2000; accepted 16 August 2000

Abstract – Recent advances in theArabidopsis sequencing project has elucidated the presence of two genesAtb561-A andAtb561-B that show limited homology to the DNA sequence encoding for the mammalian chromaffin granule cytochromeb-561(cyt b-561). Detailed analysis of the structural features and conserved residues reveals, however, that the structural homologybetween the presumptiveArabidopsis proteins and the animal proteins is very high. All proteins are hydrophobic and showhighly conserved transmembrane helices. The presumably heme-binding histidine residues in the plant and animal sequences aswell as the suggested binding site for the electron acceptor, monodehydroascorbate, are strictly conserved. In contrast, thesuggested electron donor (ascorbate) binding site is not very well conserved between the plant and animal sequences questioningthe function of this motif. Sequence analysis of theAtb561-B gene demonstrates a different splicing than that initially predictedin silico resulting in a protein with nine extra amino acids and a significantly higher homology to the other cytb-561 sequences.The homology between the plant and animal sequences is further supported by the strong similarity between a number ofbiochemical properties of the chromaffin cytb-561 and the cytb-561 isolated from bean hook plasma membranes. Since themammalian cytb-561 is considered specific to neuroadrenergic tissues, the identification of a closely related homologue in ananeural organism demonstrates that these proteins constitute a new class of widely occurring membrane proteins. Both the plantand animal cytb-561 are involved in transmembrane electron transport using ascorbate as an electron donor. The similaritybetween these proteins therefore suggests, for the first time, that this transport supports a number of different cell physiologicalprocesses. An evolutionary relationship between the plant and animal proteins is presented. © 2000 Éditions scientifiques etmédicales Elsevier SAS

ascorbic acid / chromaffin granules / cytochrome b-561 / electron transport / genome analysis / higher plants / plasmamembrane

aa, amino acid / cyt b-561, cytochrome b-561 / E'0, standard redox potential at pH 7 / EST, expressed sequence tag /MDHA, monodehydroascorbate / ORF, open reading frame

1. INTRODUCTION

Characterization of higher plant plasma membranefractions has revealed the presence of a high potential(E'0: +110 to +165 mV) ascorbate-reducibleb-typecytochrome with anα-band wavelength maximum at

561 nm (cytochromeb-561, cyt b-561) [6, 7]. Thiscytochrome constitutes a common plant plasma mem-brane protein since it was found in a variety of tissuesand in a large number of species [4, 5]. Recentexperiments have demonstrated that the cytb-561 isinvolved in the electrogenic transport of electronsacross the plasma membrane [3, 14] using cytoplasmicascorbate as an in vivo electron donor and monodehy-droascorbate (MDHA) as an extracellular (apoplastic)electron acceptor. Although the physiological function

* Correspondence and reprints: fax +32 3 218 04 17.E-mail address: hanasard@ruca.ua.ac.be (H. Asard).

Plant Physiol. Biochem. 38 (2000) 905−912© 2000 Éditions scientifiques et médicales Elsevier SAS. All rights reservedS0981942800012018/FLA

of this protein remains to be elucidated, the transmem-brane transport of electrons to MDHA suggests a rolein the regeneration of ascorbate. Ascorbate in the plantapoplast plays an important role in antioxidative defencereactions, e.g. those induced under biotic and/or abi-otic stress conditions [21]. Attempts to solubilize andpurify the protein from plasma membrane fractionshave recently been presented [8, 25].

From the early investigations on the plant plasmamembrane cyt b-561, it was recognized that thisprotein shows remarkable biochemical similarities tothe mammalian cyt b-561 characterized from chromaf-fin granules [12, 20]. The latter component (E'0:+140 mV) transfers electrons from cytoplasmic ascor-bate to MDHA present inside the chromaffin vesicles[19, 20, 32]. Intravesicular ascorbate provides reduc-ing equivalents for the monooxygenase dopamine�-hydroxylase involved in dopamine biosynthesis.Aside from hematopoietic tumour cells [26], thisparticular cytochrome has so far only been demon-strated in catecholamine and neuropeptide secretoryvesicles and its expression is coupled to neuronal celldifferentiation [27, 29]. The primary structure of thechromaffin granule cyt b-561 has been elucidated froma number of mammalian species including humans[28], sheep, pig, bovine [23] and mouse [29]. Thepresence of similar genes was demonstrated in Xeno-pus laevis [27] and in Caenorhabditis elegans [33].

On the basis of the amino acid sequence of themammalian cyt b-561, predictions on the secondaryand tertiary structure of the protein have been formu-lated. These include the prediction of five [27, 28] orsix [10, 22, 23] hydrophobic transmembrane helices,respectively. By different experimental approaches themammalian protein has been shown to contain twodistinct hemes [11, 16, 31] and suggestions have beenpresented for the binding sites of the electron donor(ascorbate) and electron acceptor (MDHA) [22].

Recent progress in the sequencing of the Arabidop-sis genome has revealed the presence of two distinctsequences showing homology to the mammalian cytb-561 genes. Although the overall homology at theDNA level is relatively low, a detailed comparison ofthe predicted protein structure, together with newsequence information presented in this paper, and thebiochemical properties of the plant plasma membranecyt b-561 strongly supports the idea that the plant andanimal proteins are highly similar. This analysis dem-onstrates that the high potential cyt b-561 constitutes anovel protein family present in neural and aneuraleukaryotic species.

2. RESULTS AND DISCUSSION

2.1. Identification of Atb561-A and Atb561-B

As a result of the participation of the Valencia groupin the Arabidopsis Genome Sequencing Project, thesequence of the genomic clone M7J2 from Arabidop-sis chromosome IV was obtained. An exhaustivesearch for similarities was performed on the 81.6 kbgenomic sequence using the BLAST program. A smallregion showed a weak similarity to the cyt b-561 genefrom different animals. That region of DNA wasanalysed to find the existing coding sequences, usingdifferent approaches (see Methods). An ORF, 1 752 bplong and containing four exons, was predicted.

The conceptual translation of the ORF produced a280-amino acid protein (GenBank accession number:AL022197), which was used to perform a new searchfor homologies using the BLAST program. This pro-tein showed a clear similarity to cyt b-561 proteinsderived from human, mouse, pig, cow, sheep, X. laevisand the nematode C. elegans. Nine Arabidopsis ESTclones were found (GenBank: AA720180, AA728528,F19849, R65413, T46072, T04182, T04807, T88282and Z37601) and the corresponding cDNAs weresequenced from both strands. The resulting sequencesconfirmed the in silico cDNA prediction and wasnamed Atb561-A.

A putative protein showing homology to Atb561-Awas found in the KAOS database [24] (GenBankaccession number AB005231), in a genomic clone(BAC clone) MBB18 from chromosome V. However,the theoretical protein of 223 aa predicted from thissequence [24] was not correct. On the basis of sequenc-ing the corresponding cDNA from the ESTs (GenBankaccession numbers: Z26702, Z33925) we determinedthat the second exon of the ORF was 27 bp longer. Theadditional nine amino acids (Ala79–Glu87 in figure 2)showed a high homology to Atb561-A and the animalproteins (see below). The new protein was namedAtb561-B (GenBank accession number: AF132115).

2.2. Intron-exon analysis

Combining the data from the genomic and ESTsequences showed a transcription unit for Atb561-Acomprised of four exons, of 267, 48, 201 and 324 bp,separated by introns of variable size (figure 1). The5'-untranslated region is approximately 100 bp longand we could not identify any sequence resembling theTATA box. In this respect, Atb561-A is similar to thehuman cyt b-561 gene which has no TATA box [26].Furthermore, two putative initiation sites are present in

906 H. Asard et al. / Plant. Physiol. Biochem. 38 (2000) 905–912

the Atb561-A ORF, with the second one 123 bp down-stream of the first (not shown). On the basis of themultiple alignment with the other cytochromes (seefigure 2), the second methionine codon in the ORF ismost likely the starting point for translation. In addi-tion, two ESTs (GenBank accession numbers: R65413,T04182), containing the full length Atb561-A, begindownstream of the first ATG. Therefore, only the shortversion of the protein is included in figure 2, starting atthe second methionine of the ORF. A similar problemwith the bovine protein, which also contains twomethionine codons in the beginning of the sequence,was solved by analysis at the protein level [27],suggesting that the protein was probably 21 aa shorterthan that previously reported [23]. The polyadenyla-tion signal for Atb561-A is 157 bp downstream of thetermination codon. In contrast to Atb561-A the 5'-UTRfor Atb561-B contains a putative TATA box 30 bpupstream of the first methionine codon.

Although the similarity between Atb561-A and -Bgenes is only 57.4 % at the DNA level, the organiza-tion of the two transcription units in A. thaliana isquite similar. Both ORFs have similar lengths, havefour exons and code for proteins of 239 and 230 aa,respectively. As could be expected, the introns havenot been conserved (figure 1). The intron-exon struc-ture of the Arabidopsis genes was compared to that ofthe human cyt b-561 homologue, which is the onlyother protein of this family described at the genomiclevel so far [26]. The human protein contains fiveexons and two of the four splicing sites are perfectlyconserved between the Arabidopsis and humansequences (figure 2).

According to our prediction of the Atb561-A and -Bstructure, the intron-exon organization does not reflect

the structural domains of the protein. There are fourexons and six predicted transmembrane domains andtheir position does not indicate that the exons reflectthe structural domains of the protein. Transmembranehelix I (TR1) is present in exon 1 (E1), helix IIexpands the end of E1, E2 and the beginning of E3,helix III is located on E3, helix IV is divided betweenexons 3 and 4 and helices V and VI are present in E4(figure 1). In contrast in the five transmembrane helixmodel suggested by Srivastava [27] for the humanprotein, each of the exons contains a single transmem-brane domain.

2.3. Multiple alignment and structural analysisBased on the Kyte and Doolittle [17] algorithm,

hydropathy plots were constructed for the two Atb561proteins (figure 3) and, as an example, compared tothat of the human protein. The hydropathy plot showssix highly hydrophobic regions for each sequenceseparated by short hydrophilic loops. The hydropho-bicity profiles of the Arabidopsis and human proteinsclosely overlap which is also reflected in the similarlocations of the predicted transmembrane helices inthe multiple alignment (figure 2). The very highhydrophobicity of the plant proteins is consistent withthe absolute need for detergent in order to solubilizethe protein from purified plasma membrane fractions[8, 25].

Recent predictions on the transmembrane structureof cyt b-561 from chromaffin granules are still contra-dictory. On one hand, the analysis of Srivastava et al.[27, 28] indicates the presence of five membranespanning helices with the N-terminus facing the chro-maffin vesicle interior. This prediction was partiallybased on the inaccessibility of the N-terminus to

Figure 1. Intron-exon structure of theArabidopsis genes. Upper and lowerstructures represent the introns (line) andexons (blocks) in the Atb561 genes. Themiddle part represents the predicted pro-tein sequences showing the putativetransmembrane (TRn) domains (dottedareas) relative to the exon (En) struc-tures.

H. Asard et al. / Plant. Physiol. Biochem. 38 (2000) 905–912 907

908 H. Asard et al. / Plant. Physiol. Biochem. 38 (2000) 905–912

antibodies and proteases [15]. On the other hand, Perinet al. [23], Degli Esposti et al. [10] and Okuyama et al.[22] suggested the presence of six transmembranehelices in the mammalian cyt b-561. The latter modelaccounts for the accommodation of two heme mol-ecules per protein. Secondary structure predictions(TMpred software, [13]), for the Arabidopsis proteinsalso suggested six transmembrane helices (figure 2). Acomparison of the predicted membrane spanningdomains for all known cyt b-561 proteins indicates thehigh conservation of this structural feature (figure 2).

Multiple alignment revealed the strict conservationof five histidine residues (His64, His98, His120, His140,His179, figure 2) located in the helices II to V. In the six

helix model for the mammalian sequence, one of thetwo heme molecules was suggested to be co-ordinatedby the pair His64-His140, whereas the other heme couldbe co-ordinated by the pair His98-His179 or, alterna-tively by the pair His102-His179 [22]. However, sincethe His102 is not conserved in the Atb561-B sequence,we suggest that this residue is not involved in hemeligation.

Another remarkable site of strict conservation amongthe Arabidopsis and animal proteins is the polarfive-amino acid stretch 138SLHSW142. This sequenceis located at the intravesicular end of α-helix IV in thechromaffin vesicle membrane. The sequence is alsostrictly conserved in a partial EST-derived Drosophilamelanogaster sequence (191 aa) (GenBank accessionnumber: AI403697) that shows a high similarity (36 %identity) to the human cyt b-561 sequence. A tentativecyt b-561 sequence identified in Mesembryanthemumcrystallinum (GenBank accession number: AF097661)contains a conservative replacement for the Ser138

residue by Thr. By sequence comparison, Tsubaki’sgroup [22] suggested that this sequence could providethe binding site for MDHA. MDHA functions as anelectron acceptor to the plant cyt b-561-mediatedelectron transport as demonstrated by using ascorbate-loaded plasma membrane vesicles [14]. The highconservation of the possible MDHA-binding site sup-ports these results and points to the functional similar-ity between the plant and animal proteins.

Similarly, a stretch of nine amino acids(79ALLVYRVFR87) was indicated as the binding sitefor ascorbate and is strictly conserved in animalsequences [22]. However, comparison with plantsequences demonstrates that this sequence is onlypartially conserved in A. thaliana (figure 2) and M.crystallinum (not shown). In addition this stretch isalso not conserved in the partial D. melanogastersequence. These observations question the function ofthis sequence and are of importance with respect to theelucidation of the electron transfer mechanism fromascorbate to one of the heme groups of cyt b-561.

Finally, comparison of potential N-glycosylationmotifs (at the PROSITE database), showed thatAtb561-A contains two potential N-glycosylation sites(Asp119, Asp223, figure 2), whereas no potential gly-cosylation sites were found in the Atb561-B or human

Figure 2. Multiple alignment analysis (Clustal W, [30]) between the predicted Arabidopsis cytochrome b-561 homologues, and other cytochromeb-561 homologues identified in animal cells. The shaded areas indicate the predicted transmembrane helices using the TMpred software [13]. Theboxed areas represent the predicted ascorbate and MDHA binding sequences. X , splicing site (first aa of new exon); N, Asp glycosylation sites;H, conserved His residues.

Figure 3. Hydropathy plot analysis for Atb561-A and -B and thehuman cytochrome b-561 protein sequence.

H. Asard et al. / Plant. Physiol. Biochem. 38 (2000) 905–912 909

sequences. Both Atb561-A and -B also showed a highprobability for a plasma membrane (64 % forAtb561-Aand B) or ER (68.5 % for Atb561-B and 37 % forAtb561-A) localization as identified by comparison toknown signal peptide sequences (PSORT version 6.4,[18]). Although these remain predictions, they areconsistent with the experimental evidence that at leastone of these proteins is probably located at the plasmamembrane. In addition, some evidence indicates thatan ascorbate-reducible cytochrome b is also found inother subcellular fractions, different from mitochon-dria and endoplasmic reticulum, in cauliflower [2] andbean hypocotyls [25].

On the basis of the conservation and similaritiesbetween the plant and animal sequences, a model isproposed for the Atb561-A protein using the six-helixmodel (figure 4). Based on the conservation of theMDHA binding site and the potential glycosylationsites, the C- and N-terminus are located on thecytoplasmic site of the membrane.

2.4. Evolutionary relationships

The evolutionary relationship between the proteinsanalysed in this study (figure 5) clearly reflects thesystematic positions of the species, suggesting that theprotein behaves as a molecular clock. Most of thesequences available so far originate from vertebratespecies. However, an increasing number of ESTs fromplants becoming available shows homology to the cytb-561 protein family and will allow a further refine-ment of the evolutionary tree. Remarkably, no homolo-

gous protein has been found in Saccharomyces cer-evisea, whose full genome has been sequenced. Theprotein is present in insects, as there are several D.melanogaster ESTs with a high degree of similarity tothe human cyt b-561.

3. CONCLUSION

This paper presents the first evidence for the exist-ence of a new protein family, consisting of b-typecytochromes common to eukaryotic species. It isobvious that the transmembrane electron transportcapacity of all of these proteins is probably verysimilar, and that this reaction may support a number ofdifferent physiological processes. Future work will bedirected towards the further characterization of thesefunctions in plants and of the regulation of the expres-sion of the cyt b-561.

4. METHODS

The Arabidopsis thaliana P1 genomic clone M7J2,corresponding to the bottom arm of chromosome IV,was completely sequenced in the frame of the Euro-pean Arabidopsis Genome Sequencing Project. A shot-gun library approach was used to determine the DNAsequence of the insert. The DNAsequence was obtainedby using the IR Taq DNA sequencing kit (BoehringerMannheim) and a LI-COR Automated DNA Sequencer.The sequencing project was managed with the StadenSoftware Package, a total of 81.6 kb were analysedand the average redundancy was 8. The search forORFs in the genomic clone was performed with theGENSCAN program [9], in collaboration with theMunich Information Centre for Protein Sequences

Figure 4. Putative model for the molecular structure of cytochromeb-561 based on the Arabidopsis sequences and the similarities to themammalian cytochrome b-561.

Figure 5. Evolutionary relationship between cytochrome b-561related sequences.

910 H. Asard et al. / Plant. Physiol. Biochem. 38 (2000) 905–912

(MIPS). The sequence databases were searched forsequence similarities using the BLAST program of theNational Center for Biotechnology Information [1].EST clones were sequenced with an ABI automaticDNA sequencer and compared to the genomicsequences using the GAP and BESTFIT programsfrom the GCG Software Package (Wisconsin Univer-sity). The Arabidopsis genes will be referred to asAtb561-A (for Arabidopsis thaliana cyt b-561) andAtb561-B.

NOTE ADDED IN PROOF

In a recent paper (Asard et al. Protoplasma, inpress), the suggested nomenclature for the plant Ara-bidopsis cyt b-561 genes is Artb561-n. A third putativeArabidopsis cyt b-561 has also been identified (Gen-Bank accession number AC006917).

Acknowledgments. The authors gratefully acknowl-edge the financial support of the Fund for ScientificResearch Flanders (FWO-Flanders), by CNR (Targetproject on Biotechnology) and by PRIN Bioenergeticsand Membrane Transport, MURST, Italy. HA and WVare Research Associate and doctoral student respec-tively at the FW-Flanders.

REFERENCES

[1] Altschul S.F., Madden T.L., Schaffer A.A., Zhang J.,Zhang Z., Miller W., Lipman D.J., Gapped BLASTand PSI-BLAST: a new generation of protein databasesearch programs, Nucleic Acids Res. 25 (1997)3389–3402.

[2] Asard H., Caubergs R.J., Renders D., De Greef J.A.,Duroquinone-stimulated NADH oxidase activity andb-type cytochromes in the plasma membrane of cauli-flower and mung beans, Plant Sci. 53 (1987) 109–119.

[3] Asard H., Horemans N., Caubergs R.J., Transmem-brane electron transport in ascorbate-loaded plasmamembrane vesicles from higher plants involves ab-type cytochrome, FEBS Lett. 306 (1992) 143–146.

[4] Asard H., Horemans N., Mertens J., Caubergs R.J.,The plant plasma membrane b-type cytochrome: anoverview, Bel. J. Bot. 127 (1994) 171–183.

[5] Asard H., Horemans N., Preger V., Trost P., Plasmamembrane b-type cytochromes, in: Asard H.,Bérczi A., Caubergs R.J. (Eds.), Plasma MembraneRedox Systems and their Role in Biological Stress andDisease, KluwerAcademic Publishers, Dordrecht, 1998,pp. 1–32.

[6] Asard H., Venken M., Caubergs R., Reijnders W.,Oltmann F.L., De Greef J.A., b-Type cytochromes inhigher plant plasma membranes, Plant Physiol. 90(1989) 1077–1083.

[7] Askerlund P., Larsson C., Widell S., Cytochromes ofplant plasma membranes, Characterization by absor-bance difference spectrophotometry and redox titra-tion, Physiol. Plant. 76 (1989) 123–134.

[8] Bérczi A., Møller I.M., Characterization and solubili-zation of residual redox activity in salt-washed anddetergent-treated plasma membrane vesicles from spin-ach leaves, Protoplasma 205 (1998) 59–65.

[9] Burge C., Karlin S., Prediction of complete genestructures in human genomic DNA, J. Mol. Biol. 268(1997) 78–94.

[10] Degli Esposti M., Kamensky Y.A., Arutjunjan A.M.,Konstantinov A.A., A model for the molecular organi-sation of cytochrome �-561 in chromaffin granulemembranes, FEBS Lett. 245 (1989) 74–78.

[11] Degli Esposti M., Palmer G., Lenaz G., Circulardichroic spectroscopy of membrane haemoproteins.The molecular determinants of the dichroic propertiesof the b cytochromes in various various ubiquinol:cy-tochrome c reductases, Eur. J. Biochem. 182 (1989)27–36.

[12] Flatmark T., Terland O., Cytochrome b561 of thebovine adrenal chromaffin granules. A high potentialb-type cytochrome, Biochim. Biophys. Acta 253 (1971)487–491.

[13] Hofmann K., Stoffel W., Tmbase - A database ofmembrane spanning protein segments, Biol. Chem.347 (1993) 166.

[14] Horemans N., Asard H., Caubergs R.J., The role ofascorbate free radical as an electron acceptor to cyto-chrome b-mediated trans-plasma membrane electrontransport in higher plants, Plant Physiol. 104 (1994)1455–1458.

[15] Kent U.T., Fleming P.J., Cytochrome b561 is fattyacylated and oriented in the chromaffin granule mem-brane with its carboxyl terminus cytoplasmicallyexposed, J. Biol. Chem. 265 (1990) 16422–16427.

[16] Kobayashi K., Tsubaki M., Tagawa S., Distinct roles oftwo heme centers for transmembrane electron transferin cytochrome b561 from bovine adrenal chromaffinvesicles as revealed by pulse radiolysis, J. Biol. Chem.237 (1998) 16038–16042.

[17] Kyte J., Doolittle R.F., A simple method for displayingthe hydropathic character of a protein, J. Mol. Biol.157 (1982) 105–132.

[18] Nakai K., Kanehisa M., A knowledge base for predict-ing protein localization sites in eukaryotic cells, Genom-ics 14 (1992) 897–911.

[19] Njus D., Kelley P.M., Harnadek G.J., Pacquing Y.V.,Mechanism of ascorbic acid regeneration mediated bycytochrome b561, Ann. N. Y. Acad. Sci. 493 (1987)108–119.

[20] Njus D., Knoth J., Cook C., Kelley P.M., Electrontransfer across the chromaffin granule membrane,J. Biol. Chem. 258 (1983) 27–30.

H. Asard et al. / Plant. Physiol. Biochem. 38 (2000) 905–912 911

[21] Noctor G., Foyer C., Ascorbate and glutathion: keep-ing active oxygen under control, Annu. Rev. PlantPhysiol. Plant Mol. Biol. 49 (1998) 249–279.

[22] Okuyama E., Yamamoto R., Ichikawa Y., Tsubaki M.,Structural basis for the electron transfer across thechromaffin vesicle membranes catalyzed by cytochromeb561: analyses of DNA nucleotide sequences and vis-ible absorption spectra, Biochim. Biophys. Acta 1383(1998) 269–278.

[23] Perin M.S., Fried V.A., Slaughter C.A., Südhof T.C.,The structure of cytochrome b561, a secretory vesicle-specific electron transport protein, EMBO J. 7 (1988)2697–2703.

[24] Sato S., Kotani H., Nakamura Y., Kaneko T.,Asamizu E., Fukami M., Miyajima N., Tabata S.,Structural analysis of Arabidopsis thaliana Chromo-some 5. I. Sequence features of the 1.6 Mb regionscovered by physically assigned twenty P1 clones,DNA Res. 4 (1997) 215–230.

[25] Scagliarini S., Rotino L., Bäurle I., Asard H., Pu-pillo P., Trost P., Initial purification study of thecytochrome b-561 of bean hypocotyls plasma mem-branes, Protoplasma 205 (1998) 66–73.

[26] Srivastava M., Genomic structure and expression ofthe human gene encoding cytochrome b561, an integralprotein of the chromaffin granule membrane, J. Biol.Chem. 270 (1995) 22714–22720.

[27] Srivastava M., Xenopus cytochrome b561: molecularconfirmation of a general five transmembrane structureand developmental regulation at the gastrula stage,DNA Cell. Biol. 15 (1996) 1075–1080.

[28] Srivastava M., Gibson K.R., Pollard H.B., Flem-ing P.J., Human cytochrome b561: a revised hypoth-esis for conformation in membranes which reconcilessequence and functional information, Biochem. J. 303(1994) 915–921.

[29] Srivastava M., Pollard H.B., Fleming P.J., Mousecytochrome b561: cDNA cloning and expression in ratbrain, mouse embryos, and human glioma cell lines,DNA Cell Biol. 17 (1998) 771–777.

[30] Thompson J.D., Higgins D.G., Gibson T.J., CLUSTALW: improving the sensitivity of progressive multiplesequence alignment through sequence weighing,position-specific gap penalties and weight matrix choice,Nucleic Acids Res. 22 (1994) 4673–4680.

[31] Tsubaki M., Nakayama M., Okuyama E., Ichikawa Y.,Hori H., Existence of two heme B centers in cyto-chrome b561 from bovine adrenal chromaffin vesiclesas revealed by a new purification procedure and EPRspectroscopy, J. Biol. Chem. 272 (1997) 23206–23210.

[32] Wakefield L.M., Cass A.E.G., Radda G.K., Functionalcoupling between enzymes of the chromaffin granulemembrane, J. Biol. Chem. 261 (1986) 9739–9745.

[33] Wilson R., Ainscough R., Anderson K., Baynes C.,Berks M., Bonfield J., Burton J., Connell M.,Copsey T., Cooper J., Coulson A., Craxton M.,Dear S., Du Z., Durbin R., Favello A., Fraser A.,Fulton L., Gardner A., Green P., Hawkins T., Hillier L.,Jier M., Johnston L., Jones M., Kershaw J., Kirsten J.,Laisster N., Latreille P., Lightning J., Lloyd C., Mor-timore B., O’Callaghan M., Parsons J., Percy C.,Rifken L., Roopra A., Saunders D., Shownkeen R.,Sims M., Smaldon N., Smith A., Smith M., Sonnham-mer E., Staden R., Sulston J., Thierry-Mieg J., Tho-mas K., Vaudin M., Vaughan K., Waterson R., Wat-son A., Weinstock L., Wilkinson-Sproat J.,Wohldman P., 2.2 Mb of contiguous nucleotide sequencefrom chromosome III of C. elegans, Nature 368 (1994)32–38.

912 H. Asard et al. / Plant. Physiol. Biochem. 38 (2000) 905–912