A phylogenetic study of cytochrome b561 proteins

Abstract

Background: As an antioxidant and cofactor to numerous metabolic enzymes, ascorbate has an essential role in plants and animals. Cytochromes b561 constitute a class of intrinsic membrane proteins involved in ascorbate regeneration. Despite their importance in ascorbate metabolism, no evolutionary analysis has been presented so far on this newly described protein family.

Results: Cytochromes b561 have been identified in a large number of phylogenetically distant species, but are absent in fungi and prokaryotes. Most species contain three or four cytochrome b561 paralogous proteins, and the encoding genes usually have four or five exons. At the protein level, sequence similarities are rather low between cytochromes b561 within a single species (34-45% identity), and among phylogenetically distant species (around 30% identity). However, particular structural features characterizing this protein family are well conserved in members from all species investigated. These features comprise six transmembrane helices, four strictly conserved histidine residues, probably coordinating the two heme molecules, and putative ascorbate and monodehydro-ascorbate (MDHA) substrate-binding sites. Analysis of plant cytochromes b561 shows a separation between those from monocotyledonous and dicotyledonous species in a phylogenetic tree.

Conclusions: All cytochromes b561 have probably evolved from a common ancestral protein before the separation of plants and animals. Their phyletic distribution mirrors the use of ascorbate as primary antioxidant, indicating their role in ascorbate homeostasis and antioxidative defense. In plants, the differentiation into four cytochrome b561 isoforms probably occurred before the separation between monocots and dicots.

Figure 1

Genomic organization of cytochrome b561-encoding genes in different species. Species and genes shown are: Arabidopsis thaliana (Artb561-1), Oryza sativa (Orsb561), Craterostigma plantagineum (Crpb561), Caenorhabdites elegans (Caeb561-1), Homo sapiens (genes encoding chromaffin granule cytochrome b561 (Hosb561-1) and duodenal cytochrome b561 (Hosb561-2) respectively), Mus musculus (gene for the newly discovered 'ubiquitous' isoform, Mumb561-3), and Drosophila melanogaster (Drmb561-1). Boxes represent exons and lines between correspond to introns or untranslated regions. The numbers in the exons represent starting and ending nucleotide positions when introns are ignored; the numbers above the lines represent the length of introns in base-pairs.

Figure 2

Correlation between transmembrane helices and exon-intron structure in plant and human chromaffin cytochrome b561 proteins and genes. For each protein-gene pair, the protein is shown in the top line with the transmembrane helices as gray bars, and the gene organization is shown below with exons as grey and introns as white bars. The first exon encodes one (Artb561-2), two (Artb561-1 and Hosb561-1) or three (Artb561-4 and Orsb561) transmembrane helices. Numbers represent nucleotide positions. Transmembrane regions were predicted with TMHMM [43].

Figure 3

Alignment of cytochrome b561 protein sequences from plant and animal species. Conserved features are marked: TMH, transmembrane helices; conserved histidine residues, gray shading; conserved aromatic residues, yellow shading; and predicted MDHA-binding site (YSLHSW) and ascorbate-binding site (ALLVYRVFR) in boxes. Other conserved residues are marked in green. Red, small hydrophobic amino acids; green, hydroxyl or amino basic side chains; blue, acidic; purple, positively charged. Conservative changes at a specific position are marked with : under the alignment, while * indicates the perfect conservation at a position in all aligned proteins. The sequence from Hosb561-3 is unpublished (H.A.). Transmembrane helices were predicted with TMHMM software [43].

Figure 4

Unrooted phylogenetic tree including all known cytochrome b561 proteins from plants and animals. The tree was derived from a Clustal W [20] alignment of the amino-acid sequences, created with Treecon software [42]. The distance scale above the tree represents the number of substitutions per site, and bootstrapping values are shown at each branch point (percentage of 200 bootstrap samples). Dicot and monocot clusters are marked as D and M, respectively, among the orthologs of Artb561-1 and Artb561-2.

Figure 5

Unrooted phylogenetic tree of the CB domains from all known cytochrome b561 proteins from plants and animals. The tree was derived from a Clustal W [20] alignment of the amino-acid sequences, created with Treecon software [42] in the same way as that in Figure 4.