PMC

Comparison of fungal and plant mitochondrial porin gene structures. Symbols are as described for Figure 6, and data were derived from the sequences described in Figure 8.

Evolutionary history of eukaryotic porin sequences – fungi and animals. The portion of the phylogenetic tree, described in Figure 2, containing the fungal and animal data is presented. Symbols and notations are as described for Figure 2.

Overview of the predicted transmembrane arrangement of the Neurospora mitochondrial porin across the mitochondrial outer membrane. The model takes into account several secondary structure predictions, and experimental probing of the structure in artificial bilayers through the use of point mutations [14], deletions [9, 15, 16], and site-specific biotinylation [13], as described in [16]. The predicted N-terminal α-helix is represented by a cylinder, and putative β-strands by zig-zag lines, and loops and β-turns by curved arrows. The arrowheads point in the direction of the C-terminus of the protein.

Alignment of representative porin sequences. Amino acid sequences from a plant (O. sativa VDAC1, Osat1,CAB82853), an animal (H. sapiens VDAC1, Hsap1, NP_003365), a fungus (N. crassa, Ncra, XP_323644) and a stramenopile (P. sojae, Psoj, C_1010023) were aligned using PRALINE, as described in Methods. Residues in grey background are predicted to reside in β-strands, overlined sequences in α-helical regions, and identical (|) and conserved (:) residues between adjacent pairs of sequences are indicated. Chemically-similar residues in all four representative sequences are shown in bold. The GLK motif (residues 87–89 of Osat1) and the existing eukaryotic porin motif (PS00558, residues 216–238 of Osat1) are indicated by dotted underlining. The VKAKV sequence (residues 224–228 of Osat1) is part of PS00558. Note that this motif is not found in the P. sojae VDAC sequence. Motifs, derived in this work, that are characteristic for plants (7–48 of Osat1), animals (4–18 of Hsap1) and fungi (1–16 and 68–74, separated by any 48–52 residues, Ncra) are underlined. A unique motif for the stramenopiles could not be discerned from the data available.

Evolutionary history of eukaryotic porin sequences – stramenopiles and plants. The phylogenetic tree, continued in Figure 3, is based on NJ and Bayesian analysis of 141 VDAC sequences. Stramenopile sequences were used as the out group. Levels of confidence of the nodes are only provided if support is above 66%. The numbers are based on posterior probability values generated by Bayesian analysis and on bootstrap analysis in combination with NJ analysis (italics). The presence of the GLK (G) and Eukaryotic porin motifs (E) are indicated towards the right of the phylogenetic tree. Note among the plants the GLK domain appears as the STK (S) motif; an X indicates the absence of the motif. Lower case g designates a G-any-K or G-any-R, where "any" refers to any other amino acid. The lower case s indicates S-any-K or S-any-R. The minus (-) indicates that the sequence was incomplete and thus the GLK and eukaryotic porin motifs could not be identified. Nodes designated by a number (1–10) are discussed within the text. Underlined accession numbers are those of VDACs that do not contain the signature motif identified in this study.

Predicted secondary structure elements in VDAC from the crown groups of plants, animals, and fungi. Predictions were made as described in Methods. For each summary diagram, the putative N-terminal α-helix is indicated by a hatched bar labelled "N" on the left, subsequent β-strands are indicated by filled rectangles, and the intervening loops are shown as thin lines. According to the model in Fig. 1, the N-terminal helix resides in the intermembrane space, and the subsequent loops and turns alternate between exposure to the cytosol and to the intermembrane space. β-strands with weak support are indicated in grey. The lower panel shows the model for Neurospora crassa VDAC structure derived in [16]. Structural elements are as described for the plant, animal and fungal models, except that the checkerboard region indicates a C-terminal segment that is exposed to the cytosol rather than forming the 19^thβ-strand (see text for discussion). Below the model of N. crassa VDAC structure, rectangles indicate segments, that when absent in porin variants, create molecules that form pores of wild-type conductivity (open), or inefficiently form pores that are either unstable or of reduced conductivity (filled) in artificial membranes. The position of the GLK sequence and the eukaryotic porin signature motif (PS00558) are also noted. Vertical lines connect regions of homology and the curved arrow indicates the discrepancy for the placement of β8 between previous models (see [4]) and the current predictions.

Comparison of animal mitochondrial porin gene structures. The exonic and intronic regions of mitochondrial porin open reading frames were mapped for a variety of organisms. The gene structure for all species was retrieved from the NCBI genome database using BLASTN or map viewer programs to analyze the ORF of each porin cDNA. Data were derived from the sequences described in the legend to Figure 8. Only regions of the gene containing ORFs are shown; chromosome number is indicated beside the species in Roman numerals and plus (+) or minus (-) strand reading frames are indicated beside chromosome number. Where multiple spliceoforms exist, as for human VDAC2, the form not listed as a splice variant was used. Lengths of exons in bp are indicated inside the boxes representing the exons; the numbers of the corresponding amino acid residues are indicated below the boxes. Lines indicate introns, the sizes of which are indicated in kb above the line; the intron phase (0, 1, or 2) is indicated directly above the intron size. Phase 0 introns do not interrupt codons, whereas phase 1 and phase 2 introns are positioned after the first and second nucleotides of the codon, respectively. The positions of the coding sequence for the GLK (diamonds) and the eukaryotic porin (circles) motifs are shown above the intron/exon map. Filled and open symbols indicate matches and imperfect matches to these motifs, respectively.

Intron placement with respect to coding sequences for conserved structural elements. Intron placement is indicated by vertical bars in the rows above the structural model for each crown group of organisms (see Fig. 4 for symbol descriptions). The positions of the GLK motif (triangles), the VDAC motifs identified in this study (stippled boxes) and the eukaryotic signature motif (open boxes) are indicated below the structural models. Note that the fungal VDAC motif is formed by two sequence elements; the intervening region is shown as a dotted line. Patterns of intron location that are found in several VDAC genes are shown only once. Animal sequences. Hs, Homo sapiens VDAC1 (NP_003365) and VDAC3 (NP_005653); Rn, Rattus norvegicus VDAC1 (NP_112643) and VDAC3 (NP_112645); Gg, Gallus gallus VDAC1 (TC10222), VDAC2 (NM_204741), VDAC3 (TC9741); Dr, Danio rerio accession numbers NM_199585 and BC065468; Hs2, H. sapiens VDAC2 (NP_003366), Rn2, R. norvegicus VDAC2 (NP_112644); Dm, Drosophila melanogaster VDAC1 (CG6647-PA), VDAC2 (NP_609462), and isoform B of CG31722 (AAF53018); Dm-2; D. melanogaster isoform A (AAF53019); Sp, Strongylocentrotus purpuratus (sea urchin, XM_775173); Ag, Anopheles gambiae (XM_318947). Middle panel: Fungal sequences. Cn, Cryptococcus neoformans (XM_569804.1); Nc, Neurospora crassa (XP_323644). Lower panel: Plant sequences. At, Arabidopsis thaliana, (NP_186777, NP_201551, NP_200057, NP_197013); At-2 A. thaliana (NP_190561); Os-1, Oryza sativa (CAB82853); Os-2, O. sativa (CAC80850); Os-3, O. sativa (CAC80851).