Cytochrome c oxidase subunit 2 from soybean. (A) Domain-structure of the nuclear-encoded Cox2 from soybean, showing the region that was transferred from the mitochondrial genome and the targeting presequence that was acquired on integration into the nuclear genome. MSS is for mitochondrial sorting signal, and TM is for transmembrane spanning domains. (B) Topology of Cox2 in the inner mitochondrial membrane based on homology with yeast and bovine where the crystal structure has been determined. (C) Amino acid alignment of the nuclear-encoded Cox2 (nGmCox2) from soybean and mitochondrial-encoded Cox2 from soybean (mtGmCox2) and Amphicarpa bracteata (mtAbCox2). Amino acid differences between the nuclear and organelle-encoded soybean proteins are boxed and TM regions are indicated. C to U editing sites in the mtGmCox2 sequence are indicated by an asterisk. Residues 169 and 171, which were changed are indicated with a filled circle (•). The start of the mature Cox2 protein, as previously experimentally determined, is indicated with a vertical arrow.

Comparison of hydropathicity profiles for the mitochondrial-encoded and nuclear encoded subunit 2 of cytochrome c oxidase from soybean. Hydropathicity was calculated by using a sliding window of 17 or 19 residues (default values) respectively for KD () and aWW (). The position of the two transmembrane regions, TM1 and TM2, are indicated with solid lines. mtGmCox2 is indicated with a dashed line, and nGmCox2 is indicated with a solid line. Only a solid line is visible where the lines overlap, indicating the hydropathy plots overlap.

The mitochondrial-encoded subunit 2 of cytochrome c oxidase (mtGmCox2) cannot be imported into mitochondria. (A) Chimeric constructs synthesized to determine whether a mitochondrial-encoded Cox2 could be imported into mitochondria. Sequence derived from the nuclear gene is marked by n, sequence derived from the mitochondrial gene is marked by mt. Gm indicates the sequence was obtained from soybean (Glycine max), and Ab indicates that the sequence was obtained from Amphicarpa bracteata. Other abbreviations are as for Fig. . (B) In vitro import of nGmCox2, nGm1-123/mtGm124-383 and nGm1-123/mtAb124-376 constructs into soybean cotyledon mitochondria. Lane 1, translation lysate containing nGmCox2 precursor protein (denoted P). Lane 2, precursor protein incubated with isolated soybean mitochondria. Lane 3, as lane 2 with proteinase K added. Lanes 4 and 5 are as for lanes 2 and 3, except in the presence of valinomycin to dissipate the membrane potential. Lanes 6–10 are as for lanes 1–5, except with the nGm1-123/mtGm124-383 precursor protein. Lanes 11–15 are as for lanes 1–5, except with the nGm1-123/mtAb124-376 precursor protein. Molecular weight standards are indicated to the right of the panel. The mature Cox2 is denoted M. (C) Western blot analysis of mitochondria and mitoplasts used in D, with antisera raised to porin (outer mitochondrial membrane), cyt c (intermembrane space), Ucp (inner mitochondrial membrane), and Hsp70 (matrix). (D) In vitro import of nGm1-123/mtAb124-376 without transmembrane spanning domains. Lane 1, translation lysate containing nGm1-123/mtAb124-376 ΔTM1 precursor protein. Lane 2, precursor protein incubated with isolated soybean cotyledon mitochondria. Lane 3, as lane 2 with proteinase K added. Lanes 4 and 5 are as for lanes 2 and 3, except that the outer mitochondrial membrane had been stripped by osmotic swelling. Lanes 6–10 are as for lanes 1–5, except with the nGm1-123/mtAb124-376 ΔTM2 precursor protein. Lanes 11–15 are as for lanes 1–5, except with the nGm1-123/mtAb124-376 ΔTM1 + 2 precursor protein. Quantitation of import efficiencies are included in Table 2. Molecular weight standards are indicated to the right of the panel.

Two residues in the first transmembrane spanning region of cytochrome c oxidase subunit 2 determine mitochondrial import. Lane 1, translation lysate containing precursor protein (denoted P). Lane 2, precursor protein incubated with isolated soybean mitochondria. Lane 3, as lane 2 with proteinase K added. Lanes 4–6 and 7–9, as for lanes 1–3 except with different precursor proteins. Precursor protein used in each instance is indicated above or below the panel. Molecular weight standards are indicated to the right of the panel.

Mesohydrophobicity vs. local hydrophobicity “scatter plots” for cytochrome c oxidase subunit 2 from different organisms. (A) Scatter plot of mitochondrial-encoded (closed symbols) and nuclear-encoded (open symbols) Cox2 sequences from legumes. Proteins are distributed on the horizontal axis according to their maximum mesohydrophobicity (〈H〉_60–80) and on the vertical axis according to the maximum value of local hydrophobicity (〈H〉₁₇) for the first transmembrane spanning domain (TM1). Hydrophobicity was calculated by using the scale of KD (). Values for functionally active proteins are indicated by circles, and protein sequences derived from pseudogenes are indicated by triangles. The maximum 〈H〉₁₇ for TM1 in legumes that is encoded by a nuclear gene, and can therefore be imported, is shown by line X. The level of 〈H〉₁₇ for TM1 in legumes that cannot be imported is show by line Y. Ab, Amphicarpa bracteata; Al, Atylosia lineata; Cc, Calopogonium caeruleum; Cl, Cologonia lemmonii; Cu, Cullen; Dv, Dumasia villosa; Ep, Eriosema psoraloides; Gm, Glycine max (soybean); Lf, Lespedeza formosa; Nw, Neonotonia weightii; Os, Otholobium sericeum; Pa, Pseudovigna argentea; Pc, Pseudiminia comosa; Pe, Pachyrhizus erosus; Tu, Teramnus uncinatus; Vu, Vigna unguiculata. (B) Scatter plot of mitochondrial-encoded (closed symbols) and nuclear-encoded (open symbols) Cox2 sequences from many species. Annotation as for A. At, Arabidopsis thaliana; Bt, Bos taurus; Cr, Chlamydomonas reinhardtii; Hs, Homo sapiens; Mm, Mus muscularis; Nc, Neurospora crassa; Ps, Polymotella sp; Rn, Rattus novegicus; Sc, Saccharomyces cerevisiae; Sk, Saccharomyces kluyveri.