Search

Two Cys residues, Cys(I) and Cys(II), are present in most plant alternative oxidases (AOXs). Cys(I) inactivates AOX by forming a disulfide bond with the corresponding Cys(I) residue on the adjacent subunit of the AOX homodimer. When reduced, Cys(I) associates with alpha-keto acids, such as pyruvate, to activate AOX, an effect mimicked by charged amino acid substitutions at the Cys(I) site. Cys(II) may also be a site of AOX activity regulation, through interaction with the small alpha-keto acid, glyoxylate. Comparison of Arabidopsis AOX1a (AtAOX1a) mutants with single or double substitutions at Cys(I) and Cys(II) confirmed that glyoxylate interacted with either Cys, while the effect of pyruvate (or succinate for AtAOX1a substituted with Ala at Cys(I)) was limited to Cys(I). A variety of Cys(II) substitutions constitutively activated AtAOX1a, indicating that neither the catalytic site nor, unlike at Cys(I), charge repulsion is involved. Independent effects at each Cys were suggested by lack of Cys(II) substitution interference with pyruvate stimulation at Cys(I), and close to additive activation at the two sites. However, results obtained using diamide treatment to covalently link the AtAOX1a subunits by the disulfide bond indicated that Cys(I) must be in the reduced state for activation at Cys(II) to occur.

The homodimeric cyanide-resistant alternative oxidase of plant mitochondria reduces oxygen to water without forming ATP. Arabidopsis thaliana alternative oxidase AOX1a is stimulated by pyruvate or other alpha-keto acids associating with a regulatory cysteine at position 78, by succinate in a serine-78 mutant, and by site-directed mutation of position 78 to glutamate. The mechanism of activation was explored with additional amino acid substitutions made at Cys-78 in AOX1a, which was functionally expressed in Escherichia coli. Oxidases with positively charged substitutions (Lys and Arg) were insensitive to pyruvate or succinate but were more active than the wild type without pyruvate. Uncharged substitutions (Gln, Leu) produced an inactive enzyme. These results indicate that activation may be due to conformational changes caused by charge repulsion between the dimer subunits and not through a direct role of alpha-keto acids in catalysis. Oxygen isotope fractionation experiments suggest that the charge of the amino acid at position 78 also affects the entry of oxygen into the active site. Therefore, the N-terminal portion of the protein containing residue 78 can indirectly affect both catalysis at the diiron active site and the path of oxygen to that site. In addition, both positively and negatively substituted alternative oxidases were stimulated by glyoxylate, suggesting the presence of a second activation site, possibly Cys-128.

Mitochondria from all plants, many fungi and some protozoa contain a cyanide-resistant, alternative oxidase that functions in parallel with cytochrome c oxidase as the terminal oxidase on the electron transfer chain. Characterization of the structural and potential regulatory features of the alternative oxidase has advanced considerably in recent years. The active site is proposed to contain a di-iron center belonging to the ribonucleotide reductase R2 family and modeling of a four-helix bundle to accommodate this active site within the C-terminal two-thirds of the protein has been carried out. The structural features of this active site are conserved among all known alternative oxidases. The post-translational regulatory features of the alternative oxidase are more variable among organisms. The plant oxidase is dimeric and can be stimulated by either alpha-keto acids or succinate, depending upon the presence or absence, respectively, of a critical cysteine residue found in a conserved block of amino acids in the N-terminal region of the plant protein. The fungal and protozoan alternative oxidases generally exist as monomers and are not subject to organic acid stimulation but can be stimulated by purine nucleotides. The origins of these diverse regulatory features remain unknown but are correlated with sequence differences in the N-terminal third of the protein.

Mitochondria isolated from maize containing cms-T cytoplasm are specifically sensitive to pathotoxins (T-toxins) produced by the fungi Bipolaris maydis race T and Phyllosticta maydis. T-toxins interact with a 13 kDa membrane-bound toxin receptor protein, URF13, to produce hydrophillic pores in the membrane. Expression of URF13 in Escherichia coli produces bacterial cells that form hydrophillic pores in the plasma membrane when exposed to T-toxin or methomyl. Topological studies have established that URF13 contains three membrane-spanning alpha-helices, two of which are amphipathic and may contribute to pore formation. URF13 specifically binds T-toxin in a cooperative manner. Oligonucleotide-directed mutagenesis of URF13 led to the isolation of methomyl/T-toxin-resistant mutations at 39 separate positions throughout the URF13 primary sequence. Chemical cross-linking of URF13 demonstrated the presence of URF13 oligomers and established that the pore-forming species is oligomeric. The ability of the carboxylate-specific reagent, dicyclohexycarbodiimide to cross-link URF13 has been used in conjunction with site-directed mutagenesis to establish that the URF13 tetramer has a central core consisting of a four-alpha-helical bundle that may undergo a conformational change after T-toxin or methomyl binding. Experimental evidence indicates that URF13 acts as a ligand-gated, pore-forming T-toxin receptor.

In addition to possessing multiple NAD(P)H dehydrogenases, most plant mitochondria contain a cyanide- and antimycin-insensitive alternative terminal oxidase. Although the general characteristics of this terminal oxidase have been known for a considerable number of years, the mechanism by which it is regulated is unclear and until recently there has been relatively little information on its exact nature. In the past 5 years, however, the application of molecular and novel voltametric techniques has advanced our understanding of this oxidase considerably. In this article, we review briefly current understanding on the structure and function of the multiple NADH dehydrogenases and consider, in detail, the nature and regulation of the alternative oxidase. We derive a kinetic model for electron transfer through the ubiquinone pool based on a proposed model for the reduction of the oxidase by quinol and show how this can account for deviations from Q-pool behaviour. We review information on the attempts to isolate and characterise the oxidase and finally consider the molecular aspects of the expression of the alternative oxidase.

A number of phenylamide herbicides are observed to uncouple electron transport in isolated chloroplasts and mitochondria and alter the H+ permeability of artificial liposomes. Several of these phenylamides were incorporated into phosphatidylcholine multilamellar and small unilamellar vesicles to measure their effects on the physical properties of membranes. X-ray diffraction analysis of the multilamellar vesicles revealed that the herbicides partitioned into the hydrocarbon chain region of the bilayer, but caused only minimal perturbations on hydrocarbon chain packing. 31P-NMR spectroscopy of these multilamellar vesicles showed both a broadening and lowering of the phase transition temperature of the bilayer lipids upon addition of the herbicides. 13C-NMR spectroscopy of small, unilamellar phosphatidylcholine vesicles was performed to measure the effects of the phenylamides on the chemical shifts and the spin-lattice relaxation times of the individual phosphatidylcholine carbon atoms. None of the added compounds had any measurable effect on the 13C-NMR chemical shifts of the phosphatidylcholine. However, the herbicides significantly modified spin-lattice relaxation times of certain of the lipid carbon atoms. These results generally indicate that the herbicides orient in the lipid bilayers such that the hydrocarbon chains of the phenylamides associate with the hydrocarbon chains of the lipid, whereas the phenyl moiety resides in the polar region of the bilayer.

The location of ubiquinone-10 in phospholipid bilayers was analyzed using a variety of physical techniques. Specifically, we examined the hypothesis that ubiquinone localizes at the geometric center of phospholipid bilayers. Light microscopy of dipalmitoylphosphatidylcholine at room temperature in the presence of 0.05-0.5 mol fraction ubiquinone showed two separate phases, one birefringent lamellar phase and one phase that consisted of isotropic liquid droplets. The isotropic phase had a distinct yellow color, characteristic of melted ubiquinone. [13C]NMR spectroscopy of phosphatidylcholine liposomes containing added ubiquinone indicated a marked effect on the 13C-spin lattice relaxation times of the lipid hydrocarbon chain atoms near the polar head region of the bilayer, but almost no effect on those atoms nearest the center of the bilayer. X-ray diffraction experiments showed that for phosphatidylcholine bilayers, both in the gel and liquid-crystal-line phases, the presence of ubiquinone did not change either the lamellar repeat period or the wide-angle reflections from the lipid hydrocarbon chains. In electron micrographs, the hydrophobic freeze-fracture surfaces of bilayers in the rippled (P beta') phase were also unmodified by the presence of ubiquinone. These results indicate that the ubiquinone which does partition into the bilayer is not localized preferentially between the monolayers, and that an appreciable fraction of the ubiquinone forms a separate phase located outside the lipid bilayer.

The effects of the quinone analog dibromothymoquinone on electron transfer in isolated mung bean mitochondria are described. Both the main, cyanide-sensitive and the alternate, cyanide-insensitive pathways are inhibited by dibromothymoquinone but in markedly different fashions. Half-maximal inhibition appeared at 40 microM and 20 microM dibromothymoquinone for the cyanide-sensitive and alternate pathways, respectively. With succinate as the electron donor, dibromothymoquinone inhibited the alternate pathway at a single site; showing a mixed, non-competitive type inhibition. On the succinate, cyanide-sensitive pathway dibromothymoquinone showed two sites of inhibition and neither coincides with the site of inhibition associated with the alternate pathway. With malate as the electron donor, two sites of inhibition by dibromothymoquinone were observed regardless of the pathway measured. Dibromothymoquinone also inhibited the rate of valinomycin-induced swelling of isolated mung bean mitochondria. Steady-state kinetics showed the inhibition to be non-competitive with respect to valinomycin. Additionally dibromothymoquinone was observed to increase the fluorescence polarization associated with the hydrophobic probe 1,6-diphenylhexatriene. The results indicated that dibromothymoquinone decreased the fluidity of the inner mitochondrial membrane and suggested that the inhibition of mitochondrial electron transfer by dibromothymoquinone may be associated with this decrease in membrane fluidity. The relationship of the multisite nature of the inhibition of electron transfer by dibromothymoquinone and the possible role of mobile electron carriers such as ubiquinone on the main and alternate respiratory pathways of higher plants is discussed.

The NO complex of lipoxygenase with EPR signals near g = 4.0 is an S = 3/2 system with D approximately 15 cm-1 similar to Fe2+-EDTA-NO. This may result from antiferromagnetic coupling of axial (D greater than E) high spin ferrous iron to NO. The other NO complex of lipoxygenase, with EPR signals below ge, may result from rhombic high spin ferrous iron coupled to NO with D greater than J. The quenching of both signals by a hydroperoxy derivative of linoleic acid probably represents replacement of NO by an oxygen ligand.