Pathogenesis of primary defects in mitochondrial ATP synthesis.

Maternally inherited mutations in the mtDNA-encoded ATPase 6 subunit of complex V (ATP synthase) of the respiratory chain/oxidative phosphorylation system are responsible for a subgroup of severe and often-fatal disorders characterized predominantly by lesions in the brain, particularly in the striatum. These include NARP (neuropathy, ataxia, and retinitis pigmentosa), MILS (maternally inherited Leigh syndrome), and FBSN (familial bilateral striatal necrosis). Of the five known pathogenic mutations causing these disorders, four are located at two codons (156 and 217), each of which can suffer mutations converting a conserved leucine to either an arginine or a proline. Based on the accumulating data on both the structure of ATP synthase and the mechanism by which rotary catalysis couples proton flow to ATP synthesis, we propose a model that may help explain why mutations at codons 156 and 217 are pathogenic.

Folded state of the integral membrane colicin E1 immunity protein in solvents of mixed polarity.

The colicin E1 immunity protein (ImmE1), a 13.2-kDa hydrophobic integral membrane protein localized in the Escherichia coli cytoplasmic membrane, protects the cell from the lethal, channel-forming activity of the bacteriocin, colicin E1. Utilizing its solubility in organic solvents, ImmE1 was purified by 1-butanol extraction of isolated membranes, followed by gel filtration and ion-exchange chromatography in a chloroform/methanol/H(2)O (4:4:1) solvent system. Circular dichroism analysis indicated that the alpha-helical content of ImmE1 is approximately 80% in 1-butanol or 2,2,2-trifluoroethanol, consistent with a previous membrane-folding model with three extended hydrophobic transmembrane helical domains, H1-H3. Each of these extended hydrophobic domains contains a centrally located single Cys residue that could be used as a probe of protein structure. The presence of tertiary structure of purified ImmE1 in a solvent of mixed polarity, chloroform/methanol/H(2)O (4:4:1) was demonstrated by (i) the constraints on Tyr residues shown by the amplitude of near-UV circular dichroism spectra in the wavelength interval, 270-285 nm; (ii) the correlation between the near-UV Tyr CD spectrum of single and double Cys-to-X mutants of the Imm protein and their in vivo activity; (iii) the upfield shift of methyl groups in a 1D NMR spectrum, a 2D- HSQC NMR spectrum of ImmE1 in the mixed polarity solvent mixture, and a broadening and disappearance of the indole (1)H proton resonance from Trp94 in H3 by a spin label attached to Cys16 in the H2 hydrophobic domain; (iv) near-UV circular dichroism spectra with a prominent ellipticity band centered at 290 nm from a single Trp inserted into the extended hydrophobic domains. It was concluded that the colicin E1 immunity protein adopts a folded conformation in chloroform/methanol/H(2)O (4:4:1) that is stabilized by helix-helix interactions. Analysis of the probable membrane folding topology indicated that several Tyr residues in the bilayer region of the three transmembrane helices could contribute to the near-UV CD spectrum through helix-helix interactions.

Structural changes linked to proton translocation by subunit c of the ATP synthase.

F1F0 ATP synthases use a transmembrane proton gradient to drive the synthesis of cellular ATP. The structure of the cytosolic F1 portion of the enzyme and the basic mechanism of ATP hydrolysis by F1 are now well established, but how proton translocation through the transmembrane F0 portion drives these catalytic changes is less clear. Here we describe the structural changes in the proton-translocating F0 subunit c that are induced by deprotonating the specific aspartic acid involved in proton transport. Conformational changes between the protonated and deprotonated forms of subunit c provide the structural basis for an explicit mechanism to explain coupling of proton translocation by F0 to the rotation of subunits within the core of F1. Rotation of these subunits within F1 causes the catalytic conformational changes in the active sites of F1 that result in ATP synthesis.

The 2.0 A structure of malarial purine phosphoribosyltransferase in complex with a transition-state analogue inhibitor.

Malaria is a leading cause of worldwide mortality from infectious disease. Plasmodium falciparum proliferation in human erythrocytes requires purine salvage by hypoxanthine-guanine-xanthine phosphoribosyltransferase (HGXPRTase). The enzyme is a target for the development of novel antimalarials. Design and synthesis of transition-state analogue inhibitors permitted cocrystallization with the malarial enzyme and refinement of the complex to 2.0 A resolution. Catalytic site contacts in the malarial enzyme are similar to those of human hypoxanthine-guanine phosphoribosyltransferase (HGPRTase) despite distinct substrate specificity. The crystal structure of malarial HGXPRTase with bound inhibitor, pyrophosphate, and two Mg(2+) ions reveals features unique to the transition-state analogue complex. Substrate-assisted catalysis occurs by ribooxocarbenium stabilization from the O5' lone pair and a pyrophosphate oxygen. A dissociative reaction coordinate path is implicated in which the primary reaction coordinate motion is the ribosyl C1' in motion between relatively immobile purine base and (Mg)(2)-pyrophosphate. Several short hydrogen bonds form in the complex of the enzyme and inhibitor. The proton NMR spectrum of the transition-state analogue complex of malarial HGXPRTase contains two downfield signals at 14.3 and 15.3 ppm. Despite the structural similarity to the human enzyme, the NMR spectra of the complexes reveal differences in hydrogen bonding between the transition-state analogue complexes of the human and malarial HG(X)PRTases. The X-ray crystal structures and NMR spectra reveal chemical and structural features that suggest a strategy for the design of malaria-specific transition-state inhibitors.

Transition-state analogs as inhibitors of human and malarial hypoxanthine-guanine phosphoribosyltransferases.

The proposed transition state for hypoxanthine-guanine phosphoribosyltransferases (HGPRTs) has been used to design and synthesize powerful inhibitors that contain features of the transition state. The iminoribitols (1S)-1-(9-deazahypoxanthin-9-yl)-1,4-dideoxy-1,4-imino-D-ribitol 5-phosphate (immucillinHP) and (1S)-1-(9-deazaguanin-9-yl)-1,4-dideoxy-1,4-imino-D-ribitol 5-phosphate (immucillinGP) are the most powerful inhibitors yet reported for both human and malarial HGPRTs. Equilibrium binding constants are >1,000-fold tighter than the binding of the nucleotide substrate. The NMR spectrum of malaria HGXPRT in the Michaelis complex reveals downfield hydrogen-bonded protons. The chemical shifts move farther downfield with bound inhibitor. The inhibitors are lead compounds for species-specific antibiotics against parasitic protozoa. The high-resolution crystal structure of human HGPRT with immucillinGP is reported in the companion paper.

Motional dynamics of the catalytic loop in OMP synthase.

In de novo pyrimidine biosynthesis, orotate phosphoribosyltransferase catalyzes the formation of orotidine 5'-monophosphate (OMP) from orotic acid and alpha-D-5-phosphoribosyl-1-pyrophosphate (PRPP). The known three-dimensional structure of the dimeric enzyme from Salmonella typhimurium is similar to that of other Type I phosphoribosyltransferases (nucleotide synthases) with a solvent-exposed active site atop a Rossman-type nucleotide binding fold. The three-dimensional structure of an enzyme-inhibitor complex [Henriksen et al. (1996) Biochemistry 35, 3803-3809] indicates that one of the two identical solvent-exposed loops can descend to cover the active site of the adjacent subunit of the dimeric enzyme. Catalytically essential residues are known to reside on this loop. In the present work, sensitivity toward limited proteolysis by trypsin confirms that the loop is solvent-exposed. Protection by PRPP and, to a lesser extent, by OMP demonstrates the existence of a second, trypsin-inaccessible, loop position. Two-dimensional 1H-15N NMR relaxation experiments on [alpha-15N]histidine-labeled WT OPRTase yielded backbone 15N T1 and T2 relaxation times and 15N[1H] NOE for His-105 (a loop residue) that are characteristic of small peptides. These results document that the surface loop is highly flexible in the unliganded enzyme. Addition of a hydrolytically stable PRPP analogue to the enzyme resulted in a significant reduction of His-105 peak intensity, indicating a dramatic change in the dynamic properties of the loop backbone in the analogue-ligated enzyme. 1H NMR titrations on histidine C2 protons, coupled with 1H and 31P titrations monitoring the C1H and 5-phosphate PRPP resonances, allowed the quantitation of the rates of loop movement during product release, and relate protein motion to enzymatic catalysis. These results suggest that loop opening and PRPP release is a two-step process, whose overall rate is partially rate-limiting in the reverse pyrophosphorolysis reaction.

Solution structure of the transmembrane H+-transporting subunit c of the F1F0 ATP synthase.

Subunit c is the H+-translocating component of the F1F0 ATP synthase complex. H+ transport is coupled to conformational changes that ultimately lead to ATP synthesis by the enzyme. The properties of the monomeric subunit in a single-phase solution of chloroform-methanol-water (4:4:1) have been shown to mimic those of the protein in the native complex. Triple resonance NMR experiments were used to determine the complete structure of monomeric subunit c in this solvent mixture. The structure of the protein was defined by >2000 interproton distances, 64 (3)JN alpha, and 43 hydrogen-bonding NMR-derived restraints. The root mean squared deviation for the backbone atoms of the two transmembrane helices was 0.63 A. The protein folds as a hairpin of two antiparallel helical segments, connected by a short structured loop. The conserved Arg41-Gln42-Pro43 form the top of this loop. The essential H+-transporting Asp61 residue is located at a slight break in the middle of the C-terminal helix, just prior to Pro64. The C-terminal helix changes direction by 30 +/- 5 degrees at the conserved Pro64. In its protonated form, the Asp61 lies in a cavity created by the absence of side chains at Gly23 and Gly27 in the N-terminal helix. The shape and charge distribution of the molecular surface of the monomeric protein suggest a packing arrangement for the oligomeric protein in the F0 complex, with the front face of one monomer packing favorably against the back face of a second monomer. The packing suggests that the proton (cation) binding site lies between packed pairs of adjacent subunit c.

Antagonist conformations with the beta(2)-adrenergic receptor ligand binding pocket.

The interactions between beta-adrenergic receptor (beta AR) antagonists and the beta(2)AR were studied with the use of photoaffinity labels. A proteolytic map of the receptor was made and confirmed through amino-terminal amino acid sequencing by locating sites of derivatization. [125I]Iodoazidothiophenylalprenolol (IAPTA) is a photoaffinity derivative of the beta AR antagonist alprenolol with a photoactivatable group on the aryloxy end of the molecule. IAPTA exclusively derivatizes a peptide consisting of transmembrane domains (TMs) 6 and 7 of the hamster lung beta(2)AR, supporting the contention that TMs 6 and 7 interact with the aryloxy portion of the beta AR antagonist pharmacophore. The beta AR antagonist photoaffinity labels [125I]iodoazidobenzylpindolol (IABP), [125I]iodoazidophenyl CGP-12177A (IAPCGP), and [125I]iodocyanopindololdiazarene (ICYPdz) are similar in that their photoactive moieties are attached to the amino end of the antagonist pharmacophore. IABP derivatized TMs 5-7 and a peptide containing TM 1 to approximately equal extents. IAPCGP derivatized Tms 6 and 7 >> TM 5 = TM 4 = TMs 2 and 3 = TM 1. ICYPdz derivatized TM 1 >> TMs 6 and 7 > Tm 4. We conclude that the aryloxy end of the beta AR antagonist pharmacophore is highly constrained within TMs 6 and 7, whereas the amino terminus is much less constrained and able to assume multiple conformations. Molecular dynamics simulations predict that IABP, IAPCGP, and ICYPdz favor a folded conformation, with both ends close together. Derivatization of TMs 6 and 7 by IABP, IAPCGP, and ICYPdz suggests the folded conformation of these compounds in the ligand binding pocket.

Determination of local protein structure by spin label difference 2D NMR: the region neighboring Asp61 of subunit c of the F1F0 ATP synthase.

Purified subunit c from the H(+)-transporting F1F0 ATP synthase of Escherichia coli folds as an antiparallel pair of extended helices in a solution of chloroform-methanol-water. A similar hairpin-like folding is predicted for the native protein in the multisubunit transmembrane Fo sector of the ATP synthase. A single Cys variant (A67C) of subunit c was created and modified with a maleimido-PROXYL [[3-(maleimidomethyl)-2,2,5,5-tetramethyl-1-pyrrolidinyl]oxy] spin label. Pairs of 1H 2D correlation and NOE spectra were collected with the nitroxide oxidized (paramagnetic) and reduced (diamagnetic). The pairs of spectra were subtracted, yielding difference spectra containing only cross-peaks from 1H within 15 A of the spin label. These greatly simplified spectra were easily analyzed to provide complete assignments for residues 10-25 and 52-79 of the protein, 150 NOE distance restraints, and 27 hydrogen-bonding restraints. The chemical shifts and NOE patterns observed in the derivatized mutant were virtually identical to those which were resolved in the unmodified wild-type protein, strongly suggesting that the spin label was not perturbing the protein structure. The restaints enabled us to calculate a detailed structure for this region of subunit c. The structure consisted of two gently curved helices, crossing at a slight (30 degrees) angle. The C-terminal helix was disrupted from Val60 to Ala62 near the essential Pro64. Asp61, the residue thought to undergo protonation--deprotonation with each H+ transported across the membrane, was in ver der Waals contact with Ala24. The proximity of these residues had been predicted from mutant analyses, where H+ translocation was retained on moving the Asp from position 61 to 24.

Hairpin folding of subunit c of F1Fo ATP synthase: 1H distance measurements to nitroxide-derivatized aspartyl-61.

Subunit c from the F1Fo ATP synthase of Escherichia coli folds in a hairpinlike structure of two alpha-helices in a solution of chloroform-methanol-H2O, and thus resembles the structure predicted for the folded protein in the membrane. The relevance of the structure in solution to the native structure was demonstrated. Asp61 in the second helical arm was shown to retain its unique reactivity with dicyclohexylcarbodiimide (DCCD) in chloroform-methanol-H2O solution. Further, the protein purified from the Ile28-->Thr DCCD-resistant mutant proved to be less reactive with DCCD in solution. This suggested that the protein folded with Ile28 of the first helical arm close to Asp61 in the second helical arm. Subunit c in wild-type E. coli membranes was specifically labeled with a nitroxide analog of DCCD (NCCD), and the derivative protein was purified. DQF COSY spectra were recorded, and the distances between the paramagnetic nitroxide and resolved protons in the spectra were calculated based upon paramagnetic broadening of the 1H resonances. The paramagnetic contribution to T2 relaxation in the NCCD-labeled sample was calculated by an iterative computer-fitting method, where a control spectrum of a phenylhydrazine-reduced sample was broadened until the line shape of one-dimensional slices through each COSY cross-peak maximally mimicked the line shape of the paramagnetic sample. The distances calculated from paramagnetic broadening indicate that Ala24 and Ala25 in helix-1 lie close (ca. 12 A) to the derivatized Asp61 in helix-2. A model for the interaction of helices in the NCCD-modified protein was generated by restrained molecular mechanics and molecular dynamics using 25 distances of < 10-20 A derived from paramagnetic broadening in combination with 15 long-range nuclear Overhauser enhancement (NOE) restraints (2-5 A) for distances between helices and the 89 intrahelical NOEs that defined helical structure in the DCCD-modified protein.

Helical structure and folding of subunit c of F1F0 ATP synthase: 1H NMR resonance assignments and NOE analysis.

Subunit c of the H(+)-transporting F1F0 ATP synthase (EC 3.6.1.34) is thought to fold across the membrane as a hairpin of two alpha-helices and function as a key component of the H(+)-translocase of F0. We report here the initial results of a structural study of purified subunit c in a chloroform-methanol-water (4:4:1) solvent mixture using standard two-dimensional NMR techniques. The spin systems of 78 of the 79 amino acid side chains have been assigned to residue type, and 44 of these have been assigned to specific residues in the sequence. Stretches of alpha-helical secondary structure were observed for Asp7-ILe26 in the first proposed transmembrane helix, and for Arg50-Ile55 and Ala67-Val78 in the second proposed transmembrane helix. Nuclear Overhauser effects (NOEs) were observed between residues at both ends of the predicted transmembrane helices. The intensities of the NOEs between helix-1 and helix-2 were not diminished by mixing of 2H-subunit c with 1H-subunit c, and therefore the NOEs must be due to intramolecular, rather than intermolecular, interactions. Hence the purified protein must fold as a hairpin in this solvent system, just as it is thought to fold in the lipid bilayer of the membrane. In native F0, dicyclohexylcarbodiimide reacts specifically with Asp61 in the second transmembrane helix of subunit c, and the rate of this reaction is reduced by substitution of Ile28 by Thr on the first transmembrane helix. The I28T substitution is shown here to alter the chemical shifts of protons at and around Asp61. This observation provides a further indication that subunit c may fold in chloroform-methanol-water solvent much like it does in the membrane.

On the question of interheme electron transfer in the chloroplast cytochrome b6 in situ.

The redox properties, the site of action of the inhibitor NQNO, and the question of interheme transfer in the chloroplast cytochrome b6 have been examined with regard to the role of the b6-f complex in quinol oxidation and H+ translocation. (i) The two hemes of the cytochrome ba and bp, have similar (delta Em less than or equal to 50 mV) oxidation-reduction midpoint potentials that are pH-independent in the range pH 6.5-8.0 (Em7 = -40 mV) but are pH dependent below this range with an estimated pK = 6.7. (ii) Only half of cytochrome b6, the stromal-side heme, ba, was reducible by NADPH and ferredoxin. (iii) The 2-3-fold increase (to 0.60 +/- 0.09 heme/600 Chl) in the amplitude of flash-induced cytochrome reduction caused by NQNO was not affected when heme ba was initially reduced, implying that NQNO affects flash reduction at the site of heme bp. (iv) Multiple light flashes did not increase the amplitude of b6 reduction in the presence or absence of NQNO or show binary oscillations. Together with localization of a site of action of NQNO near heme bp, these data provide no evidence for efficient electron transfer from heme bp to heme ba as specified by the Q cycle model. (v) NQNO interaction with heme bp does not block its oxidation, since reoxidation of the flash-reduced cytochrome in its presence or absence was 4-5 times faster (t1/2 approximately 30 ms) when heme ba was reduced. The faster oxidation of the photoreduced cytochrome after NADPH-Fd reduction of heme ba indicates that the oxidation of ba and bp may be cooperative.

Organization of the F0 sector of Escherichia coli H+-ATPase: the polar loop region of subunit c extends from the cytoplasmic face of the membrane.

The membrane-spanning F0 sector of the Escherichia coli H+-transporting ATP synthase (EC 3.6.1.34) contains multiple copies of subunit c, a 79 amino acid residue protein that is thought to insert in the membrane like a hairpin with two membrane traversing alpha-helices. The center of the protein is much more polar than the putative transmembrane alpha-helices and has been postulated to play a crucial role in coupling H+ translocation through F0 to ATP synthesis in the membrane extrinsic, F1 sector of the complex. However, the direction of insertion of subunit c in the membrane has not been established. We show here that the "polar loop" lies on the F1 binding side of the membrane. A peptide corresponding to Lys34----Ile46 of the polar loop was synthesized. Antisera were generated to the Lys34----Ile46 cognate peptide, and the polyclonal antipeptide IgG was shown to bind to a crude F0 fraction by using enzyme-linked immunosorbent assays. The antipeptide serum did not bind tightly enough to F0 to disrupt function. However, a polyclonal antiserum made to purified, whole subunit c was shown to block the binding of F1 to the F0 exposed in F1-stripped membranes. Incubation of the antisubunit c serum with the peptide reduced the inhibitory effect of the antiserum on the binding of F1 to F0. The reversal of inhibition by the peptide was specific to the antisubunit c serum in that the peptide had no effect on inhibition of F1 binding to F0 by antiserum to subunit a of F0.(ABSTRACT TRUNCATED AT 250 WORDS)

Location and mobility of ubiquinones of different chain lengths in artificial membrane vesicles.

Ubiquinone (UQn with n = 2, 3, or 10 isoprenoid groups) was incorporated into small, sonicated vesicles made of dipalmitoylphosphatidylcholine (DPPC) or dimyristoylphosphatidylcholine (DMPC). (1) The accessibility of oxidized UQ in DPPC or DMPC vesicles to the reductant sodium borohydride (NaBH4), measured by UV spectroscopy, was UQ2 greater than UQ3 greater than UQ10 (DPPC) and UQ2 greater than UQ3 approximately UQ10 (DMPC). (2) Catalysis of the reduction of entrapped ferricyanide by exogenous NaBH4 was more effective with UQ2 than UQ10 but was slower with all quinones than reduction by added dithionite. (3) The methoxy protons of UQ2 and UQ3 in DPPC and DMPC vesicles exhibited a single NMR resonance centered at approximately 3.95 ppm, whereas the methoxy groups of UQ10 gave rise to two separate proton resonances, at 3.93 ppm and a more narrow resonance at 3.78 ppm. The UQ10 population characterized by the 3.78 ppm resonance was present at a higher concentration in DPPC than in DMPC vesicles and was relatively insensitive to reduction by NaBH4. (4) UQ10 perturbed the melting temperature (Tm) of DPPC vesicles to a smaller extent (delta Tm = -1 degrees C) than did UQ2 and UQ3 (delta Tm = -3 to -4 degrees C). The combined UV and NMR data imply the following: The UQ10 pool characterized by the 3.78 ppm peak corresponds to a more mobile UQ10 fraction that is not reduced by NaBH4 in 2-3 min and is thought to be localized close to the center of the DPPC bilayer since it has little effect on the DPPC Tm.(ABSTRACT TRUNCATED AT 250 WORDS)

A redox study of the electron transport pathway responsible for generation of the slow electrochromic phase in chloroplasts.

The amplitude of the slow phase of the electrochromic bandshift and the dark redox state of cytochrome b6, as well as its flash-induced turnover, have been measured as a function of ambient redox potential between +200 and -200 mV. Formation of a quinol-like donor with an Em,7 = +100 +/- 10 mV is required for generation of the slow phase. 80-100% of the amplitude of this signal with a t 1/2 = 3-4 ms is observed at -200 mV where cytochrome b6 was almost fully reduced (Em,7 of dark and flash-induced photoreduction was -30 mV and -75 mV, respectively). The change in the photoreduction of cytochrome b6 above 0 mV had an Em,7 of +50 mV, about 50 mV more negative than the midpoint at this pH for the onset of the slow electrochromic change. At potentials below -140 mV the amplitude of b6 photoreduction becomes small or negligible. The nature of the cytochrome b6 photoresponse is changed at potentials below -140 mV from a net photoreduction with a t1/2 = approximately less than 1 ms to a photooxidation with a t1/2 = 15-20 ms that is substantially slower than the electrochromic band-shift with a t1/2 = 3-4 ms. It is concluded that the slow electrochromic phase probably does not arise from a mechanism involving a turnover of cytochrome b6. From consideration of the possible flash-induced electron-transfer steps and alternative mechanisms for generation of the slow phase, it is suggested that it may arise from a redox-linked H+ pump involving the high potential iron-sulfur protein.

Alternative Respiratory Pathway: ITS ROLE IN SEED RESPIRATION AND ITS INHIBITION BY PROPYL GALLATE.

Oxygen uptake during the first hours of imbibition in intact soybean and mung bean seeds showed a marked sensitivity to potassium cyanide but was unaffected by addition of either salicylhydroxamic acid or propyl gallate. However O(2) uptake by finely ground seed particles was very sensitive to the addition of either compound. The results indicated that O(2) uptake in intact, imbibing seeds was associated with a cyanide-sensitive process, most probably mitochondrial mediated respiration, and not the result of the cyanide-insensitive lipoxygenase activity which was readily detectable in ground seed particles.The antioxidant propyl gallate was found to inhibit specifically alternative pathway electron transfer in isolated mung bean mitochondria. Half-maximal inhibition occurred with 2 to 5 micromolar propyl gallate. Kinetic analysis indicated that propyl gallate inhibition of the alternative pathway occurred at, or very near, the site of inhibition of the alternative pathway by salicylhydroxamic acid.A high level of lipoxygenase activity was found to be associated with washed mitochondria isolated from a variety of etiolated plant tissues. Most of this lipoxygenase activity could be eliminated from mung bean mitochondria if the mitochondria were purified on a discontinuous sucrose gradient. This indicated that the mitochondrial-associated activity was probably the result of nonspecific adsorption of lipoxygenase onto the mitochondrial membranes during isolation.