Interaction of the Precursor to Mitochondrial Aspartate Aminotransferase and Its Presequence Peptide with Model Membranes
2000; Elsevier BV; Volume: 275; Issue: 44 Linguagem: Inglês
10.1074/jbc.m004494200
ISSN1083-351X
AutoresFernando Doñate, Alejandro J. Yáñez, Ana Iriarte, Marino Martinez‐Carrion,
Tópico(s)Metabolism and Genetic Disorders
ResumoThe possible contribution of the mature portion of a mitochondrial precursor protein to its interaction with membrane lipids is unclear. To address this issue, we examined the interaction of the precursor to mitochondrial aspartate aminotransferase (pmAAT) and of a synthetic peptide corresponding to the 29-residue presequence peptide (mAAT-pp) with anionic phospholipid vesicles. The affinity of mAAT-pp and pmAAT for anionic vesicles is nearly identical. Results obtained by analyzing the effect of mAAT-pp or full-length pmAAT on either the permeability or microviscosity of the phospholipid vesicles are consistent with only a shallow insertion of the presequence peptide in the bilayer. Analysis of the quenching of Trp-17 fluorescence by brominated phospholipids reveals that this presequence residue inserts to a depth of approximately 9 Å from the center of the bilayer. Furthermore, in membrane-bound pmAAT or mAAT-pp, both Arg-8 and Arg-28 are accessible to the solvent. These results suggest that the presequence segment lies close to the surface of the membrane and that the mature portion of the precursor protein has little effect on the affinity or mode of binding of the presequence to model membranes. In the presence of vesicles, mAAT-pp adopts considerable α-helical structure. Hydrolysis by trypsin after Arg-8 results in the dissociation of the remaining 21-residue C-terminal peptide fragment from the membrane bilayer, suggesting that the N-terminal portion of the presequence is essential for membrane binding. Based on these results, we propose that the presequence peptide may contain dual recognition elements for both the lipid and import receptor components of the mitochondrial membrane. The possible contribution of the mature portion of a mitochondrial precursor protein to its interaction with membrane lipids is unclear. To address this issue, we examined the interaction of the precursor to mitochondrial aspartate aminotransferase (pmAAT) and of a synthetic peptide corresponding to the 29-residue presequence peptide (mAAT-pp) with anionic phospholipid vesicles. The affinity of mAAT-pp and pmAAT for anionic vesicles is nearly identical. Results obtained by analyzing the effect of mAAT-pp or full-length pmAAT on either the permeability or microviscosity of the phospholipid vesicles are consistent with only a shallow insertion of the presequence peptide in the bilayer. Analysis of the quenching of Trp-17 fluorescence by brominated phospholipids reveals that this presequence residue inserts to a depth of approximately 9 Å from the center of the bilayer. Furthermore, in membrane-bound pmAAT or mAAT-pp, both Arg-8 and Arg-28 are accessible to the solvent. These results suggest that the presequence segment lies close to the surface of the membrane and that the mature portion of the precursor protein has little effect on the affinity or mode of binding of the presequence to model membranes. In the presence of vesicles, mAAT-pp adopts considerable α-helical structure. Hydrolysis by trypsin after Arg-8 results in the dissociation of the remaining 21-residue C-terminal peptide fragment from the membrane bilayer, suggesting that the N-terminal portion of the presequence is essential for membrane binding. Based on these results, we propose that the presequence peptide may contain dual recognition elements for both the lipid and import receptor components of the mitochondrial membrane. cytosolic aspartate aminotransferase precursor of mitochondrial aspartate aminotransferase mitochondrial aspartate aminotransferase mitochondrial aspartate aminotransferase presequence peptide diphenylhexatriene 6-carboxyfluorescein 8-aminonaphthalene-1,3,6-trisulfonic acid, disodium salt p-xylenebis(pyridinium bromide) small unilamellar vesicle large unilamellar vesicle palmitoyloleylphosphatidylcholine palmitoyloleylphosphatidylglycerol phosphatidylglycerol dimiristoylphosphatidylcholine dimiristoylphosphatidylglycerol palmitoylstearoylphosphatidylcholine dibromine palmitoylstearoylphosphatidylcholine alcohol dehydrogenase The majority of mitochondrial proteins are encoded by nuclear DNA and synthesized in the cytoplasm as precursors containing an N-terminal extension peptide called presequence or signal sequence that targets the passenger protein to mitochondria (for review, see Ref. 1Neupert W. Annu. Rev. Biochem. 1997; 66: 863-917Crossref PubMed Scopus (981) Google Scholar). Following translocation into mitochondria, the presequence is proteolytically removed to render the mature form of the protein. In addition to the presequence, a partially folded conformation is also required for efficient import into mitochondria. Molecular chaperones present in the cytosol of eukaryotic cells are involved in maintaining proteins to be translocated in an “import-competent” state (2Deshaies R.J. Koch B.D. Werner W.M Craig E.A. Schekman R. Nature. 1988; 332: 800-805Crossref PubMed Scopus (1005) Google Scholar, 3Chirico W.J. Waters M.G. Blobel G. Nature. 1988; 322: 805-810Crossref Scopus (840) Google Scholar, 4Artigues A. Iriarte A. Martinez-Carrion M. J. Biol. Chem. 1997; 272: 16852-16861Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar). Mitochondrial matrix presequences lack a consensus sequence (5von Heijne G. EMBO J. 1986; 5: 1335-1342Crossref PubMed Scopus (712) Google Scholar), suggesting that secondary structure elements may contribute to function. Presequences are 20–60 residues long and lack ordered secondary structure in aqueous solution (1Neupert W. Annu. Rev. Biochem. 1997; 66: 863-917Crossref PubMed Scopus (981) Google Scholar, 6Tamm L.K. Biochim. Biophys. Acta. 1991; 1071: 123-148Crossref PubMed Scopus (108) Google Scholar, 7Roise D. Horvath S.J. Tomich J.M. Richards J.H. Schatz G. EMBO J. 1986; 5: 1327-1334Crossref PubMed Scopus (314) Google Scholar, 8Karslake C. Piotto M.E. Pak Y.K. Weiner H. Gorenstein D.G. Biochemistry. 1990; 29: 9872-9878Crossref PubMed Scopus (85) Google Scholar, 9Endo T. Oya M. FEBS Lett. 1989; 249: 173-177Crossref PubMed Scopus (7) Google Scholar, 10Chupin V. Leenhouts J.M. de Kroon A.I. de Kruijff B. Biochemistry. 1996; 35: 3141-3146Crossref PubMed Scopus (33) Google Scholar, 11Hammen P.K. Gorenstein D.G. Weiner H. Biochemistry. 1996; 35: 3772-3781Crossref PubMed Scopus (38) Google Scholar, 12Hammen P.K. Gorenstein D.G. Weiner H. Biochemistry. 1994; 33: 8610-8617Crossref PubMed Scopus (69) Google Scholar). Their sequences are characterized by the presence of positively charged and hydroxylated residues (5von Heijne G. EMBO J. 1986; 5: 1335-1342Crossref PubMed Scopus (712) Google Scholar, 6Tamm L.K. Biochim. Biophys. Acta. 1991; 1071: 123-148Crossref PubMed Scopus (108) Google Scholar). A variety of mitochondrial presequence peptides have been shown to bind to lipid vesicles containing anionic phospholipids as amphipathic helical structures (5von Heijne G. EMBO J. 1986; 5: 1335-1342Crossref PubMed Scopus (712) Google Scholar, 6Tamm L.K. Biochim. Biophys. Acta. 1991; 1071: 123-148Crossref PubMed Scopus (108) Google Scholar, 7Roise D. Horvath S.J. Tomich J.M. Richards J.H. Schatz G. EMBO J. 1986; 5: 1327-1334Crossref PubMed Scopus (314) Google Scholar, 8Karslake C. Piotto M.E. Pak Y.K. Weiner H. Gorenstein D.G. Biochemistry. 1990; 29: 9872-9878Crossref PubMed Scopus (85) Google Scholar, 9Endo T. Oya M. FEBS Lett. 1989; 249: 173-177Crossref PubMed Scopus (7) Google Scholar, 10Chupin V. Leenhouts J.M. de Kroon A.I. de Kruijff B. Biochemistry. 1996; 35: 3141-3146Crossref PubMed Scopus (33) Google Scholar, 11Hammen P.K. Gorenstein D.G. Weiner H. Biochemistry. 1996; 35: 3772-3781Crossref PubMed Scopus (38) Google Scholar, 12Hammen P.K. Gorenstein D.G. Weiner H. Biochemistry. 1994; 33: 8610-8617Crossref PubMed Scopus (69) Google Scholar). The presence of positive charges and an amphipathic helical character of the presequence have been shown to be essential for import of most proteins into mitochondria (1Neupert W. Annu. Rev. Biochem. 1997; 66: 863-917Crossref PubMed Scopus (981) Google Scholar, 6Tamm L.K. Biochim. Biophys. Acta. 1991; 1071: 123-148Crossref PubMed Scopus (108) Google Scholar, 7Roise D. Horvath S.J. Tomich J.M. Richards J.H. Schatz G. EMBO J. 1986; 5: 1327-1334Crossref PubMed Scopus (314) Google Scholar, 8Karslake C. Piotto M.E. Pak Y.K. Weiner H. Gorenstein D.G. Biochemistry. 1990; 29: 9872-9878Crossref PubMed Scopus (85) Google Scholar, 9Endo T. Oya M. FEBS Lett. 1989; 249: 173-177Crossref PubMed Scopus (7) Google Scholar, 10Chupin V. Leenhouts J.M. de Kroon A.I. de Kruijff B. Biochemistry. 1996; 35: 3141-3146Crossref PubMed Scopus (33) Google Scholar, 11Hammen P.K. Gorenstein D.G. Weiner H. Biochemistry. 1996; 35: 3772-3781Crossref PubMed Scopus (38) Google Scholar, 12Hammen P.K. Gorenstein D.G. Weiner H. Biochemistry. 1994; 33: 8610-8617Crossref PubMed Scopus (69) Google Scholar). The molecular bases for the recognition and binding of presequences to the mitochondrial surface are controversial. It has been hypothesized that the targeting sites for import may involve the lipid phase of the mitochondrial membrane (6Tamm L.K. Biochim. Biophys. Acta. 1991; 1071: 123-148Crossref PubMed Scopus (108) Google Scholar, 7Roise D. Horvath S.J. Tomich J.M. Richards J.H. Schatz G. EMBO J. 1986; 5: 1327-1334Crossref PubMed Scopus (314) Google Scholar, 8Karslake C. Piotto M.E. Pak Y.K. Weiner H. Gorenstein D.G. Biochemistry. 1990; 29: 9872-9878Crossref PubMed Scopus (85) Google Scholar). Arguments for this view include the high affinity of presequences for anionic phospholipids, the induced changes in secondary structure required for import, and the lack of a consensus sequence. On the other hand, a multisubunit protein receptor complex has been identified in the outer mitochondrial membrane (13Sollner T. Griffiths G. Pfaller R. Pfanner N. Neupert W. Cell. 1989; 59: 1061-1070Abstract Full Text PDF PubMed Scopus (247) Google Scholar). This import receptor complex recognizes positively charged presequences that form α-helices (14Schleiff E. Heard T.S. Weiner H. FEBS Lett. 1999; 461: 9-12Crossref PubMed Scopus (14) Google Scholar). However, it has been suggested recently that Tom20, a component of the import receptor complex, preferentially recognizes presequences while membrane-bound (14Schleiff E. Heard T.S. Weiner H. FEBS Lett. 1999; 461: 9-12Crossref PubMed Scopus (14) Google Scholar, 15Schleiff E. Turnbull J.L. Biochemistry. 1998; 37: 13052-13058Crossref PubMed Scopus (23) Google Scholar). This implies that binding to anionic phospholipids in the outer membrane and subsequent induced change in secondary structure could precede recognition by the protein import complex. Although the interaction of chemically synthesized presequences with membrane-like environments has been extensively studied, very little is known of the interaction of natural mitochondrial precursor proteins with membranes (16Iriarte A. Altieri F. del Solar J. Mattingly J.R. Martinez-Carrion M. Fukui T. Kagamiyama H. Soda K. Wada H. Enzymes Dependent on Pyridoxal Phosphate and Other Carbonyl Compounds as Cofactors. Pergamon Press, New York1990: 527-529Google Scholar, 17Berezov A. Iriarte A. Martinez-Carrion M. J. Biol. Chem. 1994; 269: 22222-22229Abstract Full Text PDF PubMed Google Scholar, 18Berezov A. Iriarte A. Martinez-Carrion M. Arch. Biochem Biophys. 1996; 336: 173-183Crossref PubMed Scopus (2) Google Scholar). The hydrophobicity of the mature protein, its isoelectric point, or possible effects on the conformation of the presequence by the mature form may affect the binding of the precursor to lipids. For instance, it has been suggested that the passenger protein of an artificial chimeric protein may affect the secondary structure of the signal sequence attached to it, abolishing importin vitro (19Waltner M. Hammen P.K. Weiner H.J. J. Biol. Chem. 1996; 271: 21226-21230Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar). Thus, studies using intact authentic mitochondrial precursors are needed to further understand the interaction of targeting presequences with lipid membranes. The dimeric precursor of mitochondrial aspartate aminotransferase has been purified from bacteria in an active form (20Altieri F. Mattingly J.R. Rodriguez-Berrocal F.J. Youssef J. Iriarte A. Wu T. Martinez-Carrion M. J. Biol. Chem. 1989; 264: 4782-4786Abstract Full Text PDF PubMed Google Scholar), and represents an example of such a natural mitochondrial precursor. Aspartate aminotransferase exists in animal cells as two isoforms located in the cytosol (cAAT)1 and the matrix of mitochondria (mAAT). The mitochondrial form is synthesized in the cytoplasm as a precursor (pmAAT) with a 29-residue presequence at its N-terminal end that targets the protein to mitochondria (21Mattingly Jr., J.R. Youssef J. Iriarte A. Martinez-Carrion M. J. Biol. Chem. 1993; 268: 3925-3937Abstract Full Text PDF PubMed Google Scholar, 22Nishi T. Nagashima F. Tanase S. Fukumoto Y. Joh T. Shimada K. Matsukado Y. Ushio Y. Morino Y. J. Biol. Chem. 1989; 264: 6044-6051Abstract Full Text PDF PubMed Google Scholar). Although pmAAT is efficiently imported into mitochondria in vitro, a chimeric protein composed of the pmAAT presequence attached to cAAT was not imported, suggesting that the passenger protein can affect import into mitochondria (23Lain B. Yañez A. Iriarte A. Martinez-Carrion M. J. Biol. Chem. 1998; 273: 4406-4415Abstract Full Text Full Text PDF PubMed Scopus (11) Google Scholar).The pmAAT presequence peptide shares the same fundamental characteristics of other matrix-targeting sequences described above (see amino acid sequence in Fig. 3). The binding of pmAAT to liposomes is dependent on the presence of negatively charged phospholipids such as phosphatidylglycerol (PG) and cardiolipin (16Iriarte A. Altieri F. del Solar J. Mattingly J.R. Martinez-Carrion M. Fukui T. Kagamiyama H. Soda K. Wada H. Enzymes Dependent on Pyridoxal Phosphate and Other Carbonyl Compounds as Cofactors. Pergamon Press, New York1990: 527-529Google Scholar). The mature protein, lacking the 29-residue presequence peptide, is unable to bind to these vesicles under similar conditions, showing that the binding of pmAAT is dependent on the presence of the presequence peptide. However, at low ionic strength, the mature protein can bind to negatively charged vesicles because of its basic character (16Iriarte A. Altieri F. del Solar J. Mattingly J.R. Martinez-Carrion M. Fukui T. Kagamiyama H. Soda K. Wada H. Enzymes Dependent on Pyridoxal Phosphate and Other Carbonyl Compounds as Cofactors. Pergamon Press, New York1990: 527-529Google Scholar, 17Berezov A. Iriarte A. Martinez-Carrion M. J. Biol. Chem. 1994; 269: 22222-22229Abstract Full Text PDF PubMed Google Scholar, 18Berezov A. Iriarte A. Martinez-Carrion M. Arch. Biochem Biophys. 1996; 336: 173-183Crossref PubMed Scopus (2) Google Scholar). In this work we analyze the interaction of the pmAAT presequence with model membranes containing negatively charged phospholipids. Several pertinent questions are addressed, such as the influence of the mature portion of the pmAAT protein on the binding of the presequence to lipid vesicles, the overall conformation of the presequence region when bound to phospholipid vesicles, and elements of the topology of the presequence in the membrane and its depth of insertion. We have used both a synthetic peptide corresponding to the rat liver pmAAT presequence (mAAT-pp) as well as the full-length pmAAT precursor purified as a recombinant protein. As a fluorescence-reporting group, we introduced a tryptophan residue at position 17 in the presequence, (F17W mutant), either during chemical synthesis of the mAAT-pp peptide or by site directed mutagenesis of pmAAT. Proteolytic access to unique arginine residues within the presequence was also used to assess the topology of the presequence while bound to model membranes. pmAAT was expressed and purified fromEscherichia coli as described previously (20Altieri F. Mattingly J.R. Rodriguez-Berrocal F.J. Youssef J. Iriarte A. Wu T. Martinez-Carrion M. J. Biol. Chem. 1989; 264: 4782-4786Abstract Full Text PDF PubMed Google Scholar). mAAT was obtained by tryptic treatment of the precursor protein (trypsin:pmAAT molar ratio, 1:100) and a subsequent CM-Sepharose chromatography step (20Altieri F. Mattingly J.R. Rodriguez-Berrocal F.J. Youssef J. Iriarte A. Wu T. Martinez-Carrion M. J. Biol. Chem. 1989; 264: 4782-4786Abstract Full Text PDF PubMed Google Scholar). The mutant F17W-pmAAT was prepared by digesting the cDNA for rat liver pmAAT previously cloned in Bluescript KS (pBSKS-4) (21Mattingly Jr., J.R. Youssef J. Iriarte A. Martinez-Carrion M. J. Biol. Chem. 1993; 268: 3925-3937Abstract Full Text PDF PubMed Google Scholar) withEcoRI and BamHI and subcloning into the pALTER vector from Promega. Mutagenesis was performed following the manufacturer's instructions. The wild type and F17W-mutant mAAT-pp peptides were synthesized at the Protein Core Facility of the School of Biological Sciences and purified by reverse phase high performance liquid chromatography. Lipids were purchased from Avanti Polar Lipids (Birmingham, AL). 6-Carboxyfluorescein (CF), diphenylhexatriene (DPH), 8-aminonaphthalene-1,3,6-trisulfonic acid, disodium salt (ANTS), andp-xylenebis(pyridinium bromide) (DPX) were from Molecular Probes (Eugene, OR). All other regents were of the highest purity available. When appropriate, non-linear least-squares fitting of the data was performed using Sigma Plot 4.01 (Jandel Corp.). Small unilamellar vesicles (SUVs) were prepared by sonication with a microprobe (Heat Systems Ultrasonics, Inc.) of a suspension of lipids in the appropriate buffer for a minimum of 6 min, or until the turbidity had cleared, while maintained under an argon atmosphere. The vesicles were then centrifuged briefly in an Eppendorf centrifuge to remove titanium particles. Large unilamellar vesicles (LUVs) were prepared by the reverse phase evaporation method (24Szoka F. Papahadjopoulos D. Proc. Natl. Acad. Sci. U. S. A. 1978; 75: 4194-4198Crossref PubMed Scopus (2373) Google Scholar). Briefly, 10 μmol of lipids in chloroform were mixed to give the appropriate composition. The solution was dried under a stream of argon and dissolved in 1 ml of ether, followed by the addition of 300 μl of lipid buffer (10 mmHEPES, 0.1 mm EDTA, 100 mm NaCl, pH 7.5). The suspension was sonicated for 10 min in a bath sonicator to create a stable emulsion. The remaining ether phase was evaporated at low pressure in a rotavapor until a stable gel was formed. The gel was collapsed by a brief vortexing, and 700 μl of lipid buffer were then added. The ether phase was further evaporated for 90 min under high vacuum. The sample was passed several times through two stacked 0.2- and 0.4-μm filters (Nucleopore) to obtain a homogeneous population of liposomes and to remove multilamellar vesicles. The lipid concentration of the suspension was determined by phosphorous analysis (25Barlet G.R. J. Biol. Chem. 1959; 234: 466-468Abstract Full Text PDF PubMed Google Scholar) after hydrolysis for 30 min in 70% perchloric acid at 200 °C. Peptide-lipid association was studied by monitoring the changes in the tryptophan fluorescence of F17W-mAAT-pp or F17W-pmAAT upon addition of SUVs. Fluorescence spectra were recorded at 25 °C using an SLM 8100C spectrofluorometer set on photon counting, with excitation at 296 nm and 4-nm slits in the excitation and emission pathways. When vesicles were present, spectra were corrected for the fluorescence of vesicles alone. To determine the degree of association of mAAT-pp and pmAAT with lipid vesicles, small aliquots of one of the components (peptide or vesicles) were successively added to a fixed amount of the other component in 1.4 ml of lipid buffer. The fluorescence intensity of the mixture was recorded at 338 nm (λex = 296 nm) 5 min after mixing the vesicles and peptide/protein. The contents of the cuvette were continuously stirred. The increase in fluorescence was calculated after correction for the change in volume caused by a given addition. The binding of peptide to lipid bilayers can be described in terms of a simple bimolecular equilibrium model, which assumes that a distinct number of lipid monomers in a membrane comprise an individual peptide binding site.P+nL⇌P:LnEquation 1 P represents the peptide/protein, Lrepresents the phospholipid, and n is the number of lipid molecules that provide one binding site for the peptide. The binding can be described in terms of a dissociation constant of the lipid/peptide complex, K d. For the lipid titration experiments, the total concentration of lipid, [Lt ], is much higher than the concentration of peptide bound to the membrane, [P:Ln ] or [Pb ], as shown in Equation 2.Kd=[P][Lt/n]/[Pb]Equation 2 [P] is the equilibrium concentration of free peptide in solution. According to the law of mass conservation (total concentration of peptide, [Pt ] = [Pb ] + [P]), Equation 2 can be transformed into Equation 3.[Pb]=[Pt][Lt]/(Kd+[Lt]/n)Equation 3 The fluorescence intensity values at saturating concentrations of SUVs are an indication of maximum partitioning of peptide in the membrane. Thus, the concentration of bound peptide can be calculated using Equation 4.[Pb]=((F−F0)/(F∞−F0))[Pt]Equation 4 F 0, F, andF ∞ represent the fluorescence intensity in the absence of vesicles, at a given amount of vesicles, and in the presence of saturating amounts of vesicles, respectively. From Equation 4 and the law of mass action, we obtain Equation 5.F/F0=1+((Fmax/F0)−1)[Lt]/(nKd+[Lt])Equation 5 From a plot of F/F0 versus[Lt ] (Fig. 1 A), the value ofnK d was obtained from a fit of Equation 5 to the data. This parameter must be regarded as an apparent dissociation constant (K app), which provides an index of the affinity of the peptide for the membrane. Note that when (F/F 0) − 1 = ((F max/F 0) − 1)/2,nK d = [L], the concentration of lipid required to attain binding of 50% of the available peptide. For the peptide titration experiments, the increase in fluorescence at 338 nm was measured upon addition of small aliquots of mAAT-pp to a fixed concentration of SUVs. In the peptide titration experiments, we must also consider the conservation relationship for the lipid, [L/n] = [Lt /n] − [P:Ln ] in addition to the mass conservation of the peptide indicated before. L/nis the concentration of unoccupied binding sites, andLt /n is the total amount of lipid binding sites. Substituting for free peptide and free lipid in Equation2 and solving for bound peptide yields the quadratic equation shown in Equation 6.[Pb]=Equation 6 (Kd+[Pt]+[Lt/n]±((Kd+[Pt]+[Lt/n]) 2−4[Pt][Lt/n]) 1/2)/2In this equation, only the subtraction gives meaningful values. In order to calculate [Pb ] using Equation 4, we extrapolated F ∞ from a double-reciprocal plot of the total peptide fluorescence versus total concentration of lipid obtained in a separate experiment. CD data were collected in a Jasco spectropolarimeter model J-720 using the software provided by the manufacturer for collection and analysis of the data. All the buffers were filtered through a 0.2-μm filter to eliminate dust and other particles, and extensively degassed under vacuum. The CD spectra were recorded using a 0.1-cm cuvette at room temperature. Either the lipid buffer or a more far-UV transparent buffer (10 mm potassium phosphate, 10 mm KF, pH 7.5) was used. The spectra were corrected for the buffer contribution by subtraction of an appropriate blank and for the changes in volume due to the addition of the lipid suspension. The spectrum labeled 4 in Fig. 2 was truncated below 200 nm because the noise caused by turbidity precluded the recording of data at lower wavelengths. Molar ellipticity values were calculated according to the following expression: [θ] = (θ/10) (114/lc), where θ is the ellipticity in millidegrees, 114 is the mean residue molecular weight in g/mol, l is the path length in cm, and c is the concentration in g/cm3. [θ] has the units of degree cm2dmol−1. The factor 10 in the previous equation originates from the conversion of mol−1 to dmol−1. LUVs (POPC:POPG, 1:1) containing CF were prepared essentially according to the method described above with the following modifications. The lipid buffer was substituted with a solution containing 50 mm CF in 10 mm HEPES, pH 7.5. After extrusion, free CF was removed by gel filtration on a Sephadex G-75 column (1 × 20 cm) equilibrated in lipid buffer. The content of CF was measured by absorbance (ε492 nm =72,000 m−1) before disruption of the liposomes with 0.1% Triton X-100. The release of carboxyfluorescein from the liposomes as induced by interaction with pmAAT or mAAT-pp was followed at room temperature by monitoring the increase in fluorescence at 540 nm after excitation at 430 nm. Release of vesicular contents was also monitored by the ANTS/DPX assay (26Ellens R.M. Bentz J. Szoka F.C. Biochemistry. 1985; 24: 3099-3106Crossref PubMed Scopus (448) Google Scholar). LUVs containing 12.5 mm ANTS, 45 mm DPX, 10 mm Hepes were obtained by encapsulating this mixture in 50% PG vesicles as described above. Nonencapsulated material was removed by gel filtration on a Sephadex G-75 column. Dilution of the probes upon release of the vesicle contents would result in a relief of the quenching of the ANTS fluorescence by DPX. Fluorescence measurements were performed at 520 nm after excitation at 355 nm. In each case, the zero level of leakage corresponded to the fluorescence of vesicles alone at time 0 and 100% leakage was the maximum fluorescence intensity obtained after complete release of the probes by treatment of the vesicles with 0.1% Triton X-100. Internal and external osmolarities were adjusted with NaCl. To decrease light scattering due to the presence of LUVs, a filter (3-68) was placed between the sample and the emission monochromator. To measure changes in DPH anisotropy at a constant temperature upon interaction with mAAT-pp or pmAAT, DMPC:DMPG (1:1) SUVs, prepared as described before, were diluted to a lipid concentration of 200 μm in lipid buffer and 0.5 μl of a solution of DPH in acetonitrile was added (DPH:lipid molar ratio, 1:300). After equilibration for 30 min at room temperature, increasing concentrations of peptide or protein were added to 1.2 ml of vesicles in a 3-ml cuvette. The excitation and emission wavelengths were 360 and 490 nm, respectively. The temperature of the cell compartment in the fluorometer was maintained at 38 °C by a Haake F3 water bath and monitored by a Digi-Sense temperature control with a microprobe inside the cuvette. This temperature is above the melting temperature of the SUVs used (23 °C) as reflected in the low value of r(≈0.070) obtained for vesicles alone in the absence of peptide. The anisotropy was measured in a T-format, and it was automatically calculated using the software provided by the manufacturer according to the expression: r = (I vv −I vh)/(I vv − 2I vh), where r is the anisotropy of the sample, I vv is the fluorescence detected when the excitation and emission polarizers are set at 90°, andI vh is the fluorescence of the sample detected when the excitation polarizer is set at 90° and the emission polarizer is set at 0°. To minimize light scattering, filters (3-75) were placed between the sample and the two detectors. The samples were kept under constant stirring. The F17W mutant of pmAAT or mAAT-pp containing a tryptophan residue at position 17 of the presequence peptide was used in these experiments. The fluorescence intensity of the peptide or protein was measured at 388 nm (λex = 296 nm) in samples prepared either in lipid buffer alone, or lipid buffer containing unlabeled SUVs (20% POPC, 50% POPG, and 30% PSPC) or SUVs of similar composition but containing 30% brominated PSPC (BrPSPC) instead of PSPC. BrPSPC is commercially available brominated at either positions 6 and 7 or 11 and 12 of the stearoyl acyl chains. Vesicles were prepared by sonication as described before. The quenching efficiency was defined as (ΔF 0 − ΔF) × 100/ΔF 0, where ΔF 0and ΔF are the increase in fluorescence upon binding of peptide or protein to unlabeled vesicles and to liposomes containing brominated phospholipids, respectively. To calculate the depth of insertion using the “parallax” analysis (27Ladokhin A.S. Biophys. J. 1999; 76: 946-955Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar), the following equation was used. ZCF=Lc1+(−1/πC)lnF1/F2−L2122L21Equation 7 Z CF is the depth of the fluorophore from the center of the bilayer, L c1 is the distance from the shallow quencher to the center of the bilayer,L 21 is the distance between both quenchers,F 1 and F 2 are the fluorescence intensities of the fluorophore when incubated with membranes containing the shallow and deep quencher, respectively. Finally, C represents the two-dimensional quencher concentration in the plane of the membrane in mole fraction of quencher lipid in total lipid per unit area. The values for the constants wereL c1 = 10.8 Å, L 21 = 4.5 Å according to x-ray diffraction data (28McIntosh T.J. Holloway P.W. Biochemistry. 1987; 26: 1783-1788Crossref PubMed Scopus (165) Google Scholar); C = (0.3)/70 Å2, the average surface area per lipid molecule (29Lewis B.A. Engelman D. J. Mol. Biol. 1983; 166: 211-216Crossref PubMed Scopus (689) Google Scholar). The intact precursor of mitochondrial aspartate transferase binds to vesicles containing negatively charged phospholipids (16Iriarte A. Altieri F. del Solar J. Mattingly J.R. Martinez-Carrion M. Fukui T. Kagamiyama H. Soda K. Wada H. Enzymes Dependent on Pyridoxal Phosphate and Other Carbonyl Compounds as Cofactors. Pergamon Press, New York1990: 527-529Google Scholar). Cardiolipin-containing vesicles showed the highest affinity of all the acidic phospholipids tested. However, similar levels of binding
Referência(s)