Structural Changes of Region 1-16 of the Alzheimer Disease Amyloid β-Peptide upon Zinc Binding and in Vitro Aging
2005; Elsevier BV; Volume: 281; Issue: 4 Linguagem: Inglês
10.1074/jbc.m504454200
ISSN1083-351X
AutoresSéverine Zirah, Sergey A. Kozin, Alexey K. Mazur, Alain Blond, M. Cheminant, Isabelle Ségalas-Milazzo, Pascale Debey, Sylvie Rebuffat,
Tópico(s)Trace Elements in Health
ResumoAmyloid deposits within the cerebral tissue constitute a characteristic lesion associated with Alzheimer disease. They mainly consist of the amyloid peptide Aβ and display an abnormal content in Zn2+ ions, together with many truncated, isomerized, and racemized forms of Aβ. The region 1-16 of Aβ can be considered the minimal zinc-binding domain and contains two aspartates subject to protein aging. The influence of zinc binding and protein aging related modifications on the conformation of this region of Aβ is of importance given the potentiality of this domain to constitute a therapeutic target, especially for immunization approaches. In this study, we determined from NMR data the solution structure of the Aβ-(1-16)-Zn2+ complex in aqueous solution at pH 6.5. The residues His6, His13, and His14 and the Glu11 carboxylate were identified as ligands that tetrahedrally coordinate the Zn(II) cation. In vitro aging experiments on Aβ-(1-16) led to the formation of truncated and isomerized species. The major isomer generated, Aβ-(1-16)-l-iso-Asp7, displayed a local conformational change in the His6-Ser8 region but kept a zinc binding propensity via a coordination mode involving l-iso-Asp7. These results are discussed here with regard to Aβ fibrillogenesis and the potentiality of the region 1-16 of Aβ to be used as a therapeutic target. Amyloid deposits within the cerebral tissue constitute a characteristic lesion associated with Alzheimer disease. They mainly consist of the amyloid peptide Aβ and display an abnormal content in Zn2+ ions, together with many truncated, isomerized, and racemized forms of Aβ. The region 1-16 of Aβ can be considered the minimal zinc-binding domain and contains two aspartates subject to protein aging. The influence of zinc binding and protein aging related modifications on the conformation of this region of Aβ is of importance given the potentiality of this domain to constitute a therapeutic target, especially for immunization approaches. In this study, we determined from NMR data the solution structure of the Aβ-(1-16)-Zn2+ complex in aqueous solution at pH 6.5. The residues His6, His13, and His14 and the Glu11 carboxylate were identified as ligands that tetrahedrally coordinate the Zn(II) cation. In vitro aging experiments on Aβ-(1-16) led to the formation of truncated and isomerized species. The major isomer generated, Aβ-(1-16)-l-iso-Asp7, displayed a local conformational change in the His6-Ser8 region but kept a zinc binding propensity via a coordination mode involving l-iso-Asp7. These results are discussed here with regard to Aβ fibrillogenesis and the potentiality of the region 1-16 of Aβ to be used as a therapeutic target. Amyloid deposition in senile plaques is one of the main cerebral damages associated with Alzheimer disease (AD). 3The abbreviations used are: AD, Alzheimer disease; ACN, acetonitrile; Aβ, amyloid β-peptide; Aβ-(1-16), 1-16 N-terminal region of Aβ; CSD, chemical shift deviations; DQF-COSY, double quantum filtered correlation spectroscopy; ESI, electrospray ionization; GC, gas chromatography; HMBC, heteronuclear multiple bond correlation; HSQC, heteronuclear single quantum correlation; MS, mass spectrometry; NOE, nuclear Overhauser effect; NOESY, nuclear Overhauser effect spectroscopy; RP-HPLC, reversed phase-high performance liquid chromatography; TOCSY, total correlation spectroscopy. The amyloid β-peptide (Aβ), the major component of these extracellular deposits, is a 39-43-amino acid peptide that results from the normal proteolytic processing of the amyloid precursor protein. Considerable NMR data obtained on Aβ and on fragments argue that the soluble monomeric peptide is in an unordered conformation in aqueous solution and mainly adopts an α-helical structure in membrane-mimicking media (1Barrow C.J. Zagorski M.G. Science. 1991; 253: 179-182Crossref PubMed Scopus (548) Google Scholar, 2Zagorski M.G. Barrow C.J. Biochemistry. 1992; 31: 5621-5631Crossref PubMed Scopus (209) Google Scholar, 3Talafous J. Marcinowski K.J. Klopman G. Zagorski M.G. Biochemistry. 1994; 33: 7788-7796Crossref PubMed Scopus (143) Google Scholar, 4Lee J.P. Stimson E.R. Ghilardi J.R. Mantyh P.W. Lu Y.A. Felix A.M. Llanos W. Behbin A. Cummings M. Van Criekinge M. Timms W. Maggio J.E. Biochemistry. 1995; 34: 5191-5200Crossref PubMed Scopus (187) Google Scholar, 5Marcinowski K.J. Shao H. Clancy E.L. Zagorski M.G. J. Am. Chem. Soc. 1998; 120: 11082-11091Crossref Scopus (79) Google Scholar, 6Sticht H. Bayer P. Willbold D. Dames S. Hilbich C. Beyreuther K. Frank R.W. Rosch P. Eur. J. Biochem. 1995; 233: 293-298Crossref PubMed Scopus (277) Google Scholar, 7Coles M. Bicknell W. Watson A.A. Fairlie D.P. Craik D.J. Biochemistry. 1998; 37: 11064-11077Crossref PubMed Scopus (476) Google Scholar, 8Shao H. Jao S. Ma K. Zagorski M.G. J. Mol. Biol. 1999; 285: 755-773Crossref PubMed Scopus (279) Google Scholar, 9Zhang S. Iwata K. Lachenmann M.J. Peng J.W. Li S. Stimson E.R. Lu Y. Felix A.M. Maggio J.E. Lee J.P. J. Struct. Biol. 2000; 130: 130-141Crossref PubMed Scopus (311) Google Scholar). By contrast, the conformation of Aβ within amyloid fibrils consists mainly of a β-sheet structure (10Choo L.P. Wetzel D.L. Halliday W.C. Jackson M. LeVine S.M. Mantsch H.H. Biophys. J. 1996; 71: 1672-1679Abstract Full Text PDF PubMed Scopus (206) Google Scholar, 11Malinchik S.B. Inouye H. Szumowski K.E. Kirschner D.A. Biophys. J. 1998; 74: 537-545Abstract Full Text Full Text PDF PubMed Scopus (238) Google Scholar, 12Serpell L.C. Fraser P.E. Sunde M. Methods Enzymol. 1999; 309: 526-536Crossref PubMed Scopus (124) Google Scholar). Therefore, as for other conformational diseases (13Ellis R.J. Pinheiro T.J. Nature. 2002; 416: 483-484Crossref PubMed Scopus (124) Google Scholar), a transconformation toward β-sheet is assumed to trigger the abnormal fibrillar amyloid deposition associated with AD. The N-terminal region of Aβ has been shown to be flexible and accessible within amyloid fibrils (14Petkova A.T. Ishii Y. Balbach J.J. Antzutkin O.N. Leapman R.D. Delaglio F. Tycko R. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 16742-16747Crossref PubMed Scopus (1656) Google Scholar, 15Kheterpal I. Williams A. Murphy C. Bledsoe B. Wetzel R. Biochemistry. 2001; 40: 11757-11767Crossref PubMed Scopus (198) Google Scholar). Therefore, it constitutes an attractive therapeutic target, as illustrated by the ability of monoclonal antibodies directed toward this region to dissociate amyloid fibrils (16McLaurin J. Cecal R. Kierstead M.E. Tian X. Phinney A.L. Manea M. French J.E. Lambermon M.H. Darabie A.A. Brown M.E. Janus C. Chishti M.A. Horne P. Westaway D. Fraser P.E. Mount H.T. Przybylski M. St George-Hyslop P. Nat. Med. 2002; 8: 1263-1269Crossref PubMed Scopus (398) Google Scholar, 17Frenkel D. Balass M. Solomon B. J. Neuroimmunol. 1998; 88: 85-90Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar). Imbalances in transition metal cations such as copper, zinc, and iron are assumed to contribute to Aβ deposition (18Bush A.I. Trends Neurosci. 2003; 26: 207-214Abstract Full Text Full Text PDF PubMed Scopus (1134) Google Scholar). In particular, (i) Zn2+ ions have been shown to trigger Aβ aggregation (19Mantyh P.W. Ghilardi J.R. Rogers S. DeMaster E. Allen C.J. Stimson E.R. Maggio J.E. J. Neurochem. 1993; 61: 1171-1174Crossref PubMed Scopus (428) Google Scholar, 20Bush A.I. Pettingell W.H. Multhaup G. de Paradis M. Vonsattel J.P. Gusella J.F. Beyreuther K. Masters C.L. Tanzi R.E. Science. 1994; 265: 1464-1467Crossref PubMed Scopus (1411) Google Scholar, 21Esler W.P. Stimson E.R. Jennings J.M. Ghilardi J.R. Mantyh P.W. Maggio J.E. J. Neurochem. 1996; 66: 723-732Crossref PubMed Scopus (125) Google Scholar), and (ii) abnormally high levels of zinc have been found within amyloid deposits in AD patients (22Lovell M.A. Robertson J.D. Teesdale W.J. Campbell J.L. Markesbery W.R. J. Neurol. Sci. 1998; 158: 47-52Abstract Full Text Full Text PDF PubMed Scopus (1828) Google Scholar, 23Suh S.W. Jensen K.B. Jensen M.S. Silva D.S. Kesslak P.J. Danscher G. Frederickson C.J. Brain Res. 2000; 852: 274-278Crossref PubMed Scopus (322) Google Scholar). However, the role of zinc in amyloid deposition is subject to debate (24Cuajungco M.P. Faget K.Y. Brain Res. Brain Res. Rev. 2003; 41: 44-56Crossref PubMed Scopus (214) Google Scholar, 25Cuajungco M.P. Frederickson C.J. Bush A.I. Subcell. Biochem. 2005; 38: 235-254Crossref PubMed Google Scholar) as (i) Zn2+ ions induce the deposition of Aβ in the form of nonfibrillar aggregates (26Yoshiike Y. Tanemura K. Murayama O. Akagi T. Murayama M. Sato S. Sun X. Tanaka N. Takashima A. J. Biol. Chem. 2001; 276: 32293-32299Abstract Full Text Full Text PDF PubMed Scopus (234) Google Scholar, 27Parbhu A. Lin H. Thimm J. Lal R. Peptides. 2002; 23: 1265-1270Crossref PubMed Scopus (61) Google Scholar), and (ii) zinc has been shown to function as an antioxidant able to protect from extensive redox chemical reactions that contribute to AD-related oxidative stress (28Curtain C.C. Ali F. Volitakis I. Cherny R.A. Norton R.S. Beyreuther K. Barrow C.J. Masters C.L. Bush A.I. Barnham K.J. J. Biol. Chem. 2001; 276: 20466-20473Abstract Full Text Full Text PDF PubMed Scopus (599) Google Scholar, 29Valko M. Morris H. Cronin M.T. Curr. Med. Chem. 2005; 12: 1161-1208Crossref PubMed Scopus (3733) Google Scholar). The obligatory sequence involved in zinc binding has been mapped to the region 6-28 of Aβ (30Bush A.I. Pettingell Jr., W.H. de Paradis M. Tanzi R.E. Wasco W. J. Biol. Chem. 1994; 269: 26618-26621Abstract Full Text PDF PubMed Google Scholar), and the involvement of the His13 and His14 residues has been strongly suggested (31Liu S.T. Howlett G. Barrow C.J. Biochemistry. 1999; 38: 9373-9378Crossref PubMed Scopus (192) Google Scholar, 32Miura T. Suzuki K. Kohata N. Takeuchi H. Biochemistry. 2000; 39: 7024-7031Crossref PubMed Scopus (417) Google Scholar, 33Yang D.S. McLaurin J. Qin K. Westaway D. Fraser P.E. Eur. J. Biochem. 2000; 267: 6692-6698Crossref PubMed Scopus (113) Google Scholar). Because of marked aggregation of Aβ peptides or Aβ fragments in the presence of Zn2+, zinc-induced conformational switching has not yet been characterized. However, using Aβ-(1-16), a synthetic peptide spanning the amino acid sequence between the α- and β-secretase cleavage sites, we showed in our initial CD study that this fragment of Aβ selectively bound Zn2+ to form a 1:1 complex soluble in physiologically relevant conditions and subsequently underwent a conformational change (34Kozin S.A. Zirah S. Rebuffat S. Hoa G.H. Debey P. Biochem. Biophys. Res. Commun. 2001; 285: 959-964Crossref PubMed Scopus (114) Google Scholar). We proposed a model of zinc attachment that involved the three histidines, His6, His13, and His14, and hypothesized Arg5 as the fourth coordination ligand from the analysis of the Aβ-(1-16)-Zn2+ 1:1 complex in the gas phase by electrospray-ionization mass spectrometry (ESI-MS) (35Zirah S. Rebuffat S. Kozin S.A. Debey P. Fournier F. Lesage D. Tabet J.C. Int. J. Mass Spectrom. 2003; 228: 999-1016Crossref Scopus (60) Google Scholar). Recently, the zinc binding property of Aβ-(1-16) was studied by Mekmouche et al. (36Mekmouche Y. Coppel Y. Hochgrafe K. Guilloreau L. Talmard C. Mazarguil H. Faller P. Chembiochem. 2005; 6: 1663-1671Crossref PubMed Scopus (83) Google Scholar), and the apparent dissociation constant of the 1:1 complex was determined. In addition to its zinc binding propensity, the 1-16 N-terminal region of Aβ is subject to isomerization and racemization of the aspartate residues, which constitute very common types of aging-related protein damages. These chemical modifications occur through a common pathway involving a neutral cyclic succinimide intermediate (37Geiger T. Clarke S. J. Biol. Chem. 1987; 262: 785-794Abstract Full Text PDF PubMed Google Scholar, 38Stephenson R.C. Clarke S. J. Biol. Chem. 1989; 264: 6164-6170Abstract Full Text PDF PubMed Google Scholar). The potential contribution of such aspartyl modifications to Aβ amyloidosis has been addressed, because unusually high contents of racemized and isomerized Asp residues were found in Aβ isolated from amyloid deposits (39Roher A.E. Lowenson J.D. Clarke S. Wolkow C. Wang R. Cotter R.J. Reardon I.M. Zurcher-Neely H.A. Heinrikson R.L. Ball M.J. Greenberg B.D. J. Biol. Chem. 1993; 268: 3072-3083Abstract Full Text PDF PubMed Google Scholar). In addition, these modifications were shown to be related to an increase in β-sheet content and to in vitro fibrillation (40Kuo Y.M. Webster S. Emmerling M.R. De Lima N. Roher A.E. Biochim. Biophys. Acta. 1998; 1406: 291-298Crossref PubMed Scopus (114) Google Scholar, 41Fabian H. Szendrei G.I. Mantsch H.H. Greenberg B.D. Otvos Jr., L. Eur. J. Biochem. 1994; 221: 959-964Crossref PubMed Scopus (50) Google Scholar), leading to the protein aging hypothesis of AD (42Orpiszewski J. Schormann N. Kluve-Beckerman B. Liepnieks J.J. Benson M.D. FASEB J. 2000; 14: 1255-1263Crossref PubMed Scopus (47) Google Scholar). On the other hand, the assessment of plaque age by using antibodies targeting specifically a particular isomer of Aβ suggests that these modifications would occur rather after the amyloid deposition (43Fonseca M.I. Head E. Velazquez P. Cotman C.W. Tenner A.J. Exp. Neurol. 1999; 157: 277-288Crossref PubMed Scopus (55) Google Scholar). The N-terminal region of Aβ appears as an attractive therapeutic target, especially for active or passive immunization approaches. Indeed, only the antibodies raised against the N-terminal part of Aβ are able to reduce the plaque burden and restore cognitive deficits in the mice model of AD (16McLaurin J. Cecal R. Kierstead M.E. Tian X. Phinney A.L. Manea M. French J.E. Lambermon M.H. Darabie A.A. Brown M.E. Janus C. Chishti M.A. Horne P. Westaway D. Fraser P.E. Mount H.T. Przybylski M. St George-Hyslop P. Nat. Med. 2002; 8: 1263-1269Crossref PubMed Scopus (398) Google Scholar, 44Frenkel D. Dewachter I. Van Leuven F. Solomon B. Vaccine. 2003; 21: 1060-1065Crossref PubMed Scopus (83) Google Scholar, 45Chauhan N.B. Siegel G.J. Neurosci. Lett. 2005; 375: 143-147Crossref PubMed Scopus (36) Google Scholar). In addition, targeting this region should enable us to exert this effect without eliciting an inflammatory response, which had been critical in the first clinical tests of active immunization on humans (46Gelinas D.S. DaSilva K. Fenili D. St George-Hyslop P. McLaurin J. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 14657-14662Crossref PubMed Scopus (161) Google Scholar). We have shown previously that Aβ-(1-16) zinc binding induces an agonist effect on the 4-10 epitope recognition by different monoclonal antibodies, suggesting a folding of the peptide that would render the epitope more accessible (47Zirah S. Stefanescu R. Manea M. Tian X. Cecal R. Kozin S.A. Debey P. Rebuffat S. Przybylski M. Biochem. Biophys. Res. Commun. 2004; 321: 324-328Crossref PubMed Scopus (23) Google Scholar). Because clinical testing for passive immunization has started, it is of major importance to characterize the structural changes of the N-terminal region of Aβ upon zinc binding and protein aging-induced modifications. Here we used NMR and molecular modeling to determine the three-dimensional structure of the apoAβ-(1-16) and the Aβ-(1-16)-Zn2+ complex in aqueous solution at pH 6.5. Furthermore, we investigated the ability of Aβ-(1-16) to undergo protein aging-related modifications. In vitro aging experiments were performed on the synthetic peptide Aβ-(1-16) in the absence and in the presence of Zn2+ ions. The resultant species were isolated and identified, and the main isomers generated were analyzed by NMR. From our study, region 1-16 of Aβ would behave as an autonomous segment able to undergo both zinc binding-induced structuration and aging-related isomerizations. Synthetic Peptides—The synthetic peptides used throughout this study were purchased from Eurogentec (Angers, France) with a ≥95% purity checked by RP-HPLC and matrix-assisted desorption ionization time-of-flight mass spectrometry (Voyager-DEPro, Applied Biosystems, Courtaboeuf, France). Aβ-(1-16) (Ac-Asp-Ala-Glu-Phe-Arg5-His-Asp-Ser-Gly-Tyr10-Glu-Val-His-His-Glu15-Lys-NH2) was acetylated at the N terminus and amidated at the C terminus; Aβ-(1-16)hemi was nonacetylated at the N terminus and amidated at the C terminus; Aβ-(1-16)-l-iso-Asp7 was acetylated at the N terminus and amidated at the C terminus and contained one l-iso-Asp residue at position 7. NMR Spectroscopy—Aβ-(1-16) was dissolved either in 50 mm sodium phosphate buffer, pH 6.5, or in Tris-d11/HCl 22.5 mm, pH 5.8, prepared in H2O/D2O (9:1) using MilliQ™ water samples (Millipore, Saint-Quentin-en-Yvelines, France). The pH was checked directly in the 5-mm Wilmad NMR tubes by using a Mettler Toledo U402-M3-S7/200 electrode. pH values were uncorrected for deuterium isotope effects. Samples for experiments in the presence of Zn2+ were prepared by adding a concentrated stock solution of the highest analytical grade ZnCl2 (Aldrich) in order to reach a Zn2+/peptide ratio of 1.5. The absence of pH variation upon zinc addition was checked. Alternatively, Aβ-(1-16) and Aβ-(1-16)-l-iso-Asp7 were analyzed at 3 mm in 50 mm sodium phosphate buffer, pH 7.5, prepared in H2O/D2O (9:1). NMR experiments were carried out on AVANCE 400 and DMX 600 spectrometers (Bruker Biospin, Wissembourg, France) both equipped with shielded gradients z and set up with 1H broad band reverse gradient and triple resonance 1H-13C-15N gradient probe heads, respectively. Temperature was controlled with a BCU-05 refrigeration unit and a BVT 3000 control unit on both spectrometers. 1H and 13C chemical shifts were externally referenced to sodium 2,2-dimethyl-2-silapentane-5-sulfonate. Conventional one- and two-dimensional experiments, one-dimensional 1H, 1H-1H DQF-COSY, TOCSY (using a spin-lock field produced by an MLEV-17 spin-locking sequence for a spin-lock time of 120 ms), NOESY (with mixing times of 100, 200, 300, and 400 ms), as well as natural abundance 1H-13C HSQC and HMBC (with long-range coupling evolution delays of 70 ms and 90 ms) were performed. Water suppression was achieved by means of either selective low power irradiation for the DQF-COSY experiment or pulsed field gradients in a water suppression by a gradient tailored excitation scheme included in the pulse sequences for both TOCSY and NOESY experiments. Data were processed on Silicon Graphics Indigo 2 XL or O2 workstations, using XWINNMR 3.1 and AURELIA software (Bruker Biospin, Wissembourg, France). The determination of temperature coefficients (Δδ/ΔTHN) was achieved for each sample from five series of 1H one-dimensional and TOCSY spectra recorded between 288 and 313 K. 3JHN-Hα coupling constants were determined from 1H one-dimensional and DQF-COSY spectra. The chemical shift deviations (CSD) were calculated for H-α and C-α atoms considering the reference chemical shifts proposed by Wishart et al. (48Wishart D.S. Sykes B.D. Richards F.M. Biochemistry. 1992; 31: 1647-1651Crossref PubMed Scopus (2024) Google Scholar, 49Wishart D.S. Sykes B.D. J. Biomol. NMR. 1994; 4: 171-180Crossref PubMed Scopus (1916) Google Scholar) for each amino acid in random structure. The pKa values of the histidines were measured in the absence of Zn2+ ions by using 3.5 mm Aβ-(1-16) peptide solutions in 10 mm H2O/D2O (9:1) sodium phosphate buffer. The effects of pH on the proton chemical shifts of Aβ-(1-16) were determined from a series of 1H one-dimensional, TOCSY, and NOESY spectra recorded at different pH values ranging from 3.0 to 9.5 at 278 K. The pH values were adjusted before each NMR experiment, and the absence of pH variation over the period of acquisition was checked immediately after. The pKa values were determined by analyzing the pH titration curves by nonlinear least square fit to the equation δ = (δ1 + δ2 × 10(pH - pKa))/(1 + 10(pH - pKa)), where δ is the chemical shift of a resonance measured as a function of pH, and δ1 and δ2 are its chemical shifts at the lowest and highest pH values, respectively. This procedure was carried out using the software Curve Expert 1.3. The equation used derives from the Henderson-Hasselbalch equation, assuming a rapid equilibrium between protonated and unprotonated forms (50Forman-Kay J.D. Clore G.M. Gronenborn A.M. Biochemistry. 1992; 31: 3442-3452Crossref PubMed Scopus (121) Google Scholar) and considering a noninteracting model (51Perez-Canadillas J.M. Campos-Olivas R. Lacadena J. Martinez del Pozo A. Gavilanes J.G. Santoro J. Rico M. Bruix M. Biochemistry. 1998; 37: 15865-15876Crossref PubMed Scopus (67) Google Scholar). Distance and Dihedral Angle Constraints—Distance constraints resulting from integrated NOESY spectra and ϕ dihedral angles derived from the 3JHN-Hα coupling constants using the Pardi relation (52Pardi A. Billeter M. Wuthrich K. J. Mol. Biol. 1984; 180: 741-751Crossref PubMed Scopus (940) Google Scholar) were used for structure calculation. The NOESY experiment with 200 ms of mixing time was selected for distance calculation to get rid of spin diffusion associated with T1 relaxation. A tolerance range of ±25% of the NMR-derived distances was used to define the upper and lower values of the constraints. Structure Calculation and Analysis—The three-dimensional structures of Aβ-(1-16) at pH 6.5 in the absence and in the presence of zinc were calculated using simulated annealing and energy minimization protocols in X-PLOR 3.851. Alternatively, the conformational calculations were performed with a general purpose internal coordinate molecular dynamics program ICMD (53Mazur A.K. Becker O.M. MacKerell Jr., A.D. Roux B. Watanabe M. Computational Biochemistry and Biophysics. Marcel Dekker, Inc., New York2001: 115-131Google Scholar) by using the variable target function approach (54Braun W. Go N. J. Mol. Biol. 1985; 186: 611-626Crossref PubMed Scopus (501) Google Scholar) adopted for dynamics in the torsion angle space. The program ICMD was further used to calculate the three-dimensional structure of the Aβ-(1-16)-Zn2+ complex. The structures obtained from ICMD calculations are only presented here as final structures. The molecular dynamics calculations in X-PLOR 3.851 were performed with a target function similar to that used by Nilges et al. (55Nilges M. Gronenborn A.M. Brunger A.T. Clore G.M. Protein Eng. 1988; 2: 27-38Crossref PubMed Scopus (516) Google Scholar) and a force field adapted for NMR structure determination (parallhdg.pro and topallhdg.pro). When no stereospecific assignments could be made for methyl and methylene protons, the constraints were considered with an appropriate treatment in X-PLOR. Several rounds of structure calculation and assignment were performed to resolve ambiguities. Starting from an extended template structure, a set of 100 structures was calculated. A first phase of 400-ps dynamics (time step = 2 fs) at 1000 K was followed by 80-ps slow cooling step to 100 K (time step = 2 fs; temperature step = 20 K). A weak weight of the van der Waals repulsive term was used at high temperature in order to allow a large conformational sampling. Refinement of the structures was achieved by using the conjugate gradient Powell algorithm with 7000 cycles of energy minimization, using the CHARMM 22 force field (files topallh22x.pro and parallh22x.pro) (56Brooks B.R. Bruccoleri R.E. Olafson B.D. States D.J. Swaminathan S. Karplus M. J. Comput. Chem. 1983; 4: 187-217Crossref Scopus (14019) Google Scholar). The ICMD program was used with the standard geometry of amino acids and peptide bonds and involved multiple cycles of simulated annealing starting from an arbitrary extended peptide conformation. The AMBER99 all-atom force field parameters (57Cornell W.D. Cieplak P. Bayly C.I. GouldIan R. Merz K.M. Ferguson D.M. Spellmeyer D.C. Fox T. Caldwell J.W. Kollman P.A. J. Am. Chem. Soc. 1995; 117: 5179-5197Crossref Scopus (11640) Google Scholar) were applied, with nonbonded interactions truncated at 6 Å by a force-shift method to maintain reasonable atom-atom distances and avoid any bias from longer range interactions. The Aβ-(1-16) peptide was modeled with all torsion degrees of freedom. According to the variable target function principle (54Braun W. Go N. J. Mol. Biol. 1985; 186: 611-626Crossref PubMed Scopus (501) Google Scholar), all NOE-based distance constraints were first checked for the number of free torsions that separate every particular proton pair. The corresponding number is further referred to as torsion separation. The simulated annealing started from an arbitrary extended conformation obtained by an unrestrained MD simulation under high temperature (7000 K). At the beginning, only constraints of torsion separation one were applied. The following separation levels were added one by one in the course of the protocol when reasonable convergence was achieved for all previous torsion separation levels. As the ICMD protocol does not use pseudoatoms, instead all candidate proton pairs corresponding to a given resonance were analyzed from time to time, and the corresponding constraint was reassigned to the pair with the shortest distance in the current conformation. 4T. Malliavin et al., manuscript in preparation. The structure calculation of the Aβ-(1-16)-Zn2+ 1:1 complex was carried out using a strategy including two independent steps. In the first step, the structure of the polypeptide chain only was refined in calculations performed without explicit Zn2+ ion and without any assumption on zinc coordination. The second calculation step was used in order to obtain a specific peptide-zinc complex structure with a metal coordination geometry satisfying chemical requirements. The experimental set of constraints used in the first step was updated with additional distance and angle constraints that enforced a tetrahedral ligand coordination of the Zn2+ ion. To this end, we added four ambiguous constraints that linked the Zn2+ ion with all possible partners. These constraints were arbitrarily assigned the torsion separation one, and they were treated along with other constraints, as explained above. Their geometry was derived from zinc-binding sites of relevant high resolution x-ray structures available in the Protein Data Bank. The following atoms and groups present in the Aβ-(1-16) peptide were considered as potential zinc chelators: N-δ and/or N-ϵ atoms of His6, His13, and His14, O-δ1-O-δ2 atoms of Asp1 and Asp7, and O-ϵ1-O-ϵ2 atoms of Glu3 and Glu11. These Zn2+ ligand hypotheses were in agreement with those authorized by the NMR data obtained. The 2.0 Å harmonic distance constraints were applied for the distance between the Zn2+ ion and each of its four chelators. If one of the histidine nitrogens took part in the complex, the Zn2+ ion was kept in the plane of the imidazole ring and in the bissector plane of the corresponding nitrogen atom by using additional angle constraints. In this case the other nitrogen of the same imidazole ring was temporarily excluded from the list of possible chelators. In a similar way, an appropriate symmetrical orientation of Zn2+ with respect to the Asp and Glu carboxyl groups was ensured. The calculated conformers were analyzed using MOLMOL (58Koradi R. Billeter M. Wuthrich K. J. Mol. Graphics. 1996; 14 (29-32): 51-55Crossref PubMed Scopus (6498) Google Scholar), RASMOL (59Bernstein H.J. Trends Biochem. Sci. 2000; 25: 453-455Abstract Full Text Full Text PDF PubMed Scopus (191) Google Scholar), and PROCHECK (60Laskowski R.A. MacArthur M.W. Moss D.S. Thornton J.M. J. Appl. Crystallogr. 1993; 26: 283-291Crossref Google Scholar) programs. Their coordinates are deposited in the Protein Data Bank. Production of Aβ-(1-16) Modified Forms by in Vitro Aging—Aβ-(1-16) (50 μm or 1.50 mm) or Aβ-(1-16)hemi (50 μm) was incubated as described previously (38Stephenson R.C. Clarke S. J. Biol. Chem. 1989; 264: 6164-6170Abstract Full Text PDF PubMed Google Scholar) in 50 mm Tris-HCl, pH 7.4, containing 0.05% NaN3 (w/v) either at 37 or 70 °C in the absence or presence of 1 mm EDTA. The in vitro aging experiments in the presence of zinc were conducted with the same protocol in the absence of EDTA and in the presence of ZnCl2 with a 1:5 peptide/zinc ratio. Reactions were stopped by freezing the samples at -20 °C. The modified peptide mixtures obtained were analyzed at different incubation times by RP-HPLC on a Shiseido C-18 Capcell Pack, 5 μm, 4.6 × 250-mm column from Interchim (Montluçon, France), using a linear gradient of 10-40% MeOH in 0.1% aqueous trifluoroacetic acid, at a flow-rate of 1 ml/min. Elution was monitored by measuring the absorbance at 226 nm. Molecular masses of the separated species were determined by liquid chromatography-MS using a Discovery®HS C-18, 3 μm, 2.1 × 150-mm analytical column (Supelco, St Quentin-Fallavier, France) coupled to an ESI-hybrid quadrupole time-of-flight mass spectrometer (Q-STAR Pulsar, Applied Biosystems, Courtaboeuf, France). The separation was obtained with a linear gradient of 27-35% MeOH in 0.1% aqueous trifluoroacetic acid over 40 min at a flow rate of 200 μl/min. The same aging protocol was used for the nonacetylated peptide Aβ-(1-16)hemi. The modified peptide mixtures resulting from the incubation were analyzed at different incubation times by RP-HPLC, as described above, using a linear gradient of 13-18% ACN in 0.1% aqueous trifluoroacetic acid over 40 min at a flow rate of 1 ml/min. Isolation of Aβ-(1-16) Isomers Produced from in Vitro Aging—The modified peptides resulting from incubation of Aβ-(1-16) at 70 °C for 14 days were purified by semi-preparative RP-HPLC on a C-18 Uptisphere, 5 μm, 7.8 × 250-mm column (Interchim, Montluçon, France) with a linear gradient of 27-35% MeOH in 0.1% aqueous trifluoroacetic acid over 40 min at a flow-rate of 2 ml/min. Absorbance was monitored at 226 nm. Purity of the isolated species was checked by RP-HPLC and ESI-MS
Referência(s)