Identification of Aspartic and Isoaspartic AcidResidues in Amyloid � Peptides, Including A�1-42,Using Electron-Ion Reactions

Nadezda P. Sargaeva,† Cheng Lin,† and Peter B. O’Connor*,†,‡

Mass Spectrometry Resource, Department of Biochemistry, Boston University School of Medicine, 670 Albany Street,R504, Boston, Massachusetts 02118, and Department of Chemistry, University of Warwick, Gibbet Hill Road,Coventry, CV4 7AL, U.K.

Amyloid � peptides are the major components of thevascular and plaque amyloid filaments in individuals withAlzheimer’s disease (AD). Although it is still unclear whatinitiates the disease, isomerization of aspartic acid resi-dues in A� peptides is directly related to the pathology ofAD. The detection of isomerization products is analyticallychallenging, due to their similar chemical properties andidentical molecular mass. Different methods have beenapplied to differentiate and quantify the isomers, includingimmunology, chromatography, and mass spectrometry.Typically, those methods require comparative analysiswith the standard peptides and involve many samplepreparation steps. To understand the role of A� isomer-ization in AD progression, a fast, simple, accurate, andreproducible method is necessary. In this work, electroncapture dissociation (ECD) Fourier-transform ion cyclo-tron resonance mass spectrometry (FTICR MS) was ap-plied to detect isomerization in A� peptides. ECD gener-ated diagnostic fragment ions for the two isomers ofA�17-28, [M + 2H - 60]+• and z6• - 44 when asparticacid was present and z6• - 57 when isoaspartic acidwas present. Additionally, the zn - 57 diagnostic ionwas also observed in the electron ionization dissocia-tion (EID) spectra of the modified A�17-28 fragment.ECD was further applied toward A�1-40 and A�1-42.The diagnostic ion c6 + 57 was observed in the ECDspectra of the A�1-42 peptide, demonstrating isomer-ization at residue 7. In conclusion, both ECD and EIDcan clearly determine the presence and the positionof isoaspartic acid residues in amyloid � peptides. Thenext step, therefore, is to apply this method to analyzesamples of Alzheimer’s patients and healthy individu-als in order to generate a better understanding of thedisease.

Amyloid � (A�) peptides are the major components of thevascular and plaque amyloid filaments in individuals with Alzhe-imer’s disease (AD). Various forms of A� are proteolyticallycleaved from the A� precursor protein, with A�1-40 and A�1-42

being the most abundant forms found in amyloid deposits.1 Eversince A� was first purified and characterized, it has been stronglyassociated with the pathology of AD,2,3 although it remains unclearwhat initiates the disease. According to the most widely acceptedhypothesis, cerebral A� accumulation is the primary cause in AD.The rest of the disease process stems from an imbalance betweenA� production and clearance.4 Many attempts have been madeto measure the concentration of A� peptides in biological fluids,but it is difficult to correlate A� levels with disease stage and,thus, to use it as an AD biomarker.5 Further research anddevelopment of analytical methods is necessary for early ADdiagnosis, monitoring of disease progression, and a betterunderstanding of the disease.

The current research has focused on A� isomerization.Isomerization of aspartic acid is one of the most common post-translational modifications (PTMs) that accumulates with age inlong-lived proteins, especially in tooth, bone, cartilage, lens, andbrain tissues.6 The isomerization product is isoaspartic acid(isoAsp). It can also be formed from asparagine deamidation(Scheme 1). Both reactions proceed via formation of the five-membered succinimide ring intermediate followed by a rapidhydrolysis. As a result, aspartic and isoaspartic acid residues areformed in a ratio of 1:3.8,9 Moreover, at pH ∼7.4, IsoAsp formationis preferential due to the higher acidity of isoAsp side chainresidue.7 Under physiological conditions, both Asp isomerizationand Asn deamidation are spontaneous nonenzymatic reactions.6-11

The reaction rates depend mainly on the nature of the adjacent

* To whom correspondence should be addressed: Phone: +44 (0)2476 151008. Fax: +44 (0)2476 151 009. E-mail: p.oconnor@warwick.ac.uk.

† Boston University School of Medicine.‡ University of Warwick.

(1) Selkoe, D. J. Annu. Rev. Neurosci. 1994, 17, 489–517.(2) Glenner, G. G.; Wong, C. W. Biochem. Biophys. Res. Commun. 1984, 120,

885–890.(3) Masters, C. L.; Simms, G.; Weinman, N. A.; Multhaup, G.; McDonald, B. L.;

Beyreuther, K. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 4245–4249.(4) Hardy, J.; Selkoe, D. J. Science 2002, 297, 353–356.(5) Roher, A. E.; Esh, C. L.; Kokjohn, T. A.; Castano, E. M.; Van Vickle, G. D.;

Kalback, W. M.; Patton, R. L.; Luehrs, D. C.; Daugs, I. D.; Kuo, Y. M.;Emmerling, M. R.; Soares, H.; Quinn, J. F.; Kaye, J.; Connor, D. J.;Silverberg, N. B.; Adler, C. H.; Seward, J. D.; Beach, T. G.; Sabbagh, M. N.Alzheimer’s Dementia 2009, 5, 18–29.

(6) Ritz-Timme, S.; Collins, M. J. Ageing Res. Rev. 2002, 1, 43–59.(7) Capasso, S. Thermochim. Acta 1996, 286, 41–50.(8) Clarke, S. Int. J. Pept. Protein Res. 1987, 30, 808–821.(9) Geiger, T.; Clarke, S. J. Biol. Chem. 1987, 262, 785–794.

(10) Shimizu, T.; Matsuoka, Y.; Shirasawa, T. Biol. Pharm. Bull. 2005, 28, 1590–1596.

(11) Aswad, D. W.; Paranandi, M. V.; Schurter, B. T. Third Symposium on theAnalysis of Well Characterized Biotechnology Pharmaceuticals, January 7,1999, Washington, DC; pp 1129-1136.

Anal. Chem. 2009, 81, 9778–9786

10.1021/ac901677t CCC: $40.75 2009 American Chemical Society9778 Analytical Chemistry, Vol. 81, No. 23, December 1, 2009Published on Web 10/29/2009

residues, the higher order structure of the protein, and themolecular environment.6,8,12 Formation of isoAsp is speculated tochange protein structure as it introduces an additional methylenegroup into the polypeptide backbone. This can change proteinfunction and activity, or trigger aggregation.6,13,14 In addition,proteins containing isoAsp may not fully degrade, as the isoas-partate residue hinders proteolytic degradation.15 Nonetheless,harmful effects of isomerization can be partially repaired by theintracellular enzyme, called protein isoaspartyl methyltransferase(PIMT), which selectively converts isoAsp residues back to thesuccinimide intermediate.6,10

Isomerization of aspartic acid is directly related to the pathol-ogy of Alzheimer’s disease. A� peptides have three aspartic acidsin the sequence at residues 1, 7, and 23 and formation ofisoaspartate is enhanced in A� peptides in AD. Roher et al. foundthat Asp1 and Asp7 were isomerized in the cerebral plaquesamples of Alzheimer patients.13 Recently isoAsp7 and isoAsp23were found in the core of senile plaques and Amyloid-bearingvessels, as was shown with anti-isoasp7 and anti-isoasp23 antibod-ies.10 Moreover, the Iowa (Asn23)16 and Tottori-Japanese(Asn7)17 mutations in familial AD have the potential to accelerateformation of isoAsp, presumably due to asparagine deamidation.Accordingly, isoAsp23 containing A� peptides were preferentiallydetected in vascular deposits in Iowa cerebral amyloid angiopathybrain.18 It was further suggested that spontaneous isomerizationat position 23 induces the conformational change to form a �-turnof the polypeptide chain. This, in turn, plays a pathogenic role in

the deposition of A� peptides in sporadic AD.10 In vitro experi-ments showed increased fibrillogenesis and enhanced neurotox-icity of isoAsp23-containing A� peptides;10,19 however, otherresults suggested that A� aggregative ability and neurotoxicitywere not enhanced by this modification.20 Similar studies of A�with isoAsp7 revealed that it is primarily deposited in the core ofsenile plaques. IsoAsp7 accumulation was associated with the ageof the plaque, as the number of isoAsp-positive plaques increasedin parallel with the disease severity.21

There have been many studies of isomerization events in A�peptides, but its role in AD pathogenesis is still unclear. Formationof isoAsp in A� might be a reason for its impaired degradation,leading to accumulation. This aggregation could be a way forbiological systems to reduce the toxicity of the nondegradableA� peptides. Further research is needed to better understand therole of isoAsp formation.

The detection of isomerization products is analytically chal-lenging, due to their similar chemical properties and identicalmolecular mass. Nevertheless, various methods have been appliedto differentiate and quantify the isomers. Those include im-munological methods based on detection by specific isoAspantibodies,10,18,21 as well as liquid chromatography,11,13 PIMTenzyme utilizing assays,11,13 and tandem mass spectrometry (MS)analysis.22-24 New methods addressing isoAsp detection arequickly emerging in the field to better understand the harmfuleffects isoAsp accumulation can generate in biological systemsand to yield pharmaceutical applications. These new methods ofteninclude already known methods, such as high-performance liquidchromatography (HPLC) separation followed by MS analysis.25,26

In addition, immunochemistry analysis, capillary electrophoresis,or HPLC coupled to MS, or tandem MS alone, can be applied toanalyze L-Asp, L-isoAsp, D-Asp, and D-isoAsp isomerized/racemizedpeptides.27-29 HPLC upgraded to an ultraperformance liquidchromatography (UPLC) system could further separate isoAspand Asp located directly at the N-terminus.30

Edman degradation is a chemical cleavage mechanism usefulfor IsoAsp identification, because it does not cleave the isopeptidebond. Similarly, enzymatic digestion with the endoproteinaseAsp-N does not cleave peptide bonds N-terminal to the isoAspacid. This was utilized in combination with 15N isotope labeling

(12) Robinson, N. E.; Robinson, A. B. Proc. Natl. Acad. Sci. U.S.A. 2001, 98,944–949.

(13) Roher, A. E.; Lowenson, J. D.; Clarke, S.; Wolkow, C.; Wang, R.; Cotter,R. J.; Reardon, I. M.; Zurcherneely, H. A.; Heinrikson, R. L.; Ball, M. J.;Greenberg, B. D. J. Biol. Chem. 1993, 268, 3072–3083.

(14) Shimizu, T.; Watanabe, A.; Ogawara, M.; Mori, H.; Shirasawa, T. Arch.Biochem. Biophys. 2000, 381, 225–234.

(15) Bohme, L.; Bar, J. W.; Hoffmann, T.; Manhart, S.; Ludwig, H. H.; Rosche,F.; Demuth, H. U. Fifth General Meeting of the International-Proteolysis-Society, October 20-24, 2007, Patras, Greece, pp 1043-1053.

(16) Grabowski, T. J.; Cho, H. S.; Vonsattel, J. P. G.; Rebeck, G. W.; Greenberg,S. M. Ann. Neurol. 2001, 49, 697–705.

(17) Wakutani, Y.; Watanabe, K.; Adachi, Y.; Wada-Isoe, K.; Urakami, K.;Ninomiya, H.; Saido, T. C.; Hashimoto, T.; Iwatsubo, T.; Nakashima, K.J. Neurol. Neurosurg. Psych. 2004, 75, 1039–1042.

(18) Shin, Y.; Cho, H. S.; Fukumoto, H.; Shimizu, T.; Shirasawa, T.; Greenberg,S. M.; Rebeck, G. W. Acta Neuropathol. 2003, 105, 252–258.

(19) Fukuda, H.; Shimizu, T.; Nakajima, M.; Mori, H.; Shirasawa, T. Bioorg. Med.Chem. Lett. 1999, 9, 953–956.

(20) Murakami, K.; Uno, M.; Masuda, Y.; Shimizu, T.; Shirasawa, T.; Irie, K.Biochem. Biophys. Res. Commun. 2008, 366, 745–751.

(21) Fonseca, M. I.; Head, E.; Velazquez, P.; Cotman, C. W.; Tenner, A. J. Exp.Neurol. 1999, 157, 277–288.

(22) Gonzalez, L. J.; Shimizu, T.; Satomi, Y.; Betancourt, L.; Besada, V.; Padron,G.; Orlando, R.; Shirasawa, T.; Shimonishi, Y.; Takao, T. Rapid Commun.Mass Spectrom. 2000, 14, 2092–2102.

(23) Castet, S.; Enjalbal, C.; Fulcrand, P.; Guichou, J. F.; Martinez, J.; Aubagnac,J. L. Rapid Commun. Mass Spectrom. 1996, 10, 1934–1938.

(24) Lehmann, W. D.; Schlosser, A.; Erben, G.; Pipkorn, R.; Bossemeyer, D.;Kinzel, V. Protein Sci. 2000, 9, 2260–2268.

(25) Chelius, D.; Rehder, D. S.; Bondarenko, P. V. Anal. Chem. 2005, 77, 6004–6011.

(26) Barnes, C. A. S.; Lim, A. Mass Spectrom. Rev. 2007, 26, 370–388.(27) Takata, T.; Shimo-Oka, T.; Kojima, M.; Miki, K.; Fujii, N. Biochem. Biophys.

Res. Commun. 2006, 344, 263–271.(28) De Boni, S.; Oberthur, C.; Hamburger, M.; Scriba, G. K. E. J. Chromatogr.

A 2004, 1022, 95–102.(29) Adams, C. M.; Zubarev, R. A. Anal. Chem. 2005, 77, 4571–4580.(30) Winter, D.; Pipkorn, R.; Lehmann, W. D. J. Sep. Sci. 2009, 32, 1111–1119.

Scheme 1. Isomerization of Aspartic and IsoasparticAcids, and Deamidation of Asparagine via SuccinimideIntermediate

9779Analytical Chemistry, Vol. 81, No. 23, December 1, 2009

and MS for the detection of IsoAsp.31 The 18O labeling coupledto MS can also be used for detection of Asn deamidation andAsp isomerization;32,33 however, this can only be applied todetection of modification sites in the protein, but not to identifymodifications already existing in biological samples prior theanalysis. In addition, Alfaro et al. recently introduced a newmethod for the affinity enrichment of isoaspartyl proteins, wherechemoenzymatic detection using the PIMT enzyme and hydrazinetrapping were applied.34

Many of the methods for studying deamidation include massspectrometry, which over several decades has become one of themost powerful tools in the biological sciences. Although MS alonecannot normally distinguish the isomerssthe molecular massesof the isomers are identical and represent a single peak in massspectrumstandem MS has shown some successful results. Low-energy collisionally activated dissociation (CAD) was applied onthe basis of the difference in the abundance of immonium and band y ions,23,24 and also b + H2O and y - 46 ions (N-terminalto the isoAsp residue).22 These methods can be used todistinguish isoAsp- from Asp-containing peptides, although theabundance of all mentioned ions was strongly influenced bythe sequence of the peptides; in some cases, diagnostic ionswere not observed. Therefore, such methods require compara-tive analysis with standard peptides. This is often impossiblefor the analysis of biological systems.

An alternative tandem MS method was recently developed,where isoaspartic residues in peptides35 and proteins36 wereunambiguously differentiated and quantified37 by electroncapture dissociation (ECD). ECD was first introduced byZubarev38 and has since been widely implemented for routinestructural analysis of biological molecules.39 ECD of peptideions provides fragments (mostly c and z) complementary tothose (b and y) generated by other tandem MS methods suchas CAD40 or infrared multiphoton dissociation (IRMPD).41 TheECD fragmentation mechanism is believed to be fundamentallydifferent from other tandem methods. It is sometimes referredto as a “nonergodic” method,38 viz., energy gain upon electroncapture is used directly to cleave the N-CR bond rather thanbeing randomized along the molecule to break the weakestbond, as in CAD,40 an interpretation which is the subject ofintense debate.42-44 In an ECD experiment, multiply chargedgas-phase molecular ions produced by electrospray ionization

(ESI) react with low-energy electrons traditionally in a Fourier-transform ion cyclotron resonance mass spectrometer (FTICRMS),45,46 and recently implemented in a radio frequency iontrap.47 Electron capture is believed to occur in proximity tothe protonated site of the molecule followed by the hydrogen-bonded carbonyl group H• abstraction, which results in theN-CR bond cleavage, producing c and z• type fragments(Scheme 2a).38,48 Furthermore, radicals created by electroncapture can initiate multiple free radical rearrangements andsecondary fragmentation, which generate additional backboneas well as side chain cleavages.44,49 These can result in thecleavage of CR-C� or C�-Cγ bonds.50,51 Additionally, directCR-C� cleavage will lead to the formation of specificdiagnostic fragments in Asp- and isoAsp-containing peptides(Scheme 2b). Thus, in addition to cm and z•n-m fragments, aloss of 60 Da from the reduced species of the parent ion,[M + nH - 60](n-1)+•, will be observed; however, in the caseof isoaspartic acid, cleavage of CR-C� bond will result in abreakage of polypeptide chain and therefore cm + 57 andz•n-m - 57 fragment ions will be generated (Scheme 2).

Since the implementation of ECD, several new fragmentationtechniques based on similar electron-molecular ion interactionswere developed. For instance, electron transfer dissociation(ETD)52 was introduced as a substitute for ECD in instrumentsother than FTICR mass spectrometers. In ETD, the electronis transferred to the molecular ion from a radical anion toproduce a fragmentation pattern similar to that of ECD. Thismethod, now widely used in biological analysis,53,54 wassuccessfully applied in differentiation of aspartic and isoasparticacid in peptides.55 Furthermore, the electron ionization dis-sociation (EID)56 technique was recently developed. In EID,irradiation of the ions with high energy electrons (>∼20 eV)leads to a second ionization, followed by recapture of theelectron resulting in a highly excited species which undergoesrapid dissociation. This method was successfully applied tostandard singly and multiply charged peptides and proteins.Additionally, EID was implemented in a radio frequency linear

(31) Kameoka, D.; Ueda, T.; Imoto, T. J. Biochem. 2003, 134, 129–135.(32) Terashima, I.; Koga, A.; Nagai, H. Anal. Biochem. 2007, 368, 49–60.(33) Liu, P. R.; Regnier, F. E. Anal. Chem. 2003, 75, 4956–4963.(34) Alfaro, J. F.; Gillies, L. A.; Sun, H. G.; Dai, S. J.; Zang, T. Z.; Klaene, J. J.;

Kim, B. J.; Lowenson, J. D.; Clarke, S. G.; Karger, B. L.; Zhou, Z. S. Anal.Chem. 2008, 80, 3882–3889.

(35) Cournoyer, J. J.; Pittman, J. L.; Ivleva, V. B.; Fallows, E.; Waskell, L.; Costello,C. E.; O’Connor, P. B. Protein Sci. 2005, 14, 452–463.

(36) Cournoyer, J. J.; Lin, C.; O’Connor, P. B. Anal. Chem. 2006, 78, 1264–1271.

(37) Cournoyer, J. J.; Lin, C.; Bowman, M. J.; O’Connor, P. B. J. Am. Soc. MassSpectrom. 2007, 18, 48–56.

(38) Zubarev, R. A.; Kelleher, N. L.; McLafferty, F. W. J. Am. Chem. Soc. 1998,120, 3265–3266.

(39) Cooper, H. J.; Hakansson, K.; Marshall, A. G. Mass Spectrom. Rev. 2005,24, 201–222.

(40) Senko, M. W.; Speir, J. P.; McLafferty, F. W. Anal. Chem. 1994, 66, 2801–2808.

(41) Little, D. P.; Speir, J. P.; Senko, M. W.; Oconnor, P. B.; McLafferty, F. W.Anal. Chem. 1994, 66, 2809–2815.

(42) Turecek, F. J. Am. Chem. Soc. 2003, 125, 5954–5963.

(43) Breuker, K.; Oh, H. B.; Lin, C.; Carpenter, B. K.; McLafferty, F. W. Proc.Natl. Acad. Sci. U.S.A. 2004, 101, 14011–14016.

(44) Leymarie, N.; Costello, C. E.; O’Connor, P. B. J. Am. Chem. Soc. 2003,125, 8949–8958.

(45) Marshall, A. G. Int. J. Mass Spectrom. 2000, 200, 331–356.(46) Amster, I. J. J. Mass Spectrom. 1996, 31, 1325–1337.(47) Baba, T.; Hashimoto, Y.; Hasegawa, H.; Hirabayashi, A.; Waki, I. Anal. Chem.

2004, 76, 4263–4266.(48) Zubarev, R. A.; Haselmann, K. F.; Budnik, B.; Kjeldsen, F.; Jensen, F. Eur.

J. Mass Spectrom. 2002, 8, 337–349.(49) Cooper, H. J.; Hudgins, R. R.; Hakansson, K.; Marshall, A. G. Symposium

in Honor of Helmut Schwaz, August 2003, Berlin, Germany, pp 723-728.(50) Cooper, H. J.; Hudgins, R. R.; Hakansson, K.; Marshall, A. G. J. Am. Soc.

Mass Spectrom. 2002, 13, 241–249.(51) Falth, M.; Savitski, M. M.; Nielsen, M. L.; Kjeldsen, F.; Andren, P. E.;

Zubarev, R. A. Anal. Chem. 2008, 80, 8089–8094.(52) Syka, J. E. P.; Coon, J. J.; Schroeder, M. J.; Shabanowitz, J.; Hunt, D. F.

Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 9528–9533.(53) Mikesh, L. M.; Ueberheide, B.; Chi, A.; Coon, J. J.; Syka, J. E. P.;

Shabanowitz, J.; Hunt, D. F. Biochim. Biophys. Acta 2006, 1764, 1811–1822.

(54) Wiesner, J.; Premsler, T.; Sickmann, A. Proteomics 2008, 8, 4466–4483.(55) O’Connor, P. B.; Cournoyer, J. J.; Pitteri, S. J.; Chrisman, P. A.; McLuckey,

S. A. J. Am. Soc. Mass Spectrom. 2006, 17, 15–19.(56) Fung, E. Y. M.; Adams, C. M.; Zubarev, R. A. J. Am. Chem. Soc. 2009,

131, 9977–9985.

9780 Analytical Chemistry, Vol. 81, No. 23, December 1, 2009

ion trap instrument along with ECD and other fragmentationtechniques for complementary analysis.57

In the present study, ECD in FTICR MS was applied towardthe analysis of amyloid � peptides and their isomerized forms.Application of the top-down technique is a clever approach todistinguish isoAsp in a bigger A� (1-40 and 1-42), as it requiresno extra chemical steps that could complicate the analysis,

increase analysis time, and lead to the loss of a significant fractionof the sample. Minimization of the sample preparation is particu-larly advantageous for A�, as the peptide is very “sticky” due toits high hydrophobicity and can be completely lost duringadditional steps of the analysis. In addition to ECD, EID was alsoprobed for its ability to distinguish A� isomers.

EXPERIMENTAL SECTIONSample Preparation. Standard A�1-40 was purchased from

Sigma Aldrich (St. Louis, MO). The isomerized form of A�17-28(57) Enyenihi, A. A.; Baba, T.; Glish, G. L. Proceedings of the 57th ASMS

Conference, June 1-5, 2009, Philadelphia, PA.

Scheme 2. (a) ECD Major Pathway Produces c- and z-Fragment Ions with Identical m/z but Different ChemicalStructure of z-Fragment Isomers; (b) Proposed ECD Mechanism for the Formation of Diagnostic Fragment Ions

9781Analytical Chemistry, Vol. 81, No. 23, December 1, 2009

(LVFFAEisoDVGSNK custom synthesized) and A�1-42 withTottori-Japanese mutation at Asn7 were obtained from AnaSpec(San Jose, CA). A�1-40 was digested with sequencing-gradetrypsin (Roche Diagnostics, Indianapolis, IN) in 100 mM am-monium bicarbonate buffer at 1:50 enzyme:substrate ratio in µg/µL concentration, overnight at 37 °C with prior incubation at 65°C for 20 min. A�1-42[Asn7] was deamidated overnight at 37 °Cin 0.4% aqueous ammonium hydroxide with pH >10.2 in µg/µLconcentration. After digestion or deamidation, samples were driedin a SpeedVac system to stop the reaction and evaporate volatiles.

Mass Spectrometry. Mass spectra were acquired on a custombuilt qQq-FTICR MS with a nanospray source and a 7T activelyshielded magnet.58,59 Samples were electrosprayed at 1-5 µMconcentration in 50:50 MeOH:H2O with 1% formic acid. Ions were

isolated in the first quadrupole Q1, accumulated in the secondquadrupole Q2, and transmitted into the ICR cell, where theywere irradiated with the electrons emitted from an indirectlyheated dispenser cathode (Heatwave, Watsonville, CA) for ionfragmentation. The following ECD and EID parameters wereemployed: electron irradiation time, 35-100 ms; cathodepotential, -0.2 to -1.2 V (ECD) or -17 to -27 V (EID).

RESULTS AND DISCUSSIONDistinguishing the Isomers. ESI FTICR MS of the trypsin-

digested A�1-40 (DAEFRHDSGYEVHHQKLVFFAEDVGSNK-GAIIGLMVGGVV, hereafter abbreviated as A�40) revealed goodionization efficiency of the peptide 17-28 (Figure 1a, LVFFAED-VGSNK, abbreviated as A�17-28). The doubly charged precursorion of this peptide was isolated, accumulated, and irradiated withlow-energy electrons for the ECD event, as described in theExperimental Section. ECD mass spectra of this peptide showedabundant fragmentation with 95% sequence coverage (Figure 1a).In addition, specific neutral losses from the reduced species of

(58) O’Connor, P. B.; Pittman, J. L.; Thomson, B. A.; Budnik, B. A.; Cournoyer,J. C.; Jebanathirajah, J.; Lin, C.; Moyer, S.; Zhao, C. Rapid Commun. MassSpectrom. 2006, 20, 259–266.

(59) Jebanathirajah, J. A.; Pittman, J. L.; Thomson, B. A.; Budnik, B. A.; Kaur,P.; Rape, M.; Kirschner, M.; Costello, C. E.; O’Connor, P. B. J. Am. Soc.Mass Spectrom. 2005, 16, 1985–1999.

Figure 1. (a) ECD of A�17-28, (b) ECD of isoA�17-28, and (c) EID of isoA�17-28.

9782 Analytical Chemistry, Vol. 81, No. 23, December 1, 2009

the doubly charged tryptic peptide were observed, similar to thosedetected earlier:60 the loss of (C3H4O2 + NH3) from glutamicacid, (C3H7 + NH3) from leucine, CH3NO from asparagine, andC2H4O2 from aspartic acid, as well as NH3 from the N-terminus.

In order to perform a comparative analysis of peptide isomers,the same A�17-28 peptide was custom-synthesized with isoas-partic acid at position 23 [isoD23] (Figure 1b, LVFFAEisoDVG-SNK, abbreviated as isoA�17-28). ESI FTICR MS spectra of theisolated doubly charged A�17-28 and isoA�17-28 precursor ionsshowed identical m/z (data not shown). ECD of isoA�17-28generated the z6• - 57 diagnostic fragment ion, which was notfound in the A�17-28 peptide ECD spectrum. In contrast,carbon dioxide (CO2 ) 43.9898) was lost from the same z6•

fragment ion of A�17-28, but not from isoA�17-28 (Figure1a, b insets), indicating that CO2 loss is a diagnostic for thepresence of Asp, but not isoAsp, at a particular residue. Whilethe utility of CO2 loss as a diagnostic ion has been previouslyshown,37 it is not normally as reliable as the cn + 57/zm-n -57 diagnostic peaks, as isoAsp can also lose CO2. However,loss of CO2 from isoAsp appears to be a minor dissociationchannel (as evident by the small peak in a 100% isoAsp-containing peptide37). In the current study, loss of CO2 fromz6• is completely eliminated (

secondary structure in the gas phase and form numeroushydrogen bonds. Unless such ions are activated prior to ECDanalysis to disrupt secondary structure, strong hydrogen bondstend to keep c/z• fragment pairs in a complex, reducing thenet efficiency of ECD analysis. Nevertheless, successful analy-sis of A�42 would create a method suitable for different lengthA� peptides with the same or fewer number of amino acids.

Formation of isoaspartic acid in a standard A�42 was inducedto create a relevant test case for development of this method.Because the rate of aspartate isomerization is ∼40 times slowerthantherateofdeamidation,9thevariantcontainingTottori-Japanesemutation [Asn7] was chosen to accelerate formation of isoAsp bydeamidating the asparagine residue. The sample was incubatedin ammonium hydroxide at pH >10.2, overnight at 37 °C andanalyzed in ESI FTICR MS using ECD. [M + 5H]5+ molecularions were isolated, accumulated, and irradiated with low-energyelectrons. An ECD spectrum of the peptide is shown in Figure3. The mass of the parent ion increased by one or two units(deconvoluted mass), indicating the presence of single and doubledeamidation correspondingly (see insets). Double deamidationwas observed due to the presence of a second asparagine in thepeptide sequence at position 27. Partial deamidation of Asn27 wasalso observed, which is explained by the fact that the rate of HNS

deamidation is faster than the rate of SNK deamidation. Inparticular, according to the experimental data acquired at near-physiological conditions, the first-order deamidation half-life ofGHNSG and GSNKG model pentapeptides are 15.7 and 55.5days.12 For the purpose of this experiment, however, deamidationof Asn27 was of lesser interest, as was the completeness of thedeamidation reaction in general. Deamidation of asparagine [Asn7]was performed to induce the formation of isoaspartic acid at thisparticular residue, because in amyloid � peptides found in thecerebral plaque samples of Alzheimer patients aspartic acid is 75%isomerized to isoAsp at the residue position 7.13 As a result ofECD analysis of the deamidated A�42 peptide (Figure 3, upperspectrum), the diagnostic fragment c6 + 57 was observed (at2.2 ppm mass accuracy), demonstrating the formation ofisoaspartic acid at residue 7 and the ability of the currentmethod to detect and localize the modification. It should benoted that this model system experienced a highly nonphysi-ological (pH >10) environment, so that the deamidationmechanism is unlikely to follow the in vivo mechanism as thepH would have been likely to unfold the peptide, thus perturb-ing the native deamidation rate. However, the results of thisexperiment provide a needed baseline for future studies. It wasclearly demonstrated that ECD can be used to detect the

Figure 2. ECD of A�40.

9784 Analytical Chemistry, Vol. 81, No. 23, December 1, 2009

isomerized aspartate in a top-down experiment of the longestA� peptide. Exploring these studies further in terms ofquantitative analysis and combining this technique togetherwith HPLC separation will present a powerful tool for theanalysis of physiological fluids in Alzheimer’s disease patients.Comparative studies of isoAsp containing A� peptides in normaland diseased samples of different stages might reveal a novelperspective on Alzheimer’s disease initiation and progression.

CONCLUSIONS

The ECD diagnostic ions [M + 2H - 60]•+ and z6• - 57 wereobserved and successfully applied to differentiate the isomericforms of the amyloid � tryptic peptide 17-28. Differentiationof aspartic and isoaspartic acid residues is also possible withEID using the same diagnostic ion z6• - 57 as in ECD.Although the abundance of the diagnostic peak was relativelylow compare to ECD, EID could be particularly useful in the

Figure 3. ECD of A�42 deamidated at Asn7 and Asn27.

9785Analytical Chemistry, Vol. 81, No. 23, December 1, 2009

analysis of singly charged precursor ions, where ECD is notapplicable. Amyloid � 1-40 and 1-42 peptides can be analyzedby a top-down ECD approach without prior digestion andprovide substantial sequence coverage. Amyloid � 1-42isomerized at residue 7 can be differentiated by ECD usingthe c6 + 57 diagnostic ion. Both ECD and EID can clearlydefine the presence and the position of isoaspartic acid residuesin amyloid � peptides; therefore, they could be applied to thesamples of Alzheimer’s patients or healthy individuals for betterunderstanding of the disease.

ACKNOWLEDGMENTAuthors highly acknowledge Pavel Sargaev, Alex Cherkassky,

Konstantin Aizikov, and Eugene Moskovets for kind support andhelpful discussions. This work was supported by NIH/NCRR-P41RR10888, NIH/NHLBI-N01HV28178, and NIH/NIGMS-R01GM078293.

Received for review July 28, 2009. Accepted October 9,2009.

AC901677T

9786 Analytical Chemistry, Vol. 81, No. 23, December 1, 2009