JOURNAL OF MASS SPECTROMETRY J. Mass Spectrom. 2005; 40: 142–145 Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jms.732 Hydrolysis of the amyloid b-peptide (Ab) 1–40 between Asp23–Val24 produces non-aggregating fragments. An electrospray mass spectrometric study Waltteri Hosia, 1 William J. Griffiths 1,2 and Jan Johansson 1,3* 1 Department of Medical Biochemistry and Biophysics, Karolinska Institutet, S-171 77 Stockholm, Sweden 2 School of Pharmacy, University of London, 29/39 Brunswick Square, London, WC1N 1AX UK 3 Department of Molecular Biosciences, Swedish University of Agricultural Sciences, BMC, S-751 23 Uppsala, Sweden Received 6 April 2004; Accepted 29 July 2004 The aggregation of full-length (residues 1–40) amyloid b-peptide (Ab) and fragments corresponding to residues 1–23 and 24–40 was studied by electrospray mass spectrometry, using gramicidin as a non-aggregating reference. Following a lag period, Ab(1–40) at 140 μM concentration aggregates with apparent first-order kinetics. Under acidic conditions Ab(1–40) undergoes spontaneous cleavage between Asp23–Val24 and to a lesser extent also at two other Asp–X motifs. Incubation in acidic H 2 18 O showed incorporation of 18 O in fragment Ab(1–23), confirming that the Asp23–Val24 peptide bond had been hydrolyzed. Incubation of synthetic Ab(1–23) and Ab(24–40) peptides with Ab(1–40) showed that Ab(24–40) remained in solution for several months, that Ab(1–23) partly disappeared from solution, whereas Ab(1–40) completely disappeared. Further, treatment of sedimentable aggregates formed after co-incubation of the three peptides with hexafluoro-2-propanol or formic acid recovered the intensity of Ab(1–40). These data support previous studies showing that the region of Ab encompassing residues 16 – 24 is necessary for aggregation into amyloid fibrils. Copyright  2005 John Wiley & Sons, Ltd. KEYWORDS: amyloid fibril; protein aggregation; Alzheimer’s disease INTRODUCTION Amyloid fibrils are insoluble, proteolysis-resistant polymers resulting from aggregation of specific polypeptides. Fibril formation of the predominantly 40 or 42 amino acid long amyloid ˇ-peptide (Aˇ) is thought to be an underlying event in the development of Alzheimer’s disease. 1 Aˇ peptide is produced by cellular processing of the amyloid precur- sor protein (APP) and lysosomal components appear to be involved in this process. 2 The Aˇ aggregation process com- prises an initial lag phase during which aggregation nuclei are formed, followed by formation of prefibrillar polymers and eventually assembly into fibrils. 3–5 Recent studies sug- gest that the Aˇ fibrils, and in particular soluble prefibrillar polymers, are responsible for cell toxicity. 6–9 In solution, Aˇ adopts mainly a random structure but helical segments cov- ering approximately residues 15–23 and 30–35 are formed in the presence of unpolar solvents or detergents. 5 Aˇ(1–40) can be divided into an N-terminal hydrophilic part (residues 1 – 28, DAEFRHDSGYEVHHQKLVFFAEDVGSNK) and a C- terminal hydrophobic part (residues 29–40, GAIIGLMVG- GVV). The hydrophobic part is membrane-embedded in the APP. It has been shown that the central region around residues 16 – 24 is necessary for Aˇ fibril formation. 10 – 13 Early L Correspondence to: Jan Johansson, Department of Molecular Biosciences, Swedish University of Agricultural Sciences, BMC, S-751 23 Uppsala, Sweden. E-mail: jan.johansson@vmk.slu.se Contract/grant sponsor: Swedish Research Council. on, it was observed that C-terminally extended Aˇ(1–42) is more prone to aggregate than Aˇ(1–40). 4 This phenomenon probably underlies the association between APP mutations that result in increased production of Aˇ(1–42) and familial forms of Alzheimer’s disease. The two amino acid residue extension in Aˇ(1–42) (isoleucine and alanine) increases the hydrophobicity of the peptide. The development of mass spectrometric methods for polypeptide analysis has been rapid over the last two decades. 14 The technique not only allows the determination of peptide primary structure at the sub-picomole level, but can also be used to obtain relative quantitative data. While for obtaining absolute quantitative data isotope dilution mass spectrometry is the method of choice, useful information can be acquired by making ion abundance measurements against non-identical internal standards. For example, data on peptide aggregation have been derived by comparing the ion abundance of an aggregating peptide against a non-aggregating reference standard. 15,16 Here we report first-order kinetics for Aˇ(1–40) aggregation observed by electrospray mass spectrometry (ESMS), characterization of spontaneous Aˇ cleavage products and the stability of Aˇ(1 – 23) and in particular Aˇ(24–40). EXPERIMENTAL Synthetic Aˇ(1–40) (Bachem, Switzerland, or Anaspec, San Jose, CA, USA), Aˇ(1–23), and Aˇ(24–40) (Thermo Hybaid, Copyright  2005 John Wiley & Sons, Ltd.