- Email: [email protected]
Modification of cysteine residue in transthyretin and a synthetic peptide: analyses by electrospray ionization mass spectrometry
Biochimica et Biophysica Acta 1698 (2004) 45 – 53 www.bba-direct.com
Modification of cysteine residue in transthyretin and a synthetic peptide: analyses by electrospray ionization mass spectrometry Toyofumi Nakanishi a, Takako Sato b, Saburo Sakoda b, Masanori Yoshioka c, Akira Shimizu a,* a
Department of Clinical Pathology, Osaka Medical College, 2-7 Daigaku-cho, Takatsuki City, Osaka 569-8686, Japan Department of Neurology, Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan c Department of Analytical Chemistry, Faculty of Pharmaceutical Sciences, Setsunan University, Hirakata, Osaka 573-0101, Japan b
Received 20 August 2003; received in revised form 8 October 2003; accepted 10 October 2003
Abstract Cysteine and cystine in protein are modified to various derivatives in vitro and in vivo. By electrospray ionization mass spectrometry (ESI-MS), we previously found derivatives of serum transthyretin (TTR) in which cysteine residue at position 10 was changed to glycine residue and sulfocysteine residue. The change, cysteine to glycine, was unique and the origin of the sulfonic acid group was controversial. In the present paper, we show the molecular masses of various TTR derivatives including these two, and the modification process was studied using a synthetic peptide with the same sequence as cysteine containing part of TTR, i.e., SKCPLMVK. After incubation of the peptide at pH 8.3, various derivatives were generated, which showed changes of cysteine residue to glycine, dehydroalanine, S-thiocysteine, and Ssulfocysteine residues, which were confirmed by molecular mass and collision-induced dissociation spectra. Dehydroalanine may react with other amino acids and contribute to form cross-linking fibril, causing amyloidosis. D 2003 Elsevier B.V. All rights reserved. Keywords: Electrospray ionization mass spectrometry; S-sulfonated transthyretin; Dehydroalanine; Glycine; Amyloidosis
1. Introduction We have applied soft ionization mass spectrometry (MS) to detect variant and abnormally modified proteins in sera and in erythrocytes [1– 5]. During the course of these studies, we found a high level of S-sulfonated transthyretin (TTR) in patients with deficiency of molybdenum cofactor, which is essential for the function of enzymes including sulfite oxidase, and in patients with isolated sulfite oxidase deficiency [6,7]. In these patients, due to the lack of sulfite oxidase activity, sulfite accumulates and the presence of elevated levels of sulfite leads to accumulation of S-sulfocysteine formed by a direct reaction of sulfite with cysteine [8]. In sera without sulfite oxidase defect, S-sulfonated TTR also increased, which was accompanied by an increase in other oxidized forms of TTR, e.g., S-cysteinyl TTR, which may reflect the increased oxidation in vitro, but not the presence of elevated levels of sulfite in vivo [7]. This type of elevation of S-sulfonated TTR was also reported by The´berge et al. [9].
* Corresponding author. Tel.: +81-726-84-6448; fax: +81-726-846548. E-mail address: [email protected] (A. Shimizu). 1570-9639/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2003.10.005
We also reported another unique isoform of TTR, which was 46 u. smaller than free TTR, and the structure was identified to be a derivative in which cysteine residue at position 10 is changed to glycine residue [5]. As these two derivatives were detected simultaneously, we hypothesized that both are generated by split of mixed disulfide or accidentally formed disulfide-linked dimer of the TTR. These observations have prompted us to examine the modification process further. We analysed the TTR, and a peptide, SKCPLMVK, which corresponds to the sequence of a part of TTR, Ser 8 to Lys 15, by electrospray ionization mass spectrometry (ESI/MS). The modified structures generated by incubating in weak alkaline condition for rather long time were characterized by ESI/MS/MS.
2. Materials and methods 2.1. Specimens Purified TTR was commercially purchased (COSMO Bio, cat. no. MO-PAL-S, Lot no. OPO-4, Tokyo, Japan). A peptide, SKCPLMVK, which is the sequence of a part of TTR, Ser 8 to Lys 15, was synthesized by our order in
46
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
Fig. 1. Deconvoluted ESI mass spectra of transthyretin before (I) and after (II) incubation at 37 jC, pH 8.3, for 7 days. Incubated TTR was reduced by DTT (III). Peak a: TTR 46 (glycine), b: TTR 34 (dehydroalanine), c: free TTR, d: TTR + 32 (thiocysteine), e: TTR + 48 (cysteic acid), f: TTR + 64 (S-sulfinocysteine), g: TTR + 80 (S-sulfocysteine), h: TTR + 120(cystine), i: TTR + 177(cysteinyl-glycinyl-cysteine), j: TTR + 306 (glutathionyl-cysteine). Numbers are differences of molecular mass from unmodified TTR (in Table 1 shown by Dmass). The residues in parentheses are candidates substituting for cysteine 10 in TTR. k: TTR + 16, Mass difference corresponds to TTR with methionine sulfoxide substituted for methionine residue. Assignments were estimated according to the previous our papers [4 – 6] and by the similarity with peptide analysis shown in this paper.
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
Peptide Institute (Osaka, Japan, Lot no. 680-302031, purity: >95.6%, measured by HPLC using Zorbax). 2.2. Incubation of TTR and peptide TTR dissolved in 0.1 M NaHCO3 (pH = 8.3), 1 mg/ml, was incubated exposing to air at 37 jC, and the products were analyzed by LC/ESI mass spectrometry. The incubation condition was according to Yoshioka and Tamura [10], who reported the generation of S-sulfonated pantetheine from pantethine by incubation at pH 8.3. Preliminarily we examined the reaction at physiological pH, but the reaction was too slow to follow. The reaction at higher pH was also examined, but various hydrolyzed peptides confused the observation. The TTR incubated for 1 week was treated with dithiothreitol (DTT). DTT, 300 mM in water, was added to the solution at the volume ratio of 1:9 (DTT solution volume/ TTR solution volume). The mixture was incubated for 2 h at room temperature and was subsequently submitted to LC/ ESI mass spectrometry. The cysteine-containing model peptide, 1 mg/ml, was also incubated in the same conditions as for TTR and the products were analyzed by LC/ESI-MS/MS. 2.3. Reversed-phase HPLC and mass spectrometry For the analyses of TTR, on-line HPLC/ESI-MS was used, with a reversed-phase column (PLRP-S column, ˚ , 8 Am; obtained from Polymer 1 50 mm, 1000 A Laboratories, Shropshire, UK), as reported previously [2,4,7,11]. An ESI mass spectrometer, TSQ7000 (Thermo Quest, San Jose, CA), was utilized. TTR in an incubation buffer was acidified by adding acetic acid and was dissolved in acetonitrile – water (2:98) containing 2.0% acetic acid. The solvent system for online reversed-phase liquid chromatography was a linear gradient of solvent A mixed with solvent B from 5% B to 60% B in 40 min. Solvent A was 2.0% acetic acid and solvent B was 2.0% acetic acid in acetonitrile. The data were analyzed using a modified version of the SEQUEST computer algorithm to determine the site and type of sequence variation. Peptides were analyzed by a silica-based ODS HPLC column(XTerrak MS C18, 2.1 150 mm, 5 Am, Waters, Milford, MA) connected to a quadrupole ion trap mass spectrometer, LCQ, equipped with an ES ion source (Thermo Quest) [11]. The scanning range was m/z 500 – 2000 in 3 s. Peptides dissolved in acetonitrile – water (2:98), containing 0.1% formic acid, were injected and eluted with a gradient of 2 – 55% acetonitrile containing 0.1% formic acid over 55 min at a flow rate of 200 Al/ min. For each LC/MS analysis, 10 Al of ca. 10 pmol was loaded. To measure the CID spectra of the digests, the scanning range was set up as m/z 50– m/z 2500 in 3 s for the precursor ion scan and CID scan. ESI (LCQ) spectra
47
were recorded by scanning the quadrupole from m/z 300– 1800 u. To measure the CID spectra, the range for scanning was set from m/z of ca. 1/3 of precursor ion to m/z 1850 at 2 s/scan. Collision energy was set at 40% (of 5 mV). Calibration was performed using the peptide, MetArg-Phe-Ala and horse apo-myoglobin (Sigma, St. Louis, MO) for TSQ7000 and Met-Arg-Phe-Ala, caffeine, and Ultramark K162 (Thermo Quest) for LCQ. The structure of the peptides was assigned by MS/MS spectra using the program, Peptide Matching and Interpretation, ICIS and BIOTECH, Version 8.2.1 (Thermo Quest). All other chemicals were purchased from Nacalai Tesque (Kyoto, Japan).
3. Results Commercially obtained TTR was analyzed by LC/ESIMS before and after incubation. Fig. 1(I) shows the deconvoluted spectrum of TTR before incubation. Several peaks were seen. The molecular mass of peak c corresponded to unmodified TTR. The other ion peaks showed molecular sizes, 46 and 34 u. less than intact TTR, and 48, 120, 177, and 306 u. larger than intact TTR. We followed the generation of derivatives during incubation by ESI mass spectra every day. Fig. 1(II) shows a deconvoluted spectrum of transthyretin after 7-day incubation at 37 jC, pH 8.3. The peaks with molecular mass 120 u. larger than unmodified TTR (designated as TTR + 120, the following—the same), TTR + 155, and TTR + 306 decreased gradually and peaks TTR 46, TTR 34, TTR + 32, TTR + 48, TTR + 64 and TTR + 80 increased during incubation. The peak TTR + 80 was the most prominent at day 7. Previously we assigned TTR 46 as a derivative in which cysteine residue at position 10 is
Table 1 Modification of cysteine residue in transthyretin and the peptide, SKCPLMVK DMass
Transthyretin
Peptide
Peak (Fig. 1)
Fig. 2
Measured MH
Substitution of Cys
0 32 48 64 80 120 177
a b c d e f g h i
a b c d – – g – –
859.6 871.5 905.4 937.5 – – 985.5 – –
306 16 –
j k –
– k l
– 921.8 1807.5
glycine dehydroalanine free S-thiocysteine cysteic acid S-sulfinocysteine S-sulfocysteine cysteinyl cysteine cysteinyl-glycinyl cysteine glutathionyl cysteine methionine sulfoxide* disulfide-linked dimmer
46 34
– : not observed. * Methionine oxidation.
48
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
changed to glycine residue as described above [5]. In a previous paper, we assigned TTR + 80, and TTR + 120 as Ssulfonated TTR and S-cysteinyl-TTR, respectively [5– 7]. The molecular mass of 306 larger than free TTR corresponds to S-glutathionyl-TTR. Fig. 1(III) shows a deconvoluted spectrum of transthyretin treated with DTT after 7-day incubation. Peaks TTR + 32 and TTR + 80 decreased by reduction, and relative peak heights of unmodified TTR, TTR 46, and TTR + 48 were prominent. These changes were compatible to our proposal of assignments of these derivatives (see Table 1). The observed molecular mass of several signals around 13 600 in Fig. 1(II) and (III) marked by asterisks corresponded to a set of derivatives of TTR depleted of amino-terminal Gly-Pro (TTR 154), which were gener-
ated by cleavage in the ion source at carboxyl side of proline [12]. Fig. 2 shows mass chromatograms of the peptide before (Fig. 2(I)) and after (Fig. 2(II)) 7-day incubation. Fig. 2(III) shows the chromatogram of the peptide treated with DTT after 7-day incubation. Total ion monitoring (base peak) and selected ion chromatogram monitored with the corresponding MH+ ions are shown. Base peak monitoring shows one peak on the chromatogram before incubation, marked by c (peak designations were adjusted by mass difference to the experiment for TTR shown in Fig. 1). CID spectrum of this ion peak (data not shown), MH+ 905.4, confirmed the sequence exactly the same as that we ordered from Peptide Institute, and no modified peptides were contained before incubation. Fig. 2(II) shows the
Fig. 2. Mass chromatograms of ESI-MS of the peptide, SKCPLMVK, before (I) and after (II) incubation at 37 jC, pH 8.3, for 7 days, and after incubation followed by reduction with DTT (III). For assignment, see Table 1.
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
49
Fig. 2 (continued ).
generation of various derivatives of the peptide after 7-day incubation in 0.1 M NaHCO3, pH 8.3 at 37 jC. By the CID spectra we assigned the main components in each peak as shown in Table 1. The peptide with amino acid residue of molecular mass 34 u. less than intact peptide at cysteine position was detected in different peaks, as shown in Fig. 2(II), scan m/z 871 –872. We assumed that all are peptides with dehydroalanine by CID spectra of each peak. These peptides with the same MH+ and similar CID may be isomers separated by reversed-phase column. Peak d (m/ z 937 –938) shows also two peaks with same MH+ and essentially same CID spectra, and we also assumed that these are isomers. Peaks d, g and l in Fig. 2(II) disappeared by reduction as seen in Fig. 2(III). We concluded that these three peptides were disulfide bridged, which are in agreement with the assignment by CID spectra. MH+ and CID spectrum of peak k suggested that methionine residue is
changed to methionine sulfoxide in the corresponding peptide. This peak was clear only after reduction. Cysteine residue in the peptide with methionine sulfoxide may change to various oxidized forms before reduction, which may become uniform by reduction. In TTR, peak k (TTR + 16) was also high after reduction, we assumed the same process as the case of peptide. Fig. 3 shows the time course of alteration of the peak area ratio of each derivative of the peptide. On the abscissa, incubation days are shown and on the ordinate, ratio of peak area of mass chromatogram of MH+ ion of each component is shown. During incubation, the content of intact peptide decreased, disulfide-linked dimer increased first, followed by the increase of the other forms. Finally, peptides with dehydroalanine residue, glycine residue, Ssulfocysteine residue and S-thiocysteine residue were prominent.
50
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
Fig. 2 (continued ).
By incubation of the model peptide in saline, and in phosphate buffer, pH 7.4, at ambient temperature, peaks of disulfide dimer and peptide + 32 were increased (data not shown).
4. Discussion After incubation, the cysteine in TTR and the model peptide changed to various derivatives. The incubation at pH 8.3 we employed was according to Yoshioka and Tamura, who reported the generation of S-sulfonated pantetheine from pantethine. We detected the same oxidation products in physiological pH as at pH 8.3, but the reaction in physiological pH was too slow to follow the oxidation. In higher pH the peptide was partially hydrolyzed and the analysis was complicated. At pH 8.3 we could follow the change of oxidative forms of TTR and peptide more easily
for 1 week. We estimated that the reaction in weak alkaline pH is essentially the same for a reaction in physiological pH except reaction velocity. The reaction may occur in blood circulation, but may be slow. We usually detect small peaks of modified TTR with glycine substituted for cysteine, and S-sulfonated TTR, in samples prepared from sera by immunoprecipitation without exposure to alkaline pH as we observed these derivatives of TTR in commercially purchased TTR before incubation (Fig. 1(I)). In samples stood in room temperature at physiological pH before and after isolation, the peaks of these modified TTR were high. Therefore, the modification mainly occurs in vitro after blood collection. However, we suppose that in regions in vivo exposing to high oxygen pressure or to strong oxidative stress, the reaction may occur. The derivatives obtained from the peptide by incubation showed molecular mass differences corresponding to the
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
51
Fig. 3. Alteration of the peak area ratio of derivatives. Abscissa: incubation period, days. Ordinate: ratio of peak area of mass chromatogram of MH+ ion of each derivative and intact peptide per total. x: monomer intact, n: dimer, E: peptide with S-thiocysteine, : peptide with S-sulfocysteine, *: peptide with glycine, .: peptide with dehydroalanine.
same modification in TTR. The major derivatives were TTR or peptide 46, 34, + 34, and + 80 in which cysteine residue at position 10 is changed to glycine, dehydroalanine, S-thiocysteine, and S-sulfocysteine residues, respectively. These derivatives suggest that the transformation starts from a h-elimination of disulfide linkage to dehydroalanine and S-thiocysteine [13,14]. The h-elimination of disulfide linkage accelerates at alkaline pH and the pathway of h-elimination and oxidation in protein occurs nonenzymatically [13]. Many proteins are also reported to
be susceptible to the h-elimination in neutral pH. For example, a recent report showed that atrial natriuretic peptide is susceptible to h-elimination without exposing to alkaline pH [14]. The susceptibility to the h-elimination may depend on the neighboring amino acids or higher-order structure of protein. The probable pathway of this transformation is depicted in Fig. 4. In these pathways the step of glycine formation may be the most controversial. We found a literature to suggest the transformation from dehydroalanine to glycine, which was
Fig. 4. The proposed pathway of h-elimination and oxidation of cysteine residue in the peptide and TTR.
52
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
observed in the prosthetic group of L-phenylalanine ammonia-lyase from Rhodotorula glutinis [15]. However, the mechanism of this conversion may need to be studied further. In summary, we clarified in the present paper: (1) Observed molecular masses and CID spectra of various ion peaks generated from the peptide with amino acid sequence of a part of TTR coincided with products of helimination and their oxidation derivatives. Similar changes of molecular mass were observed also in TTR, which offers important information for application of the MS analysis to clinical diagnoses. (2) Without addition of sulfurous acid or cysteine, the S-sulfonated adduct generated, namely sulfur generated from protein or peptide itself via dimer formation. (3) We suggested for the first time the pathway from cysteine to glycine in protein and peptide. For further experiments, it may be important to elucidate the reaction of dehydroalanine-containing protein intermediate with another protein molecule [13,14]. The ion peak with dehydroalanine was as high as that with glycine in the spectrum after incubation of TTR (Fig. 1(II)), but in Fig. 1(I) and (III) it was relatively low. We have analyzed many samples of serum TTR prepared by immunoprecipitaion [4– 6] so far, but the peak with dehydroalanine was not always clear, although the peak with glycine was always clear. The analysis of serum TTR just after incubation at 37 jC for 24 h showed clear peak of TTR with dehydroalanine. These observations may suggest that dehydroalanine-containing TTR is easy to aggregate, and lost during storage. In previous report from other laboratories, the gel filtration with TTR extracted from amyloid tissue showed large-size aggregate by elution with guanidine-containing buffer (e.g. Refs. [16,17]). The existence of aggregate with larger size than dimer suggests that covalent bridge other than disulfide involves the formation of the covalent polymer. The reaction of dehydroalanine with other amino acids may be one possible mechanism for amyloid formation. One amyloidogenic TTR mutation, in which cysteine 10 is substituted to arginine, was reported [18]. Dehydroalanine is also generated from serine, and methyl-dehydroalanine from threonine [13,19]. The posttranslational modification of TTR 49 threonine to glycine was reported [17]. According to our hypothesis, this change includes the step of h-elimination, that is, change of threonine to methyl-dehydroalanine followed by methyl-dehydroalanine to glycine. Atland et al. [20,21] reported that reduction of TTR increased the stability of the folded monomeric and tetrameric conformations, and sulfite stabilized on TTR monomers and tetramers. Sulfite is a reducing reagent, decreasing the formation of disulfide conjugates. By our experiment Ssulfonated TTR was reduced to free TTR by addition of sulfite. These observations suggest that the oxidation of TTR is a factor of amyloid fibril formation, which coincides with our hypothesis. The´berg et al. [9] observed relatively high levels of S-sulfonated TTR in amyloid patients. Suhr et al. [22] showed that variant TTR is more susceptible to thiol conjugation than the wild type. Their observation also
coincides with our hypothesis, as thiol conjugation is susceptible to h-elimination. Acknowledgements This study was supported by a 2002-2003 grant-in-aid for Scientific Research (C) 14572189 (AS) and 13672332 (MY) from Ministry of Education, Culture, Sports, Science and Technology of Japan. References [1] T. Nakanishi, N. Okamoto, K. Tanaka, A. Shimizu, Laser desorption time-of-flight mass spectrometric analysis of transferrin precipitated with antiserum: a unique simple method to identify molecular weight variants, Biol. Mass Spectrom. 23 (1994) 230 – 233. [2] T. Nakanishi, M. Kishikawa, A. Miyazaki, A. Shimizu, Y. Ogawa, S. Sakoda, T. Ohi, H. Shoji, Simple and defined method to detect the SOD-1 mutants from patients with familial amyotrophic lateral sclerosis by mass spectrometry, J. Neurosci. Methods 81 (1998) 41 – 44. [3] M. Saraswathi, T. Nakanishi, A. Shimizu, Relative quantification of glycated Cu-Zn superoxide dismutase in erythrocytes by electrospray ionization mass spectrometry, Biochim. Biophys. Acta 1426 (1999) 483 – 490. [4] M. Kishikawa, T. Nakanishi, A. Miyazaki, A. Shimizu, M. Nakazato, K. Kangawa, H. Matsuo, Simple detection of abnormal serum transthyretin from patients with familial amyloidotic polyneuropathy by high-performance liquid chromatography/electrospray ionization mass spectrometry using material precipitated with specific antiserum, J. Mass Spectrom. 31 (1996) 112 – 114. [5] M. Kishikawa, T. Nakanishi, A. Miyazaki, A. Shimizu, A simple and reliable method of detecting variant transthyretins by multidimensional liquid chromatography coupled to electrospray ionization mass spectrometry, amyloid, Int. J. Exp. Clin. Invest. 6 (1999) 48 – 53. [6] M. Kishikawa, T. Nakanishi, A. Shimizu, M. Yoshino, Detection by mass spectrometry of highly increased amount of S-sulfonated transthyretin in serum from a patient with molybdenum cofactor deficiency, Pediatr. Res. 47 (2000) 492 – 494. [7] M. Kishikawa, J.O. Sass, N. Sakura, T. Nakanishi, A. Shimizu, M. Yoshioka, The peak height ratio of S-sulfonated transthyretin and other oxidized isoforms as a marker for molybdenum cofactor deficiency, measured by electrospray ionization mass spectrometry, Biochim. Biophys. Acta 1588 (2002) 135 – 138. [8] J.L. Johnson, M. Duran, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic and Molecular Bases of Inherited Disease, McGraw-Hill, New York, 2001, pp. 3163 – 3177. [9] R. The´berge, L. Connors, M. Skinner, J. Skare, C.E. Costello, Characterization of transthyretin mutants from serum using immunoprecipitation, HPLC/electrospray ionization and matrix-assisted laser desorption/ionization mass spectrometry, Anal. Chem. 71 (1999) 452 – 459. [10] M. Yoshioka, Z. Tamura, Bifidus factors in carrot: II. The structure of the factor in fraction IV, Chem. Pharm. Bull. (Tokyo) 19 (1971) 178 – 185. [11] T. Nakanishi, R. Koyama, T. Ikeda, A. Shimizu, Catalogue of soluble proteins in the human vitreous humor: comparison between diabetic retinopathy and macular hole, J. Chromatogr., B, Biomed. Sci. Appl. 776 (2002) 89 – 100. [12] K.J. Light-Walh, J.A. Loo, C.G. Esmonds, R.D. Smith, H.E. Witkowska, C.H.L. Shackleton, C.C. Wu, Collisionally activated dissociation and tandem mass spectrometry of intact hemoglobin-chain variant proteins with electrospray ionization, Biol. Mass Spectrom. 22 (1993) 112 – 120.
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53 [13] M. Friedman, Chemistry, biochemistry, nutrition, and microbiology of lysinoalanine, lanthionine, and histidinoalanine in food and other proteins, J. Agric. Food Chem. 47 (1999) 1295 – 1319. [14] S. Wu, D. Leung, L. Tretyakov, J. Hu, A. Guzzetta, Y.J. Wang, The formation and mechanism of multimerization in a freeze-dried peptide, Int. J. Pharm. 200 (2000) 1 – 16. [15] E.A. Havir, K.R. Hanson, L-Phenylalanine ammonia-lyase (maize, potato, and Rhodotorula glutinis). Studies of the prosthetic group with nitromethane, Biochemistry 14 (1975) 1620 – 1626. [16] P. Westermark, K. Sletten, B. Johansson, G.G. Cornwell III, Fibril in senile systemic amyloidosis is derived from normal transthyretin, Proc. Natl. Acad. Sci. U. S. A. 87 (1990) 2843 – 2845. [17] P.D. Goveric, F.C. Prelli, J. Wright, M. Pras, B. Frangione, Systemic senile amyloidosis identification of a new prealbumin (transthyretin) variant in cardiac tissue: Immunologic and biochemical similarity to one form of familial amyloidotic polyneuropathy, J. Clin. Invest. 83 (1989) 836 – 843.
53
[18] T. Uemichi, J.R. Murrell, S. Zeldenrust, M.D. Benson, A new mutant transthyretin (Arg 10) associated with familial amyloid polyneuropathy, J. Med. Genet. 29 (1992) 888 – 891. [19] M. Friedman, Lysinoalanine in food and in antimicrobial proteins, Adv. Exp. Med. Biol. 459 (1999) 145 – 159. [20] K. Altland, P. Winter, Potential treatment of transthyretin-type amyloidoses by sulfite, Neurogenetics 2 (1999) 183 – 188. [21] K. Altland, P. Winter, M.K. Sauerborn, Electrically neutral microheterogeneity of human plasma transthyretin (prealbumin) detected by isoelectric focusing in urea gradients, Electrophoresis 20 (1999) 1349 – 1364. [22] O.B. Suhr, I.H. Svedsen, P.-I. Ohlsson, J. Lendoire, P. Trigo, K. Tashima, P.J. Ranlov, Y. Ando, Impact of age and amyloidosis on thiol conjugation of transthyretin in hereditary transthyretin amyloidosis, amyloid, Int. J. Exp. Clin. Invest. 6 (1999) 187 – 191.
Modification of cysteine residue in transthyretin and a synthetic peptide: analyses by electrospray ionization mass spectrometry Toyofumi Nakanishi a, Takako Sato b, Saburo Sakoda b, Masanori Yoshioka c, Akira Shimizu a,* a
Department of Clinical Pathology, Osaka Medical College, 2-7 Daigaku-cho, Takatsuki City, Osaka 569-8686, Japan Department of Neurology, Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan c Department of Analytical Chemistry, Faculty of Pharmaceutical Sciences, Setsunan University, Hirakata, Osaka 573-0101, Japan b
Received 20 August 2003; received in revised form 8 October 2003; accepted 10 October 2003
Abstract Cysteine and cystine in protein are modified to various derivatives in vitro and in vivo. By electrospray ionization mass spectrometry (ESI-MS), we previously found derivatives of serum transthyretin (TTR) in which cysteine residue at position 10 was changed to glycine residue and sulfocysteine residue. The change, cysteine to glycine, was unique and the origin of the sulfonic acid group was controversial. In the present paper, we show the molecular masses of various TTR derivatives including these two, and the modification process was studied using a synthetic peptide with the same sequence as cysteine containing part of TTR, i.e., SKCPLMVK. After incubation of the peptide at pH 8.3, various derivatives were generated, which showed changes of cysteine residue to glycine, dehydroalanine, S-thiocysteine, and Ssulfocysteine residues, which were confirmed by molecular mass and collision-induced dissociation spectra. Dehydroalanine may react with other amino acids and contribute to form cross-linking fibril, causing amyloidosis. D 2003 Elsevier B.V. All rights reserved. Keywords: Electrospray ionization mass spectrometry; S-sulfonated transthyretin; Dehydroalanine; Glycine; Amyloidosis
1. Introduction We have applied soft ionization mass spectrometry (MS) to detect variant and abnormally modified proteins in sera and in erythrocytes [1– 5]. During the course of these studies, we found a high level of S-sulfonated transthyretin (TTR) in patients with deficiency of molybdenum cofactor, which is essential for the function of enzymes including sulfite oxidase, and in patients with isolated sulfite oxidase deficiency [6,7]. In these patients, due to the lack of sulfite oxidase activity, sulfite accumulates and the presence of elevated levels of sulfite leads to accumulation of S-sulfocysteine formed by a direct reaction of sulfite with cysteine [8]. In sera without sulfite oxidase defect, S-sulfonated TTR also increased, which was accompanied by an increase in other oxidized forms of TTR, e.g., S-cysteinyl TTR, which may reflect the increased oxidation in vitro, but not the presence of elevated levels of sulfite in vivo [7]. This type of elevation of S-sulfonated TTR was also reported by The´berge et al. [9].
* Corresponding author. Tel.: +81-726-84-6448; fax: +81-726-846548. E-mail address: [email protected] (A. Shimizu). 1570-9639/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2003.10.005
We also reported another unique isoform of TTR, which was 46 u. smaller than free TTR, and the structure was identified to be a derivative in which cysteine residue at position 10 is changed to glycine residue [5]. As these two derivatives were detected simultaneously, we hypothesized that both are generated by split of mixed disulfide or accidentally formed disulfide-linked dimer of the TTR. These observations have prompted us to examine the modification process further. We analysed the TTR, and a peptide, SKCPLMVK, which corresponds to the sequence of a part of TTR, Ser 8 to Lys 15, by electrospray ionization mass spectrometry (ESI/MS). The modified structures generated by incubating in weak alkaline condition for rather long time were characterized by ESI/MS/MS.
2. Materials and methods 2.1. Specimens Purified TTR was commercially purchased (COSMO Bio, cat. no. MO-PAL-S, Lot no. OPO-4, Tokyo, Japan). A peptide, SKCPLMVK, which is the sequence of a part of TTR, Ser 8 to Lys 15, was synthesized by our order in
46
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
Fig. 1. Deconvoluted ESI mass spectra of transthyretin before (I) and after (II) incubation at 37 jC, pH 8.3, for 7 days. Incubated TTR was reduced by DTT (III). Peak a: TTR 46 (glycine), b: TTR 34 (dehydroalanine), c: free TTR, d: TTR + 32 (thiocysteine), e: TTR + 48 (cysteic acid), f: TTR + 64 (S-sulfinocysteine), g: TTR + 80 (S-sulfocysteine), h: TTR + 120(cystine), i: TTR + 177(cysteinyl-glycinyl-cysteine), j: TTR + 306 (glutathionyl-cysteine). Numbers are differences of molecular mass from unmodified TTR (in Table 1 shown by Dmass). The residues in parentheses are candidates substituting for cysteine 10 in TTR. k: TTR + 16, Mass difference corresponds to TTR with methionine sulfoxide substituted for methionine residue. Assignments were estimated according to the previous our papers [4 – 6] and by the similarity with peptide analysis shown in this paper.
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
Peptide Institute (Osaka, Japan, Lot no. 680-302031, purity: >95.6%, measured by HPLC using Zorbax). 2.2. Incubation of TTR and peptide TTR dissolved in 0.1 M NaHCO3 (pH = 8.3), 1 mg/ml, was incubated exposing to air at 37 jC, and the products were analyzed by LC/ESI mass spectrometry. The incubation condition was according to Yoshioka and Tamura [10], who reported the generation of S-sulfonated pantetheine from pantethine by incubation at pH 8.3. Preliminarily we examined the reaction at physiological pH, but the reaction was too slow to follow. The reaction at higher pH was also examined, but various hydrolyzed peptides confused the observation. The TTR incubated for 1 week was treated with dithiothreitol (DTT). DTT, 300 mM in water, was added to the solution at the volume ratio of 1:9 (DTT solution volume/ TTR solution volume). The mixture was incubated for 2 h at room temperature and was subsequently submitted to LC/ ESI mass spectrometry. The cysteine-containing model peptide, 1 mg/ml, was also incubated in the same conditions as for TTR and the products were analyzed by LC/ESI-MS/MS. 2.3. Reversed-phase HPLC and mass spectrometry For the analyses of TTR, on-line HPLC/ESI-MS was used, with a reversed-phase column (PLRP-S column, ˚ , 8 Am; obtained from Polymer 1 50 mm, 1000 A Laboratories, Shropshire, UK), as reported previously [2,4,7,11]. An ESI mass spectrometer, TSQ7000 (Thermo Quest, San Jose, CA), was utilized. TTR in an incubation buffer was acidified by adding acetic acid and was dissolved in acetonitrile – water (2:98) containing 2.0% acetic acid. The solvent system for online reversed-phase liquid chromatography was a linear gradient of solvent A mixed with solvent B from 5% B to 60% B in 40 min. Solvent A was 2.0% acetic acid and solvent B was 2.0% acetic acid in acetonitrile. The data were analyzed using a modified version of the SEQUEST computer algorithm to determine the site and type of sequence variation. Peptides were analyzed by a silica-based ODS HPLC column(XTerrak MS C18, 2.1 150 mm, 5 Am, Waters, Milford, MA) connected to a quadrupole ion trap mass spectrometer, LCQ, equipped with an ES ion source (Thermo Quest) [11]. The scanning range was m/z 500 – 2000 in 3 s. Peptides dissolved in acetonitrile – water (2:98), containing 0.1% formic acid, were injected and eluted with a gradient of 2 – 55% acetonitrile containing 0.1% formic acid over 55 min at a flow rate of 200 Al/ min. For each LC/MS analysis, 10 Al of ca. 10 pmol was loaded. To measure the CID spectra of the digests, the scanning range was set up as m/z 50– m/z 2500 in 3 s for the precursor ion scan and CID scan. ESI (LCQ) spectra
47
were recorded by scanning the quadrupole from m/z 300– 1800 u. To measure the CID spectra, the range for scanning was set from m/z of ca. 1/3 of precursor ion to m/z 1850 at 2 s/scan. Collision energy was set at 40% (of 5 mV). Calibration was performed using the peptide, MetArg-Phe-Ala and horse apo-myoglobin (Sigma, St. Louis, MO) for TSQ7000 and Met-Arg-Phe-Ala, caffeine, and Ultramark K162 (Thermo Quest) for LCQ. The structure of the peptides was assigned by MS/MS spectra using the program, Peptide Matching and Interpretation, ICIS and BIOTECH, Version 8.2.1 (Thermo Quest). All other chemicals were purchased from Nacalai Tesque (Kyoto, Japan).
3. Results Commercially obtained TTR was analyzed by LC/ESIMS before and after incubation. Fig. 1(I) shows the deconvoluted spectrum of TTR before incubation. Several peaks were seen. The molecular mass of peak c corresponded to unmodified TTR. The other ion peaks showed molecular sizes, 46 and 34 u. less than intact TTR, and 48, 120, 177, and 306 u. larger than intact TTR. We followed the generation of derivatives during incubation by ESI mass spectra every day. Fig. 1(II) shows a deconvoluted spectrum of transthyretin after 7-day incubation at 37 jC, pH 8.3. The peaks with molecular mass 120 u. larger than unmodified TTR (designated as TTR + 120, the following—the same), TTR + 155, and TTR + 306 decreased gradually and peaks TTR 46, TTR 34, TTR + 32, TTR + 48, TTR + 64 and TTR + 80 increased during incubation. The peak TTR + 80 was the most prominent at day 7. Previously we assigned TTR 46 as a derivative in which cysteine residue at position 10 is
Table 1 Modification of cysteine residue in transthyretin and the peptide, SKCPLMVK DMass
Transthyretin
Peptide
Peak (Fig. 1)
Fig. 2
Measured MH
Substitution of Cys
0 32 48 64 80 120 177
a b c d e f g h i
a b c d – – g – –
859.6 871.5 905.4 937.5 – – 985.5 – –
306 16 –
j k –
– k l
– 921.8 1807.5
glycine dehydroalanine free S-thiocysteine cysteic acid S-sulfinocysteine S-sulfocysteine cysteinyl cysteine cysteinyl-glycinyl cysteine glutathionyl cysteine methionine sulfoxide* disulfide-linked dimmer
46 34
– : not observed. * Methionine oxidation.
48
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
changed to glycine residue as described above [5]. In a previous paper, we assigned TTR + 80, and TTR + 120 as Ssulfonated TTR and S-cysteinyl-TTR, respectively [5– 7]. The molecular mass of 306 larger than free TTR corresponds to S-glutathionyl-TTR. Fig. 1(III) shows a deconvoluted spectrum of transthyretin treated with DTT after 7-day incubation. Peaks TTR + 32 and TTR + 80 decreased by reduction, and relative peak heights of unmodified TTR, TTR 46, and TTR + 48 were prominent. These changes were compatible to our proposal of assignments of these derivatives (see Table 1). The observed molecular mass of several signals around 13 600 in Fig. 1(II) and (III) marked by asterisks corresponded to a set of derivatives of TTR depleted of amino-terminal Gly-Pro (TTR 154), which were gener-
ated by cleavage in the ion source at carboxyl side of proline [12]. Fig. 2 shows mass chromatograms of the peptide before (Fig. 2(I)) and after (Fig. 2(II)) 7-day incubation. Fig. 2(III) shows the chromatogram of the peptide treated with DTT after 7-day incubation. Total ion monitoring (base peak) and selected ion chromatogram monitored with the corresponding MH+ ions are shown. Base peak monitoring shows one peak on the chromatogram before incubation, marked by c (peak designations were adjusted by mass difference to the experiment for TTR shown in Fig. 1). CID spectrum of this ion peak (data not shown), MH+ 905.4, confirmed the sequence exactly the same as that we ordered from Peptide Institute, and no modified peptides were contained before incubation. Fig. 2(II) shows the
Fig. 2. Mass chromatograms of ESI-MS of the peptide, SKCPLMVK, before (I) and after (II) incubation at 37 jC, pH 8.3, for 7 days, and after incubation followed by reduction with DTT (III). For assignment, see Table 1.
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
49
Fig. 2 (continued ).
generation of various derivatives of the peptide after 7-day incubation in 0.1 M NaHCO3, pH 8.3 at 37 jC. By the CID spectra we assigned the main components in each peak as shown in Table 1. The peptide with amino acid residue of molecular mass 34 u. less than intact peptide at cysteine position was detected in different peaks, as shown in Fig. 2(II), scan m/z 871 –872. We assumed that all are peptides with dehydroalanine by CID spectra of each peak. These peptides with the same MH+ and similar CID may be isomers separated by reversed-phase column. Peak d (m/ z 937 –938) shows also two peaks with same MH+ and essentially same CID spectra, and we also assumed that these are isomers. Peaks d, g and l in Fig. 2(II) disappeared by reduction as seen in Fig. 2(III). We concluded that these three peptides were disulfide bridged, which are in agreement with the assignment by CID spectra. MH+ and CID spectrum of peak k suggested that methionine residue is
changed to methionine sulfoxide in the corresponding peptide. This peak was clear only after reduction. Cysteine residue in the peptide with methionine sulfoxide may change to various oxidized forms before reduction, which may become uniform by reduction. In TTR, peak k (TTR + 16) was also high after reduction, we assumed the same process as the case of peptide. Fig. 3 shows the time course of alteration of the peak area ratio of each derivative of the peptide. On the abscissa, incubation days are shown and on the ordinate, ratio of peak area of mass chromatogram of MH+ ion of each component is shown. During incubation, the content of intact peptide decreased, disulfide-linked dimer increased first, followed by the increase of the other forms. Finally, peptides with dehydroalanine residue, glycine residue, Ssulfocysteine residue and S-thiocysteine residue were prominent.
50
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
Fig. 2 (continued ).
By incubation of the model peptide in saline, and in phosphate buffer, pH 7.4, at ambient temperature, peaks of disulfide dimer and peptide + 32 were increased (data not shown).
4. Discussion After incubation, the cysteine in TTR and the model peptide changed to various derivatives. The incubation at pH 8.3 we employed was according to Yoshioka and Tamura, who reported the generation of S-sulfonated pantetheine from pantethine. We detected the same oxidation products in physiological pH as at pH 8.3, but the reaction in physiological pH was too slow to follow the oxidation. In higher pH the peptide was partially hydrolyzed and the analysis was complicated. At pH 8.3 we could follow the change of oxidative forms of TTR and peptide more easily
for 1 week. We estimated that the reaction in weak alkaline pH is essentially the same for a reaction in physiological pH except reaction velocity. The reaction may occur in blood circulation, but may be slow. We usually detect small peaks of modified TTR with glycine substituted for cysteine, and S-sulfonated TTR, in samples prepared from sera by immunoprecipitation without exposure to alkaline pH as we observed these derivatives of TTR in commercially purchased TTR before incubation (Fig. 1(I)). In samples stood in room temperature at physiological pH before and after isolation, the peaks of these modified TTR were high. Therefore, the modification mainly occurs in vitro after blood collection. However, we suppose that in regions in vivo exposing to high oxygen pressure or to strong oxidative stress, the reaction may occur. The derivatives obtained from the peptide by incubation showed molecular mass differences corresponding to the
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
51
Fig. 3. Alteration of the peak area ratio of derivatives. Abscissa: incubation period, days. Ordinate: ratio of peak area of mass chromatogram of MH+ ion of each derivative and intact peptide per total. x: monomer intact, n: dimer, E: peptide with S-thiocysteine, : peptide with S-sulfocysteine, *: peptide with glycine, .: peptide with dehydroalanine.
same modification in TTR. The major derivatives were TTR or peptide 46, 34, + 34, and + 80 in which cysteine residue at position 10 is changed to glycine, dehydroalanine, S-thiocysteine, and S-sulfocysteine residues, respectively. These derivatives suggest that the transformation starts from a h-elimination of disulfide linkage to dehydroalanine and S-thiocysteine [13,14]. The h-elimination of disulfide linkage accelerates at alkaline pH and the pathway of h-elimination and oxidation in protein occurs nonenzymatically [13]. Many proteins are also reported to
be susceptible to the h-elimination in neutral pH. For example, a recent report showed that atrial natriuretic peptide is susceptible to h-elimination without exposing to alkaline pH [14]. The susceptibility to the h-elimination may depend on the neighboring amino acids or higher-order structure of protein. The probable pathway of this transformation is depicted in Fig. 4. In these pathways the step of glycine formation may be the most controversial. We found a literature to suggest the transformation from dehydroalanine to glycine, which was
Fig. 4. The proposed pathway of h-elimination and oxidation of cysteine residue in the peptide and TTR.
52
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53
observed in the prosthetic group of L-phenylalanine ammonia-lyase from Rhodotorula glutinis [15]. However, the mechanism of this conversion may need to be studied further. In summary, we clarified in the present paper: (1) Observed molecular masses and CID spectra of various ion peaks generated from the peptide with amino acid sequence of a part of TTR coincided with products of helimination and their oxidation derivatives. Similar changes of molecular mass were observed also in TTR, which offers important information for application of the MS analysis to clinical diagnoses. (2) Without addition of sulfurous acid or cysteine, the S-sulfonated adduct generated, namely sulfur generated from protein or peptide itself via dimer formation. (3) We suggested for the first time the pathway from cysteine to glycine in protein and peptide. For further experiments, it may be important to elucidate the reaction of dehydroalanine-containing protein intermediate with another protein molecule [13,14]. The ion peak with dehydroalanine was as high as that with glycine in the spectrum after incubation of TTR (Fig. 1(II)), but in Fig. 1(I) and (III) it was relatively low. We have analyzed many samples of serum TTR prepared by immunoprecipitaion [4– 6] so far, but the peak with dehydroalanine was not always clear, although the peak with glycine was always clear. The analysis of serum TTR just after incubation at 37 jC for 24 h showed clear peak of TTR with dehydroalanine. These observations may suggest that dehydroalanine-containing TTR is easy to aggregate, and lost during storage. In previous report from other laboratories, the gel filtration with TTR extracted from amyloid tissue showed large-size aggregate by elution with guanidine-containing buffer (e.g. Refs. [16,17]). The existence of aggregate with larger size than dimer suggests that covalent bridge other than disulfide involves the formation of the covalent polymer. The reaction of dehydroalanine with other amino acids may be one possible mechanism for amyloid formation. One amyloidogenic TTR mutation, in which cysteine 10 is substituted to arginine, was reported [18]. Dehydroalanine is also generated from serine, and methyl-dehydroalanine from threonine [13,19]. The posttranslational modification of TTR 49 threonine to glycine was reported [17]. According to our hypothesis, this change includes the step of h-elimination, that is, change of threonine to methyl-dehydroalanine followed by methyl-dehydroalanine to glycine. Atland et al. [20,21] reported that reduction of TTR increased the stability of the folded monomeric and tetrameric conformations, and sulfite stabilized on TTR monomers and tetramers. Sulfite is a reducing reagent, decreasing the formation of disulfide conjugates. By our experiment Ssulfonated TTR was reduced to free TTR by addition of sulfite. These observations suggest that the oxidation of TTR is a factor of amyloid fibril formation, which coincides with our hypothesis. The´berg et al. [9] observed relatively high levels of S-sulfonated TTR in amyloid patients. Suhr et al. [22] showed that variant TTR is more susceptible to thiol conjugation than the wild type. Their observation also
coincides with our hypothesis, as thiol conjugation is susceptible to h-elimination. Acknowledgements This study was supported by a 2002-2003 grant-in-aid for Scientific Research (C) 14572189 (AS) and 13672332 (MY) from Ministry of Education, Culture, Sports, Science and Technology of Japan. References [1] T. Nakanishi, N. Okamoto, K. Tanaka, A. Shimizu, Laser desorption time-of-flight mass spectrometric analysis of transferrin precipitated with antiserum: a unique simple method to identify molecular weight variants, Biol. Mass Spectrom. 23 (1994) 230 – 233. [2] T. Nakanishi, M. Kishikawa, A. Miyazaki, A. Shimizu, Y. Ogawa, S. Sakoda, T. Ohi, H. Shoji, Simple and defined method to detect the SOD-1 mutants from patients with familial amyotrophic lateral sclerosis by mass spectrometry, J. Neurosci. Methods 81 (1998) 41 – 44. [3] M. Saraswathi, T. Nakanishi, A. Shimizu, Relative quantification of glycated Cu-Zn superoxide dismutase in erythrocytes by electrospray ionization mass spectrometry, Biochim. Biophys. Acta 1426 (1999) 483 – 490. [4] M. Kishikawa, T. Nakanishi, A. Miyazaki, A. Shimizu, M. Nakazato, K. Kangawa, H. Matsuo, Simple detection of abnormal serum transthyretin from patients with familial amyloidotic polyneuropathy by high-performance liquid chromatography/electrospray ionization mass spectrometry using material precipitated with specific antiserum, J. Mass Spectrom. 31 (1996) 112 – 114. [5] M. Kishikawa, T. Nakanishi, A. Miyazaki, A. Shimizu, A simple and reliable method of detecting variant transthyretins by multidimensional liquid chromatography coupled to electrospray ionization mass spectrometry, amyloid, Int. J. Exp. Clin. Invest. 6 (1999) 48 – 53. [6] M. Kishikawa, T. Nakanishi, A. Shimizu, M. Yoshino, Detection by mass spectrometry of highly increased amount of S-sulfonated transthyretin in serum from a patient with molybdenum cofactor deficiency, Pediatr. Res. 47 (2000) 492 – 494. [7] M. Kishikawa, J.O. Sass, N. Sakura, T. Nakanishi, A. Shimizu, M. Yoshioka, The peak height ratio of S-sulfonated transthyretin and other oxidized isoforms as a marker for molybdenum cofactor deficiency, measured by electrospray ionization mass spectrometry, Biochim. Biophys. Acta 1588 (2002) 135 – 138. [8] J.L. Johnson, M. Duran, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic and Molecular Bases of Inherited Disease, McGraw-Hill, New York, 2001, pp. 3163 – 3177. [9] R. The´berge, L. Connors, M. Skinner, J. Skare, C.E. Costello, Characterization of transthyretin mutants from serum using immunoprecipitation, HPLC/electrospray ionization and matrix-assisted laser desorption/ionization mass spectrometry, Anal. Chem. 71 (1999) 452 – 459. [10] M. Yoshioka, Z. Tamura, Bifidus factors in carrot: II. The structure of the factor in fraction IV, Chem. Pharm. Bull. (Tokyo) 19 (1971) 178 – 185. [11] T. Nakanishi, R. Koyama, T. Ikeda, A. Shimizu, Catalogue of soluble proteins in the human vitreous humor: comparison between diabetic retinopathy and macular hole, J. Chromatogr., B, Biomed. Sci. Appl. 776 (2002) 89 – 100. [12] K.J. Light-Walh, J.A. Loo, C.G. Esmonds, R.D. Smith, H.E. Witkowska, C.H.L. Shackleton, C.C. Wu, Collisionally activated dissociation and tandem mass spectrometry of intact hemoglobin-chain variant proteins with electrospray ionization, Biol. Mass Spectrom. 22 (1993) 112 – 120.
T. Nakanishi et al. / Biochimica et Biophysica Acta 1698 (2004) 45–53 [13] M. Friedman, Chemistry, biochemistry, nutrition, and microbiology of lysinoalanine, lanthionine, and histidinoalanine in food and other proteins, J. Agric. Food Chem. 47 (1999) 1295 – 1319. [14] S. Wu, D. Leung, L. Tretyakov, J. Hu, A. Guzzetta, Y.J. Wang, The formation and mechanism of multimerization in a freeze-dried peptide, Int. J. Pharm. 200 (2000) 1 – 16. [15] E.A. Havir, K.R. Hanson, L-Phenylalanine ammonia-lyase (maize, potato, and Rhodotorula glutinis). Studies of the prosthetic group with nitromethane, Biochemistry 14 (1975) 1620 – 1626. [16] P. Westermark, K. Sletten, B. Johansson, G.G. Cornwell III, Fibril in senile systemic amyloidosis is derived from normal transthyretin, Proc. Natl. Acad. Sci. U. S. A. 87 (1990) 2843 – 2845. [17] P.D. Goveric, F.C. Prelli, J. Wright, M. Pras, B. Frangione, Systemic senile amyloidosis identification of a new prealbumin (transthyretin) variant in cardiac tissue: Immunologic and biochemical similarity to one form of familial amyloidotic polyneuropathy, J. Clin. Invest. 83 (1989) 836 – 843.
53
[18] T. Uemichi, J.R. Murrell, S. Zeldenrust, M.D. Benson, A new mutant transthyretin (Arg 10) associated with familial amyloid polyneuropathy, J. Med. Genet. 29 (1992) 888 – 891. [19] M. Friedman, Lysinoalanine in food and in antimicrobial proteins, Adv. Exp. Med. Biol. 459 (1999) 145 – 159. [20] K. Altland, P. Winter, Potential treatment of transthyretin-type amyloidoses by sulfite, Neurogenetics 2 (1999) 183 – 188. [21] K. Altland, P. Winter, M.K. Sauerborn, Electrically neutral microheterogeneity of human plasma transthyretin (prealbumin) detected by isoelectric focusing in urea gradients, Electrophoresis 20 (1999) 1349 – 1364. [22] O.B. Suhr, I.H. Svedsen, P.-I. Ohlsson, J. Lendoire, P. Trigo, K. Tashima, P.J. Ranlov, Y. Ando, Impact of age and amyloidosis on thiol conjugation of transthyretin in hereditary transthyretin amyloidosis, amyloid, Int. J. Exp. Clin. Invest. 6 (1999) 187 – 191.