Skip to main content
Mass Spectrometry logoLink to Mass Spectrometry
. 2012 Jun 29;1(1):A0001. doi: 10.5702/massspectrometry.A0001

Influence of Secondary Structure on In-Source Decay of Protein in Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry

Mitsuo Takayama 1,*, Issey Osaka 1, Motoshi Sakakura 1
PMCID: PMC3775823  PMID: 24349902

Abstract

The susceptibility of the N–Cα bond of the peptide backbone to specific cleavage by in-source decay (ISD) in matrix-assisted laser desorption/ionization mass spectrometry (MALDI MS) was studied from the standpoint of the secondary structure of three proteins. A naphthalene derivative, 5-amino-1-naphtol (5,1-ANL), was used as the matrix. The resulting c′-ions, which originate from the cleavage at N–Cα bonds in flexible secondary structures such as turn and bend, and are free from intra-molecular hydrogen-bonded α-helix structure, gave relatively intense peaks. Furthermore, ISD spectra of the proteins showed that the N–Cα bonds of specific amino acid residues, namely Gly–Xxx, Xxx–Asp, and Xxx–Asn, were more susceptible to MALDI-ISD than other amino acid residues. This is in agreement with the observation that Gly, Asp and Asn residues usually located in turns, rather than α-helix. The results obtained indicate that protein molecules embedded into the matrix crystal in the MALDI experiments maintain their secondary structures as determined by X-ray crystallography, and that MALDI-ISD has the capability for providing information concerning the secondary structure of protein.

Keywords: matrix-assisted laser desorption/ionization, in-source decay, protein, secondary structure

Introduction

Mass spectrometry (MS) coupled with soft ionization methods, such as matrix-assisted laser desorption/ionization (MALDI)13) and electrospray ionization (ESI)4,5) represents a powerful tool that can be used for identifying thermally labile macromolecules such as proteins, nucleic acids and oligosaccharides. In particular, these methods provide indispensable approaches for proteome analysis via peptide-mass fingerprinting6,7) and amino acid sequencing with tandem mass spectrometry.8,9) In-source decay (ISD) coupled with MALDI10) is recognized as a top-down approach for obtaining internal amino acid sequences from intact proteins without any pre-degradation.1113) MALDI-ISD uses hydrogen radical transfer from the matrix to analyte molecules.14) The hydrogen radicals are generated from hydroxyl and amino groups at the 5-position of matrix molecules, which include 2,5-dihydroxybenzoic acid (2,5-DHB),15) 1,5-diaminonaphthalene (1,5-DAN)16) and 5-aminosalicylic acid (5-ASA),17) via the absorption of ultraviolet (UV) laser photons. The resulting hydrogen radicals bind to non-bonding electrons of the carbonyl oxygen of the peptide backbone. The principle processes of MALDI-ISD can be presented as follows:

graphic file with name massspectrometry-1-1-A0001-e001.jpg (1)
graphic file with name massspectrometry-1-1-A0001-e002.jpg (2)

where B and H· represent a matrix molecule and a hydrogen radical, respectively. The species M>C=O represents a carbonyl oxygen on the backbone of the peptide molecule M. The resulting radical species, M–C·–OH, causes a prompt cleavage within several tens of nanoseconds in the ion source. The unpaired electron as a radical species is localized on the carbonyl carbon, and radical-induced cleavage then follows, leading to the production of c′- and z′-series ions (Scheme 1). Therefore, the rate-determining step for MALDI-ISD is the hydrogen attachment reaction (2).

Scheme 1. Proposed mechanism for the formation of c′- and z′-series ions in MALDI-ISD.

Scheme 1. Proposed mechanism for the formation of c′- and z′-series ions in MALDI-ISD.

In a preliminary study, one of the authors proposed that in the MALDI-ISD of myoglobin the presence of intra-molecular hydrogen-bonded secondary structures such as an α-helix, may inhibit cleavage at the N–Cα bond of the peptide backbone,12) because the intra-molecular hydrogen bonded states may inhibit hydrogen radicals from being attached to the carbonyl oxygen. The assumption described above is of importance from the standpoint of protein structures embedded in a matrix crystal. It is of interest to consider whether an analyte protein embedded in a matrix crystal maintains its helical structure or not. The present report describes the influence of the secondary structure of a protein on MALDI-ISD. In order to examine the influence of secondary structure, some relatively α-helix-rich proteins, namely, myoglobin and bovine serum albumin (BSA), and an α-helix-poor protein cytochrome c, were used. Furthermore, a new matrix material 5-amino-1-naphthol (5,1-ANL) suitable for the MALDI-ISD of proteins was also introduced for the examination of the influence of the secondary structure of a protein on ISD.

Experimental

Materials

The MALDI matrices 5,1-ANL, 1,5-DAN and 2,5-DHB were purchased from Tokyo Kasei (Tokyo, Japan). Trifluoroacetic acid (TFA) and acetonitrile were purchased from Wako Pure Chemicals (Osaka, Japan). Water used in all experiments was purified using a MilliQ water purification system from Millipore (Billerica, MA, USA). Equine apo-myoglobin (Mr 16951.4), equine cytochrome c (Mr 12360.4), and BSA (Mr 66432.3) were purchased from Sigma (Milwaukee, WI, USA). All reagents were used without further purification.

Sample preparation

In a typical experiment, the analyte was dissolved in water at a concentration of 20 pmol/µL. The matrix material was dissolved in water–acetonitrile (1 : 1, v/v) containing 0.1% TFA at a concentration of 10 mg/mL. A volume of 1 µL of sample solution was deposited on a stainless steel plate and the solvents were removed by evaporation in air at room temperature. A 0.5 µL aliquot of analyte solution was deposited on a stainless-steel MALDI target and left to dry. After complete evaporation of the solvent, 0.5 µL of the matrix solution in acetone was deposited on the dried proteins.

MALDI MS

MALDI-ISD spectra were acquired on a AXIMA-CFR (Shimadzu, Kyoto, Japan) time-of-flight mass spectrometer equipped with a nitrogen laser (337 nm wavelength) operating at a pulse rate of 10 Hz. The pulse width of the laser was 4 ns. The diameter of the laser spot on the target substrate was approximately 100 µm. The ions generated by MALDI were accelerated using 20 kV with delayed extraction. The analyzer was operated in the linear mode and the ions were detected using a microchannel plate detector. A total of 500 shots were accumulated for each mass spectral acquisition. The reproducibility of the ISD spectra was confirmed.

Results and Discussion

Suitable matrix for MALDI-ISD of protein

Several matrix materials suitable for use in MALDI-ISD experiments, such as 2,5-DHB,15) 1,5-DAN16) and 5-ASA,17) have been reported to date. Of these matrices, a naphthalene derivative, 1,5-DAN, appears to be the most suitable matrix for ISD from the standpoint of hydrogen radical donating ability.18) Here we compared the ISD spectra of equine apo-myoglobin (Mr 16951.4) obtained with some conventional matrices such as 2,5-DHB and 1,5-DAN to that obtained with another naphthalene derivative 5,1-ANL used as a new matrix, as shown in Fig. 1. Since 5,1-ANL contains an amino group at the 5-position and a hydroxyl group at the 1-position on the naphthalene skeleton, it would be expected that the 5-amino and 1-hydroxyl groups would have hydrogen and proton donating properties, respectively.19) The use of 2,5-DHB gave abundant peaks corresponding to multiply protonated molecules [M+nH]n+ (n=1–3), which suggest that protonation occurs on the three relatively strong basic sites of the N-terminal amino group and two arginine residues (Arg31 and Arg139) (Fig. 1a), while the use of 1,5-DAN resulted in weak peaks of [M+nH]n+ and relatively abundant ISD fragments over an m/z range of 3000 to 7000 (Fig. 1b). The use of 5,1-ANL resulted in relatively abundant ions corresponding to [M+nH]n+ and ISD fragments. Of the matrices used, both 1,5-DAN and 5,1-ANL resulted in the steady generation of ISD fragment ions without a sweet spot, while the use of 2,5-DHB produced significant fluctuations in the generation of ISD fragments from shot to shot of the pulsed UV laser. The fluctuation in ISD fragment generation with 2,5-DHB is due to the presence of a sweet spot. Of the MALDI mass spectra of apo-myoglobin (Fig. 1), the spectrum obtained with 5,1-ANL showed excellent peak sharpness, with a wide range of ISD fragment ions from m/z 3000 to 10000 (the inset in Fig. 1c) and good reproducibility. As a result, the newly introduced matrix 5,1-ANL was deemed to be the most suitable for MALDI-ISD experiments with apo-myoglobin. Thus, here we used 5,1-ANL as a matrix to examine the influence of the secondary structure of a protein on the peak abundance of ISD fragment ions.

Fig. 1. MALDI-ISD spectra of equine apo-myoglobin (Mr 16951.4) with the following matrix materials: (a) 2,5-dihydroxybenzoic acid, (b) 1,5-diaminonaphthalene, and (c) 5-amino-1-naphthol.

Fig. 1. MALDI-ISD spectra of equine apo-myoglobin (Mr 16951.4) with the following matrix materials: (a) 2,5-dihydroxybenzoic acid, (b) 1,5-diaminonaphthalene, and (c) 5-amino-1-naphthol.

Regions of a protein that are susceptible to MALDI-ISD

Partial MALDI-ISD spectra of equine apo-myoglobin (Mr 16951.4), bovine serum albumin (Mr 66402.8) and equine cytochrome c (Mr 12360) are shown in Figs. 2, 3 and 4, respectively. All the ISD spectra showed c′-series ions originating from cleavage at the N–Cα bond of the backbone, and several discontinuously, intense peaks were observed in each spectrum, e.g., c35 and c43 originating from N–Cα bond cleavage at Gly35–His36 and Phe43–Asp44, respectively, for equine apo-myoglobin (Fig. 2). The intense c′-series ions suggest that the amino acid residues described above are in regions that are more susceptible to MALDI-ISD than other residues. For example the c35 ion originates from N–Cα bond cleavage at Gly35–His36 of myoglobin, which corresponds to the secondary structure between the end of the α-helix segment 21–35 and the start of the α-helix segment 37–40, as shown in the upper scheme of Fig. 2. The c35 ion is formed by which hydrogen radical becomes attached to the carbonyl oxygen between Gly35 and His36 residues. This suggests that the carbonyl oxygen is free from intra-molecular hydrogen bonding as would be present in α-helices and β-sheets. In fact, His36 is located in a flexible region that is free from α-helix structure (Fig. 2). It should be further noted that, in general, Gly residues are more likely to be located in a turn structure rather than an α-helix or a β-sheet.20) Therefore, it is reasonable to assume that the region of Gly35–His36 would be favorable for the hydrogen radical attachment reaction (2). Furthermore, a region with discontinuous peak abundance can be found in a boundary between the lower m/z region from c60 and the upper m/z region from c61, as shown in Fig. 2. The reason for why the upper region in m/z from c61 is unsusceptible to MALDI-ISD can be explained by the fact that an α-helix structure exists in the segment 59–76.

Fig. 2. MALDI-ISD spectrum of equine apo-myoglobin (Mr 16951.4) with 5-amino-1-naphthol as the matrix. Information concerning α-helix segments 21–35, 37–40, 52–56, and 59–76 was obtained from X-ray crystallography data in the protein data bank (PDB: 2 FRF).

Fig. 2. MALDI-ISD spectrum of equine apo-myoglobin (Mr 16951.4) with 5-amino-1-naphthol as the matrix. Information concerning α-helix segments 21–35, 37–40, 52–56, and 59–76 was obtained from X-ray crystallography data in the protein data bank (PDB: 2 FRF).

In the ISD spectrum of BSA (Fig. 3), discontinuously intense c′-series ions are observed at c12, c15, and c33 originating from cleavage at the N–Cα bond of Lys12–Asp13, Gly15–Glu16, and Gln33–Cys34, respectively. The c′-series ions corresponding to c36 to c38 were also observed, although these signals were weak. The c34 ion peak originating from cleavage at the N–Cα bond of Cys34–Pro35 was not observed due to incomplete cleavage at the N-terminal side of the proline residue. Gly15–Glu16 and Gln33–Cys34 are not involved in an α-helical structure, as shown in the secondary structure information in Fig. 3,21) which suggests that these residues are favorable for the attachment of hydrogen radicals. The Gly15 and Gln33 residues are in relatively flexible regions between α-helix segments 6–14 and 16–32, and between the segments 16–32 and 37–54, respectively. This suggests that the carbonyl oxygens of these residues are exposed to matrix molecules and are able to bind with hydrogen radicals.

Fig. 3. MALDI-ISD spectrum of bovine serum albumin (Mr 66432.3) with 5-amino-1-naphthol as the matrix. Information concerning α-helix segments 6–14, 16–32, and 37–54 was obtained from ref. 19.

Fig. 3. MALDI-ISD spectrum of bovine serum albumin (Mr 66432.3) with 5-amino-1-naphthol as the matrix. Information concerning α-helix segments 6–14, 16–32, and 37–54 was obtained from ref. 19.

Fig. 4. MALDI-ISD spectrum of equine cytochrome c (Mr 12360) with 5-amino-1-naphthol as the matrix. Information concerning α-helix regions 50–54, 61–67, and 71–74 was obtained from the X-ray crystallography data in the protein data bank (PDB: 1 HRC).

Fig. 4. MALDI-ISD spectrum of equine cytochrome c (Mr 12360) with 5-amino-1-naphthol as the matrix. Information concerning α-helix regions 50–54, 61–67, and 71–74 was obtained from the X-ray crystallography data in the protein data bank (PDB: 1 HRC).

Compared to myoglobin, equine cytochrome c is an α-helix-poor protein and its ISD spectrum gives much more intense c′-series ions from c20 to c80 than the other proteins, as shown in Fig. 3. The discontinuously intense peaks observed at c24, c30, c49, c53, and c69 originate from cleavage at the N–Cα bonds of Gly24–Lys25, Pro30–Asn31, Thr49–Asp50, Lys53–Asn54, and Gln69–Asn70, respectively. The Thr49–Asp50 and Lys53–Asn54 are in the regions of the start and end of the α-helix segment 50–54, respectively. The region of Glu69–Asn70 is in a turn structure between the α-helix segments 61–67 and 71–74. This suggests that, in the region of Glu69–Asn70, carbonyl oxygen is exposed to matrix molecule environments.

The results obtained above for three different proteins show a trend in which relatively flexible regions such as turn and bend structures are more susceptible to MALDI-ISD than intra-molecular hydrogen-bonded secondary structures such as α-helices. This suggests that protein molecules embedded in a matrix crystal still maintain the secondary structures that were originally determined by X-ray crystallography. On the other hand, the appearance of ISD fragments in α-helix segments such as 21–35, 37–40, and 52–56 of myoglobin (Fig. 2) indicates that protein molecules that are embedded in matrix crystals do not necessarily maintain their complete α-helix structures.

Amino acid residues susceptible to MALDI-ISD

The discontinuously intense c′-series ions and corresponding cleavage of amino acid residues for all the ISD spectra obtained here are summarized in Table 1. Table 1 provides information on common amino acid residues that are susceptible to MALDI-ISD as follows:

Table 1. Discontinuously intense c′-series ions and the corresponding cleavage of amino acid residues in MALDI-ISD spectra of proteins.

Protein c′-Series ions (corresponding amino acid residues)
Equine apo-myoglobin c35 (Gly35–His36), c43 (Phe43–Asp44), c60 (Asp60–Leu61)
Bovine serum albumin c12 (Lys12–Asp13), c15 (Gly15–Glu16), c33 (Gln33–Cys34)
Equine cytochrome c c24 (Gly24–Lys25), c30 (Pro30–Asn31), c49 (Thr49–Asp50), c53 (Lys53–Asn54), c69 (Glu69–Asn70)

a) Gly–Xxx for myoglobin, BSA and cytochrome c

b) Xxx–Asp for myoglobin, BSA and cytochrome c

c) Xxx–Asn for cytochrome c.

Although these amino acid residues do not necessarily exist in flexible regions that are free from α-helical structures, it would be of interest to examine the relationship between amino acid residues and susceptibility to ISD. It has been reported that certain amino acid residues are preferred for certain secondary structures such as α-helix, β-strand (or β-sheet), turn, bend or as isolated residues.20) Turn, bend and isolated residue are classified as flexible regions without the intra-molecular hydrogen-bonding that exists in α-helix and β-sheet structures, so that the carbonyl oxygen atoms on the backbone in the flexible regions are able to preferably interact with or bind to matrix molecules and hydrogen radicals. In contrast, in the case of α-helix and β-sheet structures, the carbonyl oxygen atoms are protected from interactions with matrix molecules and hydrogen radicals. According to a previous report,20) the amino acid residues Glu, Ala, Leu, Met, Gln, and Lys are preferred in the hydrogen-bonded secondary structure of an α-helix, while Pro, Gly, Ser, Cys, and Tyr residues rather tend to inhibit formation of an α-helix. In contrast, Gly, Asn, Asp, Pro, and Ser are somewhat preferred in the flexible secondary structure of a turn. It would be expected from the trend described above that the amino acid residues susceptible to MALDI-ISD, such as Gly–Xxx, Xxx–Asp, and Xxx–Asn, are preferred in turn or other flexible secondary structures, although it has been reported that the N–Cα bond of Xxx–Gly residues is less susceptible.14)

The Gly35 residue of equine apo-myoglobin is at the end of an α-helix segment 21–35, and the Gly15 of BSA and Gly24 of equine cytochrome c are in flexible regions. The Asp44 and Asp60 residues of apo-myoglobin are in a turn and at the start of the α-helix segment 59–76, respectively. The Asp13 residue of BSA is at the end of an α-helix segment 6–14. The Asp50 residue of equine cytochrome c is at the start of an α-helix segment 50–54. The Asn31 and 70 residues of equine cytochrome c are in flexible regions, and the Asn54 residue of equine cytochrome c is at the end of the α-helix segment 50–54. The Gln33 residue of BSA is located in a turn structure. Thus, the amino acid residues susceptible to MALDI-ISD, as presented in this study, are consistent with the trend described above.20) The results obtained from the ISD spectra of proteins showed that relatively intense peaks of c′-series ions originated from cleavage at N–Cα bonds of specific amino acid residues such as Gly–Xxx, Xxx–Asp, and Xxx–Asn. The question arises as to the susceptibility of amino acid residues to MALDI-ISD, e.g., why is Xxx–Asp more susceptible to ISD than Asp–Xxx? A possible explanation is that the fragmentation pathway for the cleavage at Xxx–Asp is energetically more favorable than that at Asp–Xxx. The solution to this question will require further study.

Conclusion

The MALDI-ISD spectra of proteins, obtained with an appropriate matrix 5,1-ANL, mainly showed N-terminal side c′-series ions originating from cleavage at the N–Cα bond of the peptide backbone. The peak abundance of the resulting c′-ions was dependent upon secondary structures and specific amino acid residues, suggesting that there are backbone regions and/or amino acid residues that are more susceptible to MALDI-ISD than other residues. That is, the c′-ions originating from backbone cleavage in flexible regions such as turn and bend, which are free from intra-molecular hydrogen-bonded secondary structures such as α-helix, showed discontinuously intense peaks. In contrast, the c′-ions originating from the cleavage of α-helical segments showed a relatively low abundance, because carbonyl oxygen atoms in the α-helix were protected from hydrogen radical attachment. Furthermore, the ISD spectra of the proteins showed that N–Cα bonds of specific amino acid residues, namely Gly–Xxx, Xxx–Asp, and Xxx–Asn, were more susceptible to MALDI-ISD than bonds of other amino acid residues. It should be noted that Gly, Asp, and Asn residues are rather preferred in flexible secondary structures such as turn than in intra-molecular hydrogen-bonded structures, such as α-helix.20)

The findings reported herein indicate that protein molecules that are embedded in the matrix crystal in the MALDI experiments maintain their secondary structures. Regarding this, describing the relationship between individual MALDI-ISD event and the corresponding time scale is of importance for understanding the influence of secondary structure and amino acid residues of proteins on MALDI-ISD. A series of MALDI-ISD processes may be described as follows: UV photon irradiation into matrix crystal, electronic excitation on a femtosecond (10−15 s) time scale, initial plume formation via vibronic excitation (10−14–10−11 s) on a nanosecond (<10−9 s) time scale,22) the formation of hydrogen radicals from the matrix via vibronic excitation (10−14−10−11 s), the transfer of hydrogen radicals to protein molecules and their attachment (10−14–10−11 s), and the ISD event within several tens of nanoseconds (<10−8 s) in the ion source. Based on the descriptions by Hillenkamp and Karas,22) both the initial plume formation (ablation through a crystal-phase explosion) and analyte ionization take place on a time scale of 10−9 s. On the other hand, generation of hydrogen radicals via vibronic excitation (10−14–10−11 s) occurs immediately after electronic excitation. This process may occur prior to plume formation within a nanosecond (<10−9 s). The resulting hydrogen radicals preferentially bind to carbonyl oxygens on the peptide backbone, when active hydrogens in the matrix molecules previously interact with the carbonyl oxygen on a backbone that is free from α-helix segments of protein molecules embedded in the matrix crystal. Both the attachment of hydrogen radicals to the carbonyl oxygen and proton transfer to generate analyte ions may take place simultaneously within a time scale of 10−11 s, which corresponds to a longer limit of vibronic excitation. The resulting radical and/or ionic species of protein molecules may be suspended in the initial plume similar to a dense-gas phase. Finally, ISD occurs in the plume within several tens of nanoseconds (<10−8 s) in the ion source. However, if protein molecules do not previously interact with active hydrogen in matrix molecules, the attachment of hydrogen radicals to protein molecules may occur on a time scale of 10−9 s in the plume. However, the attachment of hydrogen in the plume would be a rare event, even if the α-helix segments of protein molecules were extended to a coil or a non-rigid structure such as a turn or a bend. The folding to an α-helix structure occurs on the time scale of 10−6 s, and the helix–coil transition occurs with a relaxation time of 10−8–10−7 s.23) Although these time scales are in a solution phase, it would be difficult for a protein molecule to extend or unfold while it was embedded in the matrix crystal through plume formation on a timescale of 10−9 s. Consequently, the ISD of a protein may occur within 10−8 s in the initial plume when an α-helix structure is present, although protein molecules embedded in the matrix crystal may contain partially extended structures, and yet the MALDI-ISD method may be of use for obtaining information regarding the secondary structures of proteins.

Acknowledgments

M.T. acknowledges the support from the Creation of Innovation Centers for Advanced Interdisciplinary Research Area Program in the Special Coordination Fund for Promoting Science and Technology, and the Grant-in-Aid for Scientific Research (C) (23550101) from the Ministry of Education, Culture, Sports, Science and Technology of Japan.

References

  1. 1) M. Karas, D. Bachmann, U. Bahr, F. Hillenkamp. Int. J. Mass Spectrom. Ion Process. 78: 53–68, 1987 [Google Scholar]
  2. 2) K. Tanaka, H. Waki, Y. Ido, S. Akita, Y. Yoshida, T. Yoshida, T. Matsuo. Rapid Commun. Mass Spectrom. 2: 151–153, 1988 [Google Scholar]
  3. 3) M. Karas, F. Hillenkamp. Anal. Chem. 60: 2299–2301, 1988 [DOI] [PubMed] [Google Scholar]
  4. 4) C. M. Whitehouse, R. N. Dreyer, M. Yamashita, J. B. Fenn. Anal. Chem. 57: 675–679, 1985 [DOI] [PubMed] [Google Scholar]
  5. 5) J. B. Fenn, M. Mann, S. F. Meng, C. M. Wong, C. Whitehouse. Science 246: 60, 1989 [DOI] [PubMed] [Google Scholar]
  6. 6) W. J. Henzel, T. M. Billeci, J. T. Stults, S. C. Wong, C. Grimley, C. Watanabe. Proc. Natl. Acad. Sci. U.S.A. 90: 5011–5015, 1993 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. 7) D. J. C. Pappin, P. Hojrup, A. J. Bleasby. Curr. Biol. 3: 327–332, 1993 [DOI] [PubMed] [Google Scholar]
  8. 8) M. Mann, M. Wilm. Anal. Chem. 66: 4390–4399, 1994 [DOI] [PubMed] [Google Scholar]
  9. 9) J. K. Eng, J. R. McCormack, J. R. Yates. J. Am. Soc. Mass Spectrom. 5: 976–989, 1994 [DOI] [PubMed] [Google Scholar]
  10. 10) R. S. Brown, J. J. Lennon. Anal. Chem. 97: 3990–3999, 1995 [DOI] [PubMed] [Google Scholar]
  11. 11) D. C. Reiber, T. A. Grover, R. S. Brown. Anal. Chem. 70: 673–683, 1998 [DOI] [PubMed] [Google Scholar]
  12. 12) M. Takayama. J. Mass Spectrom. Soc. Jpn. 50: 304–310, 2002 [Google Scholar]
  13. 13) D. Suckau, A. Resemann. Anal. Chem. 75: 5817–5824, 2003 [DOI] [PubMed] [Google Scholar]
  14. 14) M. Takayama. J. Am. Soc. Mass Spectrom. 12: 1044–1049, 2001 [DOI] [PubMed] [Google Scholar]
  15. 15) M. Takayama. J. Am. Soc. Mass Spectrom. 12: 420–427, 2001 [DOI] [PubMed] [Google Scholar]
  16. 16) Y. Fukuyama, S. Iwamoto, K. Tanaka. J. Mass Spectrom. 41: 191–201, 2006 [DOI] [PubMed] [Google Scholar]
  17. 17) M. Sakakura, M. Takayama. J. Am. Soc. Mass Spectrom. 21: 979–988, 2010 [DOI] [PubMed] [Google Scholar]
  18. 18) K. Demeure, L. Quinton, V. Gabelica, E. De Pauw. Anal. Chem. 79: 8678–8685, 2007 [DOI] [PubMed] [Google Scholar]
  19. 19) D. Asakawa, M. Takayama. J. Am. Soc. Mass Spectrom. 22: 1224–1233, 2011 [DOI] [PubMed] [Google Scholar]
  20. 20) R. W. Williams, A. Chang, D. Juretić, S. Loughran. Biochim. Biophys. Acta 916: 200–204, 1987 [DOI] [PubMed] [Google Scholar]
  21. 21) B. X. Huang, H.-Y. Kim, C. Dass. J. Am. Soc. Mass Spectrom. 15: 1237–1247, 2004 [DOI] [PubMed] [Google Scholar]
  22. 22) F. Hillenkamp, J. Peter-Katalinic ed. MALDI MS, Wiley-VCH, Weinheim, Germany, 2007, pp. 1–28
  23. 23) R. H. Pain ed. Mechanisms of Protein Folding, IRL Press, Oxford, New York, 1994, pp. 1–26

Articles from Mass Spectrometry are provided here courtesy of Mass Spectrometry Society of Japan

RESOURCES