Ion Chemistry of Protonated Lysine Derivatives

JOURNAL OF MASS SPECTROMETRY, VOL. 31, 1237-1243 (1996)

Ion Chemistry of Protonated Lysine Derivatives

Talat Yalcin and Alex G. Harrison? Department of Chemistry, University of Toronto, Toronto, Ontario, MSS 3H6, Canada

Protonated lysine fragments primarily by elimination of the &-amino group as ammonia to form an ion of m/z 130 and to a minor extent by elimination of H,O to form an ion of m/z 129. Protonated lysine derivatives such as lysine fi-naphthylamide and H-Lys-Gly-OH show more pronounced formation of m/z 129 while protonated derivatives such as W-Ac-Lys-X (X = OH, OMe, NHMe) and H-Gly-Lys-X (X = OH, NHCH,COOH) also show formation of m/z 129 in both metastable ion and collision-induced fragmentation. In both the latter systems m/z 129 is formed by sequential loss of HX followed by loss of ketene for the N-acetyl derivatives or the glycine residue for the N-glycyl derivatives. Although the m/z 129 ion is nominally an acylium ion, its metastable ion characteristics and collision-induced dissociation mass spectrum are very similar to those of protonated a-amino-e- caprolactam. It is concluded that this lactam is formed from the lysine derivatives by interaction of the amino group of the lysine side-chain with the lysine carbonyl function as HX departs. Protonated N-methyllysine and N-dimethyllysine fragment exclusively by elimination of CH,NH, and (CH,),NH, respectively. Evidence is presented that the stable structure of the m/z 130 ion so formed is protonated pipecolic acid. Both the protonated a-amino-E-caprolactam and protonated pipecolic acid ions fragment further primarily to [ C,H,,N] + (m/z 84), a low mass ion commonly observed in the spectra of lysine-containing peptides.

KEYWORDS: fragmentation; lysine derivatives; fragment ion structures; cc-amino-6-caprolactam; sequential fragmentation reactions; protonated pipecolic acid

INTRODUCTION

There is an increasing use of collision-induced dissociation (CID) of protonated peptides as a means of obtaining information as to the amino acid sequence in the peptide.'-3 As a result, the primary fragmentation reactions of protonated peptides are reasonably well e~tablished,'.~-~ as illustrated schematically in Fig. 1 (although the B, ions are depicted in Fig. 1 as acyclic

.COOH

+ 53 c1 7 4 H3 N-C- C-N -C-COOH

51 51 H H H H H

82 Y;'

Figure 1. Schematic diagram of major fragmentation reactions of protonated peptides.

t Author to whom correspondence should be addressed.

acylium ions, we have shown9+'* recently that the stable structures of these ions are cyclic protonated oxazolones). In addition to the ions indicated in Fig. 1, low-mass (m/z < 160) ions, characteristic of the amino acids comprising the peptide, are frequently observed in the CID mass Indeed, parent ion scans for the precursors of m/z 86 have recently been used to identify peptides containing leucine or isoleucine at low concentration levels in rnixtures.I6

While many of these low-mass ions are immonium ions, RCH=NHz+, a particularly intriguing low-mass ion is the m/z 129 ion frequently observed as a characteristic fragment ion in the CID mass spectra of peptides containing l y ~ i n e . ~ ' ~ ~ This is illustrated by the low-energy CID mass spectrum of protonated H-Gly- Gly-Lys-Ala-Ala-OH (Fig. 2) and also is evident in the CID mass spectra of lysine-containing peptides reported by Poulter and Taylor." The m/z 129 ion also is observed' '*I9 in the fragmentation of protonated lysine and corresponds to elimination of H,O from the protonated molecule. In the absence of rearrangement this [MH+ - H,O]' ion would correspond to the acylium ion H,NCH(R)CO+. However, there is substantial e v i d e n ~ e ~ ~ ' ~ - ~ ~ that such a-amino acylium ions are unstable and exothermically eliminate CO. Since the m/z 129 ion derived from lysine and derivatives is stable, this suggests that it must have a structure other than an acyclic acylium ion. Accordingly, we have undertaken a study of the mechanism of formation and structure of the m/z 129 ion from lysine derivatives using metastable ion s t ~ d i e s ' ~ , ~ ~ and energy-resolved collisional mass

CCC 1076-5174/96/111237-07 0 1996 by John Wiley & Sons, Ltd.

Received 25 April 1996 Accepted 17 July 1996

1238 T. YALCIN AND A. G. HARRISON

H-Gly-Gly-LyS-AlO-Alo -OH 45 eV C I D

I29

I MH’ Irn’l 403

Figure 2. 45 eV CID mass spectrum of protonated H-Gly-Gly- Lys-Ala- Ala-OH.

spec t r~met ry ’~ -~~ as probes. A major fragmentation channel for protonated lysine is elimination of NH, to form m/z 13018,’9 and I5N labelling has s h ~ w n ’ ~ ~ ’ ~ that the ammonia incorporates exclusively the amino group of the side-chain and not the a-amino group. In the absence of rearrangement this would lead to an unstable primary alkyl cation and we present some limited studies relevant to the structure of the m/z 130 ion observed in the spectra of some lysine derivatives.

EXPERIMENTAL

All experimental work was carried out using a ZAB-2FQ hybrid BEqQ mass spectrometer (VG Ana- lytical, Manchester, UK), which has been described in detail previously.28 Briefly, this instrument is a reversed-geometry (BE) double-focusing mass spectrometer that is followed by a deceleration lens system, an r.f.-only quadrupole collision cell (9) and a quadrupole mass analyzer (Q). The ions of interest were prepared by fast atom bombardment (FAB) using an argon atom beam of 7-8 keV energy with the appropriate sample dissolved in a matrix consisting of thioglycerol-2,2’- dithiodiethanol(1: 1) saturated with oxalic acid.

To obtain relative abundances of fragment ions formed on the metastable ion time-scale, the precursor ion of interest (usually MH’) was mass selected by the BE double-focusing mass spectrometer at 6 keV ion energy, decelerated to 20-40 eV kinetic energy and introduced into the r.f.-only quadrupole cell in the absence of collision gas. Low-energy CID studies were carried out in the same fashion but with the addition of N, at an indicated pressure of -2 x lo-’ Torr (1 Torr = 133.3 Pa) to the quadrupole collision cell. In the CID experiments the incident ion energy typically was varied from 2-45 eV (laboratory scale). In both the unimolecular and CID studies the ionic fragments were analyzed by scanning the final quadrupole Q; 20-30 2 s scans were accumulated on a multi-channel analyzer to improve the signal-to-noise ratio. The energy-resolved CID data are presented here in the form of breakdown

graphs expressing the relative fragment ion signals as a function of the collision energy.

In several cases the precursors to selected ions were examined by reaction intermediate scans.29 In this approach the MH+ ion of interest was mass selected by the magnetic sector, the quadrupole mass analyzer (Q) was set to transmit the fragment ion of interest and the electric sector was scanned to record the intermediate ions leading to the formation of the chosen fragment ion.

Kinetic energy releases associated with the unimolecular fragmentation reactions were determined by the mass-analyzed ion kinetic energy spectrometry (MIKES) technique.’l In this technique, the precursor ion of interest was mass selected by the magnetic sector at 6 keV ion energy and the ionic products of unimolecular fragmentation reactions occurring in the drift region between the magnetic and electric sectors were identified according to their kinetic energy by scanning the electric sector. The kinetic energy releases were determined from the peak widths at half-height after correction for the inherent energy spread of the ion beam according to the relation23

(1) where w,,, is the measured width of the metastable peak and w,b is the width of the parent ion beam. The corrected half-widths were converted into T,,, values using the equation developed2’ for electric sector scans.

The lysine derivatives and pipecolic acid were obtained from Bachem Biosciences or Sigma Chemical and were used as received. a-Amino-8-caprolactam was obtained from Fluka Chemical and was used as received.

2 2 112 wcorr = (wmet - wmb)

RESULTS AND DISCUSSION

Formation and structure of m/z 129 ion

Table 1 records the results obtained in a study of the metastable ion fragmentation of the MH’ ion of lysine and a number of derivatives. For a variety of H-Lys-X species, formation of m/z 129 by loss of HX from MH’ is observed as a significant metastable ion fragmentation reaction; this fragmentation channel amounts to -20% of the total metastable ion signal for H-Lys- OMe and H-Lys-b-NA and to 40-45% for H-Lys-Gly- OH and H-Lys-Gly-Gly-OH. For all these species, which are unsubstituted on the a-amino group, loss of NH, is also observed as a significant metastable ion fragmentation channel, in agreement with the fragmentation of protonated lysine itself. It is initially surprising to observed significant yields of the m/z 129 product in fragmentation of a number of N“-substituted lysine derivatives. Thus, protonated H-Gly-Lys-OH, Ac-Lys- OH and H-Gly-Gly-Lys-OH show >40% yields of the m/z 129 product in metastable ion fragmentation of MH + . By contrast, H-Gly-Lys-Gly-OH and Ac-Lys-

PROTONATED LYSINE DERIVATIVES 1239

Table 1. Metastable ion fragmentation of protonated lysine derivatives

Fragment (% of total ion signal) Derivative -NH, -H,O mp129 H-Lys-OH.H+ Other'

H - Lys-0 H H - Lys- 0 Me H - L ~ S - P - N A ~ H - LYS- G l y - 0 H H - G ly- Lys-0 H H -Gly-Lys-G ly-0 H

Ac- Lys- N H Me

Ac-Lys-OMe Ac-Le-OH

H -G ly-G ly- Lys-OH H - Lys-Gly-Gly-OH

93 7 (-H,O) 76 4 20 77 23 57 3 40 27 49 25 5 4 2 68, -( H -Gly-OH )

21, -(H,O + NH,) 6 5 5 84, -MeNH,

11 6 52 6 32 10 33 18, -(CH,OH)

43 57 40 9 45 6, -(H-Gly-OH)

a % of total ion signal, neutral lost bfi-NA = fi-naphthylarnide.

NHMe show only low yields of the m/z 129 product in metastable ion fragmentation of MH'.

The breakdown graph for protonated Ac-Lys-NHMe is shown in Fig. 3. In agreement with the metastable ion data, the breakdown graph shows m/z 171 ([MH+ - CH,NH,]+) as the dominant primary fragment ion. With increasing collision energy this primary ion fragments further to m/z 129 by elimination of ketene, while at even higher collision energies m/z 84 becomes a significant fragment ion. The sequential fragmentation reaction

MH' -t m/z 171 + m/z 129 (2)

is also clearly indicated by the precursor ion scan for m/z 129 shown in Fig. 4. In contrast to the results for the methylamide, the breakdown graphs for protonated Ac-Lys-OH (Fig. 5 ) and protonated Ac-Lys-OMe (Fig.

6) show m/z 129 as the dominant fragment ion at low collision energies with further extensive fragmentation to m/z 84 at higher collision energies. The m/z 171 fragment ion, which dominates the low-energy CID of the methylamide (Fig. 3), is of only low abundance for the acid and the methyl ester. Loss of NH, is a significant fragmentation reaction for the methyl ester and this fragment appears to fragment further to form m/z 144 (elimination of CH,CO) and m/z 126 (elimination of CH,OH + CO). Despite the low abundance of the m/z 171 ion for the acid and the methyl ester, precursor ion scans showed this species as a precursor of the m/z 129 fragment; that for the methyl ester is shown in Fig. 4(b).

A similar difference in fragmentation behaviour is observed for protonated H-Gly-Lys-Gly-OH (Fig. 7) and protonated H-Gly-Lys-OH (Fig. 8). The former

M d , N z C I D

z 0 t z w

0 U R LL

LL

I 40

0 a-" 20

0 0 10 20 30 40

COLLISION ENERGY (eV, lab scale)

0 U R L L -

LL

m/z 84

0 ; [,Y: 10 20 30 40 COLLISION ENERGY (eV, lab scale)

0 a-" 20-

0

Figure 3. Breakdown graph for protonated Nu-Ac-Lys-NHMe.

PRECURSOR SCAN m / z 129

N*-Ac-l.ys-NH Me

d N - A c - L y s - O M e

203 MH+l I/ I1 [ Mh*- CHjOH 1'

129 171

Figure 4. Precursor ion scans for m/z 129 for protonated (a) Nu-Ac- Lys- NH Me and (b) N'-Ac- Lys-OMe.


I 1 I 1 1 1

N ~ - A C - L ~ ~ - O H

60 MH+, N,CID

w V z 0 z 3 -

40 -

a 50-

m a z E l - c 5 30- 5 3 - LL iL

iL 20 -

ENERGY (eV, lab scale)

Figure 5. Breakdown graph for protonated Na-Ac-Lys-OH.

shows dominant formation of m/z 186 by elimination of H-Gly-OH at low collision energies with further fragmentation to mjz 129 by elimination of the glycine residue as the collision energy increases. By contrast, protonated H-Gly-Lys-OH shows a relatively low ion signal for mjz 186 ([MH+ - H20]+) with a much more prominent m/z 129 fragment ion signal at low collision energies; in addition, there is much more extensive fragmentation to mjz 84 as the collision energy is increased.

Before attempting to rationalize these differences in behavior of the acetyl and glycyl derivatives, we address the question of the structure of the m/z 129 fragment ion. One possibility is that the ion corresponds to the acyclic acylium ion H2N(CH2)4CH(NH2)CO+ ;

N" -Ac-Lys-OMe

60 MH+, N,CID

COLLISION ENERGY (eV, l ab scale)

Figure 6. Breakdown graph for protonated Na-Ac-Lys-OMe.

' I ' I ' I ' I '

7 o L H-Gly-Lys-Giy-OH

COLLISION ENERGY [e< l a b scale)

Figure 7. Breakdown graph for protonated H-Gly-Lys-Gly-OH.

however, as discussed in the Introduction, there is substantial evidenceg*' '-" that such acyclic acylium ions are unstable and exothermically eliminate CO. As an alternative we propose, as illustrated in Scheme 1, that, as HX is lost from protonated RNH(CH,),CH(NH,)C(=O)X, cyclization occurs involving the &-amino group to form a protonated a- amino-e-caprolactam. To test this hypothesis, we com- pared the fragmentation behavior of the protonated a-amino-8-caprolactam with that of the m/z 129 ion formed from the lysine derivatives. Protonated a-amino- 6-caprolactam showed metastable ion fragmentation by elimination of H 2 0 (5%), elimination of CO (20%) and

I I I I 1 H- GI y- L y s -OH

MH: N,CID

W

m / i 04

z 0

COLLISION ENERGY (eV. lab scale)

Figure 8. Breakdown graph for protonated H-Gly-Lys-OH.


d-Amino-&-Coproloctom.H+

Scheme 1

elimination of (C,N,0,H3) (75%). The metastable peak for loss of CO showed a kinetic energy release TlI2 = 0.53 eV while that for loss of (C,N,O,H,) gave a TI/, of 0.18 eV. The m/z 129 ion was of sufficient intensity in the fast atom bombardment (FAB) mass spectrum of Ac-Lys-NHMe to examine its metastable ion fragmentation reactions. Identical metastable ion fragmentation reactions were observed with very similar relative inten- sities: - H,O (5%), - CO (27%), -(C,N,O,H,) (68%). The metastable peak for loss of CO gave TI,, = 0.60 eV while that for loss of (C,N,0,H3) gave T,,, = 0.18 eV.

This agreement of the metastable ion characteristics provides strong support for the hypothesis that the stable m/z 129 ion from the lysine derivatives has the protonated a-amino-8-caprolactam structure. Addi- tional support comes from CID studies. Figure 9 com- pares the 45 eV CID mass spectrum for the protonated lactam with that for the m/z 129 ion derived from protonated Ac-Lys-OH. In the latter MS/MS/MS experiment the MH+ ion was mass selected by the magnetic sector of the BEqQ instrument. The m/z 129 ions formed by unimolecular fragmentation of MH + in the field-free region between the magnetic and electric sectors were selected by the electric sector, decelerated and introduced into the quadrupole cell q where they underwent collisional activation. Although the m/z 84 fragment, also observed in unimolecular fragmentation, is the dominant ion in both CID spectra, the distribu- tions of the minor ions in the two spectra are very similar, adding additional support for the hypothesis that the stable structure of the m/z 129 fragment ion in the spectra of the lysine derivatives is the protonated a-amino-E-caprolactam (Scheme 1, R = H). In general, high-energy collisional activation is more useful in com- paring ion structures; however, we observed that the high-energy CID mass spectra of source-produced ions, in addition to ions produced by unimolecular fragmentation in the field-free region before the magnetic sector, were severely compromised by artifact peaks ; conse- quently, we were unable to obtain reliable high-energy CID mass spectra.

Accepting that the stable structure of the m/z 129 ion is protonated a-amino-E-caprolactam, we now rationalize the differing behavior of protonated Ac-Lys-NHMe and Ac-Lys-OH/Ac-Lys-OMe on unimolecular and collision-induced fragmentation. As Scheme 1 indicates, formation of the lactam requires that the proton be attached to the X group of the C-terminus. This clearly is not the thermodynamically favored site of proto- nation. As examples, Bouchonnet and H ~ p p i l l i a r d ~ ~ calculated that the proton affinity of the OH group in glycine is 43 kcal mol-' (1 kcal = 4.184 kJ) less than that of the NH, group, while Zhang et aL3' calculated that the proton affinity of the amide nitrogen in digly- cine is 17 kcal mol-' less than that of the N-terminal

amine function. Thus, formation of the X-protonated species in our systems requires that the protonated species have considerable excess internal energy. A plausible potential energy profile for fragmentation of protonated Ac-Lys-X is presented in Fig. 10. We have assumed that the most stable form of the protonated species a is the e-NH,-protonated form solvated by the carbonyl oxygen of the N"-acetyl function. As discussed above, the relative energy of the X-protonated species b depends on the identity of X. For X =NHMe it appears that the energy of the intermediate state b is below the threshold for fragmentation of c to the final product, protonated a-amino-E-caprolactam, with the result that the major metastable fragmentation reaction observed corresponds to formation of c, protonated N"- acetyl-c-caprolactam. However, for protonated Ac-Lys- O H and Ac-Lys-OMe, the X-protonated species is higher in energy with the result that c (m/z 171) is formed with sufficient internal energy to fragment further on the metastable ion time-scale to the final product protonated a-amino-E-caprolactam. A similar potential energy profile can rationalize the differing behaviour of protonated H-Gly-Lys-Gly-OH and H-Gly-Lys-OH. For the former compound, the energy of the intermediate analogous to b is sufficiently low in energy that the primary product analogous to c (m/z 186) does not have sufficient internal energy to fragment further on the metastable ion time-scale. In contrast, for protonated H-Gly-Lys-OH, the intermediate analogous to b is of sufficiently high energy that the primary m/z 186 product retains enough internal energy to fragment further on the metastable ion time-scale.

The formation of m/z 129 from the protonated acety- llysine derivatives is a further example of consecutive fragmentation reactions occurring on the metastable ion time-scale3, and it is of interest to examine the kinetic energy releases associated with these metastable ion reactions. The following data were obtained for fragmentation of protonated Ac-Lys-OMe:

MH' + m / z 171 + CH30H

TI,, = 0.054 eV (3)

MH+ + m/z 129 + CH30H + CH,CO

T,,, = 0.037 eV (4)

m/z 171 + m/z 129 + CH,CO

TI,, = 0.0068 eV ( 5 )

Beynon and c o - ~ o r k e r s ~ ~ have shown that for the fragmentation sequence

m i + +m,+ +(m, -m2).+

m3+ + (m2 - m3) + (m, - m2) (6) P(m1 - mz)Tl/mlm,11'2 + [2(mz - m3)T2/m2 m311/2

= [I2(mi - m3)T3/mim31"2 (7) where T, is the kinetic energy release for the first step m,' .+ m2+, T2 is the kinetic energy release for the second step m2+ + m 3 + and T3 is the kinetic energy release for the overall reaction ml+ --t m3+. Using our measured T,,,, values for reactions (3) and (5), Eqn (7) predicts a Tlj2 for reaction (4) of 0.040 eV, in agreement


d -Ammo-E-Coprolocram MHf, rn/z 129

45eV C l D

!

t Ac-Lys-OH Mt l+G129% MS/ MS/ MS

4 5 e V C I O

84

I I I

56

L-

Figure 9. 45 ev CID mass spectra for protonated a-amino-e-caprolactam and m p 129 originating from unimolecular fragmentation of protonated Nu-Ac-Lys-OH.

with the measured value of 0.037 eV within experimental error.

Structure and fragmentation of mlz 130 ion

As noted above, a major metastable ion fragmentation channel of protonated lysine is by elimination of NH,

'*,19 to form an ion of m/z 130; this fragmentation involves exclusively the amino group of the side- hai in.'^,'^ When this side-chain amino group is substituted to form N'-methyllysine or NE-dimethyllysine, formation of m/z 130 by loss of CH,NH, or (CH,),NH, respectively, from MH" becomes the only metastable ion fragmentation reaction observed. Figure 11 shows the 45 eV CID mass spectra of protonated N'- methyllysine and NE-dimethyllysine. Only two significant ions are observed in the CID mass spectra, m/z 130 and its fragmentation product m/z 84 [m/z 130-(C,

0

Figure 11. 45 eV CID mass spectra for protonated N"- methyllysine and Ne-dimethyllysine.

O2 ,H2)]. We believe that the stable structure of the m/z 130 ion is protonated pipecolic acid, formed, as indicated in Scheme 2, by cyclization as the &-amino function is lost. In support of this proposal we note that protonated pipecolic acid fragments on the metastable ion time-scale by elimination of (C,O, ,H2) with a kinetic energy release TI,, = 0.45 eV. The m/z 130 ion observed in the FAB mass spectrum of N'-methyllysine also showed metastable ion fragmentation by loss of (C,

PIPECOLIC ACID. H' Scheme 2

,C$- c,H2 5"2 ,CH2

CH ,NH; +CH,CO+HX

C - a - CH3

Figure 10. Potential energy profile for fragmentation of protonated Ac-Lys-X. Relative energy of b depends on identity of X (see text).


O,,H,) with an identical kinetic energy release T,,, = 0.45 eV. This fragmentation reaction is that commonly ob~erved '* '~ -~~ for fragmentation of protonated a-amino acids and the kinetic energy release is similar to those which we have measured previously for a variety of protonated a-amino acids.' The low-energy CID mass spectra of the m/z 130 ions from the two sources showed only m/z 84 as a significant fragment ion while the high-energy CID mass spectra obtained by the MIKES technique showed numerous artifact peaks and could not be used for structure character- ization. Thus, we are left with only the identity of the kinetic energy releases as evidence to support our hypothesis that the stable structure of the m/z 130 ion in the spectra of lysine derivatives is protonated pipecolic acid.

larity of the metastable ion characteristics and CID spectra of this ion with those for protonated a-amino-a- caprolactam, we conclude that the stable ion formed from the lysine derivatives is not an acyclic acylium ion but has undergone cyclization to form protonated a- amino-a-caprolactam. The major fragmentation route of this ion is by elimination of (C,O,N,H,) and this appears to be the major route to the m/z 84 ion also frequently observed**'* as a low-mass ion in the spectra of lysine-containing peptides.

We also have presented limited evidence that the m/z 130 ion observed in the spectra of some lysine derivatives also results from cyclization to form protonated pipecolic acid. This ion also fragments to form m/z 84 (CC5HlONI '1.

CONCLUSIONS Acknowledgements

The authors are indebted to the Natural Sciences and Engineering Research Council of Canada for financial support and to Fisons Instruments for the loan of the fast atom gun and FAB source.

The m/z 129 ion is Observed as a fragment ion in the spectra of many lysine derivatives. From the simi-

REFERENCES

1. K. Biernann, Methods Enzymol. 193, Chapt. 18 and 25 (1 990).

2. D. M. Desiderio (Ed.), Mass Spectrometry of Peptides. CRC Press, Boca Raton (1 991 ).

3. K. Biemann, in Biological Mass Spectrometry. Present and Future, edited by T. Matsuo, R. M. Caprioli, M. L. Gross and Y. Seyarna. Wiley, New York, p. 275 (1 993).

4. P. Roepstorff and J. Fohlrnan, Biomed. Mass Spectrom. 11, 601 (1984).

5. D. F. Hunt, J. R. Yates, Ill, J. Shabanowitz, S. Winston and C. R. Hauer, Proc. Natl. Acad. Sci. USA 83,6233 (1 986).

6. K. Biemann and S. Martin, Mass Spectrom. Rev. 6,75 (1 987). 7. K. Biernann, Biomed. Environ. Mass Spectrom. 16, 99 (1988). 8. I. A. Papayannopoulos, Mass Spectrom. Rev. 14,49 (1 995). 9. T. Yalcin, C. Khouw, I. G. Csizrnadia, M. R. Peterson and A. G.

Harrison, J. Am. SOC. Mass Spectrom. 6,1165 (1 995). 10. T. Yalcin, I. G. Csizmadia, M. R. Peterson and A. G. Harrison,

J. Am. SOC. Mass Spectrom. 7,233 (1 996). 11. W. Kausler, K. Schneider and G. Spiteller, Biomed. Environ.

Mass Spectrom. 13,405 (1 986). 12. R. S. Johnson, S. A. Martin and K. Biernann, In?. J. Mass

Spectrom. /on Processes 86, 139 (1 988). 13. R. S. Johnson and K. Biernann, Biomed. Environ. Mass

Spectrom. 18, 945 (1988). 14. T. Madden, K. J. Welharn and M. A. Baldwin, Org. Mass

Spectrom. 26,443 (1991). 15. A. M. Falick, W. M. Hines, K. F. Medzihradszky, M. A.

Baldwin and B. W. Gibson, J. Am. Soc. Mass Spectrom. 4, 882 (1 993).

16. M. Wilrn, G. Neubauer and M. Mann. Anal. Chem. 68, 527 (1996).

17. L. Poulter and L. C. E. Taylor, lnr. J. Mass Spectrom. Ion Pro- cesses 91, 1 83 (1 989).

18. W. Kulik and W. Heerrna, Biomed. Environ. Mass Spectrom. 15,419(1988).

19. N. N. Dookeran, T. Yalcin and A. G. Harrison, J. Mass Spectrom. 31,500 (1 996).

20. C. W. Tsang and A. G. Harrison, J. Am. Chem. Soc. 98, 1301 (1976).

21. G. Bouchoux, S. Bourcier, Y. Hoppilliard and C. Mauriac, Org. Mass Spectrom. 28, 1064 (1 993).

22. R. G. Cooks, J. H. Beynon, R. M. Caprioli and G. R. Lester, Metasfable lons. Elsevier, New York (1 973).

23. J. L. Holrnes and J. K. Terlouw, Org. Mass Spectrom. 15, 383 (1980).

24. S. A. McLuckey, G. L. Glish and R. G. Cooks, h t . J. Mass Specfrom. /on Phys. 39, 21 9 (1 981 ).

25. D. D. Fetterolf and R. A. Yost, lnt. J. Mass Spectrom. Ion Phys. 44,37 (1982).

26. S. A. McLuckey and R . G. Cooks, in Tandem Mass Spectrom- etry, edited by F. W. McLafferty, Wiley, New York, p. 303 (1 983).

27. G. W. A. Milne, T. Axenrod and H. M. Fales, J. Am. Chem. SOC. 92.51 70 (1 970).

28. A. G. Harrison, R. S. Mercer, E. J. Reiner, A. B. Young, R. K. Boyd, R. E. March and C. J. Porter, lnt. J. Mass Spectrom. /on Processes 74, 1 3 (1 986).

29. J. N. Louris, L. G. Wright, R. G. Cooks and A. E. Schoen,Anal. Chem. 57, 291 8 (1 985).

30. S. Bouchonnet and Y. Hoppilliard, Org. Mass Spectrom. 27, 71 (1992).

31. K. Zhang, D. M. Zirnrnerrnan, A. Chung-Phillips and C. J. Cassady, J . Am. Chem. Soc. 11 5,1081 2 (1 993).

32. C. J. Proctor, B. Kralj, A. G. Brenton and J. H. Beynon, Org. Mass Spectrom. 15, 61 9 (1 980).

Documents

Ion Chemistry of Protonated Lysine Derivatives