Substituent effects on the gas-phase fragmentation

RAPID COMMUNICATIONS IN MASS SPECTROMETRY Rapid Commun. Mass Spectrom. 2007; 21: 1230–1238 Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/rcm.2959

Substituent effects on the gas-phase fragmentation reactions of sulfonium ion containing peptides James Sierakowski1,y, Mahasilu Amunugama1,y, Kade D. Roberts1 and Gavin E. Reid1,2* 1 2

Department of Chemistry, Michigan State University, East Lansing, MI 48824, USA Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing, MI 48824, USA

Received 31 October 2006; Revised 21 January 2007; Accepted 30 January 2007

The multistage mass spectrometric (MS/MS and MS3) gas-phase fragmentation reactions of methionine side-chain sulfonium ion containing peptides formed by reaction with a series of parasubstituted phenacyl bromide (XBr where X ¼ CH2COC6H4R, and R ¼ –COOH, –COOCH3, –H, –CH3 and –CH2CH3) alkylating reagents have been examined in a linear quadrupole ion trap mass spectrometer. MS/MS of the singly (MR) and multiply ([MRRnH](nR1)R) charged precursor ions results in exclusive dissociation at the fixed charge containing side chain, independently of the amino acid composition and precursor ion charge state (i.e., proton mobility). However, loss of the methylphenacyl sulfide side-chain fragment as a neutral versus charged (protonated) species was observed to be highly dependent on the proton mobility of the precursor ion, and the identity of the phenacyl group para-substituent. Molecular orbital calculations were performed at the B3LYP/ 6-31RG level of theory to calculate the theoretical proton affinities of the neutral side-chain fragments. The log of the ratio of neutral versus protonated side-chain fragment losses from the derivatized side chain were found to exhibit a linear dependence on the proton affinity of the side-chain fragmentation product, as well as the proton affinities of the peptide product ions. Finally, MS3 dissociation of the nominally identical neutral and protonated loss product ions formed by MS/MS of the [MRRH]2R and [MRR2H]3R precursor ions, respectively, from the peptide GAILM(X)GAILK revealed significant differences in the abundances of the resultant product ions. These results suggest that the protonated peptide product ions formed by gas-phase fragmentation of sulfonium ion containing precursors in an ion trap mass spectrometer do not necessarily undergo intramolecular proton ‘scrambling’ prior to their further dissociation, in contrast to that previously demonstrated for peptide ions introduced by external ionization sources. Copyright # 2007 John Wiley & Sons, Ltd. The development of proteomic approaches for protein identification based on the database interrogation of uninterpreted tandem mass spectrometry (MS/MS) product ion spectra1–8 have been enabled by concurrent advances in our understanding of the mechanisms and other factors influencing the gas-phase fragmentation behavior of protonated peptide ions.9–14 These mechanistic studies have also facilitated the development of improved methods for controlling or directing peptide fragmentation reactions towards the formation of analytically useful product ions, from which peptides or proteins may be more readily identified and/or characterized. For example, the specificity of database search methods for the identification of aspartic acid containing peptides may be improved via inclusion of a characteristic ‘sequence’ product ion formed by cleavage at the C-terminal side of aspartic acid residues to the search parameters.15 The abundance of this characteristic product *Correspondence to: G. E. Reid, 234 Chemistry Building, Michigan State University, East Lansing, MI 48824, USA. E-mail: [email protected] y These authors contributed equally to this work. Contract/grant sponsor: National Science Foundation CAREER Award; contract/grant number: CHE 0547940.

ion is significantly enhanced by decreasing the ‘mobility’ of the ionizing charge within the peptide ion, via the introduction of a more basic functional group by conversion of lysine residues into homoarginine,15 or by introduction of a fixed charge derivative such as a quaternary alkylammonium or phosphonium ion to the N-terminus of the peptide.16–19 Under similar conditions of limited proton mobility, dominant characteristic ‘non-sequence ion’ cleavages at the side chains of oxidized methionine (–CH3SOH)14 or phosphoserine and phosphothreonine (–H3PO4) residues20 may also be observed. Recently, we have been examining the potential for side-chain fixed-charge derivatization approaches to control the formation of characteristic ‘non-sequence’ side-chain cleavage product ions in methionine- and cysteinecontaining peptides during MS/MS.21–23 The rationale for this work is that the formation of these product ions in high relative abundance, independently of amino acid composition or precursor ion charge state (i.e., proton mobility),

Copyright # 2007 John Wiley & Sons, Ltd.

Substituent effects on fragmentation reactions of peptides

dependent on the difference in proton affinity between the peptide product ion and its side-chain neutral fragment. Control over the formation of characteristic low m/z protonated ‘charged loss’ product ions could potentially be used to selectively identify the presence of fixed charge containing peptides from within complex mixtures by using precursor ion scan mode MS/MS experiments. Alternatively, the complementary high m/z charged loss peptide product ion could be used for selective identification of fixed charge containing peptides by using a ‘variable m/z gain analysis’ MS/MS scan, analogous to that described recently for phosphopeptide analysis in negative ion mode.24 The proton affinity of the neutral side-chain fragment formed by dissociation of side-chain fixed-charge sulfonium ion containing peptides may potentially be modulated as a function of the identity of meta- or para-substituents on the phenacyl alkylating reagents employed for the initial derivatization reaction. Substituents that result in the formation of a more acidic neutral molecule should result in more abundant neutral loss, while more basic substituents should result in more abundant charged losses. However, charged losses are only expected when the proton affinity of the resultant peptide product ion is relatively low, i.e., when the number of total charges on the resultant peptide product ion exceeds that of the number of basic sites within the ion. Here, in order to determine the influence of proton affinity on the relative abundance of neutral versus charged losses

enables them to be employed for selectively identifying peptide ions containing these amino acids from within complex mixtures, thereby resulting in mixture simplification and improved dynamic range for qualitative and quantitative proteome analysis.21 The mechanisms for these selective side-chain fragmentation reactions have been demonstrated to occur via neighboring group participation reactions involving a nucleophilic amide bond adjacent to the site of the fixed charge (shown in Scheme 1 for nucleophilic attack from the N-terminal amide bond of a derivatized methionine side chain).22 It was observed in these prior studies that fragmentation of the singly (Mþ) and doubly ([MþþH]2þ) charged phenacylsulfonium ion derivatives of the model peptide GAILM(X)GAILK (where X ¼ CH2COC6H5) both resulted in exclusive neutral loss of the methionine side chain (Scheme 1, pathway 1).21 In contrast, dissociation of the triply ([Mþþ2H]3þ) charged derivative of GAILM(X)GAILK resulted in the formation of protonated ‘charged loss’ product ions (Scheme 1, pathway 2), in addition to the neutral loss product. It is expected that these charged loss products are formed via intermolecular proton transfer from the triply protonated peptide product ion to the methylphenacyl sulfide side-chain fragment following the initial fragmentation reaction. Thus, the extent of neutral loss (Scheme 1, pathway 1) versus charged loss (Scheme 1, pathway 2) product ions that are formed should be highly

+nH+

O C

+

H N

N

C

H

O

+

S O

+

+

(n+1)+

[M +nH -(CH3SCH2COC6H5)]

CH3SCH2COC6H5 (Neutral loss)

Pathway 1 +

S

C

+nH+

O O

+nH+

O

CID

C

H N

C N H

C

+

H N

N

C

H

O

+

Pathway 2

[M+CH2COC6H5+nH+](n+1)+

H+ transfer (Charged loss)

+(n-1)H+

O C

S O

O

Pathway 2'

1231

+

H N

N

C

H

O

+

S + OH

Pathway 2''

+

O

+nH H N

C N

(CH3SCH2COC6H5+H+)

C O

[M++nH+-(CH3SCH2COC6H5+H+)]n+

Scheme 1. Copyright # 2007 John Wiley & Sons, Ltd.

Rapid Commun. Mass Spectrom. 2007; 21: 1230–1238 DOI: 10.1002/rcm

1232 J. Sierakowski et al.

below. All reagents were used as supplied or synthesized without further purification.

resulting from the side-chain dissociation of methionine fixed charge containing peptide ions, the gas-phase fragmentation reactions of a series of para-substituted (–COOH, –COOCH3, –H, –CH3 and –CH2CH3) phenacylsulfonium ion containing peptides have been examined. The data obtained from these experimental results have been compared with the predicted proton affinities of the side-chain cleavage products, determined by using high level density functional theory calculations of the neutral species and protonated ions at the B3LYP/6-31þG þZPVE level of theory. Finally, in order to determine the influence of the leaving group on the structure and subsequent fragmentation reactions of the resultant peptide product ions, we have compared the MS3 fragmentation reactions of the product ions formed by neutral versus charged loss from each of the para-substituted phenacylsulfonium-containing peptide ions.

4-Acetylbenzoic acid (503 mL, 3.37 mmol) was dissolved in acetic acid (20 mL) and the reaction mixture was heated at 708C until dissolution was achieved. The reaction solution was then cooled to room temperature and bromine (190 mL, 3.7 mmol) was added in one portion with stirring. After an overnight reaction with stirring at room temperature, the reaction mixture was then filtered, the precipitate washed with water and dissolved in ethyl acetate (20 mL). The insoluble material was filtered and the filtrate washed once with water (20 mL). The ethyl acetate was then removed under a gentle stream of N2, then completely dried under high vacuum to yield the product as a white solid.

EXPERIMENTAL

Synthesis of methyl-4-bromoacetylbenzoate (3)

Materials

Acetyl chloride (1.6 mL, 2 mol) was added dropwise to methanol (10 mL) with cooling and stirring in an ice bath. 40 -Acetylbenzoic acid (328 mg, 2 mmol) was then added in one portion and the reaction mixture was left to stir at room temperature for 3 days. The solvent was allowed to evaporate off overnight at room temperature to yield a white/yellow solid, which was then collected and dried under vacuum. The solid was then dissolved in acetic acid (10 mL) and bromine (190 mL, 3.7 mmol) was added in one portion with stirring at room temperature, which was continued overnight. The reaction solution was then diluted with H2O (50 mL) and extracted with diethyl ether (3 20 mL). The combined organic extracts were washed with 0.1 M NaHCO3 (2 20 mL) and H2O (1 20 mL), dried over MgSO4, concentrated in vacuo, then completely dried under high vacuum to give the product as a white solid.

Synthesis of 4-bromoacetylbenzoic acid (2)

The synthetic peptides GAILMGAILK (MK), VTMGHFDNFGR (MDR) and VTMAHFWNFGK (MWK) were obtained from Auspep (Melbourne, Victoria, Australia). The regioselectively deuterated GAIL-(d4)-MGAILK peptide was synthesized as described previously.22 Poly(4vinylpyridinium tribromide) and phenacyl bromide (compound 1 in Scheme 2) were purchased from Fluka (Switzerland). Methanol (HPLC grade), 4-acetylbenzoic acid, ethyl acetate and acetyl chloride were from Sigma-Aldrich (St. Louis, MO, USA). Glacial acetic acid (ACS grade), magnesium sulfate and sodium bicarbonate were obtained from Spectrum Chemicals (Gardena, CA, USA). Acetonitrile (HPLC grade) was purchased from OmniSolv (Gibbstown, NJ, USA). Diethyl ether and bromine were from Jade Scientific (Canton, Michigan, USA). 4-Ethylacetophenone and 4-methylacetophenone were purchased from Acros Organics (New Jersey, USA). All solutions were prepared using deionized water purified by a Barnstead nanopure diamond purification system (Dubuque, Iowa, USA). The para-substituted phenacyl bromide alklyation reagents (compounds 2–5 in Scheme 2) were prepared as described

Synthesis of 4-methylphenacyl bromide (4) 4’-Methylacetophenone (0.5g, 3.7 mmol) was added to poly(4-vinylpyridinium tribromide) (1.86 g, 3 meq Br 3 /g resin) in 20 mL methanol. The reaction mixture was then stirred at room temperature for 4 h. The reaction mixture was O

O O

Br

Br

Br O

OH O

O

(1)

(3)

(2) O

O Br

Br

(5)

(4) Scheme 2. Copyright # 2007 John Wiley & Sons, Ltd.



filtered, and the methanol was evaporated off under a gentle stream of N2 to yield the product as a white solid.

Synthesis of 4-ethylphenacyl bromide (5) 4-Ethylacetophenone was dissolved in acetic acid and bromine was added in one portion with stirring at room temperature which was continued overnight. The reaction solution went from a dark brown/orange color to a light yellow color over this time. The reaction solution was diluted with H2O (50 mL), extracted with diethyl ether (3 20mL) then H2O (1 20mL), then dried over MgSO4 and concentrated under reduced pressure to give the product as a dark yellow liquid.

Side-chain fixed-charge derivatization of methionine-containing peptides Side-chain fixed-charge sulfonium ion derivatives of synthetic methionine-containing model ‘tryptic’ peptides were produced by the addition of 10 mL of a 1 M solution of alkylating reagent (reagents 1–5 in Scheme 2) to 100 mg of either GAILMGAILK, VTMAHFWNFGK or VTM GHFDNFGR dissolved in 100 mL of aqueous 20% acetic acid containing 30% acetonitrile. The reactions were allowed to proceed for 24 h at room temperature after which the samples were diluted and introduced into the mass spectrometer without further purification.

Mass spectrometry Samples (0.02 mg/mL dissolved in 50% methanol/1% acetic acid) were introduced into a linear quadrupole ion trap mass spectrometer (Thermo model LTQ, San Jose, CA, USA) by nanoelectrospray ionization (nESI) at 0.5 mL/min. The spray voltage was set at 1.8 kV. The heated capillary temperature was 2008C. Collision-induced dissociation (CID) MS/MS and MS3 experiments were performed on monoisotopically mass-selected ions using standard isolation and excitation procedures. The peptide concentrations (approx. 10–20 pmol/uL) for this study were used in order to obtain product ion spectra with good signal-to-noise ratios, thereby allowing accurate measurements of the neutral versus charged loss product ion ratios. These concentrations are approximately an order of magnitude higher than the typical concentration of peptides eluting from a capillary highperformance liquid chromatography (HPLC) separation. The abundance of highly charged precursor ions from sulfonium

1233

ion containing peptides at low concentrations, i.e., where the number of total charges exceeds that of number of basic sites within the ion, and corresponding to precursor ions where charged losses are observed, may be more dependent on the ESI conditions than at higher signal levels. However, it is important to note that the addition of the sulfonium ion fixed charge typically leads to the formation of a precursor ion whose charge state is one greater than that of the underivatized peptide, so these charge states are expected to be observed even at low peptide signal levels.

Computational methods and proton affinity calculations Full conformational searches on neutral and protonated para-substituted S-methyl phenacyl side-chain fragments (structures 6–15 in Table 1) formed via selective gas-phase side-chain fragmentation of the phenacylsulfoniumcontaining peptide ions were initially performed at the PM3 semiempirical level of theory, followed by further optimization of low-energy structures at the B3LYP/ 6-31þG(d,p) level of theory using the GAUSSIAN 98 molecular modeling package.25 All optimized structures were subjected to harmonic vibrational frequency analysis and visualized using the computer package Gauss View 2.1 to determine the nature of the stationary points. Zero-point energies were obtained from harmonic frequency calculations without scaling. Complete structural details for each of the B3LYP/6-31þG(d,p) optimized structures are available from the authors upon request. The proton affinities (PAs) of structures 6–10 were calculated according to the negative of the enthalpy (DH) of the protonation reaction in Eqn. (1), via Eqn. (2): M þ Hþ ! MHþ

PA ¼ DH

(1)

DH298 ¼ Eelec ðMHþ Þ Eelec ðMÞ þ ZPVEðMHþ Þ ZPVEðMÞ þ DEn;298 ðMHþ Þ DEn;298 ðMÞ 5=2RT

(2)

where Eelec is the electronic energy, ZPVE is the zero-point vibrational energy correction (i.e., the difference in translational, rotational and vibrational energy from 0 K to 298 K) between the products and reactants, and DEn is the vibrational energy of the system. Values for Eelec(MHþ), Eelec(M), DEn,298 (MHþ), En,298(M) and ZPVE

Table 1. Total energies (Etotal), zero-point vibrational energies (ZPVE), vibrational energies (EVibrational) and calculated proton affinities (PA) (at the B3LYP/6-31þG level of theory) for the para-substituted methylphenacylsulfide side-chain fragments CH3SCH2COC6H4RR¼

Structure

–COOH

Neutral (6) Protonated (11) Neutral (7) Protonated (12) Neutral (8) Protonated (13) Neutral (9) Protonated (14) Neutral (10) Protonated (15)

–COOCH3 –H –CH3 –CH2CH3

Copyright # 2007 John Wiley & Sons, Ltd.

ETotal (Hartree)

ZPVE (kcal mol1)

EVibrational (kcal mol1)

Calculated PA (kcal mol1)

1011.009419 1011.354606 1050.31664 1050.662343 822.424349 822.776703 861.746192 862.104304 901.063164 901.422005

114.375630 121.810810 131.888080 139.323310 105.043650 112.701460 122.103150 129.788560 140.288790 147.935160

121.249000 128.692000 139.810000 147.298000 110.119000 117.755000 128.38900 136.04300 147.256000 154.93900

210.6 212.8 215.0 218.5 219.0



were all obtained from the output of the structural optimizations described above.

RESULTS AND DISCUSSION para-Substituent effects on the relative abundance of neutral versus charged side-chain losses from methionine phenacylsulfonium fixed-charge containing peptide ions Similar to that demonstrated previously in a 3D ion trap mass spectrometer,21 CID MS/MS of a methionine side-chain fixed-charge phenacylsulfonium ion derivative of the synthetic peptide GAILM(X)GAILK in a linear quadruple ion trap mass spectrometer resulted in exclusive neutral loss of the side chain (–CH3SX, 100% relative abundance, where X ¼ CH2COC6H5) from the singly [M]þ and doubly [MþþHþ]2þ charged precursor ions (Figs. 1(A) and 1(B), respectively). In contrast, both neutral and ‘charged’ loss product ions (–CH3SX, 100% relative abundance, –CH3SXþHþ, 47% relative abundance, CH3SXþHþ, 10% relative abundance) were observed by dissociation of the triply charged [Mþþ2Hþ]3þ precursor ion (Fig. 1(C)). This result is consistent with the expectation that the PAs of peptide product ions will decrease with increasing charge state, and that significant charged losses will be observed when the number of total charges on the peptide product ion exceeds that of the number of basic sites within the ion. The ionizing proton on the singly charged –CH3SX peptide product ion obtained from GAILM(X)GAILK is likely to reside at either the e-amino group of the lysine side chain or

along the peptide backbone, while the two ionizing protons on the doubly charged peptide product ion are likely to reside at both the lysine side chain and backbone positions. In the case of the triply charged -CH3SX peptide product ion of GAILM(X)GAILK, however, a decrease in the local PAs of the remaining backbone sites potentially available for protonation, as well as increased columbic repulsion, result in intermolecular proton transfer from the triply protonated peptide product ion to the neutral side-chain fragment following bond cleavage, but prior to separation of the initial ion-neutral complex (Scheme 1, pathway 2), thereby resulting in the formation of ‘charged’ loss protonated methylphenacyl sulfide and doubly protonated peptide product ions. The product ion spectra obtained by dissociation of the multiply charged precursor ions of the methionine side-chain fixed-charge phenacylsulfonium ion containing synthetic peptides VTM(X)AHFWNFGK (Figs. 1(D)–1(F)) and VTM(X)GHFDNFGR (data not shown) showed similar fragmentation behaviors. Exclusive neutral loss of the derivatized methionine side chain was observed for the low charge state precursor ions ([MþþHþ]2þ and [Mþþ2Hþ]3þ), while the quadruply charged ([Mþþ3Hþ]4þ) precursor ions underwent extensive charged losses. Again, this result is consistent with the expectation that only a limited number of ionizing protons (three in this case) could be readily accommodated in the resultant peptide product ions, i.e., at the side chains of the lysine or arginine residues at the C-terminus, the internal histidine residue, or along the peptide backbone. The abundance of the charged loss product ions from the quadruply charged ([Mþþ3Hþ]4þ) precursor of VTM(X)GHFDNFGR

Figure 1. CID MS/MS product ion spectra of the methionine side-chain fixed-charge phenacylsulfonium ions of GAILM(X)GAILK (MK) and VTM(X)AHFWNFGK (MWK) (X ¼ –CH2COC6H5): (A) [Mþ] ion of MK; (B) [MþþHþ]2þ ion of MK; (C) [Mþþ2Hþ]3þ ion of MK; (D) [MþþHþ]2þ ion of MWK; (E) [Mþþ2Hþ]3þ ion of MWK; and (F) [Mþþ3Hþ]4þ ion of MWK. Copyright # 2007 John Wiley & Sons, Ltd.



were slightly less than that from VTM(X)AHFWNFGK, suggestive of a higher PA in the former peptide. Previous studies carried out to examine the effect of PA on the ratios of amide bond cleavage product ion abundances formed by CID have observed a linear relationship between the log of the ratio of the b2- and y1-type product ion abundances and the PAs of the C-terminal amino acid (the leaving group) for a series of singly protonated tripeptides Gly-Gly-Xxx (where Xxx represents various amino acid residues).26 Therefore, in order to determine the effect of the leaving group PAs on the ratio of neutral versus charged loss from methionine side-chain fixed-charge sulfonium ion containing peptides, the CID MS/MS product ion spectra from a series of para-substituted (–COOH, –COOCH3, –CH3 and –CH2CH3) phenacylsulfonium ion containing peptides have been examined. For the singly and doubly charged precursor ion charge states of GAILM(X)GAILK, and the doubly and triply charged precursor ion charge states of VTM(X)GHFDNFGR and VTM(X)AHFWNFGK, exclusive neutral loss of the methionine side chain was observed for all the parasubstituted phenacylsulfonium ion derivatives examined (data not shown), indicating that the PAs of the resultant peptide product ions were all greater than that of the side-chain fragments formed (CH3SCH2COC6H4R, where R ¼ –COOH, –COOCH3, –CH3 and –CH2CH3). In contrast, the ratio of neutral versus charged losses from the methionine side chain of the [Mþþ2Hþ]3þ precursors of the para-substituted phenacylsulfonium ion derivatives of GAILM(X)GAILK (Figs. 2(A)–2(D)) and the [Mþþ3Hþ]4þ precursors of the para-substituted phenacylsulfonium ion derivatives of VTM(X)AHFWNFGK (Figs. 3(A)–3(D)) and VTM(X)GHFDNFGR (data not shown) were observed to vary dramatically as a function of the identity of the neutral side-chain fragment. For example, neutral loss of the side chain was observed as the dominant fragmentation pathway

1235

from the triply charged para-benzoic acid substituted phenacylsulfonium ion derivative (Fig. 2(A)), while almost exclusive charged loss was observed from the triply charged para-ethylphenacylsulfonium ion derivative (Fig. 3(D)). Comparison of the data in Figs. 2 and 3, as well as the data obtained from the VTM(X)GHFDNFGR peptide ions, demonstrate that the ratio of charged to neutral loss product ions from each of the para-substituted phenacylsulfonium ion derivatives increases in the order GAILM(X)GAILK [Mþþ2Hþ]3þ < VTM(X)GHFDNFGR [Mþþ3Hþ]4þ < VTM(X)AHFWNFGK [Mþþ3Hþ]4þ, and suggests that the PAs of the peptide product ions are in the order GAILMGAILK [Mþþ2Hþ–(CH3SRþHþ)]2þ > VTMGHFDNFGR [Mþþ3Hþ–(CH3SRþHþ)]3þ > VTMAHFW NFGK [Mþþ3Hþ–(CH3SRþHþ)]3þ.

PAs of the para-substituted methylphenacylsulfide side-chain fragment formed by dissociation of methionine side-chain fixed-charge containing peptide ions To determine whether a relationship does exist between the ratios of the charged versus neutral loss product ions from the fixed charge containing peptide ions described above and the PAs of the side-chain leaving group and peptide product ions, the lowest energy structures of the neutral and protonated para-substituted methylphenacylsulfide sidechain cleavage products formed via dissociation of the para-substituted phenacylsulfonium ion containing peptides were first determined at the B3LYP/6-31þG(d,p) density functional level of theory. Vibrational frequency analysis was performed to determine the nature of the optimized stationary point structure, and to obtain values of the zero-point vibrational energies (ZPVE) for subsequent PA calculations. Then, the PAs of the neutral structures were calculated according to the method described in the Experimental section above. The PAs derived from these

Figure 2. CID MS/MS product ion spectra of the [Mþþ2Hþ]3þ precursors from the parasubstituted phenacylsulfonium ions of GAILM(X)GAILK (MK) (X ¼ –CH2COC6H4R): (A) R ¼ COOH; (B) R ¼ COOCH3; (C) R ¼ CH3; and (D) R ¼ CH2CH3. Copyright # 2007 John Wiley & Sons, Ltd.



Figure 3. CID MS/MS product ion spectra of the [Mþþ3Hþ]4þ precursors from the parasubstituted phenacylsulfonium ions of VTM(X)AHFWNFGK (MWK) (X ¼ –CH2COC6H4R): (A) R ¼ COOH; (B) R ¼ COOCH3; (C) R ¼ CH3; and (D) R ¼ CH2CH3. calculations are given in Table 1. According to these results, the PAs of the neutral fragments range from 210.6 to 219.0 kcal mol1. Prior to plotting the PAs of the neutral methylphenacyl sulfide fragments calculated above against the experimentally determined neutral versus charged loss product ion ratios measured from Fig. 3, the product ion abundances were first normalized with respect to charge state to account for any charge state dependent detector response,13 in order to obtain a better approximation of the actual abundances of the neutral and charged loss side-chain cleavage product ions formed. Figure 4 shows a linear correlation between the log of the predicted PAs of the para-subtituted methylphenacyl sulfide

side-chain fragments and the experimentally observed ratios of charged versus neutral loss product ion abundances from CID MS/MS of [Mþþ2H]3þ GAILM(X)GAILK (Fig. 4(A)), [Mþþ3H]4þ VTM(X)GHFDNFGR (Fig. 4(B)), and [Mþþ3H]4þ VTM(X)AHFWNFGK (Fig. 4(C)). These results are consistent with the expected changes in the electron density of the para-substituted methylphenacyl sulfide fragments, as determined from the Hammett substituent constant s-values.27–29 The x-axis intercept of the plots shown in Fig. 4 may be used to estimate the PAs of the doubly charged product ion from GAILM(X)GAILK (approx. 216 kcal mol1), and the triply charged product ions of VTM(X)GHFDNFGR (approx. 211 kcal mol1) and VTM(X)AHFWNFGK (approx. 207.5 kcal mol1). The PAs of the amide bonds along the peptide backbone may be estimated from a simple model of the peptide amide bond, N-methylacetamide, which has a PA of 212.4 kcal mol1.30 Thus, the PA values estimated from the data above are consistent with the ionizing proton being located at a position along the peptide backbone.

Multistage tandem mass spectrometry of the neutral and charged loss product ions formed by dissociation of the methionine side-chain fixed-charge para-substituted phenacylsulfonium ion containing peptides

Figure 4. Plot of ln[–CH3SX/(–(CH3SXþHþ) þ CH3SXþHþ)] versus calculated proton affinity of the parasubstituted methylphenacylsulfide neutral loss products formed by dissociation of the methionine fixed charge containing peptide ions from (A) the [Mþþ2Hþ]3þ ion of GAILM(X)GAILK (MK); (B) the [Mþþ3Hþ]4þ ion of VTM(X)GHFDNFGR (MDR); and (C) the [Mþþ3Hþ]4þ ion of VTM(X)AHFWNFGK (MWK). Copyright # 2007 John Wiley & Sons, Ltd.

Figure 5(A) shows the MS3 product ion spectra obtained by dissociation of the neutral loss [MþþHþ–CH3SX]2þ product ion from the doubly charged phenacylsulfonium ion derivative of GAILM(X)GAILK formed in Fig. 1(B). Essentially identical spectra, including the relative abundance of the various product ions, were also obtained from dissociation of each of the neutral loss product ions of all the doubly charged para-substituted phenacylsulfonium ion derivatives of GAILM(X)GAILK (data not shown), strongly indicating that the structures of the product ions were the same in each case. In contrast, the product ion spectra Rapid Commun. Mass Spectrom. 2007; 21: 1230–1238 DOI: 10.1002/rcm


1237

Figure 5. Multistage tandem mass spectrometry (MS3) analysis of the methionine fixed-charge phenacylsulfonium ion side-chain cleavage products of GAILM(X)GAILK: (A) MS3 of the [MþþHþ-CH3SX]2þ neutral loss product ion from Fig. 1(B); (B) MS3 of the [Mþþ2Hþ–(CH3SXþHþ)]2þ charged loss product ion from Fig. 2(A); (C) MS3 of the [Mþþ2Hþ–(CH3SXþHþ)]2þ charged loss product ion from Fig. 1(C); and (D) MS3 of the [Mþþ2Hþ–(CH3SXþHþ)]2þ charged loss product ion from Fig. 2(D). obtained by dissociation of the doubly charged ‘charged loss’ [Mþþ2Hþ–(CH3SXþHþ)]2þ product ions formed from the triply charged precursor ions of GAILM(X)GAILK, X ¼ CH2COC6H4R, where R ¼ COOH (Fig. 5(B)), R ¼ H (Fig. 5(C)) and R ¼ CH2CH3 (Fig. 5(D)), revealed that the abundances of the product ions observed in each case were significantly different from each other, and from those observed in Fig. 5(A). This result suggests that the structures of the initial charged loss [Mþþ2Hþ–(CH3SXþHþ)]2þ product ions, and their subsequent MS3 fragmentation reactions, are dependent on the nature of the initial leaving group. Previously, we have demonstrated that dissociation of the derivatized side chain of methionine ‘fixed-charge’ phenacylsulfonium ion containing peptides results in the formation of cyclic five- or six-membered iminohydrofuran and oxazine product ions, via neighboring group participation reactions involving either the N- or C-terminal amide bonds adjacent to the methionine side chain, respectively.22 These cyclic product ions were also demonstrated to be stable to intramolecular ring opening during further dissociation by MS3, due to the requirement for a high-energy (predicted to be 45.0 kcal mol1) activation barrier. Furthermore, dissociation of the triply charged phenacylsulfonium ion precursor of a regioselectively deuterated GAIL-d4-MGAILK peptide indicated that intermolecular ring opening of the initial cyclic product ions did not occur upon ‘charged loss’ of the side chain (i.e., no [Mþþ2Hþ–(CH3SXþD)þ]2þ or (CH3SXþD)þ product ions observed).22 Similarly, in the current study reported here, no deuterated ‘charged loss’ product ions were observed upon dissociation of the triply charged precursor ions from each of the para-substituted phenacylsulfonium ion derivatives of the regioselectively Copyright # 2007 John Wiley & Sons, Ltd.

deuterated GAIL-d4-MGAILK peptide (data not shown), also indicating that intermolecular ring opening of the initial cyclic product ions does not occur upon ‘charged loss’ of the side chain. Thus, the possibility of a major structural difference (i.e., cyclic versus acyclic product ions) being responsible for the different MS3 fragmentation reactions observed here between the neutral and charged loss product ions may be discounted. A more likely explanation for the observed differences in the MS3 product ion abundances between the neutral and charged loss product ions is that the protonation sites in the product ions formed by neutral versus charged loss fragmentation pathways are different. For example, although fragmentation via pathway 20 and pathway 200 of Scheme 1 would both result in charged loss of the methylphenacyl sulfide side chain, only proton transfer via the former pathway would result in the peptide product ion having the same protonation site as the neutral loss product formed in pathway 1. It can be seen from the spectra in Figs. 5(B)–5(D) that the abundances of several of the products formed by MS3 dissociation of the charged loss product ions are observed to increase (e.g., the y5 and y6 ions) or decrease (e.g., the y7, y8 and y2þ 8 ions) as a function of the para-substituent on the methionine side-chain CH3SX leaving group. The trends associated with changes in the relative abundances of the product ions in these spectra were consistently observed, with variations of 5% or less, across triplicate MS3 experiments. A plot of the % abundance of these ions against the identity of the para-substituent is shown in Fig. 6. It can also be seen from the spectra in Figs. 5(B)–5(D) that the MS3 spectra of the charged loss product ions obtained from the triply charged para-substituted phenacylsulfonium precursor ions become more similar to the neutral loss product ion Rapid Commun. Mass Spectrom. 2007; 21: 1230–1238 DOI: 10.1002/rcm


for protein identification via database interrogation of the uninterpreted product ion spectrum against the available sequence databases is currently the subject of further investigation.

Acknowledgements Support for this work was provided by a National Science Foundation CAREER award to GER (Grant No: CHE 0547940).

REFERENCES

Figure 6. % abundance of the b- and y-type product ions observed by MS3 of the [Mþþ2Hþ–(CH3SXþHþ)]2þ charged loss product ions of the para-substituted phenacylsulfonium ions of GAILM(X)GAILK. spectrum of the doubly charged precursor ion as a function of increasing proton affinity of the parasubstituted CH3SX side-chain fragment (i.e., in the order of Fig. 5(B) (–COOH), Fig. 5(C) (–H) and Fig. 5(D) (–CH2CH3)). This suggests that proton transfer via pathway 20 of Scheme 1 dominates for charged losses involving basic side-chain fragments, where the lifetime of the ion-neutral complex is expected to be longest, thereby allowing the neutral fragment to ‘sample’ and subsequently transfer a proton from a position remote to that of the site of cleavage. For charged losses involving more acidic side-chain fragments, the lifetime of the ion-neutral complex is expected to be shorter, resulting predominantly in proton transfer at the site closest to the site of cleavage (i.e., pathway 200 of Scheme 1). Interestingly, these results suggest that the protonated peptide product ions formed by gas-phase CID of sulfonium ion containing precursors in the ion trap mass spectrometer do not necessarily undergo intramolecular proton ‘scrambling’ prior to their further dissociation, in contrast to that previously demonstrated for protonated peptide ions introduced by external ionization sources.31

CONCLUSIONS The results described here indicate that the ratio of neutral loss to charged loss product ion abundances from fixedcharge sulfonium ion containing peptides varies according to the differences between the proton affinities of the neutral side-chain fragment and the peptide product ions. Furthermore, the ratio of charged versus neutral loss observed for a given peptide was observed to change as a function of peptide ion charge state and amino acid composition (i.e., proton mobility). Thus, specific information regarding the number of basic sites contained within the peptide may be obtained by having control over this fragmentation reaction. The potential for utilizing this information, coupled with knowledge regarding the presence of a specific amino acid residue (e.g., methionine) within the peptide, proteolytic enzyme cleavage specificity, chromatographic retention time, or accurate peptide mass, to provide more specificity Copyright # 2007 John Wiley & Sons, Ltd.

1. Eng JK, McCormack AL, Yates JR. J. Am. Soc. Mass Spectrom. 1994; 5: 976. 2. Perkins DN, Pappin DJC, Creasy DM, Cottrell JS. Electrophoresis 1999; 20: 3551. 3. Pevzner PA, Mulyukov Z, Dancik V, Tang CL. Genome Res. 2001; 11: 290. 4. Creasy DM, Cottrell JS. Proteomics 2002; 2: 1426. 5. Keller A, Nesvizhskii A, Kolker E, Aebersold R. Anal Chem. 2002; 74: 5383. 6. Fenyo¨ D, Beavis RC. Anal. Chem. 2003; 75: 768. 7. Sadygov RG, Cociorva D, Yates JR. Nat. Methods 2004; 1: 195. 8. Hernandez P, Muller M, Appel RD. Mass Spectrom. Rev. 2006; 25: 235. 9. Dongre RA, Jones JL, Somogyi A, Wysocki VH. J. Am. Chem. Soc. 1996; 118: 8365. 10. O’Hair RAJ. J. Mass Spectrom. 2000; 35: 1377. 11. Wysocki VH, Tsaprailis G, Smith LL, Breci LA. J. Mass Spectrom. 2000; 35: 1399. 12. Paizs B, Suhai S. Mass Spectrom. Rev. 2005; 24: 508. 13. Kapp EA, Schultz F, Reid GE, Eddes JS, Moritz RL, O’Hair RAJ, Speed TP, Simpson RJ. J Anal. Chem. 2003; 75: 6251. 14. Reid GE, Roberts KD, Kapp EA, Simpson RJ. J. Proteome Res. 2004; 3: 751. 15. Sidhu KS, Sangvanich P, Brancia FL, Sullivan AG, Gaskell SJ, Wolkenhauer O, Oliver SG, Hubbard SJ. Proteomics 2001; 1: 1368. 16. Gu C, Tsaprailis G, Breci L, Wysocki VH. Anal. Chem. 2000; 72: 5804. 17. Adamczyk M, Gebler JC, Wu J. Rapid Commun. Mass Spectrom. 1999; 13: 1413. 18. Huang Z, Wu J, Roth KDW, Gage DA, Watson JT. Anal. Chem. 1997; 69: 137. 19. Sadagopan N, Watson JT. J. Am. Soc. Mass Spectrom. 2001; 12: 399. 20. Jin M, Bateup H, Padovan JC, Greengard P, Nairn AC, Chait BT. Anal Chem. 2005; 77: 7845. 21. Reid GE, Roberts KD, Simpson RJ, O’Hair RAJ. J. Am. Soc. Mass Spectrom. 2005; 16: 1131. 22. Amunugama M, Roberts KD, Reid GE. J. Am. Soc. Mass Spectrom. 2006; 17: 1631. 23. Roberts KD, Reid GE. J. Mass Spectrom. 2007; 42: 187. 24. Averbukh ME, Pipkorn R, Lehmann WD. Anal. Chem. 2006; 78: 1249. 25. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Zakrzewski VG, Montgomery JA Jr, Stratman RE, Burant JC, Dapprich S, Millam JM, Daniels AD, Kudin KN, Strain MC, Farkas O, Tomasi J, Barone V, Cossi M, Cammi R, Mennucci B, Pomelli C, Adamo C, Clifford S, Ochterski J, Petersson GA, Ayala PY, Cui Q, Morokuma K, Malick DK, Rabuck AD, Raghavachari K, Foresman JB, Cioslowski J, Ortiz JV, Baboul AG, Stefanov BB, Liu G, Liashenko A, Piskorz P, Komaromi I, Gomperts R, Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng CY, Nanayakkara A, Challacombe M, Gill PMW, Johnson B, Chen W, Wong MW, Andres JL, Gonzalez C, Head-Gordon M, Replogle ES, Pople JA. Gaussian 98, Revision A. 7. Gaussian, Inc.: Pittsburgh, PA, 1998. 26. Morgan DG, Bursey MM. Org. Mass Spectrom. 1994; 29: 354. 27. Hammett LP. J. Am. Chem. Soc. 1937; 59: 96. 28. Jaffe H. Chem. Rev. 1953; 53: 191. 29. Hansch C, Leo A, Taft RW. Chem. Rev. 1991; 97: 165. 30. Hunter EPL, Lias SG. J. Phys. Chem. Ref. Data 1998; 27: 413. 31. Johnson RS, Krylov D, Walsh KA. J. Mass Spectrom. 1995; 30: 386. Rapid Commun. Mass Spectrom. 2007; 21: 1230–1238 DOI: 10.1002/rcm