RADICAL-DRIVEN PEPTIDE BACKBONE DISSOCIATION TANDEM MASS SPECTROMETRY Han Bin Oh* and Bongjin Moon Department of Chemistry, Sogang University, Seoul 121-742, Republic of Korea Received 18 June 2012; revised 6 May 2013; accepted 20 November 2013 Published online 26 May 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/mas.21426
In recent years, a number of novel tandem mass spectrometry approaches utilizing radical-driven peptide gas-phase fragmentation chemistry have been developed. These approaches show a peptide fragmentation pattern quite different from that of collision-induced dissociation (CID). The peptide fragmentation features of these approaches share some in common with electron capture dissociation (ECD) or electron transfer dissociation (ETD) without the use of sophisticated equipment such as a Fourier-transform mass spectrometer. For example, Siu and coworkers showed that CID of transition metal (ligand)-peptide ternary complexes led to the formation of peptide radical ions through dissociative electron transfer (Chu et al., 2000. J Phys Chem B 104:3393–3397). The subsequent collisional activation of the generated radical ions resulted in a number of characteristic product ions, including a, c, x, z-type fragments and notable side-chain losses. Another example is the free radical initiated peptide sequencing (FRIPS) approach, in which Porter et al. and Beauchamp et al. independently introduced a free radical initiator to the primary amine group of the lysine side chain or N-terminus of peptides (Masterson et al., 2004. J Am Chem Soc 126:720–721; Hodyss et al., 2005 J Am Chem Soc 127: 12436– 12437). Photodetachment of gaseous multiply charged peptide anions (Joly et al., 2008. J Am Chem Soc 130:13832–13833) and UV photodissociation of photolabile radical precursors including a C–I bond (Ly & Julian, 2008. J Am Chem Soc 130:351–358; Ly & Julian, 2009. J Am Soc Mass Spectrom 20:1148–1158) also provide another route to generate radical ions. In this review, we provide a brief summary of recent results obtained through the radical-driven peptide backbone dissociation tandem mass spectrometry approach. # 2014 Wiley Periodicals, Inc. Mass Spec Rev 34:116–132, 2015 Keywords: peptides; tandem mass spectrometry; radical-driven dissociation; proteomics
I. INTRODUCTION Protein identification and characterization based on tandem mass spectrometry (MS/MS) (McLafferty, 1983) results have
Contract grant sponsor: This work was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology; Contract grant number: 2012R1A1A2006532; Contract grant sponsor: This work was also supported by the Sogang University Research Grant of 2013; Contract grant number: SRF-201314004. Correspondence to: Han Bin Oh, Department of Chemistry, Sogang University, Seoul 121-742, Republic of Korea. E-mail: [email protected]
Mass Spectrometry Reviews 2015, 34, 116–132 # 2014 by Wiley Periodicals, Inc.
played a key role in the development of the field of proteomics (Biemann & Martin, 1987; Aebersold & Goodlett, 2001). Collision-based tandem mass spectrometry methods, such as collision-activated dissociation (CAD) or collision-induced dissociation (CID), are the most widely used MS/MS methods for the purpose of proteomic studies (Hunt et al., 1986; Smith et al., 1990). In CAD/CID, protonated peptides generally accumulate internal energy through multiple collisions with buffer gas and undergo amide-bond, –C(O)–N–, dissociations to yield the characteristic b- and y-type fragment ions (McLuckey & Goeringer, 1997). In this method, the even-electron peptide ions, (M þ nH)nþ, with closed electronic structures, are generally the subject of collisional activation. Through extensive research, the peptide dissociation mechanism in CAD/CID is now relatively well understood (Wysocki et al., 2000; Paizs & Suhai, 2005; Dongre et al., 2006; Barlow & O’Hair, 2008; Harrison, 2009). In recent years, new tandem mass spectrometry methods based on odd-electron peptide/protein species have emerged as valuable tools in peptide/protein mass spectrometry studies. First, electron capture dissociation (ECD) was introduced in 1998 (Zubarev, Kelleher, & McLafferty, 1998; Zubarev et al., 2000) and was shown to be very powerful to induce extensive peptide/protein backbone dissociations (Sze et al., 2002; Zubarev, 2003; Cooper, Ha˚kansson, & Marshall, 2005; Oh & McLafferty, 2006). Capture of a low kinetic energy electron by even-electron (M þ nH)nþ peptide/protein cations in the Fourier transform-ion cyclotron resonance (FT-ICR) mass spectrometer produces the odd-electron hydrogen abundant species of (M þ nH)(n1)þ•, followed by backbone dissociations that yield characteristic c-and z-type ions as major products and some minor a- and x-type ions. Later, it was shown that electron transfer dissociation (ETD) also produces c- and z-type ions as major fragment ions when applied to peptide cations (Syka et al., 2004; Coon et al., 2005a; Pitteri et al., 2005; Swaney, McAlister, & Coon, 2008). The mechanism of ECD and ETD processes has been the subject of extensive experimental and theoretical studies; however, the mechanism is still not entirely clear, even though the understanding of the ECD and ETD processes has been deepened (Zubarev et al., 2000; Cooper et al., 2002; Oh et al., 2002; Zubarev, 2003; Turecˇek & Syrstad, 2003; Turecˇek, 2003; Leymarie, Costello, & O’Connor, 2003; Breuker et al., 2004; Cooper, Ha˚kansson, & Marshall, 2005; Fung & Chan, 2005; Anusiewicz et al., 2005; O’Connor et al., 2006; Lee et al., 2006; Pouthier & Tsybin, 2008; Sohn et al., 2009; Moore, Ly, & Julian, 2011). In ECD/ETD, the rupture of N–Ca bonds occurs while noncovalent interactions are generally conserved. Therefore, the protein regions where ECD fragmentation is silent can be considered as regions in the protein where
RADICAL-DRIVEN PEPTIDE BACKBONE DISSOCIATION MS/MS
tertiary contacts with other protein regions are intact. Based on this unique ECD property and interpretation, ECD could be used to probe the secondary and tertiary structures of intact protein cations in the gas phase (Oh et al., 2002; Adams et al., 2006; Xie et al., 2006; Lee et al., 2009b). In addition, a few electron-based tandem mass spectrometry methods, such as electron induced dissociation (EID) and electron detachment dissociation (EDD), have been introduced, but these methods are beyond our scope in this review. In parallel with ECD/ETD, a variety of methods that introduce peptide/protein radical ions in the gas phase have been developed. Siu and coworkers discovered that peptide radical cations, M•þ, could be generated by low energy CID of ternary metal complexes with peptide and auxiliary ligands (Chu et al., 2000, 2008; Bagheri-Majdi et al., 2004; Wee, O’Hair, & McFadyen, 2004; Barlow, McFadyen, & O’Hair, 2005; Barlow et al., 2006; Ke et al., 2006; Laskin, Yang, & Chu, 2008; Hopkinson, 2009; Chu & Laskin, 2011; Kalli & Hess, 2012). UV photo-excitation followed by H atom loss provides another way to generate radical peptide ions (Tabarin et al., 2005; Lemoine et al., 2006; Park et al., 2009). UV photoionization of laser-ablated peptides has also been used for the formation of aromatic amino acid-containing peptide radical ions (Grotemeyer & Schlag, 1989; Cui, Hu, & Lifshitz, 2002). Photodetachment of gaseous multiply charged peptides and amino acid anions that are prepared through electrospray ionization (ESI) also provides another route to generate radical anions (Joly et al., 2008; Larraillet et al., 2009; Brunet et al., 2011). In recent years, it was reported that UV irradiation of a photolabile radical precursor generated a site-specific phenyl-type radical with a very high quantitative yield (Ly & Julian, 2008, 2009). A crown ether-based photolabile radical precursor, which forms noncovalent complexes with peptides, has also been used for the preparation of peptide radical cations (Sun et al., 2009). Another notable way to generate peptide radical cations is the free radical initiated peptide sequencing (FRIPS) approach, in which a free radical initiator is introduced to the primary amine group of the lysine side chain or N-terminus of peptides (Masterson et al., 2004; Hodyss, Cox, & Beauchamp, 2005). Upon collisional activation of the precursor peptide ions, a radical site is generated. The fragmentation pathways and behavior of the peptide radical ions are relatively less understood than those of the protonated counterparts. Nevertheless, it has been found that the fragmentation mechanism and behavior of the peptide radical ions are quite different from those of the protonated or deprotonated ions. For example, in the fragmentation of peptide radical ions, a-, c-, x-, and z-type fragment ions are abundantly observed and the side-chain loss provides important structural information. It is also known that the radical site readily migrates to other parts of the peptide or protein, easily interconverting among various isomers (Ly & Julian, 2009). In particular, the structure and stability of small peptide cations have been extensively studied, both experimentally and theoretically, and well documented in recent reviews (Hopkinson, 2009; Chu & Laskin, 2011). Therefore, in this account we do not attempt to give a detailed description of small peptide radical cations in these aspects. Instead, we seek to shed light on the generation of peptide/protein radical ions using a variety of different methods and their radical-based peptide ion fragmentation, with a particular emphasis on larger peptide and Mass Spectrometry Reviews DOI 10.1002/mas
&
protein ion fragmentation chemistry. Considering the large number of publications on peptide radical ions, a summary of the recent results obtained through radical-driven peptide backbone dissociation tandem mass spectrometry, particularly in the analytical aspect, appears to be timely and appropriate. We would like to add that this review is not intended to be comprehensive, but to serve as a good introductory guide for readers who are interested in radical-driven peptide fragmentation chemistry, particularly graduate students and early career scholars in the field of mass spectrometry.
II. FORMATION AND MIGRATION OF A RADICAL SITE A variety of methods have been developed that can generate peptide radical ions. Each method adopts its own unique way to form a radical site and the underlying chemistry is somewhat different in each technique; therefore, we will give short descriptions of these methods. In peptide radical ion fragmentation, it is known that both its stability and the migration of the initial radical site within the peptide ions play a dominant role in dictating final fragmentation sites. In this section, the underlying mechanism with respect to its generation, stability, and migration of the radical site will be summarized.
A. CID of Ternary Metal Complexes With Peptide and Auxiliary Ligands For this method, theoretical and experimental studies have been extensively performed and recently well reviewed (Hopkinson & Siu, 2006; Hopkinson, 2009). Gas-phase redox chemistry of ternary metal complexes with peptide and auxiliary ligands has allowed the formation of cationic peptide radical ions (Chu et al., 2000), and is outlined in Scheme 1 (Barlow, McFadyen, & O’Hair, 2005). CID of the ternary complexes leads to the
SCHEME 1. Outline of a method for producing cationic peptide radicals from ternary metal complexes. [Scheme reproduced from Barlow, McFadyen, and O’Hair (2005), with permission from American Chemical Society, copyright 2005.]
117
&
OH AND MOON
separation of the peptide from the complex, with concomitant reduction of the metal and oxidation of the peptide to form the cationic peptide radical, Pþ• or Mþ•. The subsequent isolation and CID of Pþ• resulted in rich peptide radical ion fragmentation chemistry. In parallel with the formation of Pþ• (Equation 1), CID of the ternary complexes can also lead to a variety of other competitive reaction pathways. The other pathways include dissociative proton transfer to the peptide (Equation 2), dissociative proton abstraction from the peptide (Equation 3), and peptide fragmentation (Equation 4) (Chu, Lam, & Siu, 2005; Ke et al., 2006; Hopkinson, 2009). ½Mnþ ðLÞPnþ ! ½Mnþ ðLÞðn1Þþ þ Pþ
ð1Þ
½Mnþ ðLÞPnþ ! ½Mnþ ðL HÞðn1Þþ þ ½P þ Hþ
ð2Þ
½Mnþ ðLÞPnþ ! ½Mnþ ðP HÞðn1Þþ þ ½L þ Hþ
ð3Þ
½Mnþ ðLÞPnþ ! ½Mðn1Þ ðLÞðP bn Þðn1Þþ þ bn þ
ð4Þ
where Mnþ and L refer to the metal ion and auxiliary ligand, respectively, and P is the peptide species. The role of auxiliary ligands was investigated and some conclusions were drawn (Chu et al., 2004; Barlow et al., 2004; Chu, Lam, & Siu, 2005). A tri- or tetra-dentate ligand was found to be a better ligand in producing radicals because in these cases a ligand would not easily separate from the complex. A tri- or tetra-dentate was more effective in coupling oxidation and dissociation of the peptide compared to situations where bidentate ligands were used (Chu et al., 2000; Barlow et al., 2004; Turecˇek, 2007). It was also found that the formation of radical cationic tripeptides, for example, GGXþ•, was facilitated when a sterically encumbered auxiliary macrocyclic ligand was used instead of its open-chain analog (Lam et al., 2006a; Song et al., 2009). Another interesting finding was that the ligand should not have an acidic proton in order to prevent the protonated peptide from leaving the complex. An appropriate choice of auxiliary ligand, such as 12-crown-4 (1,4,7,10-tetraoxacyclododecane), allowed for the generation of peptide radical cations that contain only aliphatic residues (Chu et al., 2004). On the other hand, it was shown that in addition to Cu2þ ion, which is the best choice of transition metals for carrying out redox chemistry, Cr3þ, Mn3þ, Fe3þ, and Co3þ could be utilized to perform this type of gas-phase redox chemistry (Barlow, McFadyen, & O’Hair, 2005; Laskin, Yang, & Chu, 2008). It was also observed that either aromatic or basic amino acid residues facilitated the production of peptide radical cations (Chu et al., 2000, 2001; Bagheri-Majdi et al., 2004; Wee, O’Hair, & McFadyen, 2006a). Interestingly, it was possible to generate peptide radical ions with a well-defined initial radical site using this approach (Chu et al., 2001, 2008; Wee et al., 2006a, 2006b; Siu et al., 2009). For example, at a [CuII(L)(M)]2þ• CID, dissociative electron transfer occurred to generate Mþ•, where M is either YGG, GYG, or GGY (Chu et al., 2008, Chu & Laskin, 2011). Through Ca–Cb homolytic bond cleavage of Mþ• tyrosyl residues, that is, side chain loss (vide infra), three a-carboncentered radical cations—[G•GG]þ, [GG•G]þ, and [GGG•]þ— with a well-defined initial radical site could be generated. Due to high-energy barriers for interconversion between [G•GG]þ, 118
[GG•G]þ, and [GGG•]þ (44.7 kcal/mol relative to the structure at the global minimum), these isomers existed as isomerically pure species, and the CID spectra of these isomers were substantially different from each other. In addition, other noninterconverting radical cationic isomeric pairs, [G•GW]þ/ [GGW]•þ and [G•GY]þ/[GGY]•þ, which contain an aromatic residue, were also examined, wherein a- and p-carbon centered radicals did not interconvert (Ng et al., 2010). For the peptides in which an aromatic residue such as tyrosine, tryptophan, histidine, or phenylalanine is present, a radical site can be generated in the aromatic side chain; a p-carbon centered radical. CID of p-carbon centered [GGW]•þ/[GGY]•þ peptide radical cations mainly yielded a and c/z-type products, while CID of a-carbon centered [G•GW]þ/[G•GY]þ produced abundant b/y-type ions. The apparently different CID spectra of aand p-carbon centered peptide radical cations indicated that they did not interconvert on the time scale of the CID experiments. These results were also in agreement with the density functional theory (DFT) calculations, in which the isomerization barrier between the GGW radical cations (>35.4 kcal/mol) was higher than the barrier for competitive dissociation of these species (
In recent years, a number of novel tandem mass spectrometry approaches utilizing radical-driven peptide gas-phase fragmentation chemistry have been developed. These approaches show a peptide fragmentation pattern quite different from that of collision-induced dissociation (CID). The peptide fragmentation features of these approaches share some in common with electron capture dissociation (ECD) or electron transfer dissociation (ETD) without the use of sophisticated equipment such as a Fourier-transform mass spectrometer. For example, Siu and coworkers showed that CID of transition metal (ligand)-peptide ternary complexes led to the formation of peptide radical ions through dissociative electron transfer (Chu et al., 2000. J Phys Chem B 104:3393–3397). The subsequent collisional activation of the generated radical ions resulted in a number of characteristic product ions, including a, c, x, z-type fragments and notable side-chain losses. Another example is the free radical initiated peptide sequencing (FRIPS) approach, in which Porter et al. and Beauchamp et al. independently introduced a free radical initiator to the primary amine group of the lysine side chain or N-terminus of peptides (Masterson et al., 2004. J Am Chem Soc 126:720–721; Hodyss et al., 2005 J Am Chem Soc 127: 12436– 12437). Photodetachment of gaseous multiply charged peptide anions (Joly et al., 2008. J Am Chem Soc 130:13832–13833) and UV photodissociation of photolabile radical precursors including a C–I bond (Ly & Julian, 2008. J Am Chem Soc 130:351–358; Ly & Julian, 2009. J Am Soc Mass Spectrom 20:1148–1158) also provide another route to generate radical ions. In this review, we provide a brief summary of recent results obtained through the radical-driven peptide backbone dissociation tandem mass spectrometry approach. # 2014 Wiley Periodicals, Inc. Mass Spec Rev 34:116–132, 2015 Keywords: peptides; tandem mass spectrometry; radical-driven dissociation; proteomics
I. INTRODUCTION Protein identification and characterization based on tandem mass spectrometry (MS/MS) (McLafferty, 1983) results have
Contract grant sponsor: This work was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology; Contract grant number: 2012R1A1A2006532; Contract grant sponsor: This work was also supported by the Sogang University Research Grant of 2013; Contract grant number: SRF-201314004. Correspondence to: Han Bin Oh, Department of Chemistry, Sogang University, Seoul 121-742, Republic of Korea. E-mail: [email protected]
Mass Spectrometry Reviews 2015, 34, 116–132 # 2014 by Wiley Periodicals, Inc.
played a key role in the development of the field of proteomics (Biemann & Martin, 1987; Aebersold & Goodlett, 2001). Collision-based tandem mass spectrometry methods, such as collision-activated dissociation (CAD) or collision-induced dissociation (CID), are the most widely used MS/MS methods for the purpose of proteomic studies (Hunt et al., 1986; Smith et al., 1990). In CAD/CID, protonated peptides generally accumulate internal energy through multiple collisions with buffer gas and undergo amide-bond, –C(O)–N–, dissociations to yield the characteristic b- and y-type fragment ions (McLuckey & Goeringer, 1997). In this method, the even-electron peptide ions, (M þ nH)nþ, with closed electronic structures, are generally the subject of collisional activation. Through extensive research, the peptide dissociation mechanism in CAD/CID is now relatively well understood (Wysocki et al., 2000; Paizs & Suhai, 2005; Dongre et al., 2006; Barlow & O’Hair, 2008; Harrison, 2009). In recent years, new tandem mass spectrometry methods based on odd-electron peptide/protein species have emerged as valuable tools in peptide/protein mass spectrometry studies. First, electron capture dissociation (ECD) was introduced in 1998 (Zubarev, Kelleher, & McLafferty, 1998; Zubarev et al., 2000) and was shown to be very powerful to induce extensive peptide/protein backbone dissociations (Sze et al., 2002; Zubarev, 2003; Cooper, Ha˚kansson, & Marshall, 2005; Oh & McLafferty, 2006). Capture of a low kinetic energy electron by even-electron (M þ nH)nþ peptide/protein cations in the Fourier transform-ion cyclotron resonance (FT-ICR) mass spectrometer produces the odd-electron hydrogen abundant species of (M þ nH)(n1)þ•, followed by backbone dissociations that yield characteristic c-and z-type ions as major products and some minor a- and x-type ions. Later, it was shown that electron transfer dissociation (ETD) also produces c- and z-type ions as major fragment ions when applied to peptide cations (Syka et al., 2004; Coon et al., 2005a; Pitteri et al., 2005; Swaney, McAlister, & Coon, 2008). The mechanism of ECD and ETD processes has been the subject of extensive experimental and theoretical studies; however, the mechanism is still not entirely clear, even though the understanding of the ECD and ETD processes has been deepened (Zubarev et al., 2000; Cooper et al., 2002; Oh et al., 2002; Zubarev, 2003; Turecˇek & Syrstad, 2003; Turecˇek, 2003; Leymarie, Costello, & O’Connor, 2003; Breuker et al., 2004; Cooper, Ha˚kansson, & Marshall, 2005; Fung & Chan, 2005; Anusiewicz et al., 2005; O’Connor et al., 2006; Lee et al., 2006; Pouthier & Tsybin, 2008; Sohn et al., 2009; Moore, Ly, & Julian, 2011). In ECD/ETD, the rupture of N–Ca bonds occurs while noncovalent interactions are generally conserved. Therefore, the protein regions where ECD fragmentation is silent can be considered as regions in the protein where
RADICAL-DRIVEN PEPTIDE BACKBONE DISSOCIATION MS/MS
tertiary contacts with other protein regions are intact. Based on this unique ECD property and interpretation, ECD could be used to probe the secondary and tertiary structures of intact protein cations in the gas phase (Oh et al., 2002; Adams et al., 2006; Xie et al., 2006; Lee et al., 2009b). In addition, a few electron-based tandem mass spectrometry methods, such as electron induced dissociation (EID) and electron detachment dissociation (EDD), have been introduced, but these methods are beyond our scope in this review. In parallel with ECD/ETD, a variety of methods that introduce peptide/protein radical ions in the gas phase have been developed. Siu and coworkers discovered that peptide radical cations, M•þ, could be generated by low energy CID of ternary metal complexes with peptide and auxiliary ligands (Chu et al., 2000, 2008; Bagheri-Majdi et al., 2004; Wee, O’Hair, & McFadyen, 2004; Barlow, McFadyen, & O’Hair, 2005; Barlow et al., 2006; Ke et al., 2006; Laskin, Yang, & Chu, 2008; Hopkinson, 2009; Chu & Laskin, 2011; Kalli & Hess, 2012). UV photo-excitation followed by H atom loss provides another way to generate radical peptide ions (Tabarin et al., 2005; Lemoine et al., 2006; Park et al., 2009). UV photoionization of laser-ablated peptides has also been used for the formation of aromatic amino acid-containing peptide radical ions (Grotemeyer & Schlag, 1989; Cui, Hu, & Lifshitz, 2002). Photodetachment of gaseous multiply charged peptides and amino acid anions that are prepared through electrospray ionization (ESI) also provides another route to generate radical anions (Joly et al., 2008; Larraillet et al., 2009; Brunet et al., 2011). In recent years, it was reported that UV irradiation of a photolabile radical precursor generated a site-specific phenyl-type radical with a very high quantitative yield (Ly & Julian, 2008, 2009). A crown ether-based photolabile radical precursor, which forms noncovalent complexes with peptides, has also been used for the preparation of peptide radical cations (Sun et al., 2009). Another notable way to generate peptide radical cations is the free radical initiated peptide sequencing (FRIPS) approach, in which a free radical initiator is introduced to the primary amine group of the lysine side chain or N-terminus of peptides (Masterson et al., 2004; Hodyss, Cox, & Beauchamp, 2005). Upon collisional activation of the precursor peptide ions, a radical site is generated. The fragmentation pathways and behavior of the peptide radical ions are relatively less understood than those of the protonated counterparts. Nevertheless, it has been found that the fragmentation mechanism and behavior of the peptide radical ions are quite different from those of the protonated or deprotonated ions. For example, in the fragmentation of peptide radical ions, a-, c-, x-, and z-type fragment ions are abundantly observed and the side-chain loss provides important structural information. It is also known that the radical site readily migrates to other parts of the peptide or protein, easily interconverting among various isomers (Ly & Julian, 2009). In particular, the structure and stability of small peptide cations have been extensively studied, both experimentally and theoretically, and well documented in recent reviews (Hopkinson, 2009; Chu & Laskin, 2011). Therefore, in this account we do not attempt to give a detailed description of small peptide radical cations in these aspects. Instead, we seek to shed light on the generation of peptide/protein radical ions using a variety of different methods and their radical-based peptide ion fragmentation, with a particular emphasis on larger peptide and Mass Spectrometry Reviews DOI 10.1002/mas
&
protein ion fragmentation chemistry. Considering the large number of publications on peptide radical ions, a summary of the recent results obtained through radical-driven peptide backbone dissociation tandem mass spectrometry, particularly in the analytical aspect, appears to be timely and appropriate. We would like to add that this review is not intended to be comprehensive, but to serve as a good introductory guide for readers who are interested in radical-driven peptide fragmentation chemistry, particularly graduate students and early career scholars in the field of mass spectrometry.
II. FORMATION AND MIGRATION OF A RADICAL SITE A variety of methods have been developed that can generate peptide radical ions. Each method adopts its own unique way to form a radical site and the underlying chemistry is somewhat different in each technique; therefore, we will give short descriptions of these methods. In peptide radical ion fragmentation, it is known that both its stability and the migration of the initial radical site within the peptide ions play a dominant role in dictating final fragmentation sites. In this section, the underlying mechanism with respect to its generation, stability, and migration of the radical site will be summarized.
A. CID of Ternary Metal Complexes With Peptide and Auxiliary Ligands For this method, theoretical and experimental studies have been extensively performed and recently well reviewed (Hopkinson & Siu, 2006; Hopkinson, 2009). Gas-phase redox chemistry of ternary metal complexes with peptide and auxiliary ligands has allowed the formation of cationic peptide radical ions (Chu et al., 2000), and is outlined in Scheme 1 (Barlow, McFadyen, & O’Hair, 2005). CID of the ternary complexes leads to the
SCHEME 1. Outline of a method for producing cationic peptide radicals from ternary metal complexes. [Scheme reproduced from Barlow, McFadyen, and O’Hair (2005), with permission from American Chemical Society, copyright 2005.]
117
&
OH AND MOON
separation of the peptide from the complex, with concomitant reduction of the metal and oxidation of the peptide to form the cationic peptide radical, Pþ• or Mþ•. The subsequent isolation and CID of Pþ• resulted in rich peptide radical ion fragmentation chemistry. In parallel with the formation of Pþ• (Equation 1), CID of the ternary complexes can also lead to a variety of other competitive reaction pathways. The other pathways include dissociative proton transfer to the peptide (Equation 2), dissociative proton abstraction from the peptide (Equation 3), and peptide fragmentation (Equation 4) (Chu, Lam, & Siu, 2005; Ke et al., 2006; Hopkinson, 2009). ½Mnþ ðLÞPnþ ! ½Mnþ ðLÞðn1Þþ þ Pþ
ð1Þ
½Mnþ ðLÞPnþ ! ½Mnþ ðL HÞðn1Þþ þ ½P þ Hþ
ð2Þ
½Mnþ ðLÞPnþ ! ½Mnþ ðP HÞðn1Þþ þ ½L þ Hþ
ð3Þ
½Mnþ ðLÞPnþ ! ½Mðn1Þ ðLÞðP bn Þðn1Þþ þ bn þ
ð4Þ
where Mnþ and L refer to the metal ion and auxiliary ligand, respectively, and P is the peptide species. The role of auxiliary ligands was investigated and some conclusions were drawn (Chu et al., 2004; Barlow et al., 2004; Chu, Lam, & Siu, 2005). A tri- or tetra-dentate ligand was found to be a better ligand in producing radicals because in these cases a ligand would not easily separate from the complex. A tri- or tetra-dentate was more effective in coupling oxidation and dissociation of the peptide compared to situations where bidentate ligands were used (Chu et al., 2000; Barlow et al., 2004; Turecˇek, 2007). It was also found that the formation of radical cationic tripeptides, for example, GGXþ•, was facilitated when a sterically encumbered auxiliary macrocyclic ligand was used instead of its open-chain analog (Lam et al., 2006a; Song et al., 2009). Another interesting finding was that the ligand should not have an acidic proton in order to prevent the protonated peptide from leaving the complex. An appropriate choice of auxiliary ligand, such as 12-crown-4 (1,4,7,10-tetraoxacyclododecane), allowed for the generation of peptide radical cations that contain only aliphatic residues (Chu et al., 2004). On the other hand, it was shown that in addition to Cu2þ ion, which is the best choice of transition metals for carrying out redox chemistry, Cr3þ, Mn3þ, Fe3þ, and Co3þ could be utilized to perform this type of gas-phase redox chemistry (Barlow, McFadyen, & O’Hair, 2005; Laskin, Yang, & Chu, 2008). It was also observed that either aromatic or basic amino acid residues facilitated the production of peptide radical cations (Chu et al., 2000, 2001; Bagheri-Majdi et al., 2004; Wee, O’Hair, & McFadyen, 2006a). Interestingly, it was possible to generate peptide radical ions with a well-defined initial radical site using this approach (Chu et al., 2001, 2008; Wee et al., 2006a, 2006b; Siu et al., 2009). For example, at a [CuII(L)(M)]2þ• CID, dissociative electron transfer occurred to generate Mþ•, where M is either YGG, GYG, or GGY (Chu et al., 2008, Chu & Laskin, 2011). Through Ca–Cb homolytic bond cleavage of Mþ• tyrosyl residues, that is, side chain loss (vide infra), three a-carboncentered radical cations—[G•GG]þ, [GG•G]þ, and [GGG•]þ— with a well-defined initial radical site could be generated. Due to high-energy barriers for interconversion between [G•GG]þ, 118
[GG•G]þ, and [GGG•]þ (44.7 kcal/mol relative to the structure at the global minimum), these isomers existed as isomerically pure species, and the CID spectra of these isomers were substantially different from each other. In addition, other noninterconverting radical cationic isomeric pairs, [G•GW]þ/ [GGW]•þ and [G•GY]þ/[GGY]•þ, which contain an aromatic residue, were also examined, wherein a- and p-carbon centered radicals did not interconvert (Ng et al., 2010). For the peptides in which an aromatic residue such as tyrosine, tryptophan, histidine, or phenylalanine is present, a radical site can be generated in the aromatic side chain; a p-carbon centered radical. CID of p-carbon centered [GGW]•þ/[GGY]•þ peptide radical cations mainly yielded a and c/z-type products, while CID of a-carbon centered [G•GW]þ/[G•GY]þ produced abundant b/y-type ions. The apparently different CID spectra of aand p-carbon centered peptide radical cations indicated that they did not interconvert on the time scale of the CID experiments. These results were also in agreement with the density functional theory (DFT) calculations, in which the isomerization barrier between the GGW radical cations (>35.4 kcal/mol) was higher than the barrier for competitive dissociation of these species (
