Ping-Pong Protons: How Hydrogen-Bonding Networks Facilitate

Feb 20, 2014 - Daiki Asakawa , Asuka Yamashita , Shikiho Kawai , Takae Takeuchi , and Yoshinao Wada. The Journal of Physical Chemistry B 2016 120 (5),...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/JPCB

Ping-Pong Protons: How Hydrogen-Bonding Networks Facilitate Heterolytic Bond Cleavage in Peptide Radical Cations Konstantin O. Zhurov,† Matthew D. Wodrich,†,‡ Clémence Corminboeuf,‡ and Yury O. Tsybin†,* †

Biomolecular Mass Spectrometry Laboratory and ‡Laboratory for Computational Molecular Design, Ecole Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland S Supporting Information *

ABSTRACT: Electron capture and electron transfer dissociation (ECD/ETD) tandem mass spectrometry (MS/MS) are commonly employed techniques for biomolecular analysis. The ECD/ETD process predominately cleaves N−Cα peptide backbone bonds, leading to primary sequence information complementary to other mass spectrometry techniques. Despite frequent laboratory use, the mechanistic underpinnings surrounding N−Cα bond cleavage remain debated. While the majority of mechanisms assume a homolytic bond rupture, we recently showed that heterolytic cleavage is also thermodynamically viable. For a cleavage of this type to be feasible, the charge separation created upon breaking of the N−Cα backbone bond must be quickly annihilated. In this work, we show, using density functional computations, that specific hydrogen-bonding motifs and structural rearrangements involving proton transfers stabilize the transition state associated with heterolytic cleavage and eliminate the ensuing charge separation from the final product fragments. The movement of protons can occur either directly from the z- to c-fragment or in a more complex manner including a ping-pong-type mechanism. The nature of these diverse hydrogen-bonding motifs reveals that not only those functional groups proximate to the bond rupture site, but also the entire global chemical environment, play important roles in backbone cleavage characteristic of ECD/ETD MS/MS. For doubly charged systems, both conformation and electron localization site dictate which of the two fragments retains the final positive charge.



INTRODUCTION Hydrogen-bonding networks play key roles in the physiochemical behavior of myriad molecular systems, including (bio)polymers and polar solvents, but also less intuitive examples such as organometallic complexes. Their ubiquitous presence in diverse chemistry impacts such disparate processes as adsorption,1 luminescence,2 and superconductivity.3 Active research directions focused on developing new medicines,4,5 fuel cells,6 composite materials,7 as well as other molecular systems8−17 further reflect the importance of hydrogen bonding. Of particular interest is the derivation of details surrounding the role of hydrogen-bonding networks in physiological processes, a task frequently accomplished in the laboratory using a variety of experimental techniques including NMR, X-ray crystallography, circular dichroism, and Raman spectroscopy.18−21 The coupling of these experimental methods with computational studies (e.g., electronic structure theory or molecular dynamics simulations) provides additional avenues and support for data interpretation and rationalization.18,20,21 For instance, joint experimental/theoretical studies have demonstrated that small changes in hydrogenbonding networks can profoundly affect both biomolecular structure and the preferred reaction pathway.18,19 Mass spectrometry (MS) studies have also played a key role in © 2014 American Chemical Society

uncovering details associated with the structure of hydrogenbonding networks, including proton movements22 and their associated energetics.3,23−25 Supplementing MS experiments with Kohn−Sham density functional theory (DFT) computations may assist in the clarification and explanation of experimental observations and reveal additional details that further clarify the role of hydrogen-bonding networks in the dissociation of gaseous radical cation peptides. In 1998, McLafferty and co-workers26 developed a new tandem mass spectrometry (MS/MS) technique, termed electron capture dissociation (ECD). In ECD MS/MS, gasphase polycationic peptides or proteins capture low-energy, ∼1 eV, electrons through ion−electron interactions, leading to dissociation of N−Cα bonds in the peptide backbone. A related process, electron transfer dissociation (ETD),27 relies upon electron transfer from a radical anion to a positively charged target molecule. The cleavage specificity of ECD/ETD (ExD) complements other fragmentation techniques, such as collisioninduced dissociation (CID),28 for primary structure and labile post-translational modification analysis.29−32 For these reasons, Received: December 11, 2013 Revised: February 13, 2014 Published: February 20, 2014 2628

dx.doi.org/10.1021/jp412123h | J. Phys. Chem. B 2014, 118, 2628−2637

The Journal of Physical Chemistry B



ExD techniques are now common features of tandem mass spectrometers and are regularly employed for the sequencing and identification of peptides and proteins.33,34 Naturally, numerous mechanistic proposals followed the introduction of ExD MS/MS techniques to explain the electron capture and backbone dissociation events as well as to rationalize experimental data.26,35−42 Two of the most widely accepted mechanisms, the “Cornell”26 and “Utah−Washington”36,37 proposals, involve formation of a carbon-centered radical on a backbone amide, followed by homolytic cleavage of the proximate N−Cα bond located to the C-terminal side of the aminoketyl radical, producing c- and z-fragments. While it is believed that H-bonding networks are important in determining the location of aminoketyl radical formation,42−46 a majority of proposals involving homolytic backbone cleavage do not explore the role played by hydrogen bonding during N−Cα cleavage. On the other hand, a mechanism involving heterolytic N−Cα bond cleavage, such as our recent “enol” proposal (Scheme 1),47,48 likely requires a robust H-bonding network to

Article

COMPUTATIONAL DETAILS

The 131 peptide dication conformers investigated were obtained from a previously generated data set, the details of which are fully described elsewhere.48 In short, starting geometries were obtained from snapshots at regular time intervals from a replica-exchange molecular dynamics (REMD) simulation49−52 in Amber11.53 During the simulations, the stereochemistry of each amino acid was restrained to the Lstereoisomer, while the OC−N−H moiety was restricted to trans configuration. REMD snapshot geometries were first optimized at the PM654 semiempirical level, followed by further geometric and energy refinements using the B3LYP55,56 functional with both a small [3-21G] and medium sized [631G(d)] basis sets. This three-step process gradually reduces the number of conformers considered based on their energies,57 a process that ensures each dication precursor species conformer considered possesses well-solvated charges. The geometries of these low-lying structures were then reoptimized as radical cations, which describe the charge-reduced ground state existing after electron capture. Transition state (TS) geometries associated with peptide backbone cleavages were obtained at the B3LYP/6-31G(d) level. In this work, new minima and TS geometries of the radical cation species were obtained by optimizations at the ωB97X-D58/6-31G(d) theoretical level (in addition to the data provided in the paper; see Table S2, Supporting Information). This functional choice is expected to provide improved energetic assessments over other popular density functionals (e.g., B3LYP) as it reduces problems associated with the delocalization (or selfinteraction) error that affects radical cation peptides59 and other systems where two molecules (or fragments) share a single charge.60−64 All minima and transition states were confirmed by vibrational frequency analysis (zero and one imaginary frequencies for minima and transition states, respectively). The Gaussian0965 software package was used for all PM6 and density functional computations.

Scheme 1. Overview of the Enol Mechanism Involving Heterolytic N−Cα Bond Cleavagea



a

RESULTS AND DISCUSSION Analysis of our previously generated conformer set of model tryptic peptides48 revealed several distinguishing features of the H-bonding networks, which permitted assignment of each conformer into one of three major categories that define the mechanism of charge separation annihilation.66 The three categories, identified by their specific H-bonding motifs, involve the two amide groups flanking the N−Cα bond of interest (Scheme 2). In each category, a c-fragment amide oxygen or nitrogen (blue, Scheme 2) is H-bonded to either a protonated site or a neutral site possessing acidic hydrogen atoms (green, Scheme 2). The local negative charge produced by heterolytic cleavage of the backbone N−Cα bond, which is temporarily delocalized between the two heteroatoms (blue), is annihilated by a proton transfer, creating a final product characterized by an iminol [(OH)CNH] or an amide [OC−NH2] moiety. The existence of these types of proton transfers indicate that the zwitterionic c-fragment intermediate is likely short-lived, which minimizes the thermodynamically problematic issue of charge separation. Moreover, radical delocalization over the zfragment Cα−amide moiety (red, Scheme 2) provides an additional thermodynamic driving force that benefits heterolytic N−Cα bond cleavage. Delocalization of this type occurs either through a proton back-transfer47 of the oxygen-bound

Postfragmentation proton transfers are driven by annihilation of the negative charge that is delocalized over a portion of the c-fragment.

thermodynamically stabilize the charge separation formed upon bond cleavage and c-/z-fragment formation. For a set of prototypical tryptic peptides (AxK, x = 3−5), our previous work demonstrated the viability of a heterolytic cleavage mechanism, by showing it is frequently the kinetically preferred fragmentation pathway in comparison to the Cornell mechanism.48 However, reduction in peptide length resulted in a smaller percentage of conformers favoring a heterolytic N− Cα bond cleavage. Presumably, this arises from loss of flexibility in the H-bonding network which is unable to fully stabilize the zwitterionic transition state and remove charge separation within the nascent products via proton transfers.48 The objective of this work is to more closely examine the thermodynamic role played by postfragmentation rearrangements of the H-bonding network. We demonstrate that peptide secondary structure dictates the favored mechanistic pathway, as well as the ultimate charge distribution between the two fragments. 2629

dx.doi.org/10.1021/jp412123h | J. Phys. Chem. B 2014, 118, 2628−2637

The Journal of Physical Chemistry B

Article

been provided, they impart an overview of the magnitude of stabilization from various reaction steps. It is expected that the final thermodynamic stabilization for each individual conformer will be slightly different based on myriad factors. Finally, although each motif is described as a separate entity, many conformers possess several thermodynamically competitive pathways. Motif 1: Proton Transfer from a Protonated Site. The first motif involves a direct proton transfer from either the protonated N-terminus or lysine side chain to the c-fragment amide oxygen (blue, Scheme 2). Two subsets of motif 1 exist: in the first subset (motif 1A), the two amide oxygen atoms flanking the N−Cα bond of interest are only H-bonded to a single protonation site prior to electron capture, while in the second subset (motif 1B), the oxygen atoms of the two amide groups are H-bonded to two separate protonation sites. The first subset, motif 1A, involves the two amide oxygen atoms (blue and red, Figure 1a) surrounding the N−Cα bond being cleaved, which are solvated by a single NH3 protonation site (orange, Figure 1a). The resulting cyclic TS simultaneously cleaves the N−Cα bond and transfers two protons; the first from the protonated site to the amide oxygen (blue, Figure 1a), which eliminates the negative charge, and a second backtransfer from the aminoketyl oxygen (red, Figure 1a) to its original position on the amine group (orange, Figure 1a). The end result is zwitterion elimination and radical delocalization over a portion of the z-fragment, both of which are thermodynamically favorable processes. The entire motif 1A occurs spontaneously after overcoming the TS barrier associated with N−Cα cleavage (Figure 1b), highlighting the concerted nature of the bond cleavage and both proton transfers. A slight variation of motif 1A also exists in which the

Scheme 2. Schematic Representation of the Key Functional Groups Involved in Heterolytic N−Cα Bond Cleavagea

a

Orange indicates the hydrogen donor site after initial electron capture, while the aminoketyl radical is highlighted in red. The adjacent amide group, which bears the postcleavage negative charge, is highlighted in blue. A protonated site or an acidic hydrogen is Hbonded to the c-fragment amide group and is highlighted in green. The arrow indicates the initial N−Cα bond cleavage direction.

hydrogen (orange, Scheme 2) or by removal of an amide hydrogen directly from the backbone (red, Scheme 2). As previously stated, each conformer can be placed into one of three main H-bonding network motifs (Table S1, Supporting Information). In the first group, the c-fragment amide oxygen or nitrogen (blue, Scheme 2) solvates a protonated site (green, Scheme 2), while in the second group, one of the c-fragment amide heteroatoms (blue, Scheme 2) is H-bonded to a proximate acidic hydrogen (green, Scheme 2). The third group incorporates the C-terminus into the vicinal H-bonding network (purple, vide infra). Below, the mechanistic and thermodynamic specifics for seven prototypical subcases involving each of the three network motifs are expounded. While the thermodynamics for each of these subcases have

Figure 1. Schematic (a) and prototypical (b) example of the motif 1A proton rearrangement mechanism, where the protonated site is solvated by the two amide groups surrounding the N−Cα bond being cleaved. After N−Cα cleavage, an iminol is spontaneously formed on the c-fragment. 2630

dx.doi.org/10.1021/jp412123h | J. Phys. Chem. B 2014, 118, 2628−2637

The Journal of Physical Chemistry B

Article

Figure 2. Schematic (a) and prototypical (b) example of the motif 1B proton rearrangement mechanism, where a c-fragment amide oxygen is hydrogen-bonded directly to a protonated site, resulting in formation of an iminol on the c-fragment.

Figure 3. Schematic (a) and prototypical (b) example of the motif 2A proton rearrangement mechanism, where the c-fragment amide oxygen is hydrogen-bonded to a z-fragment backbone hydrogen. An interfragmental transfer of the backbone amide proton from the z-fragment to the cfragment delocalizes the radical over a portion of the z-fragment peptide backbone and forms an iminol on the c-fragment.

2 kcal/mol (Figure 2b). Zwitterion elimination is also accompanied by proton back-transfer from the aminoketyl radical (red, Figure 2a) to the original H-donor site (orange, Figure 2a), which permits delocalization of the radical over a portion of the z-fragment and stabilizes the final products. Motif 2: Proton Transfer from a Neutral Site. The second motif involves the c-fragment amide oxygen being Hbonded to an amide hydrogen atom located on either the zfragment (motif 2A) or the c-fragment (motif 2B). For z-

amide nitrogen (blue, Figure 1a), as opposed to the amide oxygen (blue, Figure 1a), is involved in the cyclic TS, resulting in amide formation on the final c-fragment. For motif 1B, heterolytic N−Cα cleavage is associated with a direct proton transfer from the positively charged NH3 (green) group to the amide oxygen (blue), resulting in instantaneous annihilation of the negative charge located on the c-fragment (Figure 2a). The process is highly exothermic and is generally associated with a small TS barrier that typically does not exceed 2631

dx.doi.org/10.1021/jp412123h | J. Phys. Chem. B 2014, 118, 2628−2637

The Journal of Physical Chemistry B

Article

Figure 4. Schematic (a) and prototypical (b) example of the motif 2B proton rearrangement mechanism, where the c-fragment amide oxygen is hydrogen-bonded to a backbone hydrogen within the same fragment. The final iminol product is formed on the c-fragment after a series of proton transfers from the N-terminus.

Figure 5. Schematic (a) and prototypical (b) example of the motif 3A proton rearrangement mechanism, where the c-fragment amide oxygen is hydrogen-bonded to the C-terminus, which is also hydrogen-bonded to a protonated site. A proton transfer from the C-terminus first forms an iminol on the c-fragment, followed by a second proton transfer from the protonated site to restore the free acid hydrogen to the C-terminus.

back-transfer generally seen after N−Cα cleavage, the zfragment radical is instead delocalized over a portion of the peptide backbone, while an iminol or amide is formed on the cfragment. Removal of a backbone proton from an NH group and subsequent delocalization of the radical over the z-fragment

fragment-based amide hydrogens (motif 2A), an interfragmental proton transfer occurs between a negatively charged heteroatom (blue, Figure 3a) on the c-fragment and a zfragment backbone hydrogen atom of the aminoketyl radical NH group (red, Figure 3a). Rather than undergoing the proton 2632

dx.doi.org/10.1021/jp412123h | J. Phys. Chem. B 2014, 118, 2628−2637

The Journal of Physical Chemistry B

Article

Figure 6. Schematic (a) and prototypical (b) example of the motif 3B proton rearrangement mechanism, where the c-fragment amide oxygen is hydrogen-bonded to the C-terminus and the amide nitrogen is hydrogen-bonded to a backbone hydrogen on the z-fragment. Proton transfer from the C-terminus to the amide oxygen first forms an iminol on the c-fragment. A second proton transfer from the z-fragment backbone NH group to the amide nitrogen causes the initial proton to migrate back to the C-terminus in a “ping-pong”-type motion.

Motif 3: H-Bonding to the C-Terminus. A final motif frequently found among conformers that thermodynamically prefer heterolytic N−Cα cleavage involves H-bonding between the c-fragment amide oxygen (blue) and the C-terminus free acid hydrogen (purple, vide infra). As with motifs 1 and 2, several possible scenarios exist, distinguished by the specifics of H-bonding incorporating the C-terminus. Briefly, in motif 3A, the C-terminus is additionally H-bonded directly to a protonation site; in motif 3B, the C-terminus is involved in H-bonding only with the c-fragment amide oxygen; and in motif 3C, the C-terminus is H-bonded to both a c-fragment amide oxygen and a neutral site, generally an amide hydrogen. In motif 3A, a proton spontaneously transfers from the Cterminus to the negatively charged c-fragment amide oxygen (blue, Figure 5a), leaving a negatively charged C-terminus (purple, Figure 5a) and an iminol moiety on the c-fragment. A second proton transfer, usually associated with a small TS barrier, from the remaining positively charged amine group (green, Figure 5a) restores the free acid hydrogen to the Cterminus (Figure 5b). Situations where the neutralized zwitterion product is less stable than an intermediate species also exist, suggesting possible formation of a salt bridge between the C-terminus and the protonated site, which would permit delocalization of the negative charge over the two Cterminus oxygen atoms (purple, Figure 5a) and create a strong Coulombic attraction with the positively charged protonation site (green, Figure 5a). This type of attraction could be a contributing factor in observation of the charge-reduced species in ExD mass spectra. For motif 3B, the C-terminus (purple, Figure 6a) is involved in H-bonding only with the amide oxygen (blue) adjacent to

backbone forbids proton back-transfer to the original H-donor site. As such, this situation represents a slight variation of the original enol mechanism. Thermodynamically, the TS barrier height associated with proton transfer is often small (e.g., ∼1 kcal/mol), while radical delocalization ensures that the process is exothermic overall, as illustrated in Figure 3b for a prototypical example. Proton transfer involving the removal of c-fragment backbone amide hydrogens (motif 2B) provides an example of coupled proton movements within a single fragment. A prototypical case involves a negatively charged amide oxygen (blue, Figure 4a) that is H-bonded to a backbone amide hydrogen (green, Figure 4a) of an adjacent amino acid. Since conformers fitting motif 2B tend to lack complex interfragmental H-bonding networks, a series of intrafragmental transfers along the c-fragment backbone may occur. In essence, the anion located at the c-fragment periphery (second panel, Figure 4a) “tugs” on the H-bonding network, resulting in the migration of multiple protons, which generally culminates at the still positively charged N-terminus (second and third panels, Figure 4a). The mechanism governing this type of proton movement is quite flexible: cases exist in which one or more amino acid residues with “improperly” aligned H-bonding networks are bypassed or where multiple proton movements are coupled in a concerted process. The energetics associated with this “tug” process are quite favorable (generally TS barriers do not exceed 7 kcal/mol) and could easily be overcome at experimental operating temperatures, with the final rearranged product being the most thermodynamically stable (Figure 4b).67 2633

dx.doi.org/10.1021/jp412123h | J. Phys. Chem. B 2014, 118, 2628−2637

The Journal of Physical Chemistry B

Article

Figure 7. Schematic (a) and prototypical (b) example of the motif 3C proton rearrangement mechanism, where the c-fragment amide oxygen is hydrogen-bonded to the C-terminus, which is further hydrogen-bonded to a c-fragment amide hydrogen. Proton transfer from the C-terminus to the amide oxygen first forms an iminol on the c-fragment. The negatively charged C-terminus then causes a series of proton migrations in c-fragment, which results in eventual proton transfer from a backbone NH group of the c-fragment to the C-terminus.

separation via annihilation of the c-fragment zwitterion created during heterolytic N−Cα cleavage. Each of these motifs produces the same product types: a c-fragment characterized by an amide or iminol and a z-fragment containing a delocalized radical. Importantly, motifs 2 and 3 illustrate that atoms and structures beyond the immediate chemical environment are heavily involved in heterolytic N−Cα bond cleavage. Thus, the specific c- and z-fragment secondary structures play a role in the system energetics, making the extended H-bonding network critical for the thermodynamic viability of heterolytic N−Cα cleavages. As such, the dissociation of peptides by this type of mechanism likely becomes increasingly common in larger systems and potentially significantly impacts the fragmentation of proteins, as their large structures contain diverse H-bonding networks. Motif 3 specifically highlights the enhanced complexity of available pathways, where inclusion of a single functional group opens access to multiple thermodynamically viable proton transfers spanning an entire peptide. This finding points to the potentially important roles played by side chain functional groups, such as aspartic and glutamic acids and, to a lesser extent, serine, threonine, and tyrosine that may act in a similar manner as the C-terminus. Conformer Distribution among Different H-Bonding Motifs. While each of the motifs presented above may occur in any peptide system, an additional important consideration is how relevant specific hydrogen-bonding motifs are in systems of different size. For the set of peptides used here, each system (A3K, A4K, and A5K) includes conformers representing each motif. For A3K and A4K, motifs 1 and 2 appear in roughly equal numbers, whereas motif 3 appears less frequently (Table 1). This trend is even more pronounced for the largest peptide system studied, where motif 3 represents only ∼8% of cases. This observation can be attributed to an increase in peptide length, where the numerous alternative secondary structures decrease the likelihood of the C-terminus being proximate to

the cleavage site. In this case, a proton migrates from the Cterminus to the negatively charged c-fragment amide oxygen (blue), temporarily forming an iminol (Figure 6a, second panel). A second proton then transfers from the z-fragment backbone NH group (Figure 6a, third panel in red) of the aminoketyl radical to the c-fragment iminol NH group (blue), which causes the first proton to migrate back to its original position on the C-terminus in a “ping-pong”-type movement. A slight variation of motif 3B also exists where a proton directly transfers to the c-fragment amide nitrogen. Alternatively, despite being H-bonded to the C-terminus, the c-fragment amide oxygen (blue, Figure 6a) can also accept a proton from the aminoketyl backbone hydrogen (red, Figure 6a). Finally, in motif 3C, the C-terminus (purple, Figure 7a) is Hbonded to a backbone amide group (green, Figure 7a) that is connected via additional H-bonds to a protonated site. Structures of this type give rise to an additional mechanism, where a proton first migrates from the C-terminus (purple) to the negatively charged amide oxygen atom (blue) of the cfragment (Figure 7a, first and second panels). Formation of the c-fragment iminol coupled with the close proximity of the negatively charged C-terminus causes two additional migrations: first from the positively charged amine group to an adjacent amide oxygen (Figure 7a, second and third panels) and then from the neighboring backbone amide group to the C-terminus (Figure 7a, third and fourth panels). Depending on the secondary structure of the peptide, this type of hydrogen rearrangement may occur over several amino acid residues and could bypass any residues with improperly aligned H-bonding patterns. As with each of the other motifs, a rearrangement of this type is accompanied by only modest TS barriers and would likely proceed quickly after initial N−Cα rupture (Figure 7b). As illustrated above, three main proton migration motifs (motifs 1, 2, and 3) exist involving rearrangements of the Hbonding network to eliminate energetically unfavorable charge 2634

dx.doi.org/10.1021/jp412123h | J. Phys. Chem. B 2014, 118, 2628−2637

The Journal of Physical Chemistry B

Article

findings highlight the key role played by the extended chemical environment in determining the thermodynamic viability of various backbone bond cleavages. As proton transfers appear to be common occurrences in these prototypical systems that favor heterolytic backbone cleavage, the charge location (on either the c- and z-fragment) depends both upon the electron localization site and the specific three-dimensional hydrogenbonding network of the particular conformer.

Table 1. Instances of Different H-Bonding Motifs for the 131 Conformers That Preferentially Undergo Enol Cleavage peptide

motif 1

motif 2

motif 3

Ala5-Lys Ala4-Lys Ala3-Lys

35 19 7

21 20 6

5 14 4



the backbone cleavage site. Nonetheless, it is important to note that each motif is observed in all systems studied here, a trend that likely persists throughout other peptides. Which Fragment Will Remain Charged? A final important point is establishing if the c- or z-fragment retains the positive charge after separation. As a result of the types of interfragmental proton transfers discussed above, it is clear that the final charge is not solely a function of the electron localization site. In the classical picture, all postcleavage interactions between the fragments, either rearrangements or hydrogen abstractions, involve the movement of neutral entities between the two products.68−70 It follows that, for a doubly charged system, where the two charges lie near the opposite termini, electron localization at one positively charged site (e.g., the N-terminus) results in the opposite fragment being charged (e.g., the z-fragment).26 For heterolytic N−Cα cleavage, as well as other mechanisms involving eventual electron migration onto a backbone amide group,36,37 pre- or postfragmentation proton transfers are critical steps in final product formation. Table 2 lists possible scenarios for which of the two fragments

* Supporting Information Distribution of conformers among different H-bonding motifs, along with electronic energies and Cartesian coordiantes of relevant species are provided. This material is available free of charge via the Internet at http://pubs.acs.org.



donor site

charged fragment

charge location

1a 1a 1b 1b 2a 2a 2b

Lys N-terminus Lys N-terminus Lys N-terminus Lys

Lys N-terminus N-terminus Lys N-terminus Lys Lys

ELS ELS SPS SPS SPS SPS ELS

AUTHOR INFORMATION

Corresponding Author

*E-mail: yury.tsybin@epfl.ch. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The work was supported by the Swiss National Science Foundation (Grants 20021-125147 and 147006 to Y.O.T. and 20021-12577/1 to C.C.) and the European Research Council (ERC Starting Grant 280271 to Y.O.T.) and the EPFL. Ryan Julian is acknowledged for a stimulating discussion.



Table 2. List of H-Bonding Motif Types for the AnK Peptide Seta motif

ASSOCIATED CONTENT

S

REFERENCES

(1) Li, S.-C.; Chu, L.-N.; Gong, X.-Q.; Diebold, U. Hydrogen Bonding Controls the Dynamics of Catechol Adsorbed on a TiO2(110) Surface. Science 2010, 328, 882−884. (2) Raffy, G.; Ray, D.; Chu, C.-C.; Del Guerzo, A.; Bassani, D. M. Controlling the Emission Polarization from Single Crystals Using Light: Towards Photopolic Materials. Angew. Chem., Int. Ed. 2011, 50, 9584−9588. (3) Duan, D.; Tian, F.; He, Z.; Meng, X.; Wang, L.; Chen, C.; Zhao, X.; Liu, B.; Cui, T. Hydrogen Bond Symmetrization and Superconducting Phase of HBr and HCl under High Pressure: An Ab Initio Study. J. Chem. Phys. 2010, 133, 074509. (4) Reum, N.; Fink-Straube, C.; Klein, T.; Hartmann, R. W.; Lehr, C.-M.; Schneider, M. Multilayer Coating of Gold Nanoparticles with Drug−Polymer Coadsorbates. Langmuir 2010, 26, 16901−16908. (5) Shariatinia, Z.; Mirhosseini Mousavi, H. S.; Bereciartua, P. J.; Dusek, M. Structures of a Novel Phosphoric Triamide and Its Organotin(IV) Complex. J. Organomet. Chem. 2013, 745−746, 432− 438. (6) Oh, S.-Y.; Kikuchi, T.; Kawamura, G.; Muto, H.; Matsuda, A. Proton Conductive Composite Electrolytes in the KH2PO4− H3PW12O40 System for H2/O2 Fuel Cell Operation. Appl. Energy 2013, 112, 1108−1114. (7) Glowacki, E. D.; Irimia-Vladu, M.; Bauer, S.; Sariciftci, N. S. Hydrogen-Bonds in Molecular Solids: From Biological Systems to Organic Electronics. J. Mater. Chem. B 2013, 1, 3742−3753. (8) Thirumurugan, A.; Li, W.; Cheetham, A. K. Bismuth 2,6Pyridinedicarboxylates: Assembly of Molecular Units into Coordination Polymers, CO2 Sorption and Photoluminescence. Dalton Trans. 2012, 41, 4126−4134. (9) Bako, I.; Bencsura, A.; Hermannson, K.; Balint, S.; Grosz, T.; Chihaia, V.; Olah, J. Hydrogen Bond Network Topology in Liquid Water and Methanol: A Graph Theory Approach. Phys. Chem. Chem. Phys. 2013, 15, 15163−15171. (10) Jia, Y.-Y.; Liu, B.; Liu, X.-M.; Yang, J.-H. Syntheses, Structures and Magnetic Properties of Two Heterometallic Carbonates: K2Li-

a

Donor site indicates the initial electron localization site; charged fragment column indicates which protonation site ultimately remains charged. Charge location indicates whether the charge remains on the initial electron localization site (ELS) or on the second protonation site (SPS). Note that for motif 3 the final charge location cannot be easily determined.

remain charged for the H-bonding motifs 1 and 2,71 discussed above. Since proton transfer can occur either inter- or intrafragmentally, the final charged fragment depends upon both the initial electron localization site and the specific threedimensional conformer structure.



CONCLUSION In this work, we illustrated three major classes of hydrogenbonding motifs relevant in promoting heterolytic N−Cα cleavage in the ECD/ETD MS/MS of polypeptides. Specific bonding motifs were found to play a key role in the expeditious elimination of the zwitterion formed upon heterolytic N−Cα cleavage. Various proton transfers (including ping-pong-type movements) significantly stabilize the resulting fragments not only by facilitating charge recombination but also by permitting radical delocalization over a portion of the z-fragment. These 2635

dx.doi.org/10.1021/jp412123h | J. Phys. Chem. B 2014, 118, 2628−2637

The Journal of Physical Chemistry B

Article

[Cu(H2O)2Ru2(CO3)4X2]·5H2O (X = Cl, Br). CrystEngComm 2013, 15, 7936−7942. (11) Teplyakov, A.; Zhao, Y.; Malia, T. J.; Obmolova, G.; Gilliland, G. L. IgG2 Fc Structure and the Dynamic Features of the IgG CH2− CH3 Interface. Mol. Immunol. 2013, 56, 131−139. (12) Puntus, L. N.; Lyssenko, K. A.; Pekareva, I. S.; Bünzli, J.-C. G. Intermolecular Interactions as Actors in Energy-Transfer Processes in Lanthanide Complexes with 2,2′-Bipyridine. J. Phys. Chem. B 2009, 113, 9265−9277. (13) Mishiro, K.; Furuta, T.; Sasamori, T.; Hayashi, K.; Tokitoh, N.; Futaki, S.; Kawabata, T. A Cyclochiral Conformational Motif Constructed Using a Robust Hydrogen-Bonding Network. J. Am. Chem. Soc. 2013, 135, 13644−13647. (14) Nishino, T.; Hayashi, N.; Bui, P. T. Direct Measurement of Electron Transfer through a Hydrogen Bond between Single Molecules. J. Am. Chem. Soc. 2013, 135, 4592−4595. (15) Nagornova, N. S.; Rizzo, T. R.; Boyarkin, O. V. Interplay of Intra- and Intermolecular H-Bonding in a Progressively Solvated Macrocyclic Peptide. Science 2012, 336, 320−323. (16) Lv, H.; Jia, W.-Q.; Sun, L.-H.; Ye, S. N-Heterocyclic Carbene Catalyzed [4 + 3] Annulation of Enals and o-Quinone Methides: Highly Enantioselective Synthesis of Benzo-ε-lactones. Angew. Chem., Int. Ed. 2013, 52, 8607−8610. (17) Offenbacher, A. R.; Minnihan, E. C.; Stubbe, J.; Barry, B. A. Redox-Linked Changes to the Hydrogen-Bonding Network of Ribonucleotide Reductase β2. J. Am. Chem. Soc. 2013, 135, 6380− 6383. (18) Joseph, P. R. B.; Poluri, K. M.; Gangavarapu, P.; Rajagopalan, L.; Raghuwanshi, S.; Richardson, R. M.; Garofalo, R. P.; Rajarathnam, K. Proline Substitution of Dimer Interface β-Strand Residues as a Strategy for the Design of Functional Monomeric Proteins. Biophys. J. 2013, 105, 1491−1501. (19) Gregory, M.; Mak, P. J.; Sligar, S. G.; Kincaid, J. R. Differential Hydrogen Bonding in Human CYP17 Dictates Hydroxylation versus Lyase Chemistry. Angew. Chem., Int. Ed. 2013, 52, 5342−5345. (20) Kabasakal, B. V.; Gae, D. D.; Li, J.; Lagarias, J. C.; Koehl, P.; Fisher, A. J. His74 Conservation in the Bilin Reductase PcyA Family Reflects an Important Role in Protein−Substrate Structure and Dynamics. Arch. Biochem. Biophys. 2013, 537, 233−242. (21) Light, K. M.; Hangasky, J. A.; Knapp, M. J.; Solomon, E. I. Spectroscopic Studies of the Mononuclear Non-Heme FeII Enzyme FIH: Second-Sphere Contributions to Reactivity. J. Am. Chem. Soc. 2013, 135, 9665−9674. (22) Wysocki, V. H.; Tsaprailis, G.; Smith, L. L.; Breci, L. A. Mobile and Localized Protons: A Framework for Understanding Peptide Dissociation. J. Mass. Spectrom. 2000, 35, 1399−1406. (23) Su, H.-F.; Xue, L.; Li, Y.-H.; Lin, S.-C.; Wen, Y.-M.; Huang, R.B.; Xie, S.-Y.; Zheng, L.-S. Probing Hydrogen Bond Energies by Mass Spectrometry. J. Am. Chem. Soc. 2013, 135, 6122−6129. (24) Tao, Y.; Julian, R. Factors that Influence Competitive Intermolecular Solvation of Protonated Groups in Peptides and Proteins in the Gas Phase. J. Am. Soc. Mass Spectrom. 2013, 24, 1634− 1640. (25) Bokatzian-Johnson, S. S.; Stover, M. L.; Dixon, D. A.; Cassady, C. J. Gas-Phase Deprotonation of the Peptide Backbone for Tripeptides and Their Methyl Esters with Hydrogen and Methyl Side Chains. J. Phys. Chem. B 2012, 116, 14844−14858. (26) Zubarev, R. A.; Kelleher, N. L.; McLafferty, F. W. Electron Capture Dissociation of Multiply Charged Protein Cations. A Nonergodic Process. J. Am. Chem. Soc. 1998, 120, 3265−3266. (27) Syka, J. E. P.; Coon, J. J.; Schroeder, M. J.; Shabanowitz, J.; Hunt, D. F. Peptide and Protein Sequence Analysis by Electron Transfer Dissociation Mass Spectrometry. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 9528−9533. (28) Wells, J. M.; McLuckey, S. A. Methods in Enzymology; Academic Press: New York, 2005; Vol. 402, pp 148−185. (29) Olsen, J. V.; Haselmann, K. F.; Nielsen, M. L.; Budnik, B. A.; Nielsen, P. E.; Zubarev, R. A. Comparison of Electron Capture Dissociation and Collisionally Activated Dissociation of Polycations of

Peptide Nucleic Acids. Rapid Commun. Mass Spectrom. 2001, 15, 969− 974. (30) Zubarev, R. A.; Zubarev, A. R.; Savitski, M. M. Electron Capture/Transfer versus Collisionally Activated/Induced Dissociations: Solo or Duet? J. Am. Soc. Mass Spectrom. 2008, 19, 753−761. (31) Kelleher, N. L.; Zubarev, R. A.; Bush, K.; Furie, B.; Furie, B. C.; McLafferty, F. W.; Walsh, C. T. Localization of Labile Posttranslational Modifications by Electron Capture Dissociation: The Case of γCarboxyglutamic Acid. Anal. Chem. 1999, 71, 4250−4253. (32) Savitski, M. M.; Kjeldsen, F.; Nielsen, M. L.; Zubarev, R. A. Complementary Sequence Preferences of Electron-Capture Dissociation and Vibrational Excitation in Fragmentation of Polypeptide Polycations. Angew. Chem., Int. Ed. 2006, 45, 5301−5303. (33) Zhurov, K. O.; Fornelli, L.; Wodrich, M. D.; Laskay, U. A.; Tsybin, Y. O. Principles of Electron Capture and Transfer Dissociation Mass Spectrometry Applied to Peptide and Protein Structure Analysis. Chem. Soc. Rev. 2013, 42, 5014−5030. (34) Tureček, F.; Julian, R. R. Peptide Radicals and Cation Radicals in the Gas Phase. Chem. Rev. 2013, 113, 6691−6733. (35) Zubarev, R. A.; Kruger, N. A.; Fridriksson, E. K.; Lewis, M. A.; Horn, D. M.; Carpenter, B. K.; McLafferty, F. W. Electron Capture Dissociation of Gaseous Multiply-Charged Proteins Is Favored at Disulfide Bonds and Other Sites of High Hydrogen Atom Affinity. J. Am. Chem. Soc. 1999, 121, 2857−2862. (36) Sawicka, A.; Skurski, P.; Hudgins, R. R.; Simons, J. Model Calculations Relevant to Disulfide Bond Cleavage via Electron Capture Influenced by Positively Charged Groups. J. Phys. Chem. B 2003, 107, 13505−13511. (37) Syrstad, E. A.; Tureček, F. Toward a General Mechanism of Electron Capture Dissociation. J. Am. Soc. Mass Spectrom. 2005, 16, 208−224. (38) Sobczyk, M.; Anusiewicz, I.; Berdys-Kochanska, J.; Sawicka, A.; Skurski, P.; Simons, J. Coulomb-Assisted Dissociative Electron Attachment: Application to a Model Peptide. J. Phys. Chem. A 2004, 109, 250−258. (39) Tureček, F.; Syrstad, E. A. Mechanism and Energetics of Intramolecular Hydrogen Transfer in Amide and Peptide Radicals and Cation-Radicals. J. Am. Chem. Soc. 2003, 125, 3353−3369. (40) Savitski, M. M.; Kjeldsen, F.; Nielsen, M. L.; Zubarev, R. A. Hydrogen Rearrangement to and from Radical z Fragments in Electron Capture Dissociation of Peptides. J. Am. Soc. Mass Spectrom. 2007, 18, 113−120. (41) Leymarie, N.; Costello, C. E.; O’Connor, P. B. Electron Capture Dissociation Initiates a Free Radical Reaction Cascade. J. Am. Chem. Soc. 2003, 125, 8949−8958. (42) Patriksson, A.; Adams, C.; Kjeldsen, F.; Raber, J.; van der Spoel, D.; Zubarev, R. A. Prediction of N−Cα Bond Cleavage Frequencies in Electron Capture Dissociation of Trp-Cage Dications by Force-Field Molecular Dynamics Simulations. Int. J. Mass Spectrom. 2006, 248, 124−135. (43) Breuker, K.; Oh, H. B.; Lin, C.; Carpenter, B. K.; McLafferty, F. W. Nonergodic and Conformational Control of the Electron Capture Dissociation of Protein Cations. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 14011−14016. (44) Polfer, N. C.; Haselmann, K. F.; Langridge-Smith, P. R. R.; Barran, P. E. Structural Investigation of Naturally Occurring Peptides by Electron Capture Dissociation and AMBER Force Field Modelling. Mol. Phys. 2005, 103, 1481−1489. (45) Pouthier, V.; Tsybin, Y. O. Amide-I Relaxation-Induced Hydrogen Bond Distortion: An Intermediate in Electron Capture Dissociation Mass Spectrometry of α-Helical Peptides? J. Chem. Phys. 2008, 129, 095106. (46) Ben Hamidane, H.; He, H.; Tsybin, O. Y.; Emmett, M. R.; Hendrickson, C. L.; Marshall, A. G.; Tsybin, Y. O. Periodic Sequence Distribution of Product Ion Abundances in Electron Capture Dissociation of Amphipathic Peptides and Proteins. J. Am. Soc. Mass Spectrom. 2009, 20, 1182−1192. (47) Wodrich, M. D.; Zhurov, K. O.; Vorobyev, A.; Ben Hamidane, H.; Corminboeuf, C.; Tsybin, Y. O. Heterolytic N−Cα Bond Cleavage 2636

dx.doi.org/10.1021/jp412123h | J. Phys. Chem. B 2014, 118, 2628−2637

The Journal of Physical Chemistry B

Article

in Electron Capture and Transfer Dissociation of Peptide Cations. J. Phys. Chem. B 2012, 116, 10807−10815. (48) Wodrich, M. D.; Zhurov, K. O.; Corminboeuf, C.; Tsybin, Y. O. On the Viability of Heterolytic Peptide N−Cα Bond Cleavage in Electron Capture and Transfer Dissociation Mass Spectrometry. J. Phys. Chem. B 2014, DOI: 10.1021/jp500512a. (49) Sugita, Y.; Okamoto, Y. Replica-Exchange Molecular Dynamics Method for Protein Folding. Chem. Phys. Lett. 1999, 314, 141−151. (50) Mitsutake, A.; Sugita, Y.; Okamoto, Y. Generalized-Ensemble Algorithms for Molecular Simulations of Biopolymers. Biopolymers 2001, 60, 96−123. (51) Nymeyer, H.; Gnanakaran, S.; García, A. E. Atomic Simulations of Protein Folding Using the Replica Exchange Algorithm. Methods Enzymol. 2004, 383, 119−149. (52) Cheng, X.; Cui, G.; Hornak, V.; Simmerling, C. Modified Replica Exchange Simulation Methods for Local Structure Refinement. J. Phys. Chem. B 2005, 109, 8220−8230. (53) Case, D. A.; Darden, T. A.; Cheatham, T. E., III; Simmerling, C. L.; Wang, J.; Duke, R. E.; Luo, R.; Walker, R. C.; Zhang, W.; Merz, K. M.; et al. AMBER 11; University of California, San Francisco, 2010. (54) Stewart, J. J. P. Optimization of Parameters for Semiempirical Methods V: Modification of NDDO Approximations and Application to 70 Elements. J. Mol. Model. 2007, 13, 1173−1213. (55) Becke, A. D. Density-Functional Thermochemistry. III. The Role of Exact Exchange. J. Chem. Phys. 1993, 98, 5648−5652. (56) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti Correlation-Energy Formula into a Functional of the Electron Density. Phys. Rev. B 1988, 37, 785−789. (57) For a prototypical system (A5K), the following values are present after each step: 8000 structures (obtained from REMD Snapshots), 1913 (after PM6 semiempirical refinement), 600 (after B3LYP/3-21G refinement), 237 (after B3LYP/6-31G(d) refinement). (58) Chai, J.-D.; Head-Gordon, M. Long-Range Corrected Hybrid Density Functionals with Damped Atom−Atom Dispersion Corrections. Phys. Chem. Chem. Phys. 2008, 10, 6615−6620. (59) Gilson, A. I.; van der Rest, G.; Chamot-Rooke, J.; Kurlancheek, W.; Head-Gordon, M.; Jacquemin, D.; Frison, G. Ground Electronic State of Peptide Cation Radicals: A Delocalized Unpaired Electron? J. Phys. Chem. Lett. 2011, 2, 1426−1431. (60) Standard density functionals artificially stabilize a situation in which a charge is delocalized over two molecules rather than one molecule being positively charged and the other neutral. (61) Zhang, Y.; Yang, W. A Challenge for Density Functionals: SelfInteraction Error Increases for Systems with a Noninteger Number of Electrons. J. Chem. Phys. 1998, 109, 2604−2608. (62) Mori-Sánchez, P.; Cohen, A. J.; Yang, W. Many-Electron SelfInteraction Error in Approximate Density Functionals. J. Chem. Phys. 2006, 125, 201102. (63) Ruzsinszky, A.; Perdew, J. P.; Csonka, G. I.; Vydrov, O. A.; Scuseria, G. E. Spurious Fractional Charge on Dissociated Atoms: Pervasive and Resilient Self-Interaction Error of Common Density Functionals. J. Chem. Phys. 2006, 125, 194112. (64) Ruzsinsky, A.; Perdew, J. P.; Csonka, G. I.; Vydrov, O. A.; Scuseria, G. E. Density Functionals That Are One- and Two- Are Not Always Many-Electron Self-Interaction-Free, As Shown for H+2 , He+2 , LiH+, and Ne+2 . J. Chem. Phys. 2007, 126, 104102. (65) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; et al. Gaussian 09, revision C.01; Gaussian, Inc.: Wallingford, CT, 2009. (66) Information concerning the distribution of conformers into different hydrogen-bonding motifs can be found in the Supporting Information. (67) Some proton transfers occasionally result in slightly less stable intermediate species, such that for certain conformers the proton will rest in the middle of the c-fragment rather than completely collapse the zwitterion. In such a case, the fragment is partly stabilized by the Coulombic attraction between the proton and the anionic amide.

(68) Chakraborty, T.; Holm, A. I. S.; Hvelplund, P.; Nielsen, S. B.; Poully, J.-C.; Worm, E. S.; Williams, E. R. On the Survival of Peptide Cations after Electron Capture: Role of Internal Hydrogen Bonding and Microsolvation. J. Am. Soc. Mass Spectrom. 2006, 17, 1675−1680. (69) Skurski, P.; Sobczyk, M.; Jakowski, J.; Simons, J. Possible Mechanisms for Protecting N−Cα Bonds in Helical Peptides from Electron-Capture (or Transfer) Dissociation. Int. J. Mass Spectrom. 2007, 265, 197−212. (70) Tsybin, Y. O.; He, H.; Emmett, M. R.; Hendrickson, C. L.; Marshall, A. G. Ion Activation in Electron Capture Dissociation To Distinguish between N-Terminal and C-Terminal Product Ions. Anal. Chem. 2007, 79, 7596−7602. (71) Cases involving motif 3 are more complicated. The final location of the charge cannot be determined based solely upon the H-bonding network.

2637

dx.doi.org/10.1021/jp412123h | J. Phys. Chem. B 2014, 118, 2628−2637