High-Confidence Sequencing of Phosphopeptides by Electron

Nov 18, 2016 - Daiki Asakawa† and Issey Osaka‡. † National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 2, 1-1-1 ...
0 downloads 0 Views 4MB Size
Subscriber access provided by UNIV OF NEWCASTLE

Article

High-Confidence Sequencing of Phosphopeptides by Electron Transfer Dissociation Mass Spectrometry using Dinuclear Zinc(II) Complex Daiki Asakawa, and Issey Osaka Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b03645 • Publication Date (Web): 18 Nov 2016 Downloaded from http://pubs.acs.org on December 2, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

High-Confidence

Sequencing

of

Phosphopeptides

by

Electron

Transfer

Dissociation Mass Spectrometry using Dinuclear Zinc(II) Complex

Daiki Asakawa1* and Issey Osaka2

1. National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 2, 1-1-1 Umezono, Tsukuba, Ibaraki, Japan 2. Center for Nano Materials and Technology, Japan Advanced Institute of Science and Technology, 1-1 Asahidai, Nomi, Ishikawa, Japan

Correspondence to: Daiki Asakawa National Institute of Advanced Industrial Science and Technology (AIST) Tsukuba Central 2, Umezono 1-1-1, Tsukuba, Ibaraki, 305-8568, Japan TEL: +81-29-861-0586, E-mail: [email protected]

ACS Paragon Plus Environment 1

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 39

ABSTRACT Phosphorylation is the most abundant protein modification, and tandem mass spectrometry with electron transfer dissociation (ETD) has proven to be a promising method for phosphoproteomic applications owing to its ability to determine phosphorylation sites on proteins. However, low precursor charge states hinder the ability to obtain useful information through peptide sequencing by ETD, and the presence of acidic phosphate groups contributes to a low charge state of peptide ions. In the present report, we used a dinuclear zinc complex, (Zn2L)3+ (L = alkoxide form of 1,3-bis[bis(pyridin-2-ylmethyl)amino]propan-2-ol) for electrospray ionization (ESI), followed by ETD-MS2 analysis. Since (Zn2L)3+ selectively bound to phosphopeptide with addition of a positive charge per phosphate group, the use of (Zn2L)3+ for ESI improved the ionization yield of phosphopeptides in phosphoprotein digest. Additionally, an increase in the charge state of phosphopeptides were observed by addition of (Zn2L)3+, facilitating phosphopeptide sequencing by ETD-MS2. Since the binding between (Zn2L)3+ and the phosphate group was retained during the ETD process, a comparison between the ETD mass spectra obtained using two dinuclear zinc complex derivatives containing different zinc isotopes, namely (64Zn2L)3+ and (68Zn2L)3+, provided information about the number of phosphate groups in each fragment ion, allowing the phosphorylation site to be unambiguously determined. The details of the fragmentation processes of the (Zn2L)3+-phosphopeptide complex were investigated using a density functional theory calculation. As in the case of protonated peptides, peptide backbone dissociation in the (Zn2L)3+-phosphopeptide complex proceeded through an aminoketyl radical intermediate.

Keywords: charge state, phosphorylation site, zinc-isotope, density functional theory

ACS Paragon Plus Environment 2

Page 3 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

INTRODUCTION The reversible phosphorylation of serine (Ser), threonine (Thr), and tyrosine (Tyr) side chain residues is the most abundant regulatory covalent protein modification.1 Protein phosphorylation and dephosphorylation are involved in many biological processes including signal transduction, cell division, gene expression, cytoskeletal regulation, and metabolic maintenance.1, 2 Electrospray ionization (ESI)-based mass spectrometry has been widely used as an analytical method for the characterization of protein phosphorylation. Typically, amino acid sequencing and site determination of protein phosphorylation is performed by tandem mass spectrometry (MS/MS) with collision-induced dissociation (CID).3,

4

Since the phosphate ester bond is labile, CID of

phosphopeptides results in the dominant loss of the phosphoric group (80 and/or 98 Da) from the precursor ions, which can be used as a specific marker for phosphorylated peptide identification. However, the loss of phosphoric acid(s) through CID is unfavorable for determining the location of phosphorylated sites. Fragmentation techniques involving the electron association of multiple-charged peptides, such as electron capture dissociation (ECD)5 and electron transfer dissociation (ETD)6, have been used as alternative methods to CID.7-9 Regarding the general mechanism of ECD/ETD, electron attachment/transfer occurs competitively at positively charged sites (Cornell model)10 and π* antibonding orbitals of peptide bonds (Utah–Washington model)11, 12 in multiply protonated peptides, and induces specific cleavage at N–Cα bonds on the peptide backbone. One of the main advantages of ECD/ETD is that it gives rise to fragment ions due to N–Cα bond cleavage without abundant fragmentation, allowing the determination of phosphorylation site location.8, 13, 14 Concerning the instrumentation for ECD and ETD experiments, ECD and ETD experiments are usually performed using Fourier transform-ion cyclotron resonance (FT-ICR) and quadrupole ion traps (QIT) instruments, respectively. FT-ICR mass analysis provides high-resolution and mass accuracy measurements. In contrast, QIT is widely used not only for mass analysis, but also as an intermediate storage chamber for hybrid MS instruments, such as quadrupole time-of-flight, linear

ACS Paragon Plus Environment 3

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 39

ion trap (LTQ)-FT-ICR, and LTQ-orbitrap. Thus, an advantage of ETD is that it can be performed by adapting such hybrid MS instruments, and it is widely used for proteomic applications. In terms of proteomic applications, employing ions with a higher charge state as the precursors for ETD dramatically increases the yield of radical reactive species, thereby improving the sequence coverage.15-17 Notably, the charge state of peptides is dependent on the peptide sequence, and especially on the number of basic residues.18 For ESI of tryptic peptides, the N-terminal amino group and the C-terminal basic residue are protonated, often yielding doubly protonated species. However, double protonation is not enough for the ion/electron reaction of ETD to yield better sequence coverage. Moreover, the presence of phosphate group(s) contributes to decreases in both the ion yield and the number of charge states in positive-ion ESI.18 Although ETD does not induce a significant loss of phosphate groups from phosphopeptides, it has not yet become the method of choice for large-scale phosphopeptide analysis, owing to the difficulty of producing tryptic phosphopeptides having more than three positive charges by ESI. Consequently, a method for increasing the charge state of precursor ions is desired to improve the quality of ETD mass spectra. The fixed-charge derivatization of the thiols in cysteine residues19, 20 or carboxyl groups21, 22 has been demonstrated to increase the charge state of peptides. As in the case of derivatization, we previously reported that the complexation of peptides with metal ions increases the charge state of peptides.23,

24

These methods render the ETD spectra more

informative than those obtained from protonated molecules with lower charge states. The results of our previous studies indicated that the number of excess protons in the precursor peptide does not influence the fragment ion yield, and the charge number of the precursor would be an important factor for obtaining better ETD mass spectra. To elucidate whether metal ion-protein complexation proceeded via the Cornell model or the Utah–Washington model during ETD, we used Zn2+-polyhistidine complexes in the absence of remote protons as a model system.25 According to the density functional theory (DFT) calculation, electron transfer to the Zn2+-polyhistidine complex from a fluoranthene radical anion produced a zwitterionic zinc-peptide radical, which immediately

ACS Paragon Plus Environment 4

Page 5 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

underwent N–Cα bond cleavage. As in the case of Zn2+-polyhistidine complexes, metal-peptide complexes have been found to give a c′/z• fragment pair upon ETD for most metal cations.23, 24, 26-31 Thus, it is suggested that the process described by the Utah–Washington model is dominant in the ETD of metal-peptide complexes and peptides using fixed-charge derivatization. Since ETD is mainly initiated by electron attachment to the π* antibonding orbital of the peptide bond, the most critical requirement for ETD analysis is to use relatively high charge state precursor ions, which provide the most informative fragmentation spectra. The targeted chemical derivatization of the phosphate groups in phosphopeptides is necessary to promote the wider utility of

ETD

for

large-scale

phosphoproteomics.

Dinuclear

metal

complexes

1,3-bis[bis(pyridin-2-ylmethyl)amino]propan-2-ol

with or

(2,6-bis[(N,N′-bis[2-picolyl]amino)methyl]-4-tertbutylphenol) have been reported to selectively bind to the phosphate groups of phosphopeptides and add a positive charge per phosphate group. Several dinuclear metal complexes with various metal cations, such as Co2+, Cu2+, Zn2+, Ga3+, and In3+ have been used as the phosphate capture molecules. With regards to the ETD of metal-peptide complexes, the fragmentation process is dependent on the properties of the metal cations in the complex. As described above, Zn2+-aided ETD promotes the formation of c′/z• fragment pairs. In contrast, transition metal cations with a partially filled d orbital shell, such as Co2+, Ni2+, and Cu2+, are reduced by ETD and the fragmentation is mainly induced by the substantial excitation attributable to the reduction energy of the metal cation in the complex.28, 32 The results suggest that Zn2+-containing complexes are suitable for peptide sequencing by ETD. Accordingly, we employed (Zn2L)3+ (L = deprotonated form of 1,3-bis[bis(pyridin-2-ylmethyl)amino]propan-2-ol) as a charge carrier for the ETD analysis of phosphopeptides, in order to increase the charge state of the phosphopeptides. The use of (Zn2L)3+ in the ESI processing of phosphopeptides increased the charge state of the resulting peptide ions, facilitating the production of c′/z• fragments by ETD. The details of the ETD fragmentation pathway of the (Zn2L)3+-phosphopeptide complex were investigated by a DFT calculation.

ACS Paragon Plus Environment 5

Analytical Chemistry

Page 6 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment 6

Page 7 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

EXPERIMENTS Materials Synthetic phosphopeptides designed to mimic the fragments produced by the tryptic digestion of yeast enolases. T18p (NVPLpYK), T19p (HLADLpSK), T43p (VNQIGpTLSESIK), T43pp (VNQIGTLpSEpSIK), were purchased from CS Bio Co., Ltd. (Shanghai, China). The α-casein and trypsin were purchased from Wako Pure Chemical Industries, Inc. (Tokyo, Japan) and Promega (WI, USA), respectively. The dinuclear zinc complex, (Zn2L)3+•CH3COO–•2ClO4–•H2O (L = alkoxide form of 1,3-bis[bis(pyridin-2-ylmethyl)amino]propan-2-ol, C27H29N6O) was purchased from Wako Pure Chemical Industries, Inc. (Tokyo, Japan). Dinuclear zinc complex derivatives containing

two

different

zinc

isotopes,

namely

(64Zn2L)3+•CH3COO–•2ClO4–•H2O

and

(68Zn2L)3+•CH3COO–•2ClO4–•H2O, were used. All reagents were used without further purification. All the solvents used were of high performance liquid chromatography grade, except for water, which was purified by a Milli-Q® purification system (Millipore; Billerica, MA, USA).

Table 1. Monoisotopic Mass (Mm), Sequence, and Composition of Analyte Phosphopeptides Used Analyte

Mm

Sequence

Composition

T18p

812.39

NVPLpYK

C35H57O12N8P

T19p

862.40

HLADLpSK

C34H59O14N10P

T43p

1367.68

VNQIGpTLSESIK

C55H98O23N15P

T43pp

1447.64

VNQIGTLpSEpSIK

C55H99O26N15P2

Mass spectrometry The analyte peptides were dissolved in water/acetonitrile (1/1, v/v) at a concentration of 10 µM. To produce the (Zn2L)3+-phosphopeptide complex, (Zn2L)3+ was added to the peptide solution at

ACS Paragon Plus Environment 7

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 39

a range of concentrations from 5 to 100 µM. The α-casein was digested with trypsin. The digestion was performed overnight at an enzyme to a protein ratio of 1: 250 in a 25 mM ammonium bicarbonate aqueous solution at 37 °C. The tryptic digest was diluted in water/acetonitrile (1/1, v/v) at a concentration of 10 µM. For the experiment on ETD of (Zn2L)3+-phosphopeptide complex, (Zn2L)3+ was added to the tryptic digest solution at 40 µM. Mass spectra were acquired using a 9.4 T FT-ICR mass spectrometer (SolariX FT, Bruker, Karlsruhe, Germany). The analyte solution was directly infused into the mass spectrometer using ESI as the ion source. The ion accumulation time, ion cooling time, and time-of-flight values were set to 0.5 s, 20 ms, and 7 ms, respectively. For ETD-MS2 experiments, precursor ions were mass-selected in the quadrupole filter and then reacted with fluoranthene radical anion. The times for reagent accumulation and the ion/ion reaction were set to 150 ms and 25 ms, respectively. Total ESI-MS and subsequent ETD-MS2 spectra were obtained by the accumulation of 20 and 50 single mass spectrum, respectively, except for ETD-MS2 analysis of [2(64Zn2L)-(T43pp–4H)+2H]4+ (Figure 4b) and triply-charged (64Zn2L)3+-phosphopeptide 106-119 from α-casein (Figure 5d). Figure 4b and 5d were obtained by the accumulation of 200 single mass spectrum.

Computational details All electron structure calculations were performed with the Gaussian 09 software program.33 The geometries for zinc-trihistidine complexes were optimized with DFT calculations using the M06-2X34 hybrid functional and double-zeta valence polarized basis sets that correspond to LanL-2DZ for the Zn atom and 6-31G(d) for the C, H, O, N, and P atoms. To establish the energetics for

fragmentation,

transition

state

geometries

were

also

optimized

at

the

M06-2X/LanL-2DZ/6-31G(d) level and investigated by examining their vibrational frequency analysis, showing one imaginary frequency.35 The relationship between the transition state and the reactants, as well as the intermediates, was checked by an intrinsic reaction coordinate analysis36 starting from the transition state conformation. The M06-2X/LanL-2DZ/6-31G(d)-optimized

ACS Paragon Plus Environment 8

Page 9 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

geometries were then reoptimized with the M06-2X/LanL-2DZ (Zn)/6-31+G(d,p) (C, H, O, N, and P). Single-point energies of local energy minima and transition state geometries were calculated using M06-2X functions with the 6-31++G(2d,p) basis set. Excited electronic states were calculated using a time-dependent DFT (TD-DFT) method37 with the M06-2X functions and the 6-31++G(2d,p) basis set.

Notation In the present study, Zubarev’s unambiguous notation was adopted for peptide fragment ions.38 According to this notation, homolytic N–Cα bond cleavage yields the radical c• and z• fragments, and the addition of a hydrogen atom to the c• or z• fragments produces a c' or z' fragment, respectively. The abstraction of a hydrogen atom from the c• or z• fragments produces a c or z fragment, respectively.

RESULTS AND DISCUSSION ETD mass spectra of protonated phosphopeptides First, phosphopeptides were analyzed using ESI followed by ETD-MS2 without the addition of (Zn2L)3+. The ESI mass spectra of the phosphopeptides are shown in the inset of Figure 1. Singly and doubly protonated species of phosphopeptides were observed in the ESI mass spectra, since protonation preferentially occurs at the N-terminal amino group and/or the C-terminal basic residue. Notably, triply protonated molecules, [M+3H]3+ were not observed for all phosphopeptides. Subsequently, doubly protonated phosphopeptides were used as the precursor for ETD-MS2 experiments, which produced only several fragment ions due to N–Cα bond cleavage (Figure 1). As shown in Figure 1, ETD performed with the doubly protonated precursor did not provide enough useful information for the sequencing of phosphopeptides, owing to low fragmentation efficiency under these conditions. Therefore, a method of increasing the charge state of phosphopeptides is desired for ETD.

ACS Paragon Plus Environment 9

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 39

Figure 1. ETD-MS2 spectra of doubly protonated phosphopeptides, [T18p+2H]2+, [T19+2H]2+, [T43p+2H]2+, and [T43pp+2H]2+. Black and white squares indicate the precursor ions and charge-reduced products, respectively. Inset panel: ESI-MS spectra of phosphopeptides.

Charge enhancement effect using dinuclear zinc (II) complex Herein, we used the dinuclear zinc complex, (Zn2L)3+, as a cationization reagent to increase the charge state of phosphopeptides. To avoid complications of the zinc-isotope effect, the single-isotope derivative, (64Zn2L)3+, was used for ESI-MS followed by ETD-MS2 experiments. Figure 2a shows the ESI mass spectra for a mixed sample of 10 µM mono-phosphorylated peptide, T43p, and (64Zn2L)3+ with mixing ratios, [T43p]/[(64Zn2L)3+] = 0.5, 2, and 4. The results indicated

ACS Paragon Plus Environment 10

Page 11 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

that the (Zn2L)3+-aided method produced an intense signal for (Zn2L)3+-adducted doubly charged T43p, [(64Zn2L)-(T43p–2H)+H]2+, instead of the doubly protonated T43p, due to the strong interaction between (Zn2L)3+ and the phosphate group in the T43p. Moreover, this method produced a triply charged complex, [(64Zn2L)-(T43p–2H)+2H]3+, in which positive charges are located at the N-terminal amino group, C-terminal basic residue, and (Zn2L)3+-adducted phosphate group. Figure 2b displays the relationship between the intensity ratio of the (64Zn2L)3+-T43p complex (1Zn2L) to free T43p (0Zn2L) and the concentration of

64

Zn2L3+ in the range of 0.5–60 µM. The values of the

signal ratio, 1Zn2L/0Zn2L, were calculated from the sum of the intensities of all species (H+, Na+, and K+ adducts) for the +1 and +2 charge states of free T43p and the +2 and +3 charge states of T43p bound with (64Zn2L)3+. The signal ratio, 1Zn2L/0Zn2L, increased with an increasing concentration of (64Zn2L)3+. At 40 µM (64Zn2L)3+, T43p was mostly bound to a single (64Zn2L)3+ ion, whereas T43p bound with two (64Zn2L)3+ ions was absent, indicating that (64Zn2L)3+ selectively binds to a phosphate group in T43p. Next, the enhancement of the charge state of a diphosphorylated peptide, T43pp, by the (Zn2L)3+-aided method, was investigated. The difference between T43p and T43pp is the addition of phosphate groups to the Ser residues at position 8 and 10, and the removal of a phosphate group from the Thr6 residue. Since T43pp has two phosphorylation residues, we examined the number of (Zn2L)3+ ions bound to T43pp. The ESI mass spectra for a mixed sample of 10 µM of T43pp and (64Zn2L)3+ with concentrations of metal ligand, [(64Zn2L)3+] = 5, 20, and 60 µM are shown in Figure 2c. Figure 2d displays the relationship between the relative intensity of 2(64Zn2L3+)-T43pp complex (2Zn2L), (64Zn2L)3+-T43pp complex (1Zn2L), and free T43pp (0Zn2L) and the concentration of (64Zn2L)3+ in the range of 0.5–100 µM. The relative intensity of 1Zn2L increased with the concentration of (64Zn2L)3+, reached a maximum at [(64Zn2L)3+] = 10 µM, and decreased when [(64Zn2L)3+] > 10 µM. Alternatively, when [(64Zn2L)3+] > 10 µM, most of the population of T43pp was bound to two (64Zn2L)3+ ions at 60–100 µM of (64Zn2L)3+. The results indicated that the number of (64Zn2L)3+ ions bound to the phosphopeptide corresponded to the number of phosphate groups in

ACS Paragon Plus Environment 11

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 39

the peptide. Regarding the charge state of the complex, the quadruple-charged complex, [2(64Zn2L)3+-(T43pp–4H)+2H]4+, was observed as the maximum charge state (Figure 2c). In conclusion, the maximum charge state observed by the (64Zn2L)3+-aid method corresponded to the sum of the number of basic sites and phosphorylation sites in the peptide.

Figure 2. ESI mass spectra of 10 µM of (a) T43p and (c) T43pp with

64

Zn2L3+. (b, d)

Relationship between the ESI-MS data based on the number of bound (64Zn2L)3+ for (b) T43p and (d) T43pp, and the concentration of (64Zn2L)3+.

ETD MS2 of Zn2L3+-phosphopeptide complexes As described in the previous section, the use of (Zn2L)3+ in the ESI of phosphopeptides increased the charge state of the phosphopeptide ions. Herein, we investigated the applicability of (Zn2L)3+-aided ETD-MS2 to the sequencing of phosphopeptides. The ESI mass spectra of 10 µM T18p or T19p with 40 µM (64Zn2L)3+ are shown in the left inset of Figure 3. As in the case of T43p,

ACS Paragon Plus Environment 12

Page 13 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

the (64Zn2L)3+-aided method produced triply charged complexes, [(64Zn2L)-(T18p–2H)+2H]3+ and [(64Zn2L)-(T19p–2H)+2H]3+, which were used as precursor ions for ETD-MS2 measurements. As in the case of the ETD of multi-protonated peptides, the ETD of [(64Zn2L)-(T18p–2H)+2H]3+ (Figure 3a) and [(64Zn2L)-(T19p–2H)+2H]3+ (Figure 3b) induced N–Cα bond cleavage, leading to protonation and (64Zn2L)3+-adducted c′ and z• fragments. The N–Cα bond cleavage initiated by electron association to the amide π* orbital on the peptide backbone and the potential fragmentation pathway for the (Zn2L)3+-phosphopeptide complex are shown in Scheme 1a. Regarding the ETD-MS2 of [(64Zn2L)-(T18p–2H)+2H]3+, c′3 and c′4 were observed as protonated forms, whereas their counterparts, z2• and z3• contained (64Zn2L)3+. In addition, (64Zn2L)3+-adducted c′5 was observed. Since (64Zn2L)3+ was selectively bound to the phosphate group, these results indicated that the phosphate group was located at the Tyr5 residue. As in the case of the ETD-MS2 of [(64Zn2L)-(T18p–2H)+2H]3+, the phosphorylation site in T19p was determined to be Ser5 by ETD-MS2, resulting in the production of [(64Zn2L)-(T19p–2H)+2H]3+. The ETD mass spectra of triply charged (64Zn2L)3+-phosphopeptide complexes, [(64Zn2L)-(T18p–2H)+2H]3+ (Figure 3a) and [(64Zn2L)-(T19p–2H)+2H]3+ (Figure 3b) provided better sequence information than that of the doubly protonated peptides (Figures 1a and 1b). For comparison, the doubly charged complexes, [(64Zn2L)-(T18p–2H)+H]2+ (Figure 3a) and [(64Zn2L)-(T19p–2H)+H]2+, were also used as the precursor ions for ETD-MS2 experiments (right inset of Figure 3). However, the ETD of doubly charged complexes did not provide any sequence information, as in the case of doubly protonated precursor. The doubly-charged precursor with ETD gave singly-charged intermediate, which does not facilitated extensive fragmentation, probably due to lack of Coulomb repulsion. Therefore, ETD with doubly-charged precursor does not induced enough fragmentation to obtain the sequence information in many instances. As in the case of the N–Cα bond cleavage, ETD of the (Zn2L)3+-phosphopeptide complex produced fragment ions due to the 92 Da loss, which might be assigned as C6H7N. The loss of C6H7N was probably initiated by electron attachment to the pyridine ring in (64Zn2L)3+, which

ACS Paragon Plus Environment 13

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 39

underwent proton transfer to the pyridine ring from the protonation site in the phosphopeptide. Subsequently, the generated hydrogen-abundant pyridine radical led to the loss of C6H7N, as shown in Scheme 1. Notably, ETD of triply charged complexes yielded singly charged fragment ions due to the loss of two C6H7N, [(64Zn2L)-(phosphopeptide–2H)–2C6H7N]+, whereas the corresponding doubly charged fragment, [(64Zn2L]-(phosphopeptide–2H)–2C6H7N+H]2+ was absent. The results indicated that the transfer of an electron to the pyridine ring in the (Zn2L)3+-phosphopeptide complex led to a loss of C6H7N, and subsequent electron transfer to the fragment underwent further C6H7N loss, giving rise to [(64Zn2L)-(phosphopeptide–2H)–2C6H7N]+. The results indicated that C6H7N loss was mediated by the association of an electron to the complex. Additionally, the ETD of [(64Zn2L)-(T19p–2H)+2H]3+ yielded [(64ZnL)–C6H7N]+ and [(T19p–H)+64Zn]+. Those fragment ions were generated by the degradation of [(64Zn2L)-(T19p–2H)– C6H7N]2+. To verify the composition of the fragment ions, we also repeated the ETD-MS2 experiment with (68Zn2L)3+ (Figures 3c and 3d). As expected, the fragment ions obtained were [(68ZnL)–C6H7N]+ and [(T19p–H)+68Zn]+, which are 4 Da heavier than [(64ZnL)–C6H7N]+ and [(T19p–H)+64Zn]+. It should be noted that the comparison of the ETD mass spectra of T19p with different single-isotope derivatives, (64Zn2L)3+ and (68Zn2L)3+, showed a mass difference of 8 Da for the (Zn2L)3+-containing c′ and z• fragment ions (Figure 3d). In contrast, no shift was observed for protonated fragment ions (Figure 3d). Therefore, the ETD of phosphopeptides with both (64Zn2L)3+ and (68Zn2L)3+ used in conjunction facilitates the interpretation of ETD-MS2 spectra.

ACS Paragon Plus Environment 14

Page 15 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 3. (a, b) ETD-MS2 spectra of triply charged phosphopeptide-Zn2L complexes, (a) [(64Zn2L)-(T18p–2H)+2H]3+ and (b) [(64Zn2L)-(T19p–2H)+2H]3+. The c′ and z• fragments annotated with an asterisk (c′* and z•*) correspond to the Zn2L adduct on the c′ and z• fragments. Black and white squares indicate the precursor ions and charge-reduced products, respectively. Daggers and double daggers indicate the fragment ions produced due to C6H7N and 2C6H7N losses, respectively. Left inset: ESI mass spectra of phosphopeptides containing a 4-fold molar excess of (64Zn2L)3+ relative to the phosphopeptides. Right inset: ETD-MS2 spectra of doubly charged (Zn2L)3+-phosphopeptide complexes. (c, d) Enlarged ETD spectra of [(64Zn2L)-(T19p–2H)+2H]3+ (upper panel) and [(68Zn2L)-(T19p–2H)+2H]3+ (lower panel) for (c) c′3, [ZnL–92]+ and c′4, and (d) z2•*, [T19–H+Zn]+ and z3•*.

ACS Paragon Plus Environment 15

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 39

Scheme 1. Potential ETD fragmentation pathway of Zn2L-pSerLys dication, which is a substructure of (Zn2L)3+-T19p complex. Next, we analyzed T43p and T43pp by (Zn2L)3+-aided ETD. There are three possible phosphorylation sites in T43p and T43pp, Thr6, Ser8, and Ser10. Figure 4 shows the ETD mass spectra of [(64Zn2L)-(T43p–2H)+2H]3+ and [2(64Zn2L)-(T43pp–4H)+2H]4+. Although the ETD-MS2 of [T43p+2H]2+ and [T43pp+2H]2+ yielded only one c′ and two z• fragments (Figures 1c and 1d), (Zn2L)3+-aided ETD provided complete sequence information for both T43p and T43pp (Figure 4). As in the case of T18p and T19p (Figure 3), the sequence coverage of T43p and T43pp by ETD was enhanced by increasing the charge state of the precursor peptide ions. Moreover, the locations of the phosphorylation sites were clearly identified by the ETD-MS2 spectrum of [(64Zn2L)-(T43p– 2H)+2H]3+. As an example, (64Zn2L)3+-adducted c' fragments start from c′6, which contains a phosphorylated residue. In contrast, the counterpart of c′6, i.e., z•6, was observed as a protonated form

ACS Paragon Plus Environment 16

Page 17 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

and z•7–z•11 were observed as the (64Zn2L)3+-adducted form. Therefore, Thr6 was determined to the phosphorylation site of T43p. The ETD-MS2 of (Zn2L)3+-adducted diphosphopeptide, T43pp, was investigated. Although [64Zn2L-(T43p–2H)+2H]3+ produced both protonated and Zn2L adducted fragments (Figure 4a), ETD of [2(64Zn2L)-(T43pp–4H)+2H]4+ produced only Zn2L-adducted fragment ions (Figure 4b). In consequence,

ETD

of

[2(64Zn2L)-(T43pp–4H)+2H]4+

showed

only

fragments

containing

phosphorylation residue(s). Since the binding between (Zn2L)3+ and the phosphate group was retained during the ETD process, the fragments with one and two phosphate group(s) were observed as singly and doubly charged forms, respectively. To determine phosphorylation sites, we focused on (Zn2L)3+-adducted singly charged c′ and z• ions, which start from c′8 and z•3. In addition, c′10 and z•5 were detected as doubly charged forms, i.e., [2(Zn2L)-(c′10–4H)]2+ and [2(Zn2L)-(z•5–4H)]2+, indicating two phosphorylation residues. Therefore, (Zn2L)3+-aided ETD unambiguously revealed Ser8 and Ser10 as the sites of phosphorylation. In addition, the ETD-MS2 with the single zinc isotopes of the dinuclear zinc complex derivatives, (64Zn2L)3+ and (68Zn2L)3+, provided information about the number of phosphate groups in each fragment ion, as shown in the inset of Figure 4b.

ACS Paragon Plus Environment 17

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 39

Figure 4. ETD-MS2 of (a) [(64Zn2L)-(T43p–2H)+2H]3+ and (b) [2(64Zn2L)-(T43pp–4H)+2H]4+. The c′ and z• fragments annotated with asterisk(s) (c′*, c′**, z•*, and z•**) correspond to the (Zn2L)3+ adducts on the c′ and z• fragments. Black and white squares indicate the precursor ions and charge-reduced products, respectively. Daggers and double daggers indicate the fragment ions produced due to C6H7N and 2C6H7N losses, respectively. Inset panel of (b): Enlarged ETD spectra of [2(64Zn2L)-(T43pp–4H)+2H]4+ (upper panel) and [2(68Zn2L)-(T43pp– 4H)+2H]4+ (lower panel) for z4•*, [T19–H+Zn]+ and (z8•**)2+.

Analysis of phosphoprotein digest In order to confirm the applicability of (Zn2L)3+-aided ETD to sequencing of phosphopeptides in protein digest, we analyzed the tryptic digest of α-casein. The ESI mass spectra of the tryptic digest of α-casein with or without 40 µM (64Zn2L)3+, as shown in Figures 5a and 5b. Since protonation preferentially occurs at the N-terminal amino group and the C-terminal basic residue, the doubly protonated species were mainly observed in ESI mass spectra of α-casein digest.

ACS Paragon Plus Environment 18

Page 19 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Notably, phosphopeptide (residues 104-119, YKVPQLEIVPNpSAEER) was detected as doubly protonated form. Since the phosphopeptide 104-119 contained one missed cleavage site, Lys105– Val106, the phosphopeptide have three protonation sites, N-terminal amino group, Lys105 and Arg119. However, triply protonated phosphopeptide 104-119 was absent in Figure 5a. Next, we used doubly protonated phosphopeptide 104-109 as the precursor for ETD-MS2 experiments, which produced only c'12 and z•15 fragments (Figure 5c). As in the case of standard phosphopeptides, the doubly protonated precursor did not provide the information for the sequencing of phosphopeptide by ETD-MS2. Compared with Figure 5a, (64Zn2L)3+ effectively bound to phosphopeptide 104-119 and produced and triply charged [(64Zn2L)-(M–2H)+2H]3+, as demonstrated in Figure 5b. The result indicated that the addition of (64Zn2L)3+ could increase the charge state of phosphopeptides in tryptic digest. Additionally, the phosphopeptide 106-119 (VPQLEIVPNpSAEER) without missed cleavage were observed as triply charged complex (inset of Figure 5b). It is indicated that the ionization yield of

phosphopeptides

were

improved

by

the

addition

of

(Zn2L)3+.

A triply

charged

(64Zn2L)3+-phosphopeptide 106-119 complex was used as a precursor ion for ETD-MS2 experiment. As shown in Figure 5d, a triply charged complex produced six z• and one c' fragments, whereas the doubly protonated phosphopeptide 104-119 yielded only c'12 and z•15 fragments (Figure 5c). All the fragment ions obtained from (64Zn2L)3+-phosphopeptide 106-119 complex contained phosphorylated residue, ETD-MS2 with (Zn2L)3+ provide useful information for the location of phosphorylated residue in phosphopeptides. In conclusion, (Zn2L)3+-aided ETD for the phosphopeptide enhanced fragmentation yield by increasing the charge state of the peptide ions, and the use of single zinc-isotope derivatives of (Zn2L)3+ facilitated reading of the peptide sequence, including the location of the phosphorylated residue.

ACS Paragon Plus Environment 19

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 39

Figure 5. (a, b) ESI mass spectra of tryptic digests of α-casein (a) in absence and (b) presence of 40

µM

(64Zn2L)3+.

Inset

(64Zn2L)3+-phosphopeptide

panel complex.

of

(b):

Astrisk

Enlarged indicated

ESI-MS triply

spectrum charged

of

the

complex,

(64Zn2L)3+-phosphopeptide 106-119. (c, d) ETD-MS2 spectra of (c) doubly protonated

ACS Paragon Plus Environment 20

Page 21 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

phosphopeptide 104-119 and (d) triply-charged (64Zn2L)3+-phosphopeptide 106-119 complex. Black and white squares indicate the precursor ions and charge-reduced products, respectively. The c′ and z• fragments annotated with asterisk(s) (c′*, c′**, z•*, and z•**) correspond to the (64Zn2L)3+ adducts on the c′ and z• fragments. Daggers and double daggers indicate the fragment ions produced due to C6H7N and 2C6H7N losses, respectively.

ETD mechanism of the Zn2L3+-phosphopeptide complex as determined by DFT calculation The details of the ETD of (Zn2L)3+-adducted phosphopeptide were investigated by DFT calculation. Herein, we chose the phosphorylated dipeptide, pSK and pYK, which were substructure of T19p and T18p, respectively, as the model tryptic phosphopeptides. The potential fragmentation pathway of doubly charged (Zn2L)3+-pSK complex, [(Zn2L)-(pSK–2H)+H]2+ is shown in Scheme 1. To discuss the initial step of ETD in [(Zn2L)-(pSK–2H)+H]2+ and [(Zn2L)-(pYK–2H)+H]2+, we calculated the electron structures of the precursor ions. The stable conformation of the (Zn2L)3+-phosphopeptide complex was held together by (Zn2L)3+-phosphate binding. The model complexs, [(Zn2L)-(pSK–2H)+H]2+ and [(Zn2L)-(pYK–2H)+H]2+, had positive charges located at the side-chain of the Lys residue and the (Zn2L)3+-adducted phosphate group. Figure 6 shows the lowest-energy conformation of the doubly-charged complexes, in which the phosphate group interacts with an ammonium group at a Lys residue and C-terminal carboxyl group, since the energy minimization in gas phase usually occurs through intramolecular charge solvation. As discussed in the previous section, ETD of the (Zn2L)3+-phosphopeptide complexes mainly induced N–Cα bond cleavage and C6H7N loss, which originated from the different electronic states of the charge-reduced complexes formed by these different processes. Therefore, we first discussed the electronic states of the charge-reduced complex radicals, [(Zn2L)-(pSK–2H)+H]+• and [(Zn2L)-(pYK–2H)+H]+•. Figure 6a and 6b shows the molecular orbitals for the ground state and excited states of [(Zn2L)-(pSK– 2H)+H]+• and [(Zn2L)-(pYK–2H)+H]+•, respectively, and their corresponding vertical excitation energies. The ground state and excited states, X–C in Figure 6a and X, A and C in Figure 6b,

ACS Paragon Plus Environment 21

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 39

underwent C6H7N loss via a charge-reduced complex having a pyridine radical. In contrast, the fragments due to N–Cα bond cleavage would be formed from excited state configurations, D in Figure 6a and B in Figure 6b, via aminoketyl radical intermediate.

Figure 6. Lowest-energy conformations and electronic state diagram for vertical electron attachment of doubly charged complexes, (a) [(Zn2L)-(pSK–2H)+H]2+ and (b) [(Zn2L)-(pYK– 2H)+H]2+. The optimized conformations were obtained by M06-2X/LanL-2DZ/6-31+G(d,p) and the excitation energies (eV) were obtained from TD-M06-2X/6-31++G(2d,p) calculations.

Electron transfer to multiply charged complex induces conformational rearrangement owing to the lack of Coulomb repulsion in the cation-radicals, which is one of the dominant interactions in the multiply charged peptide. Since phosphorylation of Ser and Thr residues were often observed compared with Tyr phosphorylation in proteomics analysis, thus we used [(Zn2L)-(pSK–2H)+H]2+ as

ACS Paragon Plus Environment 22

Page 23 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

a model complex for the calculation of fragmentation processes herein. The electron transfer to the pyridine ring in the complex and the subsequent geometry relaxation generated the zwitterionic radical, R1 in Scheme 2, which have a pyridine radical anion. In contrast, the charge-reduced product formed from excited state D in Figure 6a gave an aminoketyl radical intermediate, R2, as shown in Scheme 3, by hydrogen radical transfer from an ammonium radical to a carbonyl oxygen on the peptide backbone. Intermediate complex R1 was more stable than R2 by 61 kJ/mol. Herein, the formation process of fragment ions due to C6H7N loss was investigated in detail (Scheme 2). The charge-reduced complex having a pyridine radical, R1, led to a hydrogen-abundant pyridine radical by proton transfer from a protonated ammonium group in the Lys residue. The proton transfer reaction proceeded through transition state TS1-1, which was 46 kJ/mol above R1, and the generated intermediate, IM1-1, was 24 kJ/mol above R1. Subsequently, IM1-1 underwent homolytic N–C bond cleavage in (Zn2L)3+, giving [(Zn2L)-(pSK–2H)–C6H7N]+• and C6H7N. The corresponding transition state, TS1-2, was 190 kJ/mol above R1. The fragmentation due to C6H7N loss resulted in the formation of an intermediate complex, IM1-2, which was 144 kJ/mol less stable than R1. Finally, the formation energies of [(Zn2L)-(pSK–2H)–C6H7N]+• and C6H7N were 229 kJ/mol above R1. The details of the dissociation processes leading to the production of fragment ions due to C6H7N loss are summarized in Scheme 2.

ACS Paragon Plus Environment 23

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 39

Scheme 2. Mechanism of C6H7N loss from [(Zn2L)-(pSK–2H)+H]2+ cation radical. The relative energies

(kJ/mol)

were

obtained

by

single-point

energy

calculations

with

the

M06-2X/6-31++G(2d,p) basis set on M06-2X/LanL-2DZ/6-31+G(d,p)-optimized geometries including M06-2X/LanL-2DZ/6-31G(d) zero-point vibrational energies.

Next, we investigated the N–Cα bond cleavage processes of the (Zn2L)3+-phosphopeptide complex. The details of the dissociation processes leading to [(Zn2L)-(c1–2H)]+ and z1• are summarized in Scheme 3. An aminoketyl radical intermediate, R2, was generated from the excited electric state, D, in Figure 6a. The N−Cα bond cleavage of R2 proceeded through transition state TS2, which was 49 kJ/mol above R2. The bond cleavage resulted in the formation of the complex, IM2, which was 14 kJ/mol more stable than R2. Subsequently, the complete dissociation energy of [(Zn2L)-(c1–2H)]+ and z1• was 68 kJ/mol above R2.

ACS Paragon Plus Environment 24

Page 25 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Scheme 3. Mechanism of N–Cα bond cleavage of (Zn2L)-pSK cation radical. The relative energies

(kJ/mol)

were

obtained

by

single-point

energy

calculations

with

the

M06-2X/6-31++G(2d,p) basis set on M06-2X/LanL-2DZ/6-31+G(d,p)-optimized geometries including M06-2X/LanL-2DZ/6-31G(d) zero-point vibrational energies.

In the present study, the transition state energies for fragment formation were calculated on the M06-2X/LanL-2DZ/6-31G(d) and then the obtained M06-2X/LanL-2DZ/6-31G(d)-optimized geometries were then reoptimized with the M06-2X/LanL-2DZ (Zn)/6-31+G(d,p). The results are summarized in Table. 1. Comparison of two different basis sets, 6-31G(d) and 6-31+G(d,p), indicated that lower energies were obtained by 6-31G(d). This suggested that the transition state energies might be underestimated by M06-2X/LanL-2DZ/6-31G(d). In addition, single-point energy calculations with the M06-2X/6-31++G(2d,p) basis set on M06-2X/LanL-2DZ/6-31G(d)- and M06-2X/LanL-2DZ/6-31+G(d,p)-optimized geometries were performed. The results of the single point energy calculations gave similar energies, suggesting that the M06-2X/LanL-2DZ/6-31G(d)and M06-2X/LanL-2DZ/6-31+G(d,p)-optimized geometries were similar for this model system.

Table 2. Energies of Optimized Geometries Relative to R1 Including Zero-Point Vibrational

ACS Paragon Plus Environment 25

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 39

Energies (kJ/mol) M06-2X/LanL-2DZ/6-31G(d)

M06-2X/LanL-2DZ/6-31+G(d,p)

487 (514)a

499 (514)b

TS1-1

25 (46)a

39 (46)b

IM1-1

12 (24)a

27 (24)b

TS1-2

184 (190)a

191 (190)b

IM1-2

142 (148)a

149 (144)b

R2

43 (65)a

49 (61)b

TS2

91 (111)a

97 (109)b

IM2

62 (50)a

38 (47)b

R2+

a

From

single-point

energy

calculations

with

the

M06-2X/6-31++G(2d,p)

on

M06-2X/LanL-2DZ/6-31G(d)-optimized geometries including M06-2X/LanL-2DZ/6-31G(d) zero-point vibrational energies.

b

From single-point energy calculations with the M06-2X/6-31++G(2d,p) on

M06-2X/LanL-2DZ/6-31G+(d,p)-optimized geometries including M06-2X/LanL-2DZ/6-31G(d) zero-point vibrational energies.

To understand the fragmentation behaviors of the (Zn2L)3+-phosphopeptide complex in ETD in detail, we focused on the unimolecular dissociation rate. According to a recent report, the electron transfer reaction between the peptide dication and fluoranthene anion generates a peptide radical cation with 285−327 kJ/mol of vibrational excitation.39 Since the transition state barriers of C6H7N loss and N−Cα bond cleavage were 190 kJ/mol and 49 kJ/mol, respectively, the time scale of the fragmentation was much shorter than that of ETD experimental conditions (0.1 s). Consequently, the reactants R1 and R2 would immediately give the fragment ions. The experimental and computational joint study suggested that the yield of the fragments due to C6H7N loss and N–Cα bond cleavage reflect the amounts of pyridine and aminoketyl radical intermediates produced by electron transfer

ACS Paragon Plus Environment 26

Page 27 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

from the fluoranthene anion to the (Zn2L)3+-peptide complex. In order to evaluate the fragmentation yield in experimental results, the N–Cα bond cleavage yield compared with C6H7N loss was calculated as the ratio of sum of c' and z• ions divided by the sum of the fragment ions due to C6H7N losses. According to the calculation, the yield of N–Cα bond cleavage in ETD of [(64Zn2L)-(T18p– 2H)+2H]3+ (Figure 3a), [(64Zn2L)-(T19p–2H)+2H]3+ (Figure 3b), [(64Zn2L)-(T43p–2H)+2H]3+ (Figure 4a) and [2(64Zn2L)-(T43pp–4H)+2H]4+ (Figure 4b) were 0.47, 0.34, 0.48 and 0.40, respectively. The results indicate that the pyridine radical intermediate was more abundant than the aminoketyl radical intermediate at the current experimental condition. To increase the yield of the fragment ions due to N–Cα bond cleavage, the development of a method allowing the selective production of an aminoketyl radical intermediate is desired. Since the chemical modification of the ligand would affect the electronic state diagram of the charge-reduced (Zn2L)3+-phosphopeptide cation radical, it will be necessary to find a better metal-ligand complex for ETD experiments that allows efficient N–Cα bond cleavage of phosphorylated peptides and proteins.

CONCLUSION The dinuclear zinc(II) complex, (Zn2L)3+ bound to the phosphate group in phosphopeptide, and the use of (Zn2L)3+ as a charge carrier selectively increased the charge state of phosphopeptides in the phosphoprotein digest upon ESI. Since the fragmentation efficiency and peptide sequence coverage by ETD is generally enhanced by increasing the charge state of the precursor peptide ions, (Zn2L)3+-aid ETD method is useful for the sequencing of phosphopeptide. In addition, the comparison of the ETD-MS2 spectra of two dinuclear zinc complexes containing different zinc isotopes, namely (64Zn2L)3+ and (68Zn2L)3+, facilitates the assignment of a consecutive series of c′ and z• fragments, which is helpful in identifying the peptide sequence and location of phosphorylated residues. The formation of the c′ and z• fragments was initiated by electron transfer to a protonation site or an amide π* orbital on the peptide backbone. In contrast, electron transfer to the pyridine group in (Zn2L)3+ can occur, leading to the loss of C6H7N. These processes competitively occur in

ACS Paragon Plus Environment 27

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 39

the ETD of (Zn2L)3+-phosphopeptide complexes. The chemical modification of (Zn2L)3+ would suppress the yield of fragment ions due to C6H7N loss and the preferential formation of c′ and z• fragments.

ACKNOWLEDGMENTS This work was supported by JSPS KAKENHI grant number JP26505016. The computations of molecular structures in this work were supported by Research Center for Computational Science, Okazaki and Center for Computational Sciences, University of Tsukuba. The experiments were partly supported by Nanotechnology Platform Program of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan.

REFERENCES 1.

Hunter, T., Cell 2000, 100, 113-127.

2. 3.

Cohen, P., Trends Biochem. Sci. 2000, 25, 596-601. Palumbo, A. M.; Smith, S. A.; Kalcic, C. L.; Dantus, M.; Stemmer, P. M.; Reid, G. E., Mass

Spectrom. Rev. 2011, 30, 600-625. 4.

Boersema, P. J.; Mohammed, S.; Heck, A. J., J. Mass Spectrom. 2009, 44, 861-878.

5. 6.

Zubarev, R. A.; Kelleher, N. L.; McLafferty, F. W., J. Am. Chem. Soc. 1998, 120, 3265-3266. Syka, J. E.; Coon, J. J.; Schroeder, M. J.; Shabanowitz, J.; Hunt, D. F., Proc. Natl. Acad. Sci.

U S A 2004, 101, 9528-9533. 7.

Kelleher, N. L., Anal. Chem. 2004, 76, 196A–203A.

8.

Coon, J. J., Anal. Chem. 2009, 81, 3208–3215.

9. 10.

Zhou, H.; Ning, Z.; Starr, A. E.; Abu-Farha, M.; Figeys, D., Anal. Chem. 2012, 84, 720-734. Zubarev, R. A.; Kruger, N. A.; Fridriksson, E. K.; Lewis, M. A.; Horn, D. M.; Carpenter, B.

K.; McLafferty, F. W., J. Am. Chem. Soc. 1999, 121, 2857-2862. 11. Sobczyk, M.; Anusiewicz, I.; Berdys-Kochanska, J.; Sawicka, A.; Skurski, P.; Simons, J., J. Phys. Chem. A 2005, 109, 250-258. 12. 13.

Syrstad, E. A.; Tureček, F., J. Am. Soc. Mass Spectrom. 2005, 16, 208-224. Shi, S. D.-H.; Hemling, M. E.; Carr, S. A.; Horn, D. M.; Lindh, I.; McLafferty, F. W., Anal.

Chem. 2001, 73, 19-22. 14.

Kweon, H. K.; Håkansson, K., Anal. Chem. 2006, 78, 1743-1749.

15.

Good, D. M.; Wirtala, M.; McAlister, G. C.; Coon, J. J., Mol. Cell Proteomics 2007, 6,

ACS Paragon Plus Environment 28

Page 29 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1942-1951. 16.

Frese, C. K.; Altelaar, A. F.; Hennrich, M. L.; Nolting, D.; Zeller, M.; Griep-Raming, J.;

Heck, A. J.; Mohammed, S., J. Proteome Res. 2011, 10, 2377-2388. 17.

Rožman, M.; Gaskell, S. J., Rapid Commun. Mass Spectrom. 2012, 26, 282-286.

18. 19.

Osaka, I.; Takayama, M., Rapid Commun. Mass Spectrom. 2014, 28, 2222-2226. Ueberheide, B. M.; Fenyo, D.; Alewood, P. F.; Chait, B. T., Proc. Natl. Acad. Sci. U S A

2009, 106, 6910-6915. 20.

Vasicek, L.; Brodbelt, J. S., Anal. Chem. 2009, 81, 7876-7884.

21. 22.

Ko, B. J.; Brodbelt, J. S., J. Am. Soc. Mass Spectrom. 2012, 23, 1991-2000. An, M.; Zou, X.; Wang, Q.; Zhao, X.; Wu, J.; Xu, L. M.; Shen, H. Y.; Xiao, X.; He, D.; Ji, J.,

Anal. Chem. 2013, 85, 4530-4537. 23. Asakawa, D.; Takeuchi, T.; Yamashita, A.; Wada, Y., J. Am. Soc. Mass Spectrom. 2014, 25, 1029-1039. 24.

Asakawa, D.; Wada, Y., J. Phys. Chem. B 2014, 118, 12318-12325.

25. Asakawa, D.; Yamashita, A.; Kawai, S.; Takeuchi, T.; Wada, Y., J. Phys. Chem. B 2016, 120, 891-901. 26.

Iavarone, A. T.; Paech, K.; Williams, E. R., Anal. Chem. 2004, 76, 2231-2238.

27.

Fung, Y. M.; Liu, H.; Chan, T. W., J. Am. Soc. Mass Spectrom. 2006, 17, 757-771.

28.

Liu, H.; Håkansson, K., J. Am. Soc. Mass Spectrom. 2006, 17, 1731-1741.

29.

Flick, T. G.; Donald, W. A.; Williams, E. R., J. Am. Soc. Mass Spectrom. 2013, 24, 193-201.

30. Bogdanov, B.; Zhao, X.; Robinson, D. B.; Ren, J., J. Am. Soc. Mass Spectrom. 2014, 25, 1202-1216. 31. 32.

Asakawa, D.; De Pauw, E., J. Am. Soc. Mass Spectrom. 2016, 27, 1165-1175. Chen, X.; Fung, Y. M.; Chan, W. Y.; Wong, P. S.; Yeung, H. S.; Chan, T. W., J. Am. Soc.

Mass Spectrom. 2011, 22, 2232-2245. 33. Gaussian 09; Revision D.01; 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, Inc., Wallingford CT 2010. 34. 35.

Zhao, Y.; Truhlar, D. G., Theor. Chem. Acc. 2008, 120, 215-241. Pepin, R.; Laszlo, K. J.; Peng, B.; Marek, A.; Bush, M. F.; Tureček, F., J. Phys. Chem. A

2014, 118, 308-324. 36.

Fukui, K., Acc. Chem. Res. 1981, 14, 363-368.

37.

Stratmann, R. E.; Scuseria, G. E.; Frisch, M. J., J. Chem. Phys. 1998, 109, 8218-8224.

38.

Zubarev, R. A., Mass Spectrom. Rev. 2003, 22, 57-77.

39.

Pepin, R.; Tureček, F., J. Phys. Chem. B 2015, 119, 2818-2826.

ACS Paragon Plus Environment 29

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. ETD-MS2 spectra of doubly protonated phosphopeptides, [T18p+2H]2+, [T19+2H]2+, [T43p+2H]2+, and [T43pp+2H]2+. Black and white squares indicate the precursor ions and charge-reduced products, respectively. Inset panel: ESI-MS spectra of phosphopeptides. 325x337mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 30 of 39

Page 31 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 2. ESI mass spectra of 10 µM of (a) T43p and (c) T43pp with 64Zn2L3+. (b, d) Relationship between the ESI-MS data based on the number of bound (64Zn2L)3+ for (b) T43p and (d) T43pp, and the concentration of (64Zn2L)3+. 303x214mm (150 x 150 DPI)

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. (a, b) ETD-MS2 spectra of triply charged phosphopeptide-Zn2L complexes, (a) [(64Zn2L)-(T18p– 2H)+2H]3+ and (b) [(64Zn2L)-(T19p–2H)+2H]3+. The c′ and z• fragments annotated with an asterisk (c′* and z•*) correspond to the Zn2L adduct on the c′ and z• fragments. Black and white squares indicate the precursor ions and charge-reduced products, respectively. Daggers and double daggers indicate the fragment ions produced due to C6H7N and 2C6H7N losses, respectively. Left inset: ESI mass spectra of phosphopeptides containing a 4-fold molar excess of (64Zn2L)3+ relative to the phosphopeptides. Right inset: ETD-MS2 spectra of doubly charged (Zn2L)3+-phosphopeptide complexes. (c, d) Enlarged ETD spectra of [(64Zn2L)-(T19p–2H)+2H]3+ (upper panel) and [(68Zn2L)-(T19p–2H)+2H]3+ (lower panel) for (c) c′3, [ZnL–92]+ and c′4, and (d) z2•*, [T19–H+Zn]+ and z3•*. 338x299mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 32 of 39

Page 33 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 4. ETD-MS2 of (a) [(64Zn2L)-(T43p–2H)+2H]3+ and (b) [2(64Zn2L)-(T43pp–4H)+2H]4+. The c′ and z• fragments annotated with asterisk(s) (c′*, c′**, z•*, and z•**) correspond to the (Zn2L)3+ adducts on the c′ and z• fragments. Black and white squares indicate the precursor ions and charge-reduced products, respectively. Daggers and double daggers indicate the fragment ions produced due to C6H7N and 2C6H7N losses, respectively. Inset panel of (b): Enlarged ETD spectra of [2(64Zn2L)-(T43pp–4H)+2H]4+ (upper panel) and [2(68Zn2L)-(T43pp–4H)+2H]4+ (lower panel) for z4•*, [T19–H+Zn]+ and (z8•**)2+. 342x217mm (150 x 150 DPI)

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. (a, b) ESI mass spectra of tryptic digests of α-casein (a) in absence and (b) presence of 40 µM (64Zn2L)3+. Inset panel of (b): Enlarged ESI-MS spectrum of the (64Zn2L)3+-phosphopeptide complex. Astrisk indicated triply charged complex, (64Zn2L)3+-phosphopeptide 106-119. (c, d) ETD-MS2 spectra of (c) doubly protonated phosphopeptide 104-119 and (d) triply-charged (64Zn2L)3+-phosphopeptide 106-119 complex. Black and white squares indicate the precursor ions and charge-reduced products, respectively. The c′ and z• fragments annotated with asterisk(s) (c′*, c′**, z•*, and z•**) correspond to the (64Zn2L)3+ adducts on the c′ and z• fragments. Daggers and double daggers indicate the fragment ions produced due to C6H7N and 2C6H7N losses, respectively. 346x411mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 34 of 39

Page 35 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 6. Lowest-energy conformations and electronic state diagram for vertical electron attachment of doubly charged complexes, (a) [(Zn2L)-(pSK–2H)+H]2+ and (b) [(Zn2L)-(pYK–2H)+H]2+. The optimized conformations were obtained by M06-2X/LanL-2DZ/6-31+G(d,p) and the excitation energies (eV) were obtained from TD-M06-2X/6-31++G(2d,p) calculations. 425x298mm (150 x 150 DPI)

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Scheme 1. Potential ETD fragmentation pathway of Zn2L-pSerLys dication, which is a substructure of (Zn2L)3+-T19p complex. 382x272mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 36 of 39

Page 37 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Scheme 2. Mechanism of C6H7N loss from [(Zn2L)-(pSK–2H)+H]2+ cation radical. The relative energies (kJ/mol) were obtained by single-point energy calculations with the M06-2X/6-31++G(2d,p) basis set on M06-2X/LanL-2DZ/6-31+G(d,p)-optimized geometries including M06-2X/LanL-2DZ/6-31G(d) zero-point vibrational energies. 346x228mm (150 x 150 DPI)

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Scheme 3. Mechanism of N–Cα bond cleavage of (Zn2L)-pSK cation radical. The relative energies (kJ/mol) were obtained by single-point energy calculations with the M06-2X/6-31++G(2d,p) basis set on M062X/LanL-2DZ/6-31+G(d,p)-optimized geometries including M06-2X/LanL-2DZ/6-31G(d) zero-point vibrational energies. 329x151mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 38 of 39

Page 39 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Analytical Chemistry

ACS Paragon Plus Environment