University of Groningen

The influence of peptide structure on fragmentation pathwaysBari, Sadia

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2010

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Bari, S. (2010). The influence of peptide structure on fragmentation pathways. s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

The publication may also be distributed here under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license.More information can be found on the University of Groningen website: https://www.rug.nl/library/open-access/self-archiving-pure/taverne-amendment.

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 26-05-2022

Chapter 6

Peptide fragmentation by keVion–induced dissociation

We have studied multiple ionization and dissociation of a trapped protonated peptide (leucineenkephalin) as induced by keV singly and doubly charged ions(H+, He+, 2+) to demonstratethe potential of keV ions as a future tool for peptide identification. In contrast to conventionalexcitation techniques, the fragmentation pattern exhibits very strong peaks due to loss of sidechains in addition to those due to backbone scission. The results can be understood on basisof the energy deposited into the peptide via electronic stopping. A pronounced dependenceof the fragmentation pattern on the electronic structure ofthe projectile ions can be attributedto different electron capture efficiencies from localized molecular orbitals.

published:Front cover articleSadia Bari, Ronnie Hoekstra and Thomas Schlatholter,Phys. Chem. Chem. Phys.12, 3376 (2010).

67

68 Peptide fragmentation by keV ion–induced dissociation

6.1 Introduction

A recent series of studies on keV ion–induced ionization of isolated polyatomic moleculesin the gas phase [1–4] aimed for a better insight into the molecular mechanisms underlyingbiological radiation damage. These studies have greatly advanced the understanding of theunderlying dissociation dynamics. The experiments have been limited to smaller moleculeswhich are stable with respect to thermal decomposition. With the advent of soft ioniza-tion techniques such as electrospray ionization (ESI [5]) and matrix assisted laser desorp-tion/ionization [6], it has become possible to bring much larger, intact molecules such aspeptides and proteins into the gas phase. Usually these molecules are in their protonatedform.

Dissociation of mass–selected peptide ions is a basic technique for peptide sequencingand protein identification by means of mass spectrometric techniques. To drive the disso-ciation processes, excitation by multiple collisions withinert gas atoms (collision induceddissociation, CID [7]) or surfaces (surface induced dissociation, SID [7]), by electron cap-ture (electron capture dissociation, ECD [8–10]) and by blackbody infrared radiation [11, 12]or laser light [13] has been employed. CID has been the most common method for peptideidentification purposes. When CID is performed at higher peptide energies, a single collisionmay be sufficient to induce dissociation. Liu and co–workershave for instance investigatedthe fragmentation of 50 keV protonated adenosine monophosphate colliding with neutral Naor Ne atoms [14].

The reversed direct approach however, in which e.g. keV atomic ions are collided witha target of biomolecular ions has not been pursued yet. Untilnow sufficient target densi-ties for such experiments could not be reached. The use of keVatomic ions would allowfor fs–interaction times and higher excitation energies. The fs–interaction between keV ionsand larger molecules or clusters is governed by (multiple) electron capture and electronicstopping. Electron capture leads to gentle multiple ionization of a target biomolecule anddepends strongly on the charge state and electronic structure of the projectile ion. The role ofthe charge state and electronic structure of the projectilespecies has been demonstrated forcollisions on nucleobases [15] and amino acids [16]. The term electronic stopping describesprojectile ion–induced electronic excitations of the target molecule. Observed excitation en-ergies after ion impact at similar energies as used here are about 10 eV for H+ on smallbiomolecules [17] and between 30 eV and 50 eV for H+ and He+ on C60 [18, 19]. Depo-sition of such relatively high energies on fs timescales is leading to a fragmentation regimefundamentally different from CID and SID, where the excitation energies are very close todissociation thresholds of 2–4 eV [20], and thus opens the door to a variety of new applica-tions.

In addition, a detailed knowledge of the molecular dynamicstriggered by the interactionof energetic ions with complex biomolecules is of great relevance for understanding protonand heavy ion tumor therapy on a molecular level and assessing the risk of human spaceexploration due to the exposure to ions from galactic cosmicrays and solar particle events.In both cases, maximum biological damage occurs when the originally fast ions are sloweddown to MeV energies and below, the so–called Bragg peak. Biological effects of ionizingradiation are known to be due to direct and indirect damage ofcellular DNA. Pioneeringstudies on biological radiation damage on the single–molecule level have dealt with ionization

6.2 Experiment 69

b1 b4

b3b2

y4y2y3

y1

tyrosine glycine

136

120

86

N-terminalimmonium

glycine phenylalanine leucine

c1

z4

a4

Figure 6.1: Structure of leucine enkephalin with an indication of its 5 constituent amino acids, nomen-clature of the y and b fragments as well as three side chain (immonium) fragments.

and fragmentation of isolated DNA building blocks [15, 21] and amino acids [16]. Thelatter are relevant since in the nuclei of most cells, DNA is wound around protein spools,the histones. The ion interaction with these proteins is biologically relevant since secondaryparticles formed during the radiation might in turn damage the neighboring DNA. To date,almost all experiments on radiation action on isolated biomolecules have been limited tosmaller molecules that can be brought into vacuum by evaporation.

Here we demonstrate the potential of keV ion–induced ionization and dissociation (KID)as a technique for peptide structure analysis and fragmentation studies. To overcome theproblem of an insufficiently dense molecular target, molecular ions are produced by electro-spray ionization and then accumulated and cooled in an RF–trap.

The cooled sample of trapped molecular ions is then crossed by a focused ion beam. Asbeams 5–40 keV H+, He+ and He2+ ions were used; while as target molecule, the neuro-transmitter peptide leucine enkephalin was used (leu–enk,amino acid sequence: Tyr-Gly-Gly-Phe-Leu, m = 555.62, see fig. 6.1). In the following the experimental setup is brieflydescribed. Thereafter the obtained fragmentation spectraare presented and discussed. Inparticular, the different influences of the electronic excitation and electron capture on thedissociation dynamics are addressed.

6.2 Experiment

All experiments were carried out at the ZernikeLEIF facility of the KVI (University ofGroningen, The Netherlands). Recently, we have developed anew apparatus (for a sketch,see fig. 6.2) in which a home–built electrospray ionization (ESI) source is interfaced with anelectron cyclotron resonance (ECR) ion source. A brief description of the experimental setup follows, which is described in more detail in chapter 3.

The singly protonated pentapeptide leucine enkephalin ions were generated in the in–house built ESI source from a∼30 µM methanol solution with 1% formic acid. The ionswere then guided by a collisionally focusing RF–only quadrupole and an RF–quadrupole

70 Peptide fragmentation by keV ion–induced dissociation

Figure 6.2: Experimental setup.

mass analyzer through the end cap into a quadrupole ion trap.

The base pressure inside the trap chamber was 1×10−9 mbar. For collisional cooling ofthe peptide ions, a He–buffer gas pulse was applied. The estimated pressure inside the trapincreased to 1×10−3 mbar, and the trap was typically loaded with protonated peptides overa period of 300 ms. The ESI beam was then blocked by means of a 100 V skimmer bias,and at the same time the solenoid valve, controlling the He buffer gas flow, was closed. Adelay of about 500 ms allowed the pressure in the trap to decrease to about 1×10−6 mbar,necessary to keep charge–changing collisions by the keV ions below 1% before reaching thetrap center. A∼100 nA beam of keV H+, He+ or He2+ projectile ions extracted from theECR ion source intersected the Paul trap through the ring electrode.

The protonated peptides were then exposed for about 100 ms tothe ion beam. The con-ditions were chosen such that a total of about 10% of the trapped protonated peptides weredissociated by collisions with projectile ions. Directly after the projectile ion pulse, a secondHe–buffer gas pulse was applied to the trap, to cool energetic dissociation products.

Trapped protonated peptides and their cationic dissociation products were then extractedinto a linear time–of–flight (TOF) mass spectrometer (M/∆M=200) by applying a bias voltage(Ubias=± 200 V, duration: 5µs) to the RF–trap end caps. The ions were detected by asilhouette–type micro–channel–plate detector with the front plate biased to -5 kV and theanode kept at ground potential. The detector signal was recorded by a 1–GHz digitizer.

Despite the low background pressure and a liquid nitrogen cooled cryo–trap close to theRF–trap, contamination of the buffer gas or neutral molecules from the ESI source may con-tribute to the mass spectra. To extract the mass spectrum dueto peptide fragmentation only,the data acquisition was divided into successive cycles of three mass scans. In each cycle, firstthe TOF spectrum resulting from keV irradiation of trapped protonated peptides and neutralresidual gas was recorded (inclusive scan). To obtain the net effect of keV ion irradiationupon the trapped protonated peptides, in a second scan the ECR source was switched off anda TOF spectrum of the initial trap content only was recorded.For the third scan, the ESIsource was switched off and the TOF spectrum resulting from the ion–induced ionization ofresidual gas molecules was recorded. The latter two spectrawere then subtracted from theinclusive scan. A three–scan cycle took about 3 s. To obtain the final mass spectra a seriesof 2000–6000 cycles was accumulated, which assures that long term fluctuations of peptide-and projectile–ion current were averaged out.

6.3 Results 71

6.3 Results

Fragment cation spectra obtained for interactions of He2+, H+ and He+ with leu–enk are dis-played in fig. 6.3, 6.4, 6.5 respectively. In these figures we compare the dissociation patternsobtained for different projectile ion kinetic energies ranging from 5 to 40 keV. The spectraare normalized to the intensity of the m/z=107 amu peak whichis most often the strongestpeak in the spectra. Masses smaller than 50 are not trapped under the present experimentalconditions. In general, the spectra are dominated by fragments with m/z 80–140 amu withthe peaks 86, 120 and 136 being the intact immonium groups of leu, phe and tyr (see fig. 6.1).The peaks at 107 and 91 amu are common fragments of these groups. Fragment ions withmasses exceeding 350 are not observed. The peaks at masses smaller than 86, e.g. the groupbetween 75 and 80, are due to common amino acid fragments (strong peaks in this regionafter electron impact ionization of amino acids are 74 (gly,leu, phe) and 77 (tyr) [22]).

The ion–induced fragmentation spectra are fundamentally different from those observedin CID [23], SID [20] and also e.g. IR multiphoton absorption[24] in which fragmentswith m/z exceeding 350 dominate the spectra and side chain fragments are weak or veryweak. In these cases, the backbone scission leading to the heavy fragments is described inthe framework of the mobile proton model [25]. Upon an increase of vibrational excitationenergy, the proton attached to the peptide becomes mobile and samples various protonationsites within the molecule. Attachment to a backbone N atom can for instance weaken thebackbone such, that a b/y fragment is formed directly.

In the KID spectra peaks belonging to masses larger than 170 amu are of much lower in-tensity. For He+ projectiles, the peaks in this range are barely observable (fig. 6.5). For He2+

and H+ (fig. 6.3 and 6.4) the high m/z peaks can be easily assigned to fragments resultingfrom single or double scission of the peptide backbone. The peak at 278/279 amu is probablydue to the b3 and y2 fragments (for the notation, see fig. 6.1). A small contribution from thedoubly charged parent peptide cannot be excluded. At 221 amu, the b2 fragment is clearlyvisible. An internal gly–phe fragment that is due to loss of b2 and y1 from leu–enk showsup at 205 amu. The fragment with m/z=295 is likely to originate from an a4 fragment (seefig. 6.1) as will be discussed in the context of the He2+ spectra. Note that the internal gly–phefragment could alternatively also be formed in a sequence involving the cyclic intermediate,as will also be discussed later.

6.3.1 He2+

For He2+ projectiles, reducing the energy gives rise to the most pronounced change in relativepeak intensities. Formation of the intact immonium fragment ions tyr (m/z =137) and phe(m/z=120) is getting suppressed with decreasing projectile energy, as is the yield of the sidechain fragment tyr/phe (m/z=91). At 10 keV, the dominant tyrside chain fragment at m/z=107dominates over all other peaks in the spectrum by at least a factor of 2. The decrease of the pheimmonium ions may be rationalized by the fact that its formation requires a double backbonescission which becomes energetically unfavorable at lowerprojectile kinetic energy. Phe sidechain ion formation is also reduced.

Interestingly, the intensities of the internal fragments at m/z=205 (gly–phe) and 295 (phe–tyr–gly minus CO) both increase. The formation of these fragments also requires a double

72 Peptide fragmentation by keV ion–induced dissociation

50 100 150 200 250 300 3500.0

0.4

0.8

(d)

10 keV He2+

m/z (amu)

0.0

0.4

0.8

(c)

20 keV He2+

inte

nsity

0.0

0.4

0.8

(b)

30 keV He2+

0.0

0.4

0.8

295

278/279 b3/y2

221 b2

205gly-phe int

136(tyr)

120(phe)

107(tyr)

91(phe) (tyr)

86(leu)

(a)

40 keV He2+

269

Figure 6.3: Mass spectra of the product cations for He2+ with leu–enk at different projectile energies:(a) 40 keV, (b) 30 keV, (c) 20 keV and (d) 10 keV. Arrows indicate the direction of change of (b)–(d)with decreasing energy.

6.3 Results 73

backbone scission. Their increase at lower projectile kinetic energies is non–intuitive. Bothfragments originate from an intermediate a4 or b4 fragment. These fragments are most abun-dant when established dissociation techniques are employed but not observed here. Accord-ing to Laskin [20] and DFT calculations by Jalkanen [26] protonation of leucine enkephalinoccurs at the N–terminal amino group. This NH3

+ then forms a bifurcated hydrogen bondbetween the oxygen of the fourth carbonyl group and the hydrogen of the OH group of the(C–terminal) COOH group. Similarly, Polfer and co–workers[27] identify energetically fa-vored structures, where the N–terminal NH3

+ is solvated by the COOH group. In any case,the fragmentation of these secondary structures follows entropically favored pathways andcan lead to the formation of cyclic a4 or b4 intermediates which subsequently reopen andfor instance the gly–phe fragment (m/z=205) can be formed. Loss of the glycine residue,NH3 and CO from the reopened a4 intermediate lead to the formation of the m/z=295 frag-ment [28]. The increase of the two peaks with decreasing projectile energy suggests, that thedescribed process requires relatively little activation energy.

6.3.2 H+

The fragmentation patterns of leu–enk obtained with H+ are very similar to the spectrafor He2+, in particular when comparing identical projectile velocities (see 10 keV H+ infig. 6.4(c) and 40 keV He2+ in fig. 6.3 (a)). Compared to the He2+ case, the relative intensityof the strongest peak, the N–terminal immonium fragment at m/z=107, is less pronouncedand the intensities of the higher–mass fragments are also lower. Decreasing the energy ofthe H+ projectile from 20 keV to 5 keV reveals only little effect on the immonium ions anda small increase of the heavier fragments. It is worth mentioning that the intact tyr and pheimmonium ions at m/z=136 and m/z=120 increase or stay constant with decreasing projectilevelocity, at variance to what is observed for He2+.

6.3.3 He+

As mentioned before, large m/z fragments are either strongly or fully suppressed for He+

projectiles (fig. 6.5). For comparable beam intensities, the total fragment yield is much lowerfor He+ compared to the other projectiles as can also be seen from thesignal to noise ratiosof the spectra. It is interesting that for this projectile ion at 5 keV two new fragmentationchannels appear at 177 and 285. These fragments are not observed at higher He+ kinetic en-ergy or for other projectile ions. The peak at m/z = 177 can be assigned to the additional lossof CO from the internal fragment gly–phe [29]. The peak at m/z= 285 remains unassigned.

The trend of the relative intensities of the immonium ions isnot monotonically chang-ing with the He+ kinetic energy. However, at 5 keV the intensity of all immonium ions isincreased again and reveals the strongest fragmentation ofthe peptide, with the phe/tyr sidechain fragment (m/z=91) as the strongest fragmentation channel. In summary all the KIDspectra differ from established techniques in such that strong side chain features show upwhile peaks to backbone scission are weak. The differences to established techniques (andbetween different) keV projectiles must be due to the largeramount of excitation energy de-posited during KID by electronic stopping or by capture of one or more electrons from thepeptide.

74 Peptide fragmentation by keV ion–induced dissociation

50 100 150 200 250 300 3500.0

0.4

0.8

(d)

5 keV H+

m/z (amu)

0.0

0.4

0.8

(c)

10 keV H+

inte

nsity

0.0

0.4

0.8

(b)

15 keV H+

0.0

0.4

0.8

(a)

20 keV H+

91(phe) (tyr)

107(tyr)

120(phe)

136(tyr)

205gly-phe int 221

b2278/279 b3/y2 295

86(leu)

Figure 6.4: Mass spectra of the product cations for H+ with leu–enk at different projectile energies: (a)20 keV, (b) 15 keV, (c) 10 keV and (d) 5 keV. Arrows indicate thedirection of change of (b)–(d) withdecreasing energy.

6.3 Results 75

50 100 150 200 250 300 3500.0

0.4

0.8

*

(d)

5 keV He+

m/z (amu)

*

0.0

0.4

0.8

(c)

10 keV He+

inte

nsity

0.0

0.4

0.8

(b)

15 keV He+

0.0

0.4

0.8

(a)

20 keV He+

Figure 6.5: Mass spectra of the product cations for He+ with leu–enk at different projectile energies:(a) 20 keV, (b) 15 keV, (c) 10 keV and (d) 5 keV. Arrows indicatethe direction of change of (b)–(d)with decreasing energy.

76 Peptide fragmentation by keV ion–induced dissociation

Figure 6.6: Isosurface (0.006 a.u.−3) of the leu–enk valence electron density obtained from densityfunctional theory calculations (see text).

6.4 Discussion

6.4.1 Electronic stopping

As mentioned before, the term electronic stopping describes the electronic excitations of thetarget molecule induced by the projectile ion. For a quantitative analysis of the electronicstopping, here, the molecular valence electrons are treated as an electron gas. We determinedthe valence electron density distribution of protonated leu–enk by summing up the electrondensities of all valence molecular orbitals obtained by density functional theory (DFT) cal-culations (B3LYP level, 6-31+G(d,p) basis set) using Gaussian03 [30]. This was done for thelowest energy conformer as determined by IR spectroscopy and quantum chemical calcula-tions [27]. Figure 6.6 shows an isosurface (0.006/a.u.3, rs = 3.4 a.u.) of the leu–enk valenceelectron density.

In the keV energy range collective electron excitations only contribute weakly to theelectronic stopping. Individual electron excitation due to long range coupling to electron–hole pairs [31] prevails, implying that the deposited energy can be approximated as linear withprojectile velocity [32]. Interactions with the target nuclei (nuclear stopping) are negligiblefor light ions at keV energies. The stopping S being the energy lossdE per trajectory lengthdRcan then be written as a friction force

S=dEdR

= γ(rs)v (6.1)

with γ the friction coefficient depending on the local molecular electron density parameter(one–electron radius):

rs =

(

43

πn0(−→r )

)− 13

(6.2)

6.4 Discussion 77

0 25 50 75 100

inte

nsity

electronic stopping (eV)

σ=165 Å2

34 eVH+

He+

I II

Figure 6.7: Electronic stopping for 5 keV H+ and 20 keV He+ trajectories through randomly orientedprotonated leu–enk.

andv the projectile velocity [33]. This concept has been successfully applied to keV ion–induced multifragmentation of C60 [18, 34]. For H+ and He+, γ(rs) has been calculated byPuska and Nieminen [35] and can be approximated by the exponentials

γ(rs) = 0.53e−(rs−0.41)/2.13 (6.3)

and

γ(rs) = 0.755e−(rs−1.5)/0.88 (6.4)

respectively. The electronic stopping of a projectile ion along a straight line trajectory throughn0(−→r ) can now be integrated. Fig. 6.7 shows the results for 5×106 random trajectories of5 keV/amu H+ and He+ through protonated leu–enk. For both ions there is a local maximumat S≈34 eV due to trajectories traversing the peptide backbone orside chains once and undera relatively steep angle. The strong intensity increase at low stopping (I) is due to grazingtrajectories through the tails of the peptide electron density. Electronic stopping even exceed-ing 100 eV (II) is due to the few trajectories which traverse the peptide structure more thanonce or which have long pathways through high electron density regions. The cross sectionfor trajectories with electronic stopping S≥34 eV (shaded area in fig. 6.7) amounts to 165A2

and 140A2 for H+ and He+ respectively which is close to the geometric cross section of162A2 as determined by CID [36]. The region II class of trajectories will in most cases leadto multifragmentation of the peptide into small fragments which are not trapped for the trapparameters used. For the region I class of trajectories, little excitation due to electronic stop-ping occurs and the interaction is dominated by relatively gentle resonant electron capturefrom the protonated molecule. Due to the deposition of relatively little energy in the peptide,these grazing trajectories lead to less extensive fragmentation and are thus most relevant forthe fragmentation patterns observed.

78 Peptide fragmentation by keV ion–induced dissociation

HOMO

HOMO-2

HOMO-1

Dication spin density

Figure 6.8: Spatial distribution of the highest occupied molecular orbitals of the protonated leu–enkand spin density of the protonated dication radical, indicated as shaded areas.

6.4.2 Electron capture

For H+ and He+ the electronic stopping is almost identical (see fig. 6.7). Therefore thedifferences in fragmentation pattern for the two projectiles must be due to electron capturefrom the molecule. In KID, electron capture will most likely involve electrons from thehighest occupied molecular orbitals (HOMOs) of the peptidesince those electrons can beresonantly captured at large impact parameters. Using the classical over–the–barrier model[37] for symmetric systems such as C60, in an earlier study we have already reported criticalimpact parameters for resonant capture into He+ and He2+ [38].

We determined the spatial distribution of the HOMOs of the protonated leu-enk and thespin density of protonated leu–enk dication (see fig. 6.8) byDFT calculations as mentionedbefore. For the region I class of trajectories, little excitation due to electronic stopping oc-curs and the interaction is dominated by relatively gentle resonant electron capture from theprotonated molecule. The calculated HOMO and HOMO–1 haveπ character and are locatedon the phe side chain ring. The HOMO–2 is aπ–orbital located on the tyr side chain ring.

6.5 Conclusion 79

Also the spin density of the protonated leu–enk dication after non–adiabatic (because of fs–ionization) electron removal resides mainly on these side chains. The fact that we observestrong tyr and phe side chain fragments indicates dissociation close to these side chains andlittle to no distribution of excitation energy over the whole molecule. Such non–ergodicityis also believed to be the basis of ECD [8] where a low energy electron is captured by thepeptide, leading predominantly to formation of c and z fragments (see fig. 6.1) [9].

DFT is well–known for usually underestimating the true ionization energies of the cal-culated HOMOs. We therefore assume that the singly protonated peptide leu–enk has anionization energy IE of 10.9 eV according to the empirical formula by Budniket al. [39].The DFT calculations do tell us however, that the energies ofthe 3 HOMOs lie within aninterval of less than 1 eV. This implies that for H+ and He2+ [21]: electron capture into then = 1 (H+) and the n = 2 state respectively is a (near) resonant process(see fig. 6.9) alreadyoccurring at large impact parameters for which electronic stopping is small. For ion–atomcollisions, electron capture distances can be estimated from the classical over–the–barriermodel [40]. For a fixed ionization energy of the target, the capture distance depends on theion charge state and for the leu–enk ionization potential of10.9 eV, capture distances wouldbe∼7 a.u. for H+ and∼10 a.u. for He2+. Since here we are dealing with a large moleculeinstead of a target atom, the situation is more complex. However, we can translate the infor-mation from the ion–atom system and assume that electron capture sets in at a distance of∼7 a.u. (H+) and∼10 a.u. (He2+) from the leu–enk constituent atom closest to the incomingion. The capture distance defines the class of trajectories for which the traversed electrondensity and thus the molecular excitation due to electronicstopping is minimum. It is thusobvious that for He2+ with its larger capture distance the interaction is more gentle than forH+. This might also be the reason for the larger yield of intact tyr and phe immonium ions(m/z=136, m/z=120) at low projectile energies observed forH+ as compared to He2+: Forformation of a non–terminal intact immonium ion, two bonds have to be broken whereas forformation of the related ions at m/z=120 and m/z=91 only single bond scission is required.

For He+, the n = 1 state at -24.6 eV only becomes resonant in close collisions in whichelectronic stopping is high. This is consistent with the experimental findings: Large frag-ments due to backbone scission are strongly suppressed in the He+ case because of the toohigh excitation energy. Furthermore, for He+ the phe side chain peaks at m/z = 120, 91 arerelatively stronger. This is due to the fact that for production of side chain ions from the non–terminal amino acid phe, the backbone has to be broken twice whereas for the N–terminal tyrside chain, only a single bond scission and thus less excitation energy is necessary. For H+

and He2+ an increase of the projectile ion velocity and thus an increase in deposited energyleads to a relative decrease of the backbone scission peaks (m/z = 150–350) with respect tothe side chain peaks. For the non–resonant case of He+ the projectile velocity has only littleinfluence on the fragmentation.

6.5 Conclusion

To conclude we have shown the keV ion–induced fragmentationof a free protonated pep-tide. In contrast to the findings for conventional peptide dissociation techniques, amino acidside chains dominate the fragment mass spectrum. Backbone scission is a relatively weak

80 Peptide fragmentation by keV ion–induced dissociation

-30

-25

-20

-15

-10

-5

0

Ene

rgy

(eV

)

leu-enk+ H+ He+ He2+

...

n=1 n=2

n=1

n=1

HOMO

...

Figure 6.9: Relevant energy levels of the projectile ions in comparisonwith ionization energy of leu–enk. The dashed rectangle indicates the levels available for resonant electron capture (see text).

channel. This can be qualitatively explained in terms of peptide excitation due to electronicstopping and resonant electron capture from the peptide. KID is a promising new tool forpeptide dissociation, not only because fundamentally different dissociation dynamics occur,but also because ion kinetic and potential energies can be varied over a wide range. Efficientdissociation of very large peptides and small proteins is feasible due to the deposition of largeamounts of kinetic energy on fs–timescales. The use of trapped complex molecular ions asa target facilitates a whole new realm of collision experiments, for instance in the field ofbiomolecular radiation damage and physics of size selectedclusters.

In the context of biological radiation damage on the molecular level, our results showthat the large amounts of excitation energy deposited in complex biomolecules upon keV ionimpact lead to very severe fragmentation. This is in line with previous results obtained withnucleobases [15], deoxyribose [21] or amino acids [16]. Thequestion remains, how the sol-vation layers that surround biomolecules in living cells affect the fragmentation dynamics.Being based on target production by means of electrospray ionization, the developed tech-nique allows investigation of nanosolvated biomolecules [41]. In the future we plan to exploitthis potential by studying keV ion–induced dissociation ofnanosolvated oligonucleotides asnano–models of the situation in the nucleus of a living cell.

REFERENCES 81

References

[1] B. Coupier, B. Farizon, M. Farizon, M. J. Gaillard, F. Gobet, N. V. D. Faria, G. Jal-bert, S. Ouaskit, M. Carre, B. Gstir, G. Hanel, S. Denifl, L. Feketeova, P. Scheier, andT. D. Mark, Eur. Phys. J. D20, 459 (2002).

[2] J. de Vries, R. Hoekstra, R. Morgenstern, and T. Schlath¨olter, Phys. Rev. Lett.91,053401 (2003).

[3] S. Martin, R. Bredy., A. R. Allouche., J. Bernard., A. Salmoun., B. Li, and L. Chen,Phys. Rev. A77, 062513 (2008).

[4] V. Bernigaud, B. Manil, L. Maunoury, J. Rangama, and B. A.Huber, Eur. Phys. J. D51, 125 (2009).

[5] J. Fenn, Angew. Chem.-Int. Edit.42, 3871 (2003).

[6] M. Karas and F. Hillenkamp, Anal. Chem.60, 2299 (1988).

[7] A. Shukla and J. Futrell, J. Mass Spectrom.35, 1069 (2000).

[8] R. A. Zubarev, N. L. Kelleher, and F. W. McLafferty, J. Am.Chem. Soc.120, 3265(1998).

[9] P. Hvelplund, B. Liu, S. Brøndsted Nielsen, and S. Tomita, Int. J. Mass Spectrom.225,83 (2003).

[10] A. Ehlerding, C. S. Jensen, J. A. Wyer, A. I. S. Holm, P. Jorgensen, U. Kadhane,M. K. Larsen, S. Panja, J. C. Poully, E. S. Worm, H. Zettergren, P. Hvelplund, andS. Brøndsted Nielsen, Int. J. Mass Spectrom.282, 21 (2009).

[11] R. Dunbar and T. McMahon, Science279, 194 (1998).

[12] P. Schnier, W. Price, R. Jockusch, and E. Williams, J. Am. Chem. Soc.118, 7178(1996).

[13] F. Talbot, T. Tabarin, R. Antoine, M. Broyer, and P. Dugourd, J. Chem. Phys.122(2005).

[14] B. Liu, S. Brøndsted Nielsen, P. Hvelplund, H. Zettergren, H. Cederquist, B. Manil, andB. A. Huber, Phys. Rev. Lett.97, 133401 (2006).

[15] J. de Vries, R. Hoekstra, R. Morgenstern, and T. Schlatholter, J. Phys. B35, 4373(2002).

[16] S. Bari, P. Sobocinski, J. Postma, F. Alvarado, R. Hoekstra, V. Bernigaud, B. Manil,J. Rangama, B. A. Huber, and T. Schlatholter, J. Chem. Phys.128(2008).

[17] F. Alvarado, J. Bernard, B. Li, R. Bredy, L. Chen, R. Hoekstra, S. Martin, andT. Schlatholter, ChemPhysChem9, 1254 (2008).

82 REFERENCES

[18] T. Schlatholter, O. Hadjar, R. Hoekstra, and R. Morgenstern, Phys. Rev. Lett.82, 73(1999).

[19] P. Moretto-Capelle, D. Bordenave-Montesquieu, A. Rentenier, and A. Bordenave-Montesquieu, J. Phys. B34, L611 (2001).

[20] J. Laskin, J. Phys. Chem. A110, 8554 (2006).

[21] F. Alvarado, S. Bari, R. Hoekstra, and T. Schlatholter, Phys. Chem. Chem. Phys.8,1922 (2006).

[22] NIST Mass Spec Data Center, S.E. Stein, director, inNIST Chemistry WebBook, NISTStandard Reference Database Number 69, edited by P. Linstrom and W. Mal-lard, National Institute of Standards and Technology, Gaithersburg MD, 20899(http://webbook.nist.gov), June 2005.

[23] V. Rakov, O. Borisov, and C. Whitehouse, J. Am. Soc. MassSpectrom.15, 1794 (2004).

[24] R. Jockusch, K. Paech, and E. Williams, J. Phys. Chem. A104, 3188 (2000).

[25] B. Paizs and S. Suhai, Mass Spectrom. Rev.24, 508 (2005).

[26] K. Jalkanen, J. Phys.-Condes. Matter15, S1823 (2003).

[27] N. Polfer, J.Oomens, S. Suhai, and B. Paizs, J. Am. Chem.Soc.129, 5887 (2007).

[28] R. Vachet, B. Bishop, B. Erickson, and G. Glish, J. Am. Chem. Soc.119, 5481 (1997).

[29] A. Alexander and R. Boyd, Int. J. Mass Spectrom. Ion Process.90, 211 (1989).

[30] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheese-man, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A.Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox,H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala,K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford,J. Cioslowski, B. B. Ste-fanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith,M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. John-son, W. Chen, M. W. Wong, C. Gonzalez, and J. A. Pople, Gaussian 03, Revision C.02,Gaussian, Inc., Wallingford, CT, 2004.

[31] T. Ferrell, P. Echenique, and R. Ritchie, Solid State Commun.32, 419 (1979).

[32] I. Abril, R. Garciamolina, and N. Arista, Nucl. Instrum. Meth. B90, 72 (1994).

[33] A. Narmann, R. Monreal, P. M. Echenique, F. Flores, W. Heiland, and S. Schubert,Phys. Rev. Lett.64, 1601 (1990).

REFERENCES 83

[34] O. Hadjar, P. Foldi, R. Hoekstra, R. Morgenstern, and T.Schlatholter, Phys. Rev. Lett.84, 4076 (2000).

[35] M. J. Puska and R. M. Nieminen, Phys. Rev. B27, 6121 (1983).

[36] N. C. Polfer, B. C. Bohrer, M. D. Plasencia, B. Paizs, andD. E. Clemmer, J. Phys.Chem. A112, 1286 (2008).

[37] A. Barany and C. Setterlind, Nucl. Instrum. Meth. B98, 184 (1995).

[38] O. Hadjar, R. Hoekstra, R. Morgenstern, and T. Schlath¨olter, Phys. Rev. A63 (2001).

[39] B. Budnik, Y. Tsybin, P. Hakansson, and R. Zubarev, J. Mass Spectrom.37, 1141(2002).

[40] A. Niehaus, J. Phys. B19, 2925 (1986).

[41] B. Liu, N. Haag, H. Johansson, H. T. Schmidt, H. Cederquist, S. Brøndsted Nielsen,H. Zettergren, P. Hvelplund, B. Manil, and B. A. Huber, J. Chem. Phys.128(2008).