Amplified histidine effect in electron-transfer dissociation of histidine-rich peptides from histatin 5

Article abstract of DOI:10.1016/j.ijms.2010.08.021

We report electron-transfer dissociation (ETD) mass spectra of histidine-containing peptides DSHAK, FHEK, HHGYK, and HHSHR from trypsinolysis of histatin 5. ETD of both doubly and triply protonated peptides provided sequence ions of the c and z type. In a

Full text of DOI:10.1016/j.ijms.2010.08.021

International Journal of Mass Spectrometry 306 (2011) 99–107
Contents lists available at ScienceDirect
International Journal of Mass Spectrometry
journal homepage: www.elsevier.com/locate/ijms
Amplified histidine effect in electron-transfer dissociation of histidine-rich
peptides from histatin 5
Thomas W. Chung, Frantisˇek Turecˇek∗
Department of Chemistry, Bagley Hall, Box 351700, University of Washington, Seattle, WA 98195-1700, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 July 2010
Received in revised form 19 August 2010
Accepted 23 August 2010
Available online 31 August 2010
We report electron-transfer dissociation (ETD) mass spectra of histidine-containing peptides DSHAK,
FHEK, HHGYK, and HHSHR from trypsinolysis of histatin 5. ETD of both doubly and triply protonated
peptides provided sequence ions of the c and z type. In addition, electron transfer to doubly protonated
peptides produced abundant long-lived cation-radicals, (M+2H)•+, whose relative intensities depended
on the peptide sequence and number of histidine residues. CID-MS3 spectra of (M+2H)•+ cation-radicals
were entirely different from the ETD spectra of the doubly charged ions and involved radical-driven losses
of C4H6N2 neutral fragments from the histidine residues and charge-driven backbone cleavages forming
b and y ions. Product ions from CID of (M+2H)•+ were further characterized by CID-MS4 spectra to distin-
guish the histidine residues undergoing loss of C4H6N2. The ETD-CID-MSn mass spectra are interpreted
by considering radical-induced rearrangements of histidine side chains in the long-lived charge-reduced
ions.
Keywords:
Electron-transfer dissociation
Histidine effect
Histatin 5
Histidine-rich tryptic peptide
Radical rearrangement
Zwitterion
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
[3–6] and their reactive intermediates are generically sketched
in Scheme 1.
Electron-transfer dissociation (ETD) [1] mass spectra of
histidine-containing peptides have been reported to display inter-
esting effects caused by specific reactions in the histidine side chain.
Radicals formed by one-electron reduction of protonated imidazole
ring in histidine residues undergo proton-catalyzed rearrange-
ments that result in substantial stabilization of charge-reduced
intermediates. This effect has been first observed by Xia et al. as
“electron-transfer-no-dissociation” (ETnoD) and found to be spe-
cific for histidine-containing peptide ions [2]. We have studied
the reaction mechanisms of what we called the histidine effect
[3] for peptide ions ranging from singly protonated dipeptides
(GH, HG) [4], tripeptides (HAL, AHL, ALH) [5], to several singly
[6] and doubly protonated pentapeptides [3]. Charge reduction in
singly protonated peptides forms neutral radicals that cannot be
directly detected by mass spectrometry. However, peptide radi-
cals can be reionized by collisions at keV kinetic energies to form
cations, as in neutralization–reionization mass spectrometry [7,8],
or anions, as in charge-reversal mass spectrometry [9,10], that
provide information on the radical structure [6,11]. The rearrange-
ments in histidine residues were found to be catalyzed by proton
donors such as C-terminal and Asp carboxyl or N-terminal ammo-
nium groups. The presumed mechanisms for the rearrangements
Upon electron attachment to the protonated His ring, the
intermediate N-1ꢀ,N-3ꢀ-H-imidazolium radical [12] can undergo
conformational rotations that bring the N-terminal ammonium or
a carboxyl group to the proximity of the C-2ꢀ atom which car-
ries a substantial odd-electron density [5]. Proton transfer from
the ammonium group to C-2ꢀ (pathway a, Scheme 1) is highly
exothermic and produces a stable cation-radical which is detected
in the ETD spectrum [3]. Proton transfer from a carboxyl group is
mildly exothermic to form a transient zwitterion consisting of the
imidazoline cation-radical and the carboxylate anion (pathway b,
Scheme 1) [5]. The zwitterion can undergo a further facile rear-
rangement by N-3ꢀ-H proton migration onto the carboxylate group
to exothermically form an N-1ꢀ,C-2ꢀ-H radical. The latter is more
stable than the initial histidine radical and can be detected in the
ETD spectrum. Upon collisional excitation, the rearranged His rad-
ical can eliminate the C4H6N2 side-chain fragment, converting the
His residue to a Gly C␣ radical. The exact structure of the neu-
tral C4H6N2 fragment is unknown but most likely corresponds to a
methylimidazole tautomer or a mixture thereof.
Experimental evidence for the histidine rearrangement fol-
lowed from charge-reversal (+CR−) mass spectra of singly
protonated peptides and collision induced dissociation (CID-MS3)
spectra of charge-reduced but non-dissociating peptide ions. The
ETD-CID-MS3 spectra showed that long-lived charge-reduced pep-
tide cation-radicals did not undergo backbone N–C␣ bond cleavages
which are typical for ETD. Instead, CID of charge-reduced ions
∗
Corresponding author. Tel.: +1 206 685 2041; fax: +1 206 685 3478.
E-mail address: turecek@chem.washington.edu (F. Turecˇek).
1387-3806/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ijms.2010.08.021

100
T.W. Chung, F. Turecˇek / International Journal of Mass Spectrometry 306 (2011) 99–107
Scheme 1.
resulted in the loss of a C4H6N2 neutral fragment from the his-
tidine side chain that was indicative of a rearrangement [5].
Since the previous research on histidine-containing peptides con-
cerned only synthetic model compounds, it became of interest
to investigate peptides produced from natural sources, such as
histidine-containing peptides and proteins. A convenient source is
found among histidine-rich antibacterial proteins of the histatin
family [13]. Histatins 1, 3, and 5 are small cationic proteins con-
taining, respectively, 38, 32, and 24 amino acid residues. They are
secreted by human parotid and mandibular glands and are present
in saliva where they have several protective functions [14]. His-
tatin 5 has the sequence DSHAKRHHGYKRKFHEKHHSHRGY which
is a convenient source of histidine-containing peptides upon trypsi-
nolysis, e.g., DSHAK, HHGYK, FHEK, and HHSHR. These small (4–5
residues) basic peptides thus have naturally different numbers and
positions of His residues that can be expected to produce doubly
and triply charged ions by electrospray and thus be favorable for
studying the histidine effect. We now report an experimental ETD-
CID-MSn study focused on the dissociations induced by electron
transfer and the analysis of charged-reduced intermediates. We
wish to show that the histidine effect is amplified in the presence
of multiple histidine residues in a fashion which is also sequence
dependent.
proteolysis using immobilized trypsin (Catalog No. V9013, Promega
Corp., Madison, WI, USA), as described by the manufacturer. Briefly,
3.0 mg of histatin 5 was first sonicated in 100 ␮L of >18.2 ꢀ
MilliQ water to make a 30 mg/mL protein stock solution. A so-
called “protein digestion mixture” was prepared by combining
in a 2.0 mL microcentrifuge tube 16.7 ␮L of 30 mg/mL histatin 5
protein stock solution, 48 ␮L acetonitrile, and 55.3 ␮L of 50 mM
ammonium bicarbonate buffer. Immobilized trypsin, which comes
conjugated to a cellulose resin and is provided in a 1:1 (v/v)
slurry in 50 mM CH3CO2H, 1 mM CaCl2, and 0.02% NaN3, was
resuspended and 600 ␮L of it was dispensed onto a spin column,
which was then centrifuged to remove liquid from the resin. The
resin was washed with 400 ␮L of 50 mM ammonium bicarbon-
ate. This wash cycle was repeated for a total of three times. The
protein digestion mixture was added quantitatively and directly
to the resin in the spin basket and incubated at room tempera-
ture for 30 min. Peptide recovery was performed by adding 300 ␮L
of “Peptide Recovery Buffer” to the spin column and briefly cen-
trifuging to remove the peptide solution from the resin and into
a clean 2 mL microcentrifuge tube. This “Peptide Recovery Buffer”
was previously prepared by combining 400 ␮L acetonitrile, 20 ␮L
of 10% trifluoroacetic acid, and 580 ␮L of >18.2 ꢀ MilliQ water.
The peptide recovery step was repeated in order to achieve a
final tryptic peptide solution of no more than 720 ␮L for down-
stream applications. A sample solution was prepared by combining
371.5 ␮L of the tryptic peptide solution with 1.3285 mL of a 50/50
methanol–water solution that encompassed 2% acetic acid. Direct
infusion electrospray of this single, heterogeneous sample solution
afforded multiply charged peptide ions DSHAK, HHGYK, FHEK, and
HHSHR.
2. Experimental
2.1. Materials and methods
Histatin 5 protein (98% pure) was purchased from Aroz Tech-
nologies (Cincinnati, OH, USA). Histatin
5 was subjected to

T.W. Chung, F. Turecˇek / International Journal of Mass Spectrometry 306 (2011) 99–107
101
Fig. 1. ETD mass spectra of (A) (DSHAK+2H)2+ (m/z 279.2) and (B) (DSHAK+3H)3+ (m/z 186.4) ions. (C) CID-MS3 mass spectrum of mass selected (DSHAK+2H)•+ (m/z 558)
from charge-reduction of (DSHAK+2H)2+. The ion relative intensities were normalized to the sum of intensities of all charge-reduced ions but excluding the residual precursor
ion intensity.
Electron-transfer dissociation mass spectra were measured on
a Thermo Fisher Scientific (San Jose, CA, USA) LTQ XL linear ion
trap instrument outfitted with a chemical ionization source for
the production of fluoranthene anion radicals as ETD reagent. Pre-
cursor cations (M+2H)2+ and (M+3H)3+ were mass isolated with a
window of 1.6–2.5 and 0.7–2.0 m/z units, respectively, to accom-
modate nearest 13C isotopologues and allowed to react for 100, 200,
and 300 ms with fluoranthene anions. The 200 ms ETD spectra are
reported.
Survivor charge-reduced (M+2H)•+ from electron transfer were
mass selected with a 1 m/z unit window and collisionally dis-
sociated (CID-MS3) with the ion excitation energy set at 20% on
the instrument scale. This gave abundant (M+2H–C4H6N2)•+ ions,
which were subsequently mass selected and subjected to collisional

102
T.W. Chung, F. Turecˇek / International Journal of Mass Spectrometry 306 (2011) 99–107
Fig. 2. CID-MS4 mass spectra of (A) m/z 476 ion by loss of C4H6N2 from charge-reduced (DSHAK+2H)•+; (B) m/z 479 ion by loss of C4H6N2 from charge-reduced (FHEK+2H)•+
;
(C) m/z 560 ion by loss of C4H6N2 from charge-reduced (HHGYK+2H)•+, and (D) m/z 592 ion by loss of C4H6N2 from charge-reduced (HHSHR+2H)•+. The ion relative intensities
were normalized to the sum of intensities of all charge-reduced ions.
activation with the ion excitation energy set at 20% on the instru-
ment scale to yield the CID-MS4 spectra reported here. CID-MS3
mass spectra were also taken of all ETD c and z sequence fragments.
3. Results
3.1. DSHAK
The DSHAK peptide formed singly, doubly and triply charged
ions by electrospray that appeared at m/z 557.3 (48%), m/z 279.1
(48%) and m/z 186.4 (4%), respectively. Thus only about a half of the
produced ion population was useful for ETD. Electron transfer to the
Scheme 2.
m/z 279.1 precursor ion resulted in the formation of an abundant

T.W. Chung, F. Turecˇek / International Journal of Mass Spectrometry 306 (2011) 99–107
103
charge-reduced (M+2H)•+ cation-radical at m/z 558, which consti-
tuted 26% of the sum of charge-reduced ion intensities (Fig. 1A).
Fragmentation proceeded by loss of H (m/z 557), C2H4O2 (m/z 498),
and by backbone cleavages forming the c4 (m/z 428), z4 (m/z 426), c3
(m/z 357) and z3 (m/z 339) ions. The ETD mass spectrum of the triply
protonated ion (m/z 186.4) displayed similar fragments (Fig. 1B). In
particular, the singly charged backbone ions c4, z4, c3, and z3 were
formed together with the (M+2H)•+ cation-radical. In contrast, the
presence in the spectrum of the singly reduced (M+3H)•2+ ion can-
not be confirmed because of overlap with the 13C isotope satellite
of the (M+2H)2+ ion at m/z 280. As a general remark, the ETD mass
spectrum of the (M+3H)3+ ion shows more secondary fragments
(m/z 411, 371, 250, 220, 202, 186, etc.) presumably due to greater
ion excitation by electron transfer to the triply charged ion com-
pared to the same process concerning the doubly charged ion. The
origin of these singly charged secondary fragments has been estab-
lished by ETD-MS3 experiments where each of the primary c and
z fragments from ETD was mass selected and collisionally dissoci-
ated. For example, CID of the c4 fragment ion (m/z 428) produced
the m/z 411 ion as a dominant secondary fragment by loss of ammo-
nia. We note that although the m/z 411 ion formally corresponds
to a b4 sequence ion, its formation through the c4 ion is corrobo-
rated by the ETD-MS3 spectrum. Likewise, the m/z 202 ion in the
Fig. 1B spectrum, which formally is a z2(AK) fragment, is a dom-
inant secondary fragment from CID of the z4 ion at m/z 426 and
presumably is formed by a cascade elimination [15] of a neutral
Fig. 3. ETD mass spectra of (A) (FHEK+2H)2+ (m/z 280.6) and (B) (FHEK+3H)3+ (m/z 187.4) ions. (C) CID-MS3 mass spectrum of mass selected (FHEK+2H)•+ (m/z 561) from
charge-reduction of (FHEK+2H)2+. The ion relative intensities were normalized to the sum of intensities of all charge-reduced ions but excluding the residual precursor ion
intensity.

104
T.W. Chung, F. Turecˇek / International Journal of Mass Spectrometry 306 (2011) 99–107
fragment containing the deaminated serine and histidine residue
(HOCH2CHCONHCH(CH2C3H3N2)CONH). Interestingly, the forma-
tion of the m/z 202 from the z3 ion (m/z 339) is much weaker, the
main dissociation being an elimination of a C3H8N radical from the
Lys residue.
The efficiency for ET (%E) to the doubly and triply charged pep-
tide ions was 77% and 92%, respectively, when expressed as a ratio
of the sum of all charge-reduced ion intensities, ꢁIETD, to the total
ion count, ꢁIETD + IP, including the residual precursor ion intensity
IP (Eq. (1)). Both these efficiencies refer to 200 ms ion–ion interac-
tion time.
ꢀ
100
IETD
ꢀ
%E =
(1)
IP
+
IETD
The charge-reduced (M+2H)•+ cation-radicals from ETD of m/z
279.1 and m/z 186.4 gave very similar CID-MS3 spectra as repre-
sented by that from m/z 279.1 (Fig. 1C). The ETD-CID-MS3 spectrum
is dominated by a fragment ion at m/z 476 by loss of a 82 Da neu-
tral fragment, presumably C4H6N2 from the His side chain [2,16].
A conspicuous feature of Fig. 1C spectrum is the absence or very
low intensity of the c4, z4, c3, and z3 backbone fragment ions (cf.
Fig. 1A). Backbone dissociations are represented by abundant y4,
b4, y3, and b3 ions at m/z 442, 411, 355, and 340, respectively.
These charge-induced dissociations compete with the elimination
of the His C4H6N2 side-chain fragment. The pertinent fragment
ion at m/z 476 was further selected by mass and analyzed by CID.
The CID-MS4 spectrum (Fig. 2A) showed eliminations of peripheral
groups (CO2, H2O, CH2O) in combination with backbone dissocia-
tions by eliminations of amino acid residues from the N-terminus,
e.g., (AspNH2–H) (m/z 345), AspNH2 (m/z 344), and (AspSerNH2–H)
(m/z 259). These dissociations must involve N–C␣ bond cleavages
which may be triggered by radical migrations along the backbone
and side chains in the m/z 476 cation-radical [17]. The backbone
cleavages are sketched in Scheme 2, although the exact mecha-
nisms have not been established.
Scheme 3.
3.3. HHGYK
Electrospray of the HHGYK peptide produced singly, doubly
and triply charged ions at m/z 641.4 (32%), m/z 321.2 (28%)
and m/z 214.1 (40%), respectively. ETD of the doubly and triply
charged HHGYK ions proceeded with %E = 90 and 99%, respec-
tively, at 200 ms ion–ion interaction time. ETD of the doubly
charged ion at m/z 321.2 produced a dominant charge-reduced
(M+2H)•+ ion at m/z 642 that constituted 44% of total ETD ion
intensities. A series of backbone c fragments were found at m/z
292 (c2), 349 (c3), and 512 (c4), which were complemented by
the z4 ion at m/z 488 (Fig. 4A). ETD of the triply charged ion
at m/z 214.1 gave an abundant (M+2H)•+ ion at m/z 642 and
a number of fragments due to backbone and consecutive dis-
sociations (Fig. 4B). CID-MS3 of the (M+2H)•+ ion at m/z 642
showed a dominant loss of C4H6N2 (m/z 560, Fig. 4C). The other
dissociations consisted of sequential eliminations of water and
ammonia combined with loss of H-atoms to give fragment ion
clusters at m/z 623–626 and m/z 605–608. Backbone dissocia-
tions were represented by the abundant b2 fragment ion at m/z
275.
3.2. FHEK
Electrospray of the FHEK peptide produced singly, doubly, and
triply charged ions at m/z 560.3 (35%), m/z 280.6 (57%), and m/z
187.4 (8%), respectively. The ETD spectra of the doubly and triply
charged ions were analogous to those of DSHAK ions. The ETD
mass spectrum of the major doubly charged ion at m/z 280.6 gave
an abundant charge-reduced (M+2H)•+ cation-radical at m/z 561
which comprised 34% of the sum of charge-reduced ion intensities
(Fig. 3A). Backbone fragment ions appeared at m/z 431 (c3), 397
(z3), 302 (c2) and a weak peak at m/z 260 (z2). The abundant back-
bone fragment ions all contained the His residue. The ETD mass
spectrum of the triply charged ion (m/z 187.4) also gave the singly
charged c and z fragment ions, in addition to numerous fragments
from consecutive dissociations (Fig. 3B). The electron-transfer effi-
ciency at 200 ms ion–ion interaction time was %E = 80% and 85% for
the doubly and triply charged precursor ions, respectively. CID-MS3
of the charge-reduced (M+2H)•+ ion at m/z 561 resulted in losses
of water, ammonia, and backbone cleavages forming the b3 (m/z
414), b2 (m/z 285) and y2 (m/z 276) fragment ions (Fig. 3C). Loss
of C4H6N2 was also observed (m/z 479), although at lower relative
abundance than for DSHAK. CID-MS4 of the m/z 479 ion resulted
in a dominant loss of a 16 Da neutral fragment, presumably NH2,
to give a m/z 463 fragment ion (Fig. 2B). This somewhat unusual
dissociation can be facilitated by migration of a benzylic H-atom
from the Phe residue [17] to the His C␣ radical which requires a
six-membered transition state [18] and activates the N-terminal
H2N–C bond for homolytic dissociation (Scheme 3).
CID-MS4 of the m/z 560 ion due to loss of C4H6N2 formed a dom-
inant m/z 367 fragment ion which corresponds to a y3(GYK) ion
(Fig. 2C). In addition, the spectrum displayed a regular y4(HGYK)
sequence fragment ion at m/z 504, as well as a modified y4(GGYK)
ion at m/z 424. The m/z 504 fragment ion can be expected from a
m/z 560 precursor that lost the side chain from the N-terminal His1
residue, while the m/z 424 fragment ion must be formed from a
precursor that lost the side chain from the internal His2 residue.
Hence, the CID-MS4 spectrum indicates that the loss of the His side
chain was non-specific as to the His residue position in the peptide
sequence.

T.W. Chung, F. Turecˇek / International Journal of Mass Spectrometry 306 (2011) 99–107
105
Fig. 4. ETD mass spectra of (A) (HHGYK+2H)2+ (m/z 321.2) and (B) (HHGYK+3H)3+ (m/z 214.1) ions. (C) CID-MS3 mass spectrum of mass selected (HHGYK+2H)•+ (m/z 642)
from charge-reduction of (HHGYK+2H)2+. The ion relative intensities were normalized to the sum of intensities of all charge-reduced ions but excluding the residual precursor
ion intensity.
3.4. HHSHR
This last tryptic peptide contained three His residues and formed
different in that the former showed a dominant loss of C4H6N2
to give a m/z 592 fragment (Fig. 5C), while the latter involved a
dominant loss of ammonia to give a m/z 658 fragment (Fig. 5D).
These dissociations are consistent with the radical nature of the
(M+2H)•+ ion to undergo a radical-induced loss of the His side
chain on the one hand, and the closed-shell nature of the (M+3H)+
ion to eliminate ammonia via a charge-driven reaction on the
other.
singly, doubly and triply charged ions upon electrospray that
appeared at m/z 673.4 (18%), m/z 337.2 (30%), and m/z 224.9 (51%),
respectively, giving >80% of ion precursors suitable for ETD. Elec-
tron transfer to the doubly charged ion at m/z 337.2 was 91%
efficient and gave an intense charge-reduced (M+2H)•+ ion at
m/z 674 that constituted 52% of charge-reduced ion intensities
(Fig. 5A). ETD of the triply charged ion at m/z 224.8 was 99% efficient
at 200 ms of ion–ion interaction time. Interestingly, the charge-
reduced ion from the m/z 224.8 precursor appeared at m/z 675
and must have corresponded to (M+3H)+ (Fig. 5B). The CID-MS3
mass spectra of the (M+2H)•+ and (M+3H)+ ions were distinctly
CID-MS4 of the m/z 592 ion gave several products (Fig. 2D).
Among those, the m/z 536 ion can be formed by loss of a 56 Da
fragment from the truncated N-terminal His1 residue. Likewise,
the m/z 383 fragment ion can correspond to a z3 ion, indicating
the His4 residue was intact. We did not attempt to assign the other
fragments in the Fig. 5D spectrum.

106
T.W. Chung, F. Turecˇek / International Journal of Mass Spectrometry 306 (2011) 99–107
Fig. 5. ETD mass spectra of (A) (HHSHR+2H)2+ (m/z 337.2) and (B) (HHSHR+3H)3+ (m/z 224.8) ions. (C) CID-MS3 mass spectrum of mass selected (HHSHR+2H)•+ (m/z 674)
from charge-reduction of (HHSHR+2H)2+. (D) CID-MS3 mass spectrum of mass selected (HHSHR+3H)+ (m/z 675) from charge-reduction of (HHSHR+3H)3+. The ion relative
intensities were normalized to the sum of intensities of all charge-reduced ions but excluding the residual precursor ion intensity.
4. Discussion
from ETD were practically identical (31%) for the three Lys-
terminated peptides, although their distribution differed. All c
and z ions from the Lys-containing peptides contained proto-
nated His residues, whereas the complementary Lys-containing
fragments were neutral. The increased (M+2H)•+ relative inten-
sities in the ETD spectra of FHEK and HHGYK were presumably
due to decreased side-chain dissociations by loss of ammonia or
C2H4O2 from the Asp residue. The Arg-terminated peptide showed
lower c + z relative abundances (21%). In addition, the z2 and z3
fragments from HHSHR showed substantial (z + 1) satellites, and
the c4 ion was accompanied by a (c4 − 1) ion. These were due to
intramolecular H-atom migrations occurring in ion–molecule com-
The above-presented data indicate that His residues have spe-
cific effects on ETD spectra of ions from small tryptic peptides.
First, the electron-transfer efficiency weakly correlated with the
number of His residues in the peptide ion, whereby %E = 77, 80,
90, and 91% for doubly protonated DSHAK, FHEK, HHGYK, and
HHSHR, respectively. A stronger effect was found for the rel-
ative intensities of charge-reduced, non-dissociating, (M+2H)•+
cation-radicals which increased from 26% for doubly protonated
DSHAK to 34% for FHEK, 44% for HHGYK, and 52% for HHSHR.
The combined relative intensities of c and z backbone fragments

T.W. Chung, F. Turecˇek / International Journal of Mass Spectrometry 306 (2011) 99–107
107
plexes of complementary c and z fragments, as studied previously
[19].
instrument. Research support by the NSF (Grant CHE-0750048) is
gratefully appreciated.
CID of long-lived (M+2H)•+ cation-radicals showed radical-
induced losses of C4H6N2 fragments from the His residues, which
competed with proton-driven losses of water, ammonia, and back-
bone dissociations forming b and y fragment ions. The loss of
C4H6N2 upon CID is thought to originate from rearranged (M+2H)•+
charge-reduced ions according to Scheme 1 [3–6] and the extent of
this fragmentation can presumably be taken as a marker for the
population of rearranged ions. The rearrangement occurs in the
imidazole ring, requires an internal catalysis by a proton donor,
and results in a stabilization of the charge-reduced ion [3–6].
The tryptic peptides from histatin 5 contain basic residues at the
C-terminus and 1–3 histidine residues. In keeping with previ-
ous studies of histidine-containing peptide ions, we presume that
the doubly charged ions are protonated at the C-terminal basic
amino acid (Lys or Arg) and at one of the His residues. In charge-
reduced (DHSAK+2H)•+, the His rearrangement can be catalyzed
by the Asp or C-terminal COOH groups; accordingly, the loss of
C4H6N2 is a dominant fragmentation channel. In contrast, charge-
reduced (FHEK+2H)•+ shows less abundant loss of C4H6N2 which
we interpret as less efficient histidine rearrangement due to a lower
accessibility to the COOH group.
References
[1] J.E.P. Syka, J.J. Coon, M.J. Schroeder, J. Shabanowitz, D.F. Hunt, Proc. Natl. Acad.
Sci. U.S.A. 101 (2004) 9528.
[2] Y. Xia, H.P. Gunawardena, D.E. Erickson, S.A. McLuckey, J. Am. Chem. Soc. 129
(2007) 12232.
[3] F. Turecˇek, T.W. Chung, C.L. Moss, J.A. Wyer, A. Ehlerding, H. Zettergren, S.B.
Nielsen, P. Hvelplund, J. Chamot-Rooke, B. Bythell, B. Paizs, J. Am. Chem. Soc.
132 (2010) 10728.
[4] F. Turecˇek, C. Yao, Y.M.E. Fung, S. Hayakawa, M. Hashimoto, H. Matsubara, J.
Phys. Chem. B 113 (2009) 7347.
[5] F. Turecek, J.W. Jones, T. Towle, S. Panja, S.B. Nielsen, P. Hvelplund, B. Paizs, J.
Am. Chem. Soc. 130 (2008) 14584.
[6] F. Turecˇek, S. Panja, J.A. Wyer, A. Ehlerding, H. Zettergen, S.B. Nielsen, P.
Hvelplund, B. Bythell, B. Paizs, J. Am. Chem. Soc. 131 (2009) 16472.
[7] (a) P.M. Curtis, B.W. Williams, R.F. Porter, Chem. Phys. Lett. 65 (1979) 296;
(b) P.C. Burgers, J.L. Holmes, A.A. Mommers, J.K. Terlouw, Chem. Phys. Lett. 102
(1983) 1;
(c) P.O. Danis, C. Wesdemiotis, F.W. McLafferty, J. Am. Chem. Soc. 105 (1983)
7454.
[8] For recent reviews see;;
(a) F. Turecek, Top. Curr. Chem. 225 (2003) 77;
(b) D.V. Zagorevskii, in: J.A. McCleverty, T.J. Meyer (Eds.), Comprehensive Coor-
dination Chemistry II, Elsevier, Oxford, 2004, p. 381;
(c) F. Turecek, in: P.B. Armentrout (Ed.), Encyclopedia of Mass Spectrometry,
vol. 1, Elsevier, Amsterdam, 2003, p. 528 (Chapter 7);
(d) D.V. Zagorevskii, Coord. Chem. Rev. 225 (2002) 5;
(e) P. Gerbaux, C. Wentrup, R. Flammang, Mass Spectrom. Rev. 19 (2000)
367;
(f) D.V. Zagorevskii, J.L. Holmes, Mass Spectrom. Rev. 18 (1999) 87;
(g) C. Schalley, G. Hornung, D. Schröder, H. Schwarz, Chem. Soc. Rev. 27 (1998)
91.
Increasing the number of His residues in (HHGYK+2H)•+ and
(HHSHR+2H)•+ promotes the loss of C4H6N2. The backbone frag-
mentations upon CID-MS4 of the m/z 560 ion from HHGYK then
suggest that the loss of C4H6N2 had occurred from either His
residue. This indicates a statistical effect whereby either His residue
is protonated in the precursor ions and then undergoes a rearrange-
ment in the charge-reduced ion. We also note that charge-reduced
peptide ions that lack histidine residues undergo different dissoci-
ations, as studied recently [20].
[9] D. Schröder, in: P.B. Armentrout (Ed.), Encyclopedia of Mass Spectrometry, vol.
1, Elsevier, Amsterdam, 2004, pp. 521–528 (Chapter 8).
[10] S. Hayakawa, J. Mass Spectrom. 39 (2004) 111.
[11] P. Hvelplund, B. Liu, S.B. Nielsen, S. Panja, J.C. Poully, K. Støchkel, Int. J. Mass
Spectrom. 263 (2007) 66.
[12] V.Q. Nguyen, F. Turecek, J. Mass Spectrom. 31 (1996) 1173.
[13] F.G. Oppenheim, T. Xu, F.M. McMillian, S.M. Levitz, R.D. Diamond, G.D. Offner,
R.F. Troxler, J. Biol. Chem. 263 (1988) 7472.
[14] E.J. Helmerhorst, R.F. Troxler, F.G. Oppenheim, Proc. Natl. Acad. Sci. U.S.A. 98
(2001) 14637.
5. Conclusions
Histidine-rich tryptic peptides from histatin 5 form abundant
charge-reduced, non-dissociating cation-radicals upon electron
transfer. CID of the charge-reduced ions triggers elimination of
C4H6N2 which depends on the peptide sequence and is amplified
in peptide ions containing two or three His residues. This amplified
histidine effect is interpreted by facile internal rearrangement in
His radicals to form intermediates which are prone for the C4H6N2
elimination.
[15] N. Leymarie, C.E. Costello, P.B. O’Connor, J. Am. Chem. Soc. 125 (2003) 8949.
[16] Y.M.E. Fung, T.-W.D. Chan, J. Am. Soc. Mass Spectrom. 16 (2005) 1523.
[17] (a) T. Ly, R.R. Julian, J. Am. Soc. Mass Spectrom. 20 (2009) 1148;
(b) Q. Sun, H. Nelson, T. Ly, B.M. Stoltz, R.R. Julian, J. Proteome Res. 8 (2009)
958.
[18] T.W. Chung, F. Turecˇek, J. Am. Soc. Mass Spectrom. 21 (2010) 1279.
[19] (a) C. Lin, P.B. O’Connor, J.J. Cournoyer, J. Am. Soc. Mass Spectrom. 17 (2006)
1605;
(b) P.B. O’Connor, C. Lin, J.J. Cournoyer, J.L. Pittman, M. Belyayev, B.A. Budnik, J.
Am. Soc. Mass Spectrom. 17 (2006) 576;
(c) R.A. Zubarev, D.M. Horn, E.K. Fridriksson, N.L. Kelleher, N.A. Kruger, M.A.
Lewis, B.K. Carpeneter, F.W. McLafferty, Anal. Chem. 72 (2000) 563.
[20] H. Ben Hamidane, D. Chiappe, R. Hartmer, A. Vorobyev, M. Moniatte, Y.O. Tsybin,
J. Am. Soc. Mass Spectrom. 20 (2009) 567.
Acknowledgements
Thanks are due to Dr. Priska von Haller of the University of Wash-
ington Proteomics Resource Center for access to the Thermo LTQ XL