Ultraviolet photodissociation mass spectrometry of bis-aryl hydrazone conjugated peptides

Article abstract of DOI:10.1021/ac9005233

Ultraviolet photodissociation (UVPD) at 355 nm was used to rapidly identify peptides which had been chemically conjugated through bis-aryl hydrazone (BAH) moieties. The two biomolecules of interest were separately tagged to introduce either an aldehyde or

Full text of DOI:10.1021/ac9005233

Anal. Chem. 2009, 81, 4864–4872
Ultraviolet Photodissociation Mass Spectrometry
of Bis-aryl Hydrazone Conjugated Peptides
Myles W. Gardner and Jennifer S. Brodbelt*
Department of Chemistry and Biochemistry, The University of Texas at Austin, 1 University Station A5300,
Austin, Texas 78712-1167
Ultraviolet photodissociation (UVPD) at 355 nm was used
to rapidly identify peptides which had been chemically
conjugated through bis-aryl hydrazone (BAH) moieties.
The two biomolecules of interest were separately tagged
to introduce either an aldehyde or a hydrazine and then
conjugated together through these functional groups to
from the UV-chromogenic BAH-group. In a mock mixture
of peptides, UVPD was used to screen for the BAH-
conjugated peptides in direct infusion ESI-UVPD-MS and
online LC-UVPD-MS methods by comparing the abun-
dances of the ions with the laser off and with the laser
on. Only the BAH-conjugated peptides were observed to
photodissociate upon exposure to UV irradiation, thus
affording excellent selectivity for the pinpointing the
relevant conjugated peptides in a complex mixture of
nonconjugated peptides. UVPD analysis of conjugated
model peptides indicated that the UVPD efficiencies of
these species were charge state dependent. BAH-conju-
gated peptides that had a mobile proton which could
protonate the basic BAH-moiety underwent more efficient
photodissociation than the peptide ions with sequestered
protons. Ultraviolet photodissociation of BAH-cross-
linked peptides also yielded more diagnostic sequence
ions than CID to unambiguously locate the site of
conjugation.
to the hydrazide-modified ligand (or vice versa) forming a UV
chromogenic bis-aryl hydrazone (BAH) bond, which absorbs well
at 354 nm. The coupling of aldehyde functionalities to hydrazides
4
has been used to create many protein-peptide conjugates includ-
ing a follicle-stimulating hormone mutant protein-occludin peptide
5
complex, immunogenic poly glutamic acid-protein conju-
6
7
gates to anthrax, insulin-transferrin complexes, as well as
8
9
oxylipid-protein, small molecule-peptide, and many oligo-
1
0-12
nucleotide-protein complexes.
Analysis of these bioconju-
gates is typically performed using gel electrophoresis or spectro-
photometric methods, but these methods do not provide specific
information about the location of the cross-link. LC-MS/MS
methods are commonly used for more detailed structural analysis
of bioconjugates, usually in conjunction with enzymatic digestion
7
,8,13
to convert large macromolecules into smaller constituents.
Although creating more manageable components, the enzymatic
digestion step also results in more complex mixtures containing
many species that offer no direct information about the conjugation
sites.
As the interest in using protein and peptide conjugates for
therapeutic applications continues to expand, there are several
ongoing analytical challenges related to characterization of the
conjugates, especially in the context of the LC-MS/MS methods
mentioned above that are used to analyze the complex digests
and decipher the conjugation sites. Two of the most significant
challenges include the ability to rapidly distinguish the conjugated
peptides of interest from the uninformative peptides (e.g., dead-
end modified or unmodified peptides) and interpretation of the
product ion (MS/MS) spectra of the cross-linked peptide-peptide
In the past decade the conjugation of proteins and peptides to
other biomolecules has gained widespread attention as a means
to couple the independent functionalities and properties of each
individual component. One area of particular interest has been in
utilizing peptides or small proteins which can pass through
biological membranes as delivery vehicles for peptides, proteins,
(
(
4) http://www.solulink.com, accessed on July 2008.
5) Wong, C.-H.; Mruk, D. D.; Lee, W. M.; Cheng, C. Y. FASEB J. 2007, 21,
438–448
(6) Kubler-Kielb, J.; Liu, T.-Y.; Mocca, C.; Majadly, F.; Robbins, J. B.;
Schneerson, R. Infect. Immun. 2006, 74, 4744–4749
7) Kavimandan, N. J.; Losi, E.; Wilson, J. J.; Brodbelt, J. S.; Peppas, N. A.
Bioconjugate Chem. 2006, 17, 1376–1384
(8) Han, B.; Stevens, J. F.; Maier, C. S. Anal. Chem. 2007, 79, 3342–3354
9) Miranda-Olvera, A. D.; Ferro-Flores, G.; Pedraza-Lopez, M.; Arteaga de
Murphy, C.; De Leon-Rodriguez, L. M. Bioconjugate Chem. 2007, 18, 1560–
1567
10) Bailey, R. C.; Kwong, G. A.; Radu, C. G.; Witte, O. N.; Heath, J. R. J. Am.
Chem. Soc. 2007, 129, 1959–1967
11) Kozlov, I. A.; Melnyk, P. C.; Stromsborg, K. E.; Chee, M. S.; Barker, D. L.;
Zhao, C. Biopolymers 2004, 73, 621–630
.
1
drugs, oligonucleotides, or other therapeutic agents. Cell pen-
.
etrating peptides or protein transduction domains have shown
promise as an effective means for direct transcellular delivery
(
.
2
,3
.
without extensive degradation. One common method to con-
jugate biological species is a two-step protocol in which first each
biomolecule is separately labeled to introduce either aldehyde or
hydrazine moieties. The aldehyde-modified peptide is then linked
(
.
(
(
.
*
To whom correspondence should be addressed. E-mail: jbrodbelt@
.
mail.utexas.edu.
(12) Jarvius, M.; Paulsson, J.; Weibrecht, I.; Leuchowius, K.-J.; Andersson, A.-
C.; Waehlby, C.; Gullberg, M.; Botling, J.; Sjoeblom, T.; Markova, B.;
Oestman, A.; Landegren, U.; Soederberg, O. Mol. Cell. Proteomics 2007,
(
(
1) Torchilin, V. P. Annu. Rev. Biomed. Eng. 2006, 8, 343–375
2) El-Andaloussi, S.; Holm, T.; Langel, U. Curr. Pharm. Des. 2005, 11, 3597–
611
3) Zatsepin, T. S.; Turner, J. J.; Oretskaya, T. S.; Gait, M. J. Curr. Pharm. Des.
005, 11, 3639–3654
.
3
.
6, 1500–1509
(13) Negishi, J.; Shirahama, T.; Nagaoka, Y.; Takemoto, Y.; Uesato, S. Yakugaku
Zasshi 2008, 128, 1317–1323
.
(
2
.
.
4864 Analytical Chemistry, Vol. 81, No. 12, June 15, 2009
10.1021/ac9005233 CCC: $40.75  2009 American Chemical Society
Published on Web 05/18/2009

complexes to identify the site of conjugation. Several mass
spectrometric strategies have been developed to aid in the
identification of cross-linked peptides, with much work devoted
to the design and synthesis of novel cross-linkers. Some methods
have focused on distinguishing conjugated peptides from unmodi-
fied ones by incorporating affinity tags into the linkers to enrich
the same phosphate chromophore and thus would efficiently
dissociate upon IR irradiation.
2
8
In this work, ultraviolet photodissociation at 355 nm in a
quadrupole ion trap mass spectrometer is used to selectively
photodissociate only the chromogenic bis-aryl hydrazone conju-
gated peptides. In recent years, UVPD has been utilized as an
29-32
1
4-17
alternative to CID as an ion activation technique for peptides.
the peptides of interest in complex mixtures
or using
Compared to CID, UVPD in ion traps affords several advantages
such as secondary dissociation of primary product ions and a
potential reduction in ion losses due to scattering since photo-
dissociation does not alter ion trajectories. Moreover, as a
nonresonant process, UVPD can be performed at a lower rf
trapping voltage than CID which facilitates the detection of
diagnostic low-mass product ions. Photodissociation methods have
also been used as a screening tool to selectively dissociate specific
chromogenic species, such as phosphorylated peptides using
isotopically labeled cross-linking reagents to pinpoint conjugated
15,18,19
peptide ions in the mass spectra,
introducing a reporter ion
which is observed upon low-energy collision induced dissociation
2
0
(
CID), or utilizing fluorogenic cross-linkers to differentiate
between conjugated and unmodified peptides during a front-end
2
1-23
LC separation.
Other groups have aimed at simplifying
interpretation of the product ion spectra of intermolecularly cross-
linked peptides by incorporating gas-phase labile bonds which can
be cleaved upon low-energy CID, thus yielding two modified
peptides that can then be individually sequenced by a second stage
28,33,34
35
IRMPD at 10.6 µm,
containing peptides at 220 and 262 nm.
disulfide bonds at 157 nm, and tyrosyl-
3
6,37
In our work, UVPD is
1
4,24-26
of CID.
used to screen the chromogenic BAH-conjugated peptides in
complex mixtures, as demonstrated for a mixture of peptides,
using both direct infusion ESI-MS and online LC-UVPD-MS
techniques. Upon UV irradiation at 355 nm, only the BAH-
conjugated peptides are shown to undergo photodissociation. In
contrast, the nonconjugated aldehyde- and hydrazine-modified
peptides or other non-BAH peptides do not absorb at 355 nm and
thus do not photodissociate, offering a high degree of selectivity
for the rapid identification of the conjugated peptides of interest.
Each of these techniques, however, only focuses on one of
the two challenges mentioned above: either distinguishing cross-
linked peptides from uninformative ones in mixtures or sequenc-
ing the cross-linked peptides of interest. We previously reported
an IR chromogenic cross-linker which contained a phosphate
chromophore that promoted the selective infrared multiphoton
dissociation (IRMPD) of cross-linked peptides for differentiation
27
of cross-linked peptides from unmodified ones. IRMPD of these
cross-linked peptides also yielded a series of y-type fragment ions
C-terminal to the cross-linked lysine residues which allowed the
two constituent peptides to be sequenced without the need for
EXPERIMENTAL SECTION
n
Chemicals and Reagents. The peptides Ac-RFMWMK-NH
2
,
MS methods. However, this technique, as well as those
R-MSH (Ac-SYSMEHFRWGKPV-NH ), neurotensin (Pyr-
2
described above, suffered in distinguishing between cross-
linked peptides of interest and dead-end modified peptides
LYENKPRRPYIL), and substance P were purchased from
Bachem (Torrance, CA). The peptide angiotensin II was
obtained from AnaSpec (San Jose, CA). The peptides brady-
kinin, YGGFM, and disuccinimidyl suberate (DSS) were
purchased from Sigma (St. Louis, MO). The peptides Ac-
AAAKAAAAR and Ac-AAAKPAAAR were synthesized at the
Protein Microanalysis Facility at the University of Texas at
Austin and purified by reversed-phase HPLC. Benzaldehyde
and 2-hydrazinopyridine were obtained from Aldrich (Milwau-
kee, WI). Succinimidyl 4-hydrazinonicotinate acetone hydra-
zone (SANH) and succinimidyl 4-formylbenzoate (SFB) were
from Pierce Biotechnology (Rockford, IL). All other chemical
and solvents (HPLC grade) were purchased from Fisher
Scientific (Fairlawn, NJ).
(
e.g., peptides that have reacted with the cross-linker but
the other reactive site of the cross-linker has been hydro-
lyzed) which generally are not structurally informative. In
addition, phosphorylated peptides could be initially falsely
classified as IRCX-cross-linked peptides as they also contain
(
(
(
(
(
(
14) Chowdhury, S. M.; Munske, G. R.; Tang, X.; Bruce, J. E. Anal. Chem. 2006,
8, 8183–8193
15) Chu, F.; Mahrus, S.; Craik, C. S.; Burlingame, A. L. J. Am. Chem. Soc. 2006,
28, 10362–10363
16) Sinz, A.; Kalkhof, S.; Ihling, C. J. Am. Soc. Mass Spectrom. 2005, 16, 1921–
931
17) Hurst, G. B.; Lankford, T. K.; Kennel, S. J. J. Am. Soc. Mass Spectrom. 2004,
5, 832–839
18) Mueller, D. R.; Schindler, P.; Towbin, H.; Wirth, U.; Voshol, H.; Hoving,
S.; Steinmetz, M. O. Anal. Chem. 2001, 73, 1927–1934
7
.
1
.
1
.
1
.
.
19) Ihling, C.; Schmidt, A.; Kalkhof, S.; Schulz, D. M.; Stingl, C.; Mechtler, K.;
(28) Crowe, M. C.; Brodbelt, J. S. Anal. Chem. 2005, 77, 5726–5734
(29) Reilly, J. P. Mass Spectrom. Rev. 2009, 28, 425–447
(30) Wilson, J. J.; Brodbelt, J. S. Anal. Chem. 2007, 79, 7883–7892
(31) Thompson, M. S.; Cui, W.; Reilly, J. P. J. Am. Soc. Mass Spectrom. 2007,
18, 1439–1452
(32) Oh, J. Y.; Moon, J. H.; Lee, Y. H.; Hyung, S.-W.; Lee, S.-W.; Kim, M. S.
Rapid Commun. Mass Spectrom. 2005, 19, 1283–1288
(33) Crowe, M. C.; Brodbelt, J. S. J. Am. Soc. Mass Spectrom. 2004, 15, 1581–
1592
(34) Flora, J. W.; Muddiman, D. C. Anal. Chem. 2001, 73, 3305–3311
(35) Fung, Y. M. E.; Kjeldsen, F.; Silivra, O. A.; Chan, T. W. D.; Zubarev, R. A.
Angew. Chem., Int. Ed. 2005, 44, 6399–6403
(36) Lemoine, J.; Tabarin, T.; Antoine, R.; Broyer, M.; Dugourd, P. Rapid
Commun. Mass Spectrom. 2005, 20, 507–511
(37) Joly, L.; Antoine, R.; Broyer, M.; Dugourd, P.; Lemoine, J. J. Mass Spectrom.
2007, 42, 818–824
.
Haack, M.; Beck-Sickinger, A. G.; Cooper, D. M. F.; Sinz, A. J. Am. Soc.
.
Mass Spectrom. 2006, 17, 1100–1113
20) Back, J. W.; Hartog, A. F.; Dekker, H. L.; Muijsers, A. O.; de Koning, L. J.;
de Jong, L. J. Am. Soc. Mass Spectrom. 2001, 12, 222–227
21) Sinz, A.; Wang, K. Biochemistry 2001, 40, 7903–7913
22) Sinz, A.; Wang, K. Anal. Biochem. 2004, 331, 27–32
23) Wine, R. N.; Dial, J. M.; Tomer, K. B.; Borchers, C. H. Anal. Chem. 2002,
4, 1939–1945
24) Soderblom, E. J.; Goshe, M. B. Anal. Chem. 2006, 78, 8059–8068
25) Tang, X.; Munske, G. R.; Siems, W. F.; Bruce, J. E. Anal. Chem. 2005, 77,
11–318
26) Lu, Y.; Tanasova, M.; Borhan, B.; Reid, G. E. Anal. Chem. 2008, 80, 9279–
287
27) Gardner, M. W.; Vasicek, L. A.; Shabbir, S.; Anslyn, E. V.; Brodbelt, J. S.
Anal. Chem. 2008, 80, 4807–4819
.
.
(
.
.
(
(
(
.
.
.
7
.
.
(
(
.
.
3
.
.
(
(
9
.
.
.
.
Analytical Chemistry, Vol. 81, No. 12, June 15, 2009 4865

ford, MA), which were both controlled by the Millennium version
.00 software package. Reversed-phase HPLC was accomplished
using a Symmetry300 C18 column (2.1 mm × 50 mm, 3.5 µm
packing) using a gradient elution consisting of (A) H O with
.1% formic acid and (B) acetonitrile with 0.1% formic acid at a
3
2
0
flow rate of 0.300 mL/min. The mobile phase was held at 95%
A for 2 min after sample injection (typically 10 µL of ∼10 µM
solutions) and then linearly reduced to 60% over 60 min.
Analytes were detected by UV-vis absorption spectroscopy at
3
55 nm as well as by mass spectrometry.
Mass Spectrometry and Ultraviolet Photodissociation.
Mass spectrometry experiments were performed on a modified
ThermoFinnigan LCQ Deca XP quadrupole ion trap mass spec-
trometer (San Jose, CA) using the XCalibur version 1.3 software
3
0
package as previously described. For all direct infusion experi-
ments, the solutions were infused at 2.5 µL/min using an ESI
voltage of 4.5 kV. Ultraviolet photodissociation (UVPD) was
performed with a Spectra-Physics Quanta-Ray GCR-11 Nd:YAG
laser with a HG-2 harmonics generator (Mountain View, CA) to
produce the tripled harmonic photons at 355 nm. Briefly, a CF
viewport flange was welded to the vacuum manifold of the LCQ
Deca XP instrument and fitted with a quartz window to allow the
transmission of UV radiation. The unfocused laser beam was
aligned on-axis with a 5 mm hole drilled through the ring electrode
to irradiate the ion cloud. For all direct infusion UVPD-MS
experiments, the laser was operated at full power (∼65 mJ/pulse),
unless otherwise noted, at a repetition rate of 10 Hz in the
Q-switched mode. The LCQ instrument TTL signal output from
pin TP 15-3 Sync_TP was used to trigger a model 505-2C low-
voltage pulse generator (Berkeley Nucleonics Corp., San Rafael,
CA) which in turn was used to pulse the laser. Typically, the
activation time was set to 1000 ms to produce 10 laser pulses
unless indicated otherwise. The instrument control files were
modified to allow triggering of the laser during LC-MS/MS
Figure 1. Structures of the modifying reagents (A) succinimidyl
4-hydrazinonicotinate acetone hydrazone (SANH), (B) succinimidyl
4-formylbenzoate (SFB), as well as the (C) bis-aryl hydrazone cross-
link and observed bond cleavages. The solution-phase molar ab-
4
sorptivity of the BAH-cross-link is reported. The numbering scheme
R
in part C refers to the number of atoms away from C of the peptide
38
backbone of the R-peptide.
Synthesis of Benzaldehyde-2-pyridinylhydrazone and
UV-Visible Measurements. The ultraviolet chromophore ben-
zaldehyde-2-pyridinylhydrazone (B2PH) was synthesized by dis-
solving 1.0 mmol of 2-hydrazinopyridine in 2.0 mL of MeOH and
then 102 µL (1.0 mmol) of benzaldehyde was added. The solution
was heated to 55 °C and stirred for 16 h. The reaction mixture
was then cooled to room temperature, and the yellow-orange
precipitate of B2PH was vacuum filtered. The final product was
redissolved in MeOH and then diluted to 50 µM for downstream
analysis. All UV-vis measurements were obtained on an Agilent
8453 spectrophotometer (Santa Clara, CA).
z
analyses. Precursor ions were activated at a q -value of 0.25 for
Chemical Cross-Linking of Model Peptides. Model peptides
CID and 0.10 for UVPD. Energy-variable CID experiments were
performed by stepping-up the CID voltage to determine the
voltage necessary to dissociate the precursor ion to 50% relative
were cross-linked together via a two-step conjugation method. All
peptides were dissolved to 5.0 mM in H O, and the modifying
2
reagents SANH and SFB (parts A and B of Figure 1) were each
freshly prepared at 20.0 mM in DMSO. Each peptide was
individually modified at its primary amines (e.g., free N-termini
or lysine residues) by either SANH or SFB in 20 mM HEPES,
pH 8.0 buffer at a 1:10 molar ratio of peptide to modifier (200
µM:2.0 mM). The reactions were allowed to proceed at room
temperature overnight to ensure complete modification of the
peptides; the product mixture was then desalted using Waters
tC18 SepPak cartridges (Milford, MA), and peptides were eluted
with 100% MeOH. The two modified peptides were then
conjugated together at an equimolar ratio (∼100 µM) in 99:1
MeOH/HOAc acid at 55 °C overnight forming a bis-aryl
hydrazone cross-link (Figure 1C). The conjugated peptides were
27
abundance as previously described. Online LC-UVPD-MS was
performed by either isolating ions of m/z 450-1200 without
any form of activation or by subjecting all ions of m/z 450-1200
to 10 pulses of UV irradiation at 355 nm. In order to acquire
an isolation spectrum, a precursor ion of interest was selected
for activation at a normalized collision energy of 0.0% and the
isolation waveform was turned off, allowing ions to be effectively
isolated and detected.
RESULTS AND DISCUSSION
Bis-aryl Hydrazone Chromophore and UVPD Screening
Strategy. The conjugation strategy employs a two-step conjugation
method in which the two biomolecules of interest are first tagged
by either SANH that introduces a hydrazine group or by SFB
which incorporates an aromatic aldehyde (Scheme 1). The two
modified biomolecules are then cross-linked together through a
specific reaction between the hydrazine and aldehyde to form a
bis-aryl hydrazone. As shown in this study, the two-stage conjuga-
tion reaction affords a high degree of UVPD selectivity when
analyzing mixtures containing unmodified, dead-end modified, and
conjugated peptides. Ultraviolet photodissociation can be used to
subsequently diluted to ∼10 µM in 49.5:49.5:1 H
2
O/MeOH/
HOAc (v/v/v) for ESI-MS analysis without any further
purification.
Analytical High-Performance Liquid Chromatography.
Liquid chromatography was performed with a Waters 2695
Separations Module and 486 Tunable Absorbance Detector (Mil-
(38) Schilling, B.; Row, R. H.; Gibson, B. W.; Guo, X.; Young, M. M. J. Am. Soc.
Mass Spectrom. 2003, 14, 834–850.
4866 Analytical Chemistry, Vol. 81, No. 12, June 15, 2009

Scheme 1. Two-Step Conjugation and UVPD Screening Strategy
rapidly screen for BAH-conjugated peptides in which only the
peptide ions that readily absorb radiation at 355 nm dissociate.
Unmodified peptides, as well as those with dead-end SFB- or
SANH-modifications, will not photodissociate as they do not
contain the critical BAH-chromophore. The selective photodisso-
ciation of all BAH-cross-linked peptides allows ready differentiation
between diagnostic conjugated peptides and unmodified ones as
well as uninformative dead-end modified species.
Prior to conducting peptide conjugation experiments with the
SANH and SFB reagents, the chromophore of the bis-aryl
hydrazone cross-link was duplicated in a model compound by
synthesis of benzaldehyde-2-pyridinylhydrazone (B2PH, Figure
observed. In fact, the maximum peak absorption shifted from 325
to 347 nm, and the molar absorptivity at 355 nm (the wavelength
that corresponds to the tripled harmonic of a Nd:YAG laser)
increased from 6 600 L mol- cm to 19 000 L mol cm .
Neither of the two reagents, SANH nor SFB, exhibited high
absorbance near 355 nm, suggesting that peptides containing
these modifications would not undergo efficient UVPD. Al-
though the UV absorption profiles of B2PH in solution are not
expected to perfectly mirror the UV absorption profiles in the
gas phase, these initial results do suggest that the bis-aryl
hydrazone moiety exhibits a strong dependence on whether it
is protonated or not, and the high absorptivity at 355 nm in
solution shows remarkable agreement with our gas-phase
UVPD results. ESI-MS of B2PH yielded a single peak of m/z
198.1 corresponding to the protonated molecule. Benzaldehyde-
2-pyridinylhydrazone was subjected to UVPD-MS, and upon two
pulses of 355 nm radiation at ∼18 mJ/pulse (∼28% power),
the abundance of the precursor ion decreased by 60% and two
abundant product ions were observed of m/z 94 and of m/z 67
1
-1
-1
-1
2B) and characterized by UV-vis spectroscopy and UVPD-MS.
The UV-vis absorbance spectrum of this molecule at 50 µM in
MeOH did not reveal an absorption band around 355 nm, the
wavelength of a tripled Nd:YAG laser, as shown in Figure 2A. To
convert B2PH to its protonated form, acetic acid (1% v/v) was
added to the solution and a red-shift in the absorption profile was
(Figure 2B). The efficient UV photodissociation of B2PH at low
laser powers suggested that bis-aryl hydrazone cross-linked
peptides should also undergo UVPD.
For the spectra discussed in this report, all product ions of the
conjugated or cross-linked peptides were labeled according to
38,39
previously suggested nomenclature.
Briefly, the two constituent
peptides are labeled R and ꢀ, and dissociation along the backbone
of the R-peptide yields product ions such as ynR or bnR. Lysine
L
immonium ions linked to the other peptide are labeled as K R
where L refers to the linker, and fragment ions due to dissociation
of the linker moiety are labeled as LRn referring to cleavage n
bonds away from the C atom of the R-peptide (see Figure 1C for
R
observed bond cleavages and numbering scheme). Similar nomen-
clature is used for product ions of the ꢀ-peptide.
UVPD-MS of BAH-Conjugated Model Peptides. The pep-
tides Ac-AAAKAAAAR (R-peptide) and neurotensin (ꢀ-peptide) were
chosen as model peptides to evaluate the feasibility of the UVPD
strategy, and each were modified by SFB or SANH, respectively,
Figure 2. (A) UV-vis spectra of 50 µM benzaldehyde-2-pyridinyl-
hydrazone in MeOH (dashed gray line) and in 99/1 MeOH/HOAc
(v/v) (solid black line). (B) Structure of benzaldehyde-2-pyridinylhyra-
(39) Gaucher, S. P.; Hadi, M. Z.; Young, M. M. J. Am. Soc. Mass Spectrom. 2006,
zone and product ions observed by UVPD-MS.
17, 395–405
.
Analytical Chemistry, Vol. 81, No. 12, June 15, 2009 4867

Figure 3. (A) Wide m/z isolation and (B) UVPD mass spectra (10 pulses) of m/z 600-1100 of the reaction mixture of Ac-AAAKAAAAR +
SFB/SANH + neurotensin. The first peptide listed is referred to as R and the second peptide as ꢀ. BAH-cross-links are represented by X.
and then cross-linked together. The mass spectrum (with a wide
isolation range of m/z 600-1200) of the conjugated product mixture
is shown in Figure 3A in which each of the two modified peptides
were detected, corresponding to a mass addition of 132 Da for SFB
and 135 Da of SANH (after deprotection of the acetone moiety), as
well as the conjugated peptide-peptide complex in the 3+ and 4+
charge states with the BAH-linker adding 249 Da. In addition,
unmodified neurotensin (2+) and fragment ions of the R-peptide
conjugated to neurotensin were observed. (These latter ions are
attributed to in-source dissociation due to the high tube lens voltage
necessary to effectively transmit ions within the isolated m/z range
of 600-1200 and were not observed when trapping the entire range
from m/z 300 to 2000 as shown in the Supporting Information, Figure
S-1; no ions were observed above m/z 1400. The necessity of using
a high tube lens voltage is an artifact of the limited flexibility of the
LCQ software when implementing a simultaneous broad trapping
and activation experiment as opposed to conventional CID in which
a selected precursor ion is isolated and activated.) The cross-link
was formed between Lys-4 of Ac-AAAKAAAAR and Lys-6 of neuro-
tensin as each peptide only has a single primary amine that can react
with the SANH and SFB reagents. Upon 10 pulses of UV irradiation
of all ions of m/z 600-1200, only the ions which contained the BAH-
linker underwent photodissociation and were reduced in abundance
and SFB-modified peptides were individually examined by UVPD-
MS to determine if these ions would photodissociate. Neurotensin
modified by SANH was examined in the 2+ and 3+ charge states,
and both doubly charged SFB-modified neurotensin and R-MSH
were analyzed (Supporting Information, Figure S-2). For all four
of these species, the intact surviving precursor ion is the only ion of
significant abundance observed after 10 pulses of UV irradiation at
full power, confirming their lack of UV absorptivity which confers
selectivity to the UVPD strategy.
After demonstration that UVPD can be used to rapidly screen
BAH-conjugated peptides in simple peptide mixtures, UVPD was
then compared to CID to determine its utility for sequencing the
two constituent peptides and locating the site of conjugation.
UVPD of [R-MSH X BAH X Ac-RFMWMK-NH2 + 4H]4 (m/z
714.3) yielded a full series of conventional b-ions of the
+
ꢀ-peptide (Ac-RFMWMK-NH ), as well as several product ions
2
due to dissociation of the backbone of the R-peptide (R-MSH)
(Figure 4A). In addition, the cross-linker specific fragment ions
L
of K R and LR18 were observed. The presence of the b10R ion
without the BAH-linker and the adjacent b11R ion retaining the
cross-link allows one to locate the site of conjugation to Lys-11
of the R-peptide. Similarly the b5ꢀ and y1ꢀ ions can be used to
pinpoint the cross-link to Lys-6 of the ꢀ-peptide. Unambiguous
determination of the cross-link location is necessary even with
a priori knowledge of the peptide sequence, as recent work
has identified that NHS-esters not only react with lysine
residues and free primary amines but also with tyrosines,
(Figure 3B). By comparing the relative abundances of each ion in
the two spectra, one can rapidly determine which ions correspond
to BAH-conjugated peptides. The unmodified doubly charged neu-
rotensin ion actually increased in ion abundance upon UV irradiation,
most likely due to dissociation of the conjugated peptide-peptide
complexes yielding the neurotensin peptide ion as a product ion. Of
particular importance is that the dead-end modified peptides, [ꢀ +
4
0
serines, and threonines. CID of this conjugated peptide ion
4
+
([R-MSH X BAH X Ac-RFMWMK-NH + 4H] of m/z 714.3)
2
did not produce as many informative sequence fragment ions;
SANH + 3H]3 of m/z 603.5 and [R + SFB + H] of m/z 974.5
did not decrease in abundance. Previous work in our lab utilized
an IR chromogenic cross-linker and IRMPD to selectively dis-
sociate cross-linked peptides, but the uninformative dead-end
modified peptides also contained the phosphate chromophore and
would photodissociate as well, thus preventing the distinction of
+
+
the most dominant pathway was dehydration (Figure 4B). Only
a total of three product ions were detected, and while the b5ꢀ and
y1ꢀ ions can be used to unambiguously assign the location of
L
conjugation to Lys-6 of the ꢀ-peptide, the other fragment K R
does not provide any sequence information. For this conjugated
peptide-peptide complex, UVPD clearly yields a more informa-
tive product ion spectrum with diagnostic fragment ions that
can be used to pinpoint the cross-link location on both
27
cross-linked peptides from dead-end modified peptides. Since
the BAH-chromophore is formed only upon specific conjugation
of SFB- and SANH-modified peptides, the dead-end modified
peptides do not efficiently photodissociate at 355 nm. Both SANH-
(40) Kalkhof, S.; Sinz, A. Anal. Bioanal. Chem. 2008, 392, 305–312
.
4868 Analytical Chemistry, Vol. 81, No. 12, June 15, 2009

Figure 4. (A) UVPD (10 pulses) and (B) CID (0.344 V) mass spectra of [R-MSH X BAH X Ac-RFMWMK-NH
2
+ 4H]4 of m/z 714.3. The first
+
peptide listed is referred to as R and the second peptide as ꢀ. Precursor ions are signified with an asterisk (*). The schematic of bond cleavages
refers to product ions observed in the UVPD mass spectrum.
Figure 5. (A) Wide m/z range isolation and (B) UVPD (10 pulses) mass spectra of m/z 600-1100 of a mock mixture of BAH-cross-linked (X),
SFB- (ꢁ), SANH- (ψ), and unmodified peptides (∆). Ion identities are listed in Table 1.
constituent peptides, whereas this peptide predominantly dis-
sociated through a dead-end neutral loss pathway by CID.
Screening of BAH-Conjugated Peptides in Mock Mixtures.
To demonstrate the selective UVPD of BAH-conjugated peptides
in a more complex mixture, a mock mixture of peptides was
analyzed. The mock mixture contained an equimolar amount (∼10
µM) of bradykinin, angiotensin II, neurotensin, substance P,
YGGFM, and the cross-linking product mixtures of Ac-AAA-
KAAAAR + SFB/SANH + Ac-AAAKPAAAR, Ac-AAAKAAAAR +
SFB/SANH + neurotensin, and R-MSH + SANH/SFB + Ac-
unaffected by UV irradiation and were observed in both the
isolation and UVPD mass spectra. The more highly charged BAH-
conjugated peptide-peptide complexes exhibited high UVPD
efficiency and their abundances substantially decreased in the
UVPD mass spectra (Figure 5B). The abundances of these highly
protonated BAH-conjugated peptides diminished on average by
∼70%, whereas the abundances of the lower-charged conjugated
peptide ions decreased by only ∼20-30% (Table 1). The SFB-,
SANH-, and unmodified peptides did not exhibit any significant
decrease in abundance upon exposure to UV irradiation at 355
nm. The differences in ion abundances of these various species
between the wide m/z isolation and UVPD mass spectra could
be discerned with as few as 5 pulses of UV irradiation and up to
20 pulses, thus indicating a wide usable range of experimental
2
RFMWMK-NH . This mock mixture of peptides was rapidly
screened via direct infusion ESI-UVPD-MS by comparison of
a wide m/z range isolation spectrum and the UVPD mass
spectrum (Figure 5). Ions without the BAH-chromophore were
Analytical Chemistry, Vol. 81, No. 12, June 15, 2009 4869

Table 1. Changes in Relative Ion Abundances upon UV Irradiation (10 Pulses) for a Mixture of Peptides Analyzed by
Direct Infusion ESI-MS/MSa
peak area
peak area
peak area
percent change
std
deviation
b
b
b
c
label
species
m/z
isolation
UVPD
+
5
5
1
2
3
4
5
[R-MSH + SANH + 3H]3
600.5
647.4
654.1
666.6
674.4
687.5
692.0
710.1
714.3
769.5
811.3
836.9
922.0
951.5
974.3
980.3
1003.3
1020.2
1046.3
1060.3
1068.2
1086.2
1347.3
3.41 × 10
3.25 × 10
-4.8
-70.1
-74
-68
-4
-78
-83
-89
-86
-77
+44
+8
0.9
0.6
2
4
+
d
5
4
[Ac-AAAKAAAA X BAH X neurotensin + 4H]
[Ac-AAAKAAAAR X BAH X Ac-AAAKPAAAR + 3H]
2.36 × 10
7.06 × 10
3
+
5
5
4.68 × 10
1.24 × 10
3
+
5
4
[R-MSH X BAH X K + 3H]
2.45 × 10
7.87 × 10
2
2
+
5
5
[substance P + 2H]
Ac-AAAKAAAAR X BAH X neurotensin - H
8.00 × 10
7.71 × 10
3
4
+
4
4
[
2
O + 4H]
7.81 × 10
1.70 × 10
8
4
+
4
4
6
7
8
9
1
1
1
1
1
[Ac-AAAKAAAAR X BAH X neurotensin + 4H]
7.03 × 10
1.19 × 10
6
4
+
5
4
[R-MSH X BAH X Ac-RFMWMK-NH
[R-MSH X BAH X Ac-RFMWMK-NH
2
- H
2
O + 4H]
2.08 × 10
2.21 × 10
4
4
+
5
4
2
+ 4H]
1.06 × 10
1.46 × 10
1
3
+
d
4
4
[Ac-AAAK X BAH X neurotensin + 3H]
8.73 × 10
2.05 × 10
2
+
d
4
4
0
1
2
3
4
[Ac-RFMWM + H
2
O + H]
6.50 × 10
9.38 × 10
9
2
+
5
5
[neurotensin + 2H]
2.11 × 10
2.27 × 10
3
3
+
5
5
[Ac-AAAKAAAAR X BAH X neurotensin + 3H]
1.55 × 10
1.29 × 10
-17
-19
-3
4
3
+
4
4
[R-MSH X BAH X Ac-RFMWMK-NH
2
+ 3H]
6.19 × 10
4.99 × 10
2
+
5
5
[Ac-AAAKAAAAR + SFB + H]
1.21 × 10
1.17 × 10
4
2
+
4
4
[
Ac-AAAKAAAAR X BAH X Ac-AAAKPAAAR + 2H]
3.29 × 10
2.34 × 10
-29
+68
+1
4
+
4
4
1
1
5
6
[Ac-AAAKPAAAR + SANH + H]
5.02 × 10
8.43 × 10
13
5
[Ac-AAAKPAAAR + SANH + H
2
O + H]+
4.29 × 10
4
4.33 × 10
4
+
3
3
[
[
angiotensin II + H]
9.24 × 10
8.96 × 10
-3
4
+
3
4
bradykinin + H]
9.22 × 10
1.05 × 10
+13.8
0
0.5
7
+
4
4
1
7
[Ac-RFMWMK-NH
2
2
+ SFB + H]
+ SFB + H
4.41 × 10
4.41 × 10
[
[
Ac-RFMWMK-NH
2
O + H]+
1.81 × 10
4
1.87 × 10
4
+3
+2.8
8
+
3
3
substance P + H]
6.62 × 10
6.80 × 10
0.7
a
b
Ion abundances are relative to the total ion abundance in the isolation or UVPD spectrum. Values are the average of data obtained in triplicate.
c
d
Standard deviation of the percent abundance changes. Ion is isobaric with a product ion of a cross-linked peptide.
conditions for distinguishing between BAH-conjugated peptides
and uninformative species.
contains the BAH-chromophore. Only by utilizing UVPD can one
differentiate between two or more coeluting species that absorbed
well at 355 nm.
To demonstrate the utility of UVPD to screen BAH-conjugated
peptides in more complex samples, the same mock mixture was
also analyzed by online LC-UVPD-MS. In this method, a wide m/z
isolation mass spectrum and then a UVPD mass spectrum were
acquired in an alternating fashion throughout the LC separation.
Selected ion chromatograms (SICs) of the eluting species were
extracted from the total ion chromatograms of the isolation and
UVPD mass spectra. These SICs were then replotted on the same
time axis to create reconstructed ion chromatograms (RIC) as
shown in Figure 6. The peak areas of each component in the
mixture were measured in the isolation and UVPD SICs to
calculate the decreases in ion abundance upon UV irradiation
It should be noted that another type of dead-end modified
peptides were observed in some of the UVPD-MS experiments
corresponding to a reaction between free SANH (in which the
NHS-ester was hydrolyzed) with SFB-modified peptides or vice
versa creating a chromogenic dead-end modified peptide,
termed BAH-OH (+267 Da mass shift). These peptides exhib-
ited high UVPD efficiencies but lower than that of the BAH-
conjugated peptides. A UVPD mass spectrum of [R-MSH +
3
+
BAH-OH + 3H] is shown in the Supporting Information
,
Figure S-3, and several informative product ions were observed
allowing identification of the precursor as a BAH-OH modified
peptide. In fact, one product ion observed was the unmodified
(
given as percentages) (Table 2). Comparison of the two RICs
2
+
allow ready determination of which ions possess the BAH-
chromophore as these species decreased in abundance by an
average of 73%. If the BAH-containing peptide ions are further
categorized by charge state, the more highly charged species
reduce in abundance by ∼94 ± 4% and the species in the
corresponding lower charge states (i.e., without a mobile proton)
decrease by ∼43 ± 25%. The nonchromogenic peptide ions
exhibited an average decrease in peak area of only 0.7%.
UV-vis absorption measurements at 355 nm of the eluting
peptides also provided another means to aid in distinguishing the
BAH-conjugated peptides from unmodified peptides and dead-end-
modified peptides. All of the peptides with a BAH-cross-link had
an absorbance of at least 0.029; unmodified, SFB-modified, and
SANH-modified peptides showed no absorbance at 355 nm (Table
peptide, [R-MSH + 2H] , of m/z 833.5 which corresponds
to the loss of the BAH-OH modification. Since these species
do undergo relatively efficient photodissociation, they could
yield false positives in the isolation/UVPD screening pro-
cess. However, these dead-end modified peptides can be
reduced to a minimum through rigorous sample cleanup to
remove any excess free SANH or SFB from the reaction
mixtures prior to MS analysis; in addition, these species were
observed to increase in concentration in the samples over
time and thus it is recommended that the samples be
analyzed by ESI-UVPD-MS as soon as possible after the
reactions are complete.
Charge State Dependence of UVPD. Upon closer examina-
tion of the decreases in ion abundances when comparing the
wide m/z isolation mass spectra to the UVPD mass spectra in
the direct infusion ESI-MS and LC-UVPD-MS experiments, it
is clear that the more highly charged BAH-conjugated peptide
ions underwent more efficient photodissociation than those in
2). However, for coeluting species in which one peptide is
conjugated and the other is not, the UV-vis measurements in
combination with a standard mass spectrum would not provide
enough information to determine which of the two species
4870 Analytical Chemistry, Vol. 81, No. 12, June 15, 2009

Figure 6. Reconstructed total ion chromatograms from (A) wide m/z range isolation and (B) UVPD (10 pulses) spectra of LC-MS/MS analysis
of a mixture of peptides. Ions of m/z 475-1100 were either isolated or UV irradiated. BAH-conjugated peptides are labeled with X, SFB-
modified with ꢁ, and SANH-modified with ψ. Ion identities are listed in Table 2.
Table 2. Changes in Ion Abundance upon UV Irradiation from LC-UVPD-MS Reconstructed Ion Chromatograms of a
Mixture of Peptides
elution
label (min)
absorbance peak area peak area percent change
std
deviation
a
,
b
a
a
a
c
species
m/z (355 nm)
isolation
UVPD
in peak area
+
6
6
5
6
5
4
6
6
5
4
5
3
4
4
4
6
1
2
3
4
5
6
7
16.3
16.5
16.7
18.1
18.1
19.1
21.1
[Ac-AAAKAAAAR + SFB + H]
488.5
574.8
531.3
653.9
980.7
524.1
558.4
837.3
600.6
674.6
691.8
922.2
714.6
951.4
536.2
n/d
1.59 × 10
8.27 × 10
1.68 × 10
1.11 × 10
3.49 × 10
1.61 × 10
4.44 × 10
1.63 × 10
9.73 × 10
6.77 × 10
3.84 × 10
1.04 × 10
2.48 × 10
2.49 × 10
1.10 × 10
1.45 × 10
8.63 × 10
1.54 × 10
1.07 × 10
1.15 × 10
1.60 × 10
4.38 × 10
1.75 × 10
9.52 × 10
6.65 × 10
2.91 × 10
7.44 × 10
1.36 × 10
1.89 × 10
1.11 × 10
-8
+5
10
9
+
5
6
6
4
6
6
5
4
5
5
5
5
4
6
[YGGFM + H]
0.0002
n/d
2
+
[bradykinin + 2H]
-8
1
3
2
+
+
[Ac-AAAKAAAAR X BAH X Ac-AAAKPAAAR + 3H]
[Ac-AAAKAAAAR X BAH X Ac-AAAKPAAAR + 2H]
0.0372
-90
-67
-0.2
-1
+8
-5
-0.4
-99.4
-23
-94.5
-21
+0.1
3
14
12
3
2
+
[angiotensin II + 2H]
0.0007
0.0019
3
2
+
+
[neurotensin + 3H]
[neurotensin + 2H]
2
1.1
10
21
13
0.6
23
0.7
16
7
3
+
8
9
1
1
1
22.8
24.3
25.6
25.6
32.1
[R-MSH + SANH + 3H]
n/d
2
+
[substance P + 2H]
n/d
4
3
+
+
0
1
2
[Ac-AAAKAAAAR X BAH X neurotensin + 4H]
[Ac-AAAKAAAAR X BAH X neurotensin + 3H]
[R-MSH X BAH X Ac-RFMWMK-NH
[R-MSH X BAH X Ac-RFMWMK-NH
2
[Ac-RFMWMK-NH + SFB + 2H]
0.0488
4
3
+
+
2
+ 4H]
+ 3H]
0.0298
n/d
3
2.1
2
2
+
1
3
33.6
a
Percent change calculated from the difference in the peaks areas of the SICs from the isolation and UVPD spectra relative to the isolation
b
c
peak area; values are the average of data obtained in triplicate. n/d ) no peak detected. Standard deviation of the percent abundance changes
obtained in triplicate.
lower charge states. The quadruply protonated BAH-conjugated
peptide complex of m/z 692.0 in Figure 3 corresponding to [Ac-
AAAKAAAAR X BAH X neurotensin + 4H]4 decreased in
relative abundance by ∼90% whereas the triply charged
analogue was reduced by only ∼20%. This phenomenon was
further investigated by conjugating the model peptides Ac-
AAAKAAAAR + SFB with Ac-AAAKPAAAR + SANH and
individually analyzing the doubly- and triply charged cross-
linked products by UVPD-MS. Ultraviolet photodissociation
of the resulting conjugated peptide in the 3+ charge state
yielded a wide array of product ions as evident in Figure
one peptide (e.g., KLR) and those arising from cleavage of
the linker amide bond (e.g., LR18) were observed. In
addition, a full series of conventional b- and y-type ions were
also detected for each constituent peptide. However, UVPD
of the analogous doubly protonated conjugated peptide
resulted in minimal product ion formation (Figure 7B). Only
a single identifiable fragment ion was observed (y5ꢀ) at a
relative abundance of less than 1.5%. It is believed that this
charge-dependent UVPD efficiency is due to the protonation
state of the bis-aryl hydrazone moiety. The doubly protonated
conjugated peptide of Ac-AAAKAAAAR X BAH X Ac-
AAAKPAAAR is assumed to have both protons sequestered
at the two C-terminal arginine residues. However, the triply
+
7
A. Cross-linked b- and y-type ions as well as specific cross-
linked fragment ions such as lysine immonium ions linked to
Analytical Chemistry, Vol. 81, No. 12, June 15, 2009 4871

Figure 7. UVPD mass spectra of Ac-AAAKAAAAR X BAH X Ac-AAAKPAAAR in the (A) 3+ (m/z 653.7) and (B) 2+ (m/z 980.0) charge states
using 7 and 15 pulses of UV irradiation, respectively. The first peptide listed is referred to as R and the second peptide as ꢀ; R/ꢀ nomenclature
is used for product ions which could have resulted from cleavage of either peptide. Precursor ions are signified with an asterisk (*).
protonated species has one additional mobile proton, pre-
sumably partially localized at the basic BAH-chromophore,
thus protonating N12 (see Figure 1 for numbering scheme).
We speculate that the additional mobile proton alters the
resulting gas-phase absorption profile of the BAH-conjugated
peptide, in this case increasing the UVPD efficiency at 355 nm.
This effect parallels the solution-phase absorption properties of
B2PH as indicated in the UV-vis spectra of this molecule in
neutral and acid-spiked solutions (Figure 2A). The difference
in UVPD efficiencies cannot simply be explained by a decrease
in the energy necessary to dissociate the 3+ versus 2+ precursor
ions because peptides cross-linked by the non-chromogenic DSS
molecule did not exhibit any UVPD in any charge state (data not
shown). In addition, lower CID voltages were required to dissoci-
ate these DSS-cross-linked analogues than the BAH-cross-linked
peptides as determined by energy-variable CID in which the
abundances of the selected precursor peptide ions were monitored
as the CID voltage was raised. The triply charged BAH-conjugated
peptide needed a 25% greater CID voltage than the DSS-cross-
linked species to reach the 50% dissociation level (0.326 vs 0.259
V). Preliminary studies of pulse-resolved UVPD, in which fragment
ion abundances were measured as a function of the number of
UVPD pulses, suggest that the primary product ions also exhibit
this charge-dependent photodissociation efficiency (e.g., primary
product ions with a mobile proton undergo absorption and
photodissociation upon subsequent pulses of UV irradiation
whereas product ions with protons sequestered at arginine
residues do not undergo efficient secondary dissociation).
cross-linked peptides from dead-end and unmodified peptides by
UVPD-MS. Ultraviolet photodissociation of the cross-linked peptides
provided more diagnostic sequence ion information than CID,
allowing one to readily identify the cross-linked residues of the
constituent peptides. The UVPD efficiencies were also observed to
be charge-dependent as protonation of the hydrazone bond was
necessary for rapid photodissociation, mirroring the solution-phase
UV-vis absorption profile of the BAH-chromophore in neutral and
acidic solutions. Only BAH-conjugated peptides were observed to
undergo efficient UV photodissociation at 355 nm allowing for facile
and selective identification of these ions of interest. Unmodified
peptides, as well as SFB- and SANH-modified peptides, do not contain
the BAH-chromophore and thus were not observed to photodisso-
ciate under any conditions. Online LC-UVPD-MS and direct-infusion
ESI-UVPD-MS techniques were successfully applied to a mock
mixture of peptides. Ongoing work in our lab is focusing on
increasing the duty cycle of this method in order to perform on the
fly data-dependent UVPD of eluting ions by using a laser with a
higher repetition rate and/or a higher peak power.
ACKNOWLEDGMENT
Funding from the NSF (Grant CHE-0718320) and the Welch
Foundation (Grant F1155) is gratefully acknowledged. Jae Schwartz
of ThermoFisher Scientific is also gratefully acknowledged for his
assistance in the modification of the instrument control files as
well as helpful discussion on laser triggering issues.
SUPPORTING INFORMATION AVAILABLE
Additional information as noted in text. This material is
available free of charge via the Internet at http://pubs.acs.org.
CONCLUSIONS
Received for review March 11, 2009. Accepted May 4,
Using a two-step conjugation method in which the chromogenic
bis-aryl hydrazone functionality was formed upon reaction of two
modified biomolecules, we have been able to rapidly distinguish
2009.
AC9005233
4872 Analytical Chemistry, Vol. 81, No. 12, June 15, 2009