Isotope-coded and affinity-tagged cross-linking (ICATXL): An efficient strategy to probe protein interaction surfaces

Article abstract of DOI:10.1021/ja0614159

Chemical cross-linking followed by identification of the cross-linked residues by mass spectrometry provides structural information on protein interaction surfaces. Nevertheless, accurate analysis of the digested, cross-linked proteins is often challengin

Full text of DOI:10.1021/ja0614159

Published on Web 07/25/2006
Isotope-Coded and Affinity-Tagged Cross-Linking (ICATXL): An Efficient
Strategy to Probe Protein Interaction Surfaces
Feixia Chu, Sami Mahrus, Charles S. Craik, and Alma L. Burlingame*
Department of Pharmaceutical Chemistry, UniVersity of California, San Francisco, California 94143-0446
Received March 11, 2006; E-mail: alb@cgl.ucsf.edu
Scheme 1
Protein bait affinity studies on the yeast proteome have begun
to reveal the nature and complexity of cellular protein-protein
interactions.1 However, our knowledge of the actual recognition
interfaces among these macromolecules is sparse.2 To gain mecha-
nistic insights into various cellular processes, it is important to
devise strategies that enable the mapping of protein-protein
interaction surfaces even at low resolution. We and others have
recently shown that valuable spatial restraints can be obtained from
chemical cross-linking of protein assemblies through use of tandem
mass spectrometry to identify the specific cross-linked amino acid
participants. Such investigations provide information on protein
folding, connectivity of secondary structures, as well as interaction
surfaces.3
Proteolytic digestion of the cross-linked protein complexes and
structural analysis of the cross-linked species by tandem mass
spectrometry have advantages in speed, sensitivity, and capability
of handling large protein assemblies. Nevertheless, detection and
accurate assignment of the cross-linked peptides is often challeng-
ing, complicated by the heterogeneity of the peptide mixture and
low stoichiometry of the cross-linked products sought. Incorporation
of affinity tags into cross-linkers can reduce the complexity of the
specific” protease inhibitor, ecotin.6 As expected, both cross-linkers
reacted with N-terminal R-amino and lysine ꢀ-amino groups and
gave similar cross-linking and cross-linker modification results (see
Supporting Information).
analytes and optimize the detection of low abundant cross-linked
species by permitting specific enrichment of species modified with
the affinity tag.4 Amine-reactive cross-linkers are often used in
protein cross-linking studies, due to the favorable distribution of
lysine residues on protein surfaces in solution. However, formation
of the initial products with an ꢀ-aminolysyl function is subsequently
dominated by hydrolysis of the second reactive functionality,
requiring a method for distinction of legitimate cross-links from
these half-cross-linking reaction products, that is, the cross-linker-
modified peptides. Herein, we present a novel cross-linking strategy
termed ICATXL (isotope-coded and affinity-tagged cross-linking)
that relies on the use of affinity-tagged cross-linkers and isotope
coding on the cross-linker-modified species (Scheme 1). This
strategy permits sensitive and facile analysis of cross-linked protein
samples.
A novel isotope-coded strategy was developed to differentiate
cross-linked peptides from cross-linker-modified peptides (Scheme
1). The protease inhibitor was cross-linked with 1 in both H2O16
and H2O18 buffer. Hydrolysis of NHS groups of 1 in each buffer
thus resulted in O16 or O18 incorporation into hydrolyzed, nonpro-
ductive cross-links. In contrast, oxygen is not incorporated into
cross-linked or nonreactive protein residues. Protein samples cross-
linked in H2O16 and H2O18 were then combined, dialyzed, and
proteolyzed. Cross-linked peptides and cross-linker-modified pep-
tides were purified and enriched using monomeric avidin purifica-
tion. Since peptides from hydrolyzed, nonproductive cross-links
were derived equally from O16 and O18 incorporation, they appear
as characteristic “doublets” in the mass spectra. Hence, further
structural analysis is simplified by the fact that these are readily
distinguished from actual cross-linked peptides. Figure 1A shows
the Lys-C digestion mixture of 1-treated ecotin. Analysis of the
individual peaks revealed the presence of unmodified proteolytic
peptides, cross-linker-modified peptides, and cross-linked peptides.
After affinity separation on monomeric avidin, cross-linked peptides
are no longer “suppressed” by the more abundant unmodified
peptides and are thus detected with higher sensitivity. For example,
the cross-linked species at m/z 3044.43 is significantly discernible
after affinity purification (compare Figure 1A and 1B). As expected,
all abundant peptides contain the biotin moiety from 1 (Table 1).
Cross-linker-modified peptides from the half-cross-linking reaction
are characterized by a unique 2 Da ∆mass doublet isotope pattern
We first synthesized two homofunctional, affinity-tagged cross-
linkers, 1 and 2 (see Supporting Information). Biotin was selected
as the affinity tag for its high affinity and small size. Cross-linker
2 has a thiol-cleavable site to facilitate the recovery of cross-linked
peptides after an avidin affinity purification step. The distance
between the two reactive carbonyls is similar to that of the well-
documented cross-linkers, disuccinimidyl suberate (DSS) and bis-
(sulfosuccinimidyl) suberate (BS3).5 Cross-linker 1 and 2 were tested
with a synthetic peptide and a well-characterized dimeric “fold-
9
10362
J. AM. CHEM. SOC. 2006, 128, 10362-10363
10.1021/ja0614159 CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Figure 2. Crystal structure of ecotin. The cross-linked residues are
connected with dotted lines and annotated with color-coded superscripts to
indicate their chain origin.
use for the study of more complex protein assemblies where
separation and rapid recognition of cross-linked peptides are highly
desirable.
Acknowledgment. We thank Dr. R.K. Guy for providing space
during synthesis of the cross-linking reagents, and Dr. Naoaki Fuji
for helpful discussions. This work was supported by NIH NCRR
01614, NCRR 12961, and NCRR 15804 (to A.L.B.), NIH CA72006
(to C.S.C.), and Eugene Cota-Robles graduate student fellowship
(to F.C.).
Supporting Information Available: Complete ref 1a and 1b,
experimental procedures, and detailed characterizations. This material
is available free of charge via the Internet at http://pubs.acs.org.
Figure 1. MALDI TOF-MS spectrum of the Lys-C digestion mixture of
1-treated ecotin before (A) and after (B) affinity separation.
Table 1. Affinity-Separated Peptides Observed by MALDI
TOF-MS
References
obsd mass calcd mass intensity
reacted
sequence
(1) (a) Gavin, A. C.; et al. Nature 2006, 440, 631-636. (b) Krogan, N. J.; et
al. Nature 2006, 440, 637-643.
[M
+
H]+
[M
+
H]+
(%)
residuesa
assignment
924.33b
924.41
22
100
38
10
5
K94, 1 mod.
N-term, 1 mod.
K135, 1 mod.
K18, 1 mod.
K135, 1 mod.
K95, 1 mod.
K91-K94
N-term-K9
K9, 1 mod.
N-term-K135
N-term, K9, 1 mod. N-term-18
93-95
N-term-9
(2) Sali, A.; Glaeser, R.; Earnest, T.; Baumeister, W. Nature 2003, 422, 216-
1771.77b 1771.80
1811.86b 1811.86
1852.81b 1852.86
2544.32b 2544.27
225.
132-142
(3) For reviews and examples, see: (a) Back, J. W.; de Jong, L.; Muijsers,
A. O.; de Koster, C. G. J. Mol. Biol. 2003, 331, 303-313. (b) Sinz, A. J.
Mass Spectrom. 2003, 38, 1225-1237. (c) Young, M. M.; Tang, N.;
Hempel, J. C.; Oshiro, C. M.; Taylor, E. W.; Kuntz, I. D.; Gibson, B.
W.; Dollinger, G. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 5802-5806.
(d) Doyle, D. A.; Morais Cabral, J.; Pfuetzner, R. A.; Kuo, A.; Gulbis, J.
M.; Cohen, S. L.; Chait, B. T.; MacKinnon, R. Science 1998, 280, 69-
77. (e) Kopp, D. A.; Berg, E. A.; Costello, C. E.; Lippard, S. J. J. Biol.
Chem. 2003, 278, 20939-20945. (f) Fancy, D. A.; Kodadek, T. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 6020-6024. (g) Lanman, J.; Lam, T.
T.; Barnes, S.; Sakalian, M.; Emmett, M. R.; Marshall, A. G.; Prevelige,
P. E., Jr. J. Mol. Biol. 2003, 325, 759-772. (h) Chu, F.; Shan, S.-O.;
Moustakas, D. T.; Alber, F.; Egea, P. F.; Stroud, R. M.; Walter, P.;
Burlingame, A. L. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 16454-16459.
10-21
127-142
2554.43
2575.37
2554.21
2575.16
3
7
40
8
13
2
95-112
77-92, 93-95
N-term-18
N-term-18
N-term-9, 132-142
2751.25b 2751.31
2769.33
3044.43b 3044.48
3289.46 3289.48
2769.32
a 1 mod. refers to modification introduced by cross-linker 1, and N-term
refers to the N-terminal R-amino group of the protein. b Identity of the
peptides was confirmed by tandem MS.
(4) (a) Trester-Zedlitz, M.; Kamada, K.; Burley, S. K.; Fenyo, D.; Chait, B.
T.; Muir, T. W. J. Am. Chem. Soc. 2003, 125, 2416-2425. (b) Alley, S.
C.; Ishmael, F. T.; Jones, A. D.; Benkovic, S. J. J. Am. Chem. Soc. 2000,
122, 6126-6127.
(e.g., peptides at m/z 1771.77 and 1811.83 (Figure 1B)). In contrast,
the enriched cross-linked peptides have a normal isotope pattern
(e.g., peptides at m/z 2751.25 and 3044.43 (Figure 1B)). The cross-
linked species at m/z 3044.43 consists of an intermonomer cross-
link between the N-terminal R-amino group of one monomer chain
and Lys135 of the other monomer chain (Figure 2). The sequence
of these two cross-linked peptides and the identity of the cross-
linked residues were conclusively established by tandem MS.7 Not
all cross-linking products of nearby lysines were observed, presum-
ably due to the discrepancy in reactivity of primary amines and in
detection of peptides. The assignment of the cross-linked residues
agrees with its crystal structure8 (Figure 2) and that of another cross-
linking study.6 Use of O16 and O18 for isotope coding is amenable
to introduction of a liquid chromatography separation step since
isotopes of oxygen do not affect retention time (data not shown)
as those of hydrogen do.9 Therefore, we anticipated that our isotope-
coded and affinity-tagged cross-linking strategy will be of general
(5) For examples, see: (a) de Gunzberg, J.; Riehl, R.; Weiberg, R. A. Proc.
Natl. Acad. Sci. U.S.A. 1989, 86, 4007-4011. (b) Huang, B. X.; Kim, H.
Y.; Dass, C. J. Am. Soc. Mass Spectrom. 2004, 15, 1237-1247. See also
ref 3c and 3h.
(6) An ecotin mutant, GGH-ecotin D137Y, with an N-terminal Gly-Gly-His
extension and an aspartate 137 to tyrosine, was used for this study. For
more information about this protein, see: Person, M. D.; Brown, K. C.;
Mahrus, S.; Craik, C. S.; Burlingame, A. L. Protein Sci. 2001, 10, 1549-
1562.
(7) See Supporting Information for low-energy CID tandem MS spectrum of
this cross-linked species.
(8) PDB code for wild-type ecotin: 1ECZ.
(9) (a) Zhang, R.; Sioma, C. S.; Wang, S.; Regnier, F. E. Anal. Chem. 2001,
73, 5142-5149. (b) Conrads, T. P.; Alving, K.; Veenstra, T. D.; Belov,
M. E.; Anderson, G. A.; Anderson, D. J.; Lipton, M. S.; Pasa-Tolic, L.;
Udseth, H. R.; Chrisler, W. B.; Thrall, B. D.; Smith, R. D. Anal. Chem.
2001, 73, 2132-2139.
JA0614159
9
J. AM. CHEM. SOC. VOL. 128, NO. 32, 2006 10363