Improved CILAT reagents for quantitative proteomics

Article abstract of DOI:10.1016/j.bmcl.2009.02.022

Improved CILAT reagents have been developed, with which an unprecedented number of protein samples can be measured in high-throughput assays, providing a robust tool for MS-based quantitative proteomics.

Full text of DOI:10.1016/j.bmcl.2009.02.022

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2059–2061
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Improved CILAT reagents for quantitative proteomics
*
Dexing Zeng, Shuwei Li
Center for Advanced Research in Biotechnology, University of Maryland Biotechnology Institute, 9600 Gudelsky Drive, MD 20850, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Improved CILAT reagents have been developed, with which an unprecedented number of protein samples
Received 22 September 2008
Revised 30 January 2009
Accepted 3 February 2009
Available online 10 February 2009
can be measured in high-throughput assays, providing
proteomics.
a
robust tool for MS-based quantitative
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Mass spectrometry
Quantitative proteomics
Multiplex
Stable isotope labeling
Isobaric
MS-based quantitative proteomics is a robust tool for the sys-
tematic understanding of biological processes.1 Many techniques,
including both stable isotope-labeled and label-free methods, have
been developed to improve the throughput and accuracy of protein
quantitation. Stable-isotope labeled methods, such as ICAT2 and
iTRAQ3, rely on the introduction of stable isotope tags that are
chemically identical, but are distinguishable by MS, into different
biological samples. On the other hand, label-free quantitation is
achieved by spectral count and statistical analysis of unlabeled
proteolytic peptides. Despite the availability of a wide choice of
methods, there are no perfect ways for quantitative proteomics
as each technique has its own benefits and drawbacks.4 For in-
stance, label-free methods are easy to implement, but suffer from
interference from experimental variation and signal noise. Label-
based strategies can provide more accurate quantitation, yet it is
expensive and sometime impractical to label scarce biological sam-
ples. As a result, methods that can overcome limitations of present
techniques have always been sought to improve the speed and reli-
ability of quantitative analysis.
peptide, labeled peptides must be extensively fractionized to re-
duce their complexity to a level within the analytic capacity of
MS. This requirement is usually alleviated when ICAT and other
cysteine-targeting reagents are used because labeled cysteine-con-
taining peptides, which represent about 10% of all proteolytic pep-
tides from a proteome, can be enriched by affinity purification
before MS analysis.7 Nevertheless, a drawback of ICAT reagents is
that they can only be used for pair-wise measurement.
Previously, we have developed another type of reagents, named
CILAT8, with attempts to combine benefits offered by both ICAT
and iTRAQ together. However, our prototype reagents were limited
to the comparison of only two samples as well. Here, we demon-
strate the second generation CILAT reagents that allow the quanti-
tation of up to 12 samples. These reagents are a set of 12
chemically identical compounds with each containing an isobaric
tag comprised of a reporter and a balancer (compound 1, Fig. 1a).
They also share a common thiol-capture group to label cysteine-
containing peptides and a cleavable alkyne tag that allows the so-
lid-phase based enrichment of labeled peptides in a catch-and-re-
lease fashion. However, these reagents differ from one another by
incorporating different isotope atoms such as 2H, 13C and 15N into
the different positions of isobaric tags. Consequently, the molecu-
lar weight of the reporters spans from 130 to 142 (except 136)9
and that of the balancers changes concertedly from 167 to 155 to
keep the molecular weight of all isobaric tags at constant 297.
When they are applied for quantitative assay (Fig. 1b), these re-
agents can label up to 12 samples, respectively, which are then
mixed together and digested with trypsin to generate a large pop-
ulation of peptides. Labeled peptides can then be covalently immo-
bilized on beads coated with azide via a biologically compatible
A major advantage of labeled over label-free approaches is the
ability to process multiple specimens in parallel as demonstrated
by commercially available 8-plex iTRAQ5 and 6-plex TMT re-
agents.6 However, as both iTRAQ and TMT reagents react with pri-
mary amine that is ubiquitously present in each proteolytic
Abbreviations: CILAT,
cleavable isobaric labeled affinity tag; MS, mass
spectrometry; ICAT, isotope coded affinity tags; iTRAQ, isobaric tags for relative
and absolute quantitation; TMT, tandem mass tags.
* Corresponding author. Tel.: +1 240 314 6330.
E-mail address: liw@umbi.umd.edu (S. Li).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.02.022

2060
D. Zeng, S. Li / Bioorg. Med. Chem. Lett. 19 (2009) 2059–2061
same molecule weight, thus exhibiting a single peak in MS. When
a
this peak is fragmented in MS/MS, the amide bond between the
reporters and the balancers breaks apart to yield a series of strong
signature ions with m/z ranging from 130 to 142 for quantitation.
Other larger fragments can be used for de novo sequencing of this
peptide.
These CILAT reagents were prepared by a simple semi-solid-
phase synthesis (Fig. 2). First, ethanolamine was loaded onto 4-
nitrophenyl carbonate resin via its primary amine group, and its
hydroxyl group was activated by tosylation, which was displaced
with piperazine and coupled with Fmoc-leucine subsequently.
Next, after Fmoc was removed to generate an amine group to cou-
ple with a bromoacetic acid, an intermediate product (compound
2) was cleaved from beads by acid treatment and was used to react
with another intermediate (compound 3) to obtain compound 1.
By using various isotope labeled building block molecules includ-
ing ethanolamine, piperazine, Fmoc-leucine, and bromoacetic acid,
all 12 CILAT reagents were synthesized similarly.
b
To examine if the CILAT reagents could indeed be used for
12-plex measurement, bovine catalase, a 60 kD protein containing
four cysteines, was tested as an example. 12 samples of this protein
with predetermined ratios (2:2:3:3:5:6:7:6:3:3:4:4) were labeled
with individual CILAT reagents, respectively, mixed together,
precipitated to remove excess labeling reagents, and then digested
with trypsin. After the catch-and-release enrichment of labeled
peptides, three out of four cysteine-containing peptides (m/z =
1392.0, 1575.1 and 2041.4) exhibited strong signals in MS and
the last one (calculated m/z = 4043.0) was not observed due to its
poor ionization efficiency (Fig. 3a). When one of peptides
(1392.0) was selected to fragment at low collision energy condi-
Figure 1. (a) Structure of CILAT reagents (Compound 1). Each CILAT reagent
contains a thiol capture group, an isobaric tag composed of a reporter and a
balancer, and an acid-cleavable alkyne group as an enrichment tag. The molecule
weight of the reporters in all 12 reagents ranges from 130 to 142, except 136, and
the balancer in each reagent changes correspondingly to keep the molecular weight
of the isobaric tag at 297. (b) Procedure to use CILAT reagents. (1) Up to 12 protein
samples (S1 to S12) can be labeled on cysteine residues with Tag 130 to Tag 142,
respectively. (2) All samples are mixed together and digested with trypsin. (3) The
resulting peptide mixtures are treated with solid support beads coated with azide.
Only cysteine-containing peptides that are labeled with CILAT reagents can couple
to the beads while other unlabeled peptides are washed off. Then, these immobi-
lized peptides are cleaved from beads with acid treatment via the acid-cleavable
linker in CILAT reagents. (4) The labeled peptides are analyzed by MS/MS to
generate signature ions for quantitation.
tion, the sequence of this peptide (H-LC ENIAGHLK-OH)11 was eas-
*
ily identified (Fig. 3b). At high collision energy condition, a series of
signature ions from 130 to 142 appeared as designed (Fig. 3c). The
other two peptides (1575.1 and 2041.4) also performed similarly
when they were fragmented. Before these peaks were used for
quantitation, however, two issues that might potentially lead to
miscalculation of peptide ratios should be addressed: 1) are these
signature peaks exclusively generated from the isobaric tags? and
2) how can we calibrate quantitation if CILAT reagents are not
100% isotopically pure? To help answer the first question, the same
peptide (1392.0) from bovine catalase solely labeled with Tag 130
was fragmented. A single peak at 130 within the range of 130–142
(Fig. 3d) was observed, implying the signature ions were
indeed produced from the reporters, not from the peptide itself.
[3+2] cycloaddition (click chemistry) between azide and alkyne.10
After extensive washing to remove any unlabeled peptides, tagged
peptides are released from beads by acid treatment and directly
delivered for MS/MS analysis. As the name of isobaric tag suggests,
the same peptides labeled with different CILAT reagents have the
Figure 2. (a) Semi-solid-phase synthesis of CILAT reagents. (b) Isotope-labeled building block monomers used for the synthesis of Tag 130 to Tag 142. A number (2nd–5th
rows) indicates the molecular weight difference (Daltons) between this isotope-labeled monomer and its unlabeled form. The structures of these starting materials are
illustrated in Supplementary data.

D. Zeng, S. Li / Bioorg. Med. Chem. Lett. 19 (2009) 2059–2061
2061
peaks within the range will be used for a 9-plex measurement,
rather than a 12-plex quantitation. The second concern can be ad-
dressed by slightly modifying an algorithm already developed for
iTRAQ reagents, whose isotope purity also affects the quantitation
of peptides in a similar way.12 Thus, a computer program was
developed to calibrate quantitation (see Supplementary data).
Finally, when all 36 adjusted ratios of these three cysteine-contain-
ing peptides in the 12-plex measurement were plotted versus their
normalized predetermined ratios (2/7:3/7:4/7:5/7:6/7:7/7), the
linear regression relationship between them indicated the CILAT
reagents were indeed able to determine protein ratios in this
high-throughput assay (Fig. 3e).
The second generation CILAT reagents would enable us to pro-
cess an unprecedented number of samples with an additional ben-
efit to reduce biological complexity. In addition, solid-phase
enrichment could potentially eliminate problems associated with
affinity purification based on conventional biotin–avidin system,
such as high background. Although these reagents are designed
to complement, rather than replace existing techniques such as
ICAT and iTRAQ, the ease and robustness of this technique would
make it a preferred choice in many applications.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.02.022.
References and notes
Figure 3. (a) MS spectrum of labeled peptides after enrichment. Cysteine-contain-
ing peptides are underlined. Ir = % relative intensity (b) MS/MS spectrum of the
precursor ion (1392.04) labeled with all 12 tags at low collision energy condition.
The sequence of this peptide and its y/b-series fragments are shown. An asterisk on
cystein indicates it is labeled with tags. (c) MS/MS spectrum of the precursor ion
(1392.0) labeled with all 12 tags at high collision energy condition. (d) MS/MS
spectrum of the precursor ion (1392.0) solely labeled with Tag 130 at high collision
energy condition. (e) Linear regression relationship between the measured ratios
(y-axis) of labeled peptides and their normalized predetermined ratios (x-axis).
Standard error bars are shown.
1. Ong, S. E.; Mann, M. Nat. Chem. Biol. 2005, 1, 252.
2. Patel, A.; Perrin, D. M. Bioconjugate. Chem. 2004, 15, 224.
3. Ross, P. L.; Huang, Y. N.; Marchese, J. N.; Williamson, B.; Parker, K.; Hattan, S.;
Khainovski, N.; Pillai, S.; Dey, S.; Daniels, S.; Purkayastha, S.; Juhasz, P.; Martin,
S.; Bartlet-Jones, M.; He, F.; Jacobson, A.; Pappin, D. J. Mol. Cell Proteomics 2004,
3, 1154.
4. (a) Julka, S.; Regnier, F. E. Brief Funct. Genomic. Proteomic. 2005, 4, 158; (b) Yan,
W.; Chen, S. S. Brief Funct. Genomic. Proteomic. 2005, 4, 27.
5. Choe, L.; D’Ascenzo, M.; Relkin, N. R.; Pappin, D.; Ross, P.; Williamson, B.;
Guertin, S.; Pribil, P.; Lee, K. H. Proteomics 2007, 7, 3651.
6. Dayon, L.; Hainard, A.; Licker, V.; Turck, N.; Kuhn, K.; Hochstrasser, D. F.;
Burkhard, P. R.; Sanchez, J. C. Anal. Chem. 2008, 80, 2921.
7. Griffin, T. J.; Aebersold, R. J. Biol. Chem. 2001, 276, 45497.
8. Li, S.; Zeng, D. Chem. Commun. 2007, 2181.
9. Tag 136 is not used because tyrosine-containing peptides also give a peak at m/
z = 136 (immonium ion of tyrosine), interfering with quantitation.
10. (a) Galan, J. A.; Guo, M.; Sanchez, E. E.; Cantu, E.; Rodriguez-Acosta, A.; Perez, J.
C.; Tao, W. A. Mol. Cell Proteomics 2008, 7, 785; (b) Zhou, F.; Galan, J.; Geahlen,
R. L.; Tao, W. A. J. Proteome Res. 2007, 6, 133.
11. The asterisk on C in the peptide sequence indicates that this cysteine is labeled
with CILAT reagents.
12. Shadforth, I. P.; Dunkley, T. P.; Lilley, K. S.; Bessant, C. BMC Genomics 2005, 6,
145.
Nevertheless, it is noteworthy that the b1 ion of a peptide contain-
ing an N-terminus Glu/Met/His residue is in this range and may
introduce errors during quantitation. This problem can be solved
by identifying these peptides and removing interfering peaks from
quantitation. For example, the m/z of the b1 ion of a peptide con-
taining an N-terminus Glu residue is 130, so the peak at 130, to-
gether with the peak at 131 and 132 for the sake of isotope
effect, will be excluded during quantitation and the remaining