Synthesis of AX7593, a quinazoline-derived photoaffinity probe for EGFR

Article abstract of DOI:10.1021/ol048656a

(Chemical Equation Presented) The synthesis of a photoaffinity probe for EGFR is described. O-Alkylation of 4-(meta-azidoanilino)-6-methoxy-7-hydroxy- quinazoline with a protected tetraethyleneglycol linker followed by the attachment of tetramethylrhodami

Full text of DOI:10.1021/ol048656a

ORGANIC
LETTERS
2004
Vol. 6, No. 21
3715-3718
Synthesis of AX7593, a
Quinazoline-Derived Photoaffinity Probe
for EGFR
Kevin R. Shreder,* Melissa S. Wong, Tyzoon Nomanbhoy,
Phillip S. Leventhal, and Stacy R. Fuller
ActiVX Biosciences, Inc., 11025 North Torrey Pines Road, La Jolla, California 92037
keVins@actiVx.com
Received July 13, 2004
ABSTRACT
The synthesis of a photoaffinity probe for EGFR is described. O-Alkylation of 4-(meta-azidoanilino)-6-methoxy-7-hydroxy-quinazoline with a
protected tetraethyleneglycol linker followed by the attachment of tetramethylrhodamine yielded the fluorescent probe AX7593. Photoaffinity
labeling of EGFR by AX7593 (Kb
AG1478.
) 280 nM) was shown to have an efficiency of 34% and to be competitive with the EGFR inhibitors PP2 and
Photoaffinity reagents are useful biochemical tools for
investigating protein identification, structure, localization, and
active-site determination.1 Three structural components are
important in the design of such probes: a protein recognition
element that incorporates a photoreactive group, a linker,
and a tag useful for detection and quantitation (e.g., a
fluorophore). Unlike a probe that incorporates an affinity
label,2 the association of a photoaffinity probe with its target
is reversible until photoactivated. After irradiation, the
initially inert photoreactive group is converted into a highly
reactive electrophilic species that forms a covalent bond with
the protein to which it is bound. For example, photoexcitation
of aryl azides yields a reactive aryl nitrene capable of
insertion into amino acid residues with both nucleophilic and
nonnucleophilic side chains.3
thesis of a fluorophore-tagged photoaffinity probe for the
epidermal growth factor receptor (EGFR). EGFR is a
transmembrane receptor and a member of the tyrosine kinase
family that has been found to be abnormally activated in a
variety of solid tumors.5 This activation of EGFR results in
autophosphorylation, which drives signal transduction path-
ways leading to tumor growth. A number of small-molecule
inhibitors that interfere with aberrant EGFR family signaling
are currently in preclinical development or in clinical trials
for this therapeutically important kinase.6
The preparation of the (meta-azidoanilino)-quinazoline 5
(Scheme 1) began with the SNAr displacement of the chloride
group of 7-benzyloxy-4-chloro-6-methoxy-quinazoline (1)7
with meta-phenylenediamine (2) in isopropyl alcohol heated
As part of our efforts to design fluorescent probes to study
signal transduction processes,4 we present herein the syn-
(3) (a) Fleming, S. A. Tetrahedron 1995, 51 (4), 12479-12520. (b)
Schwartz, M. A. NATO ASI Ser., Ser. C., Photochem. Probes Biochem.
1989, 272, 157-168. (c) Bayley, H. Photogenerated Reagents in Biochem-
istry and Molecular Biology; Elsevier: Amsterdam, 1983.
(1) (a) Fedan, J. S.; Hogaboom, G. K.; O’Donnell, J. P. Biochem.
Pharmacol. 1984, 33, 1167-1180. (b) Tometsko, A. M.; Richards, F. M.,
Eds. Ann. N. Y. Acad. Sci. 1980, 346, 134-502.
(2) Campbell, D. A.; Szardenings, A. K. Curr. Opin. Chem. Biol. 2003,
7, 296-303.
(4) Shreder, K. R.; Liu, Y.; Nomanhboy, T.; Fuller, S. R.; Wong, M. S.;
Gai, W. Z.; Wu, J.; Leventhal, P. S.; Lill, J. R.; Corral, S. Bioconjugate
Chem. 2004, 15, 790-798.
(5) Vlahovic, G.; Crawford, J. Oncologist 2003, 8 (6), 531-538.
(6) Janmaat, M. L.; Giaccone, G. Oncologist 2003, 8 (6), 576-586.
10.1021/ol048656a CCC: $27.50
© 2004 American Chemical Society
Published on Web 09/23/2004

Scheme 1. Synthesis of the (meta-Azidoanilino)-quinazoline 5
Scheme 2. Synthesis of AX7593
at reflux, which yielded the anilino-quinazoline 3 in 97%
yield.8 Deprotection of the O-benzyl group to give compound
4 was achieved in trifluoroacetic acid (TFA) heated at reflux.
Treatment of this intermediate with sodium nitrite in 80%
acetic acid produced a yellow diazonium salt that was not
isolated. In situ treatment of this intermediate with sodium
azide generated the aryl azide 5 with the expected concomi-
tant evolution of N2 gas.
Once this aryl azide-containing precursor was in hand,
attention was turned to the attachment of the linker and
fluorophore (Scheme 2). Displacement of one of the tosylate
groups of tetraethylene glycol di-p-tosylate (6) using the
potassium salt of di-tert-butyl iminodicarboxylate (Carpino’s
acid-labile phthalimide equivalent)9 yielded the protected
amino tetraethyleneglycol linker 7 in 39% yield. O-Alkyla-
tion of the (meta-azidoanilino)-quinazoline 5 with compound
7 was achieved using K2CO3 in DMF to give the intermediate
8. Deprotection of the N,N-di-Boc group was accomplished
quantitatively in 10% TFA/dichloromethane. Higher con-
centrations of TFA resulted in the decomposition of the aryl
azide to the corresponding arylamine. AX7593 was produced
by the treatment of this ammonium salt with the mixed
succinimidyl esters of 5-(and 6)-carboxytetramethylrhodamine
(TAMRA-OSu) and diisopropylethylamine (DIEA) in DMSO.
TAMRA has previously been shown to be a superior
fluorescent tag for fluorophosphonate-based serine hydrolase
probes.10 Purification of AX7593 was achieved using C18
reversed-phase HPLC (0.1% TFA as a modifier) followed
by lyophilization. Use of this purification scheme eliminated
any significant trace of degradants in the final product. When
the HPLC eluant was removed using rotary evaporation, the
increased concentration of the less volatile TFA resulted in
product decomposition.
(7) Compound 1 was synthesized from the commercially available
4-benzyloxy-3-methoxy-vanillin in seven steps. See: (a) Julia, M.; Manoury,
P.; Voillaume M. C. Bull. Soc. Chim. Fr. 1965, 5, 1417-1423. (b) Althius,
T. H.; Hess, H.-J. J. Med. Chem. 1977, 20, 146-149. (c) Hennequin, L.
F.; Thomas, A. P.; Johnstone, C.; Stokes, E. S. E.; Ple´, P. A.; Lohmann,
J.-J. M.; Ogilvie, D. J.; Dukes, M.; Wedge, S. R.; Curwen, J. O.; Kendrew,
J.; Lambert-van der Brempt, C. J. Med. Chem. 1999, 42, 2, 5369-5389.
(8) Certain meta-substituted anilino-quinazolines have been shown to be
more potent inhibitors of EGFR when compared to ortho- and para-
substituted congeners; see: Rewcastle, G. W.; Denny, W. A.; Bridges, A.
J.; Zhou, H.; Cody, D. R.; McMichael, A.; Fry, D. W. J. Med. Chem. 1995,
38, 3482-3487.
To test whether this compound would label EGFR, the
title kinase was incubated with AX7593 over a range of
concentrations (0.01-2 µM) for 30 min.11 After this period
(10) (a) Patricelli, M. P.; Giang, D. K.; Stamp, L. M.; Burbaum, J. J.
Proteomics 2001, 1, 1067-1071. (b) Adam, G. C.; Burbaum, J.; Kozarich,
J. W.; Patricelli, M. P.; Cravatt, B. F. J. Am. Chem. Soc. 2004, 126 (4),
1363-1368.
(11) All photoaffinity labeling experiments were conducted in 50 mM
HEPES, 150 mM NaCl, 0.03% Triton X-100, pH 7.4.
(9) Carpino, L. A. J. Org. Chem. 1964, 29, 2820-2824.
3716
Org. Lett., Vol. 6, No. 21, 2004

of equilibration, the samples were illuminated with 254 nm
light to photoactivate AX7593 and tag any complexed
enzyme. Unreacted probe was separated from EGFR using
SDS-PAGE, and the fluorescent signal of TAMRA tagged
EGFR was measured using a flatbed gel scanner. Because
the degree of photoaffinity labeling is directly proportional
to the amount of probe bound prior to irradiation, the Kb
(binding constant) value could be calculated and was
determined to be 280 ( 13 nM. As a control experiment,
AX7593 (1 µM) was photoactivated in the presence of Akt1,
MAPKAP kinase 2, and Cdk2, and no labeling of these
kinases was observed.
The ability of the probe to inhibit the EGFR-catalyzed
phosphorylation of poly(Glu-Ala-Tyr) with [γ-32P]-ATP was
also examined.12 EGFR activity, as measured by the incor-
poration of 32Pi into the peptide substrate, was plotted versus
the concentration of AX7593 (0.01 µM to 10 µM, Figure
1). When the resultant curve was fit to a four-parameter Hill
Figure 2. Inhibition of AX7593 labeling of EGFR by PP2 (b)
and AG1478 (O) at various concentrations. The curves drawn are
the best fit of the data to a four-parameter Hill function, which
yielded IC50 values of 210 ( 53 nM and 8.0 ( 1.8 nM for PP2
and AG1478, respectively.
AX7593 was assigned as 34%. The labeling efficiencies of
aryl azide-containing photoaffinity probes are known to vary
widely from single digit percentages to near quantitative
values.13
Next, labeling experiments in the presence of EGFR
inhibitors were performed to determine if AX7593 could be
used to assay for small-molecule inhibition (Figure 2).14 Two
known EGFR inhibitors, PP2 (IC50 ) 480 nM)15 and AG1478
(IC50 ) 7 nM),16 were used as model compounds. After
incubation of EGFR (10 nM) with varying concentrations
of both inhibitors (0.5-10 000 nM), AX7593 (2 µM) was
added and the probe was photoactivated at 254 nm. When
the percent inhibition of photoaffinity labeling was plotted
as a function of inhibitor concentration and the resulting
curves were fitted to a four-parameter Hill function, the
calculated IC50 values for PP2 (IC50 ) 210 ( 53 nM) and
AG1478 (IC50 ) 8.0 ( 1.8 nM) were in close agreement
with literature values.
Figure 1. Inhibition of EGFR kinase activity by AX7593. The
curve drawn is the best fit of the data to a four-parameter Hill
function, which yielded an IC50 value of 350 ( 20 nM.
In summary, we developed a route for the synthesis of
AX7593, a quinazoline-based photoaffinity probe for EGFR.
Importantly, small-molecule competition of AX7593 pho-
tolabeling was shown to be a useful method of extracting
EGFR IC50 values. Such inhibition of photoaffinity labeling
has the potential to be adapted to a nonradioactive, high-
throughput screening method of new compounds as inhibitors
of EGFR. The linear nature of the synthesis described herein
function, the IC50 of AX7593 for EGFR was calculated to
be 350 ( 20 nM, a value in good agreement with the binding
constant determined by photoaffinity labeling (vide supra).
To determine the labeling efficiency of AX7593 (i.e., the
percentage of probe-complexed enzyme that is covalently
tagged), a calibration curve was generated from the labeling
of trypsin with a TAMRA-containing serine hydrolase fluoro-
phosphonate probe. Use of this calibration curve allowed the
amount of TAMRA fluorescence visualized using SDS-
PAGE to be correlated to the absolute quantity of labeled
enzyme. When EGFR was labeled by AX7593, it was de-
termined that the amount of TAMRA fluorescence observed
was 66% less than expected for the complete labeling of
AX7593-bound EGFR. Thus, the labeling efficiency of
(13) Bayley, H.; Knowles, J. R. Methods Enzymol. 1977, 46, 69-114.
(14) TAMRA-tagged activity-based probes have been shown to be useful
in the high-throughput screening of small molecule inhibitors directly in
proteomes. See: (a) Nomanbhoy, T. K.; Rosenblum, J.; Aban, A.; Burbaum,
J. J. Assay Drug DeV. Technol. 2003, 1 (2), 137-146. (b) Leung, D.;
Hardouin, C.; Boger, D. L.; Cravatt, B. F. Nat. Biotechnol. 2003 21 (6),
687-691.
(15) Hanke, J. H.; Gardner, J. P.; Dow, R. L.; Changelian, P. S.; Brissette,
W. H.; Weringer, E. J.; Pollok, B. A.; Connelly, P. A. J. Biol. Chem. 1996,
695-701.
(16) (a) Levitzki, A.; Gazit, A. Science 1995, 267, 1782-1788. (b)
Osherov, N.; Levitzki, A. Eur. J. Biochem. 1994, 225, 1047-1053.
(12) Koland, J. G.; Cerione, R. A. J. Biol. Chem. 1988, 263 (5), 2230-
2237.
Org. Lett., Vol. 6, No. 21, 2004
3717

allows for the ready incorporation of other kinds of linkers
and tags. Additional studies with AX7593 and related
photoaffinity probes for EGFR will be reported in due course.
Supporting Information Available: Synthetic experi-
mentals for compounds 3-5, 7-8, and AX7593; EGFR Kb
and IC50 determination for AX7593; determination of
AX7593 labeling efficiency. This material is available free
of charge via the Internet at http://pubs.acs.org.
Acknowledgment. The authors thank Drs. John Kozarich,
Katrin Szardenings, and David Campbell for helpful sug-
gestions concerning this work or manuscript.
OL048656A
3718
Org. Lett., Vol. 6, No. 21, 2004