Design, synthesis, and evaluation of a radicicol and geldanamycin chimera, radamide

Article abstract of DOI:10.1021/ol048266o

(Chemical equation presented) A chimera composed of the natural products radicicol and geldanamycin has been prepared through an amide linkage connecting the resorcinol moiety of radicicol to the quinone ring of geldanamycin. The inhibitory activity of th

Full text of DOI:10.1021/ol048266o

ORGANIC
LETTERS
2004
Vol. 6, No. 24
4459-4462
Design, Synthesis, and Evaluation of a
Radicicol and Geldanamycin Chimera,
Radamide
Randell C. Clevenger and Brian S. J. Blagg*
Department of Medicinal Chemistry and The Center for Protein Structure and
Function, The UniVersity of Kansas, 1251 Wescoe Hall DriVe, Malott 4070,
Lawrence, Kansas 66045-7562
bblagg@ku.edu
Received August 30, 2004
ABSTRACT
A chimera composed of the natural products radicicol and geldanamycin has been prepared through an amide linkage connecting the resorcinol
moiety of radicicol to the quinone ring of geldanamycin. The inhibitory activity of these compounds was determined by their ability to inhibit
Hsp90’s inherent ATPase activity along with degradation of the Hsp90 client protein, HER-2 in MCF-7 breast cancer cells.
The 90 kDa heat shock proteins (Hsp90) are essential for
refolding denatured proteins as well as for the conformational
maturation of nascent polypeptides into biologically active
three-dimensional structures.1 Several Hsp90 dependent client
proteins have been identified and include Src kinase, Raf,
HER-2, p185, mutant p53 (not normal p53), telomerase,
steroid hormone receptors, polo 1-kinase (PLK), protein
kinase B (AKT), death domain kinase (RIP), MET kinase,
focal adhesion kinase (FAK), as well as the aryl hydrocarbon
receptor, PKR, nitric oxide synthase, and others.2 Of these
Hsp90 client proteins, HER-2, Raf, PLK, RIP, AKT, FAK,
telomerase, and MET are directly associated with all six
hallmarks of cancer.3 Consequently, Hsp90 has emerged as
a promising biological target for the development of cancer
chemotherapeutics because multiple pathways can be simul-
taneously disrupted by inhibition of the Hsp90 protein folding
machinery.1
degradation of unfolded and partially folded client proteins
via the ubiquitin-proteasome pathway.4 Known inhibitors of
the Hsp90 protein folding process include geldanamycin
(GDA, Figure 1),5 herbimycin A (HB),6 and radicicol
(RDC).7 RDC is the most potent Hsp90 inhibitor in vitro
but has no activity in vivo.8 In contrast, GDA is less potent
than RDC in vitro with an IC50 of 1-3 µM.9 However, in
cellular studies, a derivative of GDA (17-AAG) was shown
to have greater affinity for the Hsp90 multiprotein complex
found in malignant cells with an IC50 of 5-100 nM.10
(4) Binder, R. J.; Blachere, N. E.; Srivastava, P. K. J. Biol. Chem. 2001,
276, 17163.
(5) Grenert, J. P.; Sullivan, W. P.; Fadden, P.; Haystead, T. A. J.; Clark,
J.; Mimnaugh, E.; Krutzsch, H.; Ochel, H.-J.; Schulte, T. W.; Sausville,
E.; Neckers, L. M.; Toft, D. O. J. Biol. Chem. 1997, 272, 23843.
(6) Miller, P.; DiOrio, C.; Walter, C.; Moyer, J. D. Proc. Am. Assoc.
Cancer Res. 1993, 34, 2047.
(7) Yang, Z.-Q.; Geng, X.; Solit, D.; Pratilas, C. A.; Rosen, N.;
Danishefsky, S. J. J. Am. Chem. Soc. 2004, 126, 7881.
(8) Agatsuma, T.; Ogawa, H.; Akasaka, K.; Asai, A.; Yamashita, Y.;
Mizukami, T.; Akinaga, S.; Saitoh, Y. Bioorg. Med. Chem. 2002, 10, 3445.
(9) (a) Panaretou, B.; Prodromou, C.; Roe, S. M.; O’Brien, R.; Ladbury,
J. E.; Piper, P. W.; Pearl, L. H. EMBO J. 1998, 17, 4829. (b) Chiosis, G.;
Huezo, H.; Rosen, N.; Mimnaugh, E.; Whitesell, L.; Neckers, L. Mol.
Cancer Ther. 2003, 2, 123.
The Hsp90 protein folding process is ATP dependent.
Disruption of ATP binding and hydrolysis results in the
(1) Walter, S.; Buchner, J. Angew. Chem., Int. Ed. 2002, 41, 1098.
(2) Adams, J.; Elliot, P. J. Oncogene 2000, 19, 6687.
(3) Maloney, A.; Workman, P. Expert Opin. Biol. Ther. 2002, 2, 3.
10.1021/ol048266o CCC: $27.50
© 2004 American Chemical Society
Published on Web 10/30/2004

Examination of the cocrystal structures suggested a
chimeric molecule composed of RDC’s resorcinol ring, and
GDA’s quinone may provide a molecule with high affinity
for Hsp90. Radanamycin amide (radamide, Figure 3) is such
Figure 1. Hsp90 inhibitors.
Figure 3. Radanamycin amide, a radicicol and geldanamycin
chimera.
Although 17-AAG has entered Phase I clinical trials for
the treatment of several cancers,11 17-AAG still maintains
additional toxicity and formulation problems12 that may prove
difficult to overcome. Therefore, the identification of new
inhibitors that are more accessible to structural modification
is likely to provide clinically useful alternatives to 17-AAG.
Although the entire three-dimensional structure of Hsp90
has not been elucidated, cocrystal structures of the N-terminal
region bound to GDA, RDC, and ADP have been solved
(Figure 2).13 The resorcinol moiety of RDC binds in the same
a molecule that connects the resorcinol ring of radicicol to
the quinone moiety of geldanamycin through an amide
linkage. Molecular modeling and docking experiments sup-
ported alignment of these two portions into the appropriate
locations within the ATP binding site of Hsp90.
We sought to prepare this chimera and to furnish additional
analogues that could unveil structure-activity relationships
between RDC, GDA, and radamide. Thus, compound 1 was
synthesized from methyl 2,4-dihydroxy-5-methyl benzoate14
(2, Scheme 1) by silyl protection of the phenols followed
by chlorination of the aromatic ring to provide 4. Treatment
of 4 with lithium diisopropylamide at -78 °C followed by
addition of allyl bromide provided the allylated product, 5.
Ozonolysis of the double bond and oxidation of the resultant
aldehyde gave the corresponding acid 7.
The quinone precursor 10 was prepared from 1,4-bis-
(methoxymethoxy)-2-methoxybenzene (8),15 by nitration16
and reduction of the nitro group (Scheme 2). Since the
quinone ring is redox-active17 and an excellent Michael
acceptor, we also prepared a trimethoxy phenyl derivative
11, which is not subject to the same reactivity as a quinone
but still maintains hydrogen bond-accepting capabilities. In
the event, 7 was coupled with anilines 10 and 11 to provide
the corresponding amide products, 12 and 13. Removal of
Figure 2. Binding interactions of GDA and RDC with Hsp90.
b ) one molecule of H2O.
(12) (a) Sausville, E. A.; Tomaszewski, J. E.; Ivy, P. Curr. Cancer Drug
Targets 2003, 3, 377-383. (b) Neckers, L. M. Drug Res. Updates 2000, 3,
203-205. (c) Chiosis, G.; Lucas, B.; Shtil, A.; Huezo, H.; Rosen, N. Bio.
Med. Chem. 2002, 10, 3555-3564. (d) Chiosis, G.; Lucas, B.; Huezo, H.;
Solit, D.; Bassa, A.; Rosen, N. Curr. Cancer Drug Targets 2003, 3, 371-
376. (e) Neckers, L. Trends Mol. Med. 2002, 8, S55-S61. (f) Dikalov, S.;
Landmesser, U.; Harrison, D. G. J. Biol. Chem. 2002, 277, 25480-25485.
(g) Sreedhar, A. S.; Mihaly, K.; Pato, B.; Schnaider, T.; Stetak, A.; Kis-
Petik, K.; Fidy, J.; Simonics, T.; Maraz, A.; Csermely, P. J. Biol. Chem.
2003, 278, 35231-3523140. (h) Workman, P. Curr. Cancer Drug Targets
2003, 3, 297-300.
location as the adenine ring of ADP and mimics the hydrogen
bond donor/acceptor properties of the exo- and N7 endocyclic
amine/imine, respectively. In contrast, the quinone ring of
GDA binds toward the exterior of the pocket and participates
in hydrogen bond interactions with the amino acids that
normally bind to the diphosphate region of ADP. Key
interactions observed between the quinone and ATP binding
pocket suggest that binding to this region is critical to GDA’s
affinity for Hsp90.
(13) Roe, S. M.; Prodromou, C.; O’Brien, R.; Ladbury, J. E.; Piper, P.
W.; Pearl, L. H. J. Med. Chem. 1999, 42, 260.
(14) Furstner, A.; Gastner, T. Org. Lett. 2000, 2, 2467.
(15) Andrus, M. B.; Hicken, E. J.; Meredith, E. L.; Simmons, B. L.;
Cannon, J. F. Org. Lett. 2003, 5, 3859.
(10) Kamal, A.; Thao, L.; Sensintaffar, J.; Zhang, L.; Boehm, M. F.;
Fritz, L. C.; Burrows, F. J. Nature 2003, 425, 407.
(11) Blagosklonny, M. V. Leukemia 2002, 16, 455.
(16) Crivello, J. V. J. Org. Chem. 1981, 46, 3056.
(17) Dikalov, S.; Landmesser, U.; Harrison, D. G. J. Biol. Chem. 2002,
277, 25480.
4460
Org. Lett., Vol. 6, No. 24, 2004

Scheme 1. Synthesis of Radicicol’s Resorcinol Moiety
Figure 4. Inhibition of Hsp90’s ATPase activity.
suggesting that the incorporation of methyl ethers onto the
aromatic ring disrupts a key hydrogen bond network in the
ATP binding pocket. Andrus and co-workers recently
reported a similar result for the trimethoxy derivative of
GDA, which resulted in a substantial loss in cellular
efficacy.18
As mentioned previously, Hsp90 is responsible for the
conformational maturation of several polypeptides into
biologically active, three-dimensional structures. When the
Hsp90 protein folding process is disrupted, these client
proteins are unable to adopt their native structures, which
ultimately leads to their degradation. Therefore, one can
confirm Hsp90 inhibition by observing whether Hsp90 client
proteins are degraded by analysis of cellular lysates in the
presence of varying concentrations of Hsp90 inhibitors.
When increasing concentrations of 16, 1, and 15 were
incubated with MCF-7 breast cancer cells, a decrease in the
Hsp90-dependent client protein, HER-2 was observed by
Western Blot analysis (Figure 5), providing evidence that
the silyl protecting groups gave 14 and 15, of which 14 was
a direct precursor to 1. Following the procedure of Andrus
and co-workers,15 the methoxymethyl ethers were cleaved
with in situ-derived trimethylsilyl iodide to furnish the
hydroquinone product, 16. Oxidation of the hydroquinone
with palladium on carbon12 provided the paraquinone product
1.
Upon completion of the synthesis of 1, our efforts turned
toward evaluation of its biological activity. Recombinant
yeast Hsp90 was overexpressed and purified according to
the procedure of Buchner.7 The pure protein was incubated
with ATP and either 15, 16, or 1, and the production of
inorganic phosphate was determined (Figure 4).8
The IC50 of GDA as determined by this assay was
consistent with previously reported values (2.5 µM),8 and
the inhibitory activity of 16 and 1 was 1.8 and 5.9 µM,
respectively. The IC50 of 15 was higher than 40 µM,
Scheme 2. Preparation of Radamide
Figure 5. HER-2 degradation assays. Concentrations of drug in
µM are listed above each lane. V ) vehicle (1% DMSO).
these molecules also inhibit Hsp90 in cells. P85 is not an
Hsp90-dependent client protein and was used as a control.
In contrast to GDA, radamide (1) is prepared in a minimal
number of steps and is amenable to the rapid construction
of analogues that will be useful for elucidation of Hsp90
(18) Andrus, M. B.; Meredith, E. L.; Hicken, E. J.; Simmons, B. L.;
Glancey, R. R.; Ma, W. J. Org. Chem. 2003, 68, 8162.
Org. Lett., Vol. 6, No. 24, 2004
4461

structure-activity relationships. Studies are now underway
to prepare more active inhibitors of Hsp90 based upon this
chimeric species.
In conclusion, we have designed, synthesized, and pro-
vided biological data to support the hypothesis that a chimera
comprised of RDC and GDA is a novel lead compound for
the future development of Hsp90 inhibitors.
amycin was provided by the Developmental Therapeutics
Program at the National Cancer Institute. Recombinant yeast
Hsp90 was purified by Boris Kornikaev at The University
of Kansas Biochemical Research Service Laboratory. The
authors would like to thank Professors Gunda Georg, Robert
Hanzlik, and Rick Dobrowski for helpful discussions.
Supporting Information Available: Experimental pro-
cedures and characterization for all compounds. This material
is available free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. The authors gratefully acknowledge
support of this project by the NIH COBRE RR017708, The
University of Kansas Center for Research, and The Univer-
sity of Kansas New Faculty General Research Fund. Geldan-
OL048266O
4462
Org. Lett., Vol. 6, No. 24, 2004