Orally active purine-based inhibitors of the heat shock protein 90

Article abstract of DOI:10.1021/jm0503087

Orally active Hsp90 inhibitors are of interest as potential chemotherapeutic agents. Recently, fully synthetic 8-benzyladenines and 8-sulfanyladenines such as 4 were disclosed as Hsp90 inhibitors, but these compounds are not water soluble and consequently

Full text of DOI:10.1021/jm0503087

J. Med. Chem. 2006, 49, 817-828
817
Orally Active Purine-Based Inhibitors of the Heat Shock Protein 90
Marco A. Biamonte,*,§ Jiandong Shi,§ Kevin Hong,§ David C. Hurst,§ Lin Zhang,§ Junhua Fan,§ David J. Busch,†
Patricia L. Karjian,† Angelica A. Maldonado,† John L. Sensintaffar,† Yong-Ching Yang,† Adeela Kamal,† Rachel E. Lough,†
Karen Lundgren,† Francis J. Burrows,† Gregg A. Timony,‡ Marcus F. Boehm,§ and Srinivas R. Kasibhatla§
Conforma Therapeutics Corporation, 9393 Towne Centre DriVe, Suite 240, San Diego, California 92121
ReceiVed April 5, 2005
Orally active Hsp90 inhibitors are of interest as potential chemotherapeutic agents. Recently, fully synthetic
8-benzyladenines and 8-sulfanyladenines such as 4 were disclosed as Hsp90 inhibitors, but these compounds
are not water soluble and consequently have unacceptably low oral bioavailabilities. We now report that
water-solubility can be achieved by inserting an amino functionality in the N(9) side chain. This results in
compounds that are potent, soluble in aqueous media, and orally bioavailable. In an HER-2 degradation
assay, the highest potency was achieved with the neopentylamine 42 (HER-2 IC50 ) 90 nM). In a murine
tumor xenograft model (using the gastric cancer cell line N87), the H3PO4 salts of the amines 38, 39, and
42 induced tumor growth inhibition when administered orally at 200 mg/kg/day. The amines 38, 39, and 42
are the first Hsp90 inhibitors shown to inhibit tumor growth upon oral dosage.
Introduction
Heat Shock Protein 90 (Hsp90) is a molecular chaperone that
maintains the proper conformation of many “client” proteins.1
Inhibition of Hsp90 causes these client proteins to adopt aberrant
conformations, and these abnormally folded proteins are rapidly
eliminated by the cell via ubiquitinylation and proteasome
degradation. Interestingly, the list of Hsp90 client proteins
includes a series of notorious oncogenes. Four of them are
clinically validated cancer targets: HER-2/neu (Herceptin
(trastuzumab)), Bcr-Abl (Gleevec (imatinib mesylate)), the
estrogen receptor (tamoxifen), and the androgen receptor
(Casodex (bicalutamide)), while the others play a critical role
in the development of cancer. Some of the most sensitive Hsp90
clients are involved in growth signaling (Raf-1, Akt, cdk4, Src,
Bcr-Abl, etc.). In contrast, few tumor suppressor genes, if any,
seem to be clients of Hsp90,1,2 and consequently, inhibition of
Hsp90 has an overall antiproliferative effect. In addition, some
client proteins are involved in other fundamental processes of
tumorigenesis, namely apoptosis evasion (e.g. Apaf-1, RIP, Akt),
immortality (e.g. hTert), angiogenesis (e.g. VEGFR, Flt-3, FAK,
Figure 1. Hsp90 inhibitors.
to the Hsp90 inhibitor.2b,2c Another striking feature of Hsp90
is that it occurs in an activated form in cancer cells, and in a
latent form in normal cells.4 This provides an opportunity to
specifically target cancer cells with inhibitors selective for the
activated form,4 such as the natural product geldanamycin (1a,
Figure 1) and its derivatives. What distinguishes the activated
and latent forms of Hsp90 at a molecular level is not well
understood. It is clear, however, that the activity of Hsp90 is
regulated by a highly sophisticated process involving at a
minimum (1) Hsp90 dimerization, (2) formation of multiprotein
complexes with numerous co-chaperones, and (3) ATP/ADP
binding, ATP hydrolysis being essential for the chaperone cycle
and function.1 The chaperoning function of Hsp90 can be
“switched off” by inhibiting its ATP-ase activity. The nucle-
otides ADP and ATP can bind to two sites, one located close
to the N-terminal, the other close to the C-terminal. The natural
products geldanamycin (Figure 1, 1a) and radicicol (2) bind to
the N-terminal domain, while novobiocin binds to the C-terminal
domain.5
HIF-1), and metastasis (c-Met).
The various client proteins are not equally responsive to
Hsp90 inhibitors, and some undergo degradation at lower
concentrations of the inhibitor, or with faster kinetics, depending
on the cell line. The more sensitive clients are usually those
involved in growth signaling, but some mutated proteins found
in tumor cells (mutant p53, Gleevec-resistant Bcr-Abl)3 are
particularly dependent on Hsp90 to preserve their conformation
and function. This unique feature sensitizes tumor cells to Hsp90
inhibitors, and when these factors converge, they confer on
Hsp90 inhibitors notable anticancer properties in vitro and in
vivo.
A remarkable advantage of targeting Hsp90 lies in the
simultaneous depletion of multiple oncogenic proteins, thereby
attacking several pathways necessary for cancer development
and reducing the likelihood of the tumor acquiring resistance
* Corresponding author. Tel: +1 858 795 0109. Fax: + 1 858 657 0343.
E-mail: mbiamonte@conformacorp.com.
§ Department of Medicinal Chemistry.
† Department of Biology and Pharmacology.
‡ Department of Pre-Clinical Development.
Many of the natural-product-derived Hsp90 inhibitors exhibit
pharmaceutical deficiencies. For example, the semisynthetic
10.1021/jm0503087 CCC: $33.50 © 2006 American Chemical Society
Published on Web 12/24/2005

818 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 2
Scheme 1. Preparation of 8-Benzyladeninesa
Biamonte et al.
a Reagents and conditions: (a) BuNH2, Et3N, n-BuOH, 90 °C, 16 h (83%); (b) 2,5-dimethoxyphenylacetic acid, TsCl, Et3N, DCE, 40 °C, 16 h (71%); (c)
NH3, MeOH, 120 °C, 3 d (80%); (d) NaOH, H2O, 0-5 °C (67%); (e) 3-methoxyphenylacetyl chloride, NMP, 40-50 °C (98%); (f) MeONa, n-BuOH, reflux,
2 h (75%); (g), BuI, Cs2CO3, DMF, rt, 16 h (54%); (h) SO2Cl2, Br2, or NIS (31-57%).
inhibitor 17-allylamino,17-desmethoxy-geldanamycin (17-AAG,
1b), currently in phase II clinical trials, is expensive to
manufacture, difficult to formulate,6 and at present administered
only parenterally. Although the 17-dimethylaminoethylamino
analogue (17-DMAG, 1c) is more soluble, it exhibits all of the
side effects of 17-AAG as well as gastrointestinal hemorrhaging
in preclinical toxicity studies.7 Radicicol (2, Figure 1), another
natural product Hsp90 inhibitor, is poorly water-soluble and is
inactive in tumor xenograft models. Semisynthetic oxime
derivatives of radicicol provide better solubility and substantially
improved the pharmacological profile in murine models, but
are still limited to intravenous administration.8
Fully synthetic, orally active inhibitors of Hsp90 have been
sought in order to provide more flexible dosing schedule options
and to possibly avoid the side-effects of the natural product
inhibitors. Recently, several novel nonnatural product Hsp90
inhibitors were reported.1b,9-11 The structures of these inhibitors
were designed using the crystal structures of Hsp90 in complex
with ATP, geldanamycin, or radicicol. The 8-benzyladenines
such as PU3 (3a)9 were designed to adopt the same C-shaped
conformation as geldanamycin12 with the adenine ring pointing
to the adenine-binding site (hinge region), and the trimethoxy-
benzene ring emulating the H-bond accepting nature of the
quinone ring of geldanamycin.13 The recently obtained10c crystal
structure of Hsp90 in complex with 3a confirmed that the purine
ring occupies the position normally occupied by ADP/ATP, but
the benzene ring points in a direction opposite to the predicted
one, to form a π-stacking interaction with Phe138. Nevertheless,
growth.9b Very high doses (500-1000 mg/kg) of 3b were
required to observe a similar pharmacodynamic response upon
oral administration,9b and no 8-benzyladenine has been reported
to inhibit tumor growth by the oral route. In our hands, 3b
proved to be too insoluble to be effectively formulated and
delivered orally. So far, despite extensive SAR studies to
improve potency and pharmaceutics properties, Hsp90 inhibitors
have not demonstrated activity in animal models of human
cancer (xenografts) when administered orally.
The discovery of the 8-benzyladenines led us14 and others15
to design 8-sulfanyladenines, exemplified by 8-(2-iodo-5-
methoxy-phenylsulfanyl)-9-pent-4-ynyl-9H-purin-6-ylamine (4).
Compound 4 exhibited excellent potency in several cell-based
assays but was poorly soluble in water and did not have
sufficient oral bioavailability in clinically acceptable formula-
tions. We now provide details on the SAR studies that led to
the discovery of 4 and further expand the SAR studies with the
introduction of solubilizing groups in the N(9) side-chain, which
provide water-solubility, oral bioavailability, and subsequently,
oral efficacy in tumor xenograft models.
Chemistry
Our strategy for improving the activity of purine-based Hsp90
inhibitors was to independently optimize the substituents on the
benzene ring of 8-benzyladenines and the nature of the linker
spanning between the benzene and the purine rings. We would
then combine the preferred structural elements emerging from
both optimization runs and select for compounds exhibiting
acceptable pharmaceutical properties. This plan allowed us to
take full advantage of the known methods for the preparation
of 8-benzylpurines, although in some cases refinements proved
to be necessary. The 8-benzyladenines were synthesized by
either of the two methods illustrated in Scheme 1. The first
method followed a sequence closely related to the one described
by Drysdale et al.10 and started from commercial 4,6-dichloro-
5-aminopyrimidine 5, which was treated with butylamine,
acylated with the appropriate phenacyl chloride, and cyclized
to afford the adenine 8. The second method was similar to the
3a inhibits Hsp90 (HER-2 degradation assay, HER-2 IC50
)
40 µM) and afforded a valuable starting point for further
optimization. Structure-activity studies based on 3a led to the
more active PU24FCl (3b, HER-2 IC50 ) 1.7 µM)9 which was
subsequently also cocrystallized10 with Hsp90. When 3b was
formulated in DMSO/EtOH/phosphate-buffered saline 1:1:1 and
administered intraperitoneally to mice bearing MCF-7 xenograft
tumors, it induced at 100-300 mg/kg down-regulation of HER-2
and Raf-1, a pharmacodynamic response consistent with Hsp90
inhibition, and at 200 mg/kg it significantly repressed tumor

Hsp90 Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 2 819
Scheme 3. Preparation of 8-Sulfanyladenines 23, 4, and 28a
Scheme 2. Preparation of 8-(Arylsulfanyl)adenine 18a
a Reagents and conditions: (a) BuI, Cs2CO3, DMF, rt, 16 h (62%); (b)
2,5-dimethoxybenzenethiol, 150 °C, 4 h (69%).
one of Chiosis et al.9 and started with 4,5,6-triaminopyrimidine
9, which was acylated and cyclized to give the adenine 11. The
final alkylation gave predominantly the desired N(9)-alkyl
isomer 12a, together with a minor regioisomer which was
removed by chromatography (regioselectivity ) 5:1 by 1H NMR
analysis of the crude product). The anisole 12a was halogenated
using standard reagents (SO2Cl2, Br2, NIS/AcOH). One im-
provement in the synthetic sequence pertained to the acylation
step. The published method9a involves acylation of 4,5,6-
triaminopyrimidine hemisulfate (9) in aqueous solution (9 is
soluble only in water at pH g 7) and required, in our hands,
several equivalents of the appropriate acyl fluoride to compen-
sate for the accompanying hydrolysis of the reagent. We found
that the free base of 9 was readily isolated as needles by
neutralizing and cooling to 0-5 °C an aqueous solution of the
commercial 4,5,6-triaminopyrimidine hemisulfate. The free base
proved to be soluble in N-methyl-2-pyrrolidone (NMP) and
could be efficiently acylated in this solvent with a single
equivalent of acyl chloride to give the desired amide 10, which
precipitated as its HCl salt. The use of DMF as a solvent was
less satisfactory, since it gave rise to a competitive formylation
of the 5-NH2 group via a Vilsmeier-Haack type of reaction.
Bases such as Et3N were best avoided, to prevent overacyla-
tion.16 The symmetry of the 1H NMR spectrum of 10 indicated
that the acylation had occurred selectively at the 5-position.
Finally, the cyclization of 10 to the desired purine 11 was carried
out with MeONa in refluxing n-BuOH, a minor deviation from
the original MeOH, but which provided more forceful and
generally applicable conditions.
A different approach was necessary to investigate the effect
of the linker between the purine and the benzene ring. The
compounds with a sulfur atom as a linker were prepared
according to the example shown in Scheme 2. 8-Bromoadenine
(16)17 was alkylated to give a 2:1 mixture of the N(9)- and N(3)-
alkylated isomers, from which the desired N(9)-butyl isomer
17 could be isolated by chromatography. Displacement of the
bromine atom with the desired thiophenolate gave the 8-sulfa-
nyladenine 18.
A similar approach was used to generate 8-(benzenesulfanyl)-
adenines carrying an iodo substituent on the benzene ring via
the convergent synthesis shown in Scheme 3. The nitroaniline
19 was converted in three known steps (1. NaNO2, KI. 2. H2N-
NH2/Fe. 3. NaNO2, HBF4) to the diazonium salt 20, which gave
after a two-step Leuckart synthesis (4. EtOCS2K, 5. KOH) the
thiophenolate 21.18 The purification of 21 proved to be chal-
lenging. Dissolution of 21 in MeOH and precipitation with
EtOAc gave a low recovery (<20%) of the desired thiopheno-
late, while neutralization and chromatography was complicated
by the odor of the thiophenol and its tendency to oxidize to the
corresponding disulfide. The highest overall yield for the
conversion of 20 to 23 was achieved by hydrolyzing the
intermediate xanthate with 2 equivalents of KOH in MeOH,
concentrating the reaction mixture, and using it without remov-
a Reagents and conditions: (a) EtOCS2K, -40 °C to room temperature,
30 min, then KOH, MeOH, 40 °C, 2 h; (b) 5-bromo-2-methyl-pent-2-ene,
Cs2CO3, DMF, rt, 3 h (38%); (c) 21, DMF, 100 °C, 16 h (15%); (d) 5-chloro-
1-pentyne, Cs2CO3, DMF, 70 °C, 3 h (25%); (e) 21, DMF, 70 °C, 16 h
(71%); (f) 5-chloro-1-pentyne, Cs2CO3, DMF, 85 °C, 16 h (66%); (g) 21,
DMF, 100 °C, 16 h (43%); (h) iso-AmONO, HBF4, THF, -20 °C to +40
°C, 10 min (13%).
ing the excess of potassium salts. Alkylation of 8-bromoadenine
(16) with homoprenyl bromide gave a 2:1 mixture of the desired
N(9)- vs. N(3)-alkylated products, and the desired isomer 22
was isolated by chromatography. Coupling the thiophenolate
21 with the 8-bromoadenine 22 gave the sulfanyladenine 23.
Similarly, the 8-bromoadenine 16 was alkylated with 5-chloro-
1-pentyne to give 24 and coupled to the thiophenolate 21 to
provide the pentyne analogue 4. The 2-fluoro-8-sulfanyladenine
28 was prepared by a conceptually similar route starting from
2,6-diamino-8-bromopurine (25) and using a Balz-Schiemann
reaction (iso-AmONO/HBF4) to replace the 2-NH2 group by a
fluorine substituent.
This route suffered from two drawbacks. First, the alkylation
was not regioselective and required a tedious chromatographic
separation. Second, the thiophenolate 21 was malodorous,
difficult to purify, and if used without chromatographic purifica-
tion, gave results which were very sensitive to its purity. One
improvement (Scheme 4, route A) consisted of starting from

820 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 2
Biamonte et al.
Scheme 4. Preparation of Water-Soluble 8-Sulfanyladenines of Generic Structure Aa
a Reagents and conditions: (a) AcO-(CH2)2-Br, Cs2CO3, DMF, 45 °C, 5 h (67%) or AcO-(CH2)3-Cl, Cs2CO3, DMF, 70 °C, 12 h (88%), then Br2,
AcOH, THF, H2O (34-81%); (b) 21, DMF, 50 °C, 16 h; then K2CO3, MeOH (37%); (c) MsCl, Et3N, dioxane, rt, 30 min; then amine, 55 °C, 16 h (45-
75%); (d) 21, DMF, -60° to room temperature, 2h (26%); (e) Br-(CH2)3-Cl, Cs2CO3, DMF, 50 °C, 2 h (21%); (f) amine, 90-110 °C, 16 h, (67-73%);
(g) thiourea, n-BuOH, reflux, 3 h (88%)ssee also ref 19; (h) 20, DMSO, rt, 24 h (51%); (i) K2CO3, MeOH (37%).
pure adenine (29) which underwent alkylation solely at N(9).
We chose as alkylating agents either Br-(CH2)2-OAc or Cl-
(CH2)3-OAc, in which the masked hydroxyl group provided a
handle for further functionalization. Unlike the unsubstituted
adenine, the alkylated product was easily brominated at C(8)
to give 30 (n ) 2, 3). The bromine atom of 30 was then
displaced with the potassium thiophenolate 21, and the acetyl
protecting group was cleaved in situ to give the 8-sulfanylad-
enine 31. The hydroxy group of 31 was mesylated and displaced
with amines to give the corresponding N-alkylamines of generic
structure A (Scheme 4).
This route, however, still required the disagreeable preparation
of the thiophenolate 21. We therefore investigated the direct
treatment of the diazonium salt 20 with the anion of 8-thion-
oadenine 32, which already contained the desired sulfur atom
(Scheme 4, route B), thus avoiding the preparation of 21. We
synthesized 32 by condensation of 4,5,6-triaminopyrimidine with
thiourea,19 and treatment of 32 with the diazonium salt 20 to
give the desired adduct 33. The yield (unoptimized) was low,
and the subsequent alkylation of 33 with Br-(CH2)3-Cl still
gave a 2:1 mixture of N(9)- and N(3)-regioisomers. Despite
these problems, the route was effective in generating small
amounts of amines of formula A. However, an improved route
was needed to prepare gram amounts of A. A third route was
selected (Scheme 4, route C) which eliminated the use of the
thiophenolate 21 and gave only one regioisomer. Bromoadenine
30b (route A) was converted with thiourea to the 8-thionoad-
enine 35,19 which was coupled directly with the diazonium salt
20 to give the desired, known adduct 36.19 Finally, the same
standard manipulations of the side-chain (deacetylation, mesy-
lation, amination) gave the desired amines of generic formula
A. Routes A and C gave similar overall yields and were used
interchangeably, but the latter was consistently reproducible and
avoided the use of 21.
Results and Discussion
The potency of the compounds was assessed using a HER-2
degradation assay, which was described in detail elsewhere.20
Briefly, compounds were incubated for 16 h with MCF-7 cells,
a breast cancer cell line expressing on its surface medium levels
of the Hsp90 client HER-2. Inhibition of Hsp90 induces the
degradation of HER-2, which was monitored with a combination
of phycoerythrin-labeled antibody and flow cytometry. This
assay is highly reproducible, with 17-AAG (1b) consistently
giving an HER-2 IC50 of 12.9 ( 0.3 nM, wherein the error refers
to the standard error of the mean (SEM).
We first investigated the effect of substituents on the benzene
ring. The 2,5-dimethoxy substitution pattern emerged as more
potent than the prototypic 3,4,5-trimethoxy pattern of 3a and
3b (Table 1, entries 1, 2). Replacing the 2-MeO group by Cl
marginally decreased the activity (entry 3), but replacing it with
Br or I led to an increase in activity (entries 4, 5). We next
investigated the effect of the linker. The compounds with a NH
or O as linker are inactive,9 and it was assumed that only the
CH2 linker could be tolerated. However, upon introduction of
an S linker (Table 1, entry 4), we were gratified to observe that
the sulfur atom was not only tolerated, but was superior to the
original CH2 linker (Table 1, entry 5).14
We next optimized the N(9) side-chain and screened over
100 analogues of the 2,5-dimethoxybenzyl adduct 8.21 The
homoprenyl side-chain emerged as an equipotent alternative to
the already disclosed9 pent-4-ynyl side-chain, both analogues
having an HER-2 IC50 ) 1.5 µM. Thus, having separately

Hsp90 Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 2 821
Table 3. Effect of Amino Group in Side-Chain
Table 1. Optimization of the Benzene Ring Substituentsa
entry
ID
3b
n
R
HER-2 IC50 [µM]a
MTS IC50 [µM]b
entry
ID
L
X
HER-2 IC50 [µM]
1
2
3
4
5
6
7
8
1.7
1.2
0.7
0.2
0.2
0.6
0.2
0.2
0.5
4
0.29
0.21
0.21
0.18
0.14
0.10
0.09
1
2
3
4
5
6
3a
8
12b
12c
12d
18
CH2
CH2
CH2
CH2
CH2
S
3,4,5-triMeO
2,5-diMeO
2-Cl, 5-MeO
2-Br, 5-MeO
2-I, 5-MeO
2,5-diMeO
40
12
20
8.0
5.0
3.5
37
38
39
40
41
42
3
3
3
3
2
2
Et2CH-
EtMeCH-
i-Pr-
t-Bu-
i-Bu-
t-BuCH2-
a All values represent the average of at least three independent observa-
tions. The standard errors of the mean (SEM) are 6-11% of the mean value.
a For the HER-2 degradation assay, the values represent the average of
at least three independent observations, and the standard errors of the mean
(SEM) are 6-11% of the mean value. b For the growth inhibition assay,
the values represent the average of at least three independent observations,
and the standard errors of the mean (SEM) are 9-21% of the mean value.
Table 2. Combinations of the Preferred Structural Elementsa
inhibitor that gave rise to 50% less new live cells compared to
an untreated culture. In this assay, the control 17-AAG (1b)
had an MTS IC50 of 32 ( 4 nM, and the standard error of the
mean (SEM) associated with this assay ranged from 9 to 21%
of the mean value. The 8-(sulfanyl)adenines proved to be inhibit
cell growth (Table 3), with MTS IC50 values typically in the
200-500 nM range, which is roughly within 1 logarithmic unit
of the “gold standard” 17-AAG.
The selectivity of 42 for Hsp90 over other ATP-binding
proteins was assessed with a panel of human kinases (Aurora-
A, CHK2, IKKR, MAPK1, MAPK2, MEK1, PDK1, Plk3, PI-
3K, c-Raf, c-Src), none of which were significantly inhibited
at 10 µM.
Perhaps the most important feature of these compounds,
besides their potency, was their dramatically increased water
solubility. Once converted to their H3PO4 salt, these amines
provided excellent water solubility (>10 mg/mL) and were
readily administered in standard aqueous solutions. For animal
studies, we chose to formulate the compounds in a phosphati-
dylcholine/water dispersion.22 The pharmacokinetic properties
of these compounds were determined in Balb/C mice (Table 4)
as described previously.20 When the compounds were admin-
entry
ID
X
R
HER-2 IC50 [µM]
1
2
3
4
23
28
H
H
F
pent-4-ynyl
homoprenyl
pent-4-ynyl
0.28
0.37
0.36
a All values represent the average of at least three independent observa-
tions. The standard errors of the mean (SEM) are 6-11% of the mean value.
optimized the benzene ring substituents (2-iodo-5-methoxy), the
linker (-S-), and the side-chain (homoprenyl or pent-4-ynyl),
we examined the combination of these preferred structural
features (Table 2). The homoprenyl analogue 23 and the pent-
4-ynyl analogue 28 had similar potencies (HER-2 IC50 ≈ 0.3
µM, entries 1, 2). However, the addition of a 2-F substituent
on the adenine ring, an operation known to be favorable in the
8-benzyladenine series,9 did not bring additional activity to the
8-sulfanyl series (entry 3).
Although these compounds exhibited improvements in po-
tency over previously reported Hsp90 inhibitors, they proved
to be poorly water-soluble, especially with the 2-iodo substituent.
This hampered their formulation and rendered them insuf-
ficiently orally bioavailable. We therefore sought to incorporate
ionizable amino groups in the N(9) side-chain of the inhibitor.
The introduction of the amino group not only improved the
water solubility but also increased the potency. The highest
potencies were obtained when the amino N atom was separated
by two or three methylene units from the purine ring and was
further substituted with a bulky alkyl group (Table 3). The most
active compound in the three-atom linker series proved to be
the tert-butylamine 40 (HER-2 IC50 ) 140 ( 15 nM, entry 6),
while in the two-atom linker series the neopentylamine 42
(HER-2 IC50 ) 90 ( 10 nM, entry 8) showed optimal activity.
istered orally at 100 mg/kg, peak plasma concentrations (Cmax
)
between 4.8 and 9.7 µg/mL (10-19 µM) were achieved. The
plasma concentrations peaked at Tmax ) 30 min, indicating rapid
absorption, and dropped below the detection limit (0.5 µg/mL,
1 µM) after 1-4 h to give, when integrated over a 4 h period,
AUC values of 240 to 680 min‚µg/mL (equivalent to 8-22 µM‚
h). The effect of the solubility on the oral bioavailability was
striking, and the %F increased from <10% for pentyne 4 (data
not shown) to 14-97% for the amines 37-42. When admin-
istered intravenously at 10 mg/kg, the amines 37-42 were
cleared at the rate of 33-131 mL/min/kg, which is quite high
compared to the total liver blood flow (90 mL/min/kg for mice).
We did not determine, however, if the clearance was due to
metabolism, distribution, or low protein binding. By analogy
with the structurally related adenine 3b,9b it is also possible that
the inhibitors 37-42 accumulate in the tumor to concentrations
exceeding those in the plasma. Despite their high clearance at
10 mg/kg, the oral bioavailability of compounds 39, 40, and 42
at 100 mg/kg was equal to or greater than 50%, suggesting that
the clearance was saturated at 100 mg/kg.
The amines 37-42 were tested for their ability to inhibit cell-
growth, using a previously described assay20 to quantify cell
proliferation. In brief, MCF-7 breast cancer cells were incubated
for 5 days with the compound and then treated with MTS (3-
(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-
sulfophenyl)-2H-tetrazolium). The MTS reagent is reduced only
by metabolically active cells to the formazan dye, and the
number of live cells was deduced by spectrophotometry (490
nM). The MTS IC50 was defined as the concentration of Hsp90
This is to our knowledge the first time that pharmacologically
relevant concentrations of Hsp90 inhibitors were achieved via

822 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 2
Biamonte et al.
Table 4. Pharmacokinetic Parameters of Selected Amines
iv parameters
po parameters
entry
compound
Cl [mL/min/kg]
T
1/2 [h]
Vss [L/kg]
Cmax [µg/mL]
Tmax [h]
AUC [min‚µg/mL]
T1/2 [h]
%F [%]
1
2
3
4
5
6
37
38
39
40
41
42
69
33
62
131
47
64
0.5
1.3
4.2
1.6
0.3
0.4
2.4
3.4
22
24
0.4
1.6
5.8
4.8
4.7
6.7
6.7
9.5
0.5
0.5
0.5
0.5
0.5
0.5
590
380
450
680
290
760
0.6
0.7
1.5
0.9
0.3
0.5
42
28
55
97
14
50
a The compounds were formulated as H3PO4 salts in a phosphatidylcholine/water dispersion and delivered intravenously (iv) at 10 mg/kg or orally (po)
at 100 mg/kg. Plasma concentrations were measured at six time points over 4 h, and the pharmacokinetic parameters were determined using noncompartmental
methods (WinNonlin Professional, Version 4.1). The terminal half-life was calculated using 3-4 data points.
Figure 2. (a) Levels of Hsp90 clients, Hsp70, and PI-3K p85 in murine
A549 tumor xenografts following a single oral administration of 41‚
H3PO4 at 200 mg/kg. (b) Levels of Hsp90 clients and PI-3K in murine
N87 tumor xenografts 24 h after a three-day course of 17-AAG
(intraperitoneally, 1 × 90 mg/kg/day) or 39‚H3PO4 (orally, 2 × 200 or
2 × 100 mg/kg/day).
the oral route, and the results compiled in Table 4 suggested
that these inhibitors may be orally active. For instance, a Cmax
) 5.8 µg/mL (37) corresponds to a concentration of 12 µM,
which is approximately 50-fold higher than the concentrations
required to either induce HER-2 degradation in MCF-7 cells
(HER-2 IC50 ) 0.21 µM) or to inhibit the proliferation of MCF-7
Figure 3. Tumor growth inhibition in murine N87 xenografts models
cells (MTS IC50 ) 0.2 µM). The plasma concentration of the
induced by (a) inhibitors 38‚H3PO4 and 39‚H3PO4 delivered orally (1
amines 37-42 remained above 1 µM, the detection limit, for
× 200 mg/kg/day, 5 days/week) or (b) inhibitor 42‚H3PO4 delivered
1-4 h.
orally (2 × 100 mg/kg/day, 5 days/week). Error bars ) SEM.
The amines 37-42 provided a good combination of potency,
ease of formulation, and bioavailability but displayed relatively
proteins. Once degraded, those client proteins require 6-48 h
high clearance values in mice. We next verified the ability of
to accumulate back to their normal levels, and even if the Hsp90
the arbitrarily chosen amine 41 to induce the degradation of
inhibitor is rapidly cleared as 41, its pharmacological effect can
Hsp90 clients in vivo. Nude mice were implanted with A549
be long lasting. This behavior differs significantly from that of
lung cancer cells, a cell line dependent on the Hsp90 clients
most ATP-competitive kinase inhibitors which, once cleared,
Raf-1 and Akt for cell proliferation, and were administered a
allow their target to immediately resume its function.23
single oral dose of 41‚H3PO4 (200 mg/kg). The mice were
sacrificed at 6, 24, or 48 h, the tumors were harvested, and
The pharmacodynamic effect of amine 39 was examined in
a N87 xenograft model (Figure 2b), N87 being a stomach cancer
cell line expressing high HER-2 levels. Mice were administered
39‚H3PO4 orally at two different regimens (2 × 100 or 2 ×
200 mg/kg/day) for 3 days and were sacrificed 24 h after the
last dosing. Oral administration of 39‚H3PO4 at 2 × 200 mg/
kg/day induced the degradation of the Hsp90 clients Akt, pAkt,
Raf-1, pRaf, cdk6, and pRb to levels comparable to those
obtained with 17-AAG injected intraperitoneally once daily at
90 mg/kg/day. The levels of HER-2 and pHER-2 decreased only
partially, probably reflecting the fact that HER-2 and pHER-2
are degraded and re-expressed faster (<24 h) than other Hsp90
clients. A 39‚H3PO4 dose of 2 × 100 mg/kg/day was still
Hsp90 client proteins were visualized by Western blot. The
levels of the Hsp90 clients HER-2 and pHER-2 significantly
decreased at 6 h, and then gradually reached their normal value
after 24-48 h (Figure 2a). The levels of the Hsp90 clients pAKT
and pRaf and the downstream kinase pERK decreased less
dramatically and were lowest at 24 h. Upregulation of the
chaperone Hsp70, a response characteristic of Hsp90 inhibition,
was evident and lasted 24-48 h. As expected, the kinase PI-
3K, which is not an Hsp90 client, was unaffected. These
pharmacodynamic data underscore an added benefit of targeting
Hsp90, since exposing tumor cells to an Hsp90 inhibitor for a
few hours is sufficient to induce the degradation of the client

Hsp90 Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 2 823
in research established by the Institute for Laboratory Animal
Research (ILAR).
effective at degrading Akt, pAkt, Raf-1, pMEK, cdk6, and pRb
but promoted little or no degradation of HER-2, pHER-2, and
pRaf.
The test compounds were formulated (3-20 mg/mL) in a
phospholipon 90G/water dispersion.22 The dispersion was prepared
with 86 g of sucrose and 53 g of phospholipon, brought to 1000 g
with water, homogenized for 15 min, microfluidized at 14 000-
16 000 psi (F12Y interaction chamber), and stored at 5 °C. The
test compounds were administered either orally by gavage at 5-20
mL/kg or intravenously. For pharmacokinetic studies, three mice
were administered the inhibitor orally and three others intravenously.
Six blood samples were collected with the final sample collected 4
h postdosing, and plasma concentrations were measured by HPLC.
The data were analyzed using noncompartmental methods, and the
terminal half-life was estimated using 3-4 data points (WinNonlin
Professional, Version 4.1).
Next, the ability of a subset of amines (38, 39, and 42) to
repress tumor growth was examined in murine xenograft models
using the N87 stomach cancer cell line, which grew in mice
more reproducibly than the A549 cell line. Compounds 38‚H3-
PO4 and 39‚H3PO4 were delivered orally at 200 mg/kg/day (once
daily, 5 days/week), in the same experiment (Figure 3a). Tumor
growth inhibition was observed for both compounds but with a
lower statistical significance for 38‚H3PO4 (p ) 0.07) compared
to 39‚H3PO4 (p ) 0.03). At these doses, neither mortality nor
weight loss was observed. Similarly, the compound most active
in the HER-2 degradation assay, 42‚H3PO4, was tested in a
separate experiment (Figure 3b), at 200 mg/kg/day but with a
different schedule (2 × 100 mg/kg/day, 5 days/week), and also
showed statistically significant (p ) 0.02) tumor growth
inhibition and no overt toxicity.
Tumor fragments (approximately 2 mm3) or 5 × 106 tumor cells
were inoculated subcutaneously in the right or left flank of the
animal. Mice with established tumors (50-200 mm3) were selected
for study (n ) 6-10/treatment group). Tumor dimensions were
measured using calipers, and tumor volumes were calculated using
the equation for an ellipsoid sphere (l × w2)/2 ) mm3, where l and
w refer to the larger and smaller dimensions collected at each
measurement. The vehicle alone was administered to control groups.
For pharmacology studies, animals were dosed 5 days per week
(Monday through Friday) for four to six consecutive weeks. Animals
were weighed, and the tumors were measured twice per week. Mice
were followed until tumor volumes in the control group reached
approximately 1000 mm3 and were sacrificed by CO2 euthanasia.
In those rare instances in which there was a clear outlier (over two
standard deviations from the mean), that outlier was removed. The
mean tumor volumes of each group were calculated. The change
in mean treated tumor volume was divided by the change in mean
control tumor volume, multiplied by 100, and subtracted from 100%
to give the tumor growth inhibition for each group.
Conclusion
The chaperone Hsp90 is a target of interest for the treatment
of cancer because of its central regulatory role. Inhibition of
Hsp90 induces the degradation of several client proteins and
shuts down multiple oncogenic pathways, which in turn affects
a number of critical steps implicated in the genesis of a tumor
(proliferation, angiogenesis, acquired immortality, evasion of
apoptosis, and metastasis). The simultaneous modulation of
various oncogenic effects should reduce the likelihood of the
tumor acquiring resistance to Hsp90 inhibitors.2b,2c In addition,
the existence of an activated form of Hsp90 in cancer cells offers
the possibility to develop inhibitors selective for malignant cells.
We optimized the purine-based inhibitors of Hsp90, reaching
compounds as potent as 90 nM (42) in the HER-2 degradation
assay and 200 nM in in vitro growth inhibition assays. The
introduction of an amino group in the side-chain dramatically
improved their aqueous solubility (>10 mg/kg for their H3PO4
salts), which greatly facilitated their formulation for oral
delivery. In mice, the oral bioavailability of the amines 37-42
ranged from 14 to 97%. These amines reached high plasma
concentrations (Cmax ) 10-19 µM; oral dose of 100 mg/kg)
but were cleared rapidly (Cl ) 33-131 mL/min/kg; intravenous
dose of 10 mg/kg). When administered orally to mice bearing
A549 tumor xenografts (200 mg/kg), 41‚H3PO4 induced the
pharmacodynamic response expected from Hsp90 inhibitors:
degradation of the client proteins HER-2, pHER-2, pAKT, and
pRaf and upregulation of Hsp70. Similarly, in a murine N87
xenograft model, oral administration of 39‚H3PO4 (2 × 200 mg/
kg/day) induced the degradation of Hsp90 clients but not of
PI-3K. Furthermore, in the N87 model, the H3PO4 salts of set
of amines 38, 39, and 42 inhibited tumor growth orally at 200
mg/kg/day. These are the first Hsp90 inhibitors reported to
inhibit tumor growth upon oral administration, but high doses
are currently necessary. Further work is necessary to improve
the potency and clearance of these compounds and to examine
alternate xenograft models.
Chemistry. The 1H NMR and 13C NMR spectra were re-
corded on a 400 MHz spectrometer. Analytical HPLC chroma-
tograms were obtained using a C18 column (5 µm; 4.6 mm × 150
mm). A gradient was applied between solvent A (0.1% TFA in
H2O) and solvent B (0.05% TFA in CH3CN) increasing the
proportion of B linearly from 5% to 100% over 7 min (conditions
I), 12 min (conditions II), or 15 min (conditions III) with a constant
flow rate of 1 mL/min. The samples were diluted to 0.1 mg/mL in
MeOH, and the injection volumes were 10 µL. The column was
not heated, and UV detection was effected at 254 nm. Purities were
determined by HPLC at 254 nM.
N4-Butyl-6-chloropyrimidine-4,5-diamine (6). A suspension of
4,6-dichloro-5-aminopyrimidine (5) (4.0 g, 24.4 mmol) in n-BuOH
(35 mL) was treated with BuNH2 (2.7 mL, 26.8 mmol, 1.1 equiv)
and Et3N (4.1 mL, 29.3 mmol, 1.2 equiv) and heated to 90 °C for
16 h. Evaporation, workup (EtOAc/water/brine), and drying (Na2-
SO4) gave the crude title compound as a light yellow solid (4.0 g,
83%). HPLC Purity: 99.8%. tR ) 5.08 min (Conditions I). mp:
1
81-82 °C. H NMR (DMSO-d6) δ 7.73 (s, 1H), 6.76 (t, J ) 4.8
Hz, 1H), 5.00 (s. 2H), 3.37 (q, J ) 7.0 Hz, 2H), 1.52 (quint., J )
7.4 Hz, 2H), 1.33 (sext., J ) 7.4 Hz, 2H), 0.83 (t, J ) 7.4 Hz, 3H).
13C NMR (DMSO-d6) δ 152.5, 146.2, 137.0, 123.8, 41.0, 31.3, 20.1,
14.2.
9-Butyl-8-(2,5-dimethoxy-benzyl)-9H-purin-6-ylamine (8). A
solution of 2,5-dimethoxyphenylacetic acid (6.46 g, 24.7 mmol,
1.1 equiv) in 1,2-dichloroethane (100 mL) was treated with TsCl
(4.7 g, 24.7 mmol, 1.1 equiv) and Et3N (9.4 mL, 67.2 mmol, 3.0
equiv) at room temperature for 20 min. Then, N4-butyl-6-chloro-
pyrimidine-4,5-diamine (4.5 g, 22.4 mmol, 1.0 equiv) was added,
and the solution was stirred at 40 °C for 16 h. Workup (EtOAc/
water/brine) and drying (Na2SO4) gave N-(4-butylamino-6-chloro-
pyrimidin-5-yl)-2-(2,5-dimethoxy-phenyl)-acetamide (7) as a crude
(6.0 g, 71%) which was used without further purification.
A solution of N-(4-butylamino-6-chloro-pyrimidin-5-yl)-2-(2,5-
dimethoxy-phenyl)-acetamide (7) (2.5 g, 6.6 mmol) in methanolic
ammonia (7 M NH3 in MeOH, 30 mL) was heated in a stainless
steel cylinder to 120 °C for 3 days. Flash chromatography (EtOAc/
Experimental Section
Pharmacokinetic, Pharmacodynamic, and Pharmacology. Six
to eight week old nu/nu athymic female mice were maintained in
sterilized filter topped cages in a room with a 12-h light/12-h dark
photoperiod at a temperature of 21-23 °C and a relative humidity
of 50 ( 5%. Irradiated pelleted food and autoclaved deionized water
were provided ad libitum. Animals were identified by the use of
individually numbered ear tags. Experiments were carried out under
institutional guidelines for the proper and humane use of animals

824 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 2
Biamonte et al.
hexane) gave the title compound as a white solid (1.8 g, 80%). tR
) 5.55 min (Conditions I). mp: 126-130 °C. 1H NMR (CDCl3) δ
8.34 (s, 1H), 6.86 (d, J ) 8.9 Hz, 1H), 6.80 (dd, J ) 8.9 & 3.0 Hz,
1H), 6.70 (d, J ) 3.0 Hz, 1H), 5.64 (br. s, 2H), 4.25 (s, 2H), 4.11
(t, J ) 7.7 Hz, 2H), 3.84 (s, 3H), 3.72 (s, 3H), 1.63 (quint., J )
7.6 Hz, 2H), 1.32 (sext., J ) 7.7 Hz, 2H), 0.87 (t, J ) 7.4 Hz, 3H).
13C NMR (CDCl3:CD3OD 5:1) δ 154.5, 153.5, 151.7, 151.5, 151.0,
150.4, 124.6, 117.8, 116.5, 112.6, 111.4, 55.7, 55.5, 42.7, 31.7,
27.9, 19.8, 13.4.
4,5,6-Triaminopyrimidine (9). 4,5,6-Triaminopyrimidine sulfate
(50 g, 224 mmol) was dissolved in aq 1 M NaOH (500 mL, 2.2
equiv), and the pH was adjusted to 7 with 4 M HCl. The solution
was cooled to Tint ) 0-5 °C, whereupon the 4,5,6-triaminopyri-
midine crystallized out of solution as a free base. The crystals were
collected by filtration, washed with ice-cold water, and dried to
give the title compound as white needles (18.8 g, 67%). mp: 256-
258 °C (dec) (lit: 246 °C, dec).24 1H NMR (DMSO-d6) δ 7.47 (s,
1H), 5.57 (br. s. 4H), 3.72 (br. s, 2H).
N-(5,6-diamino-pyrimidin-4-yl)-2-(3-methoxy-phenyl)-aceta-
mide hydrochloride (10). 4,5,6-Triaminopyrimidine (9) (6.25 g,
50 mmol) was dissolved in N-methyl-2-pyrrolidone (NMP, 70 mL)
at 70 °C. The solution was cooled to room temperature, and treated
with 3-methoxyphenylacetyl chloride (9.2 g, 50 mmol, 1.0 equiv)
for 3 h at 50 °C, whereupon the desired compound precipitated as
its HCl salt. The precipitate was collected, washed with EtOAc
and acetone, and dried to give the title compound as a white solid
(15.2 g, 98%). HPLC Purity: 97.4%. tR ) 4.13 min (Conditions
I). mp: 286-288 °C. 1H NMR (DMSO-d6) δ 9.22 (s, 1H), 8.20 (s,
1H), 6.75-7.58 (br. s, 4H), 7.21 (t, J ) 7.9 Hz, 1H), 6.95 (d, J )
1.3 Hz, 1H), 6.90 (d, J ) 7.6 Hz, 1H), 6.80 (dd, J ) 7.6 & 1.3 Hz,
1H), 3.74 (s, 2H), 3.73 (s, 3H). 13C NMR (DMSO-d6) δ 170.9,
159.4, 156.4 (2C), 147.9, 137.7, 129.3, 122.8, 116.2, 112.2, 93.9,
55.4, 41.8
δ 157.1, 153.6, 150.9, 149.8, 148.4, 133.1, 128.8, 123.4, 117.3,
114.7, 112.5, 54.0, 41.5, 30.4, 30.2,18.6, 12.2. HRMS: calcd for
C17H21N5ClO (MH)+ m/z 346.1429, found 346.1426
8-(2-Bromo-5-methoxy-benzyl)-9-butyl-9H-purin-6-ylamine
(12c). A solution of 9-butyl-8-(3-methoxy-benzyl)-9H-purin-6-
ylamine (12a) (35 mg, 0.11 mmol) in AcOH (1 mL) was treated
with Br2 (1 M in AcOH, 0.2 mL, 2 equiv) at room temperature for
16 h. Evaporation and preparative TLC (MeOH:CH2Cl2 10:90) gave
the title compound (24.8 mg, 57%), HPLC Purity: 99.5%. tR
)
1
7.62 min (Conditions I). mp: 158-159 °C. H NMR (CDCl3) δ
8.30 (s, 1H), 7.46 (d, J ) 8.6 Hz, 1H), 6.70 (d, J ) 2.9 Hz, 1H),
6.66 (dd, J ) 8.8 & 2.9 Hz, 1H), 6.34 (s, 2H), 4.31 (s, 2H), 4.04
(t, J ) 7.7 Hz, 2H), 3.73 (s, 3H), 1.62 (quint., J ) 7.8 Hz, 2H),
1.31 (sext., J ) 7.6 Hz, 2H), 0.86 (t, J ) 7.3 Hz, 3H). 13C NMR
(CDCl3) δ 159.2, 155.0, 152.4, 151.2, 149.9, 136.4, 133.5, 118.7,
116.3, 114.5, 114.3, 55.4, 42.9, 34.3, 31.9, 19.9, 13.7.
9-Butyl-8-(2-iodo-5-methoxy-benzyl)-9H-purin-6-ylamine (12d).
A solution of 9-butyl-8-(3-methoxy-benzyl)-9H-purin-6-ylamine
(12a) (1.24 g, 4.0 mmol) in AcOH (50 mL) was treated with NIS
(2.79 g, 12 mmol, 3.0 equiv) at room temperature for 16 h. The
reaction mixture was diluted with CH2Cl2 (100 mL), washed with
aq Na2CO3, aq Na2S2O3, and brine. Drying (Na2SO4), evaporation,
and flash chromatography (MeOH:CH2Cl2 5:95 f 10:90) gave the
title compound as a white solid (0.53 g, 31%). HPLC Purity: 91.0%.
tR ) 7.35 min (Conditions II). mp: 172-174 °C. 1H NMR (CDCl3)
δ 8.36 (s, 1H), 7.77 (d, J ) 8.7 Hz, 1H), 6.67 (d, J ) 3.0 Hz, 1H),
6.61 (dd, J ) 8.7 & 3.0 Hz), 5.67 (s, 2H), 4.33 (s, 2H), 4.06 (t, J
) 7.7 Hz, 2H), 3.71 (s, 3H), 1.66 (quint., J ) 7.4 Hz, 2H), 1.37
(sext., J ) 7.6 Hz, 2H), 0.91 (t, J ) 7.4 Hz, 3H). 13C NMR (CDCl3)
δ 160.3, 154.7, 152.5, 151.4, 150.3, 140.1, 139.8, 118.8, 116.1,
114.8, 88.8, 55.4, 43.1, 39.5, 32.0, 20.0, 13.7. HRMS: calcd for
C17H21N5ClO (MH)+ m/z 438.0785, found 438.0790.
8-Bromo-9-butyl-9H-purin-6-ylamine (17). A mixture of 8-bro-
moadenine (16) (1.87 g, 8.69 mmol),17 1-iodobutane (1.19 mL, 10.4
mmol, 1.2 equiv), Cs2CO3 (3.4 g, 10.4 mmol, 1.2 equiv), and DMF
(200 mL) was stirred at room temperature for 16 h. Workup
(EtOAc/H2O, brine), drying (Na2SO4) and flash chromatography
(MeOH:CHCl3 2:98 f 5:95) gave the title compound as a white
solid (1.46 g, 62%). mp: 152-154 °C. 1H NMR (CD3OD) δ 8.18
(s, 1H), 4.24 (t, J ) 7.4 Hz, 2H), 1.82 (quint., J ) 7.4 Hz, 2H),
1.38 (sext., J ) 7.5 Hz, 2H), 0.98 (t, J ) 7.4 Hz, 3H).
9-Butyl-8-(2,5-dimethoxy-phenylsulfanyl)-9H-purin-6-
ylamine (18). A solution of 2,5-dimethoxy-benzenethiol (251 mg,
1.48 mmol, 8.0 equiv) in DMF (3 mL) was treated with NaH (60%
in oil, 29.5 mg, 0.74 mmol, 4.0 equiv) at room temperature for 10
min. When gas evolution ceased, 8-bromo-9-butyl-9H-purin-6-
ylamine (17) (50 mg, 0.18 mmol, 1.0 equiv) was added, and the
mixture was heated to 150 °C for 4 h. Workup (EtOAc/H2O, brine),
drying (Na2SO4) and preparative TLC (MeOH/CH2Cl2 5:95) gave
the title compound (46 mg, 69%). HPLC Purity: 90.0%. tR ) 8.63
min (Conditions III). 1H NMR (DMSO-d6) δ 8.15 (s, 1H), 7.41 (s,
2H), 7.04 (d, J ) 8.9 Hz, 1H), 6.85 (d, J ) 8.9 Hz, 1H), 6.46 (s,
1H), 4.15 (t, J ) 7.3 Hz, 2H), 3.76 (s, 3H), 3.60 (s, 3H), 1.64
(quint., J ) 7.4 Hz, 2H), 1.16 (sext., J ) 7.5 Hz, 2H), 0.79 (t, J )
7.4 Hz, 3H). HRMS: calcd for C17H22N5O2S (MH)+ m/z 360.1489,
found 360.1494.
2-Iodo-5-methoxy-benzenethiophenol, Potassium Salt (21).
The title compound was prepared by a modification of the published
procedure.18b A solution of EtOCS2K (100 g, 625 mmol, 1.1 equiv)
in acetone (1.3 L) was cooled to an internal temperature Tint of
-20 °C (Tbath ) -40 °C), and 2-iodo-5-methoxybenzenediazonium
tetrafluoroborate (20) (191 g, 549 mmol, 1.0 equiv)18 was added
over 10 min. The reaction vessel was removed from the cold bath,
and when Tint reached +15 °C, N2 evolved. The reaction was stirred
for 30 min at room temperature, and the gelatinous, insoluble KBF4
was filtered off. The filtrate was concentrated, worked-up (EtOAc/
water/brine), and concentrated to give crude dithiocarbonic acid
O-ethyl ester S-(2-iodo-5-methoxy-phenyl) ester, as a brown oil
(200 g) which was used without further purification. 1H NMR
(DMSO-d6) δ 7.82 (d, J ) 8.8 Hz, 1H), 7.22 (d, J ) 3.0 Hz, 1H),
8-(3-Methoxy-benzyl)-9H-purin-6-ylamine (11). A solution of
crude N-(5,6-diamino-pyrimidin-4-yl)-2-(3-methoxy-phenyl)-aceta-
mide hydrochloride (10) (17.6 g, 57 mmol) and MeONa (12.3 g,
227 mmol, 4.0 equiv) in n-BuOH (150 mL) was heated to reflux
for 2 h, cooled to room temperature, and neutralized with 2 M HCl.
Brine was added, which gave a biphasic mixture. Concentration of
the organic layer afforded the title compound as a solid (10.7 g,
1
75%). tR ) 4.70 min (Conditions II). mp: 252-254 °C H NMR
(DMSO-d6) δ 8.06 (s, 1H), 7.43 (br. s, 1H), 7.21 (t, J ) 7.9 Hz,
1H), 7.05 (s, 2H), 6.93 (s, 1H), 6.87 (d, J ) 7.6 Hz, 1H), 6.79 (dd,
J ) 8.1 & 2.3 Hz, 1H), 4.09 (s, 2H), 3.73 (s, 3H).13C NMR (DMSO-
d6) δ 159.8, 155.5, 152.3, 151.6, 150.8, 139.3, 130.0, 121.4, 119.0.
115.0, 112.4, 55.5, 35.5.
9-Butyl-8-(3-methoxy-benzyl)- 9H-purin-6-ylamine (12a). A
mixture of 8-(3-methoxy-benzyl)-9H-purin-6-ylamine (11) (0.50 g,
2.2 mmol), BuI (0.30 mL, 2.65 mmol, 1.2 equiv), Cs2CO3 (1.43 g,
4.4 mmol, 2.0 equiv), and DMF (2.5 mL) was stirred at room
temperature for 16 h. Flash chromatography (MeOH:CH2Cl2 5:95)
gave the title compound as a white solid (370 mg, 54%). HPLC
Purity: 91.0%. tR ) 6.92 min (Conditions II). mp: 163-165 °C.
1H NMR (CDCl3:CD3OD 5:1) δ 8.13 (s, 1H), 7.16 (t, J ) 7.9 Hz,
1H), 6.73-6.67 (m, 3H), 4.13 (s, 2H), 3.95 (t, J ) 7.7 Hz, 2H),
3.68 (s, 3H), 1.48 (quint., J ) 7.7 Hz, 2H), 1.20 (sext., J ) 7.5
Hz, 2H), 0.78 (t, J ) 7.4 Hz, 3H). 13C NMR (CDCl3:CD3OD 5:1)
δ 159.9, 154.7, 152.0, 150.9, 150.7, 136.6, 129.9, 120.8, 117.8,
114.5, 112.3, 55.07, 42.9, 34.2, 31.5, 19.8, 13.4. HRMS: calcd for
C17H22N5O (MH)+ m/z 312.1819, found 312.1817.
9-Butyl-8-(2-chloro-5-methoxy-benzyl)-9H-purin-6-ylamine
(12b). A solution of 9-butyl-8-(3-methoxy-benzyl)-9H-purin-6-
ylamine (12a) (100 mg, 0.32 mmol) in THF (4 mL) was treated
with SO2Cl2 (78 µL, 0.96 mmol, 3.0 equiv) at room temperature
for 2 h. Workup and preparative TLC (MeOH:CH2Cl2 10:90) gave
the title compound (60.2 mg, 54%). mp: 138-139 °C. HPLC
Purity: 92.4%. tR ) 7.77 min (Conditions II). 1H NMR (CDCl3) δ
8.31 (s, 1H), 7.30 (d, J ) 8.8 Hz, 1H), 6.75 (dd, J ) 8.8 & 3.0 Hz,
1H), 6.67 (d, J ) 3.0 Hz, 1H), 6.26 (s, 2H), 4.32 (s, 2H), 4.04 (t,
J ) 7.7 Hz, 2H), 3.67 (s, 3H), 1.62 (quint., J ) 7.7 Hz, 2H), 1.30
(sext., J ) 7.5 Hz, 2H), 0.87 (t, J ) 7.4 Hz, 3H). 13C NMR (CDCl3)

Hsp90 Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 2 825
6.71 (dd, J ) 8.8 & 3.0 Hz, 1H), 4.62 (q, J ) 7.0 Hz, 2H), 3.80 (s,
3H), 1.34 (t, J ) 7.0 Hz, 3H).
Cs2CO3 (1.94 g, 5.95 mmol, 2.2 equiv), 5-chloro-pent-1-yne (0.56
mL, 5.26 mmol, 2.0 equiv), and DMF (5 mL) was heated to 85 °C
for 16 h. Workup and evaporation gave the title compound as a
crude solid which was used without further purification (510 mg,
66%). H NMR (DMSO-d6) δ 6.80 (s, 2H), 5.95 (s, 2H), 3.98 (t,
J ) 7.3 Hz, 2H), 2.81 (t, J ) 2.6 Hz, 1H), 2.22 (td, J ) 7.1 & 2.7
Hz, 2H), 1.96 (quint., J ) 7.2 Hz, 2H).
8-(2-Iodo-5-methoxy-phenylsulfanyl)-9-pent-4-ynyl-9H-purine-
2,6-diamine (27). A mixture of 8-bromo-9-pent-4-ynyl-9H-purine-
2,6-diamine (26) (500 mg, 1.59 mmol), 2-iodo-5-methoxy-
benzenethiol (1.34 g, 5.07 mmol, 3 equiv), t-BuOK (475 mg, 4.23
mmol, 2.5 equiv), and DMF (7 mL) was heated to 100 °C for 16
h. Workup and chromatography (MeOH:DCM 1:100 f 1:10) gave
the title compound as a solid (350 mg, 43%). tR ) 7.85 min
(Conditions II). 1H NMR (DMSO-d6) δ 7.73 (d, J ) 8.6 Hz, 1H),
7.05-6.95 (br. s., 2H), 6.64 (dd, J ) 8.6 & 2.9 Hz, 1H), 6.24 (d,
J ) 2.9 Hz, 1H), 6.02 (s, 2H), 4.01 (t, J ) 7.2 Hz, 2H), 3.59 (br.
s, 3H), 2.70 (t., J ) 2.6 Hz, 1H), 2.14 (td, J ) 7.2 & 2.6 Hz, 2H),
1.80 (quint., J ) 7.3 Hz, 2H).
The crude dithiocarbonic acid O-ethyl ester S-(2-iodo-5-methoxy-
phenyl) ester (40 g) was diluted in THF (300 mL). Separately, KOH
(12.6 g, 2 equiv) was dissolved in MeOH (300 mL). The two
solutions were combined, heated to 40 °C for 2 h, and thoroughly
evaporated. The solid was scraped from the sides of the flask, stirred
in Et2O (500 mL) for 1 h, and filtered. The solid was finely ground,
and the ether wash was repeated. Filtration and drying under
vacuum at room temperature gave potassium 2-iodo-5-methoxy-
benzenethiolate (21) as a crude brown powder (38.5 g). The crude
material still contained the excess of potassium salts, and 1H NMR
analysis indicated the presence of two impurities, tentatively
1
1
assigned as KSCO2Me (61 mol %) and KSCO2Et (6 mol %). H
NMR (DMSO-d6) δ 7.30 (d, J ) 8.8 Hz, 1H), 6.97 (d, J ) 3.0 Hz,
1H), 5.83 (dd, J ) 8.8 & 3.0 Hz, 1H), 3.58 (s, 3H).
8-Bromo-9-(4-methyl-pent-3-enyl)-9H-purin-6-ylamine (22).
A mixture of 8-bromoadenine (16) (3.0 g, 14 mmol),17 5-bromo-
2-methyl-2-pentene (2.25 mL, 16.8 mmol, 1.2 equiv), and Cs2CO3
(9.1 g, 28 mmol, 2.0 equiv) in DMF (15 mL) was stirred at room
temperature for 3 h, filtered, and evaporated. Flash chromatography
(MeOH:CH2Cl2 5:95 f 10:90) gave the title product as a solid
(1.56 g, 38%). mp: 159-162 °C. 1H NMR (CDCl3:CD3OD 5:1) δ
8.15 (s, 1H), 5.03 (t, J ) 7.5 Hz, 1H), 4.10 (t, J ) 7.2, 2H), 2.41
(q, J ) 7.2 Hz, 2H), 1.53 (s, 3H), 1.32 (s, 3H). 13C NMR (CDCl3:
CD3OD 5:1) δ 154.1, 152.6, 150.7, 135.9, 127.5, 119.4, 118.5, 44.3,
28.0, 25.4, 17.2.
2-Fluoro-8-(2-iodo-5-methoxy-phenylsulfanyl)-9-pent-4-ynyl-
9H-purin-6-ylamine (28). A mixture of 8-(2-iodo-5-methoxy-
phenylsulfanyl)-9-pent-4-ynyl-9H-purine-2,6-diamine (27) (79 mg,
0.165 mmol) and 48% aq HBF4 (0.5 mL) in THF (0.5 mL) was
treated at -20 °C with iso-amyl nitrite (22 µL, 0.165 mmol). The
reaction mixture was allowed to reach rt and was further heated to
40 °C for 10 min. Workup (DCM/aq K2CO3) and preparative plate
TLC purification (EtOAc/Hexane 1:4) gave the title compound as
1
8-(2-Iodo-5-methoxy-phenylsulfanyl)-9-(4-methyl-pent-3-enyl)-
9H-purin-6-ylamine (23). A mixture of 8-bromo-9-(4-methyl-pent-
3-enyl)-9H-purin-6-ylamine (22) (2.7 g, 9.1 mmol) and crude
potassium 2-iodo-5-methoxy-benzenethiophenolate (21) (8.3 g, 27.3
mmol, 3.0 equiv) in DMF (50 mL) was stirred at 140 °C for 3 h
and then at 100 °C for 16 h. Filtration, evaporation, and flash
chromatography (MeOH:CH2Cl2 5:95) gave the title product as a
solid (0.64 g, 15%). mp: 168-169 °C. tR ) 8.90 min (Conditions
a solid (10 mg, 13%). tR ) 9.43 min (Conditions II). H NMR
(CDCl3) δ 7.72 (d, J ) 8.7 Hz, 1H), 6.70 (d, J ) 2.9 Hz, 1H),
6.59 (dd, J ) 8.7 & 2.9 Hz 1H), 6.11-5.87 (br. s, 2H), 4.25 (t, J
) 7.3 Hz, 2H), 3.69 (s, 3H), 2.25 (td, J ) 7.1 & 2.7 Hz, 2H), 2.02
(quint., J ) 7.2 Hz, 2H), 1.97 (t, J ) 2.6 Hz, 1H).
Acetic Acid 2-(6-Amino-8-bromo-purin-9-yl)-ethyl Ester (30a).
A mixture of adenine (60.0 g, 444 mmol), Cs2CO3 (223 g, 686
mmol, 1.54 equiv), AcO-CH2-CH2-Br (75.8 mL, 686 mmol, 1.54
equiv), and DMF (187 g) was stirred at 45 °C (Tint) for 5 h, at
which point the reaction was complete as judged by analytical
HPLC of both the supernatant and the solid residues. The DMF
was evaporated, and the residue was added to a mixture of AcOH
(50 mL, 888 mmol, 2 equiv), water (100 mL), and ice (100 g).
The solid was filtered, washed with 100 mL ice-cold water, and
dried on a rotary evaporator under high vacuum to give acetic acid
2-(6-amino-purin-9-yl)-ethyl ester as a white powder (62.4 g, 67%).
The HPLC trace was clean, but the product was still contaminated
with traces of cesium bicarbonate (and/or carbonate), as evidenced
by an effervescence upon addition to an AcOH/AcONa buffer.
1
II). H NMR (CDCl3) δ 8.39 (s, 1H), 7.72 (d, J ) 8.7 Hz, 1H),
6.72 (d, J ) 2.7 Hz, 1H), 6.58 (dd, J ) 8.7 & 2.7 Hz, 1H), 5.85
(br. s, 2H), 5.15 (t, J ) 7.3 Hz, 1H), 4.25 (t, J ) 7.4 Hz, 2H), 3.69
(s, 3H), 2.50 (quint., J ) 7.3 Hz, 2H), 1.66 (s, 3H), 1.44 (s, 3H).
13C NMR (CDCl3) δ 160.4, 154.9, 153.4, 151.4, 145.0, 140.4, 138.7,
136.0, 120.4, 119.1, 116.2, 115.0, 87.2, 55.4, 43.8, 28.6, 25.7, 17.6.
Anal. (C18H20IN5OS): C, H, N.
8-Bromo-9-pent-4-ynyl-9H-purin-6-ylamine (24). A mixture
of 8-bromoadenine (16) (10.7 g, 50 mmol),17 5-chloro-1-pentyne
(6.16 g, 60 mmol, 1.2 equiv) and Cs2CO3 (32.5 g, 100 mol, 2.0
equiv) in DMF (70 mL) was stirred at 70 °C for 3 h, filtered, and
evaporated. Flash chromatography (MeOH:CH2Cl2 5:95 ) 10:90)
gave the title product as a solid (3.5 g, 25%). tR ) 5.03 min
mp: 175-177 °C.
1H NMR (DMSO-d6) δ 8.13 (s, 1H), 8.12 (s,
1H), 7.20 (s, 2H), 4.37 (br. s, 4H), 1.92 (s, 3H).13C NMR (DMSO-
d6) δ 170.5, 156.4, 152.9, 150.1, 141.5, 119.1, 62.5, 42.6, 21.0.
An acetic acid buffer was prepared with AcONa (131 g, 1.6 mol),
AcOH (96 mL, 1.6 mol), and H2O (800 mL). Acetic acid 2-(6-
amino-purin-9-yl)-ethyl ester (16.6 g, 75 mmol) was dissolved in
a mixture of the AcOH buffer (100 mL), MeOH (30 mL), and THF
(30 mL) using magnetic stirring at room temperature. Bromine (7.0
mL, 136 mmol, 1.8 equiv) was added over 1 min, and the stirring
was stopped. The desired product, acetic acid 2-(6-amino-8-bromo-
purin-9-yl)-ethyl ester (30a), partially crystallized out of solution.
After 1 h, the crystals were collected by filtration, washed (H2O),
and air-dried to give the desired bromide (11.6 g, 51%) as purple
prisms. The aqueous filtrate was combined with CHCl3 (500 mL).
The mixture was stirred, and the pH of the aqueous layer was
monitored with a pH meter. The excess bromine was quenched
with concentrated aq NH4OH (25 mL; final pH ) 7) and then
hydrazine monohydrate (3 mL, final pH ) 9). Note that the excess
Br2 can also be neutralized with Na2S2O3, which is experimentally
easier. The brown color of Br2 disappeared, and the yellow organic
layer was separated. The aqueous layer was washed with 300 mL
CHCl3, and the combined organic layers were concentrated to give
additional acetic acid 2-(6-amino-8-bromo-purin-9-yl)-ethyl ester
(30a) as a yellow powder (6.7 g, 30%). The combined yield was
81%. mp: 150-152 °C. 1H NMR (DMSO-d6) δ 8.13 (s, 1H), 7.38
1
(Conditions II). mp: 195-196 °C. H NMR (DMSO-d6) δ 8.14
(s, 1H), 7.39 (s, 2H), 4.20 (t, J ) 7.3 Hz, 2H), 2.82 (t, J ) 2.6 Hz,
1H), 2.25 (td, J ) 2.6 & 7.0 Hz, 2H), 1.94 (quint., J ) 7.3 Hz,
2H). 13C NMR (CDCl3/CD3OD 5:1) δ 154.2, 152.8, 150.9, 127.3,
119.5, 82.1, 69.5, 43.6, 28.0, 15.8.
8-(2-Iodo-5-methoxy-phenylsulfanyl)-9-pent-4-ynyl-9H-purin-
6-ylamine (4). A mixture of 8-bromo-9-pent-4-ynyl-9H-purin-6-
ylamine (24) (3.5 g, 12.5 mmol), t-BuOK (2.8 g, 25.0 mmol, 2.0
equiv) and 2-iodo-5-methoxybenzenethiol (7.5 g, 27.5 mmol, 2.2
equiv) in DMF (30 mL) was stirred at 70 °C for 16 h, filtered, and
evaporated. Flash chromatography (MeOH:CH2Cl2 5:95 f 10:90)
gave the title product as a solid which was recrystallized from
MeOH to give 2 crops (2.34 and 1.80 g, combined yield 71%). tR
) 7.82 min (Conditions II). mp: 171-172 °C. 1H NMR (DMSO-
d6) δ 8.18 (s, 1H), 7.79 (d, J ) 8.7 Hz, 1H), 7.50 (br. s, 2H), 6.70
(dd, J ) 8.7 & 2.8 Hz, 1H), 6.49 (d, J ) 2.8 Hz, 1H), 4.22 (t, J )
7.1 Hz, 2H), 3.62 (s, 3H), 2.77 (d, J ) 2.5 Hz, 1H), 2.20 (td, J )
6.9 & 4.5 Hz, 2H), 1.87 (quint., J ) 7.1 Hz, 2H). 13C NMR (CDCl3:
CD3OD 5:1) δ 160.5, 154.6, 152.7, 151.1, 146.1, 140.8, 136.4,
119.7, 118.5, 116.1, 90.1, 82.2, 69.6, 55.5, 43.1, 28.3, 15.9. Anal.
(C17H16IN5OS): C, H, N.
8-Bromo-9-pent-4-ynyl-9H-purine-2,6-diamine (26). A mixture
of 8-bromo-9H-purine-2,6-diamine (25) (600 mg, 2.63 mmol),25

826 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 2
Biamonte et al.
(s, 2H), 4.35 (br. s, 4H), 1.89 (s, 3H). 13C NMR (DMSO-d6) δ
170.5, 152.5, 150.9, 150.0, 128.6, 119.2, 61.7, 44.0, 21.0.
in hot anhydrous 1,4-dioxane (40 mL). The solution was cooled to
25-40 °C and treated with Et3N (3.0 equiv) and MsCl (1.5 equiv)
for 30 min. The mixture was concentrated, dissolved in the
appropriate amine (20 mL), and heated to 55 °C for 16 h in a
pressure vessel, whereupon the desired material precipitated and
was collected by filtration. The crude amine was purified either by
recrystallization from refluxing MeOH or by flash chromatography
(0-5% MeOH in Et3N:hexane:DCM:EtOAc 2:100:200:100) to give
the desired amine in 45-75% yield. (Note that if the chloropropane
36 was used instead of the corresponding mesylate, the aminations
were performed at 90-110 °C for 16 h to give the desired amines
in 67-73% yield after flash chromatography.)
Acetic Acid 3-(6-amino-8-bromo-purin-9-yl)-propyl Ester
(30b). The title compound was prepared by alkylation of adenine
in two steps (i. AcO-(CH2) 3-Cl. ii. Br2) as reported previously.19
2-[6-Amino-8-(2-iodo-5-methoxy-phenylsulfanyl)-purin-9-yl]-
ethanol (31a). A solution of acetic acid 2-(6-amino-8-bromo-purin-
9-yl)-ethyl ester (30a) (16.5 g, 55 mmol) and potassium 2-iodo-
5-methoxy-benzenethiolate (21) (33 g, 2 equiv) in DMF (600 mL)
was heated to 50 °C for 16 h. The reaction mixture was
concentrated, dissolved in MeOH, and treated with a catalytic
amount K2CO3 for 3 h at 50 °C to cleave the acetyl group in situ.
The mixture was concentrated again and stirred in a mixture of
water and Et2O for 16 h. The desired alcohol, which was soluble
neither in Et2O nor in water, was recovered by filtration. Washing
with ether and drying gave 2-[6-amino-8-(2-iodo-5-methoxy-
phenylsulfanyl)-purin-9-yl]-ethanol (31a) as an orange-brown pow-
der (9 g, 37%). Note that the coupling step was very clean if
performed with pure potassium thiophenolate but gave variable
results with crude potassium thiophenolate. The outcome depended
on the exact batch of thiophenolate used. In some cases, addition
of a small amount of concentrated HCl (0.1-0.2 equiv) was
sometimes necessary to adjust the pH. If it was too alkaline,
cleavage of the acetyl group occurred and other byproducts were
observed. If it was too acidic, the thiophenolate dimerized quickly
The purified amine (2 mmol) was dissolved in refluxing EtOH
(30 mL) and with very vigorous agitation a H3PO4 solution (0.84
M in EtOH, 2.3 mL, 1 equiv) was added in one portion. The salt
precipitated or crystallized immediately. After cooling, filtration
gave the desired phosphate in typically 80-85% yield as a fine
powder or white needles, which were dried on high vacuum.
9-[3-(1-Ethyl-propylamino)-propyl]-8-(2-iodo-5-methoxy-phe-
nylsulfanyl)-9H-purin-6-ylamine (37). Free Base: tR ) 6.28 min
1
(Conditions I). mp: 154-156 °C. H NMR (CDCl3/CD3OD 3:1)
δ 8.02 (s, 1H), 7.56 (d, J ) 8.8 Hz, 1H), 6.71 (d, J ) 2.9 Hz, 1H),
6.46 (dd, J ) 8.8 & 2.9 Hz, 1H), 4.14 (t, J ) 6.8 Hz, 2H), 3.52 (s,
3H), 2.56 (t, J ) 7.0 Hz, 2H), 2.40 (quint., J ) 5.9 Hz, 1H), 1.96
(quint., J ) 7.0 Hz, 2H), 1.96 (quint., J ) 7.1 Hz, 4H), 0.75 (t, J
) 7.5 Hz, 6H). 13C NMR (CDCl3:CD3OD 3:1) 160.4, 154.7, 152.5,
150.9, 145.7, 140.6, 136.0, 119.4, 118.4, 115.9, 90.0, 60.1, 55.3,
42.6, 41.2, 28.1, 23.6 (2C), 9.2 (2C). H3PO4 Salt: tR ) 6.36 min
(Conditions II). mp: 205-208 °C. 1H NMR (D2O) δ 8.02 (s, 1H),
7.66 (d, J ) 8.6 Hz, 1H), 6.81 (d, J ) 2.8 Hz, 1H), 6.62 (dd, J )
8.8 & 2.8 Hz, 1H), 4.18 (t, J ) 6.7 Hz, 2H), 3.59 (s, 3H), 2.85 (t,
J ) 8.0 Hz, 2H), 2.80 (quint., J ) 6.0 Hz, 1H), 2.01 (quint., J )
7.3 Hz, 2H), 1.47 (quint., J ) 7.1 Hz, 4H), 0.75 (t, J ) 7.5 Hz,
6H). 13C NMR (D2O) δ 160.0, 154.3, 152.4, 150.1, 146.5, 141.0,
135.1, 119.0, 118.7, 116.2, 90.6, 60.4, 55.6, 41.7, 41.1, 25.6 (2C),
21.5, 8.1 (2C). Anal. (C20H27IN6OS‚H3PO4) C, H, N.
1
to the disulfide ArSSAr. mp: 184-187 °C. H NMR (DMSO-d6)
δ 8.18 (s, 1H), 7.76 (d, J ) 8.7 Hz, 1H), 7.42 (s, 2H), 6.68 (dd, J
) 8.7 & 2.9 Hz, 1H), 6.56 (d, J ) 2.9 Hz, 1H), 4.99 (t, J ) 5.6
Hz, 1H), 4.26 (t, J ) 5.7 Hz, 2H), 3.70 (q, J ) 5.7 Hz, 2H), 3.67
(s, 3H). 13C NMR (DMSO-d6) δ 160.3, 155.9, 153.6, 151.3, 144.0,
140.6, 140.0, 120.3, 116.3, 115.1, 87.5, 59.4, 55.8, 46.5.
3-[6-Amino-8-(2-iodo-5-methoxy-phenylsulfanyl)-purin-9-yl]-
propan-1-ol (31b). A solution of crude acetic acid 3-[6-amino-8-
(2-iodo-5-methoxy-phenylsulfanyl)-purin-9-yl]-propyl ester19 (36,
20 g) in MeOH (250 mL) was treated with K2CO3 (500 mg) at 50
°C for 1 h. Filtration and concentration afforded the title compound
as a white solid (14 g, 64.5%) identical by 1H and 13C NMR to the
known compound.19
9-(3-sec-Butylamino-propyl)-8-(2-iodo-5-methoxy-phenylsul-
fanyl)-9H-purin-6-ylamine (38). Free Base: tR ) 5.94 min
1
8-(2-Iodo-5-methoxy-phenylsulfanyl)-9H-purin-6-ylamine (33).
A suspension of finely ground 6-amino-7,9-dihydro-purine-8-thione
(32) (12 g, 71 mmol)19 in DMF (250 mL) was cooled to -60 °C
and treated portionwise with 2-iodo-5-methoxybenzenediazonium
tetrafluoroborate (39.2 g, 112 mmol, 1.6 equiv). The reaction
mixture was allowed to warm to room temperature and was stirred
at room temperature for another 2 h. The reaction mixture was
neutralized to pH ) 7 with aq NaOH and evaporated. Water was
added, and the pH was adjusted to 10 with aq NaOH. The aqueous
layer was washed with CH2Cl2 (3 × 100 mL) and neutralized with
HCl 6 N, whereupon the title compound precipitated. The solid
was collected and dried under vacuum to afford the crude title
compound as a yellow powder which was used without further
purification (9.6 g, HPLC Purity: 77%, yield corrected for low
purity: 26%). tR ) 6.05 min (Conditions II). mp: 254-156 °C.
1H NMR (DMSO-d6) δ 8.13 (s, 1H), 7.79 (d, J ) 8.4 Hz, 1H),
7.38 (br. s, 2H), 6.71 (d, J ) 8.4 Hz, 1H), 6.62 (s, 2H), 3.65 (s,
3H).
(Conditions I). mp: 151-152 °C. H NMR (CD3OD) δ 8.20 (s,
1H), 7.83 (d, J ) 8.6 Hz, 1H), 6.96 (d, J ) 2.7 Hz, 1H), 6.75 (dd,
J ) 2.7 & 8.6 Hz, 1H), 4.35 (m, 2H), 3.73 (s, 3H), 2.70-2.64 (m,
3H), 2.07 (m, 2H), 1.58 (m, 1H), 1.34 (m, 1H), 1.06 (d, J ) 5.8
Hz, 3H), 0.92 (t, J ) 7.1 Hz, 3H). 13C NMR (CDCl3:CD3OD 3:1)
160.4, 154.6, 152.5, 150.9, 145.8, 140.6, 136.1, 119.4, 118.3, 115.9,
89.9, 55.3, 54.3, 42.9, 41.4, 29.1, 28.5, 18.3, 9.9. H3PO4 Salt: tR
1
) 6.10 min (Conditions I). mp: 208-212 °C. H NMR (D2O) δ
8.00 (s, 1H), 7.66 (d, J ) 8.6 Hz, 1H), 6.83 (d, J ) 2.8 Hz, 1H),
6.60 (dd, J ) 8.6 & 2.8 Hz, 1H), 4.16 (t, J ) 6.8 Hz, 2H), 3.58 (s,
3H), 2.88-2.81 (m, 3H), 1.99 (quint., J ) 6.7 Hz, 2H), 1.51-1.50
(m, 1H), 1.14 (sept., J ) 7.1 Hz, 1H), 1.05 (d, J ) 6.1 Hz, 3H),
0.77 (t, J ) 7.1 Hz, 3H). 13C NMR (D2O) δ 160.0, 154.1, 152.1,
150.1, 146.8, 141.1, 135.0, 119.0, 118.9, 116.3, 90.9, 55.7, 55.6,
41.6, 41.0, 25.6, 25.4, 14.8, 8.7. Anal. (C19H25IN6OS‚H3PO4) C,
H, N.
8-(2-Iodo-5-methoxy-phenylsulfanyl)-9-(3-isopropylamino-
propyl)-9H-purin-6-ylamine (39). Free Base: tR ) 5.61 min
1
9-(3-Chloro-propyl)-8-(2-iodo-5-methoxy-phenylsulfanyl)-9H-
purin-6-ylamine (34). A mixture 6-amino-7,9-dihydro-purine-8-
thione (33) (10 g, 25 mmol), 1-bromo-3-chloropropane (5.0 mL,
50 mmol, 2.0 equiv), Cs2CO3 (89 g, 27.5 mmol, 1.1 equiv), and
DMF (600 mL) was stirred at 50 °C for 2 h. Workup and
chromatography (CH2Cl2:EtOAc:Et3N 100:50:5) gave a 1.5:1
mixture of regioisomers, which was recrystallized twice from CH2-
Cl2:MeOH 100:5 (100 mL) to give 2.5 g of the pure title compound
(Conditions II). mp: 164-166 °C. H NMR (CD3OD) δ 8.24 (s,
1H), 7.86 (d, J ) 8.7 Hz, 1H), 6.87 (d, J ) 2.7 Hz, 1H), 6.64 (dd,
J ) 8.7 & 2.7 Hz, 1H), 4.42 (t, J ) 6.6 Hz, 2H), 3.77 (s, 3H),
3.37-3.33 (m, 1H), 3.08 (t, J ) 6.6 Hz, 2H), 2.24 (quint., J ) 6.6
Hz, 2H), 1.34 (d, J ) 6.5 Hz, 6H). 13C NMR (CDCl3:CD3OD 3:1)
160.4, 154.6, 152.5, 150.9, 145.8, 140.6, 136.1, 119.4, 118.3, 115.9,
89.8, 55.3, 48.6, 43.0, 41.3, 29.0, 21.5 (2C). H3PO4 Salt: tR
)
4.46 min (Conditions I). mp: 233-237 °C. 1H NMR (D2O) δ 8.12
(br. s, 1H), 7.83 (d, J ) 8.8 Hz, 1H), 7.00 (d, J ) 2.9 Hz, 1H),
6.75 (dd, J ) 8.8 & 2.9 Hz, 1H), 4.27 (t, J ) 6.9 Hz, 2H), 3.66 (s,
3H), 3.20 (sept., J ) 6.6 Hz, 1H), 2.94 (t, J ) 7.9 Hz, 2H), 2.10
(quint., J ) 7.4 Hz, 2H), 1.16 (d, J ) 6.6 Hz, 6H). 13C NMR (D2O)
δ 159.9, 154.1, 152.2, 150.1, 146.8, 141.0, 134.9, 119.2, 119.0,
116.4, 91.2, 55.6, 50.6, 41.7, 41.0, 25.7, 18.1 (2C). Anal. (C18H23-
IN6OS‚H3PO4) C, H, N.
1
(2.5 g, 21%). tR ) 7.93 min (Conditions II). H NMR (CDCl3) δ
8.36 (s, 1H), 7.73 (d, J ) 8.7 Hz, 1H), 6.72 (d, J ) 2.8 Hz, 1H),
6.59 (dd, J ) 8.7 & 2.8 Hz, 1H), 6.07 (br. s, 2H), 4.38 (t, J ) 7.1
Hz, 2H), 3.68 (s, 3H), 3.55 (t, J ) 7.1 Hz, 2H), 2.27 (quint., J )
6.7 Hz, 2H).
General Procedure for the Preparation of Amines 37-42.
The ethyl alcohol 31a or propyl alcohol 31b (4 mmol) was dissolved

Hsp90 Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 2 827
(3) Gorre, M. E.; Ellwood-Yen, K.; Chiosis, G.; Rosen, N.; Sawyers, C.
L. BCR-ABL point mutants isolated from patients with imatinib
mesylate-resistant chronic myeloid leukemia remain sensitive to
inhibitors of the BCR-ABL chaperone heat shock protein 90. Blood
2002, 100, 3041-3044.
(4) (a) Kamal, A.; Thao, L.; Sensintaffar, J.; Zhang. L.; Boehm, M. F.;
Fritz, L. C.; Burrows, F. J. A high-affinity conformation of Hsp90
confers tumour selectivity on Hsp90 inhibitors. Nature 2003, 425,
407-410. (b) Workman, P. Altered states: selectively drugging the
Hsp90 cancer chaperone. Trends Mol. Med. 2004, 10, 47-51.
(5) Yun, B.-G.; Huang, W.; Leach, N.; Hartson, S. D.; Matts, R. L.
Novobiocin induces a distinct conformation of Hsp90 and alters
Hsp90-cochaperone-client interactions. Biochemistry 2004, 43, 8217-
8229.
(6) The NCI clinical protocol consists of injecting a DMSO solution of
17-AAG.
(7) (a) Glaze, E. R.; Smith, A. C.; Johnson, D. W.; McCormick, D. L.;
Brown, A. B.: Levin, B. S.; Krishnaraj, R.; Lyubimov, A.; Egorin,
M. J.; Tomaszewski, J. E. Dose range-finding toxicity studies of 17-
DMAG. Proc. Am. Assoc. Cancer. Res. 2003, 44, 162-162. (b)
Eiseman, J. L.; Lan, J.; Lagatutta, T. F.; Hamburger, D. R.; Joseph,
9-(3-tert-Butylamino-propyl)-8-(2-iodo-5-methoxy-phenylsul-
fanyl)-9H-purin-6-ylamine (40). Free Base: tR ) 5.87 min
1
(Conditions II). mp: 165-166 °C. H NMR (CDCl3) δ 8.33 (s,
1H), 7.70 (d, J ) 8.7 Hz, 1H), 6.69 (d, J ) 2.7 Hz, 1H), 6.55 (dd,
J ) 8.7 & 2.7 Hz, 1H), 5.90 (br. s, 2H), 4.30 (t, J ) 6.6 Hz, 2H),
3.66 (s, 3H), 2.50 (t, J ) 6.6 Hz, 2H), 1.96 (quint., J ) 6.6 Hz,
2H), 1.05 (s, 9H). 13C NMR (CDCl3:CD3OD 3:1) 160.5, 154.8,
152.7, 151.1, 145.9, 140.8, 136.4, 119.6, 118.4, 116.1, 90.0, 55.4,
51.4, 41.7, 38.7, 29.9, 27.9 (3C). H3PO4 Salt: tR ) 4.77 min
(Conditions I). mp: 249-253 °C. 1H NMR (D2O) δ 8.03 (s, 1H),
7.65 (d, J ) 8.8 Hz, 1H), 6.86 (d, J ) 2.9 Hz, 1H), 6.61 (dd, J )
8.8 & 2.9 Hz, 1H), 4.17 (t, J ) 6.8 Hz, 2H), 3.60 (s, 3H), 2.82 (t,
J ) 8.0 Hz, 2H), 2.01 (quint., J ) 7.4 Hz, 2H), 1.15 (s, 9H). 13C
NMR (D2O) δ 159.9, 154.2, 152.3, 150.0, 146.7, 141.0, 134.9,
119.0, 118.9, 116.3, 91.0, 57.1, 55.5, 41.1, 38.4, 26.1, 24.7 (3C).
Anal. (C19H25IN6OS‚H3PO4‚0.08EtOH) C, H, N.
8-(2-Iodo-5-methoxy-phenylsulfanyl)-9-(2-isobutylamino-eth-
yl)-9H-purin-6-ylamine (41). Free Base: tR ) 6.10 min (Condi-
tions II). mp: 155-158 °C. 1H NMR (CDCl3/CD3OD 3:1) δ 8.09
(s, 1H), 7.64 (d, J ) 8.8 Hz, 1H), 6.77 (d, J ) 2.9 Hz, 1H), 6.53
(dd, J ) 8.7 & 2.9 Hz, 1H), 4.22 (t, J ) 6.5 Hz, 2H), 3.60 (s, 3H),
2.83 (t, J ) 6.5 Hz, 2H), 2.27 (d, J ) 6.9 Hz, 2H), 1.56 (non., J
) 6.7 Hz, 1H), 0.73 (d, J ) 6.6 Hz, 6H). 13C NMR (DMSO-d6)
160.4, 156.0, 153.8, 151.3, 143.5, 140.7, 140.0, 120.2, 115.8, 114.9,
E.;
Covey,
J.
and
M.;
Egorin,
M.
of
J.
Phar-
macokinetics
pharmacodynamics
17-demethoxy
17-[[(2-dimethylamino)ethyl]amino]geldanamycin (17DMAG, NSC
707545) in C. B-17 SCID mice bearing MDA-MB-231 human breast
cancer xenografts. Cancer Chemother. Pharmacol. 2005, 55, 21-
32.
(8) (a) Ikuina, Y.; Amishiro, N.; Miyata, M.; Narumi, H.; Ogawa, H.;
Akiyama, T.; Shiotsu, Y.; Akinaga, S.; Murakata, C. Synthesis and
antitumor activity of novel O-carbamoylmethyloxime derivatives of
radicicol. J. Med. Chem. 2003, 46, 2534-2541. Furthermore,
radicicol and its oximes contain an oxirane ring which has been
viewed as a liability for stability and toxicity, prompting the synthesis
of cycloproparadicicol: (b) Yang, Z. Q.; Geng, X.; Solit, D.; Pratilas,
C. A.; Rosen, N.; Danishefsky, S. J. New efficient synthesis of
resorcinylic macrolides via ynolides: establishment of cyclopro-
paradicicol as synthetic feasible preclinical anticancer agent based
on Hsp90 as the target. J. Am. Chem. Soc. 2004, 126, 7881. (b) Yang,
Z.-Q.; Danishefsky, S. J. A concise route to benzofused macrolactones
via ynolides: cycloproparadicicol. J. Am. Chem. Soc. 2003, 125,
9602-9603.
(9) (a) Chiosis, G.; Lucas, B.; Shtil, A.; Huezo, H.; Rosen, N. Develop-
ment of a purine-scaffold novel class of Hsp90 binders that inhibit
the proliferation of cancer cells and induce the degradation of HER-2
tyrosine kinase. Bioorg. Med. Chem. Lett. 2002, 10, 3555-3564. (b)
Vilenchik, M.; Solit, D.; Basso, A.; Huezo, H.; Lucas, B.; He, H.;
Rosen, N.; Spampinato, C.; Modrich, P.; Chiosis, G. Targeting wide-
range oncogenic transformation via PU24FCl, a specific inhibitor of
tumor Hsp90. Chem. Biol. 2004, 11, 787-797. (c) Chiosis, G.; Rosen,
N. Small molecule composition for binding to Hsp90. WO 0236075,
2002.
(10) (a) Drysdale, M. J.; Dymock, B. W.; Barril-Alonso, X.; Workman,
P. 3,4-Diarylpyrazoles and their use in the therapy of cancer. WO
03/055860 A1, 2003. (b) Wright, L.; Barril, X.; Dymock, B.;
Sheridan, L.; Surgenor, A.; Beswick, M.; Drysdale, M.; Collier, A.;
Massey, A.; Davies, N.; Fink, a.; Fromont, C.; Aherne, W.; Boxall,
K.; Sharp, S.; Workman, P.; Hubbard, R. Structure-activity relation-
ships in purine-based inhibitor binding to Hsp90 isoforms Chem. Biol.
2004, 11, 775-785. (c) Dymock, B.; Barril, X.; Beswick, M.; Collier,
A.; Davies, N.; Drysdale, M.; Fink, A.; Fromont, C.; Hubbard, R.
E.; Massey, A.; Surgenor, A.; Wright, L. Adenine derived inhibitors
of the molecular chaperone HSP90-SAR explained through multiple
X-ray structures. Bioorg. Med. Chem. Lett. 2004, 14, 325-328. (d)
Dymock, B. W.; Barril, X.; Brough, P. A.; Cansfield, J. E.; Massey,
A.; McDonald, E.; Hubbard, R. E.; Surgenor, A.; Roughley, S. D.;
Webb, P.; Workman, P.; Wright, L.; Drysdale, M. J. Novel, potent
small-molecule inhibitors of the molecular chaperone Hsp90 discov-
ered through structure-based design J. Med. Chem. 2005, 48, 4212-
4215. Structure of Hsp90 in complex with 3a: pdb code 1UY6, and
with 3b: pdb code 1UYF.
87.1, 57.4, 55.8, 48.8, 43.8, 28.4, 21.0 (2C). H3PO4 Salt: tR
)
5.10 min (Conditions I). mp: 214-216 °C. 1H NMR (D2O) δ 8.12
(s, 1H), 7.79 (d, J ) 8.3 Hz, 1H), 6.95 (br. s, 1H), 6.73 (d, J ) 8.3
Hz, 1H), 4.51 (m, 2H), 3.65 (s, 3H), 3.36 (m, 2H), 2.83 (d, J ) 6.6
Hz, 2H), 1.89 (sept., J ) 6.4 Hz, 1H), 0.87 (d, J ) 6.2 Hz, 6H).
13C NMR (D2O) δ 160.1, 153.9, 152.1, 150.6, 146.7, 141.2, 134.8,
119.3, 119.0, 116.6, 91.1, 55.6, 54.7, 46.1, 40.0, 25.5, 19.0 (2C).
Anal. (C18H23IN6OS‚H3PO4) C, H, N.
9-[2-(2,2-Dimethyl-propylamino)-ethyl]-8-(2-iodo-5-methoxy-
phenylsulfanyl)-9H-purin-6-ylamine (42). Free Base: tR ) 5.13
1
min (Conditions II). mp: 156-158 °C. H NMR (CDCl3/CD3OD
3:1) δ 8.21 (s, 1H), 7.74 (d, J ) 8.7 Hz, 1H), 6.87 (d, J ) 2.9 Hz,
1H), 6.63 (dd, J ) 8.7 & 2.9 Hz, 1H), 4.32 (t, J ) 6.4 Hz, 2H),
3.71 (s, 3H), 2.95 (t, J ) 6.4 Hz, 2H), 2.31 (s, 2H), 0.83 (s, 9H).
13C NMR (CDCl3:CD3OD 3:1) 160.5, 154.6, 152.7, 151.1, 146.5,
140.7, 136.9, 119.6, 118.3, 115.9, 89.7, 61.9, 55.4, 49.5, 43.9, 31.4,
27.5 (3C). H3PO4 Salt: tR ) 5.13 min (Conditions II). mp: 162-
164 °C. 1H NMR (D2O) δ 8.10 (s, 1H), 7.75 (d, J ) 8.7 Hz, 1H),
6.89 (d, J ) 2.8 Hz, 1H), 6.68 (dd, J ) 8.7 & 2.8 Hz, 1H), 4.51 (t,
J ) 5.9 Hz, 2H), 3.64 (s, 3H), 3.37 (t, J ) 5.9 Hz, 2H), 2.84 (s,
2H), 0.96 (s, 9H). 13C NMR (D2O) 160.1, 154.2, 152.2, 150.4,
147.3, 141.2, 135.0, 119.2, 119.0, 116.5, 91.0, 59.3, 55.6, 47.0,
40.2, 29.9, 26.3 (3C). Anal. (C19H25IN6OS‚H3PO4) C, H, N.
Acknowledgment. We thank T. Jiang for exploratory
synthetic work, and R. Mansfield for his expertise in formula-
tion.
References
(1) For recent reviews, see: (a) Kamal, A.; Boehm, M. F.; Burrows, F.
J. Therapeutic and diagnostic implications of Hsp90 activation. Trends
Mol. Med. 2004, 10, 283-290. (b) Dymock, B. W.; Drysdale, M. J.;
McDonald, E.; Workman, P. Inhibitors of Hsp90 and other chaper-
ones for the treatment of cancer. Expert Opin. Ther. Pat. 2004, 14,
837-847. (c) Isaacs, J. S.; Wanping, X.; Neckers, L. Heat shock
protein 90 as a molecular target for cancer therapeutics. Cancer Cell
2003, 3, 213. (d) Maloney, A.; Workman, P. Hsp90 as a new
therapeutic target for cancer therapy: the story unfolds. Expert Opin.
Biol. Ther. 2002, 2, 3-24 (e) Richter, K.; Buchner, J. Hsp90:
Chaperoning signal transduction. J. Cell. Physiol. 2001, 188, 281-
290.
(2) For lists of client proteins: (a) Pratt, W. B.; Toft, D. O. Regulation
of signaling protein function and trafficking by the hsp90/hsp70-
based machinery. Exp. Biol. Med. 2003, 228, 111-133. (b) Workman,
P. Combinatorial attack on multistep oncogenesis by inhibiting the
Hsp90 molecular chaperone. Cancer Lett. 2004, 206, 149-157. (c)
Zhang, H.; Burrows, F. Targeting multiple signal transduction
pathways through inhibition of Hsp90. J. Mol. Med. 2004, 82, 488-
499.
(11) Clevenger, R. D.; Blagg, B. S. J. Design, synthesis, and evaluation
of a radicicol and geldanamycin chimera, radamide. Org. Lett. 2004,
6, 4459-4462.
(12) Chiosis, G.; Lucas, B.; Huezo, H.; Solit, D.; Basso, A.; Rosen, N.
Development of purine-scaffold small molecule inhibitors of Hsp90.
Curr. Cancer Drug Targets 2003, 3, 371-376.
(13) The benzene ring of 3a was not designed to have exactly the same
orientation as the quinone ring of geldanamycin. Rather, the
trimethoxybenzene moiety was designed to point in the same general
direction and form a hydrogen bond with Lys112, an amino acid
which forms a hydrogen bond with the quinone ring of geldanamycin.

828 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 2
Biamonte et al.
(14) Kasibhatla, S. R.; Zhang, L.; Boehm, M. F.; Fan, J.; Hong, K.; Shi,
J.; Biamonte, M. A. Purine Analogues Having Hsp90 Inhibiting
Activity. WO 3037860, 2003.
(15) Llauger, L.; He, H.; Kim, J.; Aguirre, J.; Rosen, N.; Peters, U.; Davies,
P. Chiosis, G. Evaluation of 8-arylsulfanyl, 8-arylsulfoxyl, and
8-arylsulfonyl adenine derivatives as inhibitors of the heat shock
protein 90. J. Med. Chem. 2005, 48, 2892-2905.
(16) The free base of 2,3,5,6-tetraaminopyrimidine, a key intermediate
in the production of 2-fluoroadenines, could also be obtained from
the commercial sulfate salt and was acylated in NMP by a similar
process.
(17) Janeba, Z.; Holy, A.; Masojidkova, M. Synthesis of acyclic nucleoside
and nucleotide analogues derived from 6-amino-7H-purin-8(9H)-one.
Collect. Czech. Chem. Commun. 2000, 65, 1126-1144.
(18) (a) Ma, C.; Liu, X.; Li, X.; Flippen-Anderson, J.; Yu, S.; Cook, J.
M. Efficient asymmetric synthesis of biologically important tryp-
tophan analogues via a palladium-mediated heteroannulation reaction.
J. Org. Chem. 2001, 66, 4525-4542. (b) Flynn, B. L.; Verdier-Pinard,
P.; Hamel, E. A. Novel palladium-mediated coupling approach to
2,3-disubstituted benzo[b]thiophenes and its application to the
synthesis of tubulin binding agents. Org. Lett. 2001, 3, 651-654.
(19) Biamonte, M. A.; Shi, J.; Hurst, D.; Hong, K.; Boehm, M. F.;
Kasibhatla, S. R. Preparation of 8-(arylsulfanyl)adenines with dia-
zonium salts under mild, aerobic conditions. J. Org. Chem. 2005,
70, 717-720.
(23) Not every kinase inhibitor has a short-lasting effect. Some are
irreversible inhibitors or have slow off-rate kinetics, while others
can sequester the kinase into inactive dimers, causing their internal-
ization. For examples, see: (a) Tsou, H.-R.; Overbeek-Klumpers, E.
G.; Hallett, W. A.; Reich, M. F.; Floyd, M. B.; Johnson, B. D.;
Michalak, R. S.; Nilakantan, R.; Discafani, C.; Golas, J.; Rabindran,
S. K.; Shen, R.; Shi, X.; Wang, Y.-F.; Upeslacis, J.; Wissner, A.
Optimization of 6,7-disubstituted-4-(arylamino)quinoline-3-carboni-
triles as orally active, irreversible inhibitors of human epidermal
growth factor receptor-2 kinase activity. J. Med. Chem. 2005, 48,
1107-1131. (b) Wood, E. R.; Truesdale, A. T.; McDonald, O. B.;
Yuan, D.; Hassell, A.; Dickerson, S. H.; Pennisi, C.; Horne, E.;
Lackey, K.; Allgood, K. J.; Rusnak, D. W.; Gilmer, T. M.; Shewchuk,
L. A unique structure for epidermal growth factor receptor bound to
GW572016 (Lapatinib): relationships among protein conformation,
inhibitor off-rate, and receptor activity in tumor cells. Cancer Res.
2004, 15, 6652-6659. (c) Lichtner, R. B.; Menrad, A.; Sommer,
A.; Klar, U.; Schneider, M. R. Signaling-inactive epidermal growth
factor receptor/ligand complexes in intact carcinoma cells by
quinazoline tyrosine kinase inhibitors. Cancer Res. 2001, 61, 5790-
5795.
(24) Childress, S. J.; McKee, R. L. Chloroaminopyrimidines. J. Am. Chem.
Soc. 1950, 72, 4271-4272.
(25) (a) Beaman, A. G.; Gerster, J. F.; Robins, R. K. Potential purine
antagonists. XXVIII. The preparation of various bromopurines. J.
Org. Chem. 1962, 27, 986-990. (b) Holy, A.; Gu¨nter, J.; Dvorakova,
H.; Masojidkova, M.; Andrei, G.; Snoeck, R.; Balzarini, J.; De Clercq,
E. Structure-antiviral activity relationship in the series of pyrimidine
and purine N-[2-(2-phosphonomethoxy)ethyl] nucleotide analogues.
1. Derivatives substituted at the carbon atoms of the base. J. Med.
Chem. 1999, 42, 2064-2086.
(20) Le Brazidec, J.-Y.; Kamal, A.; Busch, D.; Thao, L.; Zhang, L.;
Timony, G.; Grecko, R.; Trent, K.; Lough, R.; Salazar, T.; Khan,
S.; Burrows, F.; Boehm, M. F. Synthesis and biological evaluation
of a new class of geldanamycin derivatives as potent inhibitors of
Hsp90. J. Med. Chem. 2004, 47, 3865-3873.
(21) Prepared in a manner analogous to 8 (Scheme 1, route A) but
replacing BuI with other alkylating agents.
(22) Purified soy-derived L-R-phosphatidylcholine (Phospholipon 90H)
was purchased from American Lecithin Company.
JM0503087