Dihydro-[1H]-quinolin-2-ones as retinoid X receptor (RXR) agonists for potential treatment of dyslipidemia

Article abstract of DOI:10.1016/j.bmcl.2007.01.049

A number of RXR modulators with novel structural features were synthesized and screened in the functional assays. The synthesis and the structure-activity relationship within the series of compounds will be presented. Some in vivo data generated in the mo

Full text of DOI:10.1016/j.bmcl.2007.01.049

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3491–3496
Dihydro-[1H]-quinolin-2-ones as retinoid X receptor (RXR)
agonists for potential treatment of dyslipidemia
Bharat Lagu,* Rimma Lebedev, Barbara Pio, Maria Yang and Patricia D. Pelton
Johnson and Johnson Pharmaceutical Research and Development, Cranbury, NJ 08512, USA
Received 16 November 2006; revised 9 January 2007; accepted 11 January 2007
Available online 25 January 2007
Abstract—A number of RXR modulators with novel structural features were synthesized and screened in the functional assays. The
synthesis and the structure–activity relationship within the series of compounds will be presented. Some in vivo data generated in the
models for dyslipidemia and diabetes will also be presented.
ꢀ 2007 Published by Elsevier Ltd.
In the accompanying communication we have disclosed
the structure-activity relationship in the compounds
shown by a general structure A (Fig. 1) based on the
structure of a known RXR modulator 1.1,2 Some com-
pounds from the series showed higher potency for
induction of ABCA1 mRNA, a target gene of the
RXR–LXR heterodimeric complex over induction of
aP2 mRNA, a target gene of RXR–PPARc heterodi-
mers. Encouraged by the results, we decided to explore
the SAR in some detail within this series. Attention
was focused on the trans double bond between the car-
boxylic acid and the phenyl ring. Almost all the RXR
agonists reported thus far in the literature contain the
carboxylic acid moiety conjugated to a double bond or
to an aryl ring as evidenced by the structures of some
of the known rexinoid agonists (2–8) shown in Figure
1.3–10 A plausible explanation for this observation might
be that most of the structures were inspired by the struc-
ture of the endogenous ligand for RXR, 9-cis-retinoic
acid.11 The cyclopropyl ring is generally considered a
double bond surrogate by organic chemists. However
to the best of our knowledge, RXR agonists incorporat-
ing cyclopropylcarboxylic acid in their structures such as
10 have not been reported. Structures of some com-
pounds that contain a cyclopropyl group (5 and 6) with
a ‘9-cis locked conformation’ have been published.10
The only example of RXR agonist containing a carbox-
ylic acid without the a,b-unsaturated double bond (8)
was reported by GlaxoSmithKline.9
In this communication the synthesis and structure–activ-
ity relationship in the analogs of 1 where the trans dou-
ble bond was replaced with a cis double bond, alkyne
(general structure A), a single bond (9), an alkoxy group
and a cyclopropyl ring (10) is described.
Compound 13 containing a cis double bond was synthe-
sized from the known intermediate 11 via Wittig olefin-
ation as shown in Scheme 1.2 The cis isomer 13 was
isolated from the trans isomer by a careful column chro-
matography, although the final compound was found to
be contaminated with 5% of the compound with a trans
double bond (14). The double bond of 14 was reduced
by hydrogenation to obtain 15a–g. Hydrogenation of
compounds containing tri-substituted double bonds 16
and 17 yielded compounds 18 and 19, respectively, as
racemic mixtures. Sonogashira coupling of propargyl
alcohols with 20 yielded intermediates 21 and 22, which
upon TEMPO oxidation gave the corresponding car-
boxylic acids 23 and 24. In the case of 24, the alkynyl
bond was hydrogenated to give compound 25. A Suzuki
coupling reaction between 3,4-dihydro-1H-quinolin-2-
one (26b) and acetic acid 3-bromo-4-trifluoroethoxyt-
phenyl ester yielded phenol 27 with the acetoxy group
having cleaved during the reaction and work-up. The
phenol was then alkylated with the a-bromo esters to
yield 28 and 29 upon hydrolysis.
Keywords: RXR; LXR; Dyslipidemia; ABCA-1; Triglycerides.
The racemic cyclopropyl analogs were synthesized by
the addition of an ylide generated from trimethylsul-
foxonium iodide and sodium hydride in DMSO, to
*
Corresponding author at Present address. Novartis Institute for
Biomedical Research, Cambridge, MA, USA. Tel.: +1 617 871 4257;
fax: +1 617 871 4081; e-mail: bharat.lagu@novartis.com
0960-894X/$ - see front matter ꢀ 2007 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2007.01.049

3492
B. Lagu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3491–3496
R2O
O
R1
N
F3CO
S
OH
O
NH
O
N
O
O
A
1
O
OH
HO
O
OH
O
N
2
N
3
4
N
OH
O
O
5
OH
7
R =
O
O
R
OH
8
O
OH
6
R2O
R2O
R1
R1
N
OH
OH
O
N
O
( )n
O
O
9
10
Figure 1.
the cinnamate 12b.12 The removal of the protecting
group under acidic conditions gave the compounds
31a–e. In all cases, only the trans diastereomer was
obtained as a racemic mixture.13 Compounds 31a and
31c were converted into a mixture of diastereomers
32 and 35, respectively, by attaching chiral auxiliaries
as shown in Scheme 2. The diastereomers of 32 were
separated by using chiral HPLC.14 The chiral auxiliary
was removed by hydrogenation to yield the cyclopro-
pylcarboxylic acids 33 and 34 in the enantiomerically
pure forms. In the case of 35, the oxazolidinone moiety
was removed by hydrolysis after separating the two
diastereomers by flash column chromatography. At
the present time, the absolute stereochemistry at the
chiral centers for 33, 34, 36, and 37 has not been
established.
co-transfection assay with the yeast GAL4 DNA bind-
ing domain fused to the ligand binding domain of
RXRa and the yeast promoter driving luciferase
expression Table 5.
In the ABCA-1 mRNA induction assay, the maximal
efficacy of the newly synthesized RXR agonists was
about half of that compared to the known LXR agonist
TO-901317.15 Compound 13 containing a cis double
bond showed significantly reduced potency to activate
the RXR–LXR heterodimer (EC50 = 886 nM) compared
to 14b (EC50 = 3.4 nM). The alkyne-containing com-
pound 23 was found to be inactive in the functional
assay. It is possible that the trajectory of the carboxylic
acid is important in order to make the key contact with
the Arg316 residue in the binding site of the RXR recep-
tor.9 Compounds 16 and 17 bearing b and a methyl
group on the double bond with respect to the carboxylic
acid showed completely different potency profiles. While
17 was found to be active in ABCA-1 gene induction as-
say and the RXR co-transfection assay, 16 was found to
be inactive in both assays at the testing concentrations.
Analog 31a where the double bond was replaced by a
cyclopropyl moiety showed potent activity in both
assays (EC50 = 17.4 and 19.2 nM, respectively). This
activity is quite interesting given the fact that these are
The compounds were initially screened for their ability
to induce expression of ABCA-1 (ATP binding cas-
sette protein), a key gene involved in the reverse cho-
lesterol transport and known to be upregulated by
LXR agonists.15,16 A few compounds were screened
in the aP2 gene induction assay indicative of the acti-
vation of RXR–PPARc heterodimer (Tables 1–4).17 In
order to determine the activity at the RXR receptor
(specifically RXRa), the compounds were tested in a

B. Lagu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3491–3496
3493
F3CO
F3CO
O
OH
F3CO
O
O
a
b
H
O
N
O
N
O
N
O
11
12a
R2O
13
F3CO
R1 R2O
N
R1
R4
F3CO
R4
c
c
OH
O
O
OH
O
N
N
N
O
COOH
COOH
O
O
R3
R3
14a-g
15a-g
16 R3= Me, R4 = H
18 R3= Me, R4 = H
19 R3 = H, R4 = Me
17 R3 = H, R4 = Me
F3CO
F3CO
F3CO
F3CO
(m = 1)
f
O
N
d
O
N
e
O
N
O
N
Br
( )n
( ) m
COOH
OH
COOH
20
21 n = 0
22 n = 1
23 m = 0
24 m = 1
25
F3CO
F3CO
R
R
O
i, j
g
h
O
N
Br
O
N
O
N
OH
O
N
B
OH
O
O
O
26a
26b
27
28 R = H
29 R = Me
Scheme 1. Reagents and conditions: (a) THF, diphenylphosphonic acid ethyl ester, Triton B, À78 to 0 ꢁC; (b) CH2Cl2, TFA, rt, 8 h, >90%; (c) 10%
Pd-C, H2, EtOAc or MeOH, >95%; (d) propargyl alcohol or 3-butyn-1-ol, NMP, Pd(PPh3)4, 120 ꢁC 15 min, microwave, 26–40%; (e) TEMPO,
CH3CN, phosphate buffer, bleach, NaClO2, 55%; (f) 10% Pd–C, H2, rt, overnight, 100%; (g) Pd(OAc)2, 2-(dicyclohexylphosphino)biphenyl, dioxane,
Et3N, pinacolborane, 85 ꢁC, 4 h, 60%; (h) acetic acid-3-bromo-4-trifluoromethoxyphenyl ester, THF–DMPU (10:1), n-BuLi, dimethyl-t-
butoxycarbonylmethylphosphonate, À78 to 0 ꢁC, 54%; (i) ethyl-2-bromoisobutyrate, CH3CN, Cs2CO3, 85 ꢁC, 12 h; (j) THF–MeOH, aq NaOH,
rt, overnight, 40%.
R1 R2O
N
R1 R2O
N
R1 R2O
N
a
b
OtBu
O
COOH
O
COOtBu
O
O
12b
30
31a-e
(+)
Only trans isomer isolated
F3CO
O
O
F3CO
F3CO
F3CO
O
N
O
O
O
d, e
c
O
N
O
N
O
N
+
OH
OH
OH
32
(-) 33
(+) 34
31a
Absolute stereochemistry not established
F3CO
F3CO
F3CO
F3CO
O
O
O
O
O
f, g
O
N
O
N
h
O
N
O
N
N
OH
O
+
OH
OH
31c
35
Ph
36
(-)
(+)
37
Absolute stereochemistry not established
Scheme 2. Reagents and conditions: (a) trimethylsulfoxonium iodide, NaH, DMSO, CH2Cl2, rt, À50 ꢁC, 3 h, 72–90%; (b) CH2Cl2, TFA, rt, 8 h,
>90%; (c) (R)-(+)-methyl-2-naphthalene methanol, EDCI, HOBt, DMF, Et3N, cat. DMAP, rt, 12 h; (d) separate diastereomers by chiral HPLC (see
Ref. 14); (e) 10% Pd–C, EtOAc, H2, rt, 8 h; (f) SOCl2, CH2Cl2, rt, overnight. Then NaH, (S)-(À)-4-benzyl-2-oxazolidinone, THF, rt; (g) separate
diastereomers by column chromatography; (h) LiOH, THF–MeOH–H2O.
the first reported RXR agonists that contain cyclopro-
pane-1-carboxylic acid. The activity of 31a raised a
possibility of reducing the double bond completely while
retaining the ABCA-1 activity. The only example of
such saturated bond to the carboxylic acid has been
reported by GlaxoSmithKline, where compound 8 was
found to be fivefold less potent than the parent com-
pound 7. The authors suggested that lower potency of

3494
B. Lagu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3491–3496
Table 3. SAR in cyclopropane-containing RXR agonists
Table 1. Linker Chain Modification
F3CO
Compound
ABCA-1
EC50 nM
RXRa co-trans
EC50 (nM)
O
a
O
N
X
OH
F3CO
O
O
N
OH
17.4
19.2 (158%)
115.8 (87%)
Compound
X
ABCA-1
RXRa co-transfection
31a
a
EC50 (nM) EC50 (nM)
a
( ) Racemate
F3C
14b
13
3.6 (81%)
886 (75%)
>3000
1.8 (115%)
39.1 (97%)
ND
O
O
>3000
O
O
O
O
N
N
N
N
OH
OH
OH
OH
31b
23
(
) Racemate
CF3
O
16
ND
>10,000
O
>3000
1524.0
>3000
5.0 (85%)
17
100 (49%)
17.4 (ND)
27.9 (81%)
19.2 (88%)
31c
(
) Racemate
Me
O
31a
( )
O
207.5 (50%)
144.0 (108%)
15b
25
47 (70%)
>3000
7.2 (79%)
527 (70%)
760 (85%)
31d
(
) Racemate
CF3
O
O
19
>3000
31e
18
28
29
347 (160%)
>3000
37.7 (99%)
ND
(
) Racemate
F3CO
O
O
O
O
O
O
O
N
N
N
N
OH
OH
OH
OH
>3000
>3000
>3000
>3000
2.3 (85%)
361 (92%)
10 (70%)
198 (50%)
>3000
ND
33
(-)-trans
ND, not determined.
a The numbers in parentheses represent %maximal activation in com-
parison to compound 1.
F3CO
F3CO
F3CO
O
O
O
34
(+)-trans
Table 2. SAR for compounds 15a–g
R2
R1
N
OH
O
36
(+)-trans
O
Compound R1
R2
ABCA-1
a
EC50 (nM) EC50 (nM)
RXRa co-trans
a
37
15a
15b
15c
15d
15e
15f
15g
Me
Et
OCF3
OCF3
OCF3
53
45.9 (91%)
7.2 (79%)
3.4 (80%)
1.4 (100%)
70.2 (90%)
36.5 (79%)
361(60%)
(+)-trans
47 (70%)
1 (59%)
>3000
ND, not determined.
a The numbers in parentheses represent %maximal activation in com-
parison to compound 1.
iPr
CH2CF3 OCF3
Et
OCH2CF3 >3000
OCH2CF3 369 (45%)
iPr
iPr
OMe
>3000
Increasing the tether length from two to three carbons
resulted in a significant loss of potency for compound
25. Compound 19 bearing a methyl group on the a car-
bon with respect to the carboxylic group was less potent
than the compound 18 bearing methyl group on the b
carbon. This observation is in contrast to the results ob-
served for the double bond containing analogs 16 and
17. Interestingly, compounds 28 and 29, which bear
the same number of atoms as 15b but contain an oxygen
atom in the linker chain, were not active up to 3 lM
concentration.
ND, not determined.
a The numbers in parentheses represent %maximal activation in com-
parison to compound 1.
8 ‘presumably resulted from the entropic cost of a more
flexible compound.’ Similarly, in the present series,
compound 15b was found to be active but fourfold less
potent than 14b in both functional assays (EC50 = 47
and 7.2 nM in the ABCA-1 and RXR co-transfection
assay, respectively).

B. Lagu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3491–3496
3495
Table 4. Selectivity for the RXR agonists
a
ABCA-1 EC50 (nM)
Compound
aP2 EC50 (nM)
RXR co-trans EC50a(nM)
O
F3CO
S
NH
O
O
N
N
180 (100%)
139
0.9 (100%)
7.2 (79%)
O
1
F3CO
OH
47 (70%)
>3000
1200
>3000
581
O
15b
F3C
O
70.2 (90%)
2.3 (85%)
10 (70%)
O
O
N
N
OH
OH
O
O
15e
F3CO
>3000
33
CF3
O
O
>3000
266
O
N
OH
36
a The numbers in parenthesis represent %maximal activation in comparison to compound 1.
Table 5. Effects of 15b on serum cholesterol and triglyceride levels following 8 days of treatment in Sprague–Dawley rats fed a high cholesterol diet
Dose (mg/kg)
HDL-C (mg/dL)
LDL-C (mg/dL)
Total Cholesterol (mg/dL)
Triglycerides (mg/dL)
Vehicle
0.3 mpk
21
2
52
53
46
46
5
5
4
4
321 20
343 18
313 18
302 14
228 18**
252 24*
267 10*
218 20
320 38
279 32
253 24
270 42
322 34
286 12
29 1**
30 2**
37 1**
39 2**
42 1**
47 1**
1 mpk
3 mpk
10 mpk
30 mpk
10 mpk 1 (0.5% methocel)
35 2**
38 2*
41
1
Data are expressed the means SEM and were analyzed using one-way ANOVA and Dunnet’s multiple comparison test (*p < 0.05, **p < 0.001). All
comparisons are made relative to the Vehicle controls.
Similar to 15b, a number of analogs with differing R1
and R2 groups (15a–g) were active in the RXR co-trans-
fection assay with EC50 values ranging from 1.4 to
361 nM (Table 2). Some of the trends were common in
the compounds with or without a double bond, e.g.
for R1 group, the ethyl or isopropyl groups were found
to give compounds that were more potent than when
R1 = Me or for R2 group, compounds with trifluoroeth-
oxy group showed more functional activity than the
compounds with trifluoroethoxy group at the R2 posi-
tion.1 Compounds 15a–c had high potency in both the
ABCA-1 induction and RXR co-transfection assays,
particularly compound 15c. Although 14b is about four
times more potent in the RXRa co-transfection assay
than 15b (an observation similar to the one reported
for 7 and 8 by the researchers from Glaxo SmithKline),9
it must be pointed out that other analogs such as 15c
and 15d are almost as potent as 14b in the functional
assays.
in the ABCA-1 gene induction and RXR co-transfection
assays. The results are summarized in Table 3. In gener-
al, the compounds (31a–e, 33–37) showed good activity
(EC50 = 2.3–361 nM) in the RXR co-transfection assay.
However, the compounds showed significant differences
in the activation of the RXR–LXR heterodimer
(EC50 = 17.4 nM to >3 lM). The individual enantio-
mers were synthesized for compounds 31a and 31c.
The (À)-enantiomers (i.e. compounds giving negative
value for optical rotation) 33 and 36 were found to be
more potent than the (+)-enantiomers (34 and 37) and
the racemates (31a and 31c) in the co-transfection assay.
At the present time the absolute stereochemistry shown
for these enantiomers has not been assigned.
With the active compounds in hand, we screened a few
compounds in the selectivity screen for their ability to
activate the RXR–LXR (ABCA-1 mRNA induction)
and RXR–PPARc aP2 mRNA induction) heterodimer
complexes. The results are summarized in Table 4. While
1 is equipotent in activating both RXR–LXR and RXR–
PPARc heterodimer complexes, 15b showed some
Compounds where the cinnamic acid double bond was
replaced with a cyclopropyl group were also screened

3496
B. Lagu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3491–3496
2. Pfahl, M.; Tachdjian, C.; Al-Shamma, H.; Giachino, A.
F.; Jakubowicz-Jaillardon, K.; Guo, J.; Boudjelal, M.;
Zapf, J. W. PCT Int. Appl. WO 2003075924 A1, 2003.
3. Boehm, M. F.; Zhang, L.; Baeda, B. A.; White, S. K.;
Mais, D. E.; Berger, E.; Suto, C. M.; Goldman, M. E.;
Heyman, R. A. J. Med. Chem. 1994, 37, 2930.
selectivity for activating the RXR-LXR complex
(EC50 = 47 nM vs EC50 = 1200 nM for RXR–PPARc).
Compound 15e had weak activity while compounds 33
and 36 were found to be selective for activating RXR–
PPARc heterodimer over the RXR–LXR heterodimer.
Compound 15b was selected as a representative example
for evaluation in the in vivo models for diabetes and dysl-
ipidemia based on the satisfactory pharmacokinetic pro-
file in rats (3 mg/kg oral dose in 0.5 methocel; oral
bioavailability = 71%; Cmax = 678 ng/mL; AUC = 6129
ng h/mL; T1/2 = 24 h; CL = 4.6 mL/min/kg).
4. Vuligonda, V.; Lin, Y.; Chandraratna, R. A. S. Bioorg.
Med. Chem. Lett. 1996, 6, 213.
5. Michellys, P. Y.; Ardecky, R. J.; Chen, J. H.; Crombie, D.
L.; Etgen, G. J.; Faul, M. M.; Faulkner, A. L.; Grese, T.
A.; Heyman, R. A.; Karanewsky, D. S.; Klausiing, K.;
Leibowitz, M. D.; Liu, S.; Mais, D. A.; Mapes, C. M.;
Marschke, K. B.; Reifel-Miller, A.; Ogiliveie; Rungta, D.;
Thompson, A. W.; Tyhonas, J. S.; Boehm, M. F. J. Med.
Chem. 2003, 46, 2683.
6. Canan Koch, S. S.; Dardashti, L. J.; Cesario, R. M.;
Croston, G. E.; Boehm, M. F.; Heyman, R. A.; Nadzan,
A. M. J. Med. Chem. 1999, 42, 742.
7. Faul, M. M.; Ratz, A. M.; Sullivan, K. A.; Trankle, W.
G.; Winneroski, L. L. J. Org. Chem. 2001, 66, 5772.
8. Kagechica, H. Curr. Med. Chem. 2002, 9, 591.
9. Haffner, C. G.; Lenhard, J. M.; Miller, A. B.; McDougald,
D. L.; Dwornik, K.; Itoop, O. R.; Gampe, R. T.; Xu, H.
E.; Blanchard, S.; Montana, V. G.; Consler, T. G.;
Bledsoe, R. K.; Ayscue, A.; Croom, D. J. Med. Chem.
2004, 47, 2010.
As a model of dyslipidemia, Sprague–Dawley male rats
were fed a high cholesterol, atherogenic diet ad libitum,
which typically induces a fivefold increase in serum cho-
lesterol after 14 days. Compound 15b was orally admin-
istered after 6 days on the diet in 0.5% methocel for a
total of 8 days. Serum cholesterol, HDL-C, LDL-C
and triglycerides were measured. Treatment with com-
pound 15b elicited a statistically significant increase in
HDL-C and a decrease in LDL-C levels as compared
to the vehicle control animals. The Cmax values ranged
from an average of 58.5 ng/mL (0.3 mg/kg) to 7510 ng/mL
(30 mg/kg). A significant increase in serum triglyceride
levels was not observed. First generation rexinoid ago-
nists such as LGD1069 (TagretinTM) elevate the triglycer-
ide levels in both rat and human.18,19 On the other hand,
15b failed to lower blood glucose in the db–db mouse
model for diabetes following 11 days of treatment
(data not shown). These results are in agreement with
the observed selectivity of the compound to activate
RXR–LXR heterodimer over the RXR–PPARc
heterodimer.
10. Vuligonda, V.; Thacher, S. M.; Chandraratna, R. A. S.
J. Med. Chem. 2001, 44, 2298.
11. Levin, A. A.; Sturzenbecker, L. J.; Kazmer, S.; Bosakow-
ski, T.; Huselton, C.; Allenby, G.; Speck, J.; Kratzeisen,
C.; Rosenberg, M.; Lovey, A.; Grippo, J. F. Nature 1992,
355, 359.
12. Corey, E. J.; Chykovsky, M. J. Am. Chem. Soc. 1965, 87,
1353.
13. Relative stereochemistry assigned based on 1H NMR.
Solladie-Cavallo, A.; Isarno, T. Tetrahedron Lett. 1999,
40, 1579.
14. The diastereomers were separated by using a Varian Prep
Star 218 HPLC instrument with Chiralpak-AD
(21.2 · 250 mL) column as the stationary phase and 80%
n-hexane-20% EtOH with 0.05% TFA as the eluent. With
the flow rate of 5 mL/min, injection volume of 0.5 mL, the
retention times (detector at 254 nm) for the two diaste-
reomers were 18.3 min and 21.4 min, respectively.
15. Repa, J. J.; Turley, S. D.; Lobaccaro, J. M. A.; Medina, J.;
Li, L.; Lustig, K.; Shan, B.; Heyman, R. A.; Dietschy, J.
M.; Mangelsdorf, D. J. Science 2000, 289, 1524.
16. Attie, A. D.; Kastelein, J. P.; Hayden, M. R. J. Lipid Res.
2001, 42, 1717.
We have reported some structurally unique and potent
RXR agonists where the double bond conjugated with
the carboxylic acid was replaced with either a single
bond or a cyclopropyl group. Some of the compounds
were found to selectively activate the RXR–LXR het-
erodimer complex over the RXR–PPARc heterodimer
complex.
Acknowledgments
17. Burris, T. P.; Pelton, P. D.; Zhou, L.; Osborne, M. C.;
Cryan, E.; Demarest, K. T. Mol. Endocrinol. 1999, 13, 410.
18. Bershad, S. A.; Rubinstein, J. R.; Paterniti, N. A.; Le, S.
C.; Poliak, B.; Heller, H. N.; Ginsberg, R.; Fleischmajer,
R.; Brown, W. V. N. Engl. J. Med. 1985, 313, 981.
19. Miller, V. A.; Benedetti, F. M.; Rigas, J. R.; Verret, A. L.;
Pfister, D. G.; Straus, D.; Kris, M. G.; Crisp, M.;
Heyman, R.; Loewen, G. R.; Truglia, J. A.; Warrell, R.
P., Jr. J. Clin. Oncol. 1997, 15, 790.
The authors thank Dr. Keith Demarest and Dr. William
Murray for their encouragment and support.
References and notes
1. Lagu, B.; Pio, B.; Lebedev, R.; Yang, M.; Pelton, P. D.
Bioorg. Med. Chem. Lett. (accompanying paper).