The synthesis and characterization of BMS-204352 (MaxiPost) and related 3-fluorooxindoles as openers of maxi-K potassium channels

Article abstract of DOI:10.1016/S0960-894X(02)00101-4

3-Aryl-3-fluorooxindoles can be efficiently synthesized in two steps by the addition of an aryl Grignard to an isatin, followed by treatment with DAST. Oxindole 1 (BMS-204352; MaxiPost) can be isolated using chiral HPLC or prepared by employing chiral res

Full text of DOI:10.1016/S0960-894X(02)00101-4

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1023–1026
The Synthesis and Characterization of BMS-204352
(MaxiPostTM) and Related 3-Fluorooxindoles as Openers of
Maxi-K Potassium Channels
Piyasena Hewawasam,a Valentin K. Gribkoff,b Yadagiri Pendri,c Steven I. Dworetzky,b
Nicholas A. Meanwell,a Eduardo Martinez,c Christopher G. Boissard,b
Debra J. Post-Munson,b Joanne T. Trojnacki,b Krishnaswamy Yeleswaram,d
Lorraine M. Pajor,d Jay Knipe,d Qi Gao,e Robert Perronef and John E. Starrett, Jr.a,*
aDepartment of Chemistry, Bristol-Myers Squibb Pharmaceutical Research Institute, 5 Research Parkway,
Wallingford, CT 06492, USA
bDepartment of Neuroscience, Bristol-Myers Squibb Pharmaceutical Research Institute, 5 Research Parkway,
Wallingford, CT 06492, USA
cDepartment of Process Research & Development, Bristol-Myers Squibb Pharmaceutical Research Institute,
5 Research Parkway, Wallingford, CT 06492, USA
dDepartment of Metabolism and Pharmacokinetics, Bristol-Myers Squibb Pharmaceutical Research Institute,
5 Research Parkway, Wallingford, CT 06492, USA
eDepartment of Analytical Research and Development, Bristol-Myers Squibb Pharmaceutical Research Institute,
5 Research Parkway, Wallingford, CT 06492, USA
fDepartment of Pharmaceutics, Bristol-Myers Squibb Pharmaceutical Research Institute, 1 Squibb Drive,
New Bruswick, NJ 08903, USA
Received 8 November 2001; accepted 1 February 2002
Abstract—3-Aryl-3-fluorooxindoles can be efficiently synthesized in two steps by the addition of an aryl Grignard to an isatin,
followed by treatment with DAST. Oxindole 1 (BMS-204352; MaxiPostTM) can be isolated using chiral HPLC or prepared by
employing chiral resolution. Cloned maxi-K channels are opened by 1, which demonstrates a brain/plasma ratio >9 in rats. # 2002
Elsevier Science Ltd. All rights reserved.
Annually, there are more than 700,000 new cases of
stroke in the US, representing a major unmet medical
need.1 Numerous neuroprotective therapies (approaches
that protect neurons ‘at risk’ following a stroke) have
been, or are currently undergoing clinical testing, but thus
far none has been approved due to lack of demonstrated
efficacy or because of poor side effect profiles.
diverse family of transmembrane proteins that serve a
variety of functions, including controlling cell excit-
ability and setting the resting membrane potential in
excitable cells.2 Calcium-activated potassium (KCa
)
channels are a family of these channels that share a
dependence on intracellular calcium ion concentration
for activity but may also be regulated by membrane
potential and phosphorylation state.3 On the basis of
their single-channel conductance in symmetrical K+
solutions, KCa channels have been broadly classified
into three subfamilies: the large conductance (BK or
maxi-K) channels, which exhibit a single channel con-
ductance of >100 picosiemens (pS), intermediate con-
ductance channels (50–100 pS) and small conductance
channels (<50 pS).4 Maxi-K channels are present in
many excitable cell types including neurons and various
We have approached the problem of neuroprotective
therapy by developing potent and specific openers of
large-conductance, calcium-activated (maxi-K) potas-
sium channels. Potassium channels are a structurally
*Corresponding author. Tel.: +1-203-677-6306; fax: +1-203-677-
7702; e-mail: john.starrett@bms.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00101-4

1024
P. Hewawasam et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1023–1026
types of smooth muscle cells. Because maxi-K channels
are thought to be important regulators of cellular excit-
ability and function, modulators of these channels have
emerged as potentially useful agents in the therapy of
various disease states associated with both the central
nervous system and smooth muscle.5
using a Chiracel-OD column, eluting with 9:1 hexanes/
isopropanol to afford (+)-1 and (ꢀ)-6.
For electrophysiological evaluation of maxi-K opening
properties, compounds were tested using two-electrode
voltage clamp recording from Xenopus laevis oocytes
injected with hSlo1 mRNA, as previously described.10
Voltage-clamp protocols ranged from a holding poten-
tial of ꢀ60 mV to a maximal potential of +140 mV
with +20 mV increments. The maxi-K current, defined
as the iberiotoxin-sensitive component of total outward
current, was measured in the absence or in the presence
of 20 mM drug. Compounds were prepared as 10 mM
stock solutions in dimethyl sulfoxide (DMSO) and
diluted in recording medium to the desired study con-
centration. When insolubility precluded evaluation at 20
mM, current was measured at 10 mM (5c). The increase
in outward current in the presence of drug is reported as
percent of drug free control, over a voltage step to
+140 mV, and the data are an average of experiments
conducted in at least five different oocytes.
In brain, pharmacological opening of maxi-K channels
protects neurons by enhancing an endogenous neuro-
protective mechanism.6 Openers of these channels
appear to synergize with calcium, making the modula-
tion of channel function largely dependent on the pre-
sence of high levels of intracellular calcium, such as is
encountered during neuronal ischemia. Maxi-K channel
opening stimulated by Ca2+ and the pharmacological
opener would function to limit further Ca2+ entry by
hyperpolarizing neurons, thereby reducing voltage-
dependent calcium entry points via some Ca2+ channels
and possibly NMDA receptors. The cascade of events
leading to ischemic neuronal death is initiated by
abnormally high Ca2+ entry and therefore limiting
intracellular access to Ca2+ via opening of maxi-K
channels should interfere with this neurotoxic cascade.
The opener would not be expected to significantly affect
maxi-K channels in non-ischemic tissue, where the
intracellular Ca2+ levels are low, reducing the potential
for undesirable side effects.
We have identified the novel fluorooxindole 1 (BMS-
204352; MaxiPostTM) as a potent opener of maxi-K
channels (Fig. 1). The electrophysiological character-
ization of 1 and activity in models of stroke have
recently been reported.7 It is of interest to note that in
addition to maxi-K activation, 1 has also been reported
to activate KCNQ4 potassium channels.8 However, in
the current study, cloned maxi-K channels were expres-
sed in cells with minimal native background potassium
current, and therefore it is reasonable to assume that the
observed effects are due to modulation of maxi-K
channels. Herein, we describe the synthesis and maxi-K
channel activity of analogues related to 1, as well as the
physical characterization, pharmacokinetic profile and
stereospecific synthesis of 1.
Scheme 1. Synthesis of 3-fluorooxindoles.
The fluorooxindoles were prepared in a straightforward
manner as depicted in Scheme 1.9 Addition of an aryl
Grignard reagent 2 to the sodium salt of isatin 3 in THF
gave the corresponding 3-aryl-3-hydroxyindol-2-ones 4
in 80–95% yield. Fluorination of the tertiary hydroxyl
moiety of 4 with DAST in dichloromethane provided
the desired 3-aryl-3-fluorooxyindol-2-ones 5 in 90–95%
yield. The presence of fluorine was confirmed by mass
spectroscopy and by fluorine NMR as depicted in Table
1. The racemate 5c was resolved by preparative HPLC
Table 1. Effects of 3-fluorooxindoles on maxi-K current in oocytes
Compda
R1
R2
R3
Maxi-K
currentb
19F N MR
(ppm)
(S)-(+)-1
5a
5b
5c
5d
5e
5f
5g
5h
5i
(R)-(ꢀ)-6
6-CF3
H
Cl
H
H
Cl
Cl
OCH
Cl
Cl
Cl
Cl
Cl
OCH3
H
H
OCH3
OCH3
I
ꢀ159.87
ꢀ154.05
ꢀ155.44
ꢀ159.87
ꢀ158.67
ꢀ160.18
ꢀ158.26
ꢀ158.81
ꢀ158.55
ꢀ157.38
ꢀ159.87
100
107
170c
106
106
95
110
104
112
I
6-CF3
6-CF3
5-CF3
6-CNCl
6-I
5-F
5-Br
5-Me
6-CF3
3
OCH3
OCH3
OCH3
OCH3
OCH3
aAll new compounds exhibited spectroscopic and elemental analyses
consistent with the assigned structures (I=insoluble at 1 mM).
b% of control of outward potassium current in oocytes injected with
hSlo. Unless otherwise noted, compounds were evaluated at 20 mM.
cDue to insolubility at 20 mM, analysis was conducted at 10 mM.
Figure 1. Structure of 1 (BMS-204352; MaxiPostTM).

P. Hewawasam et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1023–1026
1025
The maxi-K activity in Table 1 is shown as a percent of
control opening; 100% denotes a compound which had
no effect on the channel. Previous studies from our
laboratories of maxi-K activity of related 3-arylox-
indoles have revealed optimal activity with a 2-oxo-5-
chloro substitution pattern.11,12 This information was
utilized in the selection of synthetic targets in the cur-
rent study. The structure–activity relationship in this
series of oxindoles is exquisitely sensitive to the position
and properties of the substituents. Significant channel
opening was observed with 5c, which contains a 6-sub-
stitued trifluoromethyl oxindole and a 5-chloro-2-ani-
sole moiety at C-3. The unsubstituted phenyl analogues
5a and 5b had no significant effect. Moving the tri-
fluoromethyl group to the 5-position (5d) resulted in a
loss of activity, as does replacement with other electron-
withdrawing functionalities (5e–h). Interestingly, the
5-methyl derivative 5i, demonstrated modest opening
activity. The enantiomers of 5c, compounds 1 and 6,
could not be evaluated in the oocyte preparation due to
a lack of solubility in the assay medium.
current than either racemate 5c or the opposite
enantiomer 6.
To circumvent the necessity for the use of a chiral
HPLC column to prepare 1, a chiral resolution was
developed as outlined in Scheme 2.13 Starting with
commercially available 2-methoxyphenyl acetic acid (7),
chlorination with sulfuryl chloride followed by Fischer
esterification of the acid afforded ester 8. To introduce
the trifluoromethyl phenyl group, ester 8 was deproton-
ated with two equivalents of potassium hexamethyldisi-
lazide and then treated with commercially available
4-(trifluoromethyl)-2-nitroflurobenzene (9). Ipso dis-
placement of the fluorine, followed by quenching of the
benzylic anion with N-fluorobenzenesulfonimide (Accu-
fluor1), an electrophilic source of fluorine, provided
racemic ester 10. Saponification of ester 10 gave racemic
carboxylic acid 11, which was resolved into its enantio-
mers 12 and 13 through the formation of a pair of dia-
stereomeric salts with (S)-(ꢀ)-a-methyl benzyl amine in
>99% ee. Reduction of the nitro group of 12 with
sodium dithionite, followed by cyclization with HCl
cleanly afforded 1 in 85% yield with an enantiomeric
excess of >99% (as determined by chiral HPLC). In a
similar manner, reduction and cyclization of 13 afforded 6.
In order to further characterize the effects of enantio-
mers 1 and 6 on outward K+ currents, they were eval-
uated using whole-cell patch clamp recordings in HEK
cells stably-transfected with hSlo1. Utilizing a cell hold-
ing potential of ꢀ60 mV, a series of three voltage steps
(to +20, +40 and +60 mV) was applied to the cell.
The final 25 data points of the aggregate current elicited
by the +60 mV step were averaged to determine the
current amplitude. In this preparation, both enantio-
mers were very effective in increasing maxi-K current
(Table 2). Although not significantly different, 1 con-
sistently produced a more robust increase in outward
Recrystallization of 1 from hexane/CH2Cl2 afforded
large plate-like crystals (mp 198–200 ꢁC) suitable for
single-crystal X-ray analysis.14 Using the anomalous
scattering from chlorine atoms, the absolute configura-
tion at the asymmetrical carbon atom was established as
S. The trifluoromethyl group was disordered. The
molecule adopted a conformation with a dihedral angle
of 99.03(9)ꢁ between the two aromatic rings. In the
crystal, molecules form infinitely long chains along the
crystallographic b-axis through N–H. . .O¼C hydrogen
bonds, and a three dimensional network of interactions
involving fluorine atoms (F. . .CH) was observed. A
similar analysis of 6 confirmed the R-configuration for
this enantiomer. Solid-state stability studies indicated
excellent stability for 1; it retained full potency after
storage in sealed vials for 4 weeks at 70 ꢁC, 50 ꢁC/75%
relative humidity (RH), or 23 ꢁC/1000 footcandles light.
The samples stored at 70 ꢁC or 50 ꢁC/75% RH also
showed no apparent change in crystal form based on X-
ray powder diffraction analyses. To examine the poten-
tial hygroscopicity of 1, samples were stored for 4 weeks
Table 2. Effects of 1 and isomers on whole cell maxi-K current in
HEK 293 cells
Compda
10 nM
100 nM
1000 nM
(ꢂ)-5c
Not tested
127.8
105.0
131.9
185.3
164.4
169.6
247.4
205.3
(S)-(+)-1
(R)-(ꢀ)-6
aThe current amplitude in the presence of compound is presented as
the percent of pre-drug control current elicited at +60 mV. A
minimum of four cells was recorded at each drug concentration.
Scheme 2. Stereoselective synthesis of 1 and 6.

1026
P. Hewawasam et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1023–1026
Table 3. Pharmacokinetic parameters in plasma and brain following
intravenous administration of 1 to male Sprague–Dawley rats
characterized oxindole 1 as a brain penetrant maxi-K
opener with good solid state stability. A stereospecific
synthesis of 1 was developed employing chiral resolution.
Parameter
Braina
Plasma
Dose (mg/kg)
Plasma half-life (h)
Brain half-life (h)
Plasma AUC (mg h/mL)
Brain AUC (mg h/mL)
Brain AUC/plasma AUC (ratio)
Clearance (mL/min kg)
Volume of distribution (L/kg)
5.0
1.9
1.6
3.0
28.8
9.6
Nd
Nd
1.0
1.8
References and Notes
Nd
0.32
Nd
Nd
51.5
5.8
1. Williams, G. R.; Jiang, J. G.; Matchar, D. B.; Samsa, G. P.
Stroke 1999, 30, 2523.
2. Hille, B. In Ionic Channels of Excitable Membranes, 2nd
ed.; Sinaur; Sunderland, 1992.
3. Hinrichsen, R. D. In Calcium-Dependent Potassium Chan-
nels; R. G. Landes: Austin, TX, 1993.
aNd, not determined.
4. Latorre, R.; Oberhauser, A.; Labarca, P.; Alvarez, O.
Annu. Rev. Physiol. 1989, 51, 385.
5. Starrett, J. E., Jr.; Dworetzky, S. I.; Gribkoff, V. K. Curr.
Pharm. Des. 1996, 2, 413.
6. Gribkoff, V. K.; Starrett, J. E., Jr.; Dworetzky, S. I. Neu-
roscientist 2001, 7, 166.
in open glass vials at 25 ꢁC/87% RH and 50 ꢁC/75%
RH. Karl Fisher analyses indicated no significant
increase in water content.
7. Gribkoff, V. K.; Starrett, J. E., Jr.; Dworetzky, S. I.;
Hewawasam, P.; Boissard, C. G.; Cook, D. A.; Frantz, S. W.;
Heman, K.; Hibbard, J. R.; Huston, K.; Johnson, G.; Krish-
nan, B. S.; Kinney, G. G.; Lombardo, L. A.; Meanwell, N. A.;
Molinoff, P.; Myers, R. A.; Moon, S. L.; Ortiz, A.; Pajor, L.;
Pieschl, R. L.; Post-Munson, D. J.; Signor, L. J.; Srinivas, N.;
Taber, M. T.; Thalody, G.; Trojnacki, J. T.; Wiener, H.;
Yeleswaram, K.; Yeola, S. W. Nature Med. 2001, 7, 471.
8. Rikke, L. A.; Jespersen, T.; Christpheresen, P.; Strøbæk, D.;
Jensen, B. S.; Olesen, S.-P. Neuropharmacology 2001, 40, 888.
9. Hewawasam, P.; Meanwell, N. A.; Gribkoff, V. K. US.
Patent 5,602,169, 1997; Chem. Abstr. 1997, 126, 181369.
10. Gribkoff, V. K.; Lum-Ragan, J. T.; Boissard, C. G.; Post-
Munson, D. J.; Meanwell, N. A.; Starrett, J. E., Jr.;
Kozlowski, E. S.; Romine, J. L.; Trojnacki, J. T.; McKay,
M. C.; Zhong, J.; Dworetzky, S. I. Mol. Pharm. 1996, 50, 206.
11. Hewawasam, P.; Meanwell, N. A.; Gribkoff, V. K.;
Dworetkzy, S. I.; Boissard, C. G. Bioorg. Med. Chem. Lett.
1997, 7, 1255.
To characterize the pharmacokinetic disposition and
brain uptake of 1, two studies were conducted. To
determine brain uptake, groups of rats were adminis-
tered an intravenous bolus dose of 1 at 5 mg/kg and
plasma and whole brain samples were collected from
2–3 rats at each of 10 time points through 8 h after
dosing. Concentrations of 1 in plasma and whole blood
brain homogenate samples were determined by LC/MS
analysis following acetonitrile extraction. The area
under the curves of plasma or brain concentration ver-
sus time was used to assess brain uptake. To determine
pharmacokinetics, three rats were administered an
intravenous dose of 1 at 1 mg/kg and serial blood sam-
ples were withdrawn from an indwelling jugular vein
cannulae through 8 h after dosing. The results from
both studies are summarized in Table 3.
12. Hewawasam, P.; Erway, M.; Moon, S. L.; Knipe, J.;
Weiner, H.; Boissard, C. G.; Plst-Munson, D. J.; Gao, Q.;
Huang, S.; Gribkoff, V. K.; Meanwell, N. A. J. Med. Chem. In
press.
13. Pendri, Y. R.; Martinez, E. J.; Thottathil, J. K.; Hewasa-
sam, P. US Patent 5,892,106, 1998: Chem. Abstr. 1998, 129,
230636.
14. X-ray crystallographic data has been deposited at the
Cambridge Crystallographic Data Centre, University Chemi-
cal Laboratory, Lensfield Road, Cambridge CB2 1EW, UK;
Deposit number CCDC 169402.
The disposition of 1 in rats is characterized by a short
half-life, high clearance and a moderate-to-high volume
of distribution. Elimination half-lives were similar in
both plasma and brain, suggesting rapid distribution
between brain and the systemic circulation. The
observed brain/plasma AUC ratio (>9) indicates excellent
brain uptake of this compound.
In conclusion, we have developed generalized metho-
dology for the synthesis of 3-fluorooxindoles and have