Novel inhibitors of the gardos channel for the treatment of sickle cell disease

Article abstract of DOI:10.1021/jm070663s

Sickle cell disease (SCD) is a hereditary condition characterized by deformation of red blood cells (RBCs). This phenomenon is due to the presence of abnormal hemoglobin that polymerizes upon deoxygenation. This effect is exacerbated when dehydrated RBCs experience a loss of both water and potassium salts. One critical pathway for the regulation of potassium efflux from RBCs is the Gardos channel, a calcium-activated potassium channel. This paper describes the synthesis and biological evaluation of a series of potent inhibitors of the Gardos channel. The goal was to identify compounds that were potent and selective inhibitors of the channel but had improved pharmacokinetic properties compared to 1, Clotrimazole. Several triarylamides such as 10 and 21 were potent inhibitors of the Gardos channel (IC₅₀ of <10 nM) and active in a mouse model of SCD. Compound 21 (ICA-17043) was advanced into phase 3 clinical trials for SCD.

Full text of DOI:10.1021/jm070663s

976
J. Med. Chem. 2008, 51, 976–982
Novel Inhibitors of the Gardos Channel for the Treatment of Sickle Cell Disease
Grant A. McNaughton-Smith,† J. Ford Burns,† Jonathan W. Stocker,*,† Gregory C. Rigdon,† Christopher Creech,†
Susan Arrington,† Tara Shelton,† and Lucia de Franceschi‡
Icagen, Inc., Post Office Box 14487, Research Triangle Park, North Carolina 27709, and Department of Internal Medicine,
UniVersity of Verona, Italy
ReceiVed June 8, 2007
Sickle cell disease (SCD) is a hereditary condition characterized by deformation of red blood cells (RBCs).
This phenomenon is due to the presence of abnormal hemoglobin that polymerizes upon deoxygenation.
This effect is exacerbated when dehydrated RBCs experience a loss of both water and potassium salts. One
critical pathway for the regulation of potassium efflux from RBCs is the Gardos channel, a calcium-activated
potassium channel. This paper describes the synthesis and biological evaluation of a series of potent inhibitors
of the Gardos channel. The goal was to identify compounds that were potent and selective inhibitors of the
channel but had improved pharmacokinetic properties compared to 1, Clotrimazole. Several triarylamides
such as 10 and 21 were potent inhibitors of the Gardos channel (IC50 of <10 nM) and active in a mouse
model of SCD. Compound 21 (ICA-17043) was advanced into phase 3 clinical trials for SCD.
Introduction
Sickle cell disease (SCDa) is a debilitating, hereditary disease
caused by expression of a point mutation in the ꢀ-globin gene
of hemoglobin. This mutation leads to the production of
abnormal hemoglobin (HbS), which polymerizes when it is
deoxygenated, eventually leading to the formation of sickle-
shaped red blood cells (RBCs), which are involved in an
interruption of blood flow (vaso-occlusion) to various organs
and tissues. Individuals with SCD have episodes of painful vaso-
occlusive crises, acute chest syndrome, priapism, pulmonary
hypertension, stroke, and many other acute and chronic disorders
that ultimately lead to a decreased life expectancy.1,2
The rate of RBC sickling is dramatically affected by the
intracellular concentration of HbS.3,4 Studies have shown that
by preventing the dehydration of RBCs, and thus preventing
increases in the HbS concentration, RBC sickling can be
reduced.5 Because RBC dehydration is initiated by the loss of
intracellular KCl and KHCO3 followed by the osmotically driven
loss of water, inhibiting the transport of K+ from the RBCs
should reduce cell dehydration and, therefore, decrease the
formation of sickled cells.
Figure 1. Clotrimazole, 1.
Gardos channel; the second was to optimize the selectivity for
the target channel; and the third was to improve the pharma-
cokinetic properties. The initial sets of compounds were
derivatives of 1, in which the imidazole moiety was replaced
with other functional groups such as amides and reverse amides
(see Supporting Information). The compounds were prepared
using standard solution phase parallel synthesis techniques. In
the first set, four commercially available aromatic acids (tri-
phenylacetic acid, phenyl acetic acid, diphenylacetic acid, and
9-phenylflurenyl) and one isocyanate (triphenylisocyante) were
coupled with the 12 amines and 1 alcohol, resulting in a set of
65 compounds. In addition, a set of the corresponding inverse
amides/sulfonamides were generated in a similar fashion by
reacting triphenylmethylamine or flurenylamine with a set of
acid chlorides and one sulfonyl chloride.
Several additional compounds were made to further define
the structure–activity relationships. For example, compounds 4
and 7 (Scheme 1) were designed to investigate the role of the
amide and its spatial relationship to the three aryl rings. The
addition of ethanolamine to the activated acid of 2 gave
the amide 3, which was converted to the oxazolidine in high
yield using Burgess reagent.19 The constrained amide 7 was
obtained by condensing the isatin 6 with benzene using an
excess of triflic acid.20
The Gardos channel, a calcium activated K+ channel of
intermediate conductance,6–8 has been shown to be one of the
main routes for K+ loss in RBCs. Clotrimazole 1, a small
molecule inhibitor of the Gardos channel (Figure 1), was shown
to block RBC dehydration9–12 and was efficacious in a mouse
model of SCD (SAD-1 mouse model).13,14 The use of this
compound for the long-term SCD therapy is limited by several
factors including its inhibition of mammalian cytochrome P-450s
(particularly CYP3A4),15 its relatively short in vivo plasma half-
life,16 the toxicity experienced upon repetitive dosing,17 and
inhibition of several cardiac ion channels including hERG.18
Chemistry
The substituted triphenylmethane amides (Table 2) were
obtained as outlined in Scheme 2. Addition of phenylmagnesium
bromide to the commercially available benzophenones 8 af-
forded the substituted triphenylmethane alcohols, which were
transformed into the triphenylmethane chlorides using a 3–5-
fold excess of acetyl chloride. These key intermediates were
not purified, but directly converted to the cyano derivatives 9
using a slight excess of copper. Hydrolysis methods employing
a basic medium did not generate significant quantities of the
The chemistry plan had several key goals: the first was to
identify those structural features required for blocking the
* To whom correspondence should be addressed. Tel.: 919-941-5206.
Fax: 919-941-0813. E-mail: jstocker@icagen.com.
†
Icagen, Inc.
University of Verona.
‡
a Abbreviations: SCD, sickle cell disease; RBC, red blood cell; HbS,
abnormal hemoglobin.
10.1021/jm070663s CCC: $40.75
2008 American Chemical Society
Published on Web 01/31/2008

NoVel Inhibitors of the Gardos Channel
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 977
Scheme 1. Preparation of Amide Bioisosteres and Cyclic Amides Reagentsa
a Reagents and conditions: (a) HOBt, Et3N, (EtO)2POCl, 1 h, then H2NCH2CH2OH; (b) Burgess reagent, THF, 50 °C; (c) 2 mol equiv C6H6, triflic acid,
14 h.
Table 1. Inhibition of Ionomycin Stimulated 86Rb efflux (IC50 Values,
Gardos Channel) for Selected Triphenylmethane Derivatives
The monoaryl and diaryl amide derivatives that were prepared
using solution phase techniques were all inactive when tested
against the Gardos Channel (<50% inhibition at 10 µΜ as
measured by 86Rb+ effluxsdata not shown). However, those
compounds (2, 5, 15–19) resulting from the coupling of the
triphenylacetic acid with small amines and alcohols were active
(Figure 2, Table 1). The difference in activity between the acid
2 and the methyl ester 15 was striking and indicated that changes
in pKa and the ability of the carbonyl group to participate in
hydrogen bonding was a possible factor. The most significant
jump in potency was seen with compound 5, in which the
carboxylic acid was converted to the primary amide. This
compound was 7-fold more potent than 1. Introduction of
substituents on the amide nitrogen of 5 (compounds 16, 17, and
18) led to reduced activity. For example, 16 was 8-fold less
active than 5, the ethyl derivative 17 was even less active, while
the dimethylated amide 18 possessed only low micromolar
activity.
a
ID#
R1
Y
RBC’s IC50 (µM)
1
2
clotrimazole
0.36 ( 0.07
>10
H
H
H
H
H
H
Cl
OH
OMe
NH2
NHMe
NHEt
NMe2
NH2
15
5
16
17
18
19
0.33 ( 0.1
0.07 ( 0.004
0.55 ( 0.3
1.75 ( 0.8
3.15 ( 2.3
0.02 ( 0.01
a Number of measurements ) 4.
The Gardos channel inhibition data for the triphenyl-
methane derivatives indicated that small functional groups
having the ability to act as hydrogen bond acceptors were
preferred. Based upon this theory, the constrained cyclic
amide 7 and the amide bioisostere 4 were synthesized to
investigate the spatial role of the amide with the triphenyl
system and the impact of the amide tautomeric form. The
oxazolidine 4 (Scheme 1) possessed only weak inhibitory
activity (IC50 ) 4.3 ( 0.4 µM) perhaps due in part to the
increased steric bulk of the ring system, while the constrained
amide 7 was surprisingly devoid of inhibitory activity (<20%
inhibition at 10 µM).
To investigate the importance of the chlorine substituent
present in 1, the corresponding 2-chloro derivative of 5 was
prepared. Compound 19 was found to be 4-fold more potent
than 5 (IC50 of 15 nM vs 69 nM).
Prior to embarking upon the synthesis of additional analogs,
several of the more potent compounds (5, 16, and 19) were
tested against the hERG channel. In summary, none of the
compounds significantly inhibited this important cardiac ion
channel (<20% inhibition at 10 µM). Additional testing against
CYP3A4 showed that none of the compounds inhibited CYP3A4
(<20% inhibition at 10 µM).
Table 2. Summary of ADME Data for Compounds 10 and 21
cmpd RS9a Sol (µm) F% (rat) T1/2 (h) Tmax (h) Cmax (ng/ml)
10
21
40%
51%
5.8
3.8
35
51
2.5
1.0
2
4
250
400
a Percent remaining after incubation with S9.
desired amide. However, strongly acidic conditions at elevated
temperatures efficiently converted the nitriles into the amides
(10, 21–33).21
The 2,2′ and 3,3′ difluorinated benzophenones (14a and 14b)
were not commercially available and were prepared as shown
in Scheme 2. Addition of ethylformate to the appropriate
aryllithium, generated in situ via lithium-halogen exchange,22
afforded the intermediate alcohol 13a. Oxidation of 13a using
PCC gave the desired benzophenone 14a in good overall yield.
The 3-difluoro benzophenone analog 14b was prepared using
the same procedure.
Results and Discussion
The current study focused on identifying the structural
features of 1 responsible for the Gardos channel inhibitory
activity, while minimizing or eliminating CYP450 enzyme
inhibition and metabolic stability issues. The imidazole moiety
in 1 was believed to contribute to both the inhibition of the
CYP450 metabolizing enzymes and the compound’s relatively
short half-life. The short half-life was believed to result from
the known susceptibility of the C-N bond (the bond between
the triphenylmethane and imidazole group) to undergo hydro-
lysis.16
Having substantially improved both in vitro potency against
the cloned channel and ion channel selectivity compared to
1, we turned our attention to improving the in vivo stability
and, thus, bioavailability of the series. Compound 19 was
administered both intravenously and orally to rats (Figure
2). When given IV, 19 had a relatively short half-life (t1/2
15 min), while oral administration lead to very low plasma
)

978 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4
McNaughton-Smith et al.
Scheme 2. General Procedure for the Preparation of Triphenylmethane Amidesa
a Reagents and conditions for (a): (a) n-BuLi, THF, -78 °C, 1 h; (b) HCO2Et, 1 h, -78 to 0 °C; (c) PCC, DCM, 12 h, rt. Reagents and conditions for
(b): (a) fluoro-substituted phenylMgBr, THF, or t-BuOMe, rt, 4 h; (b) AcCl, DCM, rt, 12 h; (c) CuCN, 140 °C, 3 h; (d) 1:1, H2SO4/HOAc, 120 °C, 3 h.
Table 3. In Vitro Activity of Fluorinated Triphenylmethane Amides
Figure 2. Synthesis of aromatic amides related to 1.
IC50
ID# ring A ring B ring C RBCs (nM)a RS9b,c (%) Sold,e (µM)
concentrations. The bioavailability of 19 was calculated to
be <5%, presumably due to a rapid rate of elimination.
Although amide hydrolysis was a potential liability, the nature
and size of the neighboring triphenylmethane group was
considered to be a hindrance to this metabolic route.
However, the three unprotected phenyl rings were considered
potential sites for oxidative metabolism.
Initially, we examined the effect on metabolism of fluorine
substitution at the ortho and para positions of the phenyl rings.
Fluorine was chosen as one of the first substituents due to its
small size and strong electron-withdrawing capability. The 2,4′-
disubstituted and 4,4′-disubstituted compounds 10 and 21,
respectively, were prepared. Both compounds retained their
potency against the RBC Gardos channel as well as their
selectivity against a collection of ion channels. In addition, these
compounds did not have CYP3A4 activity at high concentrations
(<20% inhibition at 10 µM).
Compound 10 was found to be orally bioavailable (F ) 35%)
in a rat pharmacokinetic study with a T1/2 of 150 min. In
addition, compound 21 in which two of the para positions were
substituted with fluorine was more stable (T1/2 ) 66 min) than
19 and considerably more bioavailable in rats (F ) 51%). The
data is summarized in Table 2. In addition, the improved
pharmacokinetics were confirmed in mice.
Based upon the encouraging data obtained with 10 and 21, a
series of fluorinated analogues were synthesized (Table 3). All
of the fluorine containing amides had good inhibitory activity
versus the native Gardos channel in RBCs. However, there were
notable differences in aqueous solubility and susceptibility to
metabolism when incubated with rat S9 fractions (Table 3).
Although no clear trends were observed among mono-, di-, and
trisubstituted compounds, those containing a fluorine in the meta
position were generally found to be more prone to oxidative
metabolism.
10 2F
21 4F
22 4F
23 3F, 4F
24 2F, 4F
25 4F
26 2F
27 2F
28 3F
29 3F
30 2F
31 4F
32 3F
33 3F
1
4F
4F
4F
9
12
10
14
34
14
30
15
17
13
15
40
30
15
40
56
62
63
44
25
48
46
19
36
61
58
43
nd
nd
5.8
3.8
8.3
4.3
1.8
2.6
>31
6.8
4.1
3.4
10.6
2.9
1.2
nd
2F
4F
4F
2F
2F
4F
3F
4F
4F
3F
3F
3F
4F
0.36
nd
a Number of measurements ) 2. The error in the RBC measurement is
typically (5 nM. b Percent remaining after incubation with rat S9 fraction
for 2 h. c Number of measurements ) 1 or 2. The error in the S9 fraction
measurement is typically (15%. d Solubility was measured using the
Lipinski type method (ref 23). e Number of measurements ) 2. The error
in the solubility measurement is typically (0.1 µM.
disease.13,14 In this model, an inhibitor of the Gardos channel
should result in less RBC dehydration and would be expected
to decrease the mean corpuscular hemoglobin concentration
(MCHC) and increase the hematocrit (Hct) levels. Dosing of
SAD-1 mice with 10 (20 mg/kg) twice a day for 14 days via
oral gavage produced a reduction in the MCHC on day 7 (first
analysis day) as well as a significant increase in hematocrit. In
addition, the number of dense RBCs was reduced (Table 4).23
Most important was the fact that all of the changes in
hematological parameters were maintained throughout the course
of the experiment. Compound 21 was also tested in the model.
Due to its improved bioavailability and longer half-life, the
animals were dosed with 10 mg/kg twice a day for 21 days.
Hematological parameters were collected on days 11 and 21,
with results similar to those obtained with compound 10 (Table
4). Compound 21, ICA-17043, completed preclinical testing and
Compounds 10 and 21 remained as the prime candidates for
in vivo testing in the SAD-1 mouse model of sickle cell

NoVel Inhibitors of the Gardos Channel
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 979
Table 4. Hematological Data Obtained from Administration of 10 (20
mg/kg, bid) and 21 (10 mg/kg, bid) in the SAD-1 Mouse Modela
concentration vs time were prepared using SigmaPlot 2001 for
Windows (version 7, SPSS, Inc., Chicago, IL).
N-Methyl-2,2,2-triphenylacetamide (16). A mixture of HOBt
(281 mg, 2.1 mmol), Et3N (0.87 mL, 6.3 mmol), and (EtO)2POCl
(0.3 mL, 2.1 mmol) in dry THF (5 mL) was stirred at room
temperature for 20 min. To the reaction was added 2a (600 mg,
2.1 mmol), and the stirring continued for 1 h. Then a solution of
methylamine in methanol (2.0 M, 1.25 mL, 2.5 mmol) was added,
and the mixture was stirred for 1 h. To the resulting solution was
added EtOAc (10 mL) and brine (10 mL), and the organic layers
were separated, dried (Na2SO4), and concentrated. Purification by
column chromatography (SiO2, hexanes/EtOAc, 4:1) gave the
measure
cmpd 10
cmpd 21
Gardos channel activity
(mmol Rb/L of cells)
day 0
day 7
day 11
day 14
day 21
day 0
11 ( 1
12 ( 1
-
3 ( 1
-
4 ( 1
-
2 ( 0.2
-
1.5 ( 0.5
MCHC (g/dL)
Hct (%)
40 ( 2
34 ( 1
day 7
34 ( 1
-
day 11
day 14
day 21
day 0
-
29.5 ( 1
27 ( 0.4
-
-
1
30.3 ( 1.5
43.5 ( 0.7
-
desired product 16 as a white solid (0.587 g, 94%): H NMR δ
(CDCl3) 7.36–7.21 (15H, m), 5.82 (1H, brs), and 2.84 (3H, d, J)
6 Hz); Anal. (C22H21NO) C, H, N.
44.5 ( 0.3
day 7
44.5 ( 0.3
day 11
day 14
day 21
-
50.9 ( 2.2
The following compounds were prepared as described for 16.
N-Ethyl-2,2,2-triphenylacetamide (17). Compound 17 was
47 ( 0.6
-
-
49.9 ( 4.1
1
obtained from ethylamine, as a white solid: H NMR δ (DMSO-
a The symbol “-” indicates no data were collected on these days.
d6) 7.31–7.17 (15H, m), 3.54 (2H, q, J) 7 Hz), and 1.32 (3H, t,
J) 7 Hz); Anal. (C22H21NO) C, H, N.
has advanced into clinical trials as a potential treatment for
individuals with SCD.24
N,N-Dimethyl-2,2,2-triphenylacetamide (18). Compound 18
1
was obtained from dimethylamine, as a white solid: H NMR δ
(DMSO-d6) 7.31–7.17 (15H, m), 3.05 (3H, s), and 2.35 (3H, s).
2,2,2-Triphenylacetamide (5). Compound 5 was obtained from
ammonia in methanol, as a white solid: H NMR δ (DMSO-d6)
7.49 (1H, brs), 7.30–7.19 (15H, m), and 6.58 (1H, brs).
Triphenylacetic Acid Methyl Ester (15). Compound 15 was
obtained from methanol, as a white solid: 1H NMR δ (DMSO-d6)
7.32–7.26 (10H, m), 7.17–7.14 (5H, m), and 3.80 (3H, s); Anal.
(C21H18O2) C, H, N.
Conclusions
1
A novel series of Gardos channel inhibitors with potent
activity, selectivity, and excellent pharmacokinetic properties
has been identified. Two of these compounds, 10 and 21, were
shown to improve hemodynamic parameters in an animal model
of sickle cell disease.
N-(2-Hydroxy-ethyl)-2,2,2-triphenyl-acetamide (3). A mixture
of HOBt (135 mg, 1 mmol), Et3N (0.42 mL, 3 mmol), and
(EtO)2POCl (0.145 mL, 1 mmol) in dry THF (5 mL) was stirred at
room temperature for 20 min, after which triphenylacetic acid (289
mg, 1 mmol) was added and the stirring continued for an additional
1 h. To the mixture was added ethanolamine (0.07 mL, 1.1 mmol),
and after 1 h, EtOAc (10 mL) and brine (10 mL) were added. The
organic layer was separated, dried (Na2SO4), and concentrated.
Purification by column chromatography (SiO2, hexanes/EtOAc; 1:1)
gave 3 as a white solid (299 mg, 85%): 1H NMR δ (CDCl3)
7.34–7.21 (15H, m), 6.25 (1H, brs), 3.69 (2H, t, J ) 5.0 Hz), 3.49
(2H, q, J ) 5.5 Hz), and 2.05 (1H, brs).
2-Trityl-4,5-dihydrooxazole (4). To a solution of 3 (299 mg,
0.85 mmol) in THF (5 mL) was added (methoxycarbonylsulfa-
moyl)triethylammonium hydroxide (Burgess reagent; 300 mg, 1
mmol). The resulting solution was heated at 50 °C for 1.5 h, cooled
to room temperature, and EtOAc (20 mL) was added. The organic
layer was washed with aqueous 1 N HCl (20 mL), saturated aqueous
solution of NaHCO3 (20 mL), and brine (20 mL) and dried
(Na2SO4). The solution was concentrated in vacuo and the crude
product was purified bypassage thorough a short silica plug
(hexanes/EtOAc, 2:1) to afford the desired product as a white solid
(248 mg, 93%): 1H NMR δ (CDCl3) 7.31–7.26 (10H, m), 7.24–7.16
(5H, m), 4.34 (2H, t, J ) 9.5 Hz), and 3.98 (2H, t, J ) 9.5 Hz);
Anal. (C22H19NO) C, H, N.
Experimental Section
Melting points were determined with an Electrothermal IA9000
melting point apparatus and are uncorrected. Unless otherwise noted,
starting materials were purchased from Aldrich Chemical Co.
(Milwaukee, WI). The 2,2′- and 3,3′-difluorbenzophenones were
prepared as described below. All other fluorinated benzophenones
were commercially available. Reactions were run at room temper-
ature under an atmosphere of argon, unless otherwise indicated.
Organic phases from aqueous extractions were dried over Na2SO4,
filtered, and concentrated. Reactions were monitored by TLC
analysis on silica gel 60 F254, with detection using UV light at 254
nM. Flash chromatography was performed on silica gel (particle
size 32–63), and the column output was monitored with TLC.
Elemental analyses, performed by Atlantic Microlabs, Atlanta, GA,
were within (0.4% of theoretical values. Nuclear magnetic
resonance (NMR) spectra were recorded in either CDCl3 or DMSO-
d6 solutions with a Varian (Gemini 2000) spectrometer using TMS
as an internal standard. Mass spectral data were obtained on a PE
SCIEX API150EX mass spectrometer. The yields quoted in this
paper are isolated yields. The amount of DMSO in the solutions
tested in vitro was not greater than 1% of the total volume. Cerep,
Inc., Redmond, WA, performed the inhibitions of CYP3A4
enzymes.
(2-Fluorophenyl)-(4-fluorophenyl)phenylacetamide (10). To
a stirred solution of 2,4′-difluorobenzophenone (1.09 g, 5.0 mmol)
in t-butylmethyl ether (12 mL) at room temperature was added
dropwise phenylmagnesium bromide (1.83 mL, 5.5 mmol). After
the addition was complete, the reaction was heated at reflux for
3 h, cooled to room temperature, and then poured onto cold 1.0 M
HCl (aq; 0 °C, 20 mL). The organic layer was extracted with EtOAc
(3 × 10 mL) and dried (Na2SO4). Concentration under reduced
pressure gave the desired product (2-fluorophenyl)-(4-fluorophenyl)-
phenylmethanol as a pale brown oil, which was used in the next
reaction without any further purification.
To a 20% solution of acetyl chloride in dichloromethane (10
mL) at room temperature was added the crude (2-fluorophenyl)-
(4-fluorophenyl)phenylmethanol (1.47 g, 5.0 mmol) from above.
The resulting solution was stirred for 12 h, after which the solvent
was removed by evaporation and toluene (2 × 20 mL) was
All samples for the solubility, microsome, and in vivo studies
were analyzed using online turbulent-flow “extraction” (Cohesive
2300 HTLC system including a CTC autosampler, an Agilent 1100
Binary LC pump, and an Agilent 1100 Quaternary LC pump). Mass
spectra were obtained on a Micromass Quattro Ultima spectrometer.
All plasma samples with quantifiable concentrations were used
in the pharmacokinetic calculations. Samples measuring below the
lower limit of quantitation (LLOQ ) 1 ng/mL) were considered as
zeros. Noncompartmental analysis of plasma concentration data was
accomplished using WinNonlin Professional Software (version
4.0.1, Pharsight Corporation, Mountain View, CA). The maximum
plasma concentration (Cmax) and time of maximum plasma con-
centration (Tmax) were determined by observation of the data. The
AUC0f∞ was calculated using the linear-log trapezoidal rule. Mean,
standard deviation, minimum, median, and maximum values were
calculated using Excel2000 (Microsoft Corporation). Plots of plasma

980 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4
McNaughton-Smith et al.
sequentially added to the residue and evaporated to afford crude
2-fluorophenyl-(4-fluorophenyl) phenylchloromethane.
The following compounds were prepared using the procedure
described for compound 21.
2-(2-Chlorophenyl)-2,2-diphenylethanamide (19). A mixture
of 1 (0.5 g, 1.4 mmol) in 1.0 M HCl (2 mL) was refluxed for 2 h
and then cooled. The residue was extracted with EtOAc (2 × 10
mL), and the organic layer was washed with brine (2 × 10 mL),
dried (Na2SO4), and concentrated. The solid was recrystallized from
hexanes to provide (2-chlorophenyl)diphenylmethanol (0.4 g, 93%)).
This material was used to provide 19: 1H NMR δ (DMSO-d6)
7.49–7.52 (2H, m), 7.22–7.41 (9H, m), 6.87–6.90 (1H, m), 6.68
(1H, bs); Anal. (C20H16ClNO) C, H, N.
To this crude material was added copper cyanide (0.50 g, 5.5
mmol), and the mixture was heated at 140 °C for 2.5 h. The reaction
was cooled to approximately 110 °C, toluene (30 mL) was added,
the slurry was stirred vigorously for 10 min and filtered, and the
solvent was removed under reduced pressure. To the resulting crude
material, hot hexane (30 mL) was added and the mixture was stirred
vigorously for 30 min. Filtration and trituration with more hexane
(2 × 30 mL) gave the desired cyano intermediate 9 as a white
solid.
Bis(4-fluorophenyl)-2-fluorophenylacetamide (22). Compound
22 was prepared using 2,4′-difluorobenzophenone and 4-fluorophe-
nylmagnesium bromide. The product was a white crystalline solid
To the solid (9, 1.48 g, 5.0 mmol) at room temperature was added
a solution of concentrated sulfuric acid (10 mL) and glacial acetic
acid (10 mL). The resulting orange solution was stirred and heated
at 130 °C for 3 h, cooled to 0 °C, and was neutralized by the
dropwise addition of ammonium hydroxide. Water (30 mL) was
added, and the aqueous layer was extracted with chloroform (3 ×
30 mL). The organic fractions were combined and washed
sequentially with water (2 × 10 mL) and brine (20 mL). The organic
phase was dried (Na2SO4) and concentrated under reduced pressure
to provide a light brown oil to which hexanes (30 mL) were added
to the initiate precipitation. The precipitate was filtered and triturated
with hot hexane (2 × 30 mL). Crystallization from hexane/
dichloromethane gave the desired product 10 as a white crystalline
solid (0.45 g, 1.4 mmol, 28%, 4 steps): 1H NMR δ (CDCl3)
7.39–7.26 (8H, m), 7.15–6.90 (5H, m), 5.83 (1H, brs), 5.72 (1H,
brs); 19F NMR δ (CDCl3) -103.4 (1F, s), -115.8 (1F, s); mp
180–181 °C; Anal. (C20H15F2NO) C, H, N.
Bis(4-fluorophenyl)phenylacetamide (21). To a stirring solution
of 4,4′-difluorobenzophenone (20 g, 0.092 mol) in t-butylmethyl
ether (150 mL) at room temperature was added phenylmagnesium
bromide (100 mL, 0.1 mol) dropwise. After the addition was
complete, the reaction was heated at reflux for 3 h, cooled to room
temperature, and poured into ice cold aqueous 1.0 M HCl (100
mL). The organic layer was extracted with EtOAc (2 × 50 mL),
dried (Na2SO4), and concentrated under reduced pressure to provide
bis(4-fluorophenyl)phenylmethanol as a pale brown oil. The material
was dried in vacuo for 2 h and was used in the next reaction without
any further purification.
1
(4.98 g, 66% over 4 steps): H NMR δ (CDCl3) 7.41–7.34 (1H,
m), 7.29–7.23 (4H, m), 7.16 (1H, ddd, J ) 18.1, 8.1 and 1.2 Hz),
7.15 (1H, d, J ) 7.7 Hz), 7.05–6.97 (4H, m), 6.93–6.87 (1H, dt, J
) 8.0 and 1.4 Hz), 5.90 (1H, brs), 5.74 (1H, brs); 19F NMR δ
(CDCl3) -103.3 (1F, s), -115.5 (2F, s); m.p 168–169 °C; Anal.
(C20H15F2NO) C, H, N.
2-(3,4-Difluorophenyl)-2,2-diphenylacetamide (23). Compound
23 was prepared using 3,4-difluorobenzophenone and phenylmag-
nesium bromide. The product is a white crystalline solid (160 mg,
11% over 4 steps): 1H NMR δ (CDCl3) 7.37–7.31 (6H, m),
7.28–7.20 (5H, m), 7.12–7.04 (2H, m), 5.90 (1H, brs), 5.74 (1H,
brs); 19F NMR δ (CDCl3) -137.8 to -137.9 (1F, m), -140.3 to
-140.4 (1F, m); mp 174–175 °C; Anal. (C20H15F2NO) C, H, N.
2-(2,4-Difluorophenyl)-2,2-diphenylacetamide (24). Compound
24 was prepared using 2,4-difluorobenzophenone and phenylmag-
nesium bromide. The product is a white crystalline solid (468 mg,
32% over 4 steps): 1H NMR δ (CDCl3) 7.37–7.28 (10H, m),
6.95–6.83 (2H, m), 6.81–6.75 (1H, m), 5.92 (1H, brs), 5.80 (H,
brs); 19F NMR δ (CDCl3) -99.1 (1F, dd, J ) 19.2 and 8.5 Hz),
-111.6 (1F, m); mp 187–188 °C; Anal. (C20H15F2NO) C, H, N.
2,2,2-Tris-(4-fluorophenyl)-acetamide (25). Compound 25 was
prepared using 4,4′-difluorobenzophenone and 4-fluorophenyl-
magnesium bromide. The product was crystallized from ethanol/
water to afford a white crystalline solid (168 mg, 11% over 4
steps): 1H NMR δ (CDCl3) 7.26–7.19 (6H, dd, J ) 9.0 and 5.4
Hz), 7.20–7.01 (6H, t, J )8.7 Hz), 5.83 (1H, brs), 5.69 (1H, brs);
19F NMR δ (CDCl3) -115.3 (F, m); mp 180–181 °C; Anal.
(C20H14F3NO·0.25H2O) C, H, N.
Bis(4-fluorophenyl)phenylmethanol (0.092 mol) was added to a
20% solution of acetyl chloride in dichloromethane (50 mL) at room
temperature. The resulting purple solution was stirred for 12 h after
which time the solvent was removed by evaporation. Toluene (100
mL) was added to the residue and then evaporated to afford crude
bis(4-fluorophenyl)phenylchloromethane, which was used without
further purification.
2-Fluorophenyl-2,2-diphenylacetamide (30). Compound 30
was prepared using 2-fluorobenzophenone and phenylmagnesium
bromide. The product is a white solid (411 mg, 27% over 4
1
steps): H NMR δ (CDCl3) 7.37–7.28 (6H, m), 7.15–7.05 (2H,
m), 6.93 (1H, dt, J ) 8.0 and 2.0 Hz), 5.90 (1H, brs), 5.68 (1H,
brs); 19F NMR δ (CDCl3) -103.4 (F, m); mp 210 °C; Anal.
(C20H16FNO) C, H, N.
4-Fluorophenyl-2,2-diphenylacetamide (31). Compound 31 was
prepared using 4-fluorobenzophenone and phenylmagnesium bro-
mide. The product is a white solid (367 mg, 24% over 4 steps): 1H
NMR δ (CDCl3) 7.37–7.24 (12H, m), 6.97 (2H, t, J)8.5 Hz), 5.83
(1H, brs), 5.75 (1H, brs); 19F NMR δ (CDCl3) -116.2 (F, m); mp
193–194 °C; Anal. (C20H16FNO) C, H, N.
Copper cyanide (8.24 g, 0.11 mol) was added to the crude
residue, and the mixture was heated at 140 °C for 3 h. The reaction
was cooled to 100 °C and toluene (100 mL) was added. The
resulting slurry was stirred vigorously for 10 min, cooled to room
temperature, and filtered through a short plug of silica, and the
solvent was removed under reduced pressure to afford a brown
solid. Hot hexane (100 mL) was added to the powdered crude
material and the mixture was stirred vigorously for 4 h. Filtration
and trituration with additional hexane gave the desired bis(4-
fluorophenyl)phenylacetonitrile as a white solid (18.9 g, 67%).
A solution of concentrated sulfuric acid (50 mL) and glacial
acetic acid (50 mL) was added to bis(4-fluorophenyl)phenylaceto-
nitrile (18.9 g, 0.06 mol) at room temperature. The resulting orange
solution was stirred and heated at 130 °C for 3 h. The reaction was
cooled to 0 °C, poured into ice–water (150 mL), and neutralized
with ammonium hydroxide. The organics were extracted with
chloroform (3 × 100 mL), combined, and washed with brine (2 ×
50 mL). The organics were dried (Na2SO4) and concentrated under
reduced pressure to afford a yellow-orange solid. The solid was
stirred with hot hexane (100 mL) for 30 min and filtered.
Crystallization from dichloromethane/hexane gave 21 as a white
2,2-Bis-(3-fluorophenyl)-2-phenylacetamide (32). Compound
32 was prepared using 3,3′-difluorobenzophenone and phenylmag-
nesium bromide. The product is a white solid (423 mg, 23% over
1
4 steps): H NMR δ (CDCl3) 7.38–7.22 (7H, m), 7.09–6.96 (6H,
m), 5.83 (1H, brs), 5.77 (1H, brs); 19F NMR δ (CDCl3) -112.6
(F, dd, J ) 17.1 and 6.4 Hz); m.p 195–196 °C; Anal. (C20H15F2NO)
C, H, N.
2,2-Bis-(3-fluorophenyl)-2-(4-fluorophenyl)-acetamide (33). Com-
pound 33 was prepared using 3,3′-difluorobenzophenone and
4-fluorophenylmagnesium bromide. The product is a brown solid
(23 mg, 3% over 4 steps): 1H NMR δ (DMSO-d6) 7.60 (1H, brs),
7.34 (1H, q, J ) 8.0 Hz), 7.23–7.05 (9H, m), 7.01–6.96 (2H, m),
6.87 (1H, brs); 19F NMR δ (CDCl3) -112.8 (1F, m), -115.7 (2F,
m); Anal. (C20H14F3NO) C, H, N.
1
crystalline solid (16.9 g, 0.052 mol, 87%; 58% over 4 steps): H
NMR δ (CDCl3) 7.37–7.20 (9H, m), 7.04–6.91 (4H, m), 5.81 (1H,
brs), 5.71 (1H, brs); 19F NMR δ (CDCl3) -115.7 (F, m); mp
180–181 °C; Anal. (C20H15F2NO) C, H, N.
Preparation of 3-Fluorophenyllithium. To a stirring solution
of bromo-3-fluorobenzene (1.75 g, 10 mmol) in THF (25 mL) at

NoVel Inhibitors of the Gardos Channel
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 981
-78 °C was added n-butyllithium (4 mL, 10 mmol) dropwise. After
30 min, the solution was ready for use in the next steps.
Solubility and Rat S9-Metabolism Studies. Intrinsic microso-
mal clearance (Clint) was determined using the S9 fraction of liver
Biosciences). Compounds were incubated at 37 °C in a shaking
water bath for 15 and 30 min. The concentration of each compound
was determined by LC/MS/MS after reaction quenching with
acetonitrile (Table 2).
The solubility of compounds was determined with a modified
shake-flask method. Typically, 1 mg of compound was shaken
overnight in 1 mL of phosphate-buffered saline (pH 6.8). The
supernatant was removed and analyzed for compound by LC/
MS/MS.
In Vivo Pharmacokinetic Studies. To determine the plasma
kinetics of compounds 10, 19, and 21, the compounds were
administered either by intravenous (IV) injection (tail vein) or by
oral gavage (PO) to a set of 18 groups (n ) 3 per time point) of
male Sprague–Dawley rats weighing 251–299 g. The compounds
were administered on Day 1 at a dose level of 1 mg/Kg (IV) or 10
mg/Kg (PO), with dose volumes of 1 mL/Kg (IV), and 2 mL/Kg
(PO), respectively. A solution of each compound in a PEG-400/
cremophor-EL (both from Sigma Chemical Company, St. Louis,
Missouri) vehicle was used for the IV arm, while a 0.5% aqueous
solution of methylcellulose (from Sigma Chemical Company, St.
Louis, Missouri) was used for the oral arm of the study. Blood
samples were obtained at 0.08, 0.25, 0.5, 1, 2, 4, 6, 8, 12, and 24 h
post IV administration and at 0.5, 1, 2, 4, 6, 8, 12, and 24 h post
PO administration. Quantification of compound concentrations in
plasma was accomplished by LC-MS (Figure 2).
2,2-Bis-(4-fluorophenyl)-2-(3-fluorophenyl)-acetamide (28). Com-
pound 28 was prepared using 4,4′-difluorobenzophenone and
3-fluorophenyllithium. The product is a white solid (422 mg, 39%
1
over 4 steps): H NMR δ (CDCl3) 7.34–7.20 (6H, m), 7.06–6.97
(6H, m), 5.90 (1H, brs), 5.71 (1H, brs); 19F NMR δ (CDCl3) -112.2
(1F, dd, J ) 17.1 and 7.4 Hz), -115.1 to -115.2 (2F, m); mp
165–166 °C; Anal. (C20H14F3NO) C, H, N.
2,2,2-Tris-(3-fluorophenyl)-acetamide (29). Compound 29 was
prepared using 3,3′-difluorobenzophenone and 3-fluorophenyl-
lithium. The product is a white solid (376 mg, 11% over 4 steps):
1H NMR δ (CDCl3) 7.35–7.21 (3H, m), 7.06–6.97 (9H, m),
7.17–7.05 (4H, m), 5.96 (1H, brs), 5.76 (1H, brs); 19F NMR δ
(CDCl3) -112.2 (F, dd, J ) 17.1 and 8.5 Hz); mp 186–188 °C;
Anal. (C20H14F3NO) C, H, N.
2,2′-Difluorobenzophenone (14). To a stirred solution of bromo-
2-fluorobenzene (3.5 g, 20 mmol) in THF (50 mL) at –78 °C was
added n-butyllithium (2.5M; 8 mL, 20 mmol) dropwise. Then after
25 min ethylformate (0.74 g, 10 mmol) was added and after 1 h
the reaction was allowed to warm to 0 °C and satd NH4Cl (aq, 30
mL) was added. To the mixture was added diethylether (60 mL),
and the organic layer was washed with brine (20 mL), dried
(Na2SO4), and concentrated under reduced pressure. The residue
was purified by column chromatography (SiO2, 100% hexane to
100% CH2Cl2) to afford bis-(2-fluorophenyl)methanol as a white
solid (3.2 g, 73%).
Studies in Transgenic Sickle (SAD-1) Mice. Transgenic SAD-1
females and males between 3 and 6 months of age, weighing 25-30
g, were used for this study. The SAD-1 mice were divided into
two groups, and either vehicle or compound (10, 20 mg/kg, bid, or
21, 10/mg/kg, bid) was administered by oral gavage. Hematological
parameters were evaluated at baseline and days 7 and 14 for
compound 10, and baseline and days 11 and 21 for compound 21.
Blood sampling and vehicle administration have previously been
shown not to affect the blood parameters measured in this study
(Table 3).13
Hematological Data from SAD-1 Mice. Blood samples were
obtained at specific intervals and were used for 86Rb+ influx
measurements, determination of red blood cell (RBC) phthalate
density distribution curves, cell morphology studies, determination
of RBC cation content, and other hematological parameters. The
hemoglobin (Hb) concentration was determined by the spectroscopic
measurement of the cyanmet derivative. The hematocrit (Hct) was
determined by centrifugation in a micro-Hct centrifuge. The cells
were washed three times with PBS (330 mOsm) at 25 °C in 2 mL
tubes. Density distribution curves and median RBC densities (D50)
were obtained according to Danon and Marikovsky using phthalate
esters in microhematocrit tubes.26
A solution of bis-(2-fluorophenyl)methanol (1.09 g, 5 mmol) in
dichloromethane (3 mL) was added to a slurry of PCC (3.2 g, 14
mmol) in dichloromethane (3 mL) at room temperature and stirred
vigorously overnight. The slurry was filtered through a short
silica\celite plug to remove excess PCC. The eluent was concen-
trated and purified by column chromatography (SiO2, hexanes/
dichloromethane; 3:1). The product 14 was a white solid (0.93 g,
87%). Analysis agreed with reported data.25
2,2-Bis-(2-fluorophenyl)-2-phenylacetamide (26). Compound
26 was prepared using 2,2′-difluorobenzophenone and phenylmag-
nesium bromide. The product is a white solid (236 mg, 17% over
1
4 steps): H NMR δ (CDCl3) 7.39–7.27 (9H, m), 7.17–7.03 (4H,
m), 5.90 (1H, brs), 5.85 (1H, brs); 19F NMR δ (CDCl3) -102.9
(F, m); mp 166–167 °C; Anal. (C20H15F2NO) C, H, N.
2,2-Bis-(2-fluorophenyl)-2-(4-fluorophenyl)-acetamide (27). Com-
pound 27 was prepared using 2,2′-difluorobenzophenone and
4-fluorophenylmagnesium bromide. The product is a white solid
1
(147 mg, 14% over 4 steps): H NMR δ (CDCl3) 7.41–7.34 (2H,
m), 7.29–7.23 (4H, m), 7.17–7.05 (4H, m), 6.99 (2H, t, J ) 8.7
Hz), 5.78 (2H, brs); 19F NMR δ (CDCl3) -103.0 (2F, m), -115.9
(1F, m); mp l87–188 °C; Anal. (C20H14F3NO) C, H, N.
Washed Red Blood Cell Assay (Gardos Channel Inhibition).
Heparinized human blood was obtained from Biological Specialty
Corp. The whole blood was initially diluted 1:1 with Modified Flux
Buffer (MFB). MFB consists of 140 mM NaCl, 5 mM KCl, 10
mM Tris, 0.1 mM EGTA (pH ) 7.4). The blood was centrifuged
at 1000 rpm and the pellet consisting of predominantly red blood
cells (RBCs) was washed three times with MFB. The cells were
then loaded with 86Rb by incubating the washed cells with 86Rb at
a final concentration of 5 µCi/mL for at least 3 h at 37 °C. After
loading with 86Rb, the RBCs were washed three times with chilled
MFB. The cells were then incubated for 10 min with test compound
at concentrations that ranged between 1 and 10000 nM. Efflux of
86Rb was initiated by raising intracellular calcium levels in the
RBCs by addition of CaCl2 and A23187 to final concentrations of
2 mM and 5 µM, respectively. After 10 min, the RBCs were pelleted
in a microfuge and the supernatant was removed and counted in a
Wallac MicroBeta liquid scintillation counter. Percent inhibition
of efflux and IC50 values through the Gardos channel were then
calculated. The above protocol is a modification of the protocol
for measurement of Gardos channel inhibition in RBCs reported
previously.11 IC50 values were calculated using the Origin software
logistic function (Microcal Software, Northampton, MA).
Acknowledgment. We would like to thank Dr. Carlo
Brugnara for insightful discussion regarding SCD and the
SAD-1 mouse model.
Supporting Information Available: Supplemental analytical and
pharmacokinetic data and experimental procedures. This material
is available free of charge via the Internet at http://pubs.acs.org.
References
(1) Brittenham, G. M.; Schechter, A. N.; Noguchi, C. T. Hemoglobin S
polymerization: Primary determinant of hemolytic and clinical severity
of the sickling syndromes. Blood 1985, 65, 183–189.
(2) Fabry, M. E.; Costantini, F.; Pachnis, A.; Suzuka, S. M.; Bank, N.;
Aynedjian, H. P.; Factor, S. M.; Nagel, R. L. High expression of
human ꢀS- and R-globins in transgenic mice: Erythrocyte abnormali-
ties, organ damage, and the effect of hypoxia. Proc. Natl. Acad. Sci.
U.S.A. 1992, 89, 12155–12159.
(3) Eaton, W. A.; Hofrichter, J. Hemoglobin S gelation and cell sickle
cell disease. Blood 1987, 70, 1245–1266.
(4) Serjeant, G. R.; Serjeant, B. E.; Milner, P. F. The irreversibly sickled
cell: A determinant of hemolysis in sickle cell anemia. Br. J. Hamaetol.
1969, 17, 527–533.

982 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4
McNaughton-Smith et al.
(5) Glader, B. E.; Nathan, D. G. Cation permeability alterations during
sickling: relationship to cation composition and cellular hydration of
irreversibly sickled cells. Blood 1978, 51, 983–989.
(6) Gardos, G. The function of calcium in the potassium permeability of
human erythrocytes. Biochim. Biophys. Acta 1958, 30, 653–654.
(7) Ishii, T. M.; Siliva, C.; Hirschberg, B.; Bond, C. T.; Adelman, J. P.;
Maylie, J. A human intermediate conductance calcium-activated
potassium channel. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 11651–
11656.
required for inhibitory activity. J. Pharm. Exp. Ther. 1995, 273, 266–
272.
(17) Alvarez, J.; Montero, M.; Garica-Sancho, J. High affinity inhibition
of Ca2+-dependent K+ channels by cytochrome P-450 inhibitors.
J. Biol. Chem. 1992, 267, 11789–11793.
(18) Ashcroft, F. M. Ion Channels and Disease; Academic Press: New
York, 2000; pp 117–123.
(19) Burgess, E. M.; Penton, H. R.; Taylor, E. A. Thermal reactions of
alkyl N-carbomethoxysulfamate esters. J. Org. Chem. 1973, 38, 26–
31.
(20) Klumpp, D. A.; Yeung, K. Y.; Prakash, G. K. S.; Olah, G. Preparation
of 3,3-diaryloxindoles by superacid-induced condensations of isatins
and aromatics with a combinatorial approach. J. Org. Chem. 1998,
63, 4481–4484.
(21) Herold, F.; Wolska, I.; Helbin, E.; Krol, M.; Kleps., J. Synthesis and
structure of novel 4-arylhexahydro-1H,3H-pyrido[1,2-c]pyrimidine
derivatives. J. Heterocycl. Chem. 1999, 36, 389–396.
(22) Franco, M. L. T. M. B.; Herold, B. J.; Evans, J. C.; Rowlands, C. C.
Substituent effects on electron spin distribution and conformation of
radical ions obtained from 9-diphenylmethylenefluorenes. J. Chem.
Soc., Perkins Trans. 2 1988, 443–449.
(23) Stocker, J.; de Franceshi, L.; McNaughton-Smith, G. A.; Brugnara,
C. A novel Gardos channel inhibitor, ICA-17043, prevents red blood
cell dehydration in vitro and in a mouse model (SAD) of sickle cell
disease. Blood 2000, 96, 486A.
(24) Stocker, J.; De Franceschi, L.; McNaughton-Smith, G.; Corrocher, R.;
Beuzard, Y.; Brugnara, C. ICA-17043, a novel Gardos channel blocker,
prevents sickled red blood cell dehydration in vitro and in vivo in
SAD mice. Blood 2003, 101, 2412–2418.
(8) Grygorczyk, R.; Schwarz, W.; Passow, H. Ca2+-Activated K+ channels
in human red blood cell. Biophys. J. 1984, 45, 693–698.
(9) McGoron, A. J.; Joiner, C. H.; Palascak, M. B.; Claussen, W. J.;
Franco, R. S. Dehydration of mature and immature sickle red blood
cells during fast oxygenation/deoxygenation cycles: role of K-Cl
cotransport and extracellular calcium. Blood 2000, 95, 2164–2168.
(10) Dunn, P. M. The action of blocking agents applied to the inner face
of Ca2+-activated K+ channels from human erythrocytes. J. Membr.
Biol. 1998, 165, 133–43.
(11) Brugnara, C.; de Franceschi, L.; Alper, S. L. Inhibition of Ca2+
-
dependent K+ transport in sickle cell erythrocytes by clotrimazole and
other imidazole derivatives. J. Clin. InVest. 1993, 92, 520–526.
(12) Coupry, I.; Armsby, C. C.; Alper, S. L.; Brugnara, C.; Parini, A.
Clotrimazole and Efaroxan inhibit red cell Gardos channel indepen-
dently of imidazoline I1 and I2 binding sites. Eur. J. Pharmacol. 1996,
295, 109–112.
(13) de Franceschi, L.; Beuzard, Y.; Brugnara, C. Sulfydryl oxidation and
activation of red cell K+Cl- cotransport in the transgenic SAD mouse.
Am. J. Physiol. 1995, 269, C899–906.
(14) de Franceschi, L.; Saadane, N.; Trudel, M.; Alper, S. L. Inhibition of
Ca2+ activated K+ transport and reverses erythrocyte dehydration in
transgenic SAD mice. A model for therapy of sickle cell disease.
J. Clin. InVest. 1994, 93, 1670–1676.
(15) Fowler, S. M.; Riley, R. J.; Pritchard, M. P.; Sutcliffe, M. J.; Friedberg,
T.; Wolf, R. C. Amino acid 305 determines catalytic center acces-
sibility in CYP3A4. Biochemistry 2000, 39, 4406–4414.
(25) Riechers, H.; Albrecht, H.-P.; Amberg, W.; Baumann, E; Bernard,
H.; Böhm, H.-J.; Klinge, D.; Kling, A.; Muller, S.; Raschack, M.;
Unger, L.; Walker, N.; Wernet, W. J. Med. Chem. 1996, 39, 2123–
2128.
(26) Danon, D.; Marikovsky, Y. Determination of density distribution of
red blood population. J. Lab. Clin. Med. 1964, 64, 668–674.
(16) Brugnara, C.; Armsby, C. C.; Sakamoto, M.; Rifai, N.; Alper, S.; Platt,
O. Oral administration of clotrimazole and blockade of human
erythroctes Ca+2-activated K+ channel: The imidazole ring is not
JM070663S