Identification of potent and selective oxytocin antagonists. Part 2: Further investigation of benzofuran derivatives

Article abstract of DOI:10.1016/S0960-894X(02)00160-9

The paper covers continuing efforts to discover novel, potent and selective oxytocin antagonists. Further benzofuran derivatives with potent oxytocin antagonist activity and good pharmacokinetic parameters are reported. Efforts to improve the in vivo activity of the series are described.

Full text of DOI:10.1016/S0960-894X(02)00160-9

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1405–1411
Identification of Potent and Selective Oxytocin Antagonists.
Part 2: Further Investigation of Benzofuran Derivatives
a,
b
c
d
Paul G. Wyatt, * Michael J. Allen, Josie Chilcott, Christopher J. Gardner,
a
a,
a
David G. Livermore, Jackie E. Mordaunt, * Fabrizio Nerozzi,
a
a
d
c
Mamta Patel, Marion J. Perren, Gordon G. Weingarten, Shalia Shabbir,
c
c
Patrick M. Woollard and Ping Zhou
aDepartment of Medicinal Chemistry, GlaxoSmithKline, Medicines Research Centre, Gunnels Wood Road,
Stevenage, Herts SG1 2NY, UK
b
c
Department of Receptor Pharmacology, GlaxoSmithKline, Medicines Research Centre, Gunnels Wood Road,
Stevenage, Herts SG1 2NY, UK
Department of Research Biometabolism, GlaxoSmithKline, Medicines Research Centre, Gunnels Wood Road,
Stevenage, Herts SG1 2NY, UK
d
Neurology Centre of Excellence for Drug Discovery, GlaxoSmithKline, New Frontiers Science Park, Third Avenue,
Harlow, Essex CM19 5AW, UK
Received 22 November 2001; revised 11 February 2002; accepted 1 March 2002
Abstract—The paper covers continuing efforts to discover novel, potent and selective oxytocin antagonists. Further benzofuran
derivatives with potent oxytocin antagonist activity and good pharmacokinetic parameters are reported. Efforts to improve the in
vivo activity of the series are described. # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
of the benzoxazinone group (5), the benzofuran ring (6)
and the amide linker from the benzofuran ring to the
terminal pyridone of 4a.
In the search for oral agents that could significantly
1
delay preterm birth (the major cause of infant mortal-
ity and morbidity), potent and selective non-peptide
oxytocin (OT) antagonists have been identified, but
none has entered clinical development.2 In the first
paper of this series, use of amino moiety 1 (R=H) of
L-371,257 2 as a starting point for a chemical programme
Modification of the pyridone of 4a
,3
To optimise the activity and pharmacokinetic properties
of 4a its pyridone moiety was elaborated using base
mediated coupling (NaH or polymer supported
3
4
1
was reported. A number of oxytocin antagonists were
identified, such as the 2-substituted benzofuran deriva-
BEMP ) of bromide 4 (R =Br) with a set of hetero-
aromatic amides selected to contain the hydrogen bond
acceptor of 4a. A significant increase in antagonist
activity at the hOT receptor was obtained by substitut-
ing pyridone and other nitrogen heterocycles with both
polar groups (e.g., 4b,f,g) and lipophilic groups (e.g.,
4c,h,i) (Table 1). Also bicyclic heterocycles showed
potent activity 4d,e. The generality of the effect sug-
gested that the activity increase was caused by space
filling in the receptor site rather than any specific inter-
actions. The compounds investigated all had selectivity
over the vasopressin 1a (V ) receptor. Although the
tives 3, with the most potent compounds having general
1
structure 4 (R =amide-containing saturated or aro-
matic heterocycles). Although potent OT antagonists,
the saturated heterocyclic derivatives did not have sui-
table pharmacokinetic properties to warrant further
development. This paper reports efforts to optimise the
activity and pharmacokinetic parameters of the aro-
matic heterocyclic derivative 4a, through modification
1
a
*Corresponding author at current address: Astex-Technology, 250
increase in hOT activity was independent of the nat-
ure of the substituents, it did nevertheless affect the
Cambridge Science Park, Milton Road, Cambridge, CB4 0WE, UK.
Fax: +44-1438-763624; e-mail: jem8704@gsk.com
0
960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00160-9

1406
P. G. Wyatt et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1405–1411
pharmacokinetic properties of the compounds. Polar
groups such as the ethers 4b,f gave good dog bioavail-
ability, but conversely good rat bioavailability required
the introduction of lipophilic groups such as CF 4c,i.
3
Attempts to introduce water solublising groups such as
morpholine 4g resulted in a loss of bioavailability in
both dogs and rats. From this set of compounds, 4c was
identified as the compound with the best all round
properties, especially as rat bioavailability could be
increased to 30% by changing the dosing formulation
(self-emulsifying formulation, 20% v/v transcutol, 20%
v/v labrafil, 60% v/v labrasol). Therefore 4c has been used
a
Table 1. Inhibition of the binding of OT with human OT (hOT) and V1a (hV1a) receptors, and rat and dog pharmacokinetics
Compd
R
hOT pK
i
hV1a pK
i
Rat
Dog
Cl (mL/min/kg)
13
t1=2 (h)
Cl (mL/min/kg)
51
F%
55
t1=2 (h)
F%
29
2
8.2
8.2
<5.7
1.1
1.1
4
a
b
0.5
1.1
34
19
4
8.6
5.8
5.9
1.2
22
3
68
51
4
4
4
4
4
4
c
d
e
f
8.7
8.2
8.9
8.7
9.3
9.0
8.4
1.0
0.8
10
10
16
17
1.5
1.1
1.0
1.9
1.2
1.3
1.6
8
6
11
12
11
9
36
43
2
5.8
0.8
0.8
35
17
2
g
h
i
<1
4
5.4
0.8
20
14
6
34
a
Displacement of [H] oxytocin from hOT or vasopressin from hV1a by the test compound.5
3

P. G. Wyatt et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1405–1411
1407
F%
a
Table 2. Inhibition of the binding of OT with human OT receptor (hOT) and rat IV half-life and bioavailability
Compd
R
hOT pK
i
t1=2 (h)
Clb
F%
Compd
R
hOT pK
i
t1=2 (h)
Clb
5a
7.6
1.1
13
44
5d
8.2
5
5
a
b
c
7.0
8.2
5e
5f
8.1
8.5
1.2
1.0
11
21
67
35
0.9
17
18
3
5
Displacement of [H] oxytocin from hOT by the test compound.
Clearance (mL/min/kg).
b
a
Table 3. Inhibition of the binding of OT with human OT receptor (hOT) and rat IV half-life and bioavailability
Compd
Ring
hOT pK
i
t1=2 (h)
Clb
F%
Compd
R
hOT pK
i
t1=2 (h)
Clb
F%
6a
8.1
1.3
21
15
6d
7.3
6
6
b
c
8.5
7.0
0.6
26
2
6e
6f
7.1
8.0
a
3
5
Displacement of [H] oxytocin from hOT by the test compound.
Clearance (mL/min/kg).
b
as the lead structure in the rest of the paper to exemplify
further structure–activity relationship investigations.
potent activity they did not increase bioavailability,
however, incorporation of nitrogen into the ring 6b did
increase the water solubility 10-fold compared to 4c.
Other heterocycles such as the benzofuran isomer 6c;
indoles (e.g., 6d), synthesised from the palladium cou-
pling of the appropriate iodo-trifluoroacetamido-
benzene 8 with 7; and benzothiazoles 6e,f, from the
Modification of the benzoxazinone ring
To improve the pharmacokinetic parameters of 4c
modifications to the benzoxazinone ring were investi-
gated (Table 2). Although, amide 5a did have improved
bioavailability over 4c, its 10-fold loss of activity illus-
trates the substantial reduction of activity observed for
many analogues. The only compounds that retained
significant levels of activity were fluoro-derivatives 5c–e
and the chloro-derivative 5f. Only 5e had significantly
increased bioavailability compared to 4c, but this was
off set by a 5-fold loss of activity.
7
palladium coupling of iodo-anilines 9 with thioamide
8
10 (Scheme 1), all proved to be significantly less active.
Protein binding and in vivo activity
Due to its promising pharmacokinetic properties, 4c
was progressed into early pre-clinical development.
However, 4c proved to be less active in vivo in an OT
9
induced rat uterus contraction model (IC =3 mM)
5
0
than would have been predicted from its in vitro
0
Modification of the benzofuran ring
1
potency at the rat OT receptor (IC =2 nM). A pos-
5
0
Fluorine substituted benzofurans (e.g., 6a) and aza-
benzofurans (e.g., 6b) (Table 3) were synthesised to
block potential metabolism of the ring and for the latter
case to improve water solubility. They were synthesised
using analogous chemistry to that used to synthesise 4c,
involving the Sonogashira coupling of an appropriate
sible reason for this observation was that the con-
centration of the compounds available for interaction at
the OT receptor in vivo was limited by a high degree of
binding to plasma proteins. To investigate this possibi-
lity, in vivo and in vitro studies were carried out using a
selection of compounds of varying potency and physico-
6
1
iodo-phenol or iodo-pyridol with acetylene 7 (Scheme
1
chemical properties, including the peptide atosiban
1
1
12
). Although these modifications gave compounds with
(Tractocile), L-368,899, 2, 6b and 4c. Protein binding

1408
P. G. Wyatt et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1405–1411
ꢀ
Scheme 1. (a) CuI, PdCl
iPr
2
(PPh
3
)
EtN, DMF; (d) NaH, MeI, DMF; (e) 4 N HCl, dioxan, ether; (f) TBTU, HOBT, iPr
2
, HN¼CH(NCH
3
)
2
, DMF, 65 C; (b) KOH, MeOH, H
2
O; (c) 1, R=H, TBTU, 1-hydroxybenzotriazole (HOBT),
0
2
2
EtN, DMF; (g) 1-ethyl-3-(3 -dimethylaminopropyl)carbo-
0
.
diimide HCl (EDC HCl), HOBT, iPr
2
EtN, DMF; (h) Tris(benzylideneacetone)-palladium (0)-chloroform, 1,1 -bis(diphenylphosphino)-ferrocene,
CO tBu, NaH, DMF.
2
ꢀ
.
CaO, DMF, 60 C, DMF; (i) 4 N HCl, dioxane; (j) 11, E DCHCl, HOBT, iPr
2
EtN, DMF; (k) BrCH
2
Table 4. Comparison between in vitro and in vivo activity (rat) and
correlation with plasma protein binding
in both rat and human plasma was determined for 4c, 2
1
and atosiban . Binding of the compounds to purified
1
3
human serum albumin (HSA) was also measured, as
was their potency at the human OT receptor in the pre-
sence and absence of physiological concentrations of
Atosiban1 L-368,899
Compd
2
6b
4c
Rat in vivo IC50 (nM)a
Rat in vitro IC50 (nM)
136
40
42.6
0.81
7.4
1
164
10
82.4
1.71
7.6
4
640
25
83.7 86.2
1.66 1.85
8.1
9
744
20
3040
20
96.2
2.16
8.7
46
b
1
4
HSA. Rat serum albumin (RSA) was also examined.
The binding of compounds to HSA (or RSA) was simi-
lar to that observed with rat and human plasma pro-
teins (data not shown). When HSA was added to the in
vitro human OT potency screen a loss of activity (HSA
shift) was observed for many of the compounds and this
shift was similar when RSA was added. For technical
reasons HSA was used routinely. A close correlation was
observed between the extent of binding to HSA, increas-
ing lipophilicity, the HSA shift, and the difference
between the rat in vitro and in vivo activities (Table 4).
This led to the conclusion that a high degree of plasma
protein binding was responsible for the relatively poor
efficacy of 4c in the OT induced rat uterus contraction
model. Although we identified compounds with potent
activity and moderate HSA shifts (e.g., 4g, HSA
shift=9-fold), these compounds also had poor rat oral
bioavailabilites and so were not suitable for progression.
c
Protein binding (%)
d
CHI logD pH2
pK (human OT)
e
i
8.5
19
f
HSA shift
a
3
9
Displacement of [H] oxytocin from rat OT by the test compound.
Serum concentration for 50% inhibition of uterine contractility
b
10
response to oxytocin in rat.
c
13
Human serum albumin binding.
14
d
Lipophilicity measure.
3
e
f
5
Displacement of [H] oxytocin from hOT by the test compound.
3
HSA shift=Ratio of displacement of [H] oxytocin from hOT by the
test compound in the presence and absence of 50 mg/mL human
15
serum albumin.
molecules were designed with reduced PSA values com-
pared to 4c. As the previously described modifications
of 4c were detrimental to the properties of the series, the
amide linker between the benzofuran and pyridone ring
was investigated. As previous investigations suggested
this group was not involved in hydrogen bonding to the
receptor, non-amide containing replacements were
TM17
Modification of the amide linker
sought. Catalyst
was used to search for fragments
The accumulated data for the series suggested that
improved rat bioavailability was related to increasing
logD of the compounds, although, this also corre-
lated with increasing protein binding and HSA shift.
Conversely, reducing polar surface area (PSA)1 was
associated with an increase in rat bioavailability, but
not as closely with increasing protein binding and HSA
shift (data not shown). To exploit the latter finding
in the corporate compound database that bridged the
3-D space between the 2-position of the benzofuran ring
and the pyridone ring nitrogen using three low energy
conformations of 4c. Representative structures from
this exercise, together with ideas generated within the
group, were modelled against the conformers of 4c. A
number of compounds were then selected for synthesis
based on parameters such as; fit to one or more of the
6

P. G. Wyatt et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1405–1411
1409
ꢀ
3 3 3 2 3 2
) , E tN, CuI, PdCl (PPh ) , DMF, 45 C, 53%; (c) tetrabutyl-
, DMF, 65 C, 44%.
Scheme 2. (a) NaH, 5-trifluoromethyl-2-pyridone, DMF, 37%; (b) HCꢁCSi(CH
ꢀ
ammonium fluoride, THF, rt, 34%; (d) Et
3
N, CuI, PdCl (PPh
2
3
)
2
pharmacophores; lower PSA values than 4c; and the
presence of basic functionality. Some of the compounds
are shown in Table 5.
The aromatic linkers 3d–g, synthesised using the meth-
odology exemplified for 3g (Scheme 2), gave a range of
activity dependent on the presence and position of the
pyridyl nitrogen atom. The greater activity of 3g over
the other analogues can be rationalised by a conforma-
tional restriction caused by repulsion of the benzofuran
oxygen and the pyridine nitrogen. This observation is
supported by the lack of activity of the isomer 3e.
Although 3g had good activity and reasonable pharma-
cokinetic parameters it had a high shift in the presence
of HSA. The difference in the HSA shift seen for 3g and
the isomer 3f suggested that the magnitude of the shift
As expected, because the modelling was carried out
against a number of conformations of 4c, a range of
activity was observed for the synthesised analogues.
Compounds with potent activity 3a,b,g were identified,
but these tended to have high HSA shifts and poor
pharmacokinetic parameters, whereas closely related
structures (3c) with good pharmacokinetic parameters
had moderate levels of activity.
a
b
Table 5. Inhibition of the binding of OT with human OT receptor (hOT), effect of protein binding on in vitro activity and rat pharmacokinetics
Compd
pK
i
hOTa
8.8
HSA shiftb
46
Protein bindingc
93
t1=2 (h)
Cl (mL/min/kg)
10
AUC (h. ng/mL)
1600
%F
19
PSAd
116
4
3
c
1.5
a
7.6
8.0
222
90
96
98
0.6
28
150
2
86
82
3
3
3
3
3
b
c
d
e
f
6.1
6.1
4
93
98
96
96
1.7
2.2
6
7
12000
5700
79
93
78
90
91
89
>45
<5.5
7.2
17
87
1.4
1.2
7
5400
1500
41
22
3g
8.0
98
12
a
3
5
Displacement of [H] oxytocin from hOT by the test compound.
3
b
15
Ratio of displacement of [H] oxytocin from hOT by the test compound in the presence and absence of 50 mg/mL human serum albumin.
c
Human serum albumin binding.13
d
16
PSA calculated polar surface area.

1410
P. G. Wyatt et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1405–1411
was not just due to gross physicochemical properties,
but also had a structural element to it. Although not
suitable for progression, these highly conformationally
constrained systems have given important structural
information for further compound design.
anaesthetic, respectively. The right femoral vein was cannu-
lated for dosing of OT, OT antagonists or vehicle and a sec-
ond femoral artery cannulated for blood sampling. The left
uterine horn was exposed, a suture tied around the anterior
part of the horn ꢃ1 cm posterior to the ovary, and a second
positioned ꢃ1 cm posterior to the first tie. The anterior end of
the uterine horn was anchored in the abdominal cavity and the
posterior end connected to a strain gauge (Dymometer UF1)
under a resting tension of ꢃ2 g. Rectal temperature was
monitored and core body temperature was maintained at
Conclusion
ꢀ
The compound 4a was used as a lead to identify potent
and selective oxytocin antagonists in vitro. Investiga-
tions demonstrated that much of the molecule was
intolerant to modification. However, a significant
increase in activity and improvement of pharma-
cokinetic parameters was achieved by optimisation of
the pyridone moiety of 4a. The in vivo activity of the
compounds was significantly lower than their in vitro
activity, possibly due to protein binding. This loss of
activity was paralleled in vitro by the addition of human
serum albumin to the hOT binding screen, allowing
selection of candidates for in vivo evaluation. This later
finding and the structural information obtained during
these studies will be valuable in the design of oxytocin
antagonists with potent in vivo activity.
37 C. Following surgery, animals were allowed to stabilise for
30 min during which uterine resting tension was maintained at
ꢃ2 g. A 10-min baseline recording of the spontaneous uterine
contractile activity was then made. Two consecutive, two-
point control dose–response curves to OT (0.03 and 0.3 mg
ꢂ1
ꢂ1
kg iv; dose volume (saline) 0.3 mL kg ) were then con-
structed. OT doses were injected 15 min apart, and 45 min
allowed between dose–response curves. The onset and dura-
tion of the antagonist effect of test compounds were deter-
ꢂ1
mined by monitoring the response to 0.3 mg kg OT 3, 30, 60,
9
0, 120, 150 and 180 min after a single iv bolus dose of
antagonist. Blood samples (200 mL) were withdrawn, 5 min
before and 1, 13, 40, 70, 100, 130, 160 and 190 min following
antagonist administration and blood volume was replaced
with Haemaccel (Hoescht). Blood samples were centrifuged at
1
3,000 rpm for 2 min and plasma samples were stored at
ꢀ
ꢂ20 C until analysis of test compound by mass spectroscopy.
OT-induced uterine contractile responses were quantified by
measuring the area under the contraction–time curve for a 10
min period, beginning immediately after each dose of OT.
Area was determined using a Modular Instruments Data
Capture system. Test compounds were dissolved in DMSO,
polyethylene glycol-200 (PEG200) and distilled water (ratio
Acknowledgements
The authors would like to thank Ms. Klara Valko for
CHI logD and protein binding measurements.
of 20:50:30) to give a final drug concentration of 2–6 mg
ꢂ1
base mL
.
1
1
0. Rat Oxytocin Binding Assay. Rats that had littered within
2 h were sacrificed. The myometrium removed to a cold assay
References and Notes
2.
2
buffer (50 mM HEPES, 10 mM, MgCl 6H O, pH to 7.4 with
a saturated solution of Tris base), cut into short lengths (<5
mm) and homogenised in homogenisation buffer (250 mM
sucrose; 40 mM l-histidine; pH 7.3) on ice (20 mL) using a
Braun (Teflon/glass) homogeniser (20 strokes at 1000 rpm at
1
1
2
2
3
. Mitchell, B. F.; Schmid, B. J. Soc. Gynecol. Invest. 2001, 8,
22.
. Thornton, S.; Vatish, M.; Slater, D. Exp. Physiol 2001, 86,
97.
. Wyatt, P. G.; Allen, M. J.; Chilcott, J.; Foster, A.; Liver-
ꢀ
4 C). The suspension was centrifuged (500g) for 10 min at
4 C. The supernatant was removed and retained on ice. The
ꢀ
more, D. G.; Mordaunt, J. E.; Scicinski, J.; Woollard, P. M.
Bioorg. Med. Chem. Lett. 2002, 12, 1399 and references therein.
4. Polymer supported BEMP: 2-tert-butylimino-2-diethyla-
mino-1,3-dimethyl-perhydro-1,3,2-diazaphosphorine on poly-
pellet was resuspended in homogenisation buffer, homo-
genised and then centrifuged as above. The pooled super-
ꢀ
natants were centrifuged at 48,000g for 20 min at 4 C. The
pellet was resuspended in 8 mL assay buffer per two rats.
Radioligand binding was determined using a filtration binding
styrene; ex Fluka (cat no. 20026) Schwesinger, R. Chimia
1
3
985, 39, 269.
. Human OT binding assay method, see: Wyatt, P. G.;
assay. Each well contained membrane (100 mL), [ H]oxytocin
5
(50 mL, 1 nM, 130 Ci/mmol, Amersham UK) in the presence
of competing ligands. All reagents were prepared in assay
buffer (total assay volume 200 mL). The plates were incubated
for 1 h at room temperature before being filtered through dry
Whatman GF/C filters, using a Brandel Cell harvester. The
filters were washed with ice cold assay buffer (4ꢄ1 mL) to
remove unbound ligand, punched out into 5 mL counting vials
and Scintillant added. After >4 h to digest the filter, bound
Hickin, G.; Miller, N. D.; Allen, M. J.; Chilcott, J.; Woollard,
P. M. Bioorg. Med. Chem. Lett. 2001, 11, 1301. Human V1a
binding assay method identical to the hOT method except 10
3
mg hV1a membrane used, 1 nM [ H]vasopressin used as ligand
(
K
d
0.79nM).
. Moloney, G. P.; Martin, G. R.; Mathews, N.; Hobbs, H.;
6
Dodsworth, S.; Sang, P. Y.; Knight, C.; Maxwell, M.; Glen,
R. C. J. Chem. Soc., Perkin Trans. 1 1999, 19, 2713.
3
[ H]oxytocin was measured using a Packard Liquid Scintila-
7
8
1
9
. Edgar, K. J.; Falling, S. N. J. Org. Chem. 1990, 55, 5287.
. Moody, C. J.; Bagley, M. C. J. Chem. Soc., Perkin Trans. 1
998, 3, 601.
tion Counter. A four parameter logistic fit was used to esti-
mated IC50 and Hill slope (by non-linear least squares).
11. (a) Bossmar, T. J. Perinat. Med. 1998, 26, 458. (b)
Romergo, R.; Sibai, B. M.; Sanchez-Ramos, L.; Valenzuela,
G. J.; Veille, J.-C.; Tabor, B.; Perry, K. G.; Varner, M.;
Goodwin, T. M.; Lane, R.; Smith, J.; Shangold, G.; Creasy,
G. W. Am. J. Obstet. Gynecol. 2000, 182, 1173.
ꢂ1
. Diethylstilboestrol (250 mg kg ip; dosed in corn oil, 1.0
ꢂ1
mL kg ) pre-treated (18 h) adult female AHA rats were
ꢂ1
anaesthetised with sodium pentobarbitone (60 mg kg ip).
ꢂ1
After 60–120 min IV infusion (3 mg h ) of sodium pento-
barbitone was commenced and continued throughout the
experiment. The left femoral artery and vein were cannulated
for measurement of blood pressure/heart rate and infusion of
12. Williams, P. D.; Anderson, P. S.; Ball, R. G.; Bock, M. G.;
Carroll, L.; Chiu, S.-H. L.; Clineschmidt, B. V.; Culberson,
J. C.; Erb, J. M.; Evans, B. E.; Fitzpatrick, S. L.; Freidinger,

P. G. Wyatt et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1405–1411
1411
R. M.; Kaufman, M. J.; Lundell, G. F.; Murphy, J. S.; Paw-
luczyk, J. M.; Perlow, D. S.; Pettibone, D. J.; Pitzenberger,
S. M.; Thompson, K. L.; Veber, D. F. J. Med. Chem. 1994, 37,
15. Ratio of the IC50 hOT in the presence of 50 mg/mL HSA
to IC50 hOT in the absence of 50 mg/mL HSA. Method as in
ref 5. The ligand [ H]oxytocin was made up in buffer contain-
ing 100 mg/mL HSA. HSA was obtained from Sigma (Poole,
Dorset, UK).
16. Clark, D. E. J. Pharm. Sci. 1999, 88, 807.
17. Catalyst Accelrys Inc., a subsidiary of Pharmacopeia, Inc.
All rights resereved.
3
5
1
1
65.
3. An HSA coated HPLC column method.
4. An HPLC method based measurement of lipophilicity.
Valko, K.; Du, C. M.; Bevan, C.; Reynolds, D. P.; Abrahams,
M. H. Curr. Med. Chem. 2001, 8, 1137.