Long-chain formoterol analogues: An investigation into the effect of increasing amino-substituent chain length on the β2-adrenoceptor activity

Article abstract of DOI:10.1016/j.bmcl.2004.06.086

The synthesis of a series of long-chain formoterol analogues in which the terminal ether residue of the β-phenethyl-amino-substituent has been extended beyond the methyl ether residue present in the parent compound are described. Evaluation of these analo

Full text of DOI:10.1016/j.bmcl.2004.06.086

Bioorganic & Medicinal Chemistry Letters 14 (2004) 4705–4710
Long-chain formoterol analogues: an investigation into the effect
of increasing amino-substituent chain length on the
b2-adrenoceptor activity
Vahid Alikhani, David Beer, David Bentley, Ian Bruce, Bernard M. Cuenoud,
Robin A. Fairhurst,* Peter Gedeck, Sandra Haberthuer, Claire Hayden,
Diana Janus, Lynne Jordan, Christine Lewis, Kirsty Smithies and Elke Wissler
Novartis Horsham Research Centre, Wimblehurst Road, Horsham, West Sussex RH12 5AB, United Kingdom
Received 24 May 2004; revised 21 June 2004; accepted 25 June 2004
Available online 23 July 2004
Abstract—The synthesis of a series of long-chain formoterol analogues in which the terminal ether residue of the b-phenethyl-
amino-substituent has been extended beyond the methyl ether residue present in the parent compound are described. Evaluation
of these analogues as b2-adrenoceptor agonists was used to provide an insight into the factors controlling the magnitude and dura-
tion of receptor activation.
ꢀ 2004 Elsevier Ltd. All rights reserved.
The two currently prescribed inhaled long-acting b2-
adrenoceptor agonists formoterol 1 and salmeterol 2
play a central role as bronchodilators in the treatment
of the respiratory diseases asthma,1 and COPD.2 For-
moterol is characterised as a high-efficacy agonist with
a rapid onset of action.3 In contrast, salmeterol is a
low-efficacy agonist with a relatively slow onset of ac-
tion.4 Despite these differences both compounds provide
sustained relief of bronchoconstriction following a
twice-daily dosing regimen. In common with others,
we are interested in identifying inhaled b2-agonists with
extended durations of action, which are suitable for
once-daily dosing as the next generation of compounds
from this therapeutic class.5
lieved to enable repeated stimulation to occur without
complete dissociation of the ligand, leading to the sus-
tained effect. Key to this interaction is the location of
the ether residue within the 6-(4-phenylbutoxy)hexyl
amino chain of salmeterol, which is thought to provide
the conformational freedom for the dual binding modes
to occur.4,8 In the diffusion microkinetic theory, the ba-
sic centre and lipophilic nature of both formoterol and
salmeterol allows them to partition effectively into the
lipid bilayers of smooth muscle following local delivery
to the lung. This accumulation of material within the li-
pid bilayer functions as a depot to maintain an effective
concentration of agonist over time, resulting in the ob-
served duration of effect. Additionally for this theory,
the greater lipophilicity and nature of the membrane
interaction for salmeterol have been proposed to ac-
count for the slow onset of action due to entry into
the b2-adrenoceptor directly from the lipid bilayer as
the primary route.9 Some controversy surrounds these
two hypothese centering around the role and extent
to which the exosite interaction with salmeterol defines
the observed pharmacology.10,11 However, an interac-
tion with this region of the receptor for salmeterol
appears to be well established, through natural,12
and engineered mutations,13 even if the nature and
degree to which the exosite interaction contributes to
the molecules biological activity remains to be fully
defined.
To rationalise the duration of action of formoterol and
salmeterol two hypothese have been put forward involv-
ing a key interaction with either the receptor, termed the
exosite theory,6 or the surrounding membrane, termed
the diffusion microkinetic theory.7 In the former, a sec-
ondary interaction distant from the primary agonist
binding region is implicated in which the long-chain of
salmeterol anchors to an additional portion of the b2-
adrenoceptor, termed the exosite. This interaction is be-
*
Corresponding author. Tel.: +44-(0)-1403-323508; fax: +44-(0)-1403-
323307; e-mail: robin.fairhurst@pharma.novartis.com
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.06.086

4706
V. Alikhani et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4705–4710
Although the six-atom nitrogen-to-oxygen separation is
believed to be an essential structural element for the
activity observed with salmeterol,4 the hexyl spacer does
allow for significant conformational freedom. A conse-
quence of this is to make any accurate prediction of
the active spatial orientation in this region of the mole-
cule difficult. Our interest centred on the observation
that this proposed key six-atom separation between
the basic nitrogen and ether oxygen is also present in a
number of other well described b2-adrenoceptor agon-
ists, which include picumeterol 3, sibenadet 4, formoter-
ol 1, fenoterol 5 and TA-2005 6.11,14 The last three of
which all incorporate a para-oxygen substituted b-phen-
ethyl-amino-substituent as the common six-atom spacer.
Therefore, potentially enabling an equivalent interaction
for the common oxygen atom of the ether side chains, as
highlighted in Figure 1. The reduced conformational
mobility of the b-phenethyl moiety possibility offering
the opportunity of a more energetically favourable
interaction.
in combination with molecular modelling studies. The
key interactions are highlighted for salmeterol and for-
moterol in Figure 2. In summary, the important interac-
tions between agonists and the b2-adrenoceptor have
been established as involving; the catechol mimic and
serine residues 204 and 207 on the fifth transmembrane
spanning domain (TM5),15 the protonated amine and
Asp-113 on TM3,16 and the benzylic alcohol with the
chirally discriminating Asn-293 on TM6.17 In addition
to these primary interactions, Tyr-308 on TM7 has been
identified from molecular modelling studies as defining a
limiting-surface of the binding pocket for the amino-
substituents of formoterol, salmeterol and TA-2005.18
This interaction in the case of salmeterol enables the
much larger 4-phenylbutyl ether substituent to orientate
itself, facilitated by the presence of the ether linkage,
back into the b2-adrenoceptor.8 A consequence of this
alignment is to bring the 4-phenylbutyl ether moiety into
an interaction with the amino acids localised around
positions 149–158 on TM4, which define the key resi-
dues of the proposed exosite.13
Based upon the above correlation, we hypothesised that
the potential exists with formoterol analogues for ena-
bling the same secondary receptor binding, put forward
for salmeterol. This we anticipated could be achieved by
extension of the para-b-phenethyl terminal ether residue,
to facilitate an interaction with the exosite region of the
b2-adrenoceptor. Further support for such a commonal-
ity of binding modes with the above catecholamine-
based agonists can be derived from receptor mutagenesis
In this letter we describe our initial results from sequen-
tially increasing the size of the b-phenethyl oxygen sub-
stituent for a series of formoterol analogues to match
the equivalent terminal ether residue of salmeterol. We
anticipated that determination of the b2-adrenoceptor
activity for this series of homologues would provide an
insight into the possibility of an analogous interaction
to that proposed for salmeterol with the exosite region.
OMe
O
HN
HN
HO
HO
O
N
formoterol 1
salmeterol 2
HO
OH
H
HO
O
O
S
O
HN
HN
O
HO
S
O
N
H
Cl
N
Cl
sibenadet 4
picumeterol 3
HO
OMe
OH
HN
HN
HO
HO
fenoterol 5
TA-2005 6
N
H
O
HO
OH
HO
Figure 1. Comparison of b2-adrenoceptor agonist structures of formoterol, salmeterol, picumeterol, sibenadet, fenoterol and TA-2005.

V. Alikhani et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4705–4710
4707
Ty r - 308
Ty r - 308
OH
As p- 113
As p- 113
O
O
OH
O
Me
O
O
H
N+
H
N+
O
As n- 293
H
H
As n- 293
O
O
HO
HO
NH2
NH2
O
N
H
HO
OH
HO
HO
Res i dues 149- 158
Ser - 204
Res i dues 149- 158
OH
HO
OH
Ser - 207
Ser - 204
Ser - 207
Figure 2. Key b2-adrenoceptor/agonist interactions for salmeterol and formoterol.
Table 1. Structures and position of the b-phenethyl ether substituents
R
substituted hydroxyphenylacetic acids 7, carbodiimide
couplings yielded the corresponding N-benzylated car-
boxamides 8. The phenolic residues were then alkylated
with in situ generated alkyl iodides or, in the case of the
2-indanyl example, via a Mitsunobu reaction. Subse-
quent reduction of the amide functionality, with either
diborane or lithium aluminium hydride, gave the desired
secondary amines 9b–g. To prepare the bis-arylether
intermediate 9a an alternative sequence was employed.
Starting from commercial 4-phenoxy-b-phenylethylam-
ine, N-benzoylation followed by amide reduction with
lithium aluminium hydride produced the desired benzyl-
ated amine 9a in 48% overall yield.
Compound
Substituent: R
Regiochemistry
a
b
c
–Ph
para
para
para
para
–CH2CH2Ph
–CH2CH2CH2Ph
–CH2CH2CH2CH2Ph
d
e
para
f
–CH2CH2CH2CH2Ph
–CH2CH2CH2OPh
meta
meta
g
In addition a selection of closely related derivatives were
prepared and studied to further investigate the factors
controlling the magnitude and duration of the effect at
the b2-adrenoceptor.
The above secondary benzylated amines 9a–g were then
reacted under conditions of either conventional or
microwave heating to open the chiral epoxide 10. Reac-
tion occurring selectively at the least hindered position,
yielding the ethanolamine derivatives 11a–g.19 Subse-
quent functional group manipulation, via hydrogena-
tion of the nitro residues over platinum oxide,
followed by formylation of the liberated aniline func-
tionality gave the corresponding formamides. Finally,
hydrogenolysis of the benzyl protecting groups yielded
the target molecules 12a–g, as single enantiomers of
the more active (R)-configuration (Scheme 2).20
To explore the above possibility a series of N-benzylated
b-phenylethylamines bearing para-substituted ether res-
idues of increasing chain length and terminated with a
phenyl moiety were prepared as key intermediates, en-
tries 9a–d in Table 1. Additionally, as a way to further
explore the structure–activity relationship, related exam-
ples were prepared by varying the nature and position of
the ether residue, entries 9e–g in Table 1. The synthetic
sequence used to prepare the intermediate amines 9b–g
is shown in Scheme 1. Starting from the appropriately
Compounds 12a–g were screened in a filtration bind-
ing assay to determine their affinities for the human
O
O
i)
OH
OH
Ph
N
H
HO
7
8
ii), iii)
OR
Ph
N
H
9b-g
Scheme 1. Synthesis of the amines 9b–g: (i) BnNH2, EDCI, HOAt, CH2Cl2, rt (88–93%); (ii) R–Cl, CsCO3, KI, CH3CN, 90ꢁC (86–90%); or, R–OH,
PPh3, DEAD, THF, rt (26%); (iii) B2H6, THF, reflux, then MeOH, reflux (44–95%); or, LiAlH4, THF, reflux (29–33%).

4708
V. Alikhani et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4705–4710
phenethyl moiety in compounds 12a–g, when compared
Ph
with formoterol.20 These binding data indicate that no
marked change in affinity for the b2-adrenoceptor is
apparent for 12d where, based upon the proposed
hypothesis the terminal ether residue is in an equivalent
position to that present in salmeterol.
OR
O
N
HO
i)
NO2
BnO
The functional consequences of the above interaction
with the b2-adrenoceptor were measured in a guinea-
pig tracheal strip assay.22 We noted a greater variability
in the ability of the compounds 12a–g to agonise the re-
ceptor, measured by inhibition of contraction, when
compared with their binding affinities. IC50 values ran-
ged from being comparable with formoterol, to greater
than 200-fold less potent. In all instances greater than
an 80% inhibition of the electrically-stimulated contrac-
tion was achieved at the higher concentrations investi-
gated in the dose response studies (1lM). Thus,
indicating that all the compounds retain a significant de-
gree of the high-efficacy agonist properties associated
with formoterol. Comparing the para-substituted iso-
mers: the biphenyl ether 12a in accord with the above
binding data resulted in a marked 86-fold fall in func-
tional potency relative to formoterol. Increasing the sep-
aration between the phenyl residues by incorporation of
an alkyl chain resulted in an improvement in functional
activity to the level of formoterol for the para-substi-
tuted 2-phenethyl ether derivative 12b, with the func-
tional potency decreasing upon increasing alkyl chain
length in going from two to four methylene groups, en-
tries 12b–d. However, a satisfactory level was still re-
tained in the para-4-phenylbutyl ether derivative 12d,
with the terminal ether residue equivalent to salmeterol.
This decrease in potency of up to 51-fold relative to for-
moterol may provide a better opportunity for formula-
tion as an inhalation product with activity in this
range.24 The 2-indanyl derivative 12e, a cyclic analogue
of the potent 12b, resulted in a marked 228-fold fall in
functional activity although this was again not apparent
from the binding data. Therefore, this result potentially
indicates that steric bulk and conformational restriction
about the aryl ether linkage are relatively poorly toler-
ated for receptor activation in this series of compounds,
a rationale, which is further supported by the reduced
activity seen for the biphenyl ether 12a. Switching the
point of attachment of the ether linkage in the b-phen-
ethyl moiety to a meta-relationship and hence reducing
the six-atom separation between basic nitrogen and
ether oxygen by one atom resulted in essentially no
change in functional activity, entry 12f, compared with
the para-isomer 12d. Thus, these data suggest that the
role of the side-chain ether oxygen in this b-phenethyl
series is less critical for receptor activation compared
with salmeterol. Incorporation of a second ether residue
into the four-atom meta-linker, entry 12g, was tolerated
with little change in activity compared to the carbon
analogue 12f.
NO2
10
BnO
11a-g
OR
HN
HO
ii), iii)
O
N
H
HO
12a-g
Scheme 2. Synthesis of the b2-adrenoceptor agonists 12a–g: (i) 9a–g
toluene, reflux, 18h (53–82%); or, neat, Prolabo Synthewaveꢂ 402
microwave, 25min at 110ꢁC (63–92%); (ii) 1atm H2, PtO2, 1:1 toluene:
THF, rt (88–98%); (iii) aged mixture HCO2H/Ac2O, 1:1 toluene–THF,
rt (62–86%); or, HCO2H, oxalyl chloride, imidazole, Et3N, CH2Cl2, rt
(25%); (iv) 1atm H2, Pd/C, MeOH, rt (58–89%).
b2-adrenoceptor.21 Additionally, IC50 values for the
compounds ability to inhibit the electrically induced
contraction of superfused guinea-pig trachea were deter-
mined as a measure of the functional response at the b2-
adrenoceptor.22 Biological data and calculated logD7.4
values,23 as an assessment of the membrane partitioning
capability, are shown in Table 2. Membrane partitioning
being a key determinant of the compounds duration of
action potential based upon the diffusion microkinetic
theory.7
Comparison of the affinities of the compounds 12a–g for
the b2-adrenoceptor showed little variability in the Ki
values with the exception of the biphenyl ether 12a,
which was slightly reduced. Interestingly, all of the com-
pounds as the more active (R)-enantiomer exhibited de-
creased affinities for the b2-adrenoceptor relative to
racemic formoterol. A major proportion of this small
drop in affinity could potentially be accounted for by
the absence of the (R)-a-methyl substituent of the b-
Table 2. Biological data and calculated logD7.4 values for formoterol,
salmeterol and compounds 12a–g
Compound
b2-Adrenoceptor clogD7.4 Guinea-pig trachea
Ki (nM)
IC50 (nM)a
Formoterol 1
Salmeterol 2
2.6
1.0
0.06
1.24
1.95
1.67
2.35
2.83
2.95
2.83
2.00
1.4
15
12a
12b
12c
12d
12e
12f
12g
35.4
7.8
120
1.1
38
5.9
9.2
71
11.6
8.5
251
114
164
Comparison of the above functional data with the bind-
ing affinities highlights the need for a functional assess-
ment to understand the impact of the above structural
changes on receptor activation. The nature and position
of the ether residue in the series of compounds 12 having
3.3
a Determined from maximal response after compound present in su-
perfusate for initial 30min (n P 3).

V. Alikhani et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4705–4710
4709
been shown to have a significant effect on functional b2-
adrenoceptor activation. Analysis of the above struc-
ture–activity relationship does not highlight a trend for
an optimal interaction to indicate that the most favour-
able profile is attained with 12d, which based upon the
hypothesised structural equivalence is proposed as the
closest analogue in this series to salmeterol. However,
this approach has enabled functionally-active and
high-efficacy b2-adrenoceptor agonists to be identified
of increased lipophilicity of up to greater than 2.8 calcu-
lated logD7.4 units higher than formoterol. Based upon
the diffusion microkinetic theory, these compounds have
the potential to produce the desired increase in duration
of action. Representative duration of action data from a
superfused guinea-pig tracheal strip assay, in which a
drug-free washout phase of up to 9.5h was included,
for formoterol, 12d, and 12g, are shown in Figure 3.22
The concentrations shown for each compound are se-
lected to most closely approximate to an IC80 at the
end of the 30min compound administration phase. Du-
rations of action were taken as the time to return to 50%
of the maximal effect for each compound from the end
for the long-chain analogues 12d and 12g: a greater than
107-fold and 30-fold increased preference for a phosp-
holipids, over an aqueous phase, respectively, was deter-
mined when compared with formoterol. This increased
capacity for retention in lipid bilayers was reflected in
an increase in the duration of the relaxant effect over
formoterol of, 3.4-fold and 2.8-fold, respectively, for
12d and 12g when compared at equi-effective concentra-
tions. More than a doubling of the duration of effect
over formoterol being obtained not only for 12d where
the hypothesised key six-atom nitrogen-to-oxygen sepa-
ration is present, but also 12g where it has been reduced
by one atom. A one-atom shift in the position of the
ether oxygen having previously been shown to have a
marked impact on the duration of action for a series
of salmeterol analogues.4 Thus, when we assessed the
differences seen between 12d and 12g the extent of the in-
crease in lipophilicity appeared to best explain the de-
gree of enhancement in duration of action observed
over formoterol rather than as a consequence of a spe-
cific interaction with the exosite region of the b2-adreno-
ceptor. Additionally, from the data shown in Figure 3,
the maximal effect for 12d, like formoterol, is achieved
within 5min of the end of the initial 30min drug admin-
istration phase, which is consistent with the potential for
a rapid onset of action to be conserved.22
of the 30min compound administration phase and are
25
summarised in Table 3 with measured logD7.4
,
and
KIAM values,26 as an assessment of the membrane parti-
tioning capability.27
The measured data to assess the membrane partitioning
capabilities for formoterol, 12d, and 12g were found to
correlate with the calculated logD7.4 values presented
in Table 2. Taking the differences in the KIAM values
In summary, a series of long-chain formoterol analogues
in which the ether residue of the b-phenethyl-amino-sub-
stituent was varied have been evaluated as b2-adrenocep-
tor agonists to explore if an analogous exosite interaction
to that proposed for salmeterol was evident. In vitro
studies for this series of compounds highlighted the need
for a functional assessment to fully understand the struc-
ture–activity relationship for receptor activation.
Although evaluation of the data does not rule out an
interaction with the exosite region of the receptor for
the compounds 12, no variation in the degree and dura-
tion of agonist activity to support a dual binding mode
was apparent. However, this approach has enabled
high-efficacy b2-adrenoceptor agonists of increased lipo-
philicity to be identified, which from the diffusion micro-
kinetic theory would be anticipated to have the potential
for an extended duration of action as a result of greater
membrane partitioning. This has been demonstrated for
the compounds 12d and 12g, which produced a greater
than 2.5-fold longer duration of action compared to
the currently prescribed twice-daily b2-adrenoceptor ag-
onist formoterol. Additionally in the case of 12d, this in-
creased duration of action was also associated with the
desired rapid onset of action profile. Further studies to
explore the potential of the long-chain analogue 12d
(NVP-QAC455) as a once-daily long-acting b2-adreno-
ceptor agonist will be described in due course.
-25
0
25
50
75
100
0
1
2
3
4
5
6
7
8
9
10
Time [hrs]
Figure 3. Electrically-stimulated superfused guinea-pig tracheal strip
assay (Electrical stimulation was delivered every 2min, data shown are
for each hour following starting the superfusion of the test compound,
and at the end of the 30min drug administration phase. Data are
expressed as mean standard error of the mean) data for formoterol,
12d and 12g } formoterol 30nM (n=3), m 12d 100nM (n=5), j 12g
1000nM (n=4).
Table 3. Durations of action and measured logD7.4 values for
formoterol, 12d and 12g
References and notes
Compound
Duration-of-action (h)
logD7.4
KIAM
Formoterol 1
1.92
6.62
5.43
0.79
2.76
2.49
0.70
>2.73a
2.18
1. Global Initiative for Asthma: Global strategy for asthma
management and prevention. NHLBI/WHO Workshop
Report, Publ 02-3659, 2002.
2. Global Initiative for Obstructive Lung Disease: Global
strategy for the diagnosis, management and prevention of
12d
12g
a logKIAM value above upper limit of measurable range for assay
system.

4710
V. Alikhani et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4705–4710
chronic obstructive pulmonary disease. National Heart
lung and Blood Institute, Bethesda, 2000.
3. Matthys, H. Respiration 2001, 68, 432.
4. Johnson, M. Med. Res. Rev. 1995, 15, 225.
5. Anon Expert Opin. Ther. Pat. 2003, 13, 273.
6. Coleman, R. A.; Johnson, M.; Nials, A. T.; Vardey, C. J.
Trends Pharmacol. Sci. 1996, 17, 234.
´
7. Anderson, G. P.; Linden, A.; Rabe, K. F. Eur. Respir. J.
1994, 7, 569.
8. Lewell, X. Q. DrugDes. Discovery 1992, 9, 29.
9. Ochsner, M.; Jaekel, K.; Mutz, M.; Anderson, G. P.;
John, E. Eur. J. Med. Chem. 1994, 34, 451.
10. (a) Teschemacher, A.; Lemoine, H. J. Pharmacol. Exp.
18. (a) Furse, K. E.; Lybrand, T. P. J. Med. Chem. 2003, 46,
4450; (b) Isogaya, M.; Sugimoto, Y.; Tanimura, R.;
Tanaka, R.; Kikkawa, H.; Nagao, T.; Kurose, H. Mol.
Pharmacol. 1999, 56, 875; (c) Isogaya, M.; Yamagiwa, Y.;
Fujita, S.; Sugimoto, Y.; Nagao, T.; Kurose, H. Mol.
Pharmacol. 1998, 54, 616; (d) Kikkawa, H.; Isogaya, M.;
Nagao, T.; Kurose, H. Mol. Pharmacol. 1998, 53, 128.
19. Hett, R.; Fang, Q. K.; Gao, Y.; Hong, Y.; Butler, H. T.;
Nie, X.; Wald, S. A. Tetrahedron Lett. 1997, 38, 1125.
20. (a) Waldeck, B. Chirality 1993, 5, 350; (b) Trofast, J.;
¨
Osterberg, K.; Ka¨llstro¨m, B.-L.; Waldeck, B. Chirality
1991, 3, 443; (c) Murase, K.; Mase, T.; Ida, H.; Takahashi,
K.; Murakami, M. Chem. Pharm. Bull. 1978, 26, 1123.
21. Ki values were determined by the displacement of 3H
CGP12177 from the human b2-adrenoceptor expressed in
insect Sf21 membranes.
´
Ther. 1999, 288, 1084; (b) Bergendal, A.; Linden, A.;
Skoogh, B. E.; Gerspacher, M.; Anderson, G. P.; Lo¨fdahl,
C. G. Br. J. Pharmacol. 1996, 117, 1009.
11. Austin, R. P.; Barton, P.; Bonnert, R. V.; Brown, R. C.;
Cage, P. A.; Cheshire, D. R.; Davis, A. M.; Dougall, I. G.;
Ince, F.; Pairaudeau, G.; Young, A. J. Med. Chem. 2003,
46, 3210.
22. Nials, A. T.; Sumner, M. J.; Johnson, M.; Coleman, R. A.
J. Pharmacol. 1993, 108, 507.
23. Calculated logD7.4 values were generated using ACD
cLogD version 7.06 software.
12. Green, S. A.; Rathz, D. A.; Schuster, A. J.; Liggett, S. B.
Eur. J. Pharmacol. 2001, 421, 141.
24. Malcolmson, R. J.; Embleton, J. K. Pharm. Sci. Tech.
Today 1998, 1, 394.
13. Green, S. A.; Spasoff, A. P.; Coleman, R. A.; Johnson, M.;
Liggett, S. B. J. Biol. Chem. 1996, 271, 24029.
14. Waldeck, B. Eur. J. Pharmacol. 2002, 445, 1.
15. Strader, C. D.; Candelore, M. R.; Rands, E.; Hill, W. S.;
Dixon, R. A. F. J. Biol. Chem. 1988, 263, 10267.
16. Strader, C. D.; Candelore, M. R.; Hill, W. S.; Sigal, I. S.;
Dixon, R. A. F. J. Biol. Chem. 1989, 264, 13572.
17. Wieland, K.; Zuurmond, H. M.; Krasel, C.; IJzerman, A.
P.; Lohse, M. J. Proc. Natl. Acad. Sci. U.S.A. 1996, 93,
9726.
25. logD7.4 values were measured by potentiometric titration,
using a GlpKa instrument from Sirius Analytical Instru-
ments (Forest Row, UK).
26. logKIAM values were determined from HPLC retention
times, using a 10·3mm IAM fast screen mini column
(Regis speciality columns) with isocratic eluent;
1mLminꢀ1, 85% mixed phosphate pH7.4 buffer, 15%
acetonitrile and UV detection at 230nm.
27. Taillardat-Bertschinger, A.; Carrupt, P. A.; Barbato, F.;
Testa, B. J. Org. Chem. 2003, 46, 655.