Azole endothelin antagonists. 3. Using Δ log P as a tool to improve absorption

Article abstract of DOI:10.1021/jm9505932

The oral absorption profile of a family of azole-based ET(A)-selective antagonists has been improved through a rational series of structural modifications which were suggested by analysis of the physicochemical parameter Δ log P. Comparison of urea 2 with a series of well-absorbed compounds using A log P analysis suggested that 2 has an excess capacity for forming hydrogen bonds with solvent. A series of urea modifications were explored as a means of reducing H-bonding capacity while maintaining affinity for the ET(A)-receptor. The correlation between Δ log P values and absorption in an intraduodenal (id) bioavailability model was good; this strategy uncovered replacements for each of the urea NS groups which simultaneously improve both potency and drug absorption. A combination of these optimized modifications produces carbamate 16h, a highly-selective ET(A) antagonist with a potency/bioavailability profile consistent with an oral route of administration.

Full text of DOI:10.1021/jm9505932

982
J . Med. Chem. 1996, 39, 982-991
Azole En d oth elin An ta gon ists. 3. Usin g ∆ log P a s a Tool To Im p r ove
Absor p tion
Thomas W. von Geldern,* Daniel J . Hoffman,† J effrey A. Kester, Hugh N. Nellans,† Brian D. Dayton,
Samuel V. Calzadilla, Kennan C. Marsh,‡ Lisa Hernandez,‡ William Chiou, Douglas B. Dixon,
J inshyun R. Wu-Wong, and Terry J . Opgenorth
Aging and Degenerative Diseases Research, Gastrointestinal Pharmacology, and Drug Analysis Departments,
Pharmaceutical Products Research Division, Abbott Laboratories, Abbott Park, Illinois 60064
Received August 8, 1995X
The oral absorption profile of a family of azole-based ETA-selective antagonists has been
improved through a rational series of structural modifications which were suggested by analysis
of the physicochemical parameter ∆ log P. Comparison of urea 2 with a series of well-absorbed
compounds using ∆ log P analysis suggested that 2 has an excess capacity for forming hydrogen
bonds with solvent. A series of urea modifications were explored as a means of reducing
H-bonding capacity while maintaining affinity for the ETA-receptor. The correlation between
∆ log P values and absorption in an intraduodenal (id) bioavailability model was good; this
strategy uncovered replacements for each of the urea NH groups which simultaneously improve
both potency and drug absorption. A combination of these optimized modifications produces
carbamate 16h , a highly-selective ETA antagonist with a potency/bioavailability profile
consistent with an oral route of administration.
Sch em e 1. Evolution of Azole ET Antagonist Series
In tr od u ction
In the previous articles in this series,1,2 we have
described a strategy for modifying the backbone of the
potent and selective endothelin A-receptor (ETA) an-
tagonist FR-139317. It was our hope that such a
strategy would lead to the development of an agent with
a biochemical profile similar to that of the pseudotet-
rapeptide, but with substantially improved in vivo
stability and oral pharmacokinetics. In fact, beginning
with 1 as a lead, structure-activity studies led quickly
to compound 2, which retains the desired biochemical
properties of FR-139317 (Scheme 1). At this point we
considered it appropriate to evaluate our original hy-
pothesis that backbone modification might lead to
improved oral absorption.
We compared the absorption of 2 with that of FR-
139317 upon intraduodenal (id) injection into rats (this
model is discussed in detail in the Experimental Sec-
tion). The results (Figure 1) clearly indicate that our
hypothesis was correct; the absorption profile of 2 is
dramatically superior to that of the Fujisawa compound.
At the same time, however, it is equally clear that portal
blood levels achieved with 2 fall well short of what might
be desirable for an orally-deliverable agent. We thus
felt it necessary to embark on a structure-absorption
study, employing compound 2 as an initial lead struc-
ture.
P h ysicoch em ica l Cor r ela tion s w ith Absor p tion
In order to guide and to facilitate our search for
improvements in gut permeability within this family of
compounds, we hoped to identify a physicochemical
marker which could readily be determined for a large
series of compounds, and which might be used to predict
the absorption profile of a given class of analogs. There
is an extensive literature attempting such correlations.3
As a rule the parameter that has been employed in such
studies is log P, the octanol/water partitioning coef-
F igu r e 1. Drug levels upon id administration of azole 2 are
substantially improved over FR-139317, but further improve-
ments are still necessary.
† Gastrointestinal Pharmacology Department.
‡ Drug Analysis Department.
ficient, which is generally considered to be a measure
of hydrophobicity. However, attempts to correlate log
X Abstract published in Advance ACS Abstracts, J anuary 15, 1996.
0022-2623/96/1839-0982$12.00/0 © 1996 American Chemical Society

Azole Endothelin Antagonists. 3
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4 983
Sch em e 2. Removing H-Bonding Character
first-generation ET antagonist Ro46,20059 have both
been demonstrated to be highly orally bioavailable.
While the octanol-water partition coefficients for this
structurally diverse set of compounds span almost 3
orders of magnitude, the corresponding ∆ log P mea-
surements fall into a tighter range; more importantly,
it appears that regardless of structure these well-
absorbed species are all characterized by having ∆ log
P < 3. This result is consistent with previous reports.3
Under similar experimental conditions the ∆ log P of
compound 2 is 5.92. As ∆ log P values increase, the
amount of material present in the organic phase of the
cyclohexane/buffer partition becomes vanishingly small;
in practice a value of ∼6 represents the limit of our
ability to detect material using an HPLC-based assay.
In any event the message of this analysis is clear; it
appears that 2 has a greater ability to form hydrogen
bonds than would be expected for a well-absorbed
compound.
F igu r e 2. The world according to log P and ∆ log P.
F igu r e 3. Physicochemical comparison of 2 with benchmark
compounds.
P with absorption parameters have met with mixed
success; it appears that, while log P values may cor-
relate with absorption rates within a given structural
class, such correlations often fall apart when structures
change significantly, and as a rule have more descriptive
than predictive power.
On the basis of the synthetic strategy we have
developed for this class of compounds, two approaches
to decreasing H-bonding capacity (and thereby presum-
ably decreasing ∆ log P and improving absorption)
seemed particularly straightforward to examine (Scheme
2). First, we might expect that switching from an
imidazole nucleus to an oxazole would decrease ∆ log P
values. Unfortunately, our previous SAR studies sug-
gest that such a shift will also result in a significant
decrease in ETA receptor binding. Alternatively we
should be able to modify the urea moiety to substantially
decrease its H-bonding potential. To probe the effects
of these various modifications, we have synthesized a
series of analogs in which each urea NH is replaced by
three nondonating alternatives, NCH3, O, and CH2.
Additionally this set of seven modified ureas (including
the parent) has been prepared using both imidazole and
oxazole cores.
The failure of log P based absorption correlations may
reflect the fact that this parameter provides an inac-
curate view of compounds in the biologic milieu (see
Figure 2). In an aqueous environment the effective size
and polarity of a molecule depends not only upon its
intrinsic character, but also upon interactions with its
surroundings, for example through hydrogen-bonding.
These latter interactions may be particularly relevant
in describing a situation like intestinal transport, in
which desolvation is presumably a prerequisite to
crossing the membranes. For this reason we felt that
it might be more relevant to examine an alternative
physicochemical parameter, ∆ log P. The ∆ log P value,4
described by the formula
∆ log P ) log P(octanol/water) -
log P(cyclohexane/water)
Syn th esis
still serves as a measure of hydrophobicity. However,
by comparing the ability of a compound to partition into
two organic solvents, one of which does and one of which
does not have the ability to form hydrogen bonds, ∆ log
P analysis tends to emphasize the H-bonding capacity
of molecules. Since its initial description by Seiler in
1974,4 ∆ log P has been employed by Burton and his
co-workers3,5 to correlate absorption with substitution
in a family of tripeptides and has also been found to be
useful as an indicator of blood-brain barrier penetra-
tion.6 To the best of our knowledge, its power as a
predictor of oral absorption has never been investigated.
We began our physicochemical analysis by comparing
imidazole 2 with a number of compounds known to have
excellent pharmacokinetic profiles (Figure 3). Theo-
phylline is the prototype of a well-absorbed drug;7
similarly the Abbott AII antagonist A819888 and Roche’s
The compounds described here are prepared in a
manner analogous to that reported in our previous
articles.1,2 In particular the preparation of the hetero-
cyclic dipeptide surrogates (Scheme 3) remains un-
changed; compounds 15 and 16 differ only through
modification of the leucylurea and in the choice of an
azole core. The synthesis of the requisite modified ureas
is described in Scheme 4. The adduct of leucine benzyl
ester with 1,1′-carbonyldiimidazole (CDI) can be con-
densed with amines to produce substituted ureas like
8 after hydrogenolysis; alternatively this same starting
material may be condensed with a chloroformate (to
produce carbamates) or with an acid (to give amides).
Similar reaction of benzyl L-leucate with CDI gives an
intermediate which, after further activation with methyl
triflate, reacts rapidly with amines to provide “inverse”

984 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4
Sch em e 3.a Core Synthesis
von Geldern
a
(i) LiHMDS, THF, -78 °C; CH3COCl; (ii) HCl/H2O; (iii) P-D-Trp(Me)-OCOOiBu, THF, -20 °C; NMM (dropwise); (iv) NH4OAc, HOAc,
reflux; (v) PPh3, CCl4, pyr, CH3CN.
Sch em e 4.a Synthesis of N-Terminal Urea Replacements
a
(i) CDI, Et3N, THF; (ii) c-HexNH2 or c-HexNHCH3; (iii) c-HexOCOCl, Et3N, THF; (iv) c-HexCH2COOH, EDC, HOBt, NMM, THF/
DMF; (v) H2/10% Pd/C, EtOH; (vi) BnBr, Cs2CO3, DMF; (vii) CDI, THF; (viii) (a) MeOTf, THF, 0 °C, (b) c-HexNH2 or c-HexNHCH3 or
c-HexOH; (x) c-HexNH2 or c-HexNHCH3, EDC, HOBt, NMM, THF/DMF.
Sch em e 5.a Analog Assembly
a
(i) TFA; (ii) 30% HBr/HOAc; (iii) EDC, HOBt, NMM, THF/DMF; (iv) H2, 10% Pd/C, EtOH; (v) c-HexNCO, Et3N, THF/DMF.
carbamates 10. Finally, “inverse amide” derivatives 14
are readily assembled from acid 12.10
(see Scheme 5) and (after deprotection) forming the urea
as the penultimate step of the synthesis.
In general the final products can be assembled
through standard coupling protocols, followed by ester
hydrogenolysis (Scheme 5). In the case of analogs
containing an N-methylated leucine residue, activation
of the corresponding N-methylurea leads to rapid hy-
dantoin formation, so that no coupling occurs. This
difficulty may be circumvented by first coupling core
azole 4 or 5 with a protected N-methylleucine derivative
P h ysicoch em ica l An a lysis
We have recorded ∆ log P values for analogs 15 and
16 using a modification of the reported procedures.
Samples of test compounds are taken up in 0.05 M
PIPES buffer (pH 6.5), and their ability to partition into
an organic solvent is evaluated through HPLC analysis
of both phases after 2 h of thorough agitation. Octanol

Azole Endothelin Antagonists. 3
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4 985
is chosen as the typical organic solvent which is capable
of forming hydrogen bonds with a solute molecule, while
cyclohexane provides an organic medium without such
H-bonding capacity. After determination of the indi-
vidual log D values, ∆ log P is calculated using the
formula described above. For the purpose of these
calculations we may use log D and log P values
interchangeably; for acidic compounds these are related
by the formula
portal AUCs ranging from 4.9 to 20.8 µg min/mL. It is
clear, nonetheless, that the modification of the urea
moiety has led to improved intestinal uptake in this
series, as all of these values are significantly better than
that recorded for the parent imidazole 15a (0.26 µg min/
mL). Replacement of the imidazole with oxazole also
boosts absorption (viz. 15a vs 16a ), a trend that persists
throughout the series of urea modifications. In fact,
oxazoles 16b-g represent dramatic improvements over
15a , suggesting that it is this oxazole series that is likely
to provide a compound for further development. These
results are all consistent with our expectations on the
basis of analysis of ∆ log P values in the series.
Specifically, when ∆ log P analysis predicts a large
excess of H-bonding capacity, the compounds are poorly
absorbed; as ∆ log P values decrease, id absorption
improves. This relationship is depicted graphically in
Figure 4. It thus appears that, at least for the case at
hand, this new physicochemical parameter provides a
useful qualitative assessment of gut transport.
log P ) log D + log(1 + 10pH-pK
)
a
and the correction term, which is constant for a given
compound and pH, cancels out.
P h a r m a cok in etic Stu d ies
We have evaluated the intestinal absorption of com-
pounds 15a -g and 16a -i using the id rat model
developed previously to evaluate a series of renin
inhibitors.11 Briefly, blood samples are collected from
the portal vein of an anesthetized rat at several time
points after the test compound is injected into its
duodenum. Plasma drug levels are quantitated using
an HPLC-based assay after the samples have undergone
organic extraction. The individual data points may be
used to create a time course for drug absorption (as in
Figure 1); alternatively they may be used to estimate
the total drug absorption (“id AUC”) over the 1-h period
of the experiment. These portal drug levels provide an
indication of the ability of the analogs to cross from the
gut into the circulation.
A more detailed analysis of pharmacokinetic behavior
of 16h is accomplished through an oral dosing protocol
in rats. The plasma concentration profile of a test
compound administered at a dose of 10 mg/kg by gavage
is compared with an intravenous dose of the same
analog. Systemic blood samples, sequentially collected
from a tail vein of each animal over a 12-h period, are
analyzed by HPLC for drug levels after an extraction.
These plasma concentration values permit the estima-
tion of maximum blood levels (Cmax and Tmax), plasma
half-lives, and oral bioavailability (F).
An additional factor which is important to evaluate
in “optimizing” a urea replacement is the effect of NH
modification on the receptor affinity of the resultant
analogs. Binding data for compounds 15 and 16 are
included in Table 1. As anticipated from our previous
studies, compounds containing an imidazole core are
generally more potent than the corresponding oxazoles.
Pairwise comparison suggests that the imidazole:ox-
azole potency ratio varies from 0.5 (15c vs 16c) to 20
(15a vs 16a or 15e vs 16e), with a ratio of 5 being
typical. This result is somewhat unfortunate, since our
earlier results in the id absorption model suggest that
oxazoles will be preferred as the heterocyclic nucleus.
On the positive side, our urea substitution studies have
identified replacements that not only are tolerated by
the receptor but actually lead to improvements in ETA
binding affinity. For example, while replacement of the
first NH of the urea with a methylene group results in
a loss of potency, N-methylation leads to a substantial
improvement in ETA activity (compare 15b vs 15a or
16b vs 16a ). Substitution with oxygen at this position
is beneficial for the oxazole series, but detrimental for
the imidazoles. In a similar fashion we observe that
an oxygen atom is a beneficial replacement for the
second urea NH; on the other hand, NCH3 or CH2
substitution appears to produce a slight decrease in
binding affinity. Fortuitously, the effects of these
beneficial modifications appear to be enhanced in the
Resu lts a n d Discu ssion
Table 1 lists the ∆ log P values measured for analogs
15a -g and 16a -i. Several important trends are ap-
parent from these data. The “parent” imidazole 15a and
oxazole 16a both have ∆ log P values near 6, suggesting
not only that the presence of the intact urea is prob-
lematic, but that simply changing the nature of the
heterocycle is not enough to bring ∆ log P into the
desired range (nearer to a value of 3). Conversely, the
presence of an imidazole as the heterocyclic core seems
to present an insurmountable problem; imidazoles
15a -g all have high ∆ log P values regardless of the
modifications made to the urea group. However, in the
series of analogs 16a -g which have oxazole cores, it is
clear that changes in the urea which reduce H-bonding
character simultaneously produce lower ∆ log P values.
While these values are still outside of the range that
we consider to be optimal, we might expect, if our
predicted correlation holds, to see improvements in the
absorption profile for these compounds.
oxazole series, with 16b (IC50 ) 25 nM) and 16f (IC50
)
21 nM) each better than the parent urea by roughly a
factor of 10.
Through this multifactorial analysis we have been
able to identify a replacement for each of the potential
H-bond donors of the urea (NCH3 for the first, O for the
second) that is able simultaneously to improve the
efficiency of intestinal transport and increase affinity
for the ETA receptor. To further optimize these critical
parameters, we next prepared carbamate 16h which
contains both of the beneficial substitutions. While the
effects of the individual changes are not completely
additive, compound 16h is nonetheless more potent
(IC50 ) 4.6 nM) and better absorbed (id AUC ) 110 µg
min/mL) than any of the second-generation oxazoles.
More importantly, the pharmacokinetic profile of 16h
upon oral delivery (Figure 5) suggests that the com-
pound could be successfully administered by this route.
The results of blood level analysis of the same series
of analogs is summarized in the table. Once again,
several trends emerge. The family of imidazoles 15b-g
are characterized by poor to modest absorption, with

986 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4
Ta ble 1. Urea Replacements
von Geldern
Oral dosing of the compound at 10 mg/kg provides
modest but sustained drug concentrations in the blood;
comparison of oral and iv time course graphs provides
an estimated oral bioavailability of 13.5%. Compound
16i, another third-generation antagonist produced by
replacing each urea NH with an oxygen atom, is less
potent than 16h (IC50 ) 17 nM), but has similar id and
oral (not shown) pharmacokinetic profiles. Both 16h
and 16i have ∆ log P values < 4.
from several portions of the molecule. On the basis of
this analysis a series of modified analogs were prepared
and evaluated. Within this series a strong correlation
exists between ∆ log P values and drug absorption. By
combining absorption data with the results of receptor
binding analysis, we have been able to identify optimal
replacements for for both the heterocyclic core and the
urea moiety of 2. These optimized replacements may
be combined together to produce carbamate 16h , which
unlike any of its predecessors has a pharmacokinetic
profile that might be compatible with an oral dosing
regimen. While 16h is neither the most potent ET
antagonist reported to date nor the best absorbed, it is
to the best of our knowledge the most highly ETA-
selective compound to have a potency/bioavailability
profile consistent with an oral route of administration.
If our hypothesis regarding the importance of such
Con clu sion s
The physicochemical parameter ∆ log P has been
successfully employed as part of a strategy to improve
the pharmacokinetic profile of a series of azole-based
endothelin antagonists. A preliminary ∆ log P analysis
of compound 2 suggested a strategy for improving
absorption by removing hydrogen-bonding capability

Azole Endothelin Antagonists. 3
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4 987
Syn th esis of Acid ic P a r tn er s. N-((Cycloh exyla m in o)-
ca r bon yl)leu cin e (8, X ) NH). Leucine benzyl ester-p-
TsOH (100 mg) was dissolved in CHCl3 (2 mL). Et3N (51 mg,
75 µL) was added and the solution cooled to 0 °C in an ice
bath. Carbonyldiimidazole (41 mg) was added and the solution
stirred at 0 °C for 1 h. The bath was removed, and the solution
was stirred an additional 1 h at room temperature. Cyclo-
hexylamine (44 mg, 50 µL) was added and the solution stirred
overnight at room temperature. The solution was washed with
saturated NaHCO3 solution, 1 N H3PO4, and brine, dried with
MgSO4, and evaporated under reduced pressure to give a white
solid which was purified by flash chromatography on silica gel
eluting with 25% EtOAc-hexane to give N-((cyclohexylamino)-
carbonyl)leucine benzyl ester (75 mg, 88%). The ester was
dissolved in EtOH (5 mL), the solution was purged of oxygen,
10% Pd/C (0.10 g) was added, and the mixture was stirred
under hydrogen for 2 h. The solvent was removed in vacuo
and the residue taken up in EtOAc and filtered through Celite
to remove the catalyst. The solvent was evaporated in vacuo
to give the carboxylic acid as a white solid (50 mg, 90%).
N-((N-Cycloh exyl-N-m eth yla m in o)ca r bon yl)leu cin e (8,
X ) NCH3) was prepared as described above, substituting
N-methylcyclohexylamine.
F igu r e 4. Relationship between ∆ log P and absorption.
1-(Cycloh exyloxyca r bon yl)leu cin e (8, X ) O). Cyclo-
hexanol (0.53 mL) was dissolved in THF (10 mL) and the
mixture cooled to 0 °C. Phosgene (2.6 mL, 1.97 M in toluene)
was added and the solution stirred for 90 min. A solution of
Leu-OBn‚TsOH (1.97 g) and Et3N (0.8 mL) in THF (20 mL)
was added slowly over 5 min, followed by additional Et3N (0.8
mL). The mixture was allowed to warm to ambient temper-
ature over 2 h. The solvents were removed in vacuo, and the
residue was taken up in EtOAc and washed with 1:1 saturated
NaHCO3 solution/water, 1 N H3PO4, and brine. The organic
phase was dried over Na2SO4, filtered through Celite, and
concentrated to give a yellow oil, which was purified by flash
chromatography eluting with 6:1 hexanes-EtOAc. The prod-
uct was dissolved in ethanol (50 mL), 10% palladium on carbon
(100 mg) was added, and the mixture was purged with
nitrogen. The nitrogen line was exchanged for a balloon of
hydrogen, and the mixture was stirred at ambient temperature
for 4 h. The catalyst was removed by filtration through Celite.
The solvents were removed in vacuo to give the title compound
as a white solid.
F igu r e 5. Pharmacokinetic profile of compound 16h
(A-103739).
1-(Cycloh exyla cetyl)leu cin e (8, X ) CH2). Leucine ben-
zyl ester p-tosylate (1.97 g, 5.0 mmol) and cyclohexylacetic acid
(0.72 g, 1.0 equiv) were combined in a mixture of 10 mL of
THF and 5 mL of DMF; 1 mL of NMM was added, followed by
0.96 g of EDCI. The resultant solution was stirred at ambient
temperature for 15 h. The solvents were removed in vacuo;
the residue was taken up in EtOAc and washed with 1:1
saturated sodium bicarbonate solution-water, 1 N H3PO4, and
brine. The organic phase was dried over Na2SO4, filtered
through Celite, and concentrated to give a yellow oil. This
crude product was dissolved in ethanol (50 mL), 10% palladium
on carbon (100 mg) was added, and the mixture was purged
with nitrogen. The nitrogen line was exchanged for a balloon
of hydrogen, and the mixture was stirred at ambient temper-
ature for 4 h. The catalyst was removed by filtration through
Celite. The solvents were removed in vacuo to give the title
compound as a white solid.
(2S)-(N-Cycloh exylam in o)-2-car boxyisovaler ic acid (11,
X ) NH). A mixture of CsCO3 (10 g) and Na2CO3 (15 g) was
suspended in 80 mL of THF; L-leucic acid (10.0 g) was added,
and the mixture was stirred for 20 min. Benzyl bromide (9.5
mL, 1.05 equiv) was added, and the mixture was stirred for
15 h at ambient temperature. DMF (20 mL) was added, and
the mixture was heated at 40 °C for 3 h. The mixture was
taken up in 200 mL of water and extracted with 100 mL of
EtOAc. The organics were washed twice with water and then
once each with 1 N H3PO4 and brine. The organic extracts
were dried over Na2SO4, filtered through Celite, and concen-
trated in vacuo. A sample of the resultant ester (6.40 g) was
combined in 75 mL of THF with 4.63 g (1.0 equiv) of
carbonyldiimidazole. The solution was stirred at ambient
temperature for 6 h. The solvents were removed in vacuo; the
residue was taken up in EtOAc and then washed sequentially
with 0.2 N H3PO4 and brine. The organic layer was dried over
receptor selectivity proves to be correct, an agent with
this profile might offer significant advantages as a
therapeutic agent.
Exp er im en ta l Section
Unless otherwise specified, all solvents and reagents were
obtained from commercial suppliers and used without further
purification. THF was dried over sodium and purified by
distillation. All reactions were performed under nitrogen
atmosphere unless specifically noted. All final products are
analyzed for purity by analytical HPLC using a 25-cm Vydac
Protein and Peptide C18 column, and are >95% pure unless
otherwise stated. 1H-NMR spectra were recorded at 300 MHz;
all values are referenced to tetramethylsilane as internal
standard and are reported as shift (multiplicity, coupling
constants). Mass spectral analysis is accomplished using fast
atom bombardment (FAB-MS) or direct chemical ionization
(DCI-MS) techniques. All elemental analyses are consistent
with theoretical values to within (0.4% unless indicated.
Abbr evia tion s: CDI, 1,1′-carbonyldiimidazole; c-Hex, cy-
clohexyl; DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene; DMF, di-
methylformamide; EDCI, 1-ethyl-3-(3-(dimethylamino)propyl)-
carbodiimide hydrochloride; HOBt, 1-hydroxybenzotriazole
hydrate; NMM, N-methylmorpholine; PIPES, 1,4-piperazinebi-
s(ethanesulfonic acid); PPh3, triphenylphosphine; pyr, pyridine;
TFA, trifluoroacetic acid; THF, tetrahydrofuran; p-TsOH,
p-toluenesulfonic acid.
Syn th esis of Cor e Heter ocycles. Compounds 4 (P ) Cbz)
and 5 (P ) Boc) were prepared from the appropriate starting
materials using the procedures described in the first article
in this series.1

988 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4
von Geldern
Na2SO4, filtered through Celite, and evaporated in vacuo. A
sample of this material (400 mg) was dissolved in 10 mL of
THF and cooled to 0 °C. Methyl trifluoromethanesulfonate
(0.14 mL, 1.0 equiv) was added dropwise, and the solution was
allowed to stir for 30 min. Cyclohexylamine (0.3 mL) was
added, and the solution was allowed to warm to ambient
temperature and stir for 12 h. The solvents were removed in
vacuo; the residue was taken up in EtOAc and washed
sequentially with 1:1 saturated NaHCO3 solution/water, 1 N
H3PO4, and brine. The organic phase was dried over Na2SO4,
filtered through Celite, and concentrated to give a yellowish
oil. The crude product was purified by flash chromatography
on silica gel, eluting with 3:1 hexanes/ether, to give 300 mg of
a colorless oil. This material was dissolved in 30 mL of
ethanol; 80 mg of 10% palladium-on-carbon was added, and
the mixture was purged with nitrogen. The nitrogen line was
exchanged for a balloon of hydrogen, and the mixture was
stirred at ambient temperature for 4 h. The catalyst was
removed by filtration through a pad of Celite; the solvents were
removed in vacuo to provide 210 mg (65% overall yield) of the
title compound.
appropriate fraction was lyophilized to give the product as a
white solid (76 mg, 62%): 1H NMR (CD3OD, 300 MHz) of major
tautomer δ 0.75 (d, 3H, J ) 7 Hz), 0.78 (d, 3H, J ) 7 Hz),
1.1-1.4 (m, 8H), 1.5-1.9 (m, 6H), 2.53 (s, 3H), 3.33 (m, 1H),
3.50 (m, 2H), 3.77 (s, 3H), 4.03 (t, 1H, J ) 8 Hz), 5.40 (dd, 1H,
J ) 6, 10 Hz), 7.04 (ddd, 1H, J ) 1, 7, 8 Hz), 7.06 (s, 1H), 7.18
(ddd, 1H, J ) 1, 7, 8 Hz), 7.35 (d, 1H, J ) 8 Hz), 7.47 (d, 1H,
J ) 8 Hz); MS (FAB/NBA) m/ e 599 (M + Cu)+. Anal. for
C29H40N6O4‚1.7TFA: C, H, N.
2-{(1R)-1-[N-((N-Cycloh exyl-N-m eth ylam in o)car bon yl)-
leu cyla m in o]-2-(1-m eth ylin d ol-3-yl)eth yl}-5-m eth ylim i-
d a zole-4-ca r boxylic a cid (15b) was prepared according to
the above procedures, substituting N-((N-cyclohexyl-N-methyl-
amino)carbonyl)leucine. The crude product was triturated
with 50% diethyl ether/hexane, dissolved in acetonitrile and
water, and then lyophilized to give a white powder (26 mg):
1H NMR (CD3OD, 300 MHz) of major tautomer δ 0.72 (d, 3H,
J ) 7 Hz), 0.75 (d, 3H, J ) 7 Hz), 1.1-1.5 (m, 8H), 1.5-1.8
(m, 5H), 2.48 (s, 3H), 2.72 (s, 3H), 3.18 (m, 1H), 3.58 (m, 2H),
3.72 (s, 3H), 4.05 (t, 1H, J ) 8 Hz), 5.36 (dd, 1H, J ) 6, 10
Hz), 6.95 (s, 1H), 7.0 (t, 1H, J ) 8 Hz), 7.15 (t, 1H, J ) 8 Hz),
7.30 (d, 1H, J ) 8 Hz), 7.54 (d, 1H, J ) 8 Hz); MS (FAB/NBA)
m/ e 551 (M + H)+. Anal. for C30H42N6O4‚2 H2O: C, H, N.
2-{(1R)-1-[N-(Cycloh exyloxyca r b on yl)leu cyla m in o]-
2-(1-m eth ylin dol-3-yl)eth yl}-5-m eth ylim idazole-4-car box-
ylic a cid (15c) was prepared according to the above proce-
dures, substituting 1-(cyclohexyloxycarbonyl)leucine. The crude
material was purified by trituration with ether-EtOAc to give
a white solid which was dissolved in 0.1% aqueous TFA in
acetonitrile and lyophilized to give a white powder (106 mg):
1H NMR (CD3OD, 300 MHz) of major tautomer δ 0.78 (d, 3H,
J ) 6 Hz), 0.81 (d, 3H, J ) 6 Hz), 1.2-1.9 (m, 13H), 2.45 (s,
3H), 3.3-3.5 (m, 2H), 3.76 (s, 3H), 4.02 (m, 1H), 4.59 (m, 1H),
5.23 (t, 1H, J ) 8 Hz), 7.03 (dt, 1H, J ) 1, 7 Hz), 7.04 (s, 1H),
7.17 (dt, 1H, J ) 1, 7 Hz), 7.33 (d, 1H, J ) 8 Hz), 7.45 (d, 1H,
J ) 8 Hz); MS (FAB/NBA) m/ e 538 (M + H)+, 560 (M + Na)+.
Anal. for C29H39N5O5‚1.3TFA: C, H, N.
2-{(1R)-1-[N-(Cycloh exyla cetyl)leu cyla m in o]-2-(1-m e-
th ylin dol-3-yl)eth yl}-5-m eth ylim idazole-4-car boxylic acid
(15d ) was prepared according to the above procedures, sub-
stituting 1-(cyclohexylacetyl)leucine. The final product was
dissolved in 0.1% aqueous TFA-acetonitrile and lyophilized
to give the title compound as a white powder (11 mg): 1H NMR
(CD3OD, 300 MHz) of the major tautomer δ 0.76 (d, 3H, J )
7 Hz), 0.79 (d, 3H, J ) 7 Hz), 0.95 (m, 2H), 1.1-1.5 (m, 5H),
1.7 (m, 7H), 2.07 (d, 2H, J ) 8 Hz), 2.51 (s, 3H), 3.35 (m, 2H),
3.77 (s, 3H), 4.19 (dd, 1H, J ) 6, 7 Hz), 5.35 (dd, 1H, J ) 6, 8
Hz), 7.02 (m, 1H), 7.05 (s, 1H), 7.16 (m, 1H), 7.35 (d, 1H, J )
7 Hz), 7.45 (d, 1H, J ) 7 Hz); MS (FAB/NBA) m/ e 536 (M +
H)+. Anal. for C30H41N5O4‚1.7TFA: C, H, N.
2-{(1R)-1-(N-((2S)-((N-Cycloh exyla m in o)-2-ca r boxyis-
ovaler yl)am in o)-2-(1-m eth ylin dol-3-yl)eth yl}-5-m eth ylim -
id a zole-4-ca r boxylic a cid (15f) was prepared according to
the above procedures, substituting (2S)-(N-cyclohexylamino-
2-carboxyisovaleric acid. The crude material was purified by
trituration with ether; the resultant material was dissolved
in 0.1% aqueous TFA-acetonitrile and lyophilized to give a
white powder (43.5 mg): 1H NMR (CD3OD, 300 MHz) δ 0.85
(d, 6H, J ) 7 Hz), 1.1-1.9 (m, 13H), 2.50 (s, 3H), 3.3-3.5 (m,
4H), 3.76 (s, 3H), 5.34 (dd, 1H, J ) 7, 8 Hz), 7.03 (s, 1H), 7.04
(dt, 1H, J ) 1, 8 Hz), 7.17 (ddd, 1H, J ) 1, 7, 8 Hz), 7.34 (d,
1H, J ) 8 Hz), 7.43 (d, 1H, J ) 8 Hz); MS (DCI/NH3) m/ e 538
(M + H)+. Anal. for C29H39N5O5‚1.2TFA: C, H, N.
2-{(1R)-1-(N-((2R)-2-(((N-Cycloh exyla m in o)ca r bon yl)-
m eth yl)isova ler yl)a m in o)-2-(1-m eth ylin d ol-3-yl)eth yl}-5-
m eth ylim id a zole-4-ca r boxylic a cid (15g) was prepared
according to above procedures, substituting N-cyclohexyl-(3R)-
3-carboxy-5-methylhexanamide. The crude material was tritu-
rated with ether; the resultant material was purified by HPLC,
using a gradient of 0 f 80% acetonitrile in 0.1% aqueous TFA
to elute the product as a white solid (17 mg): 1H NMR (CD3-
OD, 300 MHz) δ 0.59 (d, 3H, J ) 6 Hz), 0.61 (d, 3H, J ) 6 Hz),
0.79 (m, 1H), 0.96 (ddd, 1H, J ) 5, 10, 14 Hz), 1.1-1.4 (m,
6H), 1.6-1.9 (m, 5H), 2.25 (dd, 1H, J ) 3, 15 Hz), 2.50 (dd,
1H, J ) 14, 15 Hz), 2.56 (s, 3H), 2.63 (m, 1H), 3.2-3.4 (m,
2H), 3.65 (m, 1H), 3.77 (s, 3H), 5.52 (dd, 1H, J ) 5, 11 Hz),
7.04 (ddd, 1H, J ) 1, 7, 8 Hz), 7.08 (s, 1H), 7.18 (ddd, 1H, J )
(2S)-(N-Cycloh exyl-N-m et h yla m in o)-2-ca r b oxyisova -
ler ic a cid (11, X ) NCH3) was prepared as described above,
substituting N-methylcyclohexylamine.
(2S)-(Cycloh exyloxy)-2-ca r boxyisova ler ic a cid (11, X
) O) was prepared as described above, substituting cyclohex-
anol.
N-Cycloh exyl-(3R)-3-car boxy-5-m eth ylh exan am ide (14,
X ) NH). To a solution of (3R)-3-(benzyloxycarbonyl)-5-
methylhexanoic acid (464 mg; prepared by the method of
Plattner et al. J . Med. Chem. 1988, 31, 2277-88, but
substituting isocaproyl chloride for 3-phenylpropionyl chloride)
and 0.23 mL (1 equiv) of cyclohexylamine in 20 mL of 3:1 THF/
DMF were added sequentially 270 mg (1 equiv) of HOBt, 1.2
mL of NMM, and 384 mg (1 equiv) of EDC. The resultant
solution was stirred at ambient temperature for 15 h; the
solvents were removed in vacuo, and the residue was taken
up in EtOAc and washed sequentially with 1:1 saturated
NaHCO3 solution-water, 1 N H3PO4 and brine. The organic
phase was dried over Na2SO4, filtered through Celite, and
concentrated in vacuo. The crude ester was dissolved in 50
mL of ethanol; 100 mg of 10% palladium-on-carbon was added,
and the mixture was purged with nitrogen. The nitrogen line
was exchanged for a balloon of hydrogen, and the mixture was
stirred at ambient temperature for 4 h. The catalyst was
removed by filtration through a pad of Celite; the solvents were
removed in vacuo to provide the title compound.
Assem bly of An a logs: 2-{(1R)-1-[N-((Cycloh exyla m i-
n o)ca r bon yl)leu cyla m in o]-2-(1-m eth yl-in d ol-3-yl)eth yl}-
5-m eth ylim id a zole-4-ca r boxylic Acid (15a ). Compound 4
(P ) Cbz, R ) Bn; 100 mg) was suspended in 1.5 mL of 30%
HBr-HOAc. The resulting solution was stirred at ambient
temperature for 60 min. The solvents were removed in vacuo;
the residue was taken up in saturated NaHCO3 solution and
extracted with EtOAc. The organic extracts were dried over
Na2SO4, filtered through Celite, and concentrated in vacuo.
This crude amine was dissolved in THF (10 mL). HOBt (36
mg), N-((cyclohexylamino)carbonyl)leucine (64 mg), and EDC
(50 mg) were added. N-Methylmorpholine (20 µL) and DMF
(1.5 mL) were added, and the mixture was stirred at room
temperature for 18 h. The solvent was evaporated under
reduced pressure and the residue taken up in EtOAc. The
solution was washed with saturated NaHCO3 solution, 1 N
H3PO4, and brine, dried with MgSO4, and evaporated in vacuo
to give a yellow oil which was purified by flash chromatography
on silica gel eluting with 50% EtOAc-hexane. This benzyl
ester was combined in 10 mL of ethanol with 30 mg of 10%
palladium on carbon. The flask was fitted with a three-way
stopcock connected to a hydrogen-filled balloon and a nitrogen/
vacuum manifold. The flask was evacuated, filled with
nitrogen, evacuated again, and then put under a hydrogen
atmosphere. The mixture was stirred at ambient temperature
for 14 h. The hydrogen was evacuated and the flask filled with
nitrogen. The catalyst was removed by filtration through a
pad of Celite and the solvent removed in vacuo. The crude
product was purified by preparative HPLC (Vydac µC18)
eluting with a 10-70% gradient of CH3CN in 0.1% TFA. The

Azole Endothelin Antagonists. 3
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4 989
1, 7, 8 Hz), 7.35 (d, 1H, J ) 8 Hz), 7.55 (d, 1H, J ) 8 Hz); MS
(DCI/NH3) m/ e 536 (M + H)+. Anal. for C30H41N5O4‚3.0TFA:
C, H, N.
7.48 (d, 1H, J ) 8 Hz); MS (FAB/NBA) m/ e 538 (M + H)+,
560 (M + Na)+, 576 (M + K)+. Anal. for C29H39N5O5‚0.6TFA:
C, H, N.
2-{(1R)-1-[(N-((N-Cycloh exyla m in o)ca r bon yl)-N-m eth -
ylleu cyl)a m in o]-2-(1-m eth ylin d ol-3-yl)eth yl}-5-m eth ylim -
id a zole-4-ca r boxylic Acid (15e). Compound 4 (P ) Cbz, R
) Bn; 80 mg) was suspended in 1 mL of 30% HBr-HOAc. The
resulting solution was stirred at ambient temperature for 60
min. The solvents were removed in vacuo; the residue was
taken up in saturated NaHCO3 solution and extracted with
EtOAc. The organic extracts were dried over Na2SO4, filtered
through Celite, and concentrated in vacuo. This crude amine
was dissolved in THF (4 mL) and DMF (2 mL). HOBt (42 mg),
N-Boc-N-methylleucine13 (64 mg), and EDC (57 mg) were
added, followed by N-methylmorpholine (8 drops), and the
mixture was stirred at ambient temperature for 18 h. The
solvent was evaporated, and the residue was taken up in
EtOAc, washed with saturated sodium bicarbonate solution,
1 N H3PO4, and brine, and evaporated to give a yellow oil. This
crude product was dissolved in TFA (5 mL). The solution was
stirred at ambient temperature for 2 h. The solvent was
evaporated and the residue taken up in saturated sodium
bicarbonate solution and extracted with EtOAc. The organic
layer was washed with brine and evaporated. The crude
product was dissolved in THF (5 mL). N-Methylmorpholine
(0.2 mL) and cyclohexyl isocyanate (5 drops) were added, and
the solution was stirred at ambient temperature for 18 h. The
solvent was evaporated, and the residue was taken up in
EtOAc, washed with saturated sodium bicarbonate solution,
1 N H3PO4, and brine, and evaporated to give a yellow oil
which was dissolved in EtOH (20 mL). Next, 10% palladium
on carbon (30 mg) was added, and the mixture was purged
with nitrogen. The nitrogen line was exchanged for a balloon
of hydrogen, and the mixture was stirred at ambient temper-
ature for 4 h. The catalyst was removed by filtration through
Celite, and the solvents were evaporated. The crude product
was triturated with 1:1 EtOAc-ether to give a white solid
which was dissolved in 0.1% aqueous TFA in acetonitrile and
lyophilized to give the title compound as a white powder (30
mg): 1H NMR (CD3OD, 300 MHz) of major tautomer δ 0.86
(d, 3H, J ) 7 Hz), 0.89 (d, 3H, J ) 7 Hz), 1.0-1.8 (m, 13H),
2.48 (s, 3H), 2.72 (s, 3H), 3.3-3.5 (m, 3H), 3.75 (s, 3H), 4.51
(dd, 1H, J ) 7, 10 Hz), 5.29 (t, 1H, J ) 7 Hz), 7.02 (s, 1H),
7.04 (dt, 1H, J ) 1, 7 Hz), 7.18 (dt, 1H, J ) 1, 7 Hz), 7.35 (d,
2H, J ) 8 Hz); MS (DCI/NH3) m/ e 551 (M + H)+; HRMS calcd
for C30H43N6O4 551.3346, found 551.3351.
2-{(1R)-1-[N-((Cycloh exylam in o)car bon yl)leu cylam in o]-
2-(1-m et h ylin d ol-3-yl)et h yl}-5-m et h yloxa zole-4-ca r box-
ylic Acid (16a ). Compound 5 above (120 mg) was dissolved
in 6 mL of trifluoroacetic acid and allowed to stir at ambient
temperature for 1 h. The solvents were removed in vacuo, the
residue was neutralized with bicarbonate solution, and the
mixture was extracted with EtOAc. The combined organic
extracts were washed with brine, dried over Na2SO4, and
concentrated under reduced pressure. The crude product was
dissolved in THF (2 mL). HOBt (42 mg), N-((cyclohexylamino)-
carbonyl)leucine, and EDC (57 mg) were added. N-Methyl-
morpholine (100 µL) was added and the mixture stirred at
room temperature for 18 h. The solvent was evaporated under
reduced pressure and the residue taken up in EtOAc. The
solution was washed with saturated NaHCO3 solution, 1 N
H3PO4, and brine, dried with MgSO4, and evaporated in vacuo
to give an orange oil which was purified by flash chromatog-
raphy on silica gel eluting with 50% EtOAc-hexane. The
resultant ester was dissolved in 30 mL of EtOH, 50 mg of 10%
palladium on carbon was added, and the mixture was purged
with nitrogen. The nitrogen line was exchanged for a balloon
of hydrogen, and the mixture was stirred at ambient temper-
ature for 4 h. The catalyst was removed by filtration through
a pad of Celite; the solvents were removed in vacuo. The crude
material was triturated with ether-hexanes, dissolved in 0.1%
aqueous TFA-acetonitrile, and lyophilized to give the title
compound as a white powder (96 mg): 1H NMR (CDCl3, 300
MHz) δ 0.70 (d, 3H, J ) 7 Hz), 0.74 (d, 3H, J ) 7 Hz), 1.1-1.9
(m, 13H), 2.50 (s, 3H), 3.40 (m, 1H), 3.76 (s, 3H), 3.90 (m, 1H),
4.35 (dd, 1H, J ) 6, 7 Hz), 5.35(m, 1H), 6.91 (s, 1H), 7.02 (t,
1H, J ) 8 Hz), 7.13 (t, 1H, J ) 8 Hz), 7.32 (d, 1H, J ) 8 Hz),
2-{(1R)-1-[N-(N-Meth yl-N-cycloh exyla m in o)ca r bon yl)-
leu cyla m in o]-2-(1-m et h ylin d ol-3-yl)et h yl}-5-m et h ylox-
a zole-4-ca r boxylic a cid (16b) was prepared according to the
above procedure, substituting N-((N-cyclohexyl-N-methylami-
no)carbonyl)leucine. The crude product was triturated with
diethyl ether-hexanes, dissolved in acetonitrile and water, and
lyophilized to give the product as a white powder (72 mg): 1H
NMR (CD3OD, 300 MHz) δ 0.71 (d, 3H, J ) 7 Hz), 0.73 (d,
3H, J ) 7 Hz), 1.1-1.8 (m, 13H), 2.54 (s, 3H), 2.71 (s, 3H),
3.45 (m, 1H), 3.72 (s, 3H), 3.90 (m, 1H), 4.32 (dd, 1H, J ) 6, 7
Hz), 5.38 (m, 1H), 6.92 (s, 1H), 7.0 (t, 1H, J ) 8 Hz), 7.13 (t,
1H, J ) 8 Hz), 7.30 (d, 1H, J ) 8 Hz), 7.52 (d, 1H, J ) 8 Hz);
MS (FAB/NBA) m/ e 552 (M + H)+. Anal. for C30H41
N5O5‚H2O: C, H, N.
-
2-{(1R)-1-[N-(Cycloh exyloxyca r b on yl)leu cyla m in o]-
2-(1-m et h ylin d ol-3-yl)et h yl}-5-m et h yloxa zole-4-ca r b ox-
ylic Acid (16c). The title compound was prepared following
the procedures described above, substituting 1-(cyclohexyloxy-
carbonyl)leucine. The crude final product was purified by
trituration with ether-hexanes; the resultant material was
dissolved in 0.1% aqueous TFA-acetonitrile and lyophilized
to give the title compound as a white powder (60 mg): 1H NMR
(CD3OD, 300 MHz) δ 0.83 (d, 6H, J ) 7 Hz), 1.3-1.9 (m, 13H),
2.54 (s, 3H), 3.3-3.5 (m, 2H), 3.73 (s, 3H), 4.10 (m, 1H), 4.51
(m, 1H), 5.38 (dd, 1H, J ) 7, 8 Hz), 6.96 (s, 1H), 7.01 (t, 1H, J
) 7 Hz), 7.13 (dt, 1H, J ) 1, 7 Hz), 7.29 (d, 1H, J ) 8 Hz),
7.45 (d, 1H, J ) 8 Hz); MS (FAB/NBA) m/ e 539 (M + H)+,
561 (M + Na)+. Anal. for C29H38N4O6‚0.3TFA: C, H, N.
2-{(1R)-1-[N-(Cycloh exyla cetyl)leu cyla m in o]-2-(1-m e-
th ylin d ol-3-yl)eth yl}-5-m eth yloxa zole-4-ca r boxylic a cid
(16d ) was prepared according to the above procedures, sub-
stituting 1-(cyclohexylacetyl)leucine. The final product was
dissolved in 0.1% aqueous TFA-acetonitrile and lyophilized
to give the title compound as a white powder (33 mg): 1H NMR
(CD3OD, 300 MHz) δ 0.79 (d, 3H, J ) 7 Hz), 0.82 (d, 3H, J )
7 Hz), 0.9 (m, 3H), 1.1-1.5 (m, 6H), 1.65 (m, 5H), 2.04 (d, 2H,
J ) 8 Hz), 2.53 (s, 3H), 3.35 (m, 2H), 3.72 (s, 3H), 4.37 (dd,
1H, J ) 6, 9 Hz), 5.40 (dd, 1H, J ) 7, 8 Hz), 6.96 (s, 1H), 7.02
(ddd, 1H, J ) 1, 7, 8 Hz), 7.14 (ddd, 1H, J ) 1, 7, 8 Hz), 7.30
(d, 1H, J ) 7 Hz), 7.45 (d, 1H, J ) 7 Hz); MS (DCI/NH3) m/ e
537 (M
N4O5‚0.5TFA: C, H; N: calcd, 9.44; found, 9.02.
+ + -
H)+, 554 (M NH4)+. Anal. for C30H40
2-{(1R)-1-[N-((2S)-(N-Cycloh exyla m in o)-2-ca r boxyiso-
va ler yl)a m in o]-2-(1-m eth ylin d ol-3-yl)eth yl}-5-m eth ylox-
a zole-4-ca r boxylic a cid (16f) was prepared according to the
above procedures, substituting (2S)-(N-cyclohexylamino)-2-
carboxyisovaleric acid. The crude material was purified by
trituration with 2:1 hexanes/ether; the resultant material was
dissolved in 0.1% aqueous TFA-acetonitrile and lyophilized
to give a white powder (103 mg): 1H NMR (CD3OD, 300 MHz)
δ 0.86 (d, 3H, J ) 7 Hz), 0.88 (d, 3H, J ) 7 Hz), 1.0-1.9 (m,
13H), 2.55 (s, 3H), 3.3-3.5 (m, 2H), 3.73 (s, 3H), 4.92 (dd, 1H,
J ) 5, 10 Hz), 5.38 (dd, 1H, J ) 7, 8 Hz), 6.93 (s, 1H), 7.00
(ddd, 1H, J ) 1, 7, 8 Hz), 7.13 (dt, 1H, J ) 1, 7 Hz), 7.28 (d,
1H, J ) 8 Hz), 7.44 (d, 1H, J ) 8 Hz); MS (FAB/NBA) m/ e
539 (M + H)+, 561 (M + Na)+. Anal. for C29H38N4O6‚0.3TFA:
C, H, N.
2-{(1R)-1-[N-((2R)-2-(((N-Cycloh exyla m in o)ca r bon yl)-
m eth yl)isova ler yl)a m in o]-2-(1-m eth ylin d ol-3-yl)eth yl}-5-
m eth yloxa zole-4-ca r boxylic a cid (16g) was prepared ac-
cording to above procedures, substituting N-cyclohexyl-(3R)-
3-carboxy-5-methylhexanamide. The crude material was
purified by trituration with hexanes-ether; the resultant
material was dissolved in 0.1% aqueous TFA-acetonitrile and
lyophilized to give a white powder (76 mg): 1H NMR (CD3-
OD, 300 MHz) δ 0.70 (d, 6H, J ) 7 Hz), 1.0-1.4 (m, 8H), 1.5-
1.8 (m, 5H), 2.12 (dd, 1H, J ) 7, 14 Hz), 2.31 (dd, 1H, J ) 8,
14 Hz), 2.54 (s, 3H), 2.71 (m, 1H), 3.2-3.4 (m, 2H), 3.50 (m,
1H), 3.73 (s, 3H), 5.45 (dd, 1H, J ) 7, 9 Hz), 6.99 (s, 1H), 7.02
(dt, 1H, J ) 1, 7 Hz), 7.13 (dt, 1H, J ) 1, 7 Hz), 7.29 (d, 1H,
J ) 8 Hz), 7.52 (d, 1H, J ) 8 Hz); MS (DCI/NH3) m/ e 537 (M
+ H)+, 554 (M + NH4)+. Anal. for C30H40N4O5‚TFA‚1.5H2O:
C, H, N.

990 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4
von Geldern
2-{(1R)-1-[N-((2S)-(N-Cycloh exyl-N-m et h yla m in o)-2-
ca r boxyisova ler yl)a m in o]-2-(1-m eth ylin d ol-3-yl)eth yl}-
5-m eth yloxa zole-4-ca r boxylic a cid (16h ) was prepared
according to above procedures, substituting (2S)-2-((N-cyclo-
hexyl-N-methylamino)carboxy)isovaleric acid. The crude final
product was triturated with hexanes-ether; the resultant
material was dissolved in 0.1% aqueous TFA-acetonitrile and
lyophilized to give a white powder (62 mg): 1H NMR (CD3-
OD, 300 MHz) δ 0.85 (d, 6H, J ) 7 Hz), 1.0-1.8 (m, 13H),
2.55 (s, 3H), 2.75 (bd s, 3H), 3.32 (m, 1H), 3.43 (dd, 1H, J ) 7,
15 Hz), 3.74 (s, 3H), 3.84 (m, 1H), 4.89 (dd, 1H, J ) 4, 10 Hz),
5.20 (dd, 1H, J ) 7, 8 Hz), 6.94 (s, 1H), 7.02 (ddd, 1H, J ) 1,
7, 8 Hz), 7.13 (dt, 1H, J ) 1, 7 Hz), 7.29 (d, 1H, J ) 8 Hz),
7.47 (d, 1H, J ) 8 Hz); MS (DCI/NH3) m/ e 553 (M + H)+, 570
(M + NH4)+. Anal. for C30H40N4O6‚0.9TFA: C, H, N.
2-{(1R )-1-[N -((2S )-2-((Cycloh e xyloxyca r b on yl)oxy)-
isova ler yl)a m in o]-2-(1-m et h ylin d ol-3-yl)et h yl}-5-m et h -
yloxa zole-4-ca r boxylic a cid (16i) was prepared according
to above procedures, substituting (2S)-2-((cyclohexyloxycar-
bonyloxy)isovaleric acid. The crude final product was tritu-
rated with hexanes-ether; the resultant material was dis-
solved in 0.1% aqueous TFA-acetonitrile and lyophilized to
give a white powder (75 mg): 1H NMR (CD3OD, 300 MHz) δ
0.84 (d, 3H, J ) 6 Hz), 0.85 (d, 3H, J ) 6 Hz), 1.0-1.8 (m,
13H), 2.55 (s, 3H), 3.32 (m, 1H), 3.43 (dd, 1H, J ) 7, 15 Hz),
3.74 (s, 3H), 4.47 (m, 1H), 4.83 (dd, 1H, J ) 5, 9 Hz), 5.20 (dd,
1H, J ) 7, 9 Hz), 6.96 (s, 1H), 7.02 (ddd, 1H, J ) 1, 7, 8 Hz),
7.14 (dt, 1H, J ) 1, 7 Hz), 7.30 (d, 1H, J ) 8 Hz), 7.48 (d, 1H,
J ) 8 Hz); MS (DCI/NH3) m/ e 540 (M + H)+, 557 (M + NH4)+.
Anal. for C29H37N3O7‚0.3TFA: C, H, N.
cells, using similar protocols. The results are qualitatively
similar, as indicated by the following data:
compound IC50, µM (hETA in CHO) IC50, µM (hETB in CHO)
16h
16i
0.0053 (n ) 2)
0.029 (n ) 2)
>100 (n ) 2)
>100 (n ) 2)
P h ysicoch em ica l Deter m in a tion s. All partition coef-
ficients were measured using standard shake-flask techniques
at 37 °C. A sample of test compound (∼1 mg) was dissolved
in 4.5 mL of 0.05 M PIPES buffer, pH 6.5. A 1-mL sample of
the resultant solution was transferred to a screw cap test tube
along with an equal volume of the appropriate organic phase;
the samples were capped and shaken for 2 h on a reciprocating
shaker at 100 cpm. Model organic phases were octanol (H-
bonding) and cyclohexane (non-H-bonding). After centrifuga-
tion to separate the phases, the concentrations of a test
compound in both organic and aqueous phases were deter-
mined by HPLC analysis. All values are the mean of at least
two determinations.
Ra t id Absor p tion Mod el. Male Sprague-Dawley rats
weighing approximately 225 g were fasted overnight. After
anesthetization with inactin (100 mg/kg, ip), PE50 catheters
were surgically implanted into the carotid artery and the
portal vein, and the pyloric sphincter was ligated. The test
compounds were dosed (10 mg/kg) by injection into the
duodenum (t ) 0), and portal and carotid blood samples were
collected at 10, 30, and 60 min after dosing. After the protocol
was complete, the animals were sacrificed by anesthesia
overdose.
2-{(1R)-1-[N-(N-((Cycloh exyla m in o)ca r bon yl)-N-Meth -
ylleu cyl)a m in o]-2-(1-m eth ylin d ol-3-yl)eth yl}-5-m eth ylox-
a zole-4-ca r boxylic Acid (16e). Compound 5 above (120 mg)
was dissolved in 6 mL of trifluoroacetic acid and allowed to
stir at ambient temperature for 1 h. The solvents were
removed in vacuo, the residue was neutralized with saturated
NaHCO3 solution, and the mixture was extracted with EtOAc.
The combined organic extracts were washed with brine, dried
over Na2SO4, and concentrated under reduced pressure. The
crude amine was dissolved in THF (2 mL). HOBt (42 mg),
N-Boc-N-methylleucine13 (80 mg), and EDC (57 mg) were
added, followed by N-methylmorpholine (100 µL), and the
mixture was stirred at ambient temperature for 15 h. The
solvent was evaporated, and the residue was taken up in
EtOAc, washed with saturated sodium bicarbonate solution,
1 N H3PO4, and brine, and evaporated to give a yellow oil. This
crude product was dissolved in EtOH (20 mL). Next, 10%
palladium on carbon (30 mg) was added, and the mixture was
purged with nitrogen. The nitrogen line was exchanged for a
balloon of hydrogen, and the mixture was stirred at ambient
temperature for 4 h. The catalyst was removed by filtration
through Celite, and the solvents were evaporated. The result-
ant acid was dissolved in TFA (6 mL). The solution was stirred
at ambient temperature for 1 h. The solvent was removed in
vacuo; the residue was taken up in toluene and concentrated
in vacuo. The resultant product was dissolved in DMF (5 mL).
N-Methylmorpholine (0.2 mL) and cyclohexyl isocyanate (5
drops) were added, and the solution was stirred at ambient
temperature for 15 h. The solvent was evaporated, and the
residue was taken up in EtOAc, washed with 1 N H3PO4, and
brine, and concentrated in vacuo. The crude product was
triturated with 1:1 hexanes-ether to give a white solid which
was dissolved in 0.1% aqueous TFA in acetonitrile and
lyophilized to give the title compound as a white powder (122
mg): 1H NMR (CD3OD, 300 MHz) of major tautomer δ 0.85
(d, 3H, J ) 7 Hz), 0.87 (d, 3H, J ) 7 Hz), 1.0-1.9 (m, 13H),
2.50 (s, 3H), 2.66 (s, 3H), 3.3-3.4 (m, 2H), 3.47 (m, 1H), 3.74
(s, 3H), 4.78 (dd, 1H, J ) 7, 9 Hz), 5.32 (dd, 1H, J ) 6, 8 Hz),
6.99 (s, 1H), 7.00 (dt, 1H, J ) 1, 7 Hz), 7.13 (dt, 1H, J ) 1, 7
Hz), 7.29 (d, 1H, J ) 8 Hz), 7.42 (d, 1H, J ) 8 Hz); MS (DCI/
Plasma samples (200 µL) and an equal volume of an internal
standard were extracted with 5 mL of dichloromethane-
ethanol (8:2). The organic phase was transferred to a 10 mL
conical centrifuge tube, and the solvent was evaporated using
a dry air stream. Samples were reconstituted in mobile phase.
Recovery of samples after extraction varied between 40 and
80%. Samples were analyzed by HPLC using a Regis Little
Champ column (50 × 4.6 mm id Spherisorb S3ODSII, 3 µm).
Chromatography was performed using a 10 min linear gradi-
ent usually consisting of 38-48% acetonitrile, 5% methanol,
and 57-47% 10 mM tetramethylammonium perchlorate(aq)
containing 0.1% trifluoroacetic acid. UV detection of analyte
was measured at 226 nm. Drug concentrations (µg/mL) were
calculated from standard curves formulated in plasma from
0.01 to 50 µg/mL in triplicate. Portal blood drug values were
used to calculate area under the curve (“id AUC”) using the
trapezoidal rule. The limit of detection using thes methods
was 15-20 ng/mL.
Or a l P h a r m a cok in etic An a lysis. The pharmacokinetic
behavior of compound 16h was evaluated in male Sprague-
Dawley rats. Briefly, the test compound was prepared as a
10 mg/mL solution in an ethanol-propylene glycol-D5W (20:
30:50, by volume) vehicle containing 1 molar equiv of sodium
hydroxide. Groups of rats (n ) 4 per group) received either a
10 mg/kg (1 mL/kg) intravenous dose administered as a slow
bolus in the jugular vein or a 10 mg/kg (1 mL/kg) oral dose
administered by gavage. Heparinized blood samples (∼0.4 mL/
sample) were obtained from a tail vein of each rat 0.1 (iv only),
0.25, 0.5, 1, 1.5, 2, 3, 4, 6, 9, and 12 h after dosing. The
samples were analyzed by reverse phase HPLC following
liquid-liquid extraction from the plasma. Initial estimates
of the pharmacokinetic parameters for NONLIN8414
were
obtained with the program CSTRIP.15 Area under the curve
(AUC) values were calculated by the trapezoidal rule over the
time course of the study. The terminal-phase rate constant
(â) was utilized in the extrapolation of the AUC from 12 h to
infinity to provide an AUC0-∞ value. Assuming dose propor-
tionality and correcting for the differences in dosing, a
comparison of the AUC following oral dosing with that
obtained following an intravenous dose provided an estimate
of the bioavailability (F).
NH3) m/ e 552 (M + H)+, 569 (M + NH4)+. Anal. for C30H41
N4O5‚0.55TFA‚1.5H2O: C, H, N.
-
Recep tor Bin d in g Assa ys. All samples were assayed
according to the protocols described in the first article in this
series.1 Representative analogs were also examined for their
binding to human ETA and ETB receptors expressed in CHO
Ack n ow led gm en t. The authors thank the Abbott
Analytical Department for assistance in acquiring H-
NMR and mass spectra.
1

Azole Endothelin Antagonists. 3
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4 991
H.; Shiosaki, K.; Sorensen, B. K.; Spina, K.; Sullivan, G. M.;
Tasker, A. S.; von Geldern, T. W.; Warner, R. B.; Opgenorth, T.
J .; Kerkman, D. J .; DeBernardis, J . F. 2-(Alkylamino)nicotinic
Acid and Analogs. Potent Angiotensin II Antagonists. J . Med.
Chem. 1993, 36, 2676-2688.
Refer en ces
(1) von Geldern, T. W.; Hutchins, C. W.; Kester, J . A.; Wu-Wong, J .
R.; Chiou, W.; Dixon, D. B.; Opgenorth, T. J . J . Med. Chem. 1996,
39, 957-967.
(2) von Geldern, T. W.; Kester, J . A.; Bal, R.; Wu-Wong, J . R.; Chiou,
W.; Dixon, D. B.; Opgenorth, T. J . J . Med. Chem. 1996, 39, 968-
981.
(3) See: Burton, P. S.; Conradi, R. A.; Hilgers, A. R. (B) Mechanisms
of Peptide and Protein Absorption. (2) Transcellular Mechanisms
of Peptide and Protein Absorption: Passive Aspects. Adv. Drug
Del. Rev. 1991, 7, 365-386 for an excellent review of the
problem, and for further references.
(4) Seiler, P. Interconversion of Lipophilicities from Hydrocarbon/
water Systems into the Octanol/water System. Eur. J . Med.
Chem. 1974, 9, 473-479.
(5) Conradi, R. A..; Hilgers, A. R.; Ho, N. F. H.; Burton, P. S. The
Influence of Peptide Structure on Transport Across Caco-2 Cells.
Pharm. Res. 1991, 8, 1453-1460. Conradi, R. A.; Hilgers, A. R.;
Ho, N. F. H.; Burton, P. S. The Influence of Peptide Structure
on Transport Across Caco-2 Cells. II. Peptide Bond Modification
Which Results in Improved Permeability. Pharm. Res. 1992, 9,
435-439.
(6) Young, R. D.; Mitchell, R. C.; Brown, T. H.; Ganellin, C. R.;
Griffiths, R.; J ones, M.; Rana, K. K.; Saunders, D.; Smith, I. R.;
Sore, N. E.; Wilks, T. J . Development of a New Physicochemical
Model for Brain Penetration and Its Application to the Design
of Centrally Acting H2 Receptor Histamine Antagonists. J . Med.
Chem. 1988, 31, 656-671. Chikale, E. G.; Ng, K.-Y.; Burton, P.
S.; Borchardt, R. T. Hydrogen Bonding Potential as a Determi-
nant of the in Vitro and in Situ Blood-Brain Barrier Permeability
of Peptides. Pharm. Res. 1994, 11, 412-419.
(7) Benet, L. Z.; Sheiner, L. B. Design and Optimization of Dosage
Regimens; Pharmacokinetic Data. In The Pharmacological Basis
of Therapeutics; Gilman, A. G., Goodman, L. S., Rall, T. W.,
Murad, F., Eds.; Macmillan Publishing Co.: New York, 1985; p
1709.
(8) Winn, M.; De, B.; Zydowsky, T. M.; Altenbach, R. J .; Basha, F.
Z.; Boyd, S. A.; Brune, M. E.; Buckner, S. A.; Crowell, D.; Drizin,
I.; Hancock, A. A.; J ae, H.-S.; Kester, J . A.; Lee, J . Y.; Mantei,
R. A.; Marsh, K. A.; Novosad, E. I.; Oheim, K. W.; Rosenberg, S.
(9) Clozel, M.; Breu, V.; Burri, K.; Cassal, J .-M.; Fischli, W.; Gray,
G. A.; Hirth, G.; Lo¨ffler, B.-M.; Mu¨ller, M.; Neidhart, W.; Ramuz,
H. Pathophysiological Role of Endothelin Revealed by the First
Orally Active Endothelin Receptor Antagonist. Nature 1993, 365,
759-761.
(10) Plattner, J . J .; Marcotte, P. A.; Kleinert, H. D.; Stein, H. H.;
Greer, J .; Bolis, G.; Fung, A. K. L.; Bopp, B. A.; Luly, J . R.; Sham,
H. L.; Kempf, D. J .; Rosenberg, S. H.; Dellaria, J . F.; De, B.;
Merits, I.; Perun, T. J . Renin Inhibitors. Dipeptide Analogues
of Angiotensinogen Utilizing a Structurally Modified Phenyl-
alanine Residue to Impart Proteolytic Stability. J . Med. Chem.
1988, 31, 2277-2288.
(11) Rosenberg, S. H.; Woods, K. W.; Kleinert, H. D.; Stein, H.;
Nellans, H. N.; Hoffman, D. J .; Spanton, S. G.; Pyter, R. A.;
Cohen, J .; Egan, D. A.; Plattner, J . J .; Perun, T. J . Azido
Glycols: Potent, Low Molecular Weight Renin Inhibitors Con-
taining an Unusual Post Scissile Site Residue. J . Med. Chem.
1989, 32, 1371-1378.
(12) Yamada, S.-I.; Shioiri, T.; Itaya, T.; Hara, T.; Matsueda, R. Ind.-
N-Alylation of Tryptophan and Synthesis of 1-Alkyltryptophan
Hydrazides. Chem. Pharm. Bull. 1965, 13, 88-93.
(13) Cheung, S. T.; Benoiton, N. L. N-Methylamino Acids in peptide
synthesis. V. The Synthesis of N-tert-butoxycarbonyl, N-Methyl-
amino Acids by N-Methylation. Can. J . Chem. 1977, 55, 906-
910.
(14) Statistical Consultants, Inc. PCNONLIN and NONLIN84: Soft-
ware for the Statistical Analysis of Nonlinear Models. Am.
Statistician 1986, 40, 52.
(15) Sedman, A. J .; Wagner, J . G. CSTRIP - A FORTRAN Computer
Program for Obtaining Initial Polyexponential Estimates. J .
Pharm. Sci. 1976, 65, 1006-1010.
J M9505932