Novel selective inhibitors of neutral endopeptidase for the treatment of female sexual arousal disorder. Synthesis and activity of functionalized glutaramides

Article abstract of DOI:10.1021/jm060133g

Female sexual arousal disorder (FSAD) is a highly prevalent sexual disorder affecting up to 40% of women. We describe herein our efforts to identify a selective neutral endopeptidase (NEP) inhibitor as a potential treatment for FSAD. The rationale for this approach, together with a description of the medicinal chemistry strategy, lead compounds, and SAR investigations are detailed. In particular, the strategy of starting with the clinically precedented selective NEP inhibitor, Candoxatrilat, and targeting low molecular weight and relatively polar mono-carboxylic acids is described. This led ultimately to the prototype development candidate R-13, for which detailed pharmacology and pharmacokinetic parameters are presented.

Full text of DOI:10.1021/jm060133g

J. Med. Chem. 2006, 49, 4409-4424
4409
Novel Selective Inhibitors of Neutral Endopeptidase for the Treatment of Female Sexual
Arousal Disorder. Synthesis and Activity of Functionalized Glutaramides
David C. Pryde,*,† Graham N. Maw,† Simon Planken,† Michelle Y. Platts,† Vivienne Sanderson,† Martin Corless,† Alan Stobie,†
Christopher G. Barber,‡ Rachel Russell,‡ Laura Foster,‡ Laura Barker,‡ Christopher Wayman,‡ Piet Van Der Graaf,‡
Peter Stacey,‡ Debbie Morren,‡ Christopher Kohl,§ Kevin Beaumont,§ Sara Coggon,§ and Michael Tute|
Departments of DiscoVery Chemistry, DiscoVery Biology, Pharmacokinetics, Dynamics and Metabolism, and Molecular Informatics and
Structure Based Design, Pfizer Global Research and DeVelopment, Sandwich, Kent CT13 9NJ, U.K.
ReceiVed February 7, 2006
Female sexual arousal disorder (FSAD) is a highly prevalent sexual disorder affecting up to 40% of women.
We describe herein our efforts to identify a selective neutral endopeptidase (NEP) inhibitor as a potential
treatment for FSAD. The rationale for this approach, together with a description of the medicinal chemistry
strategy, lead compounds, and SAR investigations are detailed. In particular, the strategy of starting with
the clinically precedented selective NEP inhibitor, Candoxatrilat, and targeting low molecular weight and
relatively polar mono-carboxylic acids is described. This led ultimately to the prototype development candidate
R-13, for which detailed pharmacology and pharmacokinetic parameters are presented.1
Introduction
is unsuitable for oral dosing as a therapy for FSAD. It is known
that VIP is degraded by endogenous neutral endopeptidase
(NEP, EC3.4.24.11), and it was our hypothesis that by selec-
tively inhibiting NEP, one could potentiate VIP levels and
thereby enhance VIP-induced increases in vaginal blood flow.
There are many inhibitors of NEP known in the literature,6 many
of which inhibit both NEP and the related enzyme, angiotensin
converting enzyme (ACE)7 with accompanying influence on
cardiovascular parameters. This highlighted ACE as a key
selectivity target to avoid any significant vasodilatory effects
in this nonlife-threatening condition. The most advanced
example of a selective NEP inhibitor is Candoxatril,8 a 5-indanyl
ester prodrug of the active constituent Candoxatrilat 1, which
was progressed into clinical trials in the 1990s for chronic heart
failure. Because 1 had already successfully completed a pre-
clinical toxicology program, had a proven benign pharmacologi-
cal profile in humans, represented a Pfizer proprietary chemical
series, and was a well-precedented chemotype, we chose this
as our starting point.
In targeting a prn treatment for FSAD, we sought a relatively
short half-life (T1/2 < 10 h), rapid onset (Tmax < 1 h) potent
inhibitor of NEP. Ideally, the target agent would not require
derivatization as a prodrug, simplifying its pharmacokinetic
profile and preclinical development, but rather the parent drug
would be completely absorbed and demonstrate high bioavail-
ability. This effectively ruled out the use of Candoxatril itself
for this program. Our strategy was to target a potent monocar-
boxylic acid based around the glutaramide template of Can-
doxatrilat by removing the non-Zn-binding carboxylic acid in
(1) and optimizing the amide substituent. This strategy is
depicted in the Figure 1. Modification of the amide substituent
X would allow the potency and physicochemistry to be easily
probed. A late-stage amide-forming step would also be amenable
to parallel synthesis techniques, greatly expediting SAR genera-
tion. Further fine-tuning of the physical properties could be
provided by changes to the P1′ side chain grouping Y.
Female sexual dysfunction (FSD)a is the difficulty or inability
of a woman to find satisfaction in sexual expression and is a
collective term for several diverse disorders including female
sexual arousal disorder (FSAD), desire disorder, and painful
intercourse disorder. FSAD is “a persistent or recurrent inability
to attain or to maintain until completion of the sexual activity
adequate lubrication-swelling response of sexual excitement”.2
There are wide variations in the reported incidence and
prevalence of FSAD, partly because of differing evaluation
criteria, but most investigators report a higher incidence of
arousal disorders in otherwise healthy women than they do
erectile dysfunction in the male population. It is therefore a
highly prevalent disorder within females, affecting up to 40%
of pre, peri, and postmenopausal women3 and is often ac-
companied by concomitant complaints such as an inability to
achieve orgasm and depression. Current therapies for treating
FSAD are limited to psychological counselling, over-the-counter
sexual lubricants, and several investigational candidates such
as hormonal agents and drug candidates approved for other
conditions such as male erectile dysfunction.
It has been proposed that the most common cause of FSAD
is decreased genital blood flow resulting in reduced vaginal,
labial, and clitoral engorgement.4 Vasoactive intestinal peptide
(VIP) is an endogenous vasodilator thought to have a key role
in the control of vaginal blood flow,5 its action being mediated
by cyclic AMP. The blood vessels supplying the vagina are
densely innervated by VIP-containing neurones. VIP is a 28
amino acid peptide, which is rapidly degraded in the liver, and
* Corresponding author. Tel: (01304) 616161. Fax: (01304) 651819.
E-mail: David.Pryde@pfizer.com.
† Department of Discovery Chemistry.
‡ Department of Discovery Biology.
§ Department of Pharmacokinetics, Dynamics and Metabolism.
| Department of Molecular Informatics and Structure Based Design.
a Abbreviations: FSD, female sexual dysfunction; FSAD, female sexual
arousal disorder; VIP, vasoactive intestinal peptide; NEP, neutral endopep-
tidase; ACE, angiotensin converting enzyme; ECE, endothelin converting
enzyzme; WSCDI, ((1-[3-(dimethylamino)propyl]-3-ethylcarbodiimde); EEDQ,
(2-ethoxy-1-ethoxycarbonyl-1,2-dihydroquinoline); Lawesson’s reagent,
(2,4-bis(4-methoxyphenyl)-1,3-dithia-2,4-diphosphetane-2,4-disulfide); Bur-
gess’ reagent, (methoxycarbonylsulfamoyl-triethylammonium hydroxide);
HOBt, 1-hydroxybenzotriazole.
By combining small groups X and Y, a substantial number
of potent NEP inhibitors were accessed relatively quickly, all
of which showed excellent selectivity (>300-fold) over related
zinc-metalloproteases, such as ACE and the related endothelin
converting enzyme (ECE). The synthesis and structure-activity
10.1021/jm060133g CCC: $33.50 © 2006 American Chemical Society
Published on Web 06/15/2006

4410 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14
Pryde et al.
Figure 1. Evolution of the strategy of targeting small monocarboxylic acid NEP inhibitors, starting from Candoxatrilat 1 and optimizing the amide
substituent X and the P1′ substituent Y.
Table 1. Elimination Half-life and Selected Properties of Short T1/2
highly bound to serum albumin and possibly even very highly
bound. We, therefore, chose to target quite high potency (hNEP
Carboxylic Acid Drugs9
T1/2
(h)
TPSA
(Å2)
IC50 of <50 nM).
drug
MWt.
cLogP
Enzyme Structure. NEP (EC.3.4.24.11) is a 750 amino acid
zinc-dependent enzyme first isolated in the 1970s. It is a Type
II membrane protein with a short N-terminal cytoplasmic domain
followed by a 23-residue hydrophobic domain and a large
extracellular domain containing the active site.11 It has a
relatively broad substrate specificity and is capable of cleaving
a variety of substrates in both the CNS and periphery, with
cleavage preferentially occurring on the amino side of hydro-
phobic residues, such as Phe, Met, and Leu,12 which suggests
the presence of a lipophilic S1′ binding region. This is reflected
in many of the potent NEP inhibitors reported in the literature
possessing a hydrophobic group adjacent to the Zn-binding
function. The Roche group has recently reported the crystal
structure of one such nonselective inhibitor, phosphoramidon
(4) co-crystallized with human recombinant NEP.13 At the outset
of our work, however, no such detailed knowledge of the NEP
structure was available, and all of the structural data in the
literature was based on the related bacterial enzyme thermolysin
(TLN).
Amoxicillin
Baclofen
1
365
214
331
318
244
411
331
171
206
254
356
257
0.6
1.6
1.3
3.3
4.1
3.6
2.9
1.2
3.7
2.8
3.6
1.6
158
63
75
49
37
83
131
63
37
54
74
59
3.8
3.7
1.5
5.5
0.7
1.4
6.5
2
Ciprofloxacin
Diclofenac
Flurbiprofen
Fluvastatin
Furosemide
Gabapentin
Ibuprofen
Ketoprofen
Sulindac
Tolmetin
1.8
7
7
relationships of a range of these selective monocarboxylic acid
NEP inhibitors are described below.
Targeting Acids as prn Agents. There are many carboxylic
acids, some of them zwitterions, exemplified in the clinic
covering a wide range of elimination half-lives.9 The majority
of them tend toward short T1/2, such as aspirin (salicyclic acid),
the penicillin-type antibiotics, and the quinolone-type antibiotics
(see Table 1). In targeting a prn treatment for FSAD, short T1/2
is precisely what is required. An analysis of the marketed acids
with T1/2 <8 h indicated that most were of small size (MWt.
<365) and/or of relatively low LogP (<3). The topological polar
surface area (TPSA) cutoff for these orally absorbed compounds
was around 140 Å2, consistent with prior experience at Pfizer
and allowing for a slightly higher value of the actively absorbed
amoxicillin.10 These agents are either excreted unchanged in
the urine (i.e., producing no circulating metabolites) or where
metabolism is seen, it is dominated by glucuronidation of the
acid function and CYP2C9-mediated oxidative metabolism.
There were very few examples where a marketed acid has both
a high molecular weight (>400) and is lipophilic (LogP > 3).
Many of the acids in Table 1 also display very high levels of
binding to serum albumin (>99%) and high unbound clearance,
especially the more lipophilic examples, leading to higher
efficacious dose requirements, a situation that is not conducive
with an ideal, low dose prn profile; indeed, some COX inhibitors
require the daily dosing of hundreds of milligrams up to more
than 1 g.
Thermolysin is a readily crystallized, thermostable 316 residue
enzyme that has a relatively poor primary sequence homology
with NEP but shares many of the same substrates, is inactivated
by the same inhibitors as NEP, and has the same stereochemical
dependence.14 Structural analogies have been drawn between
the active sites of TLN and NEP, and through the work of
Roques and others,14,15 homology models based on TLN
crystallographic data and computational overlap of sequence
data have been used to design potent NEP inhibitors. The 3D
structure of the combined ACE/NEP inhibitor thiorphan 5 bound
to thermolysin was reported in 1989, and the key contacts are
depicted below.16
The lipophilic benzyl group projects into a large and
hydrophobic S1′ subsite, whereas the central amide bond is
involved in a network of H bonds to the enzyme, and the sulfur
atom binds the zinc ion as its thiolate anion. Interestingly, a
crystallographic comparison of thiorphan and the reversed amide
retrothiorphan 6 bound to thermolysin showed that the amide
N and O atoms maintained essentially equivalent H-bond
interactions with the Asp and Arg residues of the enzyme, and
indeed, both of these agents share excellent NEP potency.
Further, a glutaramide inhibitor that contains much of the
functionality of our targets has been shown to make many of
the same contacts with thermolysin as does thiorphan.17 Acidic
groups in the S2′ region bind to a key Arg residue, and it is this
interaction that our target monoacids would be lacking. The
C-terminal region of TLN shows very low homology with NEP
and represents a relatively undefined area of the enzyme. As
Our analysis of these data was that we should aim to produce
the majority of our target structures with a molecular weight
under 400 and relatively high polarity (LogD <1). The very
polar penicillins make use of active transport mechanisms for
oral absorption,10 which we would not be able to guarantee for
our agents, and therefore, we decided to aim for around -0.5
to 0 LogD as the bottom end of the ideal lipophilicity range.
The likelihood was that our acids would probably also be quite

SelectiVe Glutaramide NEP Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14 4411
Figure 2. Structures of ACE/NEP inhibitors phosphoramidon 4, thiorphan 5, and retrothiorphan 6 for which crystal structures in either a NEP
construct or bacterial thermolysin are reported.
Scheme 1a
a Reagents and conditions: (i) 2 equiv LDA, THF, -78 °C f room
temperature, 16 h, 90%; (ii) 2 equiv LDA, allyl bromide, THF, -78 °C f
room temperature, 16 h, 92%; (iii) 15 psi H2, EtOH, 10% Pd/C, room
temperature, 91%, (iv) R1NH2, WSCDI, HOBt. NMM, solvent (see text),
25-100 °C, 16 h, 60-99%; (v) For PdBn, 30 psi H2, 10% Pd/C, room
t
temperature, 5-16 h; for P ) Bu, TFA/DCM (1:1), room temperature,
6-16 h or HCl (g), Et2O, room temperature, 15 min, 15-98%.
Figure 3. Drawing of the key contacts that thiorphan makes with the
Table 2. n-Pr-spiro-cyclopentyl Glutaramides Synthesized According to
S1′ and S2′ regions of bacterial thermolysin.
Scheme 1
such, we were uncertain at the outset of this work as to the
levels of both potency and selectivity we could expect from a
monoacid series and the manner in which neutral amide
substituents would bind to the S2′ region of NEP.18 We,
therefore, saw diversity as being crucial in our initial SAR
investigations, with subsequent targets being directed according
to emerging trends in NEP activity.
Synthetic Chemistry. A protected form of the glutaric acid
template is easily constructed by the alkylation of cyclopentane
carboxylic acid dianion (Scheme 1).19,20 The initial SAR was
generated with an n-propyl substituent adjacent to the Zn-
binding acid. This was best constructed in a 2-step allylation-
reduction sequence. The coupling of commercially available
amines to this acid was then performed using WSCDI ((1-[3-
(dimethylamino)propyl]-3-ethylcarbodiimde) with N-methyl-
morpholine (NMM) as the base. For reactive amines (e.g.,
primary alkylamines), dichloromethane (DCM) at room tem-
perature was sufficient to achieve complete reaction. For less
active amines (e.g., amino-substituted heterocycles), dimethyl-
formamide (DMF) or acetonitrile (MeCN) at elevated temper-
atures were required for good yields of the coupled products,
although the 3-ethyl pyridine (16) was consistently formed in
very poor yield. The ester protecting group could then be
removed using standard protocols to deliver the requisite acids
in excellent yields. This chemistry was sufficiently robust and
reliable to allow many of the coupling reactions to be performed
in parallel on a milligram scale in a 96-well plate format. The
reactions were monitored by TLC, the evaporation of volatiles
was carried out in a Genevac vacuum centrifuge, and the crude
residues were purified by automated HPLC using ELSD-MS
to detect products. The contents of those wells containing
product were combined, evaporated, and diluted robotically and
then sent for biological assaying.
hydride as the base (Scheme 2). The choice of DMF was found
to be critical to produce a homogeneous mixture and a complete
reaction. This reaction led to an appreciable amount of the
isomeric 3-ethyl analogue, which was easily separable by silica
gel chromatography. Hydrogenation of the separated 24 under
Pd/C catalysis then formed amino triazole 25 in quantitative
yield.
4-Amino-pyrimidine (27) was prepared starting from the
commercially available 2-amino-4-hydroxy pyrimidine. Chlo-
rination with POCl3 gave the corresponding chloroamine, which
was then aminated with propylamine to provide 27 in modest
yield.
The 4-methoxypyrazine target 29 in Scheme 4 was prepared
by sequential methoxylation and amination of 3,6-dichloropy-
rimidine, both steps proceeding in excellent yield.
A short series of pyridyl targets were synthesized according
to Schemes 5-7. First, 2-amino-4-benzyl pyridine was prepared
in three steps from 2,4-dibromopyridine (Scheme 5). Selective
The indane methanol compound 21 is known,20 and the
synthesis of both the indane methanol unit and 21 itself are
described therein. Noncommercial amines were prepared by the
procedures described in Schemes 2-11 and then coupled to
acid 7 in a fashion identical to that described in Scheme 1.
Thus, a 1-ethyl-1,2,3-triazole was synthesized by simple
alkylation of 4-nitro triazole21 with ethyl iodide using sodium
n
monolithiation at the 4-position with BuLi in ether, followed
by quenching with benzaldehyde gave benzylic alcohol 31,
which was aminated with aqueous ammonia in the presence of
copper (II) sulfate in a sealed vessel. Deoxygenation was then
carried out by hydrogenation with catalytic palladized charcoal

4412 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14
Pryde et al.
Scheme 2a
a Reagents and conditions: (i) NaH, DMF, 0 °C, 3 h, then silica gel chromatography, 57% (1-isomer), 21% (3-isomer). For the 1-isomer: (ii) 15 psi H2,
Pd/C (10%), EtOH, 4 h, 100%.
Scheme 3a
n
i
a Reagents and conditions: (i) POCl3, 100 °C, 4 h, 52%; (ii) PrNH2, Pr2EtN, DMF, 90 °C, 16 h, 26%.
Scheme 4a
a Reagents and conditions: (i) 0.88 NH3, DCM, 35 °C, 6 h, 91%; (ii) NaOMe, MeOH, steel bomb, 140 °C, 4 h, 94%.
Scheme 5a
a Reagents and conditions: (i) nBuLi, Et2O, -78 °C then PhCHO, -78 °C f room temperature, 16 h, 68%; (ii) 0.88 NH3, Cu(II)SO4‚5H2O, sealed tube,
135 °C, 24 h, 83%; (iii) 30psi H2, Pd/C (5%), 1 M HCl, EtOH, room temperature, 6 h, 78%.
Scheme 6a
a Reagents and conditions: (i) NaNH2, xylene, 150 °C, 16 h, 38%.
Scheme 7a
a Reagents and conditions: (i) NaH, LiBr, BnBr, DME/DMF, 70 °C, 16 h; (ii) Sn, cHCl, 90 °C, 1.5 h, 51%.
(5%) in acidic ethanol as solvent and the resulting amine 32
coupled to 7 as before.
tions were reported to improve the proportion of N-alkylation
relative to O-alkylation, and in our hands, none of the corre-
sponding O-benzyl analogue was detected. Mild reduction with
metallic tin in hydrochloric acid provided the 5-amino analogue
36, which was then coupled to 7 as before.
Commercially available 4-butyl pyridine was simply heated
at 150 °C in xylene with sodamide overnight using a slight
modification of the literature method.22 Chichibabin conversion
provided the requisite aminopyridine 34 in modest yield.
The benzyl-pyridone of Scheme 7 was made from 5-nitro-
2-pyridone, the sodium anion of which was simply alkylated
with benzyl bromide in the presence of lithium bromide in a
mixture of dimethoxyethane (DME) and DMF.23 These condi-
Several 5-substituted 1,3,4-thiadiazoles were prepared by
reaction of the appropriate acid chloride with thiosemicarba-
zide,24 followed by warming for several hours, until all evolution
of gas had ceased. Prolonged reaction times or higher reaction
temperatures led to a significantly reduced yield of 2-amino-

SelectiVe Glutaramide NEP Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14 4413
Scheme 8a
a Reagents and conditions: (i) Thiosemicarbazide, (no solvent), 40 °C, 4 h, then separation by flash chromatography, 28-53%.
Scheme 9a
a Reagents and conditions: (i) EEDQ, DCM, room temperature, 16 h, 37%; (ii) Lawesson’s reagent, THF, room temperature-70 °C, 3 h then room
temperature, 16 h, 89%; (iii) HCl (g), DCM, 0 °C, 15 min then N2, room temperature, 1 h, 100%.
Scheme 10a
a Reagents and conditions: (i) BOC2O, NaOH, dioxan, 0 °C f room temperature, 16 h, 81%; (ii) Me2NH, WSCDI, HOBt, NMM, DMF, room temperature,
16 h, 66%; (iii) TFA, DCM (1:1 v/v), room temperature, 4 h, 55%.
Scheme 11a
a Reagents and conditions: (i) LDA, allyl bromide, THF, -78 °C f room temperature, 16 h, 23% (49, RdPh commercially available); (ii) 30 psi H2,
Pd/C (10%), EtOH, room temperature, 16 h, 52%; (iii) R-(+)-1-Phenyl-ethylamine, p-TsOH, toluene, reflux, 24 h, 100%; (iv) 120 psi H2, RaNi, EtOH, room
temperature, 48 h, 7-15%; (v) 120 psi H2, Pd/C (10%), EtOH, 45 °C, 16 h, 89-94%.
thiadiazole products 39 and appreciable amounts of 2-amido-
thiadiazoles 38, where the requisite amine appears to form first
and then reacts with excess acid chloride in situ (Scheme 8).
A related 2-aminomethyl-1,3,4-thiadiazole (45) was synthe-
sized in excellent yield by the initial coupling of N-BOC glycine
with acylhydrazide using EEDQ (2-ethoxy-1-ethoxycarbonyl-
1,2-dihydroquinoline) as the coupling agent. Cyclodehydration
aided by Lawesson’s reagent (2,4-bis(4-methoxyphenyl)-1,3-
dithia-2,4-diphosphetane-2,4-disulfide) formed the thiadiazole
ring, and simple treatment with gaseous hydrogen chloride
revealed the amine, which was coupled to 7 as above.
Several cyclohexylamines were synthesized in one of three
ways. A 1,3-disubstituted aminocyclohexylamide (47) was
prepared from the corresponding amino-cyclohexane carboxylic
acid by an N-carbamoylation, amidation, and N-deprotection
sequence. The final product was obtained as a mixture of all
possible diastereoisomers.
that occupied by the adjacent alkyl substituent to furnish the
racemic cis diastereomer 50 in low yield.26 This was then
coupled to the glutarate template as before, and the esters
deprotected to furnish the final acids.
Finally, the 1,4-disubstituted analogue of the aminocyclo-
hexylamine described in Scheme 10 was best prepared by first
installing cyclohexane ring 53 onto substrate 7 and then
appending the amide group afterward.
A phenethyl variation at the P1′ region was prepared in a
manner entirely analogous to that described in Scheme 1. The
dianion of glutarate 5719,20 was alkylated with phenethyl bromide
in good yield and then coupled to 2-amino-5-ethyl-1,3,4-
thiadiazole according to Scheme 1. Direct amide coupling of
57 yielded compound 60 (YdH).
In the case of alkyl analogues of the n-propyl group, simple
homologation of allyl acid 6119,20 was carried out by ozonolytic
olefin cleavage, followed by Wittig condensation and hydro-
genation of the ensuing double bond to provide acids 62, which
were then coupled to 2-amino-5-ethyl-1,3,4-thiadiazole.
A small series of oxadiazole amide bioisosteres were prepared
from the corresponding carboxylic acid by the initial formation
of bis acyl hydrazides 65, followed by cyclodehydration
In contrast, two 1,2-disubstituted amino-cyclohexylamines
were formed using enolate chemistry to introduce the alkyl
substituent onto the cyclohexane ring 49.25 The cis-1,2 arrange-
ment was secured by the hydrogenation of a nonracemic
phenethylimine, which occurs specifically from the face opposite

4414 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14
Pryde et al.
Scheme 12a
a Reagents and conditions: (i) See Scheme 1; (ii) 15 psi H2, Pd/C (10%), room temperature, 16 h, 93%; (iii) R1R2NH, WSCDI, HOBt, NMM, DMF, room
temperature, 16h, 58-71%; (iv) TFA, DCM (1:1 v/v), room temperature, 4 h, 65-91%.
Scheme 13a
a Reagents and conditions: (i) 2 equiv LDA, THF, phenethyl bromide, -78 °C f room temperature, 16 h, 76%; (ii) WSCDI, 2-amino-5-ethyl-1,3,4-
thiadiazole, HOBt, NMM, DMF, 25-100 °C, 16 h, 73-89%; (iii) See Scheme 1.
Scheme 14a
a Reagents and conditions: (i) O3, DCM, -78 °C, 3 h, then DMS, room temperature, 3 h, 100%; (ii) 2 equiv Ph3PdCHR, THF, 0 °C, 16 h, 17-20%;
(iii) 15 psi H2, Pd/C (10%), EtOH, room temperature, 6 h, 100%.
Scheme 15a
a Reagents and conditions: (i) CDI, HOBt, THF, hydrazine, 0 °C, 1 h, 100%; (ii) RCO2H, WSCDI, HOBt, NEt3, DCM, room temperature, 16 h, 67-
92%; (iii) Burgess’ Reagent, THF, reflux, 16 h, 82-89%; (iv) TFA, DCM (1:1 v/v), room temperature, 3 h, 69-84%.
Scheme 16a
a Reagents and conditions: (i) (PhO)2P(O)N3, THF, reflux, then BnOH, 2.5 h, 80%; (ii) 15 psi H2, Pd/C, EtOH, room temperature, 16h, 71%; (iii))
RCO2H, WSCDI, HOBt, NEt3, DCM, room temperature, 16 h, 26-59%; (iv) TFA, DCM (1:1 v/v), room temperature, 15 h, 63-99%.
promoted by Burgess’ reagent (methoxycarbonylsulfamoyl-
triethylammonium hydroxide).
Tertiary N-methyl amides were prepared directly from the
corresponding secondary amide by a simple deprotonation/
methylation sequence (Scheme 17).
Reversed amides were synthesized from ester 7 by a Curtius
rearrangement of the derived acyl azide. The best yield was
obtained when the ensuing isocyanate was quenched with benzyl
alcohol, and the resulting benzyl carbamate then deprotected
to amine 68 by hydrogenation. The coupling of acids to this
amine then proceeded smoothly under the usual coupling
conditions.
Results and Discussion
Compounds were first evaluated as inhibitors of canine NEP
(dNEP) in vitro using a spectrophotometric assay, wherein the
fluorescent product Abz-D-Arg-Arg is formed by the cleavage
of o-aminobenzoyl-D-Arg-Arg-Leu-ethylenediamine-2,4-dini-

SelectiVe Glutaramide NEP Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14 4415
Scheme 17a
a Reagents and conditions: (i) NaH, MeI, THF, 0 °C, 6 h, 21%; (ii) 15 psi H2, Pd/C, EtOH, room temperature, 16 h, 100%.
Table 3. Canine NEP Activity (IC50) for Various Substituted Glutaric
functionality in both 8 and 12, a further 3-fold potency gain
Acid Monoamidesa
was made to give our first sub-100 nM inhibitor 21 (96 nM).
Converting this amino-carbocycle into an amino-heterocycle
gave mixed results. Although 1,3,4-thiadiazoles 10 and 13
showed good activity (176 and 60 nM, respectively), methyl
pyridine 15 was only weakly active (630 nM). Similarly,
although 1,2,3-triazole 26 demonstrated appreciable potency (82
nM), amino-pyrimidine 28 and methoxypyrazine 30 were much
weaker. A possible explanation for this SAR could lie in the
dipole moments of these heterocycles. Both thiadiazole and
1,2,3-triazole have significant dipoles toward the N atoms and
could act as mimics of a carboxylate anion and thereby bind to
the S2′ Arg suggested from the TLN modeling studies. This
dipole would also be present to a large extent in a pyrazine
ring, but the lower activity of 30 could be compromised by the
methoxyl group clashing with a hydrophobic region of the
enzyme.
Nonetheless, from this first wave of compounds, we were
very encouraged to see some good potency in these small
monoacids, and some general trends were apparent. A small
carbo or heterocycle attached directly to the amide N atom
provided several potent examples, provided there was no polar
(H-bonding) group present on the ring. Rather, a lipophilic
substituent (alkyl or phenyl) on such a small ring gave the most
potent compounds of all. These SARs pointed strongly toward
the amide S2′ substituent projecting into a hydrophobic region
of the enzyme from an appropriately configured scaffold.
Especially encouraging was the finding that the more active
compounds from Table 3 were well within our ideal range of
physical properties (e.g., 13; LogD25 0.5, MWt 339). Equally
encouraging was the finding that all compounds made showed
exquisite selectivity over the related metallopeptidase ACE.
a Mean data of at least two determinations. All entries are racemic
mixtures. All compounds showed IC50 >3 µM activity vs dACE (data not
shown) except for 15, 22, and 28, which were not tested.
trophenyl.27 Soluble NEP was obtained from the canine kidney
cortex by an adaptation of a literature method.28 Selectivity over
canine ACE (dACE) and human ECE-1 (hECE-1) was estab-
lished using o-aminobenzoyl-Gly-p-nitro-Phe-Pro-OH and big
ET-1 as substrates, respectively. Potent and selective compounds
were then assayed against human NEP (hNEP) and progressed
into pharmacokinetic evaluation. The primary amino acid
sequence of human NEP is >90% identical to that of other
mammalian species, where the sequence is known. Sequence
conservation is highest around the region that makes up the
active site, and at the outset of this work, we did not predict
any particular differences in activity of our series of glutaramides
in dog vs human-derived NEP. Promising compounds that
emerged from this analysis were assessed for efficacy in an
animal model of genital blood flow.34
Pyridyl Amides. We reasoned that the Me group attached
to the pyridine ring of 15 was possibly neither large enough
nor suitably placed to provide optimal binding. This assertion
was vindicated by the observation that unsubstituted pyridine
17 was a very weak inhibitor of NEP (>1 µM) and that both
6-ethyl pyridine 18 and 3-ethyl pyridine 16 were even less
active. Further, extending the 4-Me group of 15 to either an
nBu (35, 184 nM) or a benzyl group (33, 96 nM) improved
activity. Switching from a benzyl pyridine to an N-benzyl
pyridone 37 lost NEP potency. It appeared that an appropriately
situated lipophilic group on a pyridyl scaffold could offer good
NEP activity in this series.
Thiadiazolyl Amides. A set of results very similar to that
with the pyridyl series was seen with the 1,3,4-thiadiazole series.
Unsubstituted heterocycle 40 was only weakly active against
NEP (377 nM) and some 6-fold weaker than the ethyl case 13.
Both phenyl thiadiazole 41 and isobutyl analogue 42 were also
weaker than 13 (some 4-fold and 2-fold, respectively). However,
both benzyl expression 43 and cyclopropylmethyl compound
44 were found to be 2-fold more active than 13, which again
points to a specific lipophilic interaction between the heterocycle
substituent and the enzyme. The 3-fold difference in activity
between the isobutyl group of 42 and the cyclopropylmethyl of
Initial Amide SAR. Many of the initial series of amides
prepared were relatively weak inhibitors of NEP (Table 3).
Flexible side chains, such as those in 9 and 22 showed the least
activity against NEP (>1.5 µM). Removing some of this
flexibility by alkyl substitution on the side chain improved the
activity 3-fold (11, 530 nM). In a similar vein, the cyclopen-
tylmethanol group in 8 was only weakly active (1710 nM),
whereas when combining the hydroxymethyl function into a
benzyl template, 14 showed modest activity (384 nM). Fusing
a phenyl group onto the cyclopentane improved binding potency
approximately 5-fold (12, 313 nM), and by combining the

4416 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14
Pryde et al.
44, and the 10-fold difference between the phenyl group of 41
and the benzyl group of 43 is testament to the specificity of
this interaction. Inserting a CH2 spacer between the amide N
atom and the thiadiazole ring resulted in a large drop in activity,
similar to the trends that were seen in Table 1.
The thiadiazole series again demonstrates that achieving good
potency from a substituted, planar heterocyclic series is possible.
There was a clear spatial preference for the substituent on these
heterocycles such that small variations in the substituent or its
conformation produced quite substantial changes in activity. It
was unclear whether these features would translate to a
nonplanar carbocyclic series.
Figure 4. Plasma concentration time curve in male rats after iv or po
administration of R-13 (1 mg/kg).
Carbocyclic Amides. To allow an examination of areas of
the S2′ binding site as yet unexplored in the heterocyclic series,
several saturated five and six-membered ring amide groups were
prepared with a range of substitution, conformational preference,
and polarity. We reasoned that the alkyl substituents appended
to the thiadiazole and pyridyl rings were being directed by the
ring scaffold away from the trajectory of the amide bond and
that this trajectory was key to good potency. A carbocyclic ring
would offer a much more flexible scaffold from which to append
groups at various positions around the ring. The first couple of
targets looked at both a small alkyl and a benzyl group anchored
in a cis conformation to the amine on a cyclohexyl template.
Benzyl compound 51 was some 4-fold more active than its
propyl counterpart 52 but still some 6-fold less potent than
benzyl thiadiazole 43. Inherent in moving to a cyclohexyl
template is an increase in LogP of this substituent, and having
observed no significant breakthroughs with the first couple of
lipophilic examples from the cyclohexyl series, all further targets
featured more polar substituents. The first of these were dimethyl
amides in a 1,4 and a 1,3 arrangement (55 and 48, respectively),
which were both around 300 nM inhibitors. Moving to some
much more polar amide groups, pyrrolidine 19 showed only
micromolar potency, whereas surprisingly, a primary amide
appended to both a cyclopentyl 20 and a cyclohexyl 56 ring
proved to be the most active examples in the series at 213 and
150 nM, respectively. These cis fused ring systems were
severalfold more potent than their trans analogues (data not
shown) and seemed to refute some of the observed intolerance
of polar substituents suggested in Table 3. However, even the
most active example from this series was still 3-fold less potent
than some of the heterocyclic compounds above.
propyl chain, and its small size encouraged us to retain this as
our favored group in the P1′ position.
Secondary Amide Isosteres. The final region of our series
that was explored was the amide linkage. The literature structural
information presented above suggested that a strong H-bond
network to both the amide O atom and the NH group was
important to activity. This was borne out by methylating the N
atom of pyridone 37 to essentially destroy all activity in tertiary
amide 71. Similarly, the oxadiazole isosteres, which lacked an
H-bond donating group, were only very weak inhibitors. Given
the maintenance of NEP activity that was seen by switching
from thiorphan to retrothiorphan, two reversed amides were
synthesized (69 and 70), but this change lost almost all binding
affinity. This latter result simply confirmed that thiorphan forms
a different H-bond network to NEP and that SAR trends within
the thiorphan series would be very difficult to track over to the
current series of glutaramides. No further examples of amide-
variant compounds were pursued.
Enantiomerically Pure Material. The NEP literature clearly
establishes a stereochemical preference for substituents adjacent
to the Zn-binding moiety.14 In the case of the glutaramide-based
Candoxatrilat, this preference is R, which we fully expected to
be mirrored in our series of monoacids. This was confirmed by
resolution of several members of the series, profiling the
resolved enantiomers and confirming the absolute configuration
of the eutomer by small molecule X-ray crystallography of the
derivatives (data not shown).
The most promising compounds possessing good binding
affinity, low MWt., and LogD29 within our target range were
identified as both methyl and ethyl thiadiazoles 10 and 13, and
indane methanol 21. These compounds covered a LogD range
of between 0.1 and 1.1 with molecular weights in the range
325-373. These were resolved by chiral HPLC and the
individual enantiomers profiled.
Consistent with the Candoxatrilat series, the R enantiomers
were found to be considerably more active than the S. All
compounds were found to be exquisitely selective over hECE-1
and dACE, and the Caco-2 flux values were largely consistent
with a combination of MWt. and LogD. The favorable balance
of potency and physicochemistry led to R-13 being chosen for
further profiling in vitro and in vivo. Inhibitory IC50 values were
obtained with R-13 against several species of NEP, which were
to be used to assess either pharmacokinetic (rat) or efficacy
(rabbit) parameters, and were largely seen to agree well with
the canine NEP inhibition values, within an approximately 2-fold
window.
P1′ SAR. From the above data, several low MWt. potent NEP
inhibitors had emerged. All of these compounds were racemic
adjacent to the carboxyl group, and all incorporated an n-propyl
group at this location. It was this area where our attention was
now focused. Our choice of n-propyl had been driven largely
by its small size and precedented NEP activity within a
glutaramide template. From the thermolysin literature,14 larger
and more lipophilic groups projecting into the S1′ subsite would
be expected to provide potency gains. Taking the 5-ethyl-1,3,4-
thiadiazole as a representative example of an active S2′ amide
substituent, several analogues of the n-propyl group were
examined.
Extending the propyl group to butyl gives an essentially
equipotent compound 63 (84 nM), and a further branching Me
group in 64 loses a small amount of activity. Phenethyl analogue
59 did not produce any large potency gains, and removing the
P1′ side chain lost only 4-fold binding potency. It was gratifying
to discover that our SAR with an n-propyl substituent repre-
sented the most active P1′ group from within this limited SAR
set of targets, and this knowledge, the ease of synthesis of the
The pharmacokinetics of R-13 were evaluated in the male
and female rat (Figure 4 and Table 11). R-13 exhibited a low
clearance and a low volume of distribution resulting in a short
elimination half-life of ca. 2 h. A clear sex difference was not

SelectiVe Glutaramide NEP Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14 4417
Table 7. Canine NEP Activity (IC50) for Various P1′ Analoguesa
Table 4. Canine NEP Activity (IC50) for Various Pyridyl Amidesa
a Mean data of at least two determinations. All entries are racemic
mixtures. All compounds showed IC50 >3 µM activity vs dACE.
a Mean data of at least two determinations. All entries are racemic
mixtures. All compounds showed IC50 >3 µM activity vs dACE (data not
shown) except for 16 and 18, which were not tested.
Table 8. Canine NEP Activity (IC50) for Various Amide Isosteresa
Table 5. Canine NEP Activity (IC50) for Various Thiadiazolyl Amidesa
a Mean data of at least two determinations. All entries are racemic
mixtures. All compounds showed IC50 >3 µM activity vs dACE except
for 69, 70, and 71, which were not tested.
a Mean data of at least two determinations. All entries are racemic
mixtures. All compounds showed IC50 >3 µM activity vs dACE.
kg). Similar sex differences in the renal excretion of carboxylic
acids have been observed before30 and could be due to a sex
specific expression of the transport protein oatp1a1 (aka oatp1;
gene symbol slco1a1, formerly slc21a1) in the apical tubular
membrane.31 The insignificant PK sex difference in the rat can
possibly be attributed to the predominant elimination of R-13
via UGT enzymes, where major sex differences in expression
are not yet completely understood.32 The exact UGT enzymol-
ogy in the rat was not further investigated. There was no
difference in the quantity and quality of circulating oxidative
metabolites in male and female rats (data not shown). Conse-
quently, the oxidative metabolism of R-13 in the rat was shown
to be mainly due to the sex unspecific cyp2c6 with minor
contributions from cyp3a1 and cyp1a1 (Figure 5).
Table 6. Canine NEP Activity (IC50) for Various Cyclohexyl Amidesa
The male specific cyp3a2, cyp2c11, cyp2c13, and the female
specific cyp2c1233 did not metabolize R-13.
Given that the medicinal chemistry strategy for this program
was based on the maintenance of low molecular weight and
LogD, it was interesting to compare the above pharmacokinetic
profile with that of a larger and more lipophilic compound. R-13
had a PK profile superior to that of R-21 (Table 12). In line
with its higher lipophilicity, R-21 showed 4-5 times higher
unbound clearance than R-13. The pharmacokinetics of enan-
tiomerically pure R-21 were not significantly different from that
of the racemate rac-21 (Table 12 and Figure 6) suggesting that
both enantiomers undergo similar pharmacokinetic processes.
Efficacy. The combination of unbound clearance and potency
favored the progression of R-13 into the rabbit model of sexual
arousal.4b,34 In this model, the rabbit pelvic nerve is stimulated
a Mean data of at least two determinations. All entries are racemic
mixtures. All compounds showed IC50 >3µM activity vs dACE.
evident, although female rats showed a trend toward higher
systemic clearance and higher volume (Table 11).
The PK parameters given in Table 11 do not reflect the
complete picture of the potentially sex specific disposition of
R-13 in the rat; female rats exhibited a renal clearance
significantly higher than that of males (0.86 vs 0.03 mL/min/

4418 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14
Pryde et al.
Table 9. NEP Inhibitory Activity (IC50) and Physicochemical Data for Enantiomerically Pure Versions of 10, 13, and 21a
R-enant. Caco-2
racemate
compd
R-enant. potency,
dNEP (nM)
S-enant. potency,
dNEP (nM)
Papp, AB/BA
R-enant. HLM
R-enant. RLM
MWt
LogD
TPSA (Å2)
(×10-6 cm s-1
)
(min)
(min)
10
13
21
325
339
373
0.1
0.5
1.1
120
120
87
114
26
11
286
292
205
<1/2
9/14
6/10
>120
>120
62
ndb
>120
>120
a Mean data of at least two determinations. All compounds showed IC50 >3 µM activity vs dACE and hECE-1, except for S-10, which was not tested
against ECE-1. All compounds showed <10% inhibition against a wide panel of receptors, ion channels, and enzymes when tested at 10 µM. Caco-2 flux
was measured at pH7.4/7.4 and 25 µM. LogD was measured in 1-octanol/water at pH 7.4. Metabolic stability (substrate disappearance) was determined in
human liver microsomes at 1 µM substrate and 0.5 µM CYP. b nd not determined.
vaginal blood flow.34c The rabbit IC50 value for R-13 was
Table 10. Inhibitory NEP Potency of R-13 against Several Species of
NEPa
measured at 10.2 nM, which is in good agreement with the
hNEP
IC50
(nM)
dNEP
IC50
(nM)
ratNEP
IC50
(nM)
rabbitNEP
IC50
(nM)
recombinant
hNEP IC50
(nM)
measured EC50 value. The maximal response was seen at
approximately 10 × IC50, that is, 100 nM free drug with the
response starting to peak at potentially lower concentrations than
this one. R-13 thus demonstrates excellent efficacy in this rabbit
model of sexual arousal and was expected to be similarly
efficacious in humans. This compound was, therefore, chosen
for clinical evaluation as a potential treatment for FSAD.
compd
R-13
18.9
23.7
14.3
10.2
19.7
a Mean data of at least two determinations.
to simulate the neuronal effects of sexual arousal, and changes
to both the vaginal and clitoral blood flow are monitored using
laser doppler sonography. The effects of intravenously admin-
istered compounds can be assessed directly over extended
periods of time. The performance of R-13 in this model has
been published elsewhere.34c It is important to note that there
is no effect on the basal (i.e., unstimulated) blood flow or on
any cardiovascular parameters. The onset of the effect on vaginal
blood flow is noticeable after the first administration of
compound, in keeping with its pharmacokinetic profile. Trans-
lating these direct blood flow measurements into a dose-
response curve provides an EC50 value for R-13 of approxi-
mately 36 nM free drug, which shows a 33% potentiation of
Conclusion
We have shown that relatively small (MWt. <400) 2-ami-
nopyridines and 2-amino-1,3,4-thiadiazoles with hydrophobic
substituents at the 4- and 5-positions, respectively, are potent
and selective inhibitors of NEP. Incorporating various substitu-
tion patterns at the P1′ position is tolerated, and changes in this
region, in conjunction with those at the S2′ hydrophobic binding
region allows a range of compounds to be prepared with
disparate LogD. Pharmacokinetic profiling in animals has
suggested that a profile that is suited to prn dosing in humans
Table 11. Rat Pharmacokinetic Parameters for R-13 at 1 mg/kg (iv and po)a
CL
Vd
T1/2
Tmax
Cmax
F (oral)
sex
(mL/min/kg)
(L/kg)
(h)
(h)
(ng/mL)
fu
(%)
males
females
3.0 ( 0.3
4.7 ( 1.2
0.4 ( 0.1
0.9 ( 0.4
1.7 ( 0.3
2.1 ( 0.7
1
1
1176 ( 124
1462 ( 120
0.106 ( 0.009
0.13 ( 0.026
103 ( 46
131 ( 58
a Mean data ( standard deviation of three animals. Jugular vein cannulated rats were dosed at 1 mg/kg iv (tail vein) and po (gavage). The blood samples
were withdrawn via the indwelling catheter until 24 h post-dose. Results for CL, Vd, and T1/2 were compared with Student’s t-test and found to be not
significantly different (p >0.05).
Figure 5. Metabolic pathways of R-13. Human enzymes are written in upper case and rat enzymes in lower case letters.

SelectiVe Glutaramide NEP Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14 4419
Table 12. Rat Pharmacokinetic Parameters for 21 at 1 mg/kg (iv and po)a
CL
Vd
T1/2
Tmax
Cmax
F (oral)
compd
(mL/min/kg)
(L/kg)
(h)
(h)
(ng/mL)
fu
(%)
R-21
rac-21
34.8 ( 4.0
22.6 ( 9.8
4.7 ( 1.8
3.9 ( 2.1
1.5 ( 0.5
2.0 ( 0.2
0.25
0.25
233 ( 177
311 ( 130
0.243 ( 0.036
nd
144 ( 11
108 ( 42
a Mean data ( standard deviation of three animals. Jugular vein cannulated rats were dosed at 1 mg/kg iv (tail vein) and po (gavage). The blood samples
were withdrawn via the indwelling catheter until 24 h post-dose; nd: not determined. The results for CL, Vd, and T1/2 were compared with Student’s t-test
and found to be not significantly different (p >0.05).
g of a yellow oil, which slowly solidified on standing. This solid
was recrystallized from approximately 15 mL of hexane to provide
1.49 g of the title compound as a white solid.
1-(2-tert-Butoxycarbonyl-pent-4-enyl)-cyclopentanecarbox-
n
ylic Acid (61). A 2.5 M solution of BuLi in hexanes (3.5 mL,
105 mmol) was added dropwise to a stirred solution of diisopro-
pylamine (15 mL, 107 mmol) in 100 mL of THF at -20 °C under
nitrogen, and the whole solution was stirred at this temperature for
30 min. The solution was cooled to -65 °C and, a solution of 57
(970 mg, 4 mmol) in 5 mL THF was added dropwise over
approximately 10 min. And the whole solution was then stirred at
this temperature for 40 min. Allyl bromide (0.4 mL, 4.6 mmol)
was added in one portion, and the reaction was allowed to warm
up to room temperature slowly over 6 h. Then, 50 mL of diethyl
ether was added, followed by 9 mL of 2 N HCl solution, and the
Figure 6. Plasma concentration time curve in male rats after iv or po
administration of R-21 or rac-21(1 mg/kg).
organic layer was separated, dried over MgSO4, and evaporated in
vacuo to give a yellow oil. This oil was purified by column
chromatography using 30%, then 50% diethyl ether in pentane as
an eluant to provide the title compound as a clear oil (1.05 g).
is achievable with this class of compound, from which R-13
was chosen for further clinical evaluation.
Experimental Section
1-(2-tert-Butoxycarbonyl-pentyl)-cyclopentanecarboxylic Acid
(7). Compound 61 (900 mg, 3.2 mmol) was taken up in 15 mL of
EtOH and hydrogenated at 30 psi hydrogen pressure in the presence
of 10% palladized charcoal (90 mg) in a bomb for 6 h. The
suspension was filtered through a short plug of Arbocel, and the
filtrate was evaporated to dryness in vacuo to provide the title
compound as a clear oil (900 mg). In the case where a benzyl in
place of a tert-Bu protecting group was used, an identical procedure
was used in each of the above three steps starting with benzyl
bromopropionate, providing the benzyl ester as a clear oil.
General Procedure 1: Amide Formation. The requisite amine
derivative (1 equiv) and acid 7 (1 equiv) were dissolved in DMF,
DCM, or MeCN (1 M) and NMM added in one portion. HOBt (1
equiv) was then added, followed by 1 equiv of WSCDI. The mixture
was typically heated at 100 °C for 12 h under a nitrogen atmosphere
or until reaction was complete as judged by TLC analysis. After
cooling to room temperature, the mixture was diluted with EtOAc
and washed several times with water. The organic layer was
separated, washed with water, dried (MgSO4), and evaporated to
afford the desired amides, which were purified by flash chroma-
tography using mixtures of either pentane and EtOAc or MeOH
and DCM, depending on the polarity of the individual products.
General Procedure 2: Amide Formation in Parallel. The
requisite amine derivatives (30 µmoles) were weighed into the wells
of a solid Al 96-well block and 1 equiv of DMF solutions of the
reagents described in procedure 1 introduced into each well by a
Eppendorf micropipet. 1.1 equiv of a DMF solution of 7 was also
added to each well and the block agitated overnight while heating
to ca. 100 °C. After cooling to room temperature, 2 mL of water
was added to each well and the contents filtered through a
hydrophobic frit. Then, 1 mL of TFA was added to the filtrates
and the block once again agitated overnight at room temperature.
Volatiles were removed in a Genevac vacuum centrifuge, and the
residues purified by automated HPLC (Gilson 305 pumps, Gilson
215 injection/fraction collection, and Polymer Laboratories PL-
ELS1000 ELSD detection).
General Procedure 3: tert-Butyl Ester Deprotection. The
requisite tert-butyl ester (1 equiv) was dissolved in a 1:1 mixture
of TFA and DCM (0.25 M total) and the reaction mixture stirred
at room temperature under a nitrogen atmosphere overnight. The
reaction mixture was evaporated and then partitioned between DCM
and water. The organic layer was separated, washed several times
General. Melting points were determined on a Gallenkamp
Melting point apparatus using glass capillary tubes and are
uncorrected. Unless otherwise indicated, all reactions were carried
out under a nitrogen atmosphere, using commercially available
anhydrous solvents. Thin-layer chromatography was performed on
glass-backed precoated Merck silica gel (60 F254) plates, and flash
column chromatography was carried out using 40-63 µm silica
gel. Proton NMR spectra were measured on a Varian Inova 300 or
400 spectrometer in the solvents specified. Mass spectra were
recorded on a Fisons Trio 1000 using thermospray positive (TSP)
ionization. Combustion analyses were conducted by Exeter Analyti-
cal U.K., Ltd., Uxbridge, Middlesex. Where analyses are indicated
only by the symbols of the elements, the results obtained are within
0.4% of the theoretical values. The purity of compounds was
carefully assessed using analytical TLC and proton NMR, and the
latter technique was used to calculate the amount of solvent in
solvated samples. The pKa measurements were performed by Sirius
Analytical Instruments Ltd., Forest Row, East Sussex. In multistep
sequences, the purity and structure of intermediates were verified
1
spectroscopically by H NMR. Optical rotations were determined
using a Perkin-Elmer 341 polarimeter. Optical purity was deter-
mined by an analytical HPLC analysis of preparative HPLC-
resolved material (see below) via a comparison to the racemic
material.
1-(2-tert-Butoxycarbonyl-ethyl)-cyclopentanecarboxylic Acid
n
(57). A 2.5 M solution of BuLi in hexanes (42 mL, 105 mmol)
was added dropwise to a stirred solution of diisopropylamine (15
mL, 107 mmol) in 100 mL of THF at -20 °C under nitrogen and
the whole solution was stirred at this temperature for 30 min. A
solution of cyclopentane carboxylic acid (5.7 g, 50 mmol) in 10
mL of dry THF was then added dropwise and the resulting solution
stirred at -20 °C for 30 min, and then at room temperature for 2
h. The solution was then re-cooled to -20 °C and tert-butyl
bromopropionate (11 g, 53 mmol) was added portionwise and the
mixture then stirred at room temperature for approximately 2 h.
Then, 150 mL of diethyl ether and 110 mL of 2 N HCl solution
were added in sequence, and the organic layer was separated, dried
over MgSO4, and evaporated to a yellow oil. This oil was taken up
in diethyl ether (100 mL) and washed with saturated NaHCO3 until
almost all of the starting cyclopentane carboxylic acid was no longer
present by TLC. The organics were then washed with 2 N HCl
solution, separated, dried, and evaporated in vacuo to provide 4.5

4420 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14
Pryde et al.
with water, dried (MgSO4), then evaporated, and the residue purified
by flash chromatography on silica gel to afford the desired acid.
Alternatively, the tert-butyl ester was dissolved in either Et2O,
DCM, or EtOAc and HCl gas bubbled through the solution at room
temperature for 10 min. The solution was then purged by bubbling
through N2 gas for a further 10 min and then evaporated to dryness.
The residue was purified by flash chromatography on silica gel to
afford the desired acids.
General Procedure 4: Benzyl Ester/Carbamate Deprotection.
The requisite benzyl ester was dissolved in EtOH or MeOH (0.4M)
in a stainless steel bomb and a catalytic quantity of palladized
charcoal (either 5 or 10% concentration, 1-5% w/w with substrate)
added cautiously. H2 pressure was then applied to the vessel
(between 15 and 30 psi) and the contents magnetically stirred
overnight. The catalyst was filtered through a short plug of arbocel
(2 cm) and the filtrate evaporated to dryness under reduced pressure.
The residue was then purified by flash chromatography on silica
gel.
(1.1 equiv) in dry THF was added dropwise over 30 min, and
stirring was continued for a further 30min before the reaction was
allowed to slowly warm up to room temperature overnight. The
bright orange reaction mixture was cooled to 0 °C and quenched
with 2 N HCl. The solvent was removed in vacuo and the mixture
extracted with EtOAc. The combined extracts were washed with
brine, dried (MgSO4) and evaporated to an orange oil, which was
purified by flash chromatography (SiO2; EtOAc in pentane). The
products were isolated as pale yellow oils. Characterization data
are given in ref 20.
General Procedure 8: Installing the P1′ Substituent (Ho-
mologation of 61). Ozone (excess) was bubbled through a stirred
solution of allyl acid 6120 (1 equiv) in DCM at -50 °C for 30 min.
After this time, a persistent blue color developed, and oxygen gas
was then bubbled through the solution for 5 min before DMS (5
equiv) was added in one portion. The resulting colorless solution
was allowed to warm up slowly to room temperature overnight
under a nitrogen atmosphere and then evaporated to dryness. The
residue was redissolved in DCM, washed with water, dried
(MgSO4), and evaporated to afford a colorless oil of the aldehyde,
which was used without any further purification. To a stirred
solution of the phosphonium salt (2 equiv) in dry THF at 0 °C
under N2 was added a solution of nBuLi (2.5 M solution in hexanes,
2 equiv) and the resulting suspension stirred for 30 min. To the
resulting orange solution was added dropwise a solution of the
aldehyde from the previous step (1 equiv) in dry THF, and the
whole solution was stirred at room temperature for 4 h. The solvent
was evaporated, and the residue partitioned between DCM and 2
N HCl. The organic layer was separated, dried (MgSO4), and
evaporated to a solid, which was purified by flash column
chromatography (SiO2; EtOAc in pentane) to afford the desired
olefin; R ) CHMe. This olefin (1 equiv) was dissolved in EtOH
and hydrogenated at 30 psi H2 pressure overnight at room
temperature with a catalytic amount of 10% Pd/C. The reaction
mixture was filtered through a short plug of Arbocel, and the plug
was then washed with EtOAc (×2) and concentrated in vacuo to
afford the requisite alkane as a colorless oil, which was found to
be sufficiently pure to be used with no further purification.
General Procedure 9: Oxadiazole Formation. The acid
chloride derived from 7 was taken up in DCM and triethylamine
(2 equiv) and hydrazine hydrate (1 equiv) added in one portion.
After 3 h of stirring at room temperature, the reaction was quenched
with water, the organic layer separated, dried, and evaporated to a
yellow oil of the acyl hydrazide. The acyl hydrazide was taken up
in DCM, triethylamine (1.5 equiv), and HOBt (1.02 equiv), and
WSCDI (1.3 equiv) and the acid (1 equiv) added in one portion.
The mixture was stirred at room temperature for 16 h and then
washed with water and brine, dried (MgSO4), and evaporated to a
gum, which was purified by flash column chromatography (SiO2;
EtOAc in pentane) to give the bisacyl hydrazide. This was taken
up in THF and 1.5 equiv of Burgess’ reagent added in one portion,
and the whole solution refluxed for 16 h. The reaction mixture was
partitioned between 2% NaHCO3 solution and EtOAc, and the
organic layer was washed with water and brine, and then dried
(MgSO4) and evaporated to provide the oxadiazole analogue.
General Procedure 10: Preparation of Inverted Amides
From (7). The acid 720 (1 equiv) was taken up in dioxan (0.4 M)
under nitrogen and triethylamine (2 equiv) and DPPA (1 equiv)
added dropwise. The solution was heated at reflux for 2.5 h before
benzyl alcohol (1.5 equiv) was added, and the reflux continued
overnight. After cooling to room temperature, the mixture was
quenched with 1 N NaOH and extracted with EtOAc (×3), dried
(MgSO4), and evaporated to give a crude oil, which was purified
by flash chromatography (SiO2; EtOAc (10%) in pentane) to give
the benzyl carbamate. This was deprotected according to general
procedure 4. The resulting amine was then coupled to the requisite
acid according to general procedure 1.
General Procedure 5: 2-Amino-1,3,4-thiadiazole Ring For-
mation. The requisite acid chloride (2 equiv) (obtained either
commercially or by treatment of the acid with (COCl)2 in DCM)
was heated with thiosemicarbazide (1 equiv) in the absence of
solvent at 40 °C for 4 h. During this time, all evolution of gas
ceased, and water was added. The crude reaction mixture was
basified to pH 12 with a 50% aqueous solution of NaOH. The
solution was then concentrated in vacuo and azeotroped several
times, first with toluene and then EtOAc to remove traces of water.
The residue was then purified by silica gel chromatography using
3%, 5%, and then 10% MeOH in DCM as the eluant to provide
the amino-thiadiazole as a white solid (typically 80-90% yield)
along with small quantities of the N-acylated material. Provided
the temperature is maintained below 40 °C and reaction times below
6 h, the formation of this side-product is largely avoided.
General Procedure 6: cis-1,2-Disubstituted Cyclohexane
Formation. Cyclohexanone (1 equiv) was added to a stirred solution
of lithium diisopropylamide (1 equiv) in dry THF (0.4 M total
concentration) at -78 °C under nitrogen and the mixture stirred at
this temperature for 45 min. A solution of the electrophile (1 equiv)
in dry THF was added dropwise over 30 min, and stirring was
continued for a further 30 min before the reaction was allowed to
slowly warm up to room-temperature overnight. The bright orange
reaction mixture was cooled to 0 °C and quenched with 2 N HCl.
The solvent was removed in vacuo and the mixture extracted with
EtOAc. The combined extracts were washed with brine, dried
(MgSO4), and evaporated to an orange oil, which was purified by
flash chromatography (SiO2; Et2O in pentane). The purified oil was
taken up in toluene and R-1-phenyl-ethylamine (1 equiv) added in
one portion. p-TsOH (cat.) was added in one portion and the whole
solution refluxed overnight in a flask fitted with a Dean and Stark
apparatus. The solvent was then evaporated to a white solid, which
was taken up in EtOH and RaNi (cat.) added in one portion and
the mixture hydrogenated at 4 bar H2 pressure and room temperature
for 2 days. A further portion of RaNi was added, and hydrogenation
continued for a further 2 days. Water was added and the crude
mixture filtered through a short plug of Arbocel, and the solid was
washed with water and EtOH. Evaporation provided an oil, which
was purified by flash chromatography (SiO2; MeOH (5%) in DCM)
to yield a clear oil, which was dissolved in EtOH, and HCl gas
was bubbled through the solution for 30min. The mixture was
evaporated to low volume and then flooded with ether to precipitate
a white solid, which was collected by filtration. This solid was
dissolved in EtOH and hydrogenated at 4 bar H2 pressure with 10%
Pd/C (cat.) as the catalyst at 45 °C overnight. Filtration through a
short plug of Arbocel and evaporation of the filtrate provided a
white solid of the cis-2-alkyl-1-amino-cyclohexylamine, which
could be coupled to 7 with no further purification.
General Procedure 7: Installing the P1′ Substituent (Al-
kylation of 57). Glutarate 5719,20 (1 equiv) was added to a stirred
solution of lithium diisopropylamide (2.2 equiv) in dry THF (0.27
M total concentration) at -78 °C under nitrogen and the mixture
stirred at this temperature for 45 min. A solution of the electrophile
4-Amino-1-ethyl-1,2,3-triazole (25). Step 1: 24. 4-Nitro-1,2,3-
triazole (96 mg, 0.84 mmol) was dissolved in dry DMF (4 mL) at
0 °C under a nitrogen atmosphere before the portionwise addition
of a 60% suspension of NaH in mineral oil (35 mg, 0.88 mmol)

SelectiVe Glutaramide NEP Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14 4421
over 10 min, followed by excess ethyl iodide. The reaction was
allowed to reach room temperature and stirred for a further 3 h
and then quenched by the addition of water (3 mL). The quenched
mixture was extracted with EtOAc (2 × 30 mL), and the combined
organic fractions were dried (MgSO4) and evaporated to an oily
residue. This residue was purified by silica gel chromatography
(SiO2; EtOAc (20%) in pentane), to provide the desired 1-ethyl
isomer (284 mg, 57%) and the 3-isomer (106 mg, 21%), both as
clear oils.
Step 2: 25. The product from step 1 (100 mg, 0.70 mmol) was
dissolved in EtOH (8 mL) and placed in a stainless steel
hydrogenation vessel. Then, 10 mg of 10% palladized charcoal was
added carefully to the solution, and the vessel was then pressurized
to 15 psi with hydrogen gas. The vessel was warmed to ca. 40 °C,
the contents stirred for 4 h, and then allowed to cool to room
temperature and vented. The reaction mixture was filtered through
a 2 cm plug of arbocel, and the filtrate evaporated to a clear oil.
This oil was purified by flash chromatography (SiO2; MeOH (5%)
in DCM) to afford the desired amino triazole as a clear oil (92 mg,
98%).
2-Amino-4-propylamino-pyrimidine (27). Step 1. 2-Amino-
4-chloro-pyrimidine. 2-Amino-4-hydroxy-pyrimidine (10 g, 80
mmol) and POCl3 (15 mL, 160 mmol) were refluxed together for
1.5 h. The reaction mixture was poured onto ice and then basified
to pH 9 by the addition of solid Na2CO3. The solution was extracted
with EtOAc (2 × 400 mL) and DCM (300 mL). The combined
organics were dried over anhydrous MgSO4, filtered, and evaporated
to give an off-white solid, which was purified by flash chroma-
tography (SiO2; EtOAc (40 f 100%) in pentane) to give a white
solid of the chloropyrimidine (5.9 g, 52%).
Step 3: 32. The product from step 2 (700 mg, 3.5 mmol) and
5% Pd/C (70 mg) in 1 M HCl (5 mL) and EtOH (20 mL) were
hydrogenated (30 psi H2, room temperature) for 6 h. The mixture
was filtered through a 2 cm plug of Arbocel and the filtrate
concentrated in vacuo and basified using aqueous NaHCO3 solution.
The organics were then extracted with DCM (3 × 50 mL), dried
(MgSO4), and evaporated under reduced pressure. The crude product
was purified by flash chromatography (SiO2; MeOH (8%) in DCM
containing 0.4% NH3 solution) to give the amine as a solid (500
mg, 78%); mp 107-109 °C.
2-Amino-4-butyl pyridine (34). A mixture of 4-butyl pyridine
(5 g, 37 mmol) and 95% NaNH2 (1.7 g, 41 mmol) in xylene (10
mL) was heated at 150 °C for 18 h. The cooled mixture was diluted
with ether (100 mL) and extracted with 2 N HCl (×2). The extracts
were basified with NaOH and re-extracted with ether. The combined
extracts were dried (MgSO4) and evaporated under reduced pressure
to an oil, which was purified by flash chromatography (SiO2; MeOH
(3%) in DCM containing 0.15% NH3 solution) to a white solid
(2.1 g, 38%).
5-Amino-1-benzyl-2(1H)-pyridone (36). A mixture of 1-benzyl-
5-nitro-1H-pyridin-2-one (1 g, 4.4 mmol) and granulated Sn (3.5
g, 30 mmol) were taken up in cHCl (14 mL) and heated at 90 °C
for 1.5 h. The cooled solution was diluted with water (25 mL),
neutralized with a Na2CO3 solution, and extracted with EtOAc (250
mL). The extract was dried (MgSO4) and evaporated to give the
aminopyridone as a pale green solid (440 mg, 51%).
2-Aminomethyl-5-methyl-1,3,4-thiadiazole Hydrochloride (45).
Step 1. N-BOC glycine (5 g, 28.5 mmol) was dissolved in DCM
(75 mL) at room temperature. EEDQ (7.96 g, 28.5 mmol) was added
in one portion, followed by acetic hydrazide (2.6 g, 34.2 mmol)
and the solution left to stir overnight at room temperature. The
resulting solid was filtered off and dried under high vacuum to
provide 2.42 g (37%) of the bisacyl hydrazide as white crystals.
Step 2: 45. The hydrazide from step 1 (500 mg, 2.16 mmol)
was dissolved in dry THF (40 mL) at room temperature and
Lawesson’s reagent (960 mg, 2.4 mmol) added in one portion. The
resulting suspension was then heated to reflux for 3 h and then
allowed to cool to room temperature over 16 h. The volatiles were
evaporated at reduced pressure and the resulting clear oil purified
by flash chromatography (SiO2; EtOAc (70%) in pentane) to provide
the thiadiazole as a clear oil (483 mg, 91%), which crystallized on
standing. This was further purified by dissolving in 100 mL of
EtOAc and adding approximately 2 g of decolorizing charcoal. After
stirring for 10 min, the mixture was filtered under a slight vacuum
and the solvent evaporated to provide pure product (441 mg) as a
white crystalline solid. This was taken up in DCM at 0 °C, and a
slow stream of HCl gas bubbled through the solution via a
micropipet for 15 min. The HCl stream was turned off and replaced
with a slow stream of N2 gas to purge most of the HCl from the
solution. After 1 h of N2 bubbling, the solvent was evaporated in
vacuo to provide a white solid of the amine hydrochloride salt in
a quantitative yield.
3-Amino-cyclohexane Carboxylic Acid Dimethyl Amide (47).
3-Amino-1-cyclohexane carboxylic acid (1 g, 6.9 mmol) was added
to a mixture of 7 mL of 1 N NaOH solution in 7 mL of dioxan.
The mixture was then cooled to 0 °C and di-tert-butyl-dicarbonate
(1.67 g, 7.68 mmol) was added in one portion, followed by a slow
warming up to room temperature overnight. The mixture was
evaporated to a low volume (approximately 5 mL) in vacuo, and
the aqueous solution was acidified with 2 N HCl solution to pH 1
and extracted with EtOAc (3 × 10 mL), and the extracts were
combined, dried (MgSO4), and concentrated in vacuo to afford a
white solid (1.43 g). This solid was combined with dimethylamine
(1.5 mL, 33% solution in absolute ethanol, 5.6 mmol), WSCDI
(1.19 g, 6.2 mmol), HOBt (0.84 g, 6.2 mmol), and NMM (1.11
mL, 10.1 mmol) in DMF (30 mL) and stirred at room temperature
for 16 h. The mixture was then evaporated to dryness in vacuo
and chromatographed directly on silica gel using 5% MeOH in
DCM as the eluant to provide 997 mg of a clear oil. This oil was
dissolved in DCM (8 mL) and TFA (8 mL) and the whole solution
was stirred at room temperature for 4h. The reaction mixture was
Step 2: 27. The product from step 1 (900 mg, 6.3 mmol) was
n
heated at 90 °C with Pr amine (655 µL), Hunigs base (2.2 mL,
12.6 mmol), and DMF (10 mL) under N2 for 16 h. The cooled
reaction mixture was diluted with EtOAc (20 mL) and then washed
with water (4 × 10 mL). The organic layer was dried (MgSO4),
filtered, and evaporated to give an orange oil, which was purified
by flash column chromatography (SiO2; EtOAc (50%) in pentane,
then MeOH (5%) in DCM) to a colorless solid (245 mg, 26%) of
the amine.
3-Amino-6-methoxy-pyrazine (29). Step 1: 3-Amino-6-chloro-
pyrazine. 3,6-Dichloropyrazine (500 mg, 3.36 mmol) was taken
up in a mixture of DCM (4 mL) and NH4OH solution (2 mL) and
warmed at 35 °C for 6 h. The mixture was evaporated to dryness
in vacuo and used in the next step without further purification.
Step 2: 29. The product from step 1 was taken up in MeOH
(10 mL) and NaOMe (5 eq) added in one portion, and the whole
solution was heated at 140 °C in a stainless steel bomb for 4 h.
The reaction mixture was evaporated to dryness in vacuo and then
purified by column chromatography using 2% MeOH in DCM as
the eluant to give the product as an off-white solid. Yield 94%.
n
2-Amino-4-benzyl-pyridine (32). Step 1: 31. BuLi (17 mL,
2.5 M in hexanes, 43 mmol) was added dropwise to a cooled (-78
°C) solution of 3,5-dibromopyridine (10 g, 42 mmol) in ether (200
mL) to maintain an internal temperature of < -70 °C. The mixture
was stirred for 15 min before a solution of benzaldehyde (4.5 g,
43 mmol) in 20 mL of ether was added dropwise and the mixture
stirred for 15 min before the cooling bath was removed and the
contents allowed to warm to room temperature over 1 h. The
reaction was quenched by the addition of an aqueous NH4Cl
solution (200 mL) and the organics extracted with ether, dried, and
evaporated. The residual yellow oil was purified by flash chroma-
tography (SiO2; MeOH (5 f 20%) in DCM) to give the alcohol as
a pale yellow oil (7.6 g, 68%).
Step 2. The product from step 1 (2 g, 7.6 mmol) and CuSO4‚
5H2O (350 mg, 1.4 mmol) in 0.88 NH3 (18 mL) was heated at 135
°C in a sealed tube for 24 h. A NaOH solution (1 M, 10 mL) was
added to the cooled solution and the mixture extracted with ether
(6 × 20 mL). The combined extracts were dried (MgSO4) and
evaporated to low volume. The resulting precipitate was filtered,
washed with ether, and dried to give the aminopyridine (1.3 g, 83%)
as a white solid; mp 92-94 °C.

4422 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14
Pryde et al.
Table 13. Variability of the Primary NEP Inhibition Assay
concentrated in vacuo, DCM (20 mL) was added, and the solution
was washed with water (2 × 25 mL) and a saturated solution of
NaHCO3 (25 mL). The aqueous washes were combined, concen-
trated in vacuo to provide a white solid, which was purified by
chromatography using 84:14:2 DCM/MeOH/NH3 as the eluant to
provide 47 as a colorless oil (346 mg).
2-[1-(5-Ethyl-[1,3,4]thiadiazol-2-ylcarbamoyl)-cyclopentyl-
methyl]-4-phenyl-butyric Acid (59). This compound was prepared
in a fashion identical to that used to prepare compound 61 but using
phenethyl bromide in place of allyl bromide to give intermediate
58, which was then processed according to Scheme 1 to give the
title compound as a clear oil.
3-[1-(5-Ethyl-[1,3,4]thiadiazol-2-ylcarbamoyl)-cyclopentyl]-
propionic Acid (60). This compound was prepared directly from
57 according to Scheme 1 to give the title compound as a clear oil.
2-[(1-{[(1-Benzyl-6-oxo-1,6-dihydro-3-pyridinyl)-(methyl)-
amino]-carbonyl}-cyclopentyl)-methyl]-pentanoic Acid (71). The
benzyl ester of (37) (120 mg, 0.24 mmol) was dissolved in dry
THF (3 mL) and cooled to 0 °C under nitrogen. A suspension of
NaH (19 mg, 0.48 mmol, 60% suspension in oil) was added in one
portion and the mixture stirred for 1 h at 0 °C. Methyl iodide (1.1
equiv, 0.30 mmol) was added in one portion, and the reaction
mixture was stirred at 0 °C for 30 min and then allowed to warm
up to room temperature over 3 h. The solution was concentrated to
low volume (ca. 1 mL) and treated with 4 mL of saturated aqueous
NaHCO3 solution. The organics were extracted with EtOAc (2 ×
10 mL), dried (MgSO4), and evaporated to give a gum. This was
purified by flash column chromatography (SiO2; EtOAc (35 f
70%) in pentane) to give the methyl amide as a clear oil (26 mg,
21%). This was then deprotected according to general procedure
4.
General Procedure 11: HPLC Resolutions. HPLC resolutions
were performed with either a Chiralpak AD or OD column, 25 ×
2 cm at ambient temperature and using mixtures of hexane, IPA,
and TFA as the solvent (typically, 70% hexane, 30% IPA, and 0.1%
TFA). The flow rate was typically 10 mL/min, and the runs lasted
for 20 min, and detection was carried out at 220 nm. HPLC resolved
material was typically of >98%ee and was obtained from the HPLC
fractions by evaporation to low volume, azeotroping with several
aliquots of toluene followed by slow crystallization from diisopropyl
ether at -4 °C. Simple filtration and drying in vacuo provided the
R-enantiomeric material in typically 40-45% yield from the
racemate.
The racemic acid (13, 824 mg) was resolved by HPLC using an
AD column (hexane/IPA/TFA (85:15:0.2), 10 mL/min) as eluant
to give the S enantiomer (RT 7.6 min) as a white powder (400 mg)
in 99.5%ee; [R]D +3.8° (c 0.1 in MeOH) and the R enantiomer
(RT 6.2min) as a white powder (386 mg) in 99.5%ee; [R]D -9.0°
(c 0.1 in MeOH).
The racemic acid (10, 620 mg) was resolved by HPLC using an
AD column (hexane/IPA/TFA (85:15:0.1), 10 mL/min) as the eluant
to give the S enantiomer (RT 7.5 min) as a white powder (298 mg)
in 94.5%ee and the R enantiomer (RT 6.6min) as a white powder
(261 mg) in 99.2%ee; [R]D -11.9° (c 1.0 in MeOH).
The racemic acid (21, 204 mg) was resolved by HPLC using an
OD column (hexane/IPA/TFA (95:5:0.1), 10 mL/min) as the eluant
to give the R enantiomer (RT 11.8 min) as a white powder (83 mg)
in 99.5%ee; [R]D -10.9° (c 0.46 in MeOH) and the S enantiomer
(RT 9.8 min) as a white powder (62 mg) in 99.5%ee; [R]D +10.4°
(c 0.67 in MeOH).
NEP Inhibition Assay. In a 96-well microtiter plate, 100 mL
of fluorescent NEP substrate peptide (50 mM o-aminobenzoyl-D-
Arg-Arg-Leu-ethylenediamine 2,4-dinitrophenyl) was added to 50
mL of 4 times the test concentration of inhibitor solutions. The
assays were initiated by the addition of 50 mL of enzyme, which
had been previously diluted such that with an incubation time of 1
h the percentage of substrate converted to product was ap-
proximately 10% or less. The reaction was incubated for 1 h at 37
°C in a shaking incubator and then stopped by the addition of 100
mL of 300 nM phosphoramidon. Samples corresponding to 100%
substrate to product conversion were included to enable the percent
substrate proteolyzed to be determined. Each assay point was in
duplicate. Each IC50 experiment used a 10 point dilution series at
half-logarithmical inhibitor concentration increments, with the
highest concentration either 10, 1, or 0.1 mM. After the 1 h of
incubation, the fluorescence of the samples was measured (Ex320
/
Em420) using a BMG Fluostar Galaxy reader and an IC50 value
calculated using a customized Microsoft Excel add-in. The enzyme
and substrate were prepared in 50 mM Tris‚Cl, (pH 7.4 at 37 °C),
and 4× inhibitor dilutions were made in 50 mM Tris‚Cl at pH 7.4
at 37 °C/4% DMSO. Standard compounds assessed in this assay
were as follows: phosphoramidon 0.71 nM, thiorphan 3.85 nM,
and Candoxatrilat 2.61 nM.
The IC50 estimates stated for NEP represent the geometric mean
of at least 2 values obtained using separately weighed and
solubilized compound samples. The NEP assay was highly repro-
ducible. We demonstrated this from a statistical analysis of 25 IC50
values obtained from a single compound, which was tested multiple
times over an 8 month period. By assessing the assay-to-assay
variation, we were able to conclude that two replicates would be
sufficient to ensure that the geometric mean IC50 was estimated
with less than 2-fold error, on the basis of the width of the 95%
confidence interval.
ACE Inhibition Assay. This assay was performed as described
above for the NEP assay, except that the substrate used was
o-aminobenzoyl-Gly-Gly-p-nitro-Phe-Pro-OH at a final assay con-
centration of 10 mM, and the enzyme and substrate buffer were 50
mM Tris‚Cl at pH 7.4 (at 37 °C) and 300 mM NaCl, and the
reaction was terminated using 100 mL of a 2 mM EDTA solution.
ECE-1 Inhibition Assay. Human recombinant ECE-1 was
harvested from the growth media of Chinese Hamster Ovary (CHO)
cells transiently transfected with an expression plasmid encoding
the extracellular domain of ECE-1 (amino acids 73-753). ECE-1
was assayed in 100 mM HEPES-KOH at pH 7.0/1% DMSO using
1 µM big ET-1 peptide as the substrate. Following a 1 h incubation
at 37 °C, the assay was terminated by the addition of an equal
volume of 10 mM EDTA and the conversion of big ET-1 to ET-1
measured using an ELISA (Amersham Pharmacia RPN 228). The
quantity of enzyme to use was determined as the dilution giving a
final ELISA OD reading of approximately 1.0. Inhibitors were
included in the assay at a range of concentrations, and their effects
measured as % inhibition.
Pharmacokinetic Studies. Male CD rats (Charles River, Man-
ston, UK) were used. These were jugular vein cannulated under
general anaesthesia, with the cannula exteriorized at the back of
the neck. At specified times after dosing, 0.2 mL samples of blood
were withdrawn from the cannula and transferred to heparin tubes.
Following adequate mixing, the blood samples were centrifuged,
and plasma was generated. Immediately after that the plasma was
stabilized by the addition of 0.01 parts of phosphoric acid (85%,
v/v) and stored deep frozen until analysis. The fraction unbound
in rat plasma was determined at 1 µg/mL using equilibrium dialysis
against an isotonic Krebs-Ringer bicarbonate buffer (pH 7.4). The
concentrations of glutaramides in rat plasma were determined by
LC-MS-MS (Hewlett-Packard HP1100 binary HPLC pump, CTC

SelectiVe Glutaramide NEP Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14 4423
(10) Dantzig, A. H. Oral Absorption of â-Lactams by Intestinal Peptide
Transport Proteins. AdV. Drug DeliVery ReV. 1997, 23, 63-76.
(11) Devault, A.; Lazure, C.; Nault, C.; Le Moual, H.; Seidah, N. G.;
Chretien, M.; Kahn, P.; Powell, J.; Mallet, J.; Beaumont, A.; Roques,
B. P.; Crine, P.; Boileau, C. Amino Acid Sequence of Rabbit Kidney
Neutral Endopeptidase 24.11 (Enkephalinase) Deduced from a
Complementary DNA. EMBO J. 1987, 6, 1317-1322.
(12) Kerr, M. A.; Kenny, A. J. Purification and Specificity of a Neutral
Endopeptidase from Rabbit Kidney Brush Border. Biochem J. 1974,
137, 477-488.
(13) Oefner, C.; D’Arcy, A.; Henning, M.; Winkler, F. K.; Dale, G. E.
Structure of Human Neutral Endopeptidase (Neprilysin) Complexed
with Phosphoramidon. J. Mol. Biol. 2000, 296, 341-349.
(14) (a) Roderick, S. L.; Fournie-Zaluski, M. C.; Roques, B. P.; Matthews,
B. W. Thiorphan and Retro-Thiorphan Display Equivalent Interac-
tions When Bound to Crystalline Thermolysin. Biochemistry 1989,
28, 1493-1497. (b) Matthews, B. W. Structural Basis of the Action
of Thermolysin and Related Zinc Peptidases. Acc. Chem. Res. 1988,
21, 333-340.
(15) Tiraboschi, G.; Jullian, N.; Thery, V.; Antonczak, S.; Fournie-Zaluski,
M. C.; Roques, B. P. A Three-Dimensional Construction of the Active
Site (Region 507-749) of Human Neutral Endopeptidase
(EC.3.4.24.11). Protein Eng. 1999, 12, 141-149.
Pal Autosampler, Sciex API 2000 mass spectrometer with Tur-
boIonSpray interface) at a flow rate of 1 mL/min, split 50:1
postcolumn using an Acurate flow splitter. Test compounds were
extracted from 100 µL of acidified plasma mixed with 400 µL of
50 mM ammonium formate buffer (pH 3.5) using activated C18
IST cartridges in 96-well format (Porvair). The cartridges were
washed with 1 mL of 50 mM ammonium formate (pH 3.5). The
compounds were eluted with 1 mL of acetonitrile containing 5%
(v/v) formic acid and evaporated to dryness under nitrogen at 20
°C. The residues were resuspended in 200 µL of 2 mM ammonium
acetate in 70:30 (v/v) methanol/water (pH 3.5), and 180 µL was
injected into the HPLC mass spectrometer. The mobile phase was
2 mM ammonium acetate in methanol/water (90/10; v/v; pH 3.5),
and the column was a Hypersil HS100 C18, 5 µm, 50 × 4.6 mm.
Detection was by positive ion multiple reaction monitoring (MRM)
at Q1 and Q3 resolution of ca. 0.7 Da peak width at half-height
(for R-13, Q1 340, Q3 130). The curtain gas, nebulizer gas, and
TurboIonSpray gas were nitrogen at settings of 30 (CUR), 25 (GS1),
and 40 (GS2), respectively. TurboIonSpray temperature was 100
°C. The collision gas was nitrogen at a setting of 2. The collision
energy was 25 eV (OR 30 V), and the dwell time was 200 ms with
50 ms pause. All other data was obtained using standard methods.
(16) For a recent study of thiorphan bound in a human NEP construct
see Sahli, S.; Frank, B.; Schweizer, W. B.; Diederich, F.; Blum-
Kaelin, D.; Aebi, J. D.; Bohm, H.-J.; Oefner, C; Dale, G. E. Second
Generation Inhibitors for the Metalloprotease Neprilysin Based on
Bicyclic Heteroaromatic Scaffolds: Synthesis, Biological Activity
and X-ray Crystal Structure Analysis. HelV. Chim. Acta 2005, 88,
731-750.
Supporting Information Available: Full experimental details
and characterization data. This material is available free of charge
via the Internet at http://pubs.acs.org.
(17) Holland, D. R.; Barclay, P. L.; Danilewicz, J. C.; Matthews, B. W.;
James, K. Inhibition of Thermolysin and Neutral Endopeptidase 24.11
by a Novel Glutaramide Derivative: X-ray Structure Determination
of the Thermolysin-Inhibitor Complex. Biochemistry 1994, 33, 51-
56.
(18) Subsequent studies have investigated cocrystal structures of propri-
etary inhibitors bound in recombinant human NEP. These studies
and compound designs derived from them will be disclosed else-
where.
(19) Challenger, S.; Derrick, A.; Mason, C. P.; Silk, T. Stereoselective
Synthesis of a Candoxatril Intermediate via Asymmetric Hydrogena-
tion Tetrahedron Lett. 1999, 40, 2187-2190.
(20) Barnish, I. T.; James, K.; Terrett, N. K.; Danilewicz, J. C.; Samuels,
G. M. R.; Wythes, M. J. Eur. Pat. Appl. EP 274234, 1988.
(21) Maiorana, S.; Pocar, D.; Dalla Croce, P. Studies in the Enamine
Field: Reactions of Sulfonyl and Nitro-Enamines with Azides.
Tetrahedron Lett. 1966, 48, 6043-6045.
References
(1) Maw, G. N.; Stobie, A.; Planken, S.; Pryde, D. C.; Sanderson, V.;
Platts, M. Y.; Corless, M.; Stacey, P.; Wayman, C.; Van Der Graaf,
P.; Kohl, C.; Coggon, S.; Beaumont, K. The Discovery of Small
Molecule Inhibitors of Neutral Endopeptidase: Structure-Activity
Studies on Functionalised Glutaramide. Chem. Biol. Drug Des. 2006,
1, 74-77.
(2) Leiblum, S. R. Definition and Classification of Female Sexual
Disorders. Int. J. Impotence Res. 1998, 10, S104-S106.
(3) (a) Frank, E.; Anderson, C.; Rubinstein, D. Frequency of Sexual
Dysfunction in ‘Normal’ Couples. N. Eng. J. Med. 1978, 299, 111-
115. (b) Rosen, R. C.; Taylor, J. F.; Leiblum, S. R.; Bachmann, G.
A. Prevalence of Sexual Dysfunction in Women. J. Sex Marital Ther.
1993, 19, 171-188.
(4) (a) Goldstein, I.; Berman, J. R. Vasculogenic Female Sexual
Dysfunction: Vaginal Engorgement and Clitoral Erectile Insuf-
ficiency Syndromes. Int. J. Impotence Res. 1998, 10, S84-S90. (b)
Park, K.; Goldstein, I.; Andry, C.; Siroky, M. B.; Krane, R. J.;
Azadzoi, K. M. Vasculogenic Female Sexual Dysfunction: The
Haemodynamic Basis for Vaginal Engorgement and Clitoral Erectile
Insufficiency. Int. J. Impotence Res. 1997, 9, 27-37.
(22) Viscardi, G.; Savarino, P.; Quagliotto, P.; Barni, E.; Botta, M. Novel
Heterocyclic Ligands with Tuned Hydrophobicity. J. Heterocycl.
Chem. 1996, 33, 1195-1200.
(23) Meerwein, H.; Hinz, G. Isomers of 2-Oxypyridine Derivatives. Justus
Liebigs Ann. Chem. 1930, 484, 52-64.
(5) (a) Ottesen, B.; Gerstenberg, T.; Ulrichsen, H.; Manthorpe, T,
Fahrenkrug, J.; Wagner, G. Vasoactive Intestinal Polypeptide (VIP)
Increases Vaginal Blood Flow and Inhibits Uterine Smooth Muscle
Activity in Women. Eur. J. Clin. InVest. 1983, 13, 321-324. (b)
Ottesen, B.; Wagner, G.; Fahrenkrug, J. In Handbook of Sexology;
Sitsen, J. M. A., Ed.; Elsevier: Amsterdam, 1988; Vol. 6, pp 66-
97. (c) Ottesen, B.; Pedersen, B.; Nielsen, J.; Dalgaard, D.; Wagner,
G.; Fahrenkrug, J. Vasoactive Intestinal Polypeptide (VIP) Provokes
Vaginal Lubrication in Normal Women. Peptides 1987, 8, 797-800.
(6) Loffler, B.-M. ACE, NEP, ECE-1: Selective and Combined Inhibi-
tion. Curr. Opin. CardioVasc., Pulm. Renal InVest. Drugs 1999, 1,
352-364.
(7) See for example, (a) De Lombaert, S.; Chatelain, R. E.; Fink, C. A.;
Trapani, A. J. Design and Pharmacology of Dual Angiotensin-
Converting Enzyme and Neutral Endopeptidase Inhibitors. Curr.
Pharm. Des. 1996, 2, 443-462. (b) Robl, J. A.; Sulsky, R.; Sieber-
McMaster, E.; Ryono, D. E.; Cimarusti, M. P.; Simpkins, L. M.;
Karanewsky, D. S.; Chao, S.; Asaad, M. M.; Seymour, A. A.; Fox,
M.; Smith, P. L.; Trippodo, N. C. Vasopeptidase Inhibitors: Incor-
poration of Geminal and Spirocyclic Substituted Azepinones in
Mercaptoacyl Dipeptides. J. Med. Chem. 1999, 42, 305-311 and
references therein.
(24) (a) Tadakazu, T.; Keiko, T. Convenient Synthesis of 2,7-Disubstituted
5H-1,3,4-Thiadiazolo[3,2-a]-Pyrimidin-5-ones and Related Com-
pounds. Bull. Chem. Soc. Jpn. 1982, 55, 637-638. (b) Steahly, G.
W. US Patent 2,422,050, 1947.
(25) Frahm, A. W.; Knupp, G. Asymmetric Synthesis of cis-2-Substituted
Cyclohexanamines with High Optical Purity. Tetrahedron Lett. 1981,
22, 2633-2636.
(26) Knupp, G.; Frahm, A. W. Synthesis and Absolute Configuration of
2-Substituted Cyclohexylamine. Chem. Ber. 1984, 117, 2076-2098.
(27) Medeiros, M. A.; Franca, M. S.; Boileau, G.; Juliano, L.; Carvalho,
K. M. Specific Fluorogenic Substrates for Neprilysin (Neutral
Endopeptidase, EC 3.4.24.11) which are Highly Resistant to Serine-
and Metalloproteases. Braz. J. Med. Biol. Res. 1997, 30, 1157-1162.
(28) Booth, A. G.; Kenny, A. J. A Rapid Method for the Preparation of
Microvilli from Rabbit Kidney. Biochem. J. 1974, 142, 575-581.
(29) (a) All LogD values were measured at pH 7.4. (b) Stopher, D.;
McClean, S. An Improved Method for the Determination of Distribu-
tion Coefficients. J. Pharm. Pharmacol. 1990, 42, 144.
(30) (a) Tanaka, Y.; Fujiwara, T.; Esumi, Y. Sex Difference in the
Excretion of Zenarestat in Mice, Rats, Dogs and Humans. Xenobiotica
1992, 22, 941-947. (b) Terashita, S.; Sawamoto, T.; Deguchi, T.;
Tokuma, Y.; Hata, T. Sex-Dependent and Independent Renal
Excretion of Nilvadipine Metabolites in Rat: Evidence for a Sex-
Dependent Active Secretion in Kidney. Xenobiotica 1995, 25, 37-
47. (c) Hanada, K.; Fujisawa, R.; Kataoka, R.; Nakamura, S.; Ogata,
H. Sex-Related Differences in the Renal Disposition of the Acidic
Metabolite of Clentiazem in Rats. Xenobiotica 2001, 31, 725-731.
(8) McDowell, G.; Nicholls, D. P. The Endopeptidase Inhibitor, Can-
doxatril, and Its Therapeutic Potential in the Treatment of Chronic
Cardiac Failure in Man. Expert Opin. InVest. Drugs 1999, 8, 79-
84.
(9) Goodman and Gillman’s The Pharmacological Basis of Therapeutics,
10th ed.; Hardman, J., Limbird, L., Eds.; McGraw-Hill Medical
Publishing Division: New York, 2001.

4424 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 14
Pryde et al.
(31) (a) Kato, Y.; Kuge, K.; Kusuhara, H.; Meier, P. J.; Sugiyama, Y.
Gender Difference in the Urinary Excretion of Organic Anions in
Rats. J. Pharmacol. Exp. Ther. 2002, 302, 483-489. (b) Gotoh, Y.;
Kato, Y.; Stieger, B.; Meier, P. J.; Sugiyama, Y. Gender Difference
in the Oatp1-Mediated Tubular Reabsorption of Estradiol 17â-D-
Glucuronide in Rats. Am. J. Physiol. 2002, 282, E1245-E1254.
(32) Mugford, C. A.; Kedderis, G. L. Sex-Dependent Metabolism of
Xenobiotics. Drug Metab. ReV. 1998, 30, 441-498.
(34) (a) Munarriz, R.; Kim, S. W.; Kim, N. N.; Abdulmaged, T; Goldstein,
I. A Review of the Physiology and Pharmacology of Peripheral
Female Genital Arousal in the Animal Model. 2003, J. Urol. 170,
S40-S45. (b) Maw, G. N; Wayman, C. P. Eur. Pat. Appl. EP
1097719, 2001. (c) Wayman, C.; Morren, D.; Turner, L.; Van der
Graaf, P.; Naylor, A. Inhibition of Neutral Endopeptidase Potentiates
Female Genital Blood Flow. Int. J. Impotence Res. 2002, 14, S7-S8.
(33) Czerniak, R. Gender-Based Differences in Pharmacokinetics in
Laboratory Animal Models. Int. J. Toxicol. 2001, 20, 161-163.
JM060133G