Novel complex crystal structure of prolyl hydroxylase domain-containing protein 2 (PHD2): 2,8-Diazaspiro[4.5]decan-1-ones as potent, orally bioavailable PHD2 inhibitors

Article abstract of DOI:10.1016/j.bmc.2013.08.046

We have discovered a novel complex crystal structure of the PHD2 enzyme with its inhibitor, the 2,8-diazaspiro[4.5]decan-1-one analogue 4b. The widely reported salt bridge between Arg383 of the enzyme and its inhibitors in all complex structures published thus far was not observed in our case. In our complex structure compound 4b forms several novel interactions with the enzyme, which include a hydrogen bond with Arg322, a π-cation interaction with Arg322, a π-π stacking with Trp389, and a π-π stacking with His313. Guided by the structural information, SAR studies were performed on the 2,8-diazaspiro[4.5]decan-1-one series leading to the discovery of compound 9p with high potency and good oral pharmacokinetic profile in mice.

Full text of DOI:10.1016/j.bmc.2013.08.046

Bioorganic & Medicinal Chemistry 21 (2013) 6349–6358
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Novel complex crystal structure of prolyl hydroxylase domain-
containing protein 2 (PHD2): 2,8-Diazaspiro[4.5]decan-1-ones as
potent, orally bioavailable PHD2 inhibitors
Guanghui Deng, Baowei Zhao, Yingli Ma, Qiongfeng Xu, Hailong Wang, Liuqing Yang, Qing Zhang,
Taylor B. Guo, Wei Zhang, Yang Jiao, Xin Cai, Jinqiang Zhang, Houfu Liu, Xiaoming Guan, Hongtao Lu,
⇑
Jianing Xiang, John D. Elliott, Xichen Lin, Feng Ren
Research and Development, GlaxoSmithKline Pharmaceuticals, 898 Halei Road, Zhangjiang Hi-tech Park, Pudong, Shanghai 201203, China
a r t i c l e i n f o
a b s t r a c t
Article history:
We have discovered a novel complex crystal structure of the PHD2 enzyme with its inhibitor, the 2,8-
diazaspiro[4.5]decan-1-one analogue 4b. The widely reported salt bridge between Arg383 of the enzyme
and its inhibitors in all complex structures published thus far was not observed in our case. In our com-
plex structure compound 4b forms several novel interactions with the enzyme, which include a hydrogen
Received 8 August 2013
Revised 24 August 2013
Accepted 24 August 2013
Available online 3 September 2013
bond with Arg322, a p-cation interaction with Arg322, a p–p stacking with Trp389, and a p–p stacking
with His313. Guided by the structural information, SAR studies were performed on the 2,8-diazaspi-
ro[4.5]decan-1-one series leading to the discovery of compound 9p with high potency and good oral
pharmacokinetic profile in mice.
Keywords:
2,8-Diazaspiro[4.5]decan-1-one
Prolyl hydroxylase domain-containing
protein 2 (PHD2)
Ó 2013 Elsevier Ltd. All rights reserved.
PHD inhibitors
Hypoxia inducible factor (HIF)
PHD2 crystal structure
SAR study
1. Introduction
of red blood cells (RBCs).9–13 Thus, inhibition of PHD enzymes has
emerged as an attractive strategy for the treatment of anemia.14
Prolyl hydroxylase domain-containing protein 2 (PHD2), also
known as egg-laying defective nine homologue 1 (EGLN1) or HIF
prolyl hydroxylase 2 (HPH2), is one of three known 2-oxoglutarate
(2OG) and Fe(II)-dependent oxygenases (PHD1-3).1 Primarily local-
ized in cytoplasm, PHD2 is highly expressed in heart and testis
while moderately expressed in brain, kidney, and liver.2 PHD2
plays an important role in cell adaptation and survival, responding
to the oxygen environment, by regulating the hypoxia-inducible
transcription factor (HIF).3,4 Under normoxia, hypoxia inducible
In this work, we have focused on PHD2, the most extensively
studied PHD enzymes and also the key oxygen sensor keeping
low steady-state levels of HIF1a
in vivo under normoxia.5,14 Quan-
tities of PHD2 enzyme were isolated and purified according to the
literature report15 for the high-throughput screening (HTS) of an
internal compound library. Spiro compounds 1 and 2 (Fig. 1) with
moderate/low inhibition potencies (pIC50 6.0 and 4.9, respectively)
in the PHD2 HTRF (homogeneous time resolved fluorescence) assay
were identified from the HTS. By searching the literature, we found
that similar structures (e.g., compound 3) have been reported11,16
and the compound (3) demonstrated potent enzymatic inhibition
(pIC50 9.3). As a continuous effort to improve compounds’ physico-
chemical properties such as reduced number of aromatic rings and
increased 3-dementionality,17 we decided to remove the phenyl
ring fused to the central spiro core of compound 3 for simplified
structures such as the 2,8-diazaspiro[4.5]decan-1-one derivatives
4a and 4b. To our delight, the new structure series showed reason-
able PHD2 inhibitory potency (pIC50 7.6 and 7.8 for 4a and 4b,
respectively) thus served as our chemistry starting point. Herein,
we will report the novel complex crystal structure of PHD2 enzyme
with 4b and the SAR studies to discover the 2,8-diazaspiro
factor 1a (HIF1a), an a-subunit of HIF, was reported to be rapidly
degraded after oxidation by PHD enzymes, which were identified
to use O2 and 2OG as co-substrates to generate 4-hydroxyprolines
at residues Pro402 and/or Pro564 of HIF1a.
5–8 Once PHD1-3 activ-
ities are inhibited, as they would be under hypoxia conditions or by
small molecules competing for 2OG or Fe(II) binding, the oxidative
degradation of HIFa subunits would be impeded resulting in the
up-regulation of HIF, which in turn leads to the up-regulation of
erythropoietin (EPO), an essential growth factor for the production
⇑
Corresponding author. Tel.: +86 021 61590721; fax: +86 021 61590730.
E-mail address: feng.q.ren@gsk.com (F. Ren).
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.08.046

6350
G. Deng et al. / Bioorg. Med. Chem. 21 (2013) 6349–6358
O
NH
N
N
N
N
N
N
N
N
S
O
O
N
N
O
1
, PHD2 pIC50: 6.0
2
, PHD2 pIC50: 4.9
4a
4b
, PHD2 pIC50: 7.6
N
N
N
N
N
N
N
O
O
, PHD2 pIC50: 7.8
3
, PHD2 pIC50: 9.3
Figure 1. HTS hits 1 and 2, literature compound 3 and newly designed compounds 4a and 4b.
[4.5]decan-1-one derivatives as potent, orally bioavailable PHD2
inhibitors.
soaking. In the complex structure, PHD2 has a double-stranded b-
helix fold common for 2OG-dependent dioxygenases, comprising
eight b-strands packed around by three a-helices similar to previ-
ous reports.15 There is a single Ni(II), in place of Fe(II), coordinated
in an octahedral manner by His313, Asp315, His374, compound 4b,
and a water molecule. Compound 4b chelates with Ni(II) through
the 3-N atom of the imidazole (1.9 Å from the metal) and the ter-
tiary amine of the piperidine (2.3 Å), and this metal chelation inter-
action is an important binding interaction between the PHD2
enzyme and its inhibitors.12,15
Besides metal chelation, the salt bridge interaction in the active
site between Arg383 of the PHD2 enzyme and the carboxylic acid
functional group of its inhibitors/natural ligand has been consis-
tently observed in all reported complex structures (Fig. 3), and this
interaction is considered to be critical for achieving potent inhibi-
tion activity.12,15,18 In our complex structure of PHD2 enzyme and
compound 4b, the salt bridge was not observed. Instead, we ob-
served several novel interactions. Compound 4b forms a strong
hydrogen bond (donor–acceptor distance at 3.0 Å, donor-hydro-
gen-acceptor angle at 162°) with Arg322 through its carbonyl
2. Results and discussion
2.1. Chemistry
The synthetic route for compounds 4a and 4b (and their close
analogues) was presented in Scheme 1. Copper(I)-catalyzed Ull-
mann coupling of commercial starting materials spiro lactam 5
and biphenyl iodide 6 provided intermediate 7. The tert-butoxycar-
bonyl (Boc) protective group was removed under the acidic
condition to afford amine 8. Reductive amination reactions of 8
with 3-methylpicolinaldehyde or 1-methyl-1H-imidazole-2-carb-
aldehyde provided the desired products 4a and 4b, respectively.
Compounds 4c–l, 9a–s were prepared under similar protocol. All
compounds (4a–l, 9a–s) were purified by mass directed auto-puri-
fication system (HPLC purity >95%) and their spectroscopic data
(HRMS, 1H NMR, and/or 13C NMR) were consistent with the
structure assignments.
group on the spiro core. Arg322 also forms a cation–
(distances between 3.3 and 3.9 Å) with the phenyl ring directly at-
tached to the spiro core. In addition, a stacking (edge-to-face,
p interaction
2.2. Crystallography
p–p
hydrogen–carbon distance at 2.6 Å, two rings are perpendicular)
between the same phenyl ring of 4b with Trp389 of PHD2 was
We firstly undertook structural studies to elucidate the mode of
inhibition of the series to guide SAR studies. Fortunately, we were
able to obtain the X-ray crystal structure of a truncated PHD2 en-
zyme (residues 181–418) complexed with compound 4b (Fig. 2) by
observed. Furthermore, a
p–p stacking (face-to-edge, nitrogen–
hydrogen distance at 2.6 Å, two rings are nearly perpendicular)
N
a
N
Boc
I
N
HN
Boc
O
b
O
7
6
5
c
N
N
NH
R
N
O
O
8
4a,
R =
R =
N
N
4b,
N
Scheme 1. General synthetic procedure for the 2,8-diazaspiro[4.5]decan-1-ones 4a, 4b, and analogues. Reagents and conditions: (a) CuI, K2CO3, N1,N2-dimethylethane-1,2-
diamine, Acetonitrile, 80 °C, 36%; (b) TFA, DCM, 23 °C, 5 h, 100%; (c) sodium triacetoxyborohydride, 3-methylpicolinaldehyde or 1-methyl-1H-imidazole-2-carbaldehyde, 2 h,
26% for 4a and 41% for 4b.

G. Deng et al. / Bioorg. Med. Chem. 21 (2013) 6349–6358
6351
Figure 2. Crystal structure of truncated PHD2 enzyme complexed with compound 4b (PDB: 4JZR). The structure and interaction diagram are drawn in Maestro [Schrödinger
2012R1].
of the imidazole moiety of 4b with His313 of PHD2 was also iden-
tified from the crystal structure. These
p–p stacking interactions
are usually weaker than hydrogen bonds or salt bridges. The bi-
phenyl moiety of 4b is exposed to solvent, and not much hydro-
phobic interaction can be identified. Compound 4b clearly does
not form salt bridges with Arg383 (more than 5 Å away in distance,
Fig. 2) which, besides the chelation interaction, was reported to be
critical for the potency of PHD2 inhibitors (Fig. 3).12,15,18 Without
this interaction, it is thus intriguing how compound 4b can achieve
such a high potency (pIC50: 7.8). To understand this, we studied the
crystal structure of PHD2 complexed with its substrate HIF-1
see whether compound 4b can interfere with their binding
(Fig. 4). HIF1 has several key interactions with PHD2, including
but not limited to the salt bridge between Asp571 of HIF1 and
Arg396 of PHD2 and also the hydrogen bonds between Pro564
a to
a
a
Figure 3. Crystal structure of PHD2 with co-substrate 2OG (PDB: 3OUJ12).

6352
G. Deng et al. / Bioorg. Med. Chem. 21 (2013) 6349–6358
Table 1
SAR on the right-hand side pyridine substitutions (4)
N
N
R
O
4
a
Compound
R
PHD2 pIC50
7.6
N
4a
N
4b
7.8
N
Figure 4. Crystal structure of PHD2 with substrate HIF1a
[PDB: 3HQR19] in overlay
4c
<5
<5
with compound 4b. PHD2 is shown in white, HIF1
a peptide in blue, and compound
4b in green.
N
4d
4e
4f
<5
(HIF1a hydroxylation site) and Arg322 of PHD2. By overlaying this
crystal structure with that of PHD2 complexed with compound 4b,
one can clearly see that compound 4b overlaps with both of these
two interactions as well as with Ile566 of HIF1a. This indicates
that, although compound 4b might not compete for the co-sub-
strate binding site (2OG) as strongly, it may compete extensively
N
N
6.8
6.7
N
4g
Cl
for the substrate (HIF1a) binding site. We speculate that it is the
N
combination of these two properties that confer high potency to
4h
6.1
F3C
this compound.
N
4i
4j
6.2
8.6
2.3. In vitro pharmacology
MeO
N
N
2.3.1. SAR study on the right-hand side pyridine substitutions
It was challenging to get the complex structure of PHD2 enzyme
with compound 4a. Given the structural similarity, we proposed
that the pyridine nitrogen of 4a formed a metal chelation interac-
tion with PHD2 which is critical for potency. This hypothesis was
supported by the loss of potency when the pyridine nitrogen was
replaced with carbon (4c, Table 1). In addition, moving the nitro-
gen atom away from the chelation position of the pyridine also
provided inactive analogues (4d and 4e). The 3-methyl substitu-
tion on the pyridine proved to be important for potency as well
and the compound without this methyl substitution (4f) demon-
strated ꢀ10-fold decrease in potency. Replacements of the 3-
methyl group with electron withdrawing (chloro, trifluoromethyl)
and electron donating (methoxyl) groups (4g, 4h, and 4i) resulted
in 10- to 15-fold decreases in potency, which might be due to re-
duced ability of the pyridine to form chelation interactions with
CO2H
4k
4l
9.0
7.1
CO2H
N
CO2H
PHD2 assay data is the average of at least two determinations.
a
Fe(II) or decreased
p–p stacking (edge-to-face) interaction with
His313. To further improve potency, compounds 4j, 4k, and 4l with
carboxylic acid side chains were designed to form salt bridge
interactions with Arg383 of the PHD2 enzyme which have been
identified as one of the key interactions responsible for high poten-
cies of 2OG mimics.12,15,18 As expected, the carboxylic acid ana-
logues (4j and 4k) demonstrated more than 10-fold increase in
potency compared with 4a, indicating the formation of the salt
bridge. This interaction was further supported by docking com-
pound 4k with the enzyme (Fig. 5). Compound 4l, however,
showed much decreased potency (ꢀ100-fold) compared with 4i
and 4k likely due to the longer carboxylic side chain making the
salt bridge interaction less optimal. Although the carboxylic acid
analogue (4k) demonstrated sub-nanomolar potency, it is not a
developable compound due to its poor PK profile. Very low periph-
eral exposure (DNAUC0ꢀt: 22.6 (ng h/mL)/(mg/kg)) and high hepa-
tic extraction (EH = 0.68) were observed for compound 4k in the
mouse short oral absorption (SOA) study, a mouse in vivo oral PK
Figure 5. Docking study of the carboxylic acid analogue 4k indicating salt bridge
interactions with the PHD2 enzyme. Docking was performed using Glide [Schrö-
dinger 2012R1].

G. Deng et al. / Bioorg. Med. Chem. 21 (2013) 6349–6358
6353
Table 2
study. We then focused our efforts on the left-hand side biphenyl
moiety to improve the series’ oral exposure while maintaining
potency.
SAR on the left-hand-side biphenyl moiety (9)
N
N
N
R
O
2.3.2. SAR study on the left-hand side bi-phenyl moiety
Our SAR efforts were focused on reducing series’ lipophilicity
(cLogP) for better ADMET profile.19 As shown in Table 2, a variety
of analogues with reduced cLogP were prepared and evaluated.
When the terminal phenyl group was replaced with a more polar
and smaller ethoxyl group (9a), the potency was decreased by
ꢀ100-fold. Increasing the size of the ethyl group to a sterically
bulkier isopropyl group (9b) provided only negligible increase in
potency compared to 9a. The oxygen atom of 9a was then removed
to explore the hydrophobic effect at this position, and the resultant
compound 9c showed marginal activity. Replacement of the termi-
nal phenyl group of 4a with a saturated cyclohexyl ring (9d) re-
sulted in ꢀ80-fold decrease in potency. The more polar
morpholine analogue (9e) was almost inactive. These SAR data
suggested that the aromatic ring was preferred at the terminal po-
sition. We then designed different six-membered heteroaromatic
rings such as pyridine (9f) and pyrimidine (9g) but both com-
pounds proved to be barely active. In addition, the five-membered
pyrazole analogue (9h) did not provide any better potency. The ter-
minal phenyl group proved to be very critical for potency, though
9
a
Compound
R
PHD2 pIC50
cLogPb
4a
7.6
4.75
9a
9b
9c
5.7
5.9
5.4
3.31
3.62
3.89
O
O
9d
9e
9f
6.0
5.5
6.4
5.48
2.32
3.25
N
O
no specific interaction(s) (e.g.,
p–p interaction or hydrophobic
interactions) could be identified from the complex structure. On
the other hand, the internal phenyl ring was observed to form a
p
–cation interaction with Arg322 and
a
p
–
p
stacking with
N
N
Trp389. Thus, introduction of a nitrogen atom into this phenyl ring
(9i and 9j) resulted in significant decreases in potencies, likely due
to reduced ability to form p–cation and p–p stacking interactions.
9g
9h
5.5
6.3
2.29
2.86
N
N
The SAR studies were then focused on substitutions on the bi-phe-
nyl rings. Introduction of a methoxy group to the ortho-position of
the internal phenyl ring (9k) resulted in an improved potency
(pIC50: 8.1) comparing to the 4a (pIC50: 7.6). Moving the methoxyl
substitution to the meta-position (9l) caused a slight decrease in
potency (pIC50: 7.5). Different ortho-susbstitutions with a variety
of electronic properties including methyl (9m), fluoro (9n), and cy-
ano (9o) groups were then explored and the methyl analogue (9m)
proved to be the most potent (pIC50: 8.4). Substitutions on the ter-
minal phenyl ring were also evaluated. Introduction of a carboxylic
acid group at para-position (9p) provided ꢀ5-fold increase of
potency (pIC50: 8.3) compared with the parent compound (4a),
probably due to ion pair interactions of the carboxylic acid group
of 9p with Arg396 of PHD2 enzyme. The acetic acid analogue
(9q) demonstrated slightly decreased potency (pIC50: 7.8). Moving
the carboxylic acid to the meta-position (9r) resulted in 10-fold de-
crease in potency (pIC50: 7.3) compared with 9p, suggesting a less
optimal polar interaction. The ortho-substituted analogue (9s) was
totally inactive.
N
N
9i
9j
5.7
6.2
3.25
3.46
N
OMe
9k
8.1
4.67
9l
7.5
8.4
4.11
4.85
OMe
2.4. Pharmacokinetics
9m
Between the most potent compounds discovered (9m and 9p),
compound 9p was considered more developable due to its better
physicochemical properties such as lower lipophilicity (cLogP:
2.2) compared with that of 9m (cLogP: 4.8). Compound 9p was
thus evaluated in the mouse short oral absorption (SOA) study.
To our delight, 9p demonstrated good oral bioavailability with high
peripheral exposure (DNAUC: 634 (ng h/mL)/(mg/kg)). In addition,
low hepatic extraction (EH = 0.10) for 9p was observed, indicating
low liver clearance of the compound.
F
9n
9o
7.6
7.4
4.72
4.18
CN
(continued on next page)

6354
G. Deng et al. / Bioorg. Med. Chem. 21 (2013) 6349–6358
introduced via infusion at a flow rate of 5 lL/min to acquire accu-
Table 2 (continued)
rate mass. LCMS (Agilent 1200SL-6110) analysis was conducted for
all assayed compounds on either acidic or basic conditions: (1)
acidic condition refers to water containing 0.05% TFA/acetonitrile
a
Compound
R
PHD2 pIC50
8.3
cLogPb
as mobile phase on Agilent SB-C18 column (1.8 lm, 4.6
9p
9q
2.22
ꢁ 30 mm) with MS and photodiode array detector (PDA). The fol-
lowing conditions were used: a gradient from 5% to 95% in 4 min
(or 6 min) and held at 95% for 1 min; UV detection at 214 and
254 nm; a flow rate of 1.5 mL/min; full scan; mass range from
100 to 1000 amu; (2) basic condition refers to water containing
10 mM NH4HCO3 aqueous/acetonitrile as mobile phase on Waters
HO2C
7.8
7.3
<5
1.75
2.22
1.42
HO2C
HO2C
XBridge C18 column (3.5
lm, 4.6 ꢁ 50 mm) with MS and photodi-
ode array detector (PDA). The following conditions were used: a
gradient from 5% to 95% in 5 min and held at 95% for 1 min; UV
detection at 214 and 254 nm; a flow rate of 2 mL/min; full scan;
mass range from 100 to 1000 amu. All the assayed compounds pos-
sess P95% purity determined using LC/MS analysis. Column chro-
matography was performed on Isco or Biotage using a pre-packed
silica gel column, a detector with UV wavelength at 254 nm and
mixed solvents. MDAP equipped with 2489 UV detector, 2767 sam-
ple manager, 2545 pump and 3100 single quadrupole mass spec-
9r
9s
CO2H
a
PHD2 assay data is the average of at least two determinations.
b
cLogP was calculated using the Daylight software.
trometer was performed on Sunfire Prep C18 column (5
lm, 19
ꢁ 50 mm) using water containing 0.05% TFA/acetonitrile as mobile
phase. The following condition was used: a gradient from 5% to
95% in 8 min and held in 95% for 2 min; a flow rate 30 mL/min.
3. Conclusion
4.2. General procedure for the preparation of compounds 4a–l
In conclusion, we have discovered 2,8-diazaspiro[4.5]decan-1-
one series as potent, orally bioavailable PHD2 inhibitors. To under-
stand the binding mode of the series, the complex crystal structure
of an analogue 4b with the PHD2 enzyme was resolved. In our
structure, the metal chelation interaction was maintained whereas
the salt bridge with Arg383 of the enzyme was not observed. This
is in contrast to all previous literature complex structure reports.
Instead, compound 4b was found to form several novel interac-
tions with the enzyme. These interactions include a hydrogen bond
4.2.1. tert-Butyl 2-([1,10-biphenyl]-4-yl)-1-oxo-2,8-
diazaspiro[4.5]decane-8-carboxylate (7)
A mixture of tert-butyl 1-oxo-2,8-diazaspiro[4.5]decane-8-car-
boxylate 5 (200 mg, 0.79 mmol) and 4-iodo-1,10-biphenyl 6 (337
mg, 1.2 mmol) in acetonitrile (10 mL) at 40 °C was degassed with
stream of nitrogen for 20 min. Then K2CO3 (326 mg, 2.36 mmol),
copper(I) iodide (37.4 mg, 0.20 mmol) and N1,N2-dimethyle-
thane-1,2-diamine (20 mg, 0.23 mmol) were added sequentially.
The reaction mixture was heated at 80 °C for 15 h. After cooling,
the solution was diluted with ethyl acetate and washed with
0.1 M HCl aqueous solution. The organic layer was concentrated
and purified by column chromatography to afford compound 7
(115 mg, 36%) as a white solid. MS (ES): C21H23N2O3 (MꢂtBu+H)+
calcd 351.2; found 351.1. LC/MS: tR = 3.95 min, >95%, m/z: 351.1
(MꢂtBu+H)+.
with Arg322, a
p–cation interaction with Arg322, a
p–p stacking
with Trp389, and a
p–p
stacking with His313. Guided by the struc-
tural information, extensive SAR studies were performed and com-
pound 9p was discovered with high enzymatic potency and a good
oral in vivo PK profile.
4. Experimental
4.1. Chemistry Methods
4.2.2. 2-([1,10-Biphenyl]-4-yl)-2,8-diazaspiro[4.5]decan-1-one (8)
To
tert-butyl
2-([1,10-biphenyl]-4-yl)-1-oxo-2,8-diazaspi-
Compounds not described below were purchased from com-
mercial vendors or reported. Compounds’ purity was determined
using LC/MS analysis. Purification of compounds was carried out
by conventional methods such as chromatography and/or
recrystallization using suitable solvents. Chromatographic meth-
ods include column chromatography and MDAP (mass-directed
auto-purification system).
ro[4.5]decane-8-carboxylate 7 (115 mg, 0.28 mmol) was added
HCl (4.0 M in MeOH) (10 mL, 27.4 mmol). The reaction mixture
was stirred at room temperature for overnight then was concen-
trated under reduced pressure to afford compound 8 (157 mg) as
a HCl salt. 1H NMR (400 MHz, DMSO-d6) d ppm 1.83–1.95 (m,
2H), 2.15 (ddd, J = 4.0, 10.0, 14.4 Hz, 2H), 2.24 (t, J = 7.0 Hz, 2H),
3.21 (ddd, J = 3.5, 9.7, 12.9 Hz, 2H), 3.55 (ddd, J = 4.3, 6.1,
12.9 Hz, 2H), 3.96 (t, J = 7.0 Hz, 2H), 7.30–7.37 (m, 1H), 7.43 (t,
J = 7.7 Hz, 2H), 7.62 (d, J = 7.3 Hz, 2H), 7.64–7.70 (m, J = 8.8 Hz,
2H), 7.70–7.76 (m, J = 8.8 Hz, 2H). MS (ES): C20H23N2O (M+H)+
calcd 307.2; found 307.2. LC/MS: tR = 2.44 min, >95%, m/z: 307.2
(M+H)+.
1H NMR spectral data were recorded on a Bruker 400 NMR spec-
trometer operating at 400 MHz. 13C NMR spectra were recorded at
100 MHz. CDCl3 is deuteriochloroform, DMSO-d6 is hexadeuterio-
dimethylsulfoxide. Chemical shifts are given in parts per million
(d) downfield from the internal standard tetramethylsilane (TMS)
or the NMR solvent. Abbreviations for NMR data are as follows: s
= singlet, d = doublet, t = triplet, q = quartet, m = multiplet, dd =
doublet of doublets, dt = doublet of triplets, app = apparent, br =
broad. J indicates the NMR coupling constant measured in Hertz.
High resolution mass (HRMS) was operated in positive mode of
electrospray ionization (ESI) at an orthogonal acceleration time-
of-flight (oa-TOF), SYNAPT G2 HDMSTM (Waters, Manchester,
UK). Solutions (500 ng/mL in acetonitrile: H2O (1:1, v/v)) were
4.2.3. 2-([1,10-Biphenyl]-4-yl)-8-((3-methylpyridin-2-
yl)methyl)-2,8-diazaspiro[4.5]decan-1-one (4a)
To a solution of 2-([1,10-biphenyl]-4-yl)-2,8-diazaspiro[4.5]dec-
an-1-one 8 (HCl salt, 100 mg, 0.29 mmol) and 3-methylpicolinalde-
hyde (70.7 mg, 0.58 mmol) in DCM (10 mL) was added sodium
triacetoxyborohydride (371 mg, 1.75 mmol). The reaction mixture
was stirred at room temperature for overnight. The reaction was

G. Deng et al. / Bioorg. Med. Chem. 21 (2013) 6349–6358
6355
quenched with water (0.2 mL) and the reaction mixture was con-
centrated. The residue was purified via MDAP to afford compound
4a (41 mg, 27%) as a TFA salt. 1H NMR (400 MHz, MeOH-d4) d ppm
1.87 (d, J = 14.56 Hz, 2H), 2.13–2.25 (m, 4H), 2.33 (s, 3H), 3.32 (t, J =
6.78 Hz, 2H), 3.58 (d, J = 12.30 Hz, 2H), 3.90 (t, J = 6.78 Hz, 2H), 4.58
(s, 2H), 7.31–7.42 (m, 2H), 7.46 (t, J = 7.65 Hz, 2H), 7.64–7.78 (m, 5
H), 7.78–7.83 (m, 2H), 8.53 (d, J = 3.76 Hz, 1H). 13C NMR (100 MHz,
DMSO-d6) d ppm 176.3, 149.4, 146.8, 139.9, 139.3, 136.4, 132.6,
129.4, 127.8, 127.3, 126.8, 124.1, 120.4, 57.2, 49.8, 45.0, 29.5,
17.7. HRMS C27H30N3O (M+H)+ calcd 412.2389, found: 412.2395.
LC/MS: tR = 2.96 min, 96.6%, m/z: 412.1 (M+H)+.
(t, J = 6.65 Hz, 2H), 3.27 (br s, 2H), 3.43 (br s, 2H), 3.90 (t, J = 6.78
Hz, 2H), 4.45 (br s, 2H), 7.32–7.38 (m, 1H), 7.41–7.49 (m, 3H),
7.64–7.74 (m, 4H), 7.76–7.82 (m, 2H), 8.55 (d, J = 5.02 Hz, 1H),
8.68 (s, 1H). HRMS C27H30N3O (M+H)+ calcd 412.2389, found:
412.2396. LC/MS: tR = 2.35 min, 100%, m/z: 412.3 (M+H)+.
4.2.7. 2-([1,10-Biphenyl]-4-yl)-8-((2-methylpyridin-3-
yl)methyl)-2,8-diazaspiro[4.5]decan-1-one (4e)
To a solution of 2-methylnicotinaldehyde (6.92 mg, 0.057
mmol) and 2-([1,10-biphenyl]-4-yl)-2,8-diazaspiro[4.5]decan-1-
one 8 (TFA salt, 20 mg, 0.048 mmol) in dichloromethane (DCM)
(3 mL) and N,N-dimethylformamide (1 mL) was added sodium tri-
acetoxyborohydride (10.08 mg, 0.048 mmol) at room temperature.
The reaction mixture was stirred at room temperature for over-
night. Water (10 mL) was added to the reaction mixture and then
the mixture was extracted with DCM (2 ꢁ 10 mL). The combined
fractions were dried over anhydrous sodium sulfate. The dried
solution was filtered and the filtrate was concentrated. The residue
was purified by MDAP to give 4e (2.8 mg, 11.1% yield) as a TFA salt.
1H NMR (400 MHz, DMSO-d6) d ppm 1.82 (d, 2H), 2.04 (t, J = 12.30
Hz, 2H), 2.22 (br s, 2H), 2.68 (s, 3H), 3.19–3.31 (m, 2H), 3.43 (br s,
2H), 3.90 (t, J = 6.53 Hz, 2H), 4.45 (br s, 2H), 7.31–7.38 (m, 1H), 7.46
(t, J = 7.65 Hz, 2H), 7.48–7.55 (m, 1H), 7.64–7.73 (m, 4H), 7.76–7.82
(m, 2H), 8.07 (d, J = 7.03 Hz, 1H), 8.59–8.65 (m, 1H). HRMS
4.2.4. 2-([1,10-Biphenyl]-4-yl)-8-((1-methyl-1H-imidazol-2-
yl)methyl)-2,8-diazaspiro[4.5]decan-1-one (4b)
To a solution of 2-([1,10-biphenyl]-4-yl)-2,8-diazaspiro[4.5]dec-
an-1-one 8 (HCl salt, 57 mg, 0.17 mmol) and 1-methyl-1H-imidaz-
ole-2-carbaldehyde (36.7 mg, 0.33 mmol) in DCM (10 mL) was
added sodium triacetoxyborohydride (211 mg, 1.0 mmol). The
solution was stirred at room temperature for overnight then the
reaction was quenched with MeOH (0.2 mL). The reaction mixture
was concentrated. The residue was purified via MDAP to afford
compound 4b (36 mg, 41%) as a TFA salt. 1H NMR (400 MHz,
DMSO-d6) d ppm 1.65–1.78 (m, 2H), 1.89–2.03 (m, 2H), 2.12 (t,
J = 6.40 Hz, 2H), 2.87–2.90 (m, 1H), 3.30–3.33 (m, 2H), 3.81 (s,
3H), 3.87 (t, J = 6.90 Hz, 2H), 4.22–4.27 (m, 3H), 7.32–7.38 (m,
1H), 7.40–7.42 (m, 1H), 7.46 (t, J = 7.65 Hz, 2H), 7.53–7.57 (m,
1H), 7.64–7.74 (m, 4H), 7.76–7.81 (m, 2H). 13C NMR (100 MHz,
DMSO-d6) d ppm 176.9, 139.9, 139.3, 136.2, 129.4, 127.7, 127.3,
126.8, 124.4, 120.3, 50.5, 49.5, 45.0, 43.5, 34.3, 30.9. HRMS
C
27H30N3O (M+H)+ calcd 412.2389, found: 412.2391. LC/MS: tR =
2.27 min, 99.4%, m/z: 412.2 (M+H)+.
4.2.8. 2-([1,10-Biphenyl]-4-yl)-8-((3-methylpyridin-2-
yl)methyl)-2,8-diazaspiro[4.5]decan-1-one (4f)
C
25H29N4O (M+H)+ calcd 401.2341, found: 401.2335. LC/MS: tR =
To a solution of 2-([1,10-biphenyl]-4-yl)-2,8-diazaspiro[4.5]dec-
an-1-one 8 (TFA salt, 27 mg, 0.064 mmol) and picolinaldehyde
(6.88 mg, 0.064 mmol) in DCM (20 mL) was added sodium triacet-
oxyborohydride (40.8 mg, 0.19 mmol). The reaction mixture was
stirred at room temperature for overnight. The reaction was
quenched with MeOH (0.2 mL) and the mixture was concentrated.
The residue was purified by MDAP to afford compound 4f (12 mg,
37%) as a TFA salt as an off-white solid. 1H NMR (600 MHz, DMSO-
d6) d ppm 1.84 (d, J = 13.55 Hz, 2H), 2.06–2.22 (m, 4H), 3.22–3.26
(m, 2H), 3.49 (d, J = 12.05 Hz, 2H), 3.88 (t, J = 6.59 Hz, 2H), 4.53
(s, 2H), 7.30–7.39 (m, 1H), 7.46 (t, J = 7.34 Hz, 2H), 7.50 (t, J =
6.02 Hz, 1H), 7.56 (d, J = 7.53 Hz, 1H), 7.67 (d, J = 7.53 Hz, 2H),
7.71 (m, J = 7.91 Hz, 2H), 7.79 (m, J = 7.91 Hz, 2H), 7.95 (t, J =
7.72 Hz, 1H), 8.67–8.74 (m, 1H), 10.00 (br s, 1H). 13C NMR
(100 MHz, DMSO-d6) d ppm 176.2, 150.9, 150.1, 139.9, 139.2,
138.2, 136.3, 129.4, 127.8, 127.3, 126.9, 125.4, 124.7, 120.4, 59.9,
49.3, 45.0, 40.6, 29.4. HRMS C26H28N3O (M+H)+ calcd 398.2232,
found: 398.2246. LC/MS: tR = 2.66 min, 100%, m/z: 398.1 (M+H)+.
2.53 min, 100%, m/z: 401.2 (M+H)+.
4.2.5. 2-([1,10-Biphenyl]-4-yl)-8-benzyl-2,8-
diazaspiro[4.5]decan-1-one (4c)
To a solution of 2-([1,10-biphenyl]-4-yl)-2,8-diazaspiro[4.5]dec-
an-1-one 8 (TFA salt, 27 mg, 0.064 mmol) and benzaldehyde (6.82
mg, 0.064 mmol) in dichloromethane (20 mL) was added sodium
triacetoxyborohydride (40.8 mg, 0.193 mmol) at room tempera-
ture. The reaction mixture was stirred at room temperature for
overnight. The reaction was quenched with MeOH (200
lL). The
reaction mixture was concentrated and the residue was by MDAP
to afford 4c (16 mg, 46.4%) as a TFA salt as an off-white solid. 1H
NMR (400 MHz, DMSO-d6) d ppm 1.73–1.91 (m, 2H), 1.91–2.10
(m, 2H), 2.18 (t, J = 6.8 Hz, 2H), 3.05–3.21 (m, 2H), 3.45–3.49 (m,
2H), 3.82–3.95 (m, 2H), 4.29–4.46 (m, 2H), 7.30–7.40 (m, 1H),
7.40–7.53 (m, 7H), 7.62–7.85 (m, 6H), 9.50 (s, 1H). 13C NMR
(100 MHz, DMSO-d6) d ppm 176.2, 139.9, 139.2, 136.3, 131.9,
131.7, 130.2, 130.1, 129.4, 127.8, 127.3, 126.8, 120.5, 120.3, 59.6,
48.6, 45.0, 43.7, 29.4, 27.4. HRMS
C
27H29N2O (M+H)+ calcd
4.2.9. 2-([1,10-Biphenyl]-4-yl)-8-((3-chloropyridin-2-yl)methyl)-
2,8-diazaspiro[4.5]decan-1-one (4g)
397.2280, found: 397.2292. LC/MS: tR = 2.95 min, 100%, m/z:
397.2 (M+H)+.
A solution of 2-([1,10-biphenyl]-4-yl)-2,8-diazaspiro[4.5]decan-
1-one 8 (30 mg, 0.098 mmol) and 3-chloropicolinaldehyde (20.8
mg, 0.15 mmol) in DCM (3 mL) and DMF (1 mL) was stirred at room
temperature for 30 min. Sodium triacetoxyborohydride (31 mg,
0.15 mmol) was then added. The reaction mixture was stirred at
room temperature for overnight. The reaction was quenched with
water (0.2 mL) and the reaction mixture was extracted with DCM
(20 mL). The organic fraction was separated and dried over anhy-
drous sodium sulfate. The dried solution was filtered and the fil-
trate was concentrated. The residue was purified by MDAP to
afford compound 4g (7.5 mg, 14%) as a TFA salt. 1H NMR (400
MHz, DMSO-d6) d ppm 1.84 (d, J = 13.55 Hz, 2H), 2.14–2.27 (m,
4H), 3.30–3.38 (m, 2H), 3.61 (d, J = 11.29 Hz, 2H), 3.90 (t, J = 6.65
Hz, 2H), 4.73 (s, 2H), 7.32–7.39 (m, 1H), 7.46 (t, J = 7.65 Hz, 2H),
7.57 (dd, J = 8.28, 4.77 Hz, 1H), 7.64–7.75 (m, 4H), 7.77–7.84 (m,
2H), 8.12 (dd, J = 8.03, 1.25 Hz, 1H), 8.69 (d, J = 3.76 Hz, 1H),
4.2.6. 2-([1,10-Biphenyl]-4-yl)-8-((4-methylpyridin-3-
yl)methyl)-2,8-diazaspiro[4.5]decan-1-one (4d)
To a solution of 2-([1,10-biphenyl]-4-yl)-2,8-diazaspiro[4.5]dec-
an-1-one 8 (TFA salt, 25 mg, 0.059 mmol) and 4-methylnicotinal-
dehyde (8.64 mg, 0.071 mmol) in dichloromethane (3 mL) and
N,N-dimethylformamide (1 mL) was added sodium triacetoxyboro-
hydride (12.6 mg, 0.059 mmol) at room temperature. The reaction
mixture was stirred at room temperature for overnight. Water (10
mL) was added to the reaction mixture and then the mixture was
extracted with DCM (2 ꢁ 10 mL). The combined fractions were
dried over anhydrous sodium sulfate. The dried solution was fil-
tered and the filtrate was concentrated. The residue was purified
by MDAP to give 4d (4.9 mg, 15.8% yield) as a TFA salt. 1H NMR
(400 MHz, DMSO-d6) d ppm 1.83 (d, 2H), 1.96–2.10 (m, 2H), 2.22

6356
G. Deng et al. / Bioorg. Med. Chem. 21 (2013) 6349–6358
10.00 (br s, 1H). HRMS C26H27ClN3O (M+H)+ calcd 432.1843, found:
purified by column chromatography to afford 2-([1,10-biphenyl]-
4-yl)-8-((5-bromo-3-methylpyridin-2-yl)methyl)-2,8-diazaspiro
[4.5]decan-1-one (437 mg, 74.9%). MS (ES): C27H29BrN3O (M+H)+
calcd 490.1, found: 490.2.
432.1845. LC/MS: tR = 2.71 min, 99.9%, m/z: 432.0 (M+H)+.
4.2.10. 2-([1,10-Biphenyl]-4-yl)-8-((3-(trifluoromethyl)pyridin-
2-yl)methyl)-2,8-diazaspiro[4.5]decan-1-one (4h)
To a dried three-necked flask were added 2-([1,10-biphenyl]-4-yl)-
8-((5-bromo-3-methylpyridin-2-yl)methyl)-2,8-diazaspiro[4.5]
decan-1-one (100 mg, 0.20 mmol) and THF (5 mL) under nitrogen.
The mixture was heated to reflux and it became a clear solution. To
the mixture were added tri-tert-butylphosphine (TFA salt, 23.7 mg,
0.082 mmol), cesium carbonate (26.6 mg, 0.082 mmol), Pd2(dba)3
(28.0 mg, 0.031 mmol) and (2-ethoxy-2-oxoethyl)zinc(II) bromide
(474 mg, 2.04 mmol). Then the reaction mixture was stirred at re-
flux for 30 min. After cooling, saturated aqueous NH4Cl solution
(15 mL) was added. The aqueous fraction was separated and ex-
tracted with DCM (20 mL). The combined organic fractions were
dried over anhydrous sodium sulfate. The dried solution was fil-
tered and the filtrate was concentrated. The residue was purified
by column chromatography to afford ethyl 2-(6-((2-([1,10-biphe-
nyl]-4-yl)-1-oxo-2,8-diazaspiro[4.5]decan-8-yl)methyl)-5-methyl-
pyridin-3-yl)acetate (101 mg, 100%). MS (ES): C31H36N3O3 (M+H)+
calcd 498.3, found: 498.3.
A solution of 2-([1,10-biphenyl]-4-yl)-2,8-diazaspiro[4.5]decan-
1-one 8 (TFA salt, 30 mg, 0.071 mmol) and 3-(trifluoromethyl)pic-
olinaldehyde (18.74 mg, 0.107 mmol) in DCM (3 mL) and DMF (1
mL) was stirred at room temperature for 30 min. Sodium triacet-
oxyborohydride (22.7 mg, 0.11 mmol) was then added. The reac-
tion mixture was stirred at room temperature for overnight. The
reaction was quenched with water (0.2 mL) and the reaction mix-
ture was extracted with DCM (20 mL). The organic fraction was
separated and dried over anhydrous sodium sulfate. The dried
solution was filtered and the filtrate was concentrated. The residue
was purified by MDAP to afford compound 4h (13.5 mg, 33%) as a
TFA salt. 1H NMR (400 MHz, DMSO-d6) d ppm 1.82 (d, J = 13.80 Hz,
2H), 2.17–2.29 (m, 4H), 3.27–3.45 (m, 2H), 3.51–3.59 (m, 2H), 3.90
(t, J = 6.53 Hz, 2H), 4.80 (s, 2H), 7.31–7.39 (m, 1H), 7.46 (t, J = 7.53
Hz, 2H), 7.63–7.77 (m, 5 H), 7.77–7.85 (m, 2H), 8.36 (d, J = 8.28 Hz,
1H), 9.00 (d, J = 4.27 Hz, 1H), 10.07 (br s, 1H). 13C NMR (100 MHz,
DMSO-d6) d ppm 153.0, 139.8, 139.2, 136.3, 129.5, 127.8, 127.3,
126.8, 124.4, 120.3, 49.8, 45.0, 43.6, 29.4, 27.6. HRMS C27H27F3N3O
(M+H)+ calcd 466.2106, found: 466.2112. LC/MS: tR = 2.94 min,
99.9%, m/z: 466.0 (M+H)+.
To a solution of ethyl 2-(6-((2-([1,10-biphenyl]-4-yl)-1-oxo-2,8-
diazaspiro[4.5]decan-8-yl)methyl)-5-methylpyridin-3-yl)acetate
(101 mg, 0.20 mmol) in THF (10 mL), ethanol (5 mL) and water (2
mL) was added KOH (11.4 mg, 0.20 mmol). The reaction mixture
was stirred at room temperature for 3 h. Then the mixture was
acidified to pH 6 using 2 N HCl aqueous solution. The organic sol-
vents were evaporated under reduced pressure. The resultant
aqueous solution was extracted with DCM (2 ꢁ 20 mL). The organic
fractions were combined and dried over anhydrous sodium sulfate.
The dried solution was filtered and the filtrate was concentrated.
The residue was purified by MDAP to afford 4j (9 mg, 9.6%). 1H
NMR (400 MHz, DMSO-d6) d ppm 1.44–1.61 (m, 2H), 1.68–1.87
(m, 2H), 2.02–2.30 (m, 4H), 2.38 (s, 3H), 2.68–2.92 (m, 2H), 3.54–
3.74 (m, 4H), 3.83 (t, J = 6.78 Hz, 2H), 7.31–7.37 (m, 1H), 7.42–
7.51 (m, 3H), 7.63–7.72 (m, 4H), 7.75–7.81 (m, 2H), 8.20 (s, 1H),
12.49 (br s, 1H). 13C NMR (100 MHz, DMSO-d6) d ppm 172.8,
156.0, 146.9, 139.9, 139.5, 132.8, 129.4, 127.7, 127.3, 126.8,
120.2, 48.0, 45.0, 37.7, 32.1, 28.4, 18.3. HRMS C29H32N3O3 (M+H)+
calcd 470.2444, found: 470.2451. LC/MS: tR = 2.69 min, 100%, m/
z: 470.3 (M+H)+.
4.2.11. 2-([1,10-Biphenyl]-4-yl)-8-((3-methoxypyridin-2-
yl)methyl)-2,8-diazaspiro[4.5]decan-1-one (4i)
To a mixture of 2-([1,10-biphenyl]-4-yl)-2,8-diazaspiro[4.5]dec-
an-1-one 8 (TFA salt, 100 mg, 0.24 mmol) and 3-methoxypicolinal-
dehyde (100 mg, 0.73 mmol) in dichloromethane (5 mL) and
N,N-dimethylformamide (5 mL) was added three drops of acetic
acid. The reaction mixture was stirred at room temperature for
overnight. Sodium triacetoxyborohydride (151 mg, 0.71 mmol)
was added to the reaction mixture. The reaction mixture was stir-
red at room temperature for overnight. The reaction mixture was
extracted with DCM (2 ꢁ 20 mL). The organic fractions were com-
bined and the combined solution was dried over anhydrous so-
dium sulfate. The dried solution was filtered and the filtrate was
concentrated. The residue was purified by MADP to give compound
4i (15.5 mg, 15%). 1H NMR (400 MHz, DMSO-d6) d ppm 1.46 (d, J =
12.55 Hz, 2H), 1.73 (td, J = 12.61, 3.64 Hz, 2H), 2.01 (t, J = 6.90 Hz,
2H), 2.13–2.23 (m, 2H), 2.83 (d, J = 11.80 Hz, 2H), 3.60 (s, 2H), 3.78–
3.84 (m, 5 H), 7.26–7.37 (m, 2H), 7.39–7.48 (m, 3H), 7.67 (t, J = 8.16
Hz, 4H), 7.74–7.81 (m, 2H), 8.10 (d, J = 4.52 Hz, 1H). 13C NMR
(100 MHz, DMSO-d6) d ppm 178.0, 154.7, 147.4, 140.5, 140.0,
139.6, 135.9, 129.4, 127.7, 127.2, 126.8, 123.8, 120.1, 118.6, 58.6,
56.1, 49.9, 44.9, 44.6, 32.4, 28.4. HRMS C27H30N3O2 (M+H)+ calcd
428.2338, found: 428.2347. LC/MS: tR = 2.78 min, 100%, m/z:
428.2 (M+H)+.
4.2.13. 3-(6-((2-([1,10-Biphenyl]-4-yl)-1-oxo-2,8-
diazaspiro[4.5]decan-8-yl)methyl)-5-methylpyridin-3-
yl)propanoic acid (4k)
To a dried three-neck flask was added 2-([1,10-biphenyl]-4-yl)-
8-((5-bromo-3-methylpyridin-2-yl)methyl)-2,8-diazaspiro[4.5]
decan-1-one (150 mg, 0.31 mmol) and THF (5 mL) under nitro-
gen. The mixture was heated to reflux and it became a clear
solution. To the mixture were added tri-tert-butylphosphine (TFA
salt, 35.5 mg, 0.122 mmol), cesium carbonate (40.0 mg, 0.12
mmol), Pd2(dba)3 (42.0 mg, 0.046 mmol), and (3-ethoxy-3-oxopro-
pyl)zinc(II) bromide (6.12 mL, 3.1 mmol). The reaction mixture was
stirred at reflux for 30 min. After cooling, saturated aqueous NH4Cl
solution (10 mL) was added. The mixture was extracted with DCM
(2 ꢁ 20 mL). The organic fractions were combined and the com-
bined solution was dried over anhydrous sodium sulfate. The dried
solution was filtered and the filtrate was concentrated. The residue
was purified by column chromatography to afford ethyl 3-(6-((2-
([1,10-biphenyl]-4-yl)-1-oxo-2,8-diazaspiro[4.5]decan-8-yl)methyl-
5-methylpyridin-3-yl)propanoate (63.7 mg, 40.7%). MS (ES):
4.2.12. 2-(6-((2-([1,10-Biphenyl]-4-yl)-1-oxo-2,8-
diazaspiro[4.5]decan-8-yl)methyl)-5-methylpyridin-3-yl)acetic
acid (4j)
A mixture of 2-([1,10-biphenyl]-4-yl)-2,8-diazaspiro[4.5]decan-
1-one 8 (TFA salt, 500 mg, 1.19 mmol), 5-bromo-3-methylpicolin-
aldehyde (285 mg, 1.43 mmol), and sodium acetate (98 mg, 1.19
mmol) in DCM (20 mL) and DMF (4 mL) was added five drops of
HOAc. The reaction mixture was stirred at room temperature for
16 h. To the mixture was then added sodium triacetoxyhydrobo-
rate (250 mg, 1.19 mmol). After stirring for 1 h, more sodium tria-
cetoxyhydroborate (250 mg, 1.19 mmol) was added, and the
resultant mixture was stirred for another 2 h. Saturated aqueous
NaHCO3 solution (20 mL) was added. The organic fraction was sep-
arated and dried over anhydrous sodium sulfate. The dried solution
was filtered and the filtrate was concentrated. The residue was
C
32H38N3O3 (M+H)+ calcd 512.3, found: 512.3.
To a solution of ethyl 3-(6-((2-([1,10-biphenyl]-4-yl)-1-oxo-2,8-
diazaspiro[4.5]decan-8-yl)methyl)-5-methylpyridin-3-yl)propano-
ate (63.7 mg, 0.124 mmol) in THF (10 mL), ethanol (10 mL) and
water (2 mL) was added KOH (2 ml, 6.00 mmol, 3 M in water).

G. Deng et al. / Bioorg. Med. Chem. 21 (2013) 6349–6358
6357
The reaction mixture was stirred at room temperature for 2 h. Then
the mixture was acidified to pH 5 by addition of 2 N HCl aqueous
solution. The mixture was extracted with DCM (2 ꢁ 20 mL). The or-
ganic fractions were combined and dried over anhydrous sodium
sulfate. The dried solution was filtered and the filtrate was concen-
trated. The residue was purified by MDAP to afford 4k (11.3 mg,
18.9%). 1H NMR (400 MHz, DMSO-d6) d ppm 1.47 (d, J = 12.55 Hz,
2H), 1.72 (t, J = 10.92 Hz, 2H), 2.03–2.09 (m, 2H), 2.15 (t, J =
11.67 Hz, 2H), 2.37 (s, 3H), 2.55 (t, J = 7.53 Hz, 2H), 2.65–2.75 (m,
2H), 2.75–2.81 (m, 2H), 3.56 (s, 2H), 3.82 (t, J = 6.90 Hz, 2H),
7.31–7.37 (m, 1H), 7.41–7.48 (m, 3H), 7.63–7.71 (m, 4H), 7.78 (d,
J = 8.78 Hz, 2H), 8.16 (s, 1H), 12.18 (br. s, 1H). HRMS C30H34N3O3
(M+H)+ calcd 484.2600, found: 484.2602. LC/MS: tR = 2.77 min,
100%, m/z: 484.2 (M+H)+.
drop of reservoir solution supplemented with 1 mM inhibitor com-
pound and 10% glycerol before freezing in liquid nitrogen.
X-ray diffraction data were collected at beamline 17U at the
Shanghai Synchrotron Radiation Facility and were processed using
HKL2000.20 The complex crystallized in the space group with unit
cell dimensions of a = 111.455, b = 111.455, c = 39.974, and
c = 120,
and one complex per asymmetric unit. The structure was solved by
molecular replacement using the apo PHD2 structure (GSK pro-
prietary) as a search model. Model refinement was done against
full range of data using REFMAC521 and rebuilt using COOT.22
The model was refined to a final Rwork = 23.0 and Rfree = 25.9 and
contained 1 copy of PHD2, 1 copy of compound 4b, and water mol-
ecules. The bond length and angle root-mean-square deviation
from ideality are 0.008 Å and 1.388°, respectively.
The coordinates of the PHD2 complex with compound 4b have
been deposited in the Protein Data Bank with the access number
4JZR.
4.2.14. 4-(6-((2-([1,10-Biphenyl]-4-yl)-1-oxo-2,8-
diazaspiro[4.5]decan-8-yl)methyl)-5-methylpyridin-3-
yl)butanoic acid (4l)
4.4. Docking
To a dried three-necked flask were added 2-([1,10-biphenyl]-4-yl)-
8-((5-bromo-3-methylpyridin-2-yl)methyl)-2,8-diazaspiro[4.5]dec-
an-1-one (100 mg, 0.20 mmol) and THF (5 mL) under nitrogen. The
mixture was heated to reflux and it became a clear solution. To the
mixture were added tri-tert-butylphosphine (TFA salt, 23.7 mg,
0.082 mmol), cesium carbonate (26.6 mg, 0.082 mmol), Pd2(dba)3
(28.0 mg, 0.031 mmol), and (4-ethoxy-4-oxobutyl)zinc(II) bromide
(0.41 ml, 0.20 mmol). The reaction mixture was stirred at reflux for
30 min. After cooling, saturated aqueous NH4Cl solution (10 mL)
was added. The mixture was extracted with DCM (2 ꢁ 20 mL).
The organic fractions were combined and the combined solution
was dried over anhydrous sodium sulfate. The dried solution was
filtered and the filtrate was concentrated. The residue was purified
by column chromatography to afford ethyl 4-(6-((2-([1,10-biphe-
nyl]-4-yl)-1-oxo-2,8-diazaspiro[4.5]decan-8-yl)methyl)-5-methyl-
Ligand was prepared using LigPrep [Schrödinger 2012R1] to
generate tautomers at pH 7 (using Epik) and steoreoisomers.
Receptor was prepared using Protein Preparation Wizard [Schrö-
dinger 2012R1] to assign bond orders, add hydrogen atoms, refine
h-bond assignment, and perform energy minimization, etc. Then
docking grid was generated around compound 4b in the crystal
structure using Glide [Schrödinger 2012R1]. Eventually ligand
was docked using Glide with two constraints: h-bond with
Arg383 and h-bond with Arg322. All other settings took the default
values.
As compound 4k is large with the biphenyl which may hinder
conformational sampling during docking, we replaced the biphenyl
with a methyl before docking and added it back to the top docking
pose afterwards.
pyridin-3-yl)butanoate (71 mg, 67.6%). MS (ES):
C33H40N3O3
(M+H)+ calcd 526.3, found: 526.3.
4.5. Homogeneous time resolved fluorescence (HTRF) assay for
measurement of prolyl hydroxylase domain 2 (PHD2) activity
To a solution of ethyl 4-(6-((2-([1,10-biphenyl]-4-yl)-1-oxo-2,8-
diazaspiro[4.5]decan-8-yl)methyl)-5-methylpyridin-3-yl)butano-
ate (71 mg, 0.135 mmol) in tetrahydrofuran (THF) (10 mL), ethanol
(5 mL) and water (2 mL) was added 3 M potassium hydroxide
aqueous solution (2 mL, 6.00 mmol). The reaction mixture was stir-
red at room temperature for 3 h. Then the mixture was acidified by
addition of 2 N HCl aqueous solution to pH 6.0. The mixture was
extracted with DCM (2 ꢁ 20 mL). The organic fractions were com-
bined and combined solution was dried over anhydrous sodium
sulfate. The dried solution was filtered and the filtrate was concen-
trated. The residue purified by MDAP to give 4l (1.2 mg, 1.5% yield)
as a TFA salt MS C31H36N3O3 (M+H)+ calcd 498.28, found: 498.3. LC/
MS: tR = 2.77 min, 100%, m/z: 498.3 (M+H)+.
10 nL of compound was shot by Echo (Labcyte Inc.) into NUNC
384-well Black PP Plate. 5
buffer (50 mM HEPES, 50 mM KCl, pH 7.5) with 1 mM TCEP HCl,
0.2 mg/mL BSA, 5 g/mL FeCl2, 0.4 mM ascorbic acid, 0.3 mM
lL of enzyme working solution (assay
l
CHAPS and 16 nM His-streptag-tev-EGLN1) was added into com-
pound wells and high control wells in compound-containing NUNC
384-well assay plate and 5 lL of control working solution (enzyme
working solution without EGLN1 enzyme) was added into low con-
trol wells. The plate was then incubated at room temperature for
30 min. 5
nM biotin-VBC, 1
Cy5CODD, 1 M 2-oxoglutaratic acid and 0.3 mM CHAPS) was
l
L of substrate working solution (assay buffer with 50
l
g/mL Europium/Streptavidin, 3.8 M HIF1a-
l
l
4.3. Crystallography
added into all wells of assay plate. After another incubation at
room temperature for 30 min, the plate was loaded into ViewLux
imager (PerkinElmer) and read with LANCE protocol.
Purification of human PHD2 is as described previously.15 PHD2
inhibitor complex was made by mixing purified PHD2 and inhibi-
tor compound at 1:2 molar ratio. Co-crystals of PHD2 complexed
with a 100-fold weaker PHD inhibitor compound (unpublished
4.6. Short oral absorption (SOA) study in male mice
GSK proprietary data) grew spontaneously by mixing 2
l
L of
All animal studies were ethically reviewed and carried out in
accordance with European Directive 2010/63/EU and the GSK Pol-
icy on the Care, Welfare and Treatment of Animals.
Six male C57BL/6 mice received oral gavage administration
(p.o.) of the test compound at 10 mg/kg. The dose was prepared
on the day of administration in suspension in DMSO:1% methylcel-
lulose (v/v, 1:99), at a concentration of 1 mg/mL (dose volume = 10
mL/kg). At 15 min, 30 min, 1 h, 3 h, 5 h and 7 h post p.o. adminis-
tration, fifty microliters of blood were collected via portal vein and
cardiac puncture under anesthesia, respectively, and were diluted
PHD2 complex (20 mg/mL) and 2 L of a reservoir solution (30%
l
PEG 3350, 200 mM ammonium sulfate, 100 mM sodium acetate,
pH 4.8) in sitting drop vapor diffusion method at 20 °C. Structure
of this complex was solved and revealed a clean empty pocket
(unpublished GSK proprietary data). Co-crystals of PHD2 and this
proprietary compound were transferred to a fresh drop of reservoir
solution without inhibitor compound before 1 mM of compound
4b was introduced to the transfer drop and soaking was allowed
for 2 days at 20 °C. Soaked crystals were quickly transferred to a

6358
G. Deng et al. / Bioorg. Med. Chem. 21 (2013) 6349–6358
7. Ivan, M.; Haberberger, T.; Gervasi, D. C.; Michelson, K. S.; Gunzler, V.; Kondo, K.;
with 150
lL of water. There was one animal for each time point.
Yang, H.; Sorokina, I.; Conaway, R. C.; Conaway, J. W.; Kaelin, W. G. Proc. Natl.
Acad. Sci. U.S.A. 2002, 99, 13459.
8. Bruick, R. K.; McKnight, S. L. Science 2001, 294, 1337.
9. Boulahbel, H.; Duran, R. V.; Gottlieb, E. Biochem. Soc. Trans. 2009, 37, 291.
10. Rocha, S. Trends Biochem. Sci. 2007, 32, 389.
11. Vachal, P.; Miao, S.; Pierce, J. M.; Guiadeen, D.; Colandrea, V. J.; Wyvratt, M. J.;
Salowe, S. P.; Sonatore, L. M.; Milligan, J. A.; Hajdu, R.; Gollapudi, A.; Keohane,
C. A.; Lingham, R. B.; Mandala, S. M.; DeMartino, J. A.; Tong, X.; Wolff, M.;
Steinhuebel, D.; Kieczykowski, G. R.; Fleitz, F. J.; Chapman, K.; Athanasopoulos,
J.; Adam, G.; Akyuz, C. D.; Jena, D. K.; Lusen, J. W.; Meng, J.; Stein, B. D.; Xia, L.;
Sherer, E. C.; Hale, J. J. J. Med. Chem. 2012, 55, 2945.
12. Rosen, M. D.; Venkatesan, H.; Peltier, H. M.; Bembenek, S. D.; Kanelakis, K. C.;
Zhao, L. X.; Leonard, B. E.; Hocutt, F. M.; Wu, X.; Palomino, H. L.; Brondstetter, T.
I.; Haugh, P. V.; Cagnon, L.; Yan, W.; Liotta, L. A.; Young, A.; Mirzadegan, T.;
Shankley, N. P.; Barrett, T. D.; Rabinowitz, M. H. ACS Med. Chem. Lett. 2010, 1,
526.
Blood sample was stored at ꢂ80 °C until LC/MS/MS bioanalysis.
The measured concentrations were corrected for dilution factor,
and then the peripheral exposure was determined based on the
area under the curve (AUC) systemic blood concentrations. Hepatic
extraction (EH) was calculated based on the AUC of systemic and
hepatic portal vein blood concentrations:
AUCSYS
EH ¼ 1 ꢂ
AUCHPV
Acknowledgment
13. Katschinski, D. M. Acta Physiol. 2009, 195, 407.
We thank Drs. Kelly Dong and Ruina Gao for measuring HRMS
data.
14. (a) Berchner-Pfannschmidt, U.; Tug, S.; Trinidad, B.; Oehme, F.; Yamac, H.;
Wotzlaw, C.; Flamme, I.; Fandrey, J. J. Biol. Chem. 2008, 283, 31745; (b) Yan, L.;
Colandrea, V. J.; Hale, J. J. Expert Opin. Ther. Pat. 2010, 20, 1219.
15. McDonough, M. A.; Li, V.; Flashman, E.; Chowdhury, R.; Mohr, C.; Lienard, B. M.
R.; Zondlo, J.; Oldham, N. J.; Clifton, I. J.; Lewis, J.; McNeill, L. A.; Kurzeja, R. J. M.;
Hewitson, K. S.; Yang, E.; Jordan, S.; Syed, R. S.; Schofield, C. J. Proc. Natl. Acad.
Sci. U.S.A. 2006, 103, 9814.
Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.bmc.2013.08.
046.
16. Fletcher, J. M.; Hale, J. J.; Miao, S.; Vachal, P. WO 20,081,442,66A, 2008.
17. Meanwell, N. A. Chem. Res. Toxicol. 2011, 24, 1420.
18. (a) Chowdhury, R.; McDonough, M. A.; Mecinovic, J.; Loenarz, C.; Flashman, E.;
Hewitson, K. S.; Domene, C.; Schofield, C. J. Structure 2009, 17, 981; (b)
Demetriades, M.; Leung, I. K. H.; Chowdhury, R.; Chan, M. C.; McDonough, M.
A.; Yeoh, K. K.; Tian, Y.-M.; Claridge, T. D. W.; Ratcliffe, P. J.; Woon, E. C. Y.;
Schofield, C. J. Angew. Chem., Int. Ed. 2012, 51, 6672; (c) Evdokimov, A. G.;
Kawamoto, M.; Boyer, A. S.; Mekel, M. J.; Pokross, M. E.; Walter, R. L.
US2008,012,474,0A1, 2008.
19. (a) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Delivery
Rev. 2001, 46, 3; (b) Lipinski, C. A. Drug Discovery Today: Technol. 2004, 1, 337;
(c) Waring, M. J. Expert Opin. Drug Discovery 2010, 5, 235; (d) Hann, M. M.;
Keseru, G. M. Nat. Rev. Drug Disc. 2012, 11, 355.
References and notes
1. Fong, G.-H.; Takeda, K. Cell Death Differ. 2008, 15, 635.
2. Karuppagounder, S. S.; Ratan, R. R. J. Cereb. Blood Flow Metab. 2012, 1.
3. Poppe, L.; Tegley, C. M.; Li, V.; Lewis, J.; Zondlo, J.; Yang, E.; Kurzeja, R. J. M.;
Syed, R. J. Am. Chem. Soc. 2009, 131, 16654.
4. Schofield, C. J.; Ratcliffe, P. J. Nat. Rev. Mol. Cell Biol. 2004, 40, 343.
5. Berra, E.; Benizri, E.; Ginouves, A.; Volmat, V.; Roux, D.; Pouyssegur, J. EMBO J.
2003, 22, 4082.
20. Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307.
21. Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Acta Crystallogr. 1997, D53, 240.
22. Emsley, P.; Cowtan, K. Acta Crystallogr. 2004, D60, 2126.
6. Natarajan, R.; Salloum, F. N.; Fisher, B. J.; Kukreja, R. C.; Fowler, A. A. Circ. Res.
2006, 98, 133.