Optimization of a pyrazole hit from FBDD into a novel series of indazoles as ketohexokinase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.06.067

A series of indazoles have been discovered as KHK inhibitors from a pyrazole hit identified through fragment-based drug discovery (FBDD). The optimization process guided by both X-ray crystallography and solution activity resulted in lead-like compounds with good pharmaceutical properties.

Full text of DOI:10.1016/j.bmcl.2011.06.067

Bioorganic & Medicinal Chemistry Letters 21 (2011) 4762–4767
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Optimization of a pyrazole hit from FBDD into a novel series of indazoles
as ketohexokinase inhibitors
⇑
Xuqing Zhang , Fengbing Song, Gee-Hong Kuo, Amy Xiang, Alan C. Gibbs, Marta C. Abad,
Weimei Sun, Lawrence C. Kuo, Zhihua Sui
Johnson & Johnson Pharmaceutical Research and Development, Welsh & McKean Roads, PO Box 776, Spring House, PA 19477, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of indazoles have been discovered as KHK inhibitors from a pyrazole hit identified through frag-
ment-based drug discovery (FBDD). The optimization process guided by both X-ray crystallography and
solution activity resulted in lead-like compounds with good pharmaceutical properties.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 2 May 2011
Revised 14 June 2011
Accepted 15 June 2011
Available online 29 June 2011
Keywords:
FBDD (fragment based drug discovery)
KHK (ketohexokinase)
Pyrazole
Indazole
Ketohexokinase (KHK, also known as fructokinase) catalyzes,
with adenosine triphosphate (ATP) and a potassium ion (K+), the
conversion of the furanose form of D-fructose to fructose-1-phos-
vanced into lead optimization or late lead generation according
to their bio-activity, pharmacokinetic and selectivity profiles.
Fragment evolution has been by far the most successful method
of fragment optimization.6–9 When allied with a high degree of
structural information such as X-ray crystallography, it gives
medicinal chemists valuable tools in the validation and subsequent
optimization of a hit. Pyrazole 1 (Fig. 1) was identified as a hit
binding to the ATP binding site of KHK by X-ray crystallographic
screening of the fragment libraries. Without KHK enzyme activity
tested, fragment 1 was optimized in the follow-up chemistry.
Figure 1 illustrates the combination of fragment fusion and evolu-
tion of pyrazole 1 to yield indazole 2. Evidence supporting incorpo-
ration of an indazole core was an intra-molecular H-bonding
between the 4-carboxamide and 5-amine functional groups of
pyrazole 1, as revealed by X-ray crystallography, molecular
mechanics and virtual docking experiments. Furthermore, as
indicated by X-ray crystallography, these two functional groups
phate.1 It initiates the intracellular catabolism of a large proportion
of dietary carbohydrate and is an important regulator of hepatic
glucose metabolism.2 Due to its role in dietary fructose metabo-
lism, inhibition of KHK would suppress carbon supply for fatty acid
and very low density lipoprotein synthesis. Hence, modulation of
KHK will help relieve some metabolic syndromes such as obesity,
hypertriglyceridemia, insulin resistance and hypertension, which
makes KHK an important target in drug discovery.3
Recently, a novel series of pyrimidinopyrimidines have been
discovered by using a combination of high-throughput screening
(HTS) and structure-based drug design (SBDD).4 Along with HTS
approach, we also described a different approach for lead genera-
tion of KHK inhibitors through a FBDD protocol.5 The entire process
includes X-ray crystallographic screening of three stages of itera-
tive design and synthesis of fragment libraries. We have described
the discovery of unique and structurally diverse hits through this
pathway. Herein we focus on a medicinal chemistry approach to
drive one hit into lead-like structures. Unlike the early libraries
solely guided by electron density information from X-ray
crystallography, the follow-up compounds were evaluated in a
KHK-mediated enzymatic assay. Selected compounds were ad-
MeS
N
O
MeS
N
NH2
NH2
N
R2
N
1
2
FBDD hit
⇑
Corresponding author. Tel.: +1 215 488 7866.
E-mail address: xzhang5@its.jnj.com (X. Zhang).
Figure 1. From pyrazole to indazole.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.06.067

X. Zhang et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4762–4767
4763
did not engage in obvious H-bonding with the protein. Thus, 6,5-
fused systems such as an indazole core became one of the logical
designs provided that indazole could be tolerated in the binding
pocket. This strategy was also supported by the observation that
some phenyl fragments, while not hits from X-ray screening, over-
lapped well with the amine/carboxamide moieties of pyrazole 1.
the binding site. Substituted indazole analogs could be easily syn-
thesized and quickly derivatized. Thus, indazole 2 was considered
as a major chemotype in the follow-up chemistry. Docking of inda-
zole 2 with a X-ray crystal structure of pyrazole 1 revealed that 2
overlapped closely with 1 and was ꢀ6–7 Å from the Asp27 carbox-
ylate of the b subunit that forms the KHK dimer (Asp27B). A series
of indazole analogs were synthesized and evaluated as KHK
inhibitors. Considering the acidic nature of Asp27B, we designed
From medicinal chemistry perspective, indazole template
2
allowed further substitution (R2) on the phenyl ring to extend into
Me
Y
R2
N
e), i)
N
8a 8c 8d 8p
,
,
,
Me
Y
N
HN
f), i)
N
O
R2
Me
O
NH
Y
O
X
5
N
c) for
8h 8m
X
-
a)
b)
N
O
N
X
Me
Y
6
N
H
O
d) and c) for
g), i)
h), i)
N
N
3
N
R2
5
4
, Y = O
6, Y = S
7a 8b 8e 8n 8o
,
,
,
,
Me
Y
N
O
N
R2
8f 8g 8q 8r
,
,
,
Scheme 1. Reagents and conditions: (a) PhNHNH2, DIPEA, rt; (b) NaNO2, HCl; (c) Cs2CO3, MeI; (d) P2S5, xylene, 130 °C; (e) for 8c, 8d, 8p, X = Br: Pd(PPh3)4, Boc-R2-B(OH)2,
Na2CO3; H2,Pd/C (for 8a); (f) for 8h–m, X = NO2: H2, Pd/C; HATU, COOH-R2-Boc; (g) for 7a, 8b, 8e, 8n, 8o, X = Br: Pd(OAc)2, Cs2CO3, BINAP, Boc-R2NH, toluene, reflux; (h) for 8f,
8g, 8q, 8r, X = OMe: BBr3; Cs2CO3, TsO-R2-Boc; and (i) HCl.
SMe
SMe
O
O
a)
d), e)
N
b), c)
B
N
N
N
N
N
N
H
+
N
H
N
H
Br
HN
BocN
BocN
9
12
10
Boc
11
R3
Scheme 2. Reagents and conditions: for 12a–g (a) Pd(PPh3)4, Na2CO3; (b) KOH, I2; (c) NaSMe, t-Bu-ONa, Pd(OAc)2, Josiphos; (d) Boc-R3-Ph-B(OH)2, Cu(OAc)2; and (e) HCl; for
12g, Fe, HOAc (R3 from NO2 to NH2).
R1
N
R1
N
R1
F
c) or d) or e)
followed by f)
R2
O
F
X
H
N
a)
N
b)
N
R1
N
X
X
1
1
1
1
15,
16,
13,
14,
R = Me
R = Me
17,
18,
R = Me
19,
20,
R = Me
1
1
1
1
R = Et
R = Et
R = Et
R = Et
Scheme 3. Reagents and conditions: (a) PhNHNH2, TSA; (b) K2CO3, DMF, 100 °C; (c) for 19a–b, 20a–h, X = Br: Pd(OAc)2, Cs2CO3, BINAP, Boc-R2-NH, toluene, reflux; (d) for 20i,
X = Br: Pd(PPh3)4, Boc-R2-B(OH)2, Na2CO3; (e) for 20k, X = NO2: H2, Pd/C; HATU, Boc-R2-COOH; for 20j, HCl (gas), MeOH followed by NH4OAc; and (f) HCl.

4764
X. Zhang et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4762–4767
Table 1
the indazoles containing basic amino groups to target potentially
favorable ionic ligand–protein interactions. The size, length, rigid-
ity and geometry of the R2 group were modified for favorable inter-
actions with negatively charged Asp27B.
SAR on R2 group of indazole 8
MeS
N
R2
The general synthetic route to indazole analogs 7 and 8 is
shown in Scheme 1. Phenylhydrazine was reacted with substituted
N
1H-benzo[d][1,3]oxazine-2,4-diones
3
either commercially
available or prepared according to the literatures followed by
intramolecular cyclization via diazotization to provide 1-phenyl-
1H-indazol-3(2H)-ones 4 in 35–52% yield. Indazolones 4 were
methylated with cesium carbonate and methyl iodide to afford
3-methoxy-1-phenyl-1H-indazoles 5 in 65–71% yield. Indazolones
4 were converted into thio-indazolones by treatment with P2S5 in
toluene followed by thio-alkylations with methyl iodide to afford
8
ID
8a
8b
8c
8d
8e
8f
R2 (6-substitution)
KHK enzymea IC50
(lM)
NH
0.33
0.56
0.08
>10
N
NH
NH
N
the corresponding 3-(methylthio)-1-phenyl-1H-indazoles
6 in
28–40% yield. Compounds 5 and 6 were functionalized at 6- or
7-position (R2) of indazole core via Suzuki coupling, Buchwald ami-
nation, amide coupling or displacement strategies. Bromo-indazole
6 was coupled with aryl boronic acid under standard Suzuki condi-
tions, followed by de-protection of N-Boc group to afford 8c, 8d
and 8p. Compound 8a was obtained by hydrogenation of 8c. Bro-
mo-indazole 6 was also converted into 7a, 8b, 8e, 8n and 8o
through Buchwald amination followed by de-protection of the
N-Boc group. For synthesis of the amide side chain on the indazole
core, nitro-substituted indazole 6 was reduced under hydrogena-
tion to give an amino indazole intermediate, which was then cou-
pled with carboxylic acid by HATU followed by de-protection to
afford the corresponding amide 8h–m. De-methylation of meth-
oxy-substituted indazole 6 followed by displacement of various
tosylates, after de-protection, yielded the corresponding ether-
linked adducts 8f, 8g, 8q and 8r. To explore SAR on 1-phenyl
substitution (R3) of indazole scaffold 12, an efficient divergent syn-
thesis was also developed as shown in Scheme 2. Suzuki coupling
of commercially available 6-bromo-indazole 9 with 4-(4,4,5,5-
tetramethyl-[1,3,2]dioxaborolan-2-yl)-3,6-dihydro-2H-pyridine-
1-carboxylic acid tert-butyl ester gave indazole 10 in 75% yield.
3-Iodination of 10 with iodine and potassium hydroxide followed
by palladium catalyzed methylthio-ether formation provided
3-methylthio-indazole 11 as the key intermediate in 45% yield.
Thio-indazole 11 was then coupled with various phenyl-boronic
acids by Cu(OAc)2, after de-protection of the N-Boc group under
acidic condition, to afford 3-(methylthio)-1-phenyl-1H-indazole
12 in moderate to good yield. 3-Methyl or 3-ethyl substituted inda-
zoles 19 and 20 were synthesized according to the routes shown in
Scheme 3.10 Ketones 13 and 14, available through literature known
methods,11 were reacted with PhNHNH2 under acid catalysis to
give the corresponding hydrazones 15 and 16 in 75–100% yield.
Intramolecular cyclizations occurred upon treatment of 15 and
16 with potassium carbonate to form indazole cores 17 and 18 in
45–60% yield. Compounds 19a–b and 20a–j were obtained through
Suzuki, Buchwald or amide coupling procedures similar to the pro-
cedures described in Scheme 1.
NH
N
0.71
0.71
O
NH
NH
NH
8g
8h
0.44
0.33
O
O
NH
O
NH
8i
0.31
0.59
0.22
NH
O
NH
NH
8j
NH
O
8k
NH
O
HN
8l
0.68
NH
O
8m
8n
0.38
0.67
NH
N
O
NH
NH
N
N
N
N
8o
0.83
NH
R2 (7-substitution)
NH
8p
8q
11.5
1.68
O
NH
NH
8r
0.24
The in vitro biological activity of all target compounds was
evaluated in an enzyme assay that was developed to measure
O
a
High Throughput Mass Spectrometry (HTMS, BioTrove RapidFire™) format. IC50
was determined in a 12 point dose–response curve under the established steady-
state conditions of 200 M fructose, 100 M ATP and 2 nM KHK for 60 min at 25 °C.
KHK-mediated conversion of
D-fructose to fructose-1-P (F-1-P)
using High Throughput Mass Spectroscopy (HTMS) as a means of
product detection.12 Our SAR studies initially focused on the pen-
dant amino group (R2) of indazole 8. Table 1 showed KHK enzyme
activity of key compounds 8a–r. In general, modifications at 6-po-
sition of indazole 8 could generate compounds with good activity
provided that a basic amino group existed on the terminal position,
with an exception of a terminal pyridinyl group noted in 8d. After
examining a range of substituents at 6-position of indazole scaffold
8, we found that terminal piperidinyl or piperazinyl group (8a & 8b
in Table 1) provided good KHK enzyme activity (IC50 of 0.33 and
l
l
dine (THP) substitution presented better potency with IC50 of
0.08 M. The steric hindrance of the piperazinyl ring was tolerated
as illustrated by 8e (IC50, 0.71 M) containing a bulky bicyclo-
[2,2,1] ring system. We also modified the terminal basic amino
group by introducing R2 of variable length at the 6-position of
indazole core. Compounds 8f and 8g bearing one- and two-atoms
between the core and the piperidine showed comparable activities
l
l
0.56 lM, respectively). Compound 8c bearing a tetrahydro-pyri-

X. Zhang et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4762–4767
4765
as 8a with direct linkage. Varying the terminal basic amino group
played IC50 of 1.68 and 0.24
lM, respectively. Molecular modeling
at either 3- or 4-position of piperidine ring (8h & 8i in Table 1) re-
sulted in equipotent compounds. The ring size was also well toler-
ated as illustrated by 8h, 8j and 8k containing corresponding 6, 5
and 4 membered rings. We also observed that the length of the
linkers at 6-position of indazole scaffold 8 was tolerated for KHK
enzyme activity. Homologs of 8h with longer pendant side chains
of 7-substituted indazoles suggested the terminal basic amino
group of 8p was too rigid to form a tight charge–charge interaction
with carboxylate side-chain of b-clasp residue on the enzyme. Un-
der this circumstance, installation of linkers on the side chain such
as in 8r could extend the side chain deeper into the ATP binding re-
gion to reach the key interaction with the Asp27B.
were also tolerated for sub-
(IC50, 0.68 and 0.38 M, respectively). This was further demon-
strated by 8n and 8o bearing bicyclic pendant basic side chains
(IC50, 0.67 and 0.83 M, respectively). SAR clearly indicated that
l
M activities, as shown in 8l and 8m
We then sought to investigate the substitution effect (R3) of a
1-phenyl group on KHK enzyme activity. This sub-set of
compounds (12a–g) all have a THP group at the 6-position of the
indazole core. KHK enzyme activities were summarized in Table
2. Most compounds bearing small 3- or 4-substitution on the
phenyl ring retained good potency whereas 4-nitro and 4-amino
substituted phenyl analogs 12f and 12g tended to be less potent.
Whilst fluoro analogs 12a and 12b were just slightly more potent
l
l
there is room for modifications in this region for addressing
physical properties of this scaffold. We next turned our attention
to 7-position of indazole 8. Analog 8p with a 7-THP substitution
displayed considerably lower KHK enzyme activity (IC50
11.5 M) compared with the corresponding 6-THP counterpart 8c
(IC50, 0.08 M). Insertion of some linkers between the indazole
,
l
than 8c (IC50, 0.042, 0.023 lM vs 0.08 lM), both compounds
l
showed better metabolic stability in rat liver microsomes than 8c
(12a, 96% remaining at 10 min; 12b, 105% remaining at 10 min;
8c, 72% remaining at 10 min).
core and the pendant amino side chain resulted in significant
improvement in potency. For example, compound 8q with a one
atom oxy linker and 8r with a two atom methyleneoxy linker dis-
According to X-ray crystallographic information from library
screening, replacement of 3-thiomethyl group on pyrazole 1 with
other substitutions was not well tolerated. Thiomethyl groups
are usually considered as a metabolically unfavorable functional
group due to high tendency toward chemical or enzymatic
oxidation. To complete SAR of the indazole series and to improve
metabolic stability of indazole 2, we replaced 3-thiomethyl with
methoxy, methyl or ethyl groups (Table 3). Not surprisingly,
3-methoxy and 3-methyl substituted analogs showed reduced
Table 2
SAR on R3 group of indazole 12
MeS
N
N
NH
R3
Table 4
12
SAR on R2 group of indazole 20
ID
R3
KHK enzyme IC50 (lM)
Et
N
8c
H
3-F
4-F
3-Me
4-Me
4-OMe
4-NO2
4-NH2
0.080
0.042
0.023
0.034
0.04
0.087
0.48
0.16
12a
12b
12c
12d
12e
12f
12g
R2
N
20
ID
R2
KHK enzyme IC50 (lM)
N
N
NH
20a
20b
0.57
0.56
Table 3
SAR on R1 and R2 groups of indazole I
N
NH
R1
Me
20c
0.86
N
N
N
N
N
N
N
NH
N
R2
N
Me
Et
20d
20e
20f
20g
20h
20i
4.52
11.5
0.59
0.92
3.52
9.80
N
I
ID
R1
R2
KHK enzyme IC50
(
lM)
NH
NH
N
N
N
N
N
N
N
N
N
N
NH
N
8b
SMe
SMe
Me
Me
OMe
Et
0.56
0.67
1.28
3.20
1.38
0.57
0.56
NH
8n
N
NH
NH
19a
19b
7a
HN
N
NH
NH
NH2
NH
O
20j
0.68
0.94
N
NH
O
N
NH
N
20a
20b
20k
NH
Et
NH

4766
X. Zhang et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4762–4767
potency compared with the corresponding 3-thiomethylated
indazoles. Compounds 7a, 19a and 19b only showed low
KHK enzyme activity. In contrast to the observation from library
screening, some 3-ethylated analogs were equipotent to their cor-
responding 3-thiomethyl indazoles. For example, replacement of
at 6-position was well tolerated for sub-
l
M KHK enzyme activity
lM
(IC50, 0.59
lM). According to SAR findings in Table 1, it is further
suggested that the size and length of the R2 group were well toler-
ated in this region. While the basic amino group was crucial for
KHK enzyme activity for indazoles 20, a well positioned linker
had great impact for the favorable ionic interaction. Replacement
of the more flexible piperidine ring of 20b with the rigid phenyl
thiomethyl in 8b (IC50 of 0.56
lM) and 8n (IC50 of 0.67
lM) with
ethyl gave 20a and 20b with IC50 of 0.57 and 0.56
lM, respectively,
which demonstrated that the ethyl group was an effective bioiso-
ring of 20i reduced activity almost twenty fold (IC50, 9.80 lM),
stere for the thiomethyl group.
which indicated the planarity of the phenyl ring directed the ter-
minal basic amino group to an unfavorable distance from Asp27B.
It is interesting to observe that compounds bearing terminal basic
groups other than secondary amines could still maintain decent
potency. For example, compound 20j with a terminal amidate
and 20k with a terminal imidazole exhibited good KHK enzyme
We then turned our attention on optimization of the 3-Et-inda-
zole on R2 substitution (Table 4). Both steric effects and the basicity
of the terminal amine play important roles for efficient interaction
with the acidic ASP27B. While 20c bearing a
terminal amine showed a slight potency drop compared with 20a
(IC50 of 20a, 0.57 M vs 20c, 0.86 M), methylation of the second-
ary amino group of 20a resulted in more than seven fold decrease
in potency in 20c (IC50, 4.52 M). Ethylation led to even more
activity loss as shown in 20d (IC50, 11.5 M). Pyridinyl analog
20h bearing a relatively weaker basic group displayed weaker
KHK enzyme activity (IC50, 3.52 M) as compared with 20b and
20g (IC50, 0.56 and 0.92 M), although the rigidity of the pyridine
a methyl group to the
l
l
activity with IC50 of 0.68 and 0.94 lM, respectively.
We also incorporated structurally similar scaffolds to expand
the scope around the indazole template. To this end, isomeric
indazole 21, pyrazolo-pyrimidinone 22 and imidazopyridine 23
were synthesized and their KHK enzyme activities were evalu-
ated as shown in Figure 2. Although these isosteric heterocyclic
6,5-fused core systems showed striking structural similarity for
maintaining H-bonding between 2-nitrogen atom of the five
membered heterocycle with the conserved water, they may pro-
vide very different physicochemical property. The data indicated
l
l
l
l
ring could be another factor on its reduced activity. It was not sur-
prising to observe that 20f bearing a 5,5-fused amino substitution
that compound 21 maintains KHK enzyme activity (IC50
1.28 M), while compound 22 with pyrazolo-pyrimidinone skel-
eton lost five fold activity (IC50, 6.10 M) compared to indazole
19a. The imidazopyridine core shown in 23 was not tolerated
at all.
The X-ray crystal structure of 20f bound to the active site of
KHK revealed that 20f binds in the ATP site following the same
fashion as FBDD hit 1 (Fig. 3).13 The 2-nitrogen of indazole core
formed key H-bonding to the conserved water. The 3-ethyl group
of indazole 20f projected toward the aromatic side chain of
Phe260 in the pocket to form a hydrophobic interaction. The flat
,
Me
N
l
Me
N
Et
N
l
O
N
N
N
N
NH
N
N
NH
NH
NH
22
23
21
IC50 1.28 µM
IC50 6.10 µM
IC50 > 20 µM
indazole core bound to the lipophilic residues from both the
a/b/
Figure 2.
a
and b-sheet domains of KHK at the narrow central region of
Figure 3. X-ray co-crystal structure of 20f bound to the active site of KHK.

X. Zhang et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4762–4767
4767
Table 5
In vivo rat DMPK profiles of selected indazolesa
Cpd
POa
t1/2 (h)
Cmax (ng/lL)
AUClast (h à ng/
l
L)
IVb
Vss (L/kg)
CL (mL/min/kg)
F (%)
20f
20g
10 mpk
10 mpk
3.41
19.6
619
399
5679
4887
2 mpk
2 mpk
7.81
13.9
28.9
21.2
98.8
72.4
a
PO (10 mg/kg) in 0.5% methocel.
IV (2 mg/kg) in 20% HPBCD.
b
4. Maryanoff, B. E.; O’Neill, J. C.; McComsey, D. F.; Yabut, S. C.; Luci, D. K.; Jordan,
A. D.; Masucci, J. A.; Jones, W. J.; Abad, M. C.; Gibbs, A. C.; Petrounia, I. ACS Med.
Chem. Lett. ACS ASAP.
5. Gibbs, A. C.; Abad, M. C.; Zhang, X.; Tounge, B. A.; Lewandowski, F. A.; Struble,
G. T.; Sun, W.; Sui, Z.; Kuo, L. C. J. Med. Chem. 2010, 53, 7979.
6. Warr, W. A. J. Comput. Aided Mol. Des. 2009, 23, 453.
7. Congreve, M.; Chessari, G.; Tisi, D.; Woodhead, A. J. J. Med. Chem. 2008, 51,
3661.
8. Orita, M.; Warizaya, M.; Amano, Y.; Ohno, K.; Niimi, T. Expert Opin. Drug Discov.
2009, 4, 1125.
9. De Kloe, G. E.; Bailey, D.; Leurs, R.; de Esch, I. J. P. Drug Discovery Today 2009, 14,
630.
the active site. Additionally, a small hydrophobic pocket was
formed between the 1-substituted phenyl group and three sequen-
tial proline residues, Pro246-248. The tight binding mode of the
central region of the active site allowed the carboxylate side-chain
of b-clasp residue Asp27B to pack next to the fragment. The [3.3.0]
bicyclic amino side chain attached to the indazole core in the
closed catalytic site and the terminal amino group was engaged
in both ionic interactions with the carboxylate residue of Asp27B
and hydrogen bonding with Asn107.
Selected compounds according to their eADME profiles were
evaluated in a 24 h rat pharmacokinetic study at an oral dose of
10 mg/kg and iv dose of 2 mg/kg (Table 5). Compounds 20f
and 20g demonstrated good oral bioavailability (98.8% and
72.4%) and good oral exposure in plasma (AUClast = 5679 and
4887 ng h/mL). Both compounds had moderate clearance (28.9
and 21.2 mL/min kg) and volume of distribution at steady state
(Vss = 7.81 and 13.9 L/kg) within our target range. In addition, sev-
eral compounds from this scaffold were submitted to the Cerep
‘Comprehensive Pharmacological Profile’ panel, including more
than 100 biological targets (GPCRs, ion channels, transporters
and enzymes). All compounds showed clean selectivity profiles at
10. 3-Ethyl-6-(hexahydropyrrolo[3,4-c]pyrrol-2(1H)-yl)-1-phenyl-1H-
indazolehydrochloride (20f):
Into a 1000-mL round-bottom flask, was placed a solution of 1-(4-bromo-2-
fluorophenyl)propan-1-one (32 g, 138.53 mmol, 1.00 equiv) in ethanol
(300 mL), 1-phenylhydrazine (15.0 g, 138.89 mmol, 1.00 equiv), TSA-H2O
(1.32 g, 6.95 mmol, 0.05 equiv). The resulting solution was heated to reflux
for 1 h in an oil bath. The resulting mixture was concentrated under vacuum to
yield 1-(1-(4-bromo-2-fluorophenyl)propylidene)-2-phenylhydrazine as
a
yellow solid (44.3 g, yield: 99%). MS: 322 (MH+).
Into a 1000-mL 3-necked round-bottom flask, was placed a solution of 1-(1-(4-
bromo-2-fluorophenyl)propylidene)-2-phenylhydrazine (44.3 g, 138.01 mmol,
1.00 equiv) in N,N-dimethylformamide (400 mL), potassium carbonate (83 g,
601.45 mmol, 4.40 equiv). The resulting solution was stirred for 2 days at
100 °C in an oil bath. The resulting mixture was concentrated under vacuum.
The resulting solution was diluted with water (500 mL). The resulting solution
was extracted with ethyl acetate (2 Â 300 mL) and the organic layers
combined. The resulting mixture was washed with water (2 Â 300 mL) and
brine (1 Â 300 mL). The resulting mixture was dried over anhydrous sodium
sulfate and concentrated under vacuum. The residue was applied onto a silica
gel column with ethyl acetate/petroleum ether (1:50) to yield 6-bromo-3-
ethyl-1-phenyl-1H-indazole as a yellow solid (20.8 g, yield: 50%). MS (m/z):
301 [M+H]+; 1H NMR (400 MHz, CDCl3, ppm): d 1.42–1.46 (3H, t), 3.01–3.07
(2H, dd), 7.25–7.86 (8H, m).
Into a 100-mL 3-necked round-bottom flask purged and maintained with an
inert atmosphere of nitrogen, was placed a solution of 6-bromo-3-ethyl-1-
phenyl-1H-indazole (300 mg, 1.00 mmol, 1.00 equiv) in toluene (30 mL), tert-
butyl hexahydropyrrolo[3,4-c]pyrrole-2(1H)-carboxylate (212 mg, 1.00 mmol,
1.00 equiv), Pd(OAc)2 (2.24 mg, 0.01 mmol, 0.01 equiv), Cs2CO3 (482 mg,
2.50 mmol, 2.50 equiv), BINAP (18.7 mg, 0.03 mmol, 0.03 equiv). The
resulting solution was heated to reflux overnight in an oil bath. The resulting
mixture was concentrated under vacuum. The residue was applied onto a silica
gel column with dichloromethane/methanol (100:1). The residue was
dissolved in hydrogen chloride/MeOH (50 mL) and then stirred for 2 h at
room temperature. The resulting mixture was concentrated under vacuum and
the residue by re-crystallization from diethyl ether (50 mL). The solids were
collected by filtration, was dried in an oven under reduced pressure to yield 3-
ethyl-6-(hexahydropyrrolo[3,4-c]pyrrol-2(1H)-yl)-1-phenyl-1H-indazole-
10
l
M in the full panel.14
In summary, a lead generation process has been conducted for
discovering novel indazoles as KHK inhibitors. Optimization of an
indazole series according to KHK enzyme activity produced com-
pounds with nanomolar KHK enzyme activity. Furthermore, the
PK profiles of lead compounds proved to be acceptable with mod-
erate blood clearance, high volume of distribution, and high oral
bioavailability. The lead compounds from this series therefore de-
serve to be further explored in vivo to exploit the therapeutic po-
tential as KHK inhibitors for metabolic diseases.
Acknowledgments
The authors would like to thank ADME/PK, Analytical Research
and Lead Generation Biology teams at Johnson & Johnson Pharma-
ceutical Research and Development for service support. The
authors would also like to thank the staff at IMCA-CAT for their
assistance during X-ray data collection. The use of the IMCA-CAT
beam lines 17-ID and 17-BM at the Advanced Photon Source was
supported by the companies of the Industrial Macromolecular
Crystallography Association through a contract with Hauptman–
Woodward Medical Research Institute. Use of the Advanced
Photon Source was supported by the US Department of Energy,
Office of Science, Office of Basic Energy Sciences, under Contract
No. W-31-109-Eng-38.
hydrochloride as
a white solid (170 mg, yield: 46%). MS (m/z): 333
[MÀHCl+H]+; 1H NMR (400 MHz, DMSO, ppm): d 1.33 (3H, t), 2.92 (2H, q),
3.08 (4H, m), 3.42 (6H, m), 3.89 (3H, s), 6.71–7.72 (8H, m), 9.35 (2H, s).
11. Wang, C. H.; Feng, T. C.; Christensen, B. E. J. Am. Chem. Soc. 1950, 72, 4887.
12. An enzymatic assay was developed to measure KHK-mediated conversion of D-
fructose to fructose-1-P (F-1-P) using High Throughput Mass Spectroscopy
(HTMS) as a means of product detection. This assay served as a primary screen to
evaluate the ability to inhibit KHK enzyme activity and it has been adapted to
high throughput mass spectrometry (HTMS, BioTrove RapidFire™) format for
higher throughput. The compounds to be tested were dosed in 12-points.
Inhibition of the fragment, IC50, was determined in a dose–response curve under
the established steady-state conditions of 200 lM fructose, 100 lM ATP and
2 nM KHK for 60 min at 25 °C. The assay was carried out in 384-well plate format.
13. The atomic coordinate and structure factor for the KHK complexes with
compound 20f have been deposited in the Protein Data Bank under accession
code (pdbid: 3NBW).
References and notes
1. Raushel, F. M.; Cleland, W. W. Biochemistry 1977, 16, 2169.
2. Gross, L. S.; Li, L.; Ford, E. S.; Liu, S. Am. J. Clin. Nutr. 2004, 79, 774.
3. Basciano, H.; Federico, L.; Adeli, K. Nutr. Metab. 2005, 2, 5.
14. A representative compound 20b in Cerep panel: only hit, Na channel, 61% at
10 lM.