Discovery of N-(6-(5-fluoro-2-(piperidin-1-yl)phenyl)pyridazin-3-yl)-1-(tetrahydro-2H-pyran-4-yl)methanesulfonamide as a brain-permeable and metabolically stable kynurenine monooxygenase inhibitor

Article abstract of DOI:10.1016/j.bmcl.2021.128115

Kynurenine monooxygenase (KMO) is expected to be a good drug target to treat Huntington's disease (HD). This study presents the structure-activity relationship of pyridazine derivatives to find novel KMO inhibitors. The most promising compound 14 resolved the problematic issues of lead compound 1, i.e., metabolic instability and reactive metabolite-derived side-effects. Compound 14 exhibited high brain permeability and a long-lasting pharmacokinetics profile in monkeys, and neuroprotective kynurenic acid was increased by a single administration of 14 in R6/2 mouse brain. These results demonstrated 14 may be a potential drug candidate to treat HD.

Full text of DOI:10.1016/j.bmcl.2021.128115

Bioorg. Med. Chem. Lett. 44 (2021) 128115
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of N-(6-(5-fluoro-2-(piperidin-1-yl)phenyl)
pyridazin-3-yl)-1-(tetrahydro-2H-pyran-4-yl)methanesulfonamide as a
brain-permeable and metabolically stable kynurenine
monooxygenase inhibitor
Katsunori Tsuboi, Hidenori Kimura, Yoshie Nakatsuji, Momoe Kassai, Yoko Deai,
Yoshiaki Isobe*
Drug Research Division, Sumitomo Dainippon Pharma. Co., Ltd., 3-1-98, Kasugade-naka, Konohana-ku, Osaka 554-0022, Japan
A R T I C L E I N F O
A B S T R A C T
Keywords:
Kynurenine monooxygenase (KMO) is expected to be a good drug target to treat Huntington’s disease (HD). This
study presents the structure-activity relationship of pyridazine derivatives to find novel KMO inhibitors. The
most promising compound 14 resolved the problematic issues of lead compound 1, i.e., metabolic instability and
reactive metabolite-derived side-effects. Compound 14 exhibited high brain permeability and a long-lasting
pharmacokinetics profile in monkeys, and neuroprotective kynurenic acid was increased by a single adminis-
tration of 14 in R6/2 mouse brain. These results demonstrated 14 may be a potential drug candidate to treat HD.
Kynurenine monooxygenase
KMO
Kynurenine pathway
Huntington’s disease
KYNA
3-HK
Brain permeable
BBB
R6/2
Huntington’s disease (HD) is a genetic neurodegenerative condition
caused by a CAG repeat expansion in the huntingtin gene, and is a
dominantly inherited progressive neurological disorder and presents a
combination of motor, cognitive, and psychiatric problems that progress
over a 20-year period until death.1 Tetrabenazine is a dopamine-
depleting agent that is effective for reducing chorea, although it risks
potentially serious adverse effects as depression, suicide, and car-
diotoxicity. Furthermore, tetrabenazine is a symptomatic therapy, and
the disease modifying agents are not available in clinic. Similar to drug
development for other neurological diseases, failure rates in HD drug
development remain high.2
(KMO) activity shifts the flux between the two branches of the KP to-
ward increased KYNA formation, thus generating a neuroprotective
environment (Fig. 1). KMO inhibitors can correct the 3-HK/KYNA
imbalance in HD patients and are therefore considered in strategies to
prevent or arrest neurodegeneration.
A number of KMO inhibitors have been reported, but no compounds
have entered into clinical development.6 All compounds in Fig. 2
contain an acidic functional group as a common feature. The acidic
compounds carry a negative charge at physiological pH, making it
difficult for them to cross the blood–brain barrier (BBB). Ro-61-8048
demonstrated KMO inhibition both in vitro and ex vivo, but its efficacy
in the brain was poorer than that in the liver.7 We also confirmed the
brain penetration of Ro-61-8048 and CHDI340236 in a mouse phar-
macokinetic (PK) study, but their brain penetration was <1%, indicating
negligible brain exposure. Zwilling et al. reported that JM-6, a non-brain
permeable KMO inhibitor, was effective at regulating brain kynurenine
metabolism.8 However, the PK/PD relationship was unclear, since the
The kynurenine pathway (KP) is the major route for tryptophan
metabolism in mammals and produces neuroactive metabolites. KP
metabolite imbalances lead to elevated levels of the free-radical gener-
ator 3-hydroxykynurenine (3-HK) and the excitotoxin quinolinic acid
(QA) relative to the neuroprotective metabolite kynurenic acid (KYNA),
contributing to the pathogenesis of HD.3,4 Early-stage HD patients have
an increased 3-HK/KYNA ratio in the striatum and cortex compared with
healthy controls.5 The downregulation of kynurenine monooxygenase
KMO inhibition by JM-6 was very weak (IC50 = 20 μM) and JM-6 was
metabolically unstable.9 We previously reported compound 1 as a brain
* Corresponding author.
E-mail address: yoshiaki-isobe@ds-pharma.co.jp (Y. Isobe).
https://doi.org/10.1016/j.bmcl.2021.128115
Received 26 March 2021; Received in revised form 4 May 2021; Accepted 13 May 2021
Available online 17 May 2021
0960-894X/© 2021 Elsevier Ltd. All rights reserved.

K. Tsuboi et al.
Bioorganic & Medicinal Chemistry Letters 44 (2021) 128115
Fig. 1. Kynurenine pathway of tryptophan metabolism.
O O
S
NH
O O
S
NH
N
N
N
OMe
OMe
OMe
OMe
N
S
N
S
O
H
CO2
N
N
O2
O2
Cl
Ro-61-8048
CHDI-340246
JM-6
:
:
:
µ
M
µ
M
µ
20
M
IC50 0.037
IC50
0.0005
IC50
Brain/Plasma=0.004
Brain/Plasma=0.005
H
N
N
O
O
N
N
N
S
N
O
O
Cl
H
CO2
1
µ
Compound
IC50 0.0024
GSK-065
:
M
:
µ
M
IC50 0.0023
Brain/Plasma=0.282
Fig. 2. Chemical structures of known KMO inhibitors and compound 1.
permeable and strong KMO inhibitor: its IC50 toward human KMO
(hKMO) was 2.4 nM, and the mouse PK study indicated its brain/plasma
concentration ratio was 0.282, much higher than the ratios of Ro-61-
8048 and CHDI-340246.10 Compound 1 corrected the 3-HK/KYNA
imbalance in the brain striatum and improved the cognition impair-
ment in R6/2 mice by oral administration, but the improvement of
motor function was not statistically significant. We hypothesized that
observation of the long-lasting PK profile is required to show potent in
vivo efficacy.
H
N
N
N
N
S
O
O
Reduce the electron
by
density
introduction of F atom
Reduce the
lipophylicity
1
Compound
MS
µ
1.07/0.395
(human/rats):
dGSH
L/min/mg protein
µ
0.852 mol/L
conjugation:
PAMPA: 50.6 10-6
(x
Solubility (pH 1.2/7.4):
Mouse PK: T1/2=0.3
cm/sec)
0.163/<0.001
Detailed evaluation of the drug metabolism and PK revealed that
compound 1 has two problematic issues to be resolved. First is that
compound 1 is not metabolically stable: its intrinsic clearance is 1.07
mg/mL
ml/min/kg
Cl=67
hr,
μ
L/min/mg protein for human and 0.395
μ
L/min/mg protein for rat
Fig. 3. DMPK profile of compound 1 and optimization strategy.
following incubation with hepatic microsomes and NADPH at 37 ◦C for
30 min. We concluded that this instability profile led to the short plasma
half-life (T1/2 = 0.3 h) of 1 in the above mouse PK study. The second
issue is that compound 1 risks the formation of reactive metabolites.
Reactive metabolites have been putatively linked to drug-related side
effects including liver dysfunction, agranulocytosis, and aplastic ane-
mia, etc. Dansylated glutathione (dGSH) has been used as a trapping
agent for the quantitative estimation of reactive metabolites.11 Com-
pound 1 was incubated with dGSH and human hepatocyte microsomes,
and the amount of compound 1-dGSH conjugate was measured using a
fluorescence detector and HPLC analysis. High dGSH-conjugate forma-
on 1, we planned to introduce a fluorine (F) atom at the phenyl moiety to
reduce the electron density and to replace the toluoyl group with a low-
lipophilic one to resolve the two issues (Fig. 3).
The synthetic route of pyridazine derivatives is shown in Scheme 1.
The reductive amination of commercially available 2-bromoaniline (A1-
4) and glutaraldehyde, followed by the reaction with triisopropyl borate
gave boronate (B1-4). Suzuki-Miyaura coupling of the boronate with 3-
amino-6-bromopyridazine in the presence of Pd(PPh3)4 afforded a pyr-
idazine intermediate (C1-4). Subsequently, the reaction of (C1-4) with
p-toluene sulfonyl chloride in pyridine gave targets 2–5 in moderate
yields. Compounds 6–16 were prepared by the reaction of C-2 with the
appropriate sulfonyl chloride in pyridine. The compound screening was
carried out by the hKMO assay: the mitochondrial fraction of human
liver was pre-incubated with the test compounds for 5 min, L-kynurnine
was added and incubated for 30 min, and then the amount of generated
3-HK (converted from L-kynurenine by hKMO) was measured by LC/MS/
tion was seen in compound 1 (0.852 μmol/L). In order to improve both
the metabolic stability and reactive metabolite formation, we planned
the following optimization strategy. The electron-rich benzene ring
often becomes the target site for CYP enzyme metabolism and dGSH-
conjugate formation. In general, a high degree of correlation between
the lipophilicity and metabolism is seen in drug discovery.12 Thus, based
2

K. Tsuboi et al.
Bioorganic & Medicinal Chemistry Letters 44 (2021) 128115
N
NH2
H
N
N
OH
B
N
N
c
a,
b
N
N
d
S
Br
6
3
OH
O
O
F
5
4
F
F
F
N
NH2
:
:
:
:
A-1
A-2
A-3
A-4
3-F
:
:
:
:
:
:
:
:
B-1
y.
25%
32%
32%
41%
C-1
y.
29%
3-F,
4-F,
5-F,
6-F,
3-F,
4-F
5-F
6-F
B-2
B-3
B-4
y.
y.
y.
C-2
C-3
C-4
y.
y.
y.
48%
48%
38%
;
2
;
3
;
4
;
5
y.
y.
y.
y.
50%
42%
65%
38%
4-F,
5-F,
6-F,
3-F,
4-F,
5-F,
6-F,
H
N
R1
N
R1
1
O
:
6 R
y.
y.
8%
14%
y.
=Et,
=Pr,
N
S
1
1
O
e
:
7
O
O
R
:
C-2
8 R
35%
y. 30%
=i-Bu,
13
14
16
(y. 50%)
(y. 45%)
1=CH
F
N
:
9 R
2-c-Pen,
O
,
1=CH2CF3 y. 9%
O
:
10 R
1=(CH2)2
:
:
11
12
R
R
y.
58%
OMe,
6-17
1=(CH2)3
y.
15
17%
(y. 65%)
OMe,
(y. 37%)
Scheme 1. Synthetic route for compounds 2–16. Reagents and conditions: (a) OHC(CH2) 3CHO, NaBH(OAc)3, THF, (b) (i) BuLi, THF, (ii) B(Oi-Pr)3, THF, ꢀ 78 ◦C, (c)
3-amino-6-bromopyridazine, Pd(PPh3)4, EtOH/toluene, 120 ◦C, (d) p-toluene sulfonyl chloride, pyridine, (e) appropriate sulfonyl chloride, pyridine.
Table 1
SAR exploration of compounds 1–5.
Compd
F-position
hKMO
microsomal stabilitya
dGSH
(μmol/L)
mice PK (2 mg/kg, po)
Brain (B)c
IC50 (nM)
human
rat
0.395
Plasma (A)b
808 ± 117 d
Ratio (B/A)
0.28
1
2.4
20.0
9.1
1.071
0.178
0.41
0.852
0.146
0.073
0.669
0.637
0.221
230 ± 51 d
2
3
4
5
6
0.518
0.272
0.342
0.339
<0.05
3
183 ± 69
267 ± 57
41 ± 12
43 ± 2
0.23
0.17
0.28
0.005
4
4.2
0.178
0.478
0.49
5
6.1
284 ± 56
79 ± 11
Ro-61-8048
33.8
7303 ± 800 d
39 ± 10d
a: μL/min/mg protein, b: ng/mL, c: ng/g tissue, d: 10 mg/kg, po.
MS.
We replaced the toluoyl moiety in 3 with an alkyl group. The hKMO
inhibitory activity of ethyl (6) was 115 nM, approximately 13-fold
weaker than 3, but the inhibitory activity was enhanced by the alkyl
chain elongation: propyl (7: 37 nM), isobutyl (8: 27 nM), cyclo-
pentylmethyl (9: 29 nM). On the other hand, human MS may also
correlate with lipophilicity, and clogP should be <3.5 to obtain meta-
bolically stable compounds. Interestingly, the hKMO inhibitory activity
of trifluoroethyl (10) was 3-fold stronger than 6, suggesting the acidity is
important for hKMO inhibition. In contrast, the human MS worsened
with 10, because the lipophilicity of 10 was higher than of 6. In order to
reduce the lipophilicity (clogP: <3) we designed and prepared ether-
type compounds (11 and 12), finding they had favorable human MS
but not strong hKMO inhibition. The results of 6–12 suggested that it is
difficult to balance good human MS and hKMO inhibition. The results of
compounds 9 and 12 led us to prepare a cyclic ether analogue (13). The
human/rat MS of 13 was much improved compared with 9 while
maintaining the hKMO inhibitory activity. Thus, we investigated cyclic
ether analogues (14–16) and identified compound 14 with the best
balance. Although the hKMO inhibitory activity of 14 was 5-fold weaker
than 1, compound 14 showed good human/rat MS. Furthermore, the d-
GSH-conjugate formation of 14 was much lower than 1, suggesting 14
was a low-risk compound in terms of reactive metabolite-derived side-
effects. Therefore, compound 14 can be considered a candidate that
resolves the problematic issues of 1.
First, we introduced a F atom onto the phenyl moiety of 1 to reduce
the electron density. The results are summarized in Table 1. Compounds
3–5 showed single nM level potencies toward hKMO, but the IC50 of 2
was 20 nM, approximately 8-fold weaker than 1, suggesting the 3rd
position was unfavorable in terms of hKMO inhibition. The reduction of
electron density of phenyl group by F atom was often effective to
improve microsomal stability (MS) in drug discovery.13 Human MS was
significantly improved by all compounds, as we expected, but rat MS
was not. Introduction of the F atom at the 3rd or 4th position was
effective at reducing the conjugate formation by compounds 2 and 3 but
not for compounds 4 and 5. Unfortunately, it was difficult to understand
these results from the view point of electron density. We evaluated the
brain permeability of test compounds 3–5 by a mouse PK study, finding
that the brain/plasma concentration ratios of the compounds were
similar to that of compound 1, indicating the F atom did not influence on
brain permeability. Considering a balance of hKMO inhibition, human/
rat MS, and the dGSH conjugate, the 4th position could be considered
best for introduction of the F atom. The biological profile of 3 is superior
to compound 1; however, further improvement in the MS is needed to
obtain long-lasting compounds. Compound 3 is a lipophilic compound
(clogP = 4.4). Therefore, we speculated that further reduction of the
lipophilicity may be useful to improve the MS. Accordingly, we explored
optimization at the toluoyl moiety in compound 3 (see Table 2).
3

K. Tsuboi et al.
Bioorganic & Medicinal Chemistry Letters 44 (2021) 128115
should be able to cross the BBB to reach the target tissue. We tested the
brain permeability of 14 in cynomolgus monkeys. Blood and brain
samples were collected two hours after the oral dosing of 14. As shown
in Table 3, compound 14 exhibited good brain penetration, and the
striatum/plasma concentration ratio reached 0.43. In addition, we
tested the brain permeability of CHDI-340246 as a reference. The
plasma concentration of CHDI-340246 was very high by oral adminis-
tration, but the striatum/plasma ratio was only 0.04, 100-fold lower
than that of compound 14.
Table 2
SAR exploration of compounds 6–16.
Compound
R1
hKMO
IC50
clogP
microsomal
stabilitya
dGSH
(μ
mol/
L)
R6/2 transgenic mice are a well-characterized animal model
mimicking many of the histopathological aspects of HD.14 In R6/2 mice,
motor symptoms, such as dyskinesia, ataxia, and clasping behavior, start
at the age of 6 weeks. From the age of 9–10 weeks, the mice show sig-
nificant neuronal dysfunction and display neuronal atrophy in the
striatum.15 An increase in 3-HK levels was also reported in the brains of
R6/2 mice.16,17 We tested the efficacy of 14 on the localization of KP-
metabolites in R6/2 mice. Compound 14 was orally dosed in 8-week-
old R6/2 mice. The amount of KP metabolites in the striatum was
measured at 1 h after the dosing. The results are summarized in Fig. 5.
Although the level of neurotoxic 3-HK did not change with the dose,
neuroprotective KYNA was increased >10-fold in a dose-dependent
manner. Thus, the 3-HK/KYNA ratio was improved, indicating a neu-
roprotective environment. In another experimental, 45% reduction of 3-
HK by 14 administration (100 mg/kg, po) was observed at 24 h after
dosing. This result suggested the half-life of 3-HK in brain might be long,
and we speculated that repeated administration is required to show clear
3-HK reduction in brain. Our pharmacology colleagues have a plan to
test 14 on the motor/cognition functions and histopathological findings
of R6/2 mice by repeated administration, and the detail results will be
reported in the future.
(nM)
human
rats
3
6
4-Me-Ph
Et
9.1
115.5
37.1
27.2
28.6
38.7
83.9
61.6
41
4.4
0.41
0.272
0.073
3.01
3.54
3.94
4.57
3.28
2.6
0.156
0.179
0.375
0.703
0.569
0.179
0.026
0.179
<0.05
0.168
0.483
1.6
0
0
7
n-Pr
8
i-Bu
0.11
0.49
0
9
CH2-c-Pen
CH2CF3
(CH2)2OMe
(CH2)3OMe
10
11
12
13
0.265
0.05
0
2.98
3.1
0.045
0.086
0
0.289
14
15
16
12.8
17.9
34.1
2.73
2.73
3.53
0.068
0.081
0.139
0.218
0.23
0.092
ntb
0.435
0
Ro-61-
8048
33.8
3.23
0.49
<0.05
0.22
a:
μ
L/min/mg protein, b: nt means not tested.
The PK profile of compound 14 was characterized in cynomolgus
In conclusion, we investigated brain-permeable KMO inhibitors with
long-lasting PK profiles for the treatment of HD. We conducted an
optimization study based on compound 1. The introduction of an F atom
onto the 3rd position of the phenyl moiety improved human MS and d-
GSH-conjugate formation. Replacement of the toluoyl moiety in 3 with a
monkeys. Compound 14 demonstrated good clearance (8.0 mL/min/kg)
with a favorable plasma half-life of 6.2 h when intravenously adminis-
tered. Compound 14 also exhibited good oral exposure (BA = 25%) and
a long-lasting profile when dosed orally, as shown in Fig. 4. In order to
be an effective neuroprotective agent for HD treatment, a compound
i.v. 0.482 mg/kg
p.o. 5 mg/kg
10000
1000
100
10
1
0.1
0
4
8
12
16
20
24
time (hr)
Fig. 4. Plasma concentration curve of 14 in cynomolgus monkeys.
Table 3
Striatum/plasma concentration ratio of 14 and CHDI-340246 in cynomolgus monkeys.
Compound
Time
(hr)
Animal No.
Plasma
Striatum
(ng/g)
Striatum/Plasma
Ratio
(ng/mL)
14
2
1
2
873
318
940
0.36
1930
0.49
Mean
1
1402
629 (24 nMa)
56.0
0.43
CHDI-340246
2
14,200
17,000
15,600
0.004
0.004
2
73.5
Mean
64.8
a) Brain free concentration (protein binding ratio = 98.4%).
4

K. Tsuboi et al.
Bioorganic & Medicinal Chemistry Letters 44 (2021) 128115
Fig. 5. 3-HK and KYNA concentrations in R6/2 mice brain.
cyclic ether group further improved human and rat MS. As a result of the
optimization, we succeeded to identify compound 14 as a well-balanced
compound in terms of hKMO inhibition, human/rat MS, and dGSH-
conjugate formation. A cynomolgus monkey PK study indicated that
14 has good brain permeability with a long-lasting profile and was su-
perior to CHDI-340246 in terms of brain penetration. Neuroprotective
KYNA was significantly increased in R6/2 mouse brain by a single
administration of 14. Overall, compound 14 resolved the problematic
issues of 1 and is expected to be drug candidate for HD treatment.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bmcl.2021.128115.
References
´
1
2
3
4
5
Bachoud-Levi AC, Ferreira J, Massart R, et al. Front Neurol. 2019;10:710.
Caron NS, Dorsey ER, Hayden MR. Nat Rev Drug Discovery. 2018;17:729–750.
Maddison DC, Giorgini F. Semin Cell Dev Biol. 2015;40:134–141.
Chen Y, Guillemin GJ. Int J Trp Res. 2009;2:1–19.
Schwarz MJ, Guillmin GJ, Teipel SJ, Buerger K, Hampel H. Eur Arch Psychiatry Clin
Neurosci. 2013;263:345–352.
Declaration of Competing Interest
6
7
Zhang S, Sakuma M, Deora GS, et al. Commun Biol 2019;2:271–271.
Rover S, Cesura AM, Huguenin P, Kettler R, Szente A. J Med Chem. 1997;40:
4378–4385.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
8
9
Zwilling D, Huang SY, Sathyasaikumar KV, et al. Cell. 2011;145:863–874.
Beconi MG, Yates D, Lyons K, et al. Drug Metab Dispos. 2012;40:2297–2306.
10 Kimura H, Suda H, Kassai M, et al. Bioorg Med Chem Lett. 2021;33, 127753.
11 Attwa MW, Kadi AA, Abdelhameed AS. J Separation Sci. 2020;43:708–718.
12 Waring MJ. Expert Opin Drug Discov. 2010;5:235–248.
Acknowledgement
13 Clader JW. J Med Chem. 2004;47:1–9.
The authors would like to thank Ms. Mari Miyajima, Mr. Itaru Nat-
sutani and Dr. Kazuto Yamada for their technical support and useful
advice.
14 Mangiarini L, Sathasivam K, Seller M, et al. Cell. 1996;87:493–506.
15 Stack EC, Kubilius JK, Smith K, et al. J Comp Neurol. 2005;490:354–370.
16 Guidetti P, Bates GP, Graham RK, et al. Neurobiol Dis. 2006;23:190–197.
17 Sathyasaikumar KV, Stachowski EK, Amori L, Guitti P, Muchowski PJ, Schwarcz R.
J Neurochem. 2010;113:1416–1425.
5