Structure-activity relationship study of a series of caspase inhibitors containing γ-amino acid moiety for treatment of cholestatic liver disease

Article abstract of DOI:10.1016/j.bmcl.2018.04.002

A series of caspase inhibitors containing γ-amino acid moiety have been synthesized. A systemic study on their structure-activity relationship of anti-apoptotic cellular activity is presented. These efforts led to the discovery of compound 20o as a potent caspase inhibitor, which demonstrated preclinical ameliorating total bilirubin efficacy with a significantly improved pharmacokinetic profile.

Full text of DOI:10.1016/j.bmcl.2018.04.002

Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure-activity relationship study of a series of caspase inhibitors
containing
c-amino acid moiety for treatment of cholestatic liver disease
a
a
Jianfeng Mou a, Songliang Wu a, Zhi Luo a, Fengying Guo a, Haiying He a, , Jianhua Wang , Fusen Lin ,
Fengxun Guo a, Jianping Sun a, Liang Shen a, Minggao Zeng a, Chuan Wang a, Deming Xu a, Zhengxian Gu a,
Xin Tian b, Aiming Zhang b, Hongjiang Xu b, Ling Yang b, Xiquan Zhang b, Jian Li a, Shuhui Chen a
⇑
a WuXi AppTec (Shanghai) Co. Ltd., 288 Fute Zhong Road, Shanghai 200131, China
b Chia Tai Tianqing Pharmaceutical Group Co. Ltd., Building 9, No. 699-8, Xuanwu Road, Nanjing, Jiangsu 210023, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of caspase inhibitors containing c-amino acid moiety have been synthesized. A systemic study on
Received 5 December 2017
Revised 30 March 2018
Accepted 2 April 2018
Available online xxxx
their structure-activity relationship of anti-apoptotic cellular activity is presented. These efforts led to the
discovery of compound 20o as a potent caspase inhibitor, which demonstrated preclinical ameliorating
total bilirubin efficacy with a significantly improved pharmacokinetic profile.
Ó 2018 Published by Elsevier Ltd.
Keywords:
Apoptosis
Caspase inhibitor
c
-Amino acid
Chronic liver disease
Cell death is a fundamental process that removes unwanted or
damaged cells and ensures recycling of cellular constituents pro-
moting further growth and differentiation to maintain tissue
homeostasis. Based on the morphology of dying cells, two distinct
modes of cell death are studied including apoptosis and necrosis.
Apoptosis, or programmed cell death, is characterized by distinc-
tive morphological changes such as cell shrinkage and nuclear con-
densation.1–3 Dysregulation of apoptosis can lead to many
pathological conditions.4,5 Caspases are a family of cysteine pro-
teases that are involved in cellular apoptosis through both the
intrinsic and the extrinsic pathways.6 Besides, the release of pro-
inflammatory cytokines like IL-1b or IL-18 is also highly regulated
by caspase.7 In the past few decades, these enzymes have been
recognised as attractive targets for the treatment of diseases where
programmed cell death or the pro-inflammatory cytokines plays a
major role.8
liver in murine models of acute liver damage12, and attenuates
inflammation and fibrosis in murine models of cholestatic-13
,
ethanol-14 and methionine choline-deficient (MCD) diet-
induced15,16 liver injury. Thus, the use of caspase inhibitors could
be an attractive therapeutic approach for various types of liver dis-
ease (Fig. 1).
Two apoptosis-regulating pan-caspase inhibitors, IDN-655617,18
and GS-945019,20, have been studied in clinic for the treatment of
liver disease. IDN-6556, a first-in-class, orally active pan-caspase
inhibitor, has been studied in sixteen clinical trials across a broad
range of liver disease etiologies and stages of progression. During
its 6 months phase II clinical trial in patients with cirrhosis, IDN-
6556 had shown beneficial effects in patients with cirrhosis and
mild to moderate hepatic impairment.21 GS-9450 is also an orally
active caspase inhibitor with a unique scaffold that was shown to
have a fast onset and long duration of action as a potential hepato-
protective agent targeted for indications such as HBV, HCV hepati-
tis, and NASH.7 More recently, VX-166 was reported as a potential
therapy for sepsis22 and NASH23 in preclinical animal models.
Based on these encouraging results, we embarked on a journey
to develop novel caspase inhibitors to treat the unmet medicinal
needs, such as NASH and cholestatic liver diseases.
It has been established that apoptosis is a cardinal pathological
feature of many liver disease such as NASH, liver fibrosis and cir-
rhosis.9 The phagocytosis of apoptotic bodies by Kupffer cells10
and hepatic stellate cell (HSC)11 is one of the mechanism that pro-
mote inflammation and fibrogenesis. Consistent with this concept,
reducing hepatocytes apoptosis by caspase inhibitors protects the
According to a thorough analysis of the possible warheads,24 we
believe that tetrafluorophenoxymethylketone, as shown in IDN-
6556 or VX-166, is a desired moiety to interact irreversibly with
the catalytic cysteine residue. By incorporating unique moieties
⇑
Corresponding author.
E-mail address: he_haiying@wuxiapptec.com (H. He).
https://doi.org/10.1016/j.bmcl.2018.04.002
0960-894X/Ó 2018 Published by Elsevier Ltd.
Please cite this article in press as: Mou J., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.04.002

2
J. Mou et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
a plasma stability issue with 4, as only 67% of 4 left after 2 h incu-
bation with rat plasma, compared to 93% left for 6. Besides, one
more chiral center in 4 made the synthetic route more complex
and difficult. Indeed, compound 6 exhibited good oral systemic
exposure in mouse, with an area under the curve (AUC) of 3657
nMꢀh, and a maximum concentration (Cmax) of 1636 nM at a dose
of 10 mg/kg. The good cellular activities, combined with evidence
of good oral PK profiles, established this series as an excellent
starting point for optimization.
The synthesis of compound 6 was very convergent, as outlined
in Scheme 1. Commercially available Z-Asp(O-tBu)-OH was con-
verted to the bromomethyl ketone.25 Displacement with the
sodium salt of 2,3,5,6-tetrafluorophenol gave compound 11.
Reduction of the ketone and hydrogenolysis afforded the key inter-
mediate 12, which could then be coupled to the protected amino
acid of choice. The oxamic acid (14) was prepared using ethyl
chlorooxoacetate and 2-tert-butylaniline (13), followed by ester
hydrolysis. Then compound 14 was coupled with ethyl isonipeco-
Fig. 1. Chemical structures of selected irreversible caspase inhibitors.
connected to the warhead, which are different from a-amino acids,
we developed some novel molecules with excellent cellular activ-
ities, improved liver exposure and ADME properties. Additionally,
compound 20o, one of the best compounds we got, demonstrated
efficacy in a rodent model of cholestatic liver disease.
As shown in Table 1, our expedition began with the modifica-
tion of the medium part close to the warhead by incorporating dif-
tate. For other derivatives, different protected b or c-amino acids
were used. After hydrolysis, the carboxylic acid of interest could
be coupled with the key intermediate 12, and followed by oxida-
tion of the alcohol and subsequent tertbutyl ester hydrolysis pro-
vides 6 and other amino acid analogues in Table 1.
Next we set out to determine whether steric or electrostatic
interactions were important in the oxamide moiety. As shown in
Table 2, the tert-butyl group was incorporated to different posi-
tions of the terminal phenyl group (6–18b). All these compounds
could be synthesized similarly to that of 6, starting from different
ferent unnatural
evaluated by cell data, namely following potency in the Jurkat/
Fas (JFas) assay. Compared with -amino acid, direct one carbon
homologation (1) was not potent enough. Then different cyclic b
or -amino acids were introduced (2–8) and some of them did give
b or c-amino acids. All compounds were
a
c
very impressive results, such as compound 4 and 6. It is worth-
while to note that a clear stereochemical preference was observed
in the piperidine-3-carboxylic acid scaffold by comparison of
diastereomers 4 and 5, as the R diastereomer 4 was more active.
Compound 4 was 3-fold more potent than 6. However, there was
Table 1
SAR of the medium part, different b or c-amino acids.
Compd.
A
JFas IC50 (nM)
6
IDN-6556
1
138
2
3
213
142
4
21
5
6
142
64
Scheme 1. Reagents and conditions: (a) i. isobutyl chloroformate (1.5 equiv), N-
methylmorpholine (NMM) (1.6 equiv), THF, ꢁ10 °C, 40 min; ii. CH2N2/Et2O (2.5
equiv), 0 °C – room temperature (rt), 2.5 h; iii. HBr (aq 35%) THF, 0 °C – rt, 1 h; (b)
2,3,5,6-tetrafluorophenol (1.2 equiv), NaHCO3 (2 equiv), KI (0.1 equiv), DMF, 0 °C,
17 h; (c) NaBH4 (4 equiv), MeOH/THF = 1:1 (v/v), 0 °C – rt, 1 h; (d) H2 (1 atm), 10%
Pd/C, MeOH, rt, 4 h; (e) Ethyl chlorooxoacetate (1 equiv), TEA (1.1 equiv), 0 °C – rt,
16 h; (f) LiOH (1 equiv), 1,4-dioxane/H2O = 2:1 (v/v), rt, 1.5 h; (g) ethyl isonipecotate
(1 equiv), EDCI (1.5 equiv), HOBt (1.5 equiv), DIPEA (2 equiv), DCM, rt, 18 h; (h) LiOH
(2 equiv), THF/H2O = 1:1 (v/v), 0 °C – rt, 1.5 h; (i) compd.12 (1 equiv), EDCI (1.4
equiv), HOBt (1.4 equiv), NMM (3 equiv), DCM, rt, 60 h; (j) iodobenzene diacetate (4
equiv), TEMPO (0.2 equiv), DCM, rt, 24 h; (k) TFA/DCM = 1:2 (v/v), rt, 2 h.
7
8
75
244
Please cite this article in press as: Mou J., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.04.002

J. Mou et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
3
Table 2
Table 3
SAR of the oxamide moiety.
SAR of the carbamate moiety.
Compd.
R
JFas IC50 (nM)
Compd.
R
JFas IC50 (nM)
14
6
2-tert-butyl
3-tert-butyl
4-tert-butyl
2-H
2-F
2-Cl
64
19a
18a
18b
18c
18d
18e
18f
18g
112
108
40
27
79
19b
9
19c
19d
35
26
2-CN
2-CF3
27
25
19e
16
19f
14
24
19g
19h
19i
19j
20
30
14
Table 4
SAR of the amide moiety.
Fig. 2. Overlay of the structure of compound 6 (in green) and IDN-6556 (in purple)
with caspase-1 by molecular docking.
Compd.
R
JFas IC50 (nM)
14
20a
substituted anilines (More details reported in the supporting mate-
rial). It was obvious that the ortho-position was the best point to
find a favorable substitute to improve the activity. When the
tert-butyl group was removed, the corresponding compound 18c
provided some kind of improvement in potency, which gave a clue
that the 2-tert-butyl group was a little big and might not suitable
20b
22
20c
20d
23
11
for this new c
-amino acid scaffold. The incorporation of fluorine26
20e
44
was envisioned to increase cellular permeability and therefore
increase activity as well as modify the electronics without altering
its size. We were delighted to find that 18d provided further
improvement in potency, and other halogen analogues (18e–18g)
also gave similar results. Compared with IDN-6556 (JFas IC50 = 6
nM in the same assay), the best compound in the oxamic acid ser-
ies (18g) was still about 4-fold less potent. We believed that when
20f
7
20g
20
20h
18
the linker or P2 part was changed from
a-amino acid to c-amino
acid, it was a little longer for the terminal oxamide moiety to inter-
act with the P4 pocket of the corresponding caspase enzymes,
which was further confirmed by overlay of the docking model of
compound 6 and IDN-6556 with caspase-1 (Fig. 2). To investigate
the hypothesis, a little truncated carbamate or amide group was
introduced to improve the cellular activities.
20i
20j
20
19
20k
12
Table 3 and Table 4 list a set of diverse analogues with trun-
cated terminal moieties. The majority of these compounds showed
much improved activities and some of them (20f, 20m, 20n, 20o)
(continued on next page)
Please cite this article in press as: Mou J., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.04.002

4
J. Mou et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
(20i) or benzocyclic ring system incorporated (20e, 20h & 20j) all
Table 4 (continued)
resulted in a loss in activities. In consistent with the carbamate ser-
ies, alkyl substituted amide gave comparable or even better activ-
ities. Notably, cycloalkyl substituted analogues (20m-20o) showed
the best cellular activities. To our surprise, the adamantyl group
(20r) was also tolerant. The alkyl amide series suffered from low
lipophilicity which might preclude them from passive diffusion.
In an attempt to improve the Log D value, 3,3-difluorocyclobutyl
(20s) or 4,4-difluorocyclohexyl (20t) group was introduced and
the activity was restored.
Compd.
R
JFas IC50 (nM)
12
20l
20m
20n
20o
6
6
6
20p
30
After two rounds of modification of this novel
c-amino acid
20q
20r
10
20
scaffold, some compounds with comparable or improved activities
were found. But it is worthwhile to note that the liver exposure
was as important as the cellular activity. As such, representative
compounds were advanced to in vivo PK assessments. As shown
in Fig. 3, liver exposure of 20o was comparable to IDN-6556. The
thienyl amide compound (20f) was potent enough, but its expo-
sure in liver was not very satisfactory. Besides, the incorporation
of 3,3-difluorocyclobutyl (20 s) group did not result in improve-
ment in liver exposure. Taken together, 20o showed similar activ-
ity and comparable liver exposure as IDN-6556 which deserved
20s
20t
10
18
further evaluation in PK profiles and ANIT (
cyanate) model for cholestatic liver disease.
a-naphtylisothio-
Table 5 compares the mouse pharmacokinetics of IDN-6556 and
compound 20o and indicates that the latter showed a much slower
clearance which was consistent with prolonged plasma half-life,
better systemic exposure and improved bioavailability. It was sup-
posed that once-daily oral dosing should be taken for 20o based on
its dramatically improved PK profile, whereas bid dosing was
required for IDN-6556 in clinical trials. As for chronic liver disease,
a reduction in the frequency of administration can be beneficial to
patient compliance.
Excellent activity seen in the Jurkat cellular assay and improve-
ment in PK profiles prompted in vivo assessment of 20o in an ANIT
animal model. In this model, animals exposed chronically to low
doses of ANIT develop periportal inflammation, mild hepatocellu-
lar injury, significant bile duct proliferation, and progressive fibro-
sis.27 Besides, compared to model group, ANIT treatment group
resulted in significant elevation in serum total bilirubin (TB), which
is a key diagnostic indicator for liver injury. INT-747, which has
been used as a drug to treat primary biliary cholangitis, was used
as a positive control. When dosing 6 mg/kg once daily for only
one week, compound 20o could significantly reduce the content
of serum total bilirubin which indicates an improvement in liver
function (Fig. 4). Correspondingly, treatment with 20o improved
direct bilirubin (DB) and indirect bilirubin (IB), which are the com-
ponents of total bilirubin.
Fig. 3. Liver exposure comparison of representative compounds in mouse via oral
cassette dosing at 1 mg/kg for each compound.
provided comparable potency as IDN-6556, which further corrob-
orated our hypothesis. In general, 2-chloro-substituted phenyl car-
bamate (19b) showed the best overall activity, as well as other
halogenated phenyl carbamate (19e). Electronic effects were not
very important in this carbamate series (19e and 19g v.s. 19f and
19h), but the steric effect (19i) did result in reduced activities.
Though this aryl carbamate series showed better results than the
oxamide series, the majority of them suffered from rodent plasma
stability issue. For example, none of 19b was left after incubation
with rat or mouse plasma for 2 h. When it comes to alkyl substi-
tuted (19j), the cellular activity restored and fortunately the
plasma stability issue was devoid. As for the amide series, phenyl
ring substituted alone gave comparable result, and chlorine incor-
porated (20b-20d) did not contribute to further improvement, but
its isostere (20f) did. It seemed that this series had a narrow struc-
ture-activity relationship as exemplified by isoxazolyl replacement
A novel series of caspase inhibitors were discovered from a
promising replacement of the
a-amino acids existed in most of
the caspase inhibitors with -amino acids. After sophisticatedly
c
tuning the length of the analogues, we finally identified com-
pound 20o as a potent caspase inhibitor with excellent in vitro
and preclinical in vivo efficacy which is suitable for further
development.
Table 5
Mouse PK profiles for Compound 20o and IDN-6556.
Compd.
Dose (mg/kg)
Routea
Cmax (nM)
Vdss (L/kg)
CL (mL/min/kg)
T1/2 (h)
AUC0-1 (nMꢀh)
F (%)
IDN-6556
IDN-6556
20o
1
10
1
iv
Oral
iv
–
227
–
0.82
–
0.65
–
106
–
39
–
0.25
–
0.87
–
446
197
852
1708
–
4.4
–
20o
10
Oral
405
20.1
a
For iv and po dosing, 1% Tween80, 30% PEG400, 69% water was used as a vehicle.
Please cite this article in press as: Mou J., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.04.002

J. Mou et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
5
T B
D B
1 0 0
8 0
6 0
4 0
2 0
0
8 0
6 0
**
4 0
2 0
0
**
**
**
**
**
l
l
l
l
g
g
g
g
o
g
g
o
o
k
ro
k
k
r
k
k
r
t
/
/k
g
tr
/
t
g /
g
g /
n
n t
g
g /
n
n
o
m
m
m
6
6 m
co
0
6
6 m
-
co
c
-
6
e
co
h
lt
h
2 0 m
-2
le
t
l
cl
0 o -
7
5
i
7 -
5 6 -
5
a
ic
2
4
5
6
2 0 o
4
h
h
ea
7
6
-
H e
-7
H
V e
T -
V e
N
T
N -
IN
ID
IN
ID
IB
2 5
2 0
1 5
1 0
5
**
**
**
0
l
l
g
g
o
k g
r
k
k
/
ro
t
t
g /
g /
n
o
n
m
m g
6
co
6 m
c
0
e
h
t
-2
6 -
5
l
cl
7
i
a
2 0 o -
4
5
6
-
eh
-7
H e
V
T
N
IN
ID
Fig. 4. Evaluation of compound 20o and IDN-6556 in a rat ANIT cholestatic liver disease model measuring the reduction in total bilirubin (TB), direct bilirubin (DB) and
indirect bilirubin (IB). INT-747 was used as positive control. Data were presented as Mean SEM, ***p < 0.001 vs vehicle group, **p < 0.01 vs vehicle group, *p < 0.05 vs vehicle
group, analyzed by unpaired t-test.
12. Ueno Y, Ohmi T, Yamamoto M, et al. J Pharmacol Sci. 2007;105:201.
13. Canbay A, Feldstein A, Baskin-Bey E, Bronk SF, Gores GJ. J Pharmacol Exp Ther.
A. Supplementary data
2004;308:1191.
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.bmcl.2018.04.002.
14. Deaciuc IV, D’Souza NB, de Villiers WJ, et al. Alcohol Clin Exp Res. 2001;25:935.
15. Witek RP, Stone WC, Karaca FG, et al. Hepatology. 2009;50:1421.
16. Anstee QM, Concas D, Kudo H, et al. J Hepatol. 2010;53:542.
17. Hoglen NC. In: O’Brien T, Linton SD, eds. Design of caspase inhibitors as potential
clinical agents. CRC Press; 2008:211–224.
References
18. Linton SD, Aja T, Armstrong RA, et al. J Med Chem. 2005;48:6779.
19. Manns, M. P., Lawitz, E., Hoepelman, A. I. M. Abstract of papers, 45th Annual
Meeting of the European Association for the Study of the Liver (EASL 2010), Vienna,
Austria, 2010.
1. Galluzzi L, Bravo-San Pedro JM, Vitale I, et al. Cell Death Differ. 2015;22:58.
2. Kroemer G, Galluzzi L, Vandenabeele P, et al. Cell Death Differ. 2009;16:3.
3. Galluzzi L, Vitale I, Abrams JM, et al. Cell Death Differ. 2012;19:107.
4. Galle PR. J Hepatol. 1997;27:405.
20. Waters, R. Bloomberg News. April 19, 2010.
21. Frenette C, Morelli GM, Shiffman M, et al. In: 67th annual meeting of the
American Association for the Study of Liver Diseases (AASLD), Boston, America,
2016; Poster 2099.
22. Weber P, Wang P, Maddens S, et al. Crit Care. 2009;13:R146.
23. Witek RP, Stone WC, Karaca FG, et al. Hepatology. 2009;50:1.
24. Brady KD, Giegel DA, Grinnell C, et al. Bioorg Med Chem. 1999;7:621.
25. Dolle RE, Hoyer D, Prasad CVC, et al. J Med Chem. 1994;37:563.
26. Park BK, Kitteringham NR, O’Neill PM. Ann Rev Pharmacol Toxicol. 2000;41:443.
27. Sullivan BP, Weinreb PH, Violette SM, Luyendyk JP. Am J Pathol. 2010;177:2837.
5. Reed J. Nat Rev Drug Discovery. 2002;1:111.
6. Li J, Yuan J. Oncogene. 2008;27:6194.
7. Kersse K, Vanden Berghe T, Lippens S, Declercq W, Vandenabeele P. In: O’Brien
SD, Linton SD, eds. Design of caspase inhibitors as potential clinical agents. CRC
Press; 2008:19–58.
8. Eda H. In: O’Brien T, Linton SD, eds. Design of caspase inhibitors as potential
clinical agents. CRC Press; 2008:251–287.
9. Luedde T, Kaplowitz N, Schwabe RF. Gastroenterology. 2014;147:765.
10. Canbay A, Feldstein AE, Higuchi H, et al. Hepatology. 2003;38:1188.
11. Canbay A, Taimr P, Torok N, Higuchi H, Friedman S, Gores GJ. Lab Invest.
2003;83:655.
Please cite this article in press as: Mou J., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.04.002