Discovery of Novel and Highly Selective Inhibitors of Calpain for the Treatment of Alzheimer's Disease: 2-(3-Phenyl-1H-pyrazol-1-yl)-nicotinamides

Article abstract of DOI:10.1021/acs.jmedchem.7b00731

Calpain overactivation has been implicated in a variety of pathological disorders including ischemia/reperfusion injury, cataract formation, and neurodegenerative diseases such as Alzheimer's disease (AD). Herein we describe our efforts leading to the ide

Full text of DOI:10.1021/acs.jmedchem.7b00731

Subscriber access provided by AUBURN UNIV AUBURN
Article
Discovery of Novel and Highly Selective Inhibitors of Calpain for the
Treatment of Alzheimer´s Disease: 2-(3-Phenyl-1H-pyrazol-1-yl)nicotinamides
Andreas Kling, Katja Jantos, Helmut Mack, Wilfried Hornberger, Karla Drescher, Volker Nimmrich,
Ana Relo, Karsten Wicke, Charles W. Hutchins, Yanbin Lao, Kennan C Marsh, and Achim Moeller
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00731 • Publication Date (Web): 31 Jul 2017
Downloaded from http://pubs.acs.org on July 31, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of Novel and Highly Selective Inhibitors
of Calpain for the Treatment of Alzheimer´s
Disease: 2ꢀ(3ꢀPhenylꢀ1Hꢀpyrazolꢀ1ꢀyl)ꢀ
nicotinamides
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†,*
†
†
†
†,#
Andreas Kling, Katja Jantos, Helmut Mack, Wilfried Hornberger, Karla Drescher, Volker
†
,
⊥
†
†
‡
‡,#,
⊥
‡
Nimmrich, Ana Relo, Karsten Wicke, Charles W. Hutchins, Yanbin Lao,
Kennan Marsh,
†
,#
Achim Moeller
†
AbbVie Deutschland GmbH & Co. KG, Neuroscience Research, Knollstrasse, 67061
Ludwigshafen, Germany
AbbVie Inc., 1 North Waukegan Road, North Chicago, Illinois 60064ꢀ6125, United States of
America
‡
ABSTRACT: Calpain overactivation has been implicated in a variety of pathological disorders
including ischemia/reperfusion injury, cataract formation, and neurodegenerative diseases such
as Alzheimer´s Disease (AD). Herein we describe our efforts leading to the identification of
ketoamideꢀbased 2ꢀ(3ꢀphenylꢀ1Hꢀpyrazolꢀ1ꢀyl)nicotinamides as potent and reversible inhibitors
of calpain with high selectivity versus related cysteine protease cathepsins, other proteases and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
receptors. Broad efficacy in a set of preclinical models relevant to AD suggests that inhibition of
calpain represents an attractive approach with potential benefit for the treatment of AD.
INTRODUCTION AND RATIONALE
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The calpains are a family of proteases that are constitutively expressed in the inactive form,
1,2
some of which are activated by Calcium. Calpain 1 and 2 belong to the Ca2+ꢀactivated
calpains and together with calpains 3, 8, 9 constitute a subfamily of cysteine proteases that are
responsible for limited proteolysis of a variety of target substrates. The two isoforms, calpain 1
and 2, differ in their Ca2+ꢀsensitivities required for activation and are expressed ubiquitously,
whereas the other members of the subfamily show a tissue specific expression pattern. The
activity of calpain 1 and 2 is tightly controlled by the natural specific protein inhibitor
calpastatin, which only binds to activated calpains. Under physiological conditions calpains are
involved in processes like cytoskeletal remodeling, modulation of signal transduction pathways,
platelet activation, cell differentiation and apoptotic cell death. Overactivation of both calpain 1
and calpain 2 results in unregulated proteolysis and aberrant activation of signaling cascades,
leading to cellular damage and ultimately to cell death. Inhibition of calpain after pathological
insult has been shown to exert general cellꢀ and organꢀprotective effects. Therefore calpain has
been proposed as attractive target for several human pathological disorders such as chronic
3
,4
kidney disease, ischemia/reperfusion injury, cataract formation, and neurological diseases.
Recent evidence suggests that disruption of calcium homeostasis in neuronal cells leads to a
dysregulation of calpain. Activated calpain 1 and 2 have been found in plaques and
neurofibrillary tangles of the brain of Alzheimer´s Disease (AD) patients, and are critically
5
ꢀ7
involved in the multiple pathogenic processes of AD. Calpains modulate processes that
regulate the function of proteins key to the pathogenesis of AD, such as dynaminꢀdependent
ACS Paragon Plus Environment

Page 3 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
8ꢀ11
synaptic signaling, cytoskeletal turnover and remodeling.
In addition, calpain is part of the
1
2,13
NMDAꢀdependent excitoxic cascade
and modulates the activity of protein kinases such as
CDK5, which play a role in tau pathology, Aβ generation, and dysregulation of mitochondrial
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
function. All of these events lead to impaired synaptic function, neuroinflammation/degeneration
14,15
and finally neuronal death.
Inhibition of calpain (via overexpression of the natural inhibitor
calpastatin) was shown to restore synaptic function and prevent neurodegeneration in different
transgenic mouse models related to AD, whereas overexpression of calpain promoted amyloid
1
6ꢀ18
plaque formation and neuronal degeneration.
In summary, these findings suggest that inhibition of calpain 1 and 2 is a compelling approach
to address synaptic dysfunction and neurodegeneration, and that an appropriate calpain inhibitor
1
9ꢀ21
could have therapeutic potential for the treatment of neurodegenerative diseases like AD.
Calpain inhibitors: The majority of calpain inhibitors described so far are active site directed
and can be grouped into reversible and irreversible inhibitors. A common feature of these
inhibitors is the presence of an electrophilic center, which is required for covalent interaction
with the catalytic cysteine residue of the protease. Residues on the adjacent sides of the warhead
interact with pockets at the subsites of calpain in a manner similar to substrate binding with most
inhibitors presenting their backbones in a βꢀstrand conformation (Figure 1).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1 Protease substrates are designated according to their residues extending from the
scissile bond. Backbone side chains P3 to P2´ are supposed to interact with the corresponding
protease sub sites S3 to S2´ (adapted from lit. 24).
Over the past decades many groups have been engaged in the field of calpain inhibitors, and
2
2ꢀ25
corresponding patent applications and scientific literature have been extensively reviewed.
Although a variety of potent calpain inhibitors has been disclosed, none of these compounds
were suitable for further advancement into clinical development. Major limitations were the lack
of specificity versus other proteases and the proteasome, low cellular penetration, limited
metabolic stability and no or insufficient bioavailability. In 2003, we disclosed a series of nonꢀ
peptidic ketoamideꢀbased calpain inhibitors comprising an orthoꢀsubstituted benzoyl residue in
2
6
the P2 position. Analogues from this series were potent inhibitors of calpain 1 and 2 with Ki
values in the nanomolar range, selectivity versus the proteasome, high cell permeability, water
solubility, oral bioavailability and were shown to exert protective effects in several organ injury
2
7ꢀ29
models.
In particular, Nꢀ(1ꢀbenzylꢀ2ꢀcarbamoylꢀ2ꢀoxoꢀethyl)ꢀ2ꢀ[(E)ꢀ2ꢀ(4ꢀdiethylaminomethylphenyl)ꢀ
vinyl]benzamide 1 (A-705253) has emerged as prototypic calpain inhibitor and was used by us
and others as a tool compound to demonstrate efficacy in ADꢀrelevant animal models covering
3
0ꢀ34
synaptic dysfunction, neurodegeneration and tau hyperphosphorylation.
Furthermore, chronic
ACS Paragon Plus Environment

Page 5 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
administration of 1 in aged 3xTg mice significantly mitigated the ADꢀlike pathology and
3
5
cognitive decline. Altogether, these data strongly support our hypothesis that inhibition of
calpain might present a promising novel approach for the treatment of AD.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2 Structure and in-vitro inhibition profile of compound 1
Although compound 1 is a highly efficacious calpain inhibitor, its lack of selectivity versus the
related cysteine protease cathepsins B, K, L, S, and C prevented further advancement of the
compound (Figure 2). Analysis of the physiological and pathological functions of cysteine
protease cathepsins revealed that inhibition of cathepsin C and in particular L and S is expected
to result in immunosuppression as side effect. In contrast, inhibition of cathepsins B and K at
3
6ꢀ40
supraꢀefficacious exposure levels is not considered detrimental.
We concluded that high
selectivity versus cathepsins L, S and C is indispensable, whereas moderate selectivity versus
cathepsins B and K might be tolerable.
Therefore, our objective was to identify novel calpain inhibitors with enhanced selectivity
starting from 1 as a lead. Initial efforts were focused on exploring the P3ꢀP2 linker to address the
inherent lightꢀsensitivity observed for this chemotype, which probably originates from the
presence of the stilbene unit. Comparison of 1 with pyridyl analogue 2 revealed that introduction
of N in P2 has a positive impact on cathepsin S selectivity while only slightly affecting calpain
potency (Scheme 1). Parallel studies on a related chemotype showed that introduction of
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
piperazine or pyrrolidine as P3ꢀP2 linker in combination with P2 nicotinamide and phenyl in P3
4
1
(compounds 3 and 4) also led to calpain inhibitors with Ki values in range comparable to 1.
Based on these results we considered an alternative approach towards novel calpain inhibitors:
replacement of P3ꢀP2 vinyl by 5ꢀmember hetaryl should mimic the conformational constraint of
the stilbene unit and retain the overall orientation of phenyl towards the hydrophobic P3 pocket
of the enzyme.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 1 Starting points for optimization
a
Tested as racemates.
Herein we describe the identification and synthesis of Nꢀ(4ꢀaminoꢀ3,4ꢀdioxoꢀ1ꢀphenylbutanꢀ2ꢀ
yl)ꢀ2ꢀ(3ꢀphenylꢀ1Hꢀpyrazolꢀ1ꢀyl)nicotinamide 19 (A-933548) as a potent and selective inhibitor
of calpain. Compound 19 and analogues feature enhanced selectivity versus related cysteine
protease cathepsins, favorable microsomal stability and efficacy in cellular assays. Selected
compounds such as 4ꢀfluoro derivative Nꢀ(4ꢀaminoꢀ3,4ꢀdioxoꢀ1ꢀphenylbutanꢀ2ꢀyl)ꢀ2ꢀ(3ꢀ(4ꢀ
fluorophenyl)ꢀ1Hꢀpyrazolꢀ1ꢀyl)nicotinamide 38 (A-953227) were advanced further and
ACS Paragon Plus Environment

Page 7 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
characterized for PK properties and preclinical safety profile. In addition, compound 38 was also
shown to reproduce the in-vivo profile demonstrated with 1 in ADꢀrelated models.
RESULTS AND DISCUSSION
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis: Scheme 2 outlines the synthesis of Nꢀ(4ꢀaminoꢀ3,4ꢀdioxoꢀ1ꢀphenylbutanꢀ2ꢀyl)ꢀ2ꢀ
((3ꢀphenylꢀ1Hꢀpyrazolꢀ1ꢀyl)nicotinamide 19 as general example for the synthesis of compounds
17-45 comprising different P3ꢀP2 hetaryl linkers. Starting from ethyl 2ꢀchloronicotinate,
nucleophilic substitution using 3ꢀphenylꢀ1Hꢀpyrazole led to ethyl 2ꢀ(3ꢀphenylꢀ1Hꢀpyrazolꢀ1ꢀ
yl)nicotinate 5. Ester cleavage and coupling with 3ꢀaminoꢀ2ꢀhydroxyꢀ4ꢀphenylbutanamide
yielded the corresponding hydroxyamide intermediate Nꢀ(4ꢀaminoꢀ3ꢀhydroxyꢀ4ꢀoxoꢀ1ꢀ
phenylbutanꢀ2ꢀyl)ꢀ2ꢀ(3ꢀphenylꢀ1Hꢀpyrazolꢀ1ꢀyl)nicotinꢀ amide 7, which in the final step was
4
2
oxidized using PfitznerꢀMoffat conditions to give ketoamide 19. Final compounds 17–45 were
prepared as outlined for 19 starting from the appropriate P3ꢀP2 acids, and were provided as
racemic mixtures (unless noted otherwise).
Scheme 2 Synthetic route to Nꢀ(4ꢀaminoꢀ3,4ꢀdioxoꢀ1ꢀphenylbutanꢀ2ꢀyl)ꢀ2ꢀ((3ꢀphenylꢀ1Hꢀ
a
pyrazolꢀ1ꢀyl)nicotinamide 19 and analogues
a
Reagents and conditions: a) 3ꢀphenylꢀ1Hꢀpyrazole, K CO , KI, DMF, 130°C; b) NaOH,
2
3
MeOH/H O, rt.; c) 3ꢀaminoꢀ2ꢀhydroxyꢀ4ꢀphenylbutanamide, EDCI, HOBt, triethlyamine,
2
CH Cl , 0°C – rt.; d) EDCI, dichloroacetic acid, DMSO, rt.
2
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Substituted pyrazoles employed in the synthesis were either commercially available or
prepared as exemplified for the diethylaminoꢀ and morpholinobenzyl analogue 8 in Scheme 3.
Condensation of appropriate phenylethanꢀ2ꢀones with DMF dimethylacetal under microwave
conditions gave the corresponding 1ꢀarylꢀ3ꢀdimethylaminoꢀ2ꢀpropenꢀ1ꢀones, which upon
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
43
treatment with hydrazine hydrate were converted into the 3ꢀsubstituted pyrazoles.
a
Scheme 3 Synthetic route to 4ꢀmethylaminosubstituted 3ꢀphenylꢀ1Hꢀpyrazoles 8
a
Reagents and conditions: a) DMF dimethylacetal, reflux; b) N H x H O, MeOH, reflux.
2
4
2
2ꢀ(2ꢀPhenylthiazolꢀ4ꢀyl)ꢀ and 2ꢀ(4ꢀphenylthiazolꢀ2ꢀyl)nicotinic acid 9 and 11 were prepared
4
4
via classical Hantzsch thiazole synthesis (Scheme 4). Thus, methyl 2ꢀacetylꢀnicotinate was
converted into the αꢀbromo analogue and reacted with benzothioamide to give thiazole 9.
Analogous reaction using the commercially available ethyl 2ꢀcarbamothioylnicotinate and 2ꢀ
bromoꢀ1ꢀphenylethanꢀ1ꢀone led to 4ꢀphenylꢀsubstitued thiazole 11. Synthesis of 2ꢀ(2ꢀ
45
phenyloxazolꢀ4ꢀyl)nicotinic acid 14 was achieved via SuzukiꢀMiyaura cross coupling of ethyl
2ꢀchloronicotinate and 2ꢀphenylꢀ4ꢀ(4,4,5,5,ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)oxazole to give
13. Ester cleavage then gave the desired nicotinic acid analogues 10, 12 and 14, which were
reacted further as described for the synthesis of compound 19.
ACS Paragon Plus Environment

Page 9 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 4 Synthetic route to 2ꢀ and 4ꢀphenylthiazolylꢀsubstituted nicotinic acid derivatives 10
a
and 12 and 2ꢀphenyloxazolylꢀanalogue 14
O
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a, b
OR
9 R= OMe
10 R= OH
OMe
O
c
N
N
S
N
O
O
S
e
OR
S
OEt
11 R= OEt
c
NH2
N
N
12 R= OH
N
O
N
O
f
OR
13 R= OEt
14 R= OH
OEt
c
N
N
Cl
O
a
Reagents and conditions: a) PyBr , HBr/HOAc, rt; b) benzothioamide, DMF, rt.; c) NaOH,
MeOH/H O, rt; e) 2ꢀbromoꢀ1ꢀphenylethanꢀ1ꢀone, DMF, rt; f) 2ꢀphenylꢀ4ꢀ(4,4,5,5,ꢀtetramethylꢀ
,3,2ꢀdioxaborolanꢀ2ꢀyl)oxazole, PEPPSIꢀIPr, K CO , dioxane, 65°C.
3
2
1
2 3
The P1 building blocks commonly employed, 3ꢀaminoꢀ2ꢀhydroxyꢀ4ꢀphenylbutanamide and 3ꢀ
2
6,46
aminoꢀ2ꢀhydroxyheptanamide, were prepared according to literature procedures.
(3S)ꢀ and
(3R)ꢀ3ꢀaminoꢀ2ꢀhydroxyꢀ4ꢀphenylbutanamide with defined configuration at the P1 stereogenic
47
center were prepared following a route described by Pellacini et al. Benzyloxycarbonylꢀ
protected (S)ꢀ or (R)ꢀ2ꢀaminoꢀ3ꢀphenylpropanꢀ1ꢀol was oxidized to the aldehyde and directly
converted into the corresponding cyanohydrin (Scheme 5). Hydrolysis to amides 15a and b and
hydrogenolytic cleavage of the protecting group yielded the envisaged P1 hydroxyamides, which
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
were employed in the synthesis of compounds (S)ꢀ and (R)ꢀ19, respectively. Optical purity of the
final products was monitored via chiral HPLC.
a
Scheme 5 Preparation of (2R, 3S)ꢀ3ꢀaminoꢀ2ꢀhydroxyꢀ4ꢀphenylbutanamide 16a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: a) DMSO, oxalylchloride, CH Cl , ꢀ70°C ꢀ rt.; b) trimethylsilylꢀ
2
2
cyanide, THF, rt.; c) conc. HCl, dioxane, 10°C – rt.; d) H , PdꢀC, MeOH, rt.
2
Screening: Compounds 17 - 45 were evaluated for enzyme inhibition using kinetic
fluorescence assays. Primary activity screening was performed using natural human calpain 1
isolated from human erythrocytes. Detailed structural analysis of related cysteine protease
families revealed cathepsins B, K, L and S as closest active site homologues of calpain, hence
selectivity for these cathepsins was assessed routinely (Table 1-4). Cellular efficacy was
determined by measuring a compounds ability to prevent NMDAꢀinduced cleavage of the
calpain substrate spectrin into the corresponding spectrin degradation products (SDP 150 kDa
and SDDP 145 kDa) in rat hippocampal slice cultures. In addition, all compounds were routinely
submitted to ADME tier 1 screening comprising rat and human microsomal stability and cell
permeability.
SAR: As outlined above, we started exploring various phenylꢀsubstituted hetaryl linkers in P3ꢀ
P2 position in combination with P2 pyridyl (Table 1). Whereas 1,4ꢀ and 1,3ꢀsubstituted
pyrazoles 18 and 19, respectively, displayed enhanced calpain potency, 1,3ꢀsubstituted imidazole
17 and 2,4ꢀsubstituted thiazole 20 showed inhibition of calpain in a range comparable to 1. The
corresponding oxazole analogue 21 and the isomeric thiazole 22 showed 4ꢀ5 fold reduced
ACS Paragon Plus Environment

Page 11 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
potency. In terms of selectivity only 1,3ꢀsubstituted pyrazole 19 and 2,4ꢀsubstituted thiazole 20
had the desired selectivity profile with high selectivity versus cathepsins S and L and moderate
versus cathepsins B and K.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Inhibition of calpain 1 and cathepsin selectivity for compounds 17 – 24
ꢀꢀꢀ selectivity ratios Cat/Cal ꢀꢀꢀ
Cal 1
compd linker
Ki [nM]
CatB CatK CatL
CatS
59
1
ꢀ
56
45
< 1
< 1
< 0.1
< 1
2.5
23
17
54
18
19
20
21
< 1
26
39
11
10
1
12
14
18
21
11
20
11
176
139
29
>500
310
127
105
38
2
1
2
177
234
2
37
2
3
4
ꢀ
43
82
7
1
38
82
2
ꢀ
23
< 1
116
186
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Investigation of the corresponding P2 phenyl analogues 23 and 24 confirmed the positive
impact of pyridyl in P2 position. Both compounds showed diminished cathepsin selectivitity, in
particular for cathepsin K. However, in the case of thiazole 24 the observed changes in
selectivity for cathepsins B, L and S are mainly caused by the reduction in calpain potency, since
the absolute Ki values for these cathepsins are similar between 24 and 20. For selected examples
we examined the impact of P1 variation and prepared the respective Nꢀ(1ꢀaminoꢀ1,2ꢀ
dioxoheptanꢀ3ꢀyl)nicotinamides comprising nꢀbutyl as P1 side chain (Table 2). Compounds 25-
27 retained inhibition of calpain in a range comparable to the P1 benzyl analogues 17-19,
whereas thiazole 28 displayed a 3 fold reduction in potency. However, cathepsin selectivity was
found to be significantly reduced for all of these analogues and not in an acceptable range.
Further exploration of the SAR for P1 revealed that P1 benzyl remained optimal in terms of
selectivity profile (data not shown).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 13 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 2. Inhibition of calpain 1 and cathepsin selectivity for compounds 25 - 28
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ꢀꢀꢀ selectivity ratios Cat/Cal ꢀꢀꢀ
Cal 1
compd linker
Ki [nM] CatB CatK
CatL
CatS
2
5
6
38
20
< 1
< 1
< 1
< 1
22
22
2
4
2
2
7
8
32
1
2
2
1
25
13
61
27
2
122
To better understand the structural basis of the SAR for inhibitors 17-28 a homology model of
human calpain 1 was developed using the crystal structure of leupeptin in rat calpain as
48
template. Superposition placed the ketoamideꢀbased inhibitors in the binding cleft in a pose
comparable to leupeptin, with the P2 pyridine fitting well into the S2 site and the phenylꢀ
substituted 5ꢀmember heterocycle into S3. For the selected human cathepsins the compounds
could be modeled into similar poses. Potency and selectivity of the various ligands investigated
can be explained primarily by the differences between calpain and cathepsins in the S3 and the
S1 region, this is depicted using compound 19 as representative (Figure 3).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
8,49
Figure 3 Model of compound 19 in the active site of human calpain 1 and cathepsin K.
Left:
Surface model of calpain 1 with 19 in the active cleft; shown is the postulated waterꢀmediated
interaction between S3 Ser209 and P3 pyrazole N. Center and Right: Closeꢀup on S1: in calpain
P1 phenyl aligns in parallel to Gly207/Ser208 amide (center), whereas in cathepsin K (right) the
corresponding amide carbonyl is rotated about 90° towards the phenyl residue.
In general, in the distant part of the S3 site the secondary structure of calpain differs from the
cathepsins: whereas calpain 1 has an open pocket in this region, the cathepsin backbone forms a
βꢀstrand (see Supporting Information). Therefore substituents at the P3 hetaryl residue are
expected to show an impact on overall selectivity. In addition, the pyrazole N in compound 19 is
able to form a waterꢀmediated hydrogen bond with Ser209 at the S3 site of calpain, thereby
contributing to the overall binding affinity of this molecule (Figure 3 left). This hypothesis is
supported by the fact that all hetaryl residues with N in this position show selectivity. In contrast,
the cathepsins studied have larger residues in this position (Tyr for cathepsins B and K, Leu for
cathepsin L, and Phe for cathepsin S), which are not able to form stabilizing interactions.
In addition, at the S1 site in calpain 1 the P1 phenyl residue in 19 aligns parallel with the
backbone amide at Gly207 in an amideꢀπ interaction, and thus fits well into this sub pocket
ACS Paragon Plus Environment

Page 15 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
Figure 3 center). The cathepsins show a different folding of the backbone in this region with
the corresponding amide carbonyl rotated towards the P1 phenyl, which leads to an unfavorable
interaction and thus contributes to selectivity. (Figure 3 right, cathepsin K depicted as example).
When changing P1 from benzyl to nꢀbutyl, the alkyl side chain is able to rotate torsions to
interact well with both the calpain or the cathepsin backbone amide, which explains the loss of
selectivity for this P1 modification, while calpain potency is retained.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Based on the data obtained we focused our efforts on pyrazole 19 and moved on to further
explore the SAR for P3. Table 3 summarizes the results obtained. As expected, omitting P3
substitution led to significant reduction in calpain inhibition (compound 29). We then evaluated
whether nonꢀaryl substituents in P3 would be tolerated, since this modification was also
3
50
supposed to enhance the fraction sp , which for compound 19 is not in an optimal range. In
terms of calpain inhibition, P3 benzyl derivative 35 exhibited the highest potency, followed by
cyclohexyl and tetrahydropyranyl analogues 30 and 33, respectively. The latter two derivatives
3
also showed significantly enhanced fraction sp . However, none of the analogues 29-36 could
compete with compound 19 in terms of calpain potency and selectivity profile.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 3. Inhibition of calpain 1 and cathepsin selectivity for compounds 29 – 36
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ
ꢀꢀ selectivity ratios Cat/Cal ꢀꢀ
Cal 1 Ki
[nM]
3
compd
R
CatB
CatK
CatL
CatS
fsp
19
ꢀ
18
26
1
21
3
176
20
>500 0.08
2
9
0
H
164
21
92
0.11
0.32
3
56
81
94
9
10
24
7
33
10
34
31
16
4
56
50
0.26
0.23
32
33
34
48
14
2
13
1
56
2
147
89
0.29
0.32
146
3
5
6
38
9
2
9
2
50
45
75
10
0.12
0.26
3
ꢀ
101
Subsequently, we explored substitution at P3 phenyl, with Table 4 presenting the results for
selected examples. Introduction of chloro, methoxy and trifluoromethyl in 4ꢀposition (37, 39 and
40) led to diminished calpain inhibition, whereas 4ꢀfluoro analogue 38 and 3ꢀtrifluoromethyl
derivative 41 retained calpain potency and the overall selectivity profile, although with varied
ACS Paragon Plus Environment

Page 17 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
selectivities for the individual cathepsins. As demonstrated by indazole 44, annelation of P3
phenyl also resulted in reduced selectivity, most notably in the case of cathepsin S. In analogy to
the structure of compound 1 we also explored the impact of additional basic moieties in P3.
Benzyldiethylamine 42 and its respective morpholine analogue 43 were both found to act as
nonselective inhibitors of calpain and cathepsins B, K and L with moderate to favorable
selectivity versus cathepsin S, respectively. Remarkably, chromene analogue 45 represented the
most potent calpain inhibitor from this series and also featured the highest cathepsin selectivity.
However, the extremely low solubility of 45 in aqueous and nonꢀaqueous vehicles prevented us
from further advancing this compound.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Selected derivatives were submitted to screening for cellular efficacy. As depicted in Table 4
compounds 19 and 38 showed the highest efficacy in a functional cellular assay using prevention
of NMDAꢀinduced intracellular spectrin cleavage as readꢀout. Both compounds have IC values
5
0
in a range comparable to 1 (rat HSC IC = 461 nM total and 36 nM free).
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 4. Inhibition of calpain 1 and cathepsin selectivity for compounds 37 – 45
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀꢀꢀꢀ selectivity ratios Cat/Cal ꢀꢀꢀ
a
Cal 1
Ki [nM] CatB
rat HSC
IC [nM] total/free
b
compd
R
CatK
CatL
CatS
5
0
1
3
3
3
4
9
7
8
9
0
H
18
88
34
83
72
40
100
41
26
6
21
9
176
89
>500
>300
>300
>500
87
971/39
nd
4ꢀCl
4ꢀF
10
11
11
17
<1
18
17
22
28
<1
127
81
591/41
nd
4ꢀOMe
4ꢀCF
210
>300
3
nd
3
3
41
42
3ꢀCF
>500
69
5179/155
>10.000/ꢀ
nd
4
3
<1
<1
5
240
4
4
5
ꢀ
ꢀ
46
8
6
6
124
278
51
3200/288
2020/65
4
29
47
>500
a
Cellular efficacy was determined by measuring prevention of NMDAꢀinduced cleavage of
b
spectrin in rat hippocampal slice cultures. IC values relating to total and free levels; free levels
5
0
calculated based on fraction unbound. nd= not determined
ACS Paragon Plus Environment

Page 19 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
P1-Configuration: In general, racemization is an inherent feature of ketoamideꢀbased
inhibitors due to the electrophilic group adjacent to the chiral enter. However, although calpain
inhibitors have been studied extensively, only Harbeson et al. have reported on studies regarding
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
46
P1 configuration and the epimerization rates of peptidic calpain inhibitors. In order to
determine the potency of the individual enantiomers and to investigate the rate of racemization,
we prepared the distinct enantiomers (S)ꢀ and (R)ꢀ19. Determination of activity in-vitro revealed
a >500 fold difference in calpain inhibition with the P1 Sꢀconfiguration as the more active
5
1
enantiomer (K Cal 1 (S)ꢀ19= 10 nM). As expected, the distinct enantiomers were
i
configurationally stable in solid form, but underwent pHꢀdependent racemization in
.
solution. Rapid racemization was observed under physiological conditions at 37°C (assay buffer
pH 7.4 or hirudinꢀanticoagulated rat plasma; calpain inhibition used as readꢀout), which was
completed within two hours (see Supporting information). The individual enantiomers showed a
halfꢀlife of ~30 min in buffer and 19 min in rat plasma, respectively.
Covalent reversible inhibition: Taking compound 19 as general representative, we studied
the dissociation from calpain 1. Assessment of inhibitor offꢀkinetics by recovery of calpain 1
enzymatic activity after preꢀincubation with the inhibitor revealed that that the compound acts as
reversible covalent inhibitor of calpain with a halfꢀlife t = 41 min and a dissociation rate k =
1
/2
ꢀ1
2
.85 x 10ꢀ4 s (see Supporting Information).
Addressing potential unspecific reactivity of ketoamide warhead: Overall, we did not find
any evidence for unspecific interaction of ketoamideꢀbased calpain inhibitors with cysteine
thiols. Inhibition of calpain in-vitro was not affected by the presence of glutathione up to 10 mM
concentrations. In addition, none of the compounds showed nonspecific reactivity towards the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
human La antigen in ALARM NMR, which is a standard inꢀhouse assay used to identify thiol
5
2
reactive compounds.
Selectivity profile: Compounds 19 and 38 did not show cross reactivities for other GPCRs, ion
channels and transporters up to 10µM in the CEREP profile (except for the peripheral BZD
receptor), and displayed excellent selectivity versus a comprehensive set of different enzymes
and proteases including caspases, the proteasome and cathepsins C and H (MDS
EnyzmeProfilingScreen at 10µM concentration). Furthermore, none of the compounds displayed
any liability against hERG, AMES and micronucleus tests, and rat cardiovascular safety.
Physchemical properties: In terms of physicochemical properties the novel calpain inhibitors
are in a chemical space typical for protease inhibitors. Ketoamides 19 and 38 show clogP values
of 0.97 and 1.12, respectively, a TPSA of 120, and pass the Lipinski Ro5 and the 3/75 rule.
ADME and PK profile: The in-vitro ADME profile for compounds 19 and 38 was
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ6
characterized by high permeability (P > 30 x 10 cm/s) in Cacoꢀ2 cells and an efflux ratio of
app
approximately 2. Low aqueous solubility in combination with high permeability puts both
compounds in the BCS II class. Plasma protein binding is moderate in both rat and human with
approximately 93% bound. Both compounds 19 and 38 had good liver microsomal stability
across rat, dog, monkey and human (mClint< 23 µL/min/mg). Hepatocyte stability of 38 was
moderate to good in human, monkey, rat and dog (hepClint, µL/min/million cells: 19.4, 13.9,
3
8.9, and 1.3, respectively). Further metabolite characterization studies using [ H]ꢀlabelled 38
revealed significant species differences in its metabolism. In the blood and hepatocytes of human
and monkey, 38 is predominantly metabolized by cytosolic carbonyl reductases reducing the
ketoamide to the inactive hydroxyamide. In contrast, in rat, 38 is rapidly hydrolyzed in blood to
the corresponding keto acid and undergoes oxidation and glucuronidation in hepatocytes. In the
ACS Paragon Plus Environment

Page 21 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
dog, minimal metabolism was observed in vitro. Both compounds 19 and 38 demonstrated low
potential to elicit clinically significant CYPꢀmediated drugꢀdrug interactions (DDIs): no direct or
timeꢀdependent inhibition of major CYPs and no CYP3A4 mRNA induction was observed. In
addition, nonꢀCYPꢀmediated carbonyl reduction is the primary metabolism pathway.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The pharmacokinetic behaviour of 38 was evaluated in Sprague Dawley rats, Beagle dogs, and
Cynomolgus monkeys (Table 5). Pharmacokinetics were characterized by high mean plasma
clearance (CL ) in both rat and monkey (CL = ~ 2 L/hr•kg), but lower in dog (CL = 0.32
p
p
p
L/hr•kg), consistent with in vitro hepatocyte stability data. Mean steady state volumes of
distribution (V ) were moderate in all species: 0.65 L/kg in dog and 1.8 L/kg in rat and monkey.
ss
The mean elimination halfꢀlives range from 1.1 hours in monkey to 2.9 hours in dog. Oral bioꢀ
availability values were moderate to good in dog (F = 30%) and rat (F = 68%), but low in
monkey (F =4.5%). Low oral bioavailability and high Cl value in monkey are consistent with
p
the ubiquitous distribution of carbonyl reductase activities, the primary metabolizing enzymes
5
3
for 38. Compound 38 partitioned into the brain, with brain to plasma drug concentration ratio
of ~ 0.2 in rat.
Table 5. Pharmacokinetics of 38 in preclincial species following a single IV or PO dose
IV
PO
Tmax
species dose t ° AUC0ꢀinf
CLp
Vss
1.8
dose t ° Cmax
AUC0ꢀinf
F
1
/2
1/2
rat
2
1.7
0.95
0.13)
2.1
(0.3)
10
2.6 1.28
2.2 0.53
6.3 0.02
0.4
(0.1)
3.22
68
(
(
(
(0.46)
(9.8)
dog
0.5 2.9
1.56
0.25)
0.3
(0.05)
0.6
5
1
0.3
(0.1)
0.94
30
(0.50)
(15.8)
monkey
1
1.1
0.52
0.08)
2.0
(0.3)
1.8
10
13.3
(10.1)
0.23
4.5
(0.032)
(0.6)
Data provided as mean (standard deviation); ° harmonic mean; Units: Dose (mg/kg); t1/2 (hr);
Vss (L/kg); AUC0ꢀinf (µg•hr/mL); CL (L/hr•kg); Cmax (µg/mL); Tmax (hr); F (%); IV:
p
intravenous; PO: oral.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Preclinical efficacy: Based on the overall profile compound 38 was selected for extended
characterization in a set of AD relevant models. First, we examined the effect of 38 in a panel of
in-vitro models of neurodegeneration. As already mentioned, compound 38 was shown to
prevent neuronal death in rat hippocampal slice cultures via inhibition of NMDAꢀinduced
spectrin degradation with an IC value of 41 nM (Table 4). In the same system 38 blocked the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
calpainꢀmediated proteolysis of intracellular tau with comparable efficacy (IC of 46 nM free
5
0
5
4
concentration). Application of Aβ globulomer, a specific Aβ
oligomer preparation described
ꢀ42
1
5
5
as neuropathological factor in AD, strongly reduces synaptic transmission in the CA1 area in
rat hippocampal slice cultures, and is thus considered as an in-vitro model reflecting synaptic
degeneration in AD. Coꢀapplication of compound 38 (10 nM) completely prevented Aβ
globulomerꢀinduced synaptic deficits and restored transmission to a level not significantly
different from control (Figure 5).
5
control
4
3
2
1
0
Aβ globulomer
Aβ globulomer
+ compound 38
1
2
3
4
5
6
stimulation intensity (V)
Figure 5 Compound 38 at 10 nM concetration prevents Aβ oligomerꢀinduced deficits in
synaptic transmission in rat. Data points are shown as mean +/ꢀ SEM. The strength of synaptic
transmission is illustrated by the input/output relation in hippocampal slice cultures after
stimulation of the Schaffer collateral (p=0.016 when compared to untreated globulomer group;
p=0.851 when compared to control).
ACS Paragon Plus Environment

Page 23 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Since degeneration/dysfunction of cholinergic neurons is closely associated with cognitive
deficits in AD, an increase of cholinergic function is widely thought to improve cognitive
performance. Inhibitors of acetylcholineꢀesterase (AChEIs) like donepezil increase extracellular
concentration of acetylcholine, thereby i) improving cognition and ii) modifying sleep EEG
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
6
pattern. We therefore investigated the in-vivo effects of 38 on neurotransmission and synaptic
function in a microdialysis study. Acute, subcutaneous administration of 38 resulted in a doseꢀ
dependent increase in acetylcholine levels in the medial prefrontal cortex reaching significance
after administration of 3 and 10 mg/kg, respectively (Figure 6), without affecting glutamate and
dopamine levels (data not shown).
Figure 6 Compound 38 doseꢀdependently increases microdialysate acetylcholine (ACh) levels
in the medial prefrontal cortex. All data are means ± S.E.M. in arbitrary units based on percent
change of baseline levels. Oneꢀway ANOVA revealed a significant treatment effect for ACh
levels [F(3,41) = 5.82, p < 0.0023]. Subsequent Dunnett´s multiple comparisons demonstrated
an increase of ACh levels following administration of 38 at 3 and 10 mg/kg, s.c. (*p < 0.05; **p
< 0.01).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Furthermore, the effects of compound 38 on rat rapid eye movement (REM) sleep pattern were
investigated by recording a sleep electroencephalogram (EEG). When given acutely (1ꢀ10
mg/kg, p.o.), 38 specifically affected the REM sleep pattern of Fisher rats in a doseꢀrelated
manner without affecting total sleep (Table 6). This is similar to what has been observed with
AChEIs such as donepezil (see Supporting Information) and suggests that this effect is a
functional consequence of increased extracellular ACh.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 6. Summary of doseꢀdependent effects of 38 on rat sleep
Dose (mg/kg)
REM time
1
3
10
105 ± 5 131 ± 6** 125 ± 8**
Latency to REM
Total sleep time
100 ± 8 99 ± 10
98 ± 3 105 ± 2
80 ± 7**
102 ± 2
All parameters are expressed as % of vehicle controls (mean ± SEM) (** p < 0.01 from
control).
To show that the improvement of cholinergic function translates into enhanced behavioral
performance we also explored 38 in a model of cognitive performance in rats. Compound 38
significantly attenuated cognitive impairment induced by the muscarinic antagonist scopolamine
in a rat 24 hour passive avoidance paradigm. Administration of scopolamine impaired
acquisition of this task, while application of compound 38 (1 mg/kg, s.c. prior to scopolamine)
resulted in a significant attenuation of the scopolamineꢀinduced deficits. The dose of 3 mg/kg
showed a trend without reaching statistical significance (Figure 7).
ACS Paragon Plus Environment

Page 25 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. Effect of compound 38 on scopolamineꢀinduced deficits in rats (doses in mg/kg;
n=14ꢀ47/group). Veh. (vehicle); Scop. (scopolamine, 0.1 mg/kg, s.c.); *** significantly (p <
0.001) different from vehicle controls; # (p < 0.05) significantly different from scopolamine
alone.
Efficacious doses for 38 in-vivo were found to be consistent across models with free exposures
ranging from approximately 6ꢀ25 ng/mL in plasma and 0.2ꢀ2.1 ng/g in brain.
In summary, compound 38 showed robust efficacy in various in-vitro and in-vivo models
related to AD. Inhibition of calpain using selective inhibitor 38 resulted in prevention of
neurodegeneration and improved neurotransmission, thereby attenuating behavioral deficits.
Altogether, these results suggest that novel calpain inhibitors such as 38 have a potential for the
treatment of neurodegenerative diseases such as AD.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CONCLUSION
Our goal was to identify novel inhibitors of calpain with enhanced selectivity and favorable PK
properties, in particular brain penetration, for the potential treatment of AD. Starting from
compound 1 as lead we investigated phenylꢀsubstituted 5ꢀmember hetaryls in P3ꢀP2 position in
combination with nicotinamide as P2 core. These efforts led to the identification of Nꢀ(4ꢀaminoꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
,4ꢀdioxoꢀ1ꢀphenylbutanꢀ2ꢀyl)ꢀ2ꢀ((3ꢀphenylꢀ1Hꢀpyrazolꢀ1ꢀyl)nicotinamides as novel series of
calpain inhibitors with potencies in the nanomolar range and medium to excellent selectivity
versus closely related cysteine protease cathepsins. In particular, compounds 19 and 38 combine
favorable cathepsin selectivity, microsomal stability, high cell permeability, and functional
efficacy in cellular assays. Potency and enhanced selectivity observed in the 1,3ꢀpyrazole series
can be explained via modeling comparing the active sites of calpain 1 and the most relevant
cathepsins, respectively. Taking 19 as representative example, we confirmed that potency is
dependent on the P1 configuration with the Sꢀenantiomer being more active. Due to the adjacent
ketoamide the individual enantiomers undergo rapid racemization under in-vivo conditions. In
addition, 19 was shown to act as covalent reversible inhibitor of calpain, and despite the presence
of the ketoamide warhead, no evidence for unspecific reactivity was found. Both compounds
were shown to have an excellent profile without liabilities in standard enzyme and receptor
panels, DDI, AMES and micronucleus test.
The overall PK profile and brain penetration enabled further advancement of compound 38. In
preclinical models relevant to AD 38 was shown to prevent NMDAꢀinduced neurodegeneration
and Aβꢀinduced synaptic dysfunction. In addition, 38 was shown to enhance neurotransmission
which translates into improved cognitive performance. Finally, extended studies in rat and dog
such as cardiovascular safety and 14ꢀday tolerability demonstrated an excellent safety
ACS Paragon Plus Environment

Page 27 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
pharmacology profile. Based on the overall favorable results compound 38 was progressed into
5
7
early clinical development, results will be disclosed in a subsequent publication.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
Chemistry: All commercially available chemicals and solvents were used without further
purification. In general, reaction mixtures were magnetically stirred at the respective temperature
under nitrogen atmosphere. Organic solutions were concentrated under reduced pressure using a
Buechi rotovap. Reactions under microwave irradiation conditions were carried out in a Biotage
Initiator EXP instrument. Normal phase chromatography was performed on an ISCO
CombiFlash Companion MPLC system using RediSep preꢀpacked columns with silica, reverse
phase chromatography was performed using Chromabond C18 cartridges. Reactions were
monitored by thin layer chromatography using HPTLC Silicagel 60 F254 plates from Merck
KGaA, and were visualized using 254 nm ultraviolet light and/or exposure to silica gel
impregnated with iodine. Optical rotations were obtained on a PerkinElmer polarimeter 341. All
1
new compounds gave satisfactory HNMR and LC/MS. On the basis of LCMS, all final
compounds were >95% pure unless otherwise noted. NMR spectra were obtained on a Bruker
Avance I 400 MHz, Avance III 500 MHz or 600 MHz NMR spectrometer using residual signal
of deuterated NMR solvent as internal reference; chemical shifts for protons are reported in a
parts per million scale downfield from tetramethylsilane. Analytical LCMS data were obtained
using an Agilent 1100 series HPLC system with DAD and SQ mass spectrometer with ESI in
positive mode and a scan range of 100ꢀ700 amu. Samples were run on a Chromasil 80 ODSꢀ7pH
column 4µm, 40x2mm; gradient elution 5ꢀ100% B over 10 minutes; solvent A: H2O/0.1% TFA;
solvent B: acetonitrile/0.1% TFA; flow rate: 0.5 mL/min; temperature: 60°C. Chiral compounds
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
were analyzed by chiral HPLC using an Agilent 1100 HPLC with samples run on Chiralpak ADꢀ
H 4.6mmID x 250 mm; 5µ; with nꢀhexane/EtOH/MeOH 20/40/40 as mobile phase; flow rate: 0.6
/
min and detection at 300nm wavelength. Chemical names were generated using ChemDraw
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Professional 15.0 (PerkinꢀElmer Informatics). Modeling studies were performed using InsightII
from Accelrys/MSI.
N-(4-(1H-Pyrazol-3-yl)benzyl)-N-ethylethanamine (8a). A mixture of 1ꢀ(4ꢀ((diethylamino)ꢀ
methyl)phenyl)ethanꢀ1ꢀone (11.9 g, 57.96 mmol) and DMF dimethylacetal (7.6 g, 57.96 mmol)
was heated in a microwave to 200°C for 20 minutes. After cooling to rt first EtOH (200 mL) and
then hydrazine hydrate (10.3 g, 316.1 mmol ꢀ dropwise) were added and the reaction heated to
reflux for 2 hours. Afterwards the mixture was concentrated, HCl (1 M in water) added to adjust
pH~1 and extracted twice with MTBE. The aqueous layer was basified with NaOH (1 M in
water), the product extracted into MTBE (x2), and the combined organic extracts were dried
(
Na SO ) and concentrated to give a pale yellow oil (13.3 g, 91.7%) which was used in the next
2 4
1
reaction without further purification. H NMR (400 MHz, DMSOꢀd – mixture of tautomers) δ
6
13.22 and 12.80 (s, 1H), 7.80–7.60 (m, 3H), 7.32 (d, J = 7.9 Hz, 2H), 6.65 (d, J = 8.6 Hz, 1H),
3.52 (s, 2H), 2.50 – 2.36 (q, overlap with DMSO, 2 H), 0.98 (t, J = 7.1 Hz, 3H). ESI MS (m/z):
+
calcd for C H N , 229.2; found 230.1 [M + H ].
1
4
19
3
4-(4-(1H-Pyrazol-3-yl)benzyl)morpholine (8b). 1ꢀ(4ꢀ(Morpholinomethyl)phenyl)ethanꢀ1ꢀone
(10.5g, 47.9 mmol) was converted into the pyrazole in analogy to the preparation of 8a giving
1
the title compound as pale yellow oil (12.3 g, 97%). H NMR (400 MHz, DMSOꢀd , mixture of
6
tautomers) δ 13.24 and 12.82 (s, 1H), 7.80–7.58 (m, 3H), 7.323 (d, J = 7.8 Hz, 2H), 6.66 (d, J =
7.5 Hz, 1H), 3.57 (m, 4H), 3.46 (s, 2H), 2.36 (m, 4H). ESI MS (m/z): calcd for C H N O,
14 17 3
+
2
43.1; found 244.1 [M + H ].
ACS Paragon Plus Environment

Page 29 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Methyl 2-(2-phenylthiazol-4-yl)nicotinate (9). PyBr (4.5 g, 14.07 mmol) – suspended in
3
acetic acid (15 mL) – was added to a solution of methyl 2ꢀacetylnicotinate (2.4 g, 13.39 mmol) in
hydrogen bromide (33% solution in acetic acid, 6 mL), and stirred for 3 hours at rt. The mixture
was filtered, the precipitate obtained washed with nꢀpentane and then dried to give 3.45 g
(68.4%) of the bromide as offꢀwhite solid, which was used in the next step without further
purification. The mixture of 2ꢀ(2ꢀbromoacetyl)ꢀ3ꢀ(methoxycarbonyl)pyridinium bromide (1.15 g,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
.05 mmol) and benzothioamide (0.5 g, 3.64 mmol) in DMF (12 mL) was stirred for 1 hr at rt
and then concentrated to give a brownish oil. Purification by chromatography (silica gel,
1
dichloromethane/MeOH as eluent) gave thiazole 9 as a pale yellow oil (0.8 g, 88%). H NMR
(
400 MHz, DMSOꢀd ) δ 8.77 (dd, J = 4.9, 1.7 Hz, 1H), 8.37 (s, 1H), 8.02 (dd, J = 7.8, 1.7 Hz,
6
1H), 7.99–7.88 (m, 2H), 7.62–7.37 (m, 4H), 3.75 (s, 3H). ESI MS (m/z): calcd for C H N O S,
1
6
12
2
2
+
2
96.1; found 297.1 [M + H ].
2-(2-Phenylthiazol-4-yl)nicotinic acid (10). NaOH (2 M in water, 4 mL) was added dropwise
to a solution of methyl 2ꢀ(2ꢀphenylthiazolꢀ4ꢀyl)nicotinate 9 (0.62 g, 2.1 mmol) in MeOH (15
mL), and the mixture heated for 1 hr to 70°C. The mixture then was concentrated, the obtained
residue dissolved in water, acidified to pH5 using 2 M HCl (4 mL), the precipitate collected,
1
washed twice with water and dried to give 10 as offꢀwhite solid (0.35 g, 59.3%). H NMR (400
MHz, DMSOꢀd ) δ 13.14 (s, 1H), 8.72 (d, J = 4.6 Hz, 1H), 8.28 (s, 1H), 8.04–7.95 (m, 3H),
6
+
7.58–7.45 (m, 4H). ESI MS (m/z): calcd for C H N O S, 282.1; found 283.0 [M + H ].
1
5
10
2
2
Ethyl 2-(4-phenylthiazol-2-yl)nicotinate (11). Reaction of ethyl 2ꢀcarbamoylnicotinate (2.8
g, 10.65 mmol) and 2ꢀbromoꢀ1ꢀphenlyethanone (2.3 g, 11.56 mmol) in DMF (35 mL) in analogy
1
to 9 and purification by chromatography gave thiazole 11 as offꢀwhite solid (2.79 g, 84%). H
NMR (400 MHz, DMSOꢀd ) δ 8.77 (dd, J = 4.8, 1.6 Hz, 1H), 8.33 (s, 1H), 8.07 (dd, J = 7.8, 1.7
6
ACS Paragon Plus Environment