Design of Potent and Druglike Nonphenolic Inhibitors for Catechol O-Methyltransferase Derived from a Fragment Screening Approach Targeting the S-Adenosyl- l -methionine Pocket

Article abstract of DOI:10.1021/acs.jmedchem.6b00927

A fragment screening approach designed to target specifically the S-adenosyl-l-methionine pocket of catechol O-methyl transferase allowed the identification of structurally related fragments of high ligand efficiency and with activity on the described orthogonal assays. By use of a reliable enzymatic assay together with X-ray crystallography as guidance, a series of fragment modifications revealed an SAR and, after several expansions, potent lead compounds could be obtained. For the first time nonphenolic and small low nanomolar potent, SAM competitive COMT inhibitors are reported. These compounds represent a novel series of potent COMT inhibitors that might be further optimized to new drugs useful for the treatment of Parkinson's disease, as adjuncts in levodopa based therapy, or for the treatment of schizophrenia.

Full text of DOI:10.1021/acs.jmedchem.6b00927

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Design of Potent and Drug Like Non Phenolic inhibitors for
Catechol O-Methyltransferase Derived from a Fragment Screening
Approach Targeting the S-Adenosyl-L-Methionine Pocket
Christian D. Lerner, Roland Jakob-Roetne, Bernd Buettelmann, Andreas
Ehler, Markus G. Rudolph, and Rosa Maria Rodriguez Sarmiento
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00927 • Publication Date (Web): 29 Sep 2016
Downloaded from http://pubs.acs.org on September 30, 2016
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 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design of Potent and Drug Like Non Phenolic
Inhibitors for Catechol OꢀMethyltransferase Derived
from a Fragment Screening Approach Targeting the
SꢀAdenosylꢀLꢀMethionine Pocket
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
Christian Lerner, Roland Jakob-Roetne, Bernd Buettelmann, Andreas Ehler, Markus Rudolph,
Rosa María Rodríguez Sarmiento*
Pharmaceutical Research and Early Development (pRED), F. HoffmannꢀLa Roche AG, Roche
Innovation Center Basel, Grenzacherstrasse 124, 4070 Basel, Switzerland
ABSTRACT
A fragment screening approach designed to target specifically the SꢀadenosylꢀLꢀmethionine
pocket of catechol Oꢀmethyl transferase, allowed the identification of structurally related
fragments of high ligand efficiency and with activity on the described orthogonal assays. Using a
reliable enzymatic assay together with Xꢀray crystallography as guidance, a series of fragment
modifications revealed an SAR and, after several expansions, potent lead compounds could be
obtained. For the first time non phenolic and small low nanomolar potent, SAM competitive
COMT inhibitors are reported. These compounds represent a novel series of potent COMT
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 47
1
2
3
4
5
6
inhibitors that might be further optimized to new drugs useful for the treatment of Parkinson’s
disease, as adjuncts in levodopa based therapy, or for the treatment of schizophrenia.
7
8
9
INTRODUCTION
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 enzyme catechol Oꢀmethyltransferase (COMT) catalyzes the Mg2+ dependent methyl
transfer from its cofactor SꢀadenosylꢀLꢀmethionine (SAM), to one of the hydroxyl groups of
endogenous catechols and neurotransmitters such as dopamine and noradrenaline, resulting in
termination of their biological activity. In the brain, COMTꢀdependent dopamine degradation is
of high relevance in regions with low expression of the presynaptic dopamine transporter (DAT),
in particular in the prefrontal cortex. For this reason COMT inhibition offers a unique
opportunity to influence pathologies characterized by low dopamine levels in the prefrontal area
such as cognitive impairment associated with schizophrenia.1
Most known COMT inhibitors are nitroꢀcatechol based compounds, such as tolcapone
and entacapone (Figure 1) which are successfully used as adjuncts to the LꢀDopa/based treatment
of Parkinson’s Disease.2 They are highly metabolized and mainly peripherally acting, protecting
LꢀDopa from methylation and deactivation mostly in the liver.3 While entacapone is almost
exclusively peripherally acting, there is evidence for central effects with tolcapone, despite the
limited brain penetration (brain to plasma ratio up to 0.01).4
Bisubstrate and hydroxyꢀpyridone based COMT inhibitors (Figure 1) have also been
described.5,6 They do not require the presence of a nitro group, however like the catechols, they
have chelating motives of phenolic nature binding to the Mg2+ site. Such phenolic compounds
are prone to high metabolic clearance, notably glucuronidation and sulfatation causing low oral
bioavailability.3,7 In addition, their acidity and high polarity make it challenging to develop them
ACS Paragon Plus Environment

Page 3 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
as centrally acting drugs. Accordingly, to date the focus has been on peripheral COMT
inhibition,8 while evidence for an important role of COMT in schizophrenia is accumulating.
7
8
9
We want to report the identification of structurally differentiated COMT inhibitors
targeting exclusively the SAM pocket instead of the Mg2+ site with its strict limitations for
binding.9 Ideally, the adenosine replacing motive should be of low molecular weight, with an
acceptable polarity reflected in an adequate number of Hꢀbond donors (typically HBD <3), with
less groups able to function as HBA (<7) and with a suitable lipophilicity for CNS penetration
(clogP, best between 1ꢀ4),10 in contrast to previous examples targeting the SAM pocket.11
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
Tolcapone
Tasmar
Entacapone
Comtan
Roche/Valeant
Orion/Novartis
++
Bisubstrate inhibitor
H-bond interactions for bisubstrates inhibitors
Non nitrocathecol inhibitor (hydroxy-pyridone)
Figure 1. A) Reported COMT inhibitors. B) Interactions for bisubstrate inhibitors binding to
SAM and Mg2+ pockets.5a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 47
1
2
3
4
5
6
7
8
9
In the recent past, fragmentꢀbased drug discovery was successfully applied to the
generation of hits or chemical tools for drug targets. A critical challenge is to further elaborate
the fragment hits to potent leads with balanced properties12 with particular examples such as
Zelboraf, which could be further developed into a marketed drug.13 Fragment hits typically show
very low affinity to the targets, so sensitivity of the screening assay and confirmation of binding
by orthogonal methods is required.14
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
RESULTS AND DISCUSSION
We performed a fragment screening for COMT targeting the SAM pocket. We used a
library consisting of 6000 compounds with molecular weights below 250 Da, up to three
hydrogen bond donors, up to three hydrogen bond acceptors, and calculated logP (clogP) of ≤3,
in compliance with the rule of three.14 Affinity screening was first based on surface plasmon
resonance (SPR).15 Binding of the fragments to wild type human COMT apoprotein was
measured at a single concentration of 200 ꢀM, in absence of SAM and Mg2+ to exclude
chelators.
Results obtained with this protocol were compared with the binding to six enzyme
mutants harboring variations in the SAM binding domain.16 This counter screen ensured the
specificity of the interaction with the desired coꢀfactor pocket.
ACS Paragon Plus Environment

Page 5 of 47
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 2. Selection of specific mutants on the adenosine binding pocket. Amino acids in bold
(Glu90, Ser119 and His142) were mutated
Upon selection of 600 fragments based on affinity and site specificity, hit confirmation
was performed with a 10 point SPR dose response measurement.15 For 200 confirmed binders
1
identified, one dimensional (1D) H NMR Spin Lock variation of CarrꢀPurcellꢀMeiboom Gill
(CPMG)14,17 ligand detected method was performed, by measuring the NMR spectra of the
compound without/with hCOMT protein and observing line broadening due to fragment binding
to the slow tumbling hCOMT and the differences in relaxation. Binding of fragments confirmed
by 1D 1H NMR, was further characterized by twoꢀdimensional (2D) 1H NMR target detect
experiments method performed with 15N labeled hCOMT protein by use of heteronuclear singleꢀ
quantum coherence (1H/15N HSQC) NMR with chemical shifts mapping, where fragmentꢀ
1
induced chemical shift differences of H/15N cross peaks of hCOMT amide backbones involved
in binding, are observed.14,17a,18 Additionally, all 600 fragments were tested in a functional
enzymatic assay19 to determine the IC50 values for COMT inhibition. Very few fragments
showed significant binding in the coꢀfactor pocket combined with a weak enzymatic inhibition
and only four fragments were active in all assays. Following this screening/validation cascade we
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 47
1
2
3
4
5
6
7
8
9
identified three pyrazole derivatives as fragment hits (Table 1) with high ligand efficiencies
(LE).20 It is worth noting that we found similar starting points to the ones recently reported, in
particular fragment 1.21 This supports our findings and it is not surprising since, unlike the highꢀ
throughput screening libraries, fragment libraries typically contain commercially available
molecules selected with similar property criteria.14
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. Selected fragment screening hits. Binding, functional and ligand efficiency data.
3´
3´
3´
1
2
2'
Fragment hit
SPR KD (ꢀM)
IC50 (ꢀM)
LE
1
2
18
2´
18
9
69
80
85
0.44
0.37
0.36
In order to avoid interspecies differences in our hit to lead optimization process, we
performed coꢀcrystallization efforts with humanized rat COMT as this surrogate crystallizes
better that human COMT and yields data relevant to human.5c The crystal structure of humanized
rat COMT in complex with fragment 2 (PDB code 5k03) confirmed its binding in the SAM
pocket (Figure 3).
ACS Paragon Plus Environment

Page 7 of 47
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 3. Superposition of the crystal structures of rat COMT9 (orange) in complex with Sꢀ
adenosylmethionine (SAM, yellow), dinitrocatechol (DNC, yellow), magnesium (Mg2+) and
humanized rat COMT (gray) bound to fragment 2 (green). Met91 of the rat COMT is mutated in
the humanized protein to Ile91.
Most striking is the Hꢀbond (290 pm) between the backbone NH of Ser119 and the
nitrogen of the imidazopyridine from 2. It corresponds to the Hꢀbond interaction of Ser119 to the
Nꢀ1 of adenine in the classical SAM binding mode. The pyrazole occupies the ribose binding
region and the NH serves as Hꢀbond donor (280 pm) to the carboxylate of Glu90, mimicking the
interaction of the hydroxyl groups of the ribose in SAM which are normally the partners for the
Glu90 carboxylate. There is a novel element in the binding of 2 to the adenosine pocket which is
not present in the SAM bound structure, the Hꢀbond (310 pm) between the backbone NH of
Ile91 and the unprotonated Nꢀ1 of the pyrazole moiety. Additionally, the imidazopyridine ring
forms a sandwichꢀlike interaction with the aliphatic side chain of Ile91 on one side and with the
benzylic CH2 of Trp143 on the other side. The benzylic CH2 of His142 interacts with both the
pyrazole ring and the imidazopyridine. Finally, the pyrazole also interacts with the aromatic ring
of Tyr68 in an edgeꢀtoꢀface interaction. Comparing the Xꢀray structures with SAM and fragment
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 47
1
2
3
4
5
6
7
8
9
2, the side chains of Trp143 and to a lesser extend of Arg146, move to offer space to the pyridine
ring. Also, the previously mentioned Tyr68 flips over to close the methionine channel indicating
significant flexibility.
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
Fragment expansion using interactive substructure searches and similarity searches
within the Roche compound library, followed by chemical synthesis, led to the identification of a
series of thiazole analogs (Table 2) with affinities in the micromolar range.
Table 2: IC50 values obtained from the enzymatic assay for representative first fragment
modifications.
4
2
3
4
5
3'
5'
R1=
6
7
8
9
Compd
3
4
5
6
7
8
9
a
21±3
16±2
>625
24±2
>626
11±2.5
>425
IC50 (ꢀM)
a IC50: half maximal inhibitory concentration in enzymatic functional assay. IC50 values
are the average of two to four independent measurements ±standard deviation.
A formal exchange of the corresponding pyrazole ring in the original fragment 1 (IC50 =
69 ꢀM) by an identically substituted thiazole 3 (IC50 = 21 ꢀM), led to a threefold increase in
affinity. In general, the thiazolylpyrazoles were slightly more potent. Another group
significantly affecting affinity is the 4ꢀmethyl substitution at the thiazole ring. 4ꢀMethyl
substituted 6 (IC50 = 24 ꢀM) compares favorably with its unsubstituted analog 7 (>625 ꢀM).
ACS Paragon Plus Environment

Page 9 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Indeed, 4ꢀmethyl seems to be optimal, illustrated by 4 (16 ꢀM), and 5 (>625 ꢀM). At position 2
of the thiazole ring, only a 2ꢀtolyl substituted compound 8 was similarly potent as its 2ꢀmethyl
analog 4. Other alkylꢀ or aryl substituents with improved potency at this position, could not be
identified (data not shown). The more sterically demanding benzylic derivative 9 was inactive.
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
To better understand the affinity differences, we examined the crystal structure of
compound 8 in complex with humanized rat COMT (PDB code: 5k05) (Figure 4).
Figure 4. Crystal structure of humanized rat COMT with fragment 8
The crystal structure shows that the nitrogen atom of the thiazole ring is involved in
hydrogen bonding to the backbone NH (280 pm) of Ser119. The pyrazole ring, as described for
compound 2, occupies the ribose binding region and both nitrogen atoms form hydrogen bonds
to Glu90 (270 pm) and the backbone NH of Ileꢀ91 (310 pm).
For compound 8, the flexible Trp143 is now in close contact to the thiazole, binding
edgeꢀtoꢀface to its sulfur. Extensions at position 2 in the thiazole ring, are at the surface of the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 47
1
2
3
4
5
6
7
8
9
enzyme pointing to the solvent. Larger substituents could even reach out, through the surface of
the enzyme into the surrounding water. This may partially explain why no significant gain in
affinity is observed with addition of aromatic rings comparing compound 4 with 6, 8 and the lost
of activity for compound 9 in Table 2.
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 methyl group at position 4 in the thiazole (4, 6, 8) fits perfectly into a small pocket
deep in the binding site (Figure 5). A larger substituent such as cyclopropyl (5) is already too big
to accommodate, while a hydrogen (7) leaves an unoccupied cavity.
Figure 5. Crystal structure of humanized rat COMT with fragment 8, oriented in order to show
the small pocket at position 4 in the thiazole.
Although the coꢀ crystal structures of 2 and 8 (Figures 3 and 4) indicate that there is
limited space due to the flipped conformation of Tyr68 compared to the SAMꢀbound state,
considering the observed flexibility, we decided to explore substituent effects at position 3’ (R2)
of the pyrazole while keeping constant the 5ꢀ(1Hꢀpyrazolꢀ5ꢀyl)ꢀ2,4ꢀdimethylꢀthiazole motif of
compound 4 (Table 3 with key representatives).
ACS Paragon Plus Environment

Page 11 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
Table 3: IC50 values and physicochemical parameters to assess brain penetration for 3´ꢀ
substituted derivatives (R2).
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
10
11
12
13
17
14
4
2
15
16
3'
5'
R2 =
18
19
20
21
22
23
24
Solubility
(ꢀg/mL)
compd
IC50 (ꢀM)
PSA[a] clogP [b] logD[c]
130±20
34±5
37.0
37.1
37.1
36.9
37.1
45.5
37.0
53.9
37.3
37.3
37.3
37.3
35.8
4.2
3.2
3.0
3.4
3.2
3.3
3.6
3.2
3.5
3.5
3.5
3.4
3.5
ꢀ
ꢀ
8
4
15
ꢀ
10
11
4.4
4.0
3.6
2.6
3.7
ꢀ
8±1
12
2±0.2
13
20±1
14
1.0±0.1
3.2±0.5
1.3±0.1
0.9±0.1
1.1±0.001
0.7±0.05
21±3
2
ꢀ
15
16
3.9
3.9
ꢀ
4
3
ꢀ
17
18
18S
18R
19
4.0
4.3
4.1
5
24
4
1.1±0.3
20
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 47
1
2
3
4
5
6
7
8
9
0.21±0.04
0.86±0.16
0.47±0.14
0.075±0.01
36.8
45.9
45.8
45.6
3.2
3.4
3.4
3.0
4.2
3.9
4.0
4.1
1
5
5
2
21
22
23
24
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] PSA: Polar surface area, (for CNS penetration< 60ꢀ70Å2), [b] clogP = calculated
logarithm of the partition coefficient, [c] logD = logarithm of the octanolꢀwater distribution
coefficient @ pH 7.4. Note: compounds on table 3 have: MW<350 Da, Hydrogen bond donors
(HBD):1 (NHs+OHs), and Hydrogen bond acceptors (HBA): 2ꢀ3 (Ns+Os).
Exploring substituent effects at position 3’ (R2) on Table 3, we observed that direct
attachment of a sterically demanding phenyl ring at the pyrazole, compound 10, led to a strong
decrease in potency. Variation on the length of a slim linker revealed a key trend for potency
improvement (compounds 11 to 13), with 2 ꢀM binding affinity for the shortest benzylic
compound 13. However, the lack of a phenyl ring attached to the linker (alkyl compound 14
compared with 13) reduced potency.
Further modifications were performed to understand the binding requirements at the (R2)
side of the molecule for the benzylic compounds.
We observed a tendency for potency increase with higher electron densities at the phenyl ring
and in particular with a small alkoxy group (15) in para position. In this regard, alternatives with
lower electron density such as a pyridine or chloro substituted rings have IC50 values > 3 ꢀM,
(data not shown). Increase of size at the benzylic aromatic ring does not contribute to potency
improvement (16), however heterocyclic biaryls such as a methyl pyrazole 17 are tolerated and
ACS Paragon Plus Environment

Page 13 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
show similar potency to the alkoxy derivative 15. Significant changes in potency were achieved
with different substitutions at the benzylic position (18ꢀ21). Introducing a methyl substituent
leads to submicromolar potency (18) while for example an ethyl group, is already too large (19).
For the monosubstitution at the benzylic side, the R enantiomers are consistently more potent by
about 2ꢀfold (e.g 23 vs 22).22 Bridging of the dimethyl (20) to a cyclopropyl group (21) leads to a
further gain in affinity. By combining the cyclopropyl at the benzylic position with an optimally
substituted aromatic ring, potencies below 100 nM are obtained for the first time (24). Pyrazoloꢀ
pyrazole analogs follow a similar increase in potency as the thiazoloꢀpyrazoles with introduction
of the benzyl and cyclopropyl functions and are less potent.23 We were able to obtain a crystal
structure of the most active compound (24) coꢀcrystallized with humanized rat COMT (PDB
code: 5k0l) to clarify key interactions (Figure 6).
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 47
1
2
3
4
Figure 6. Crystal structure of humanized rat COMT with inhibitor 24: hydrogen bonds (black)
5
6
and hydrophobic interactions (blue) indicated.
7
8
9
The thiazoloꢀpyrazole Hꢀbond pattern is confirmed to be as described for the previous
compounds. Trp143 is in close contact to the thiazole sulfur atom (330 pm) and the
methoxybenzene ring (350 pm). The methoxybenzene ring additionally interacts with the side
chain of Met40 and is thus in a sandwich constellation.
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
These interactions rationalize the importance of the aromatic ring as observed in the
comparison of compounds 13 with 14 (Table 3). Importantly, the side chain of Tyr68 moves out
to create space for the binding of the cyclopropylene linker. The two cyclopropyl carbon atoms
make 3 vdW interactions with the side chain of Tyr68.
Although the binding potencies for the monomethyl analogs are not of big difference, we
observed that the R enantiomers show about 2 fold higher potencies than the S enantiomers. The
difference in potency becomes larger when comparing the weakest Sꢀmethyl enantiomers (e.g 22
with IC50: 0.86 ꢀM) with the cyclopropyl derivative 24 (IC50: 0.075 ꢀM) or the 18 Sꢀmethyl
enantiomer 18S (IC50: 1.1 ꢀM) with the cyclopropyl 21 (0.21 ꢀM). There is also a five fold
reduction in potency from the cyclopropyl derivative 21 to the dimethyl analog 20 (IC50: 1.1 ꢀM)
which binds with similar potency compared to the weakest 18S enantiomer. Interestingly,
superpositions for the most rigid cyclopropyl (24 in green), the dimethyl (20 in yellow), and the
R and S monomethyl enantiomers (racemic 18 in blue and 22+23 racemic mixture in orange)
reveals different conformational trends for the most differentiated compounds (Figure 7). The
dimethyl substitution is sterically more demanding than cyclopropyl and seems to fit less well
into the pocket. This dimethyl substitution could also affect the preferred torsional angle of the
ACS Paragon Plus Environment

Page 15 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
aryl group leading to different and probably worse, interactions of the terminal phenyl. These
less favorable interactions for the terminal aromatic ring are also observed for the weakest S
enantiomers (see superpositions of the crystal structures on Figure 7). Furthermore, for the
dimethyl and for the R/S monomethyl analogs at the benzylic side, the number of vdW
interactions with the Tyr 68 (indicated on Figure 6) are reduced with respect to the best fitting
two carbon interactions on the cyclopropyl analog.
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. Superposition of the crystal structures of humanized rat COMT with inhibitors
reported in Table 3: 22/23 (both in orange, coꢀcrystallized as racemate PDB code 5k0f), 18 (blue,
coꢀcrystallized as racemate, however one enantiomer (18R) observed in the electron density,
PDB code 5k0b), 20 (yellow, PDB code 5k0c) and 24 (green, PDB code 5k01).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 47
1
2
3
4
5
6
The role of key amino acids (in particular Ty68) in the enzymatic methyl transfer reaction
catalyzed by COMT, has recently been analyzed in great detail.24
7
8
9
CHEMISTRY
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 synthetic strategy to obtain compounds 10ꢀ16 is depicted in Scheme 1.
Deprotonation of ketone 25 using NaH followed by addition of the corresponding ester gave the
diketone intermediates 26aꢀg as a mixture of tautomers. Attempts to use acid chlorides instead of
esters in this transformation gave no product or very poor yields. Final compounds 10ꢀ16 were
obtained by condensation of intermediates 26aꢀg with hydrazine hydrate in ethanol with catalytic
HCl.
Suzuki coupling with a suitable bromo phenyl substituted intermediate 28 (Scheme 2)
prepared similarly as compounds 10ꢀ16, allowed the preparation of bicyclic analogs such as
compound 17.
Scheme 1. Synthesis of compounds 10ꢀ16 described in Table 3a
ii
i
-
10 16
25
26a-g
26a
26b
26c
26d
26e
R2=
26f
26g
aReagents and conditions: (i) NaH, 0 °C, 15 min then R2CO2Et, 0ꢀ100 °C, 1ꢀ3 h. 56ꢀ65%; (ii)
N2H4·H2O, EtOH/25% aq. HCl (80:1), 60 °C, overnight 60ꢀ90%.
ACS Paragon Plus Environment

Page 17 of 47
Journal of Medicinal Chemistry
1
2
3
4
Scheme 2. Synthesis of compound 17 in Table 3a
5
6
7
8
9
27
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
i
ii
25
26h
28
iii
29
17
aReagents and conditions: (i) NaH, 0 °C, 30 min then 27, 0ꢀ100 °C, 1 h, 63%; (ii) N2H4·H2O,
EtOH/25% aq. HCl (80:1), 60 °C, 3 h, 99%; (iii) 29, Cs2CO3, (PPh3)4Pd, dioxane, H2O (4:1), 165
°C, microwave, 20 min, 51%.
For the preparation of the substituted methylene bridged benzylic derivatives (18ꢀ24)
(Scheme 3) the addition to the deprotonated methyl ketone 25, required aldehydes 32aꢀf and very
low temperatures (ꢀ78 °C) in order to obtain acceptable yields. Deprotonation was best
performed using a lithium base. The corresponding starting aldehydes 32aꢀf (many of them
commercially available) were obtained by reduction of available acids 30aꢀf to alcohols 31aꢀf,
followed by oxidation using DessꢀMartin periodinane. Oxidation of the βꢀhydroxy ketone
intermediates 33aꢀf to diketones 34aꢀf was also performed with DessꢀMartin periodinane.
Reaction of intermediates 34aꢀf with hydrazine gave the final compounds 18ꢀ24 with the central
pyrazole ring in one unique tautomeric form in DMSO, as corroborated by NOESYꢀNMR with
some observed NOEs between the NH of the imidazole and the benzylic CH or CH2.
Enantiomers such as 22 and 23, were initially obtained from racemic precursor by
chromatographic separation using a chiral column.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 47
1
2
3
4
Scheme 3. Synthesis of compounds 18ꢀ24 described in Table 3a
5
6
7
8
9
25
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
iv
iii
i
ii
30a-f
31a-f
32a-f
33a-f
34a-f
v
R2 =
18, 30-34a
30-34e
19, 30-34b
20, 30-34c
21, 30-34d
24, 30-34f
18-24
22
23
aReagents and conditions: (i) LiAlH4, THF, 0ꢀ20 °C, 70ꢀ98%; (ii) DessꢀMartin periodinane,
CH2Cl2, 0ꢀ25 °C, 10 h, 70ꢀ83%; (iii) 25, LiN(SiMe3)2, THF, ꢀ78 °C, 0.5 h; then 32, 3 h, ꢀ78 °C,
60ꢀ99%; (iv) DessꢀMartin periodinane, CH2Cl2, 0ꢀ25 °C, 5 h, 61ꢀ88%; (v) N2H4·H2O,
EtOH/25% aq. HCl (80:1), 60 °C, 4 h, 61ꢀ92%.
CONCLUSION
This study illustrates the successful application of a fragmentꢀbased approach followed
by fragment optimization, to obtain for the first time potent and novel inhibitors that bind to the
SAM pocket of COMT and do not require a catechol or phenol like motif for high potency.
These compounds behave as competitive inhibitors of SAM.
In the reported lead optimization process, a challenging fragment derivatization was
performed by observing key binding pocket characteristics, described in a stepwise manner with
ACS Paragon Plus Environment

Page 19 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the important use of Xꢀray crystallography. Key interactions are described such as hydrogen
bonding between the Ser118 backbone NH and the thiazole or the pyrazole nitrogen for the
thiazoloꢀpyrazole and pyrazoloꢀpyrazole subꢀseries, as well as interactions of the central pyrazole
with Glu90 and with the backbone NH of Ile91. A small lipophilic pocket was identified
occupied ideally with a 4ꢀmethyl substituent in the thiazole ring, that has an edgeꢀtoꢀface contact
with Trp143.
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
Flexibility of the enzyme was confirmed, as previously described,25 for several amino
acids such as the Trp143 side chain movement. In addition, we provide insights for a relevant
movement and interactions of Tyr68 that opens a novel pocket, offering best contacts with a
cyclopropylene linker at a benzylic position. This allows the aromatic ring, to fit best into the
pocket created by Trp143 and Met40 with some preferential substitution patterns.
The designed potent lead compounds have CNS drug like properties. Taken together, the
reported study and observations represent an advance towards the rational design of CNS drug
like and potent non phenolic COMT inhibitors.
EXPERIMENTAL SECTION
COMT Enzyme Inhibition Assay.
The principle of the assay relies on methylation of a 4ꢀnitrocatecholꢀAlexa Fluor 488
substrate. After transfer of the methyl group from Sꢀadenosylmethionine (SAM) to the substrate,
an increase of fluorescence intensity is observed because the methylation disturbs the
intramolecular static quenching of Alexa Fluor 488. This increase of fluorescence intensity can
be followed in a kinetic measurement. An inhibitor compound of COMT decreases the slope.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 47
1
2
3
4
5
6
7
8
9
In detail, 10 ꢀl of recombinant hCOMT (expressed in E.coli.) diluted to 80 nM in assay
buffer, was pipetted into a 384 well microtiter plate (black with flat clear bottom, nonꢀbinding
surface treated polystyrene, Corning ref. 3655). 2 ꢀl of compound dilution in 100% DMSO was
then added to the enzyme solution and the plate was shaken with 1500 rpm for 1 min. After
adding 20 ꢀl of a mixture of 320 nM 4ꢀnitrocatecholꢀAlexa488 (produced in house) and 800 nM
SAM (SigmaꢀAldrich A2804) in assay buffer the plate was shaken for 5 min at 1500 rpm. The
plate was then transferred into a microtiter plate reader and the fluorescence was measured every
60s for 180 times at an excitation wavelength of 475 nm and an emission wavelength of 535 nm.
The slope was then calculated from the linear range of the kinetic. The assay buffer used was 40
mM phosphate buffer at pH 7.6 containing 2.88 mM MgCl2, 0.9 mM DTT and 0.25 mM CaCl2.
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 patented COMT Enzyme Inhibition Assay is a very sensitive homogenous assay,
suitable for screening large numbers of compounds for their COMT modulating activity.19 It
provides a robust read out, with IC50 values determined by the average of two to four
independent repeat measurements. The average coefficient of variation (%CV) of the IC50 values
is 15%.
Enzyme Cloning, Expression and Purification.
The synthetically made gene for human, soluble COMT encoding the sequence from
M51 to P271 (UniProt: P21964) was purchased from GenScript Corp. (NJ, USA) and subcloned
in pET32a. The recombinant protein was expressed in BL21 (DE3) cells. The cDNA encoding
for rat COMT was PCR cloned into the E. coli expression vector pDS56/RBSII. The construct
encoded for the soluble part of rat COMT from M44 to S264 (UniProt: 22734). Recombinant
rCOMT was produced in E. coli W3110[pREP4]. In our numbering, M51 for human and M44
ACS Paragon Plus Environment

Page 21 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
for rat of the soluble COMT enzymes were set as first amino acids in the alignment of the
sequences. Cell lysis was made in 50 mM Tris/HCL pH 7.6 containing 10 mM 2ꢀ
mercaptoethanol, inhibitor complete (Roche Applied Science), lysozyme, DNAse and RNAse.
After one passage through the cell disruptor at 700 bar the solution was centrifuged (20 min,
30’000 x g, 4°C). The supernatant was collected and ammonium sulfate added to 55% saturation.
The protein pellet obtained after centrifugation at 30’000 x g was dissolved in 20 mM Bisꢀ
Tris/HCl pH 6.2 containing 10 mM DTT and 1 mM MgCl2, dialyzed against the same buffer and
then again centrifuged. The clear supernatant was applied on a Superdex Sꢀ75 column (100 x 2.6
cm). Enzyme containing fractions were pooled and further purified on a HiTrap Q FF (5 ml)
column in the same BisꢀTris buffer. The protein was eluted in the course of a linear NaCl
gradient. Finally, the protein was again applied on a Superdex Sꢀ75 column (60 x 1.6 cm) in 50
mM Tris/HCl pH 7.5, 50 mM NaCl, 10 mM DTT, 2 mM MgCl2. As shown by SDSꢀPAGE and
HPLC, >98% pure protein was obtained. The purified protein was monomeric as detected by
analytical ultracentrifugation and active in a specific enzyme assay.
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
Protein engineering by site directed mutagenesis, according to the Statagene Quik
Change Mutagenesis Method, was applied to mutate specific amino acids in the active site of the
enzyme. In order to block the Sꢀadenosylꢀmethionine pocket, bulky amino acids like
tryptophane, arginine and lysine were chosen. In fact the mutations E90R, E90W, S119W,
H142R, H142K, H142W were made. Using the same mutagenesis strategy a “humanized” rat
protein was made. Thereby the methionine M91 in the so called methionine pocket of the wild
type rat enzyme was mutated to isoleucin as it is found in the wild type human enzyme. All
mutant proteins could be expressed and purified as described above for the wild type enzymes.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 47
1
2
3
4
Compound Synthesis and Characterization. Chemistry.
5
6
7
8
General. All compounds from Table 3 are described below. Reactions were carried out under
argon atmosphere. Unless otherwise stated, the reactions were stopped when the starting material
was fully consumed according to thin layer chromatography (TLC). The solvent was then
removed and the residue was purified by column chromatography. Unless otherwise mentioned,
all reagents and chemicals were obtained from commercial suppliers and used without further
purification. The purity of final compounds as measured by HPLC was at least above 95%.
Column chromatography was carried out either using cartridges packed with silica gel (Isolute
Columns, Telos Flash Columns) or on glass columns on silica gel 60 (32ꢀ60 mesh, 60Å) with the
eluent mixtures given for the corresponding procedures. LC high resolution spectra were
recorded with a Agilent LCꢀsystem consisting of Agilent 1290 high pressure system, a CTC PAL
auto sampler and a Agilent 6520 QTOF. The separation was achieved on a Zorbax Eclipse Plus
C18 1,7 µm 2.1*50mm column at 55°C; A=0.01% formic acid in Water; B= 0.01% formic acid
in acetonitrile at flow 1 mL/min. gradient: 0 min 5%B, 0.3 min 5%B, 4.5 min 99 %B 5 min
99%B. The NMR spectras were measured on a Bruker 600 MHz machine in a 5 mm TCI
cryoprobe at 298 K. TMS was used for referencing. Chemical shifts are reported relative to the
solvent in ppm as reference. Coupling constants (J) are given in Hz. The resonance multiplicity
is described as s (singlet), d (doublet), t (triplet), q (quadruplet), m (multiplet), and br. (broad).
1H NMR is reported for the major tautomer.
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
1ꢀ(2,4ꢀDimethylthiazolꢀ5ꢀyl)ꢀ3ꢀphenylꢀpropaneꢀ1,3ꢀdione (10, intermediate 26a).
Sodium hydride (60 %, 186 mg, 7.73 mmol) was suspended in THF (20 mL). At 0 °C a solution
of 1ꢀ(2,4ꢀdimethylthiazolꢀ5ꢀyl)ethanone 25 (CAS 38205ꢀ60ꢀ6, 600 mg, 3.87 mmol) in THF (4
mL) was added dropwise. The reaction mixture was allowed to warm to room temperature and
ACS Paragon Plus Environment

Page 23 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
stirred for 15 minutes. A solution of ethyl benzoate (1.16 g, 1.11 mL, 7.73 mmol) in THF (4 mL)
was added dropwise at 0 °C. After 5 minutes at 0 °C the mixture was stirred at 100 °C for 2 hrs
and then cooled to 0 °C. A saturated aqueous solution of ammonium chloride (25 mL) was
added. Extraction with water/ethyl acetate and chromatography (silica, 0ꢀ50% EtOAc in heptane)
yielded 602 mg (60% yield) of 26a (CAS1467020ꢀ71ꢀ8) as a yellow solid. Rf: 0.46
(EtOAc/heptane 1:1); 1H NMR (600 MHz, CDCl3) δ ppm 16.53 (s, 1H), 7.91 (dd, J=8.2, 1.1 Hz,
2H), 7.53 ꢀ 7.57 (m, 1H), 7.47 ꢀ 7.50 (m, 2H), 6.47 (s, 1H), 2.78 (s, 3H), 2.72 (s, 3H) (compound
present only as enolꢀform); LCꢀHRMS: m/z = 260.0746 [(M+H)+ calculated for C14H13NO2S =
260.074; Diff = 0.60mD].
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
2,4ꢀDimethylꢀ5ꢀ(5ꢀphenylꢀ1Hꢀpyrazolꢀ3ꢀyl)thiazole (10). 1ꢀ(2,4ꢀDimethylthiazolꢀ5ꢀyl)ꢀ3ꢀ
phenylꢀpropaneꢀ1,3ꢀdione (26a) (300 mg, 1.16 mmol) was combined with ethanol (15 ml).
Hydrazine monohydrate (99.5 mg, 98 µl, 1.27 mmol) and hydrochloric acid (25%, 75.0 µl) were
added. The solution was stirred at 60 °C overnight. For compounds 11ꢀ17 with sterically less
demanding linkers, the reaction was typically finished after 2ꢀ3 hrs. The reaction mixture was
basified with saturated aqueous sodium bicarbonate and concentrated. Chromatography (silica,
0ꢀ5% MeOH in dichloromethane) and trituration with diethyl ether/pentane gave compound 10
1
as offꢀwhite solid (105 mg, 36% yield). Rf: 0.43 (dichloromethane/methanol 9:1); H NMR (600
MHz, DMSOꢀd6 ) δ ppm 13.46 (br s, 1H), 7.82 (br d, J=6.6 Hz, 2H), 7.47 (br d, J=6.3 Hz, 2H),
7.38 (br d, J=6.5 Hz, 1H), 6.94 (s, 1H), 2.60 (br s, 3H), 2.53 (br s, 3H); LCꢀHRMS: m/z =
256.0909 [(M+H)+ calculated for C14H13N3S = 256.0903 ; Diff = 0.60mD]
1ꢀ(2,4ꢀDimethylthiazolꢀ5ꢀyl)ꢀ6ꢀphenylꢀhexaneꢀ1,3ꢀdione (11, intermediate 26b). 1ꢀ(2,4ꢀ
Dimethylthiazolꢀ5ꢀyl)ethanone 25 (150 mg, 966 µmol), NaH (60% in mineral oil, 46.4 mg, 1.16
mmol) and ethyl 4ꢀphenylbutanoate (372 mg, 1.93 mmol) were used to synthesize 26b following
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 47
1
2
3
4
5
6
7
8
9
the procedure described for the preparation of 26a. Compound 26b (40.0 mg, 13% yield) was
obtained as a yellow oil. Rf: 0.54 (EtOAc/heptane 1:1); (600 MHz, CDCl3) δ ppm 15.95 (s, 1H),
7.31 ꢀ 7.27 (m, 2H), 7.23 ꢀ 7.15 (m, 1H), 7.21 ꢀ 7.17 (m, 2H), 5.77 (s, 1H), 2.70 (s, 3H), 2.69 (s,
3 H), 2.71 – 2.68 (m, 2H), 2.37 (t, J=7.6 Hz, 2H), 2.00 (quin, J=7.6 Hz, 2 H) (9:1 tautomeric
mixture of enol and diꢀketone); LCꢀHRMS: m/z = 302.1206 [(M+H)+ calculated for C17H19NO2S
= 302.1209 ; Diff = ꢀ0.30mD].
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
2,4ꢀDimethylꢀ5ꢀ[5ꢀ(3ꢀphenylpropyl)ꢀ1Hꢀpyrazolꢀ3ꢀyl]thiazole
(11).
1ꢀ(2,4ꢀ
Dimethylthiazolꢀ5ꢀyl)ꢀ6ꢀphenylꢀhexaneꢀ1,3ꢀdione 26b (170 mg, 564 µmol), hydrazine
monohydrate (48.5 mg, 47 µL, 620 µmol) and hydrochloric acid (25%, 10 µL) were used to
synthesize 11 following the procedure described for the preparation of 10. After
chromatography, compound 11 (115 mg, 68% yield) was obtained as an offꢀwhite solid. Rf: 0.21
1
(EtOAc/heptane 1:1); H NMR (600 MHz, CDCl3) δ ppm 11.46ꢀ7.9 (br s, 1H), 7.32 ꢀ 7.28 (m,
2H), 7.23 ꢀ 7.19 (m, 1H), 7.20 ꢀ 7.17 (m, 2H), 6.23 (s, 1H), 2.73 ꢀ 2.69 (m, 2H), 2.72 ꢀ 2.68 (m,
2H), 2.67 (s, 3H), 2.55 (s, 3H), 2.03 (quin, J=7.6 Hz, 2H); LCꢀHRMS: m/z = 298.138 [(M+H)+
calculated for C17H19N3S = 298.1373 ; Diff = 0.70mD].
1ꢀ(2,4ꢀDimethylꢀthiazolꢀ5ꢀyl)ꢀ5ꢀphenylꢀpentaneꢀ1,3ꢀdione (12, intermediate 26c). 1ꢀ
(2,4ꢀDimethylthiazolꢀ5ꢀyl)ethanone 25 (100 mg, 0.644 mmol), NaH (60% in mineral oil, 31 mg,
0.77 mmol) and ethyl 3ꢀphenylpropanoate (230 mg, 0.774 mmol) were used to synthesize 26c
following a similar procedure as described for the preparation of 26a. In this case 1ꢀ(2,4ꢀ
dimethylthiazolꢀ5ꢀyl)ethanone and NaH were stirred at 0 °C for 30 minutes in THF (0.5 mL)
before adding the ester. After chromatography (silica gel, 0ꢀ20% EtOAc/heptane), compound
1
26c (185 mg, 57% yield) was obtained as a light yellow oil. Rf: 0.39 (EtOAc/heptane 1:1); H
NMR (600 MHz, CDCl3) δ ppm 15.93 (br. s., 1H), 7.32 ꢀ 7.28 (m, 2H), 7.27 ꢀ 7.17 (m, 3H), 5.73
ACS Paragon Plus Environment

Page 25 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
(s, 1H), 3.01 ꢀ 2.97 (m, 2H), 2.68 (s, 3H), 2.66 (s, 3H), 2.69 ꢀ 2.65 (m, 2H); LCꢀHRMS: m/z =
288.1066 [(M+H)+ calculated for C16H17NO2S = 288.1058 ; Diff = 1.30 mD].
7
8
9
2,4ꢀDimethylꢀ5ꢀ(5ꢀphenethylꢀ2Hꢀpyrazolꢀ3ꢀyl)ꢀthiazole (12). 1ꢀ(2,4ꢀDimethylꢀthiazolꢀ
5ꢀyl)ꢀ5ꢀphenylꢀpentaneꢀ1,3ꢀdione 26c (40 mg, 0.14 mmol), hydrazine monohydrate (7.7 mg, 7.6
µL, 0.150 mmol) and hydrochloric acid (25%, 10 µL) were used to synthesize 12 following the
procedure described for the preparation of 10. After chromatography, compound 12 (36 mg, 91%
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
1
yield) was obtained as a light brown solid. Rf: 0.13 (EtOAc/heptane 1:1); H NMR (600 MHz,
CDCl3) δ ppm 7.34 ꢀ 7.28 (m, 2H), 7.25 ꢀ 7.22 (m, 1H), 7.21 ꢀ 7.17 (m, 2H), 6.20 (s, 1H), 3.07ꢀ
2.90 (m, 4H), 2.66 (s, 3H), 2.53 (s, 3H); LCꢀHRMS: m/z = 284.1233 [(M+H)+ calculated for
C16H17N3S = 284.1216 ; Diff = 1.70 mD].
1ꢀ(2,4ꢀDimethylꢀthiazolꢀ5ꢀyl)ꢀ4ꢀphenylꢀbutaneꢀ1, 3ꢀdione (13, intermediate 26d). 1ꢀ
(2,4ꢀDimethylthiazolꢀ5ꢀyl)ethanone 25 (50 mg, 0.32 mmol), NaH (60% in mineral oil, 26 mg,
0.64 mmol) and ethyl 2ꢀphenylacetate (10.6 mg, 0.64 mmol) were used to synthesize 26d
following the procedure described for the preparation of 26c. Compound 26d (49 mg, 56% yield)
1
was obtained as a white powder. Rf: 0.49 (EtOAc/heptane 1:1); H NMR (600 MHz, CDCl3) δ
ppm 7.36 ꢀ 7.34 (m, 2H), 7.31 ꢀ 7.27 (m, 3H), 5.72 (s, 1H), 3.66 (s, 2H), 2.67 (s, 3H), 2.61 (s,
3H); LCꢀHRMS: m/z = 274.0893 [(M+H)+ calculated for C15H15NO2S = 274.0896 ; Diff = ꢀ0.30
mD].
2,4ꢀDimethylꢀ5ꢀ(5ꢀbenzylꢀ2Hꢀpyrazolꢀ3ꢀyl)ꢀthiazole (13). 1ꢀ(2,4ꢀDimethylꢀthiazolꢀ5ꢀ
yl)ꢀ4ꢀphenylꢀbutaneꢀ1,3ꢀdione 26d (100 mg, 0.37 mmol), hydrazine monohydrate (20 mg, 20 µL,
407 µmol) and hydrochloric acid (25%, 10 µL) were used to synthesize 13 following the
procedure described for the preparation of 10. The crude product was purified by trituration with
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 47
1
2
3
4
5
6
7
8
9
diethyl ether to afford 13 as an off white solid (50 mg, 51% yield). Rf: 0.19 (EtOAc/heptane 1:1);
1H NMR (600 MHz, CDCl3) δ ppm 7.36 ꢀ 7.32 (m, 2H), 7.28 (d, J = 7.3 Hz, 1H), 7.26 ꢀ 7.23 (m,
2H), 6.25 (s, 1H), 4.05 (s, 2H), 2.66 (s, 3H), 2.53 (s, 3H); LCꢀHRMS: m/z = 270.1070 [(M+H)+
calculated for C15H15N3S = 270.1060 ; Diff = 1.00 mD].
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
1ꢀ(2,4ꢀDimethylꢀthiazolꢀ5ꢀyl)ꢀpentaneꢀ1,3ꢀdione (14, intermediate 26e). 1ꢀ(2,4ꢀ
Dimethylthiazolꢀ5ꢀyl)ethanone 25 (100 mg, 0.32 mmol), NaH (60% in mineral oil, 52 mg, 1.3
mmol) and ethyl proprionate (132 mg, 1.29 mmol) were used to synthesize 26e following the
procedure described for the preparation of 26c. Compound 26e (80 mg, 59% yield) was obtained
1
as a light yellow oil. Rf: 0.51 (EtOAc/heptane 1:1); H NMR (600 MHz, CDCl3) δ ppm 5.79 (s,
1H), 2.70 (s, 3H), 2.69 (s, 3H), 2.39 (q, J=7.6 Hz, 2H), 1.19 (t, J=7.6 Hz, 3H); LCꢀHRMS: m/z
= 212.0745 [(M+H)+ calculated for C10H13NO2S = 212.074 ; Diff = 0.50mD].
2,4ꢀDimethylꢀ5ꢀ(5ꢀethylꢀ2Hꢀpyrazolꢀ3ꢀyl)ꢀthiazole (14). 1ꢀ(2,4ꢀDimethylꢀthiazolꢀ5ꢀyl)ꢀ
pentaneꢀ1,3ꢀdione 26e (30 mg, 0.38 mmol), hydrazine monohydrate (21 mg, 21 µL, 408 µmol)
and hydrochloric acid (25%, 10 µL) were used to synthesize 14 following the procedure
described for the preparation of 10. The crude product was purified by trituration with diethyl
1
ether to afford 14 as an off white solid (27 mg, 92% yield). Rf: 0.16 (EtOAc/heptane 1:1); H
NMR (600 MHz, CDCl3) δ ppm 6.23 (s, 1H), 2.72 (q, J = 7.7 Hz, 2H), 2.67 (s, 3H), 2.56 (s, 3H),
1.32 (t, J=7.7 Hz, 3H); LCꢀHRMS: m/z = 208.0924 [(M+H)+ calculated for C10H13N3S =
208.0903 ; Diff = 2.10 mD].
1ꢀ(2,4ꢀDimethylꢀ1,3ꢀthiazolꢀ5ꢀyl)ꢀ4ꢀ(4ꢀmethoxyphenyl)butaneꢀ1,3ꢀdione
(15,
intermediate 26f). 1ꢀ(2,4ꢀDimethylthiazolꢀ5ꢀyl)ethanone 25 (200 mg, 1.29 mmol), NaH (60%
in mineral oil, 103 mg, 2.58 mmol) and ethyl 2ꢀ(4ꢀmethoxyphenyl)acetate (501 mg, 0.456 mL)
ACS Paragon Plus Environment

Page 27 of 47
Journal of Medicinal Chemistry
1
2
3
4
were used to synthesize 26f following the procedure described for the preparation of 26c.
5
6
1
Compound 26f (202 mg, 52% yield) was obtained as a white solid. Rf: 0.51 (EtOAc); HꢀNMR
7
8
9
(600 MHz, CDCl3) (enol tautomer) δ ppm 15.84 (br s, 1H), 7.19 (br d, J=8.7 Hz, 2H), 6.88 (d,
J=8.7 Hz, 2H), 5.71 (s, 1H), 3.81 (s, 2H), 3.60 (s, 2H), 2.67 (s, 4H), 2.62 (s, 3H); LCꢀHRMS:
m/z = 304.1019 [(M+H)+ calculated for C16H17NO3S = 304.1002 ; Diff = 1.70mD].
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
5ꢀ[5ꢀ[(4ꢀMethoxyphenyl)methyl]ꢀ1Hꢀpyrazolꢀ3ꢀyl]ꢀ2,4ꢀdimethylꢀthiazole (15). 1ꢀ(2,4ꢀ
Dimethylꢀthiazolꢀ5ꢀyl)ꢀ4ꢀ(4ꢀmethoxyꢀphenyl)ꢀbutaneꢀ1,3ꢀdione (200 mg, 659 ꢁmol) 26f,
hydrazine monohydrate (36.3 mg, 35 ꢁL, 725 ꢁmol) and HCl 25% (26 ꢀL) were used to
synthesize 15 following the procedure described for the preparation of 10. After
chromatography, compound 15 (159 mg, 81% yield) was obtained as a white solid. Rf: 0.343
1
(EtOAc); HꢀNMR (600 MHz, CDCl3) δ ppm 7.17ꢀ7.16 (m, 2H); 6.88 ꢀ 6.87 (m, 2H), 6.23 (s,
1
1H), 3.99 (s, 2H), 3.81 (s, 3H), 2.66 (s, 3H), 2.53 (s, 3H); HꢀNMR (600 MHz, DMSOꢀd6) δ
13.39 (br s, 1H, NH imidazole); 7.19 (d, J=8.66, 2H), 6.87 (d, J=8.66, 2H), 6.22 (s, 1H), 3.91 (s,
2H), 3.72 (s, 3H), 2.56 (s, 3H), 2.42 (s, 3H); NOESY (600 MHz, DMSOꢀd6): Interaction
between δ 13.39 (br s, 1H, NH), and 3.91 (s, 2H, CH2) indicating the imidazole tautomer is the
one indicated. LCꢀHRMS: m/z = 300.1176 [(M+H)+ calculated for C16H17N3OS = 300.1165 ;
Diff = 1.10mD].
4ꢀ(Biphenylꢀ4ꢀyl)ꢀ1ꢀ(2,4ꢀdimethylꢀ1,3ꢀthiazolꢀ5ꢀyl)butaneꢀ1,3ꢀdione (16, intermediate
26g). 1ꢀ(2,4ꢀDimethylthiazolꢀ5ꢀyl)ethanone 25,(100 mg, 0.644 mmol, 87 ꢀL), NaH (60% in
mineral oil, 51.5 mg, 1.29 mmol) and ethyl biphenylꢀ4ꢀylacetate (CAS number 14062ꢀ23ꢀ8)
(310 mg, 1.29 mmol) were used to synthesize 26g following the procedure described for the
preparation of 26c. Compound 26g (137 mg, 61% yield) was obtained as a white solid. Rf: 0.61
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 47
1
2
1
3
(EtOAc); H NMR (600 MHz, CDCl3) δ 15.86 (br s, 1H, OH enol), 7.52 ꢀ 7.63 (m, 5H), 7.40 ꢀ
7.49 (m, 2H), 7.36 (br d, J=8.1 Hz, 2H), 5.78 (s, 1H), 3.70 (s, 2H), 2.67 (s, 3H), 2.63 (s, 3H);
LCꢀHRMS: m/z = 350.121 [(M+H)+ calculated for C21H19NO2S = 350.1209 ; Diff = 0.10mD].
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
5ꢀ[5ꢀ(Biphenylꢀ4ꢀylmethyl)ꢀ1Hꢀpyrazolꢀ3ꢀyl]ꢀ2,4ꢀdimethylꢀ1,3ꢀthiazole (16).
4ꢀ(Biphenylꢀ4ꢀyl)ꢀ1ꢀ(2,4ꢀdimethylꢀ1,3ꢀthiazolꢀ5ꢀyl)butaneꢀ1,3ꢀdione 26g (116 mg, 332 ꢁmol),
hydrazine monohydrate (18.3 mg, 18 ꢁL, 365 ꢁmol) in ethanol (1.3 mL) and HCl 25% (13 ꢀL)
were used to synthesize 16 following the procedure described for the preparation of 10. After
chromatography, compound 16 (81.4 mg, 71% yield) was obtained as a light yellow solid. Rf:
1
0.16 (pentane/EtOAc 1:1); H NMR (600 MHz, CDCl3): 7.61 ꢀ 7.55 (m, 4H). 7.46 ꢀ 7.41 (m,
2H), 7.37 ꢀ 7.34 (m, 1H), 7.32 (br d, J=8.4 Hz, 2H), 6.29 (s, 1H), 4.10 (br s, 2H), 2.67 (s, 3H),
2.55 (s, 3H); LCꢀHRMS: m/z = 346.1378 [(M+H)+ calculated for C21H19N3S = 346.1373 ; Diff =
0.50mD].
4ꢀ(4ꢀBromophenyl)ꢀ1ꢀ(2,4ꢀdimethylꢀ1,3ꢀthiazolꢀ5ꢀyl)butaneꢀ1,3ꢀdione
(17,
intermediate 26h). 1ꢀ(2,4ꢀDimethylthiazolꢀ5ꢀyl)ethanone 25 (200 mg, 1.29 mmol, 174 ꢀL),
sodium hydride (60% in mineral oil, 103 mg, 2.58 mmol) and ethyl (4ꢀbromophenyl)acetate
(626 mg, 2.58 mmol) were used to synthesize 26h following the procedure described for the
preparation of 26c. Compound 26h was obtained as a light yellow solid (283.8 mg, 63% yield).
Rf: 0.54 (EtOAc); 1H NMR (600 MHz, CDCl3): δ 15.82 (br s, 1H, OH enol), 7.43 ꢀ 7.51 (m, 2H),
7.16 (br d, J=8.5 Hz, 2H), 5.71 (s, 1H), 3.61 (s, 2H), 2.68 (s, 3H), 2.64 (s, 3H); GCꢀEIꢀMS: m/z
= 351 (M·+; calculated: for C15H14BrNO2S m/z = 351)
ACS Paragon Plus Environment

Page 29 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5ꢀ[5ꢀ(4ꢀBromobenzyl)ꢀ1Hꢀpyrazolꢀ3ꢀyl]ꢀ2,4ꢀdimethylꢀ1,3ꢀthiazole (17, intermediate
28). 4ꢀ(4ꢀBromophenyl)ꢀ1ꢀ(2,4ꢀdimethylꢀ1,3ꢀthiazolꢀ5ꢀyl)butaneꢀ1,3ꢀdione 26h (200 mg, 568
ꢁmol), hydrazine monohydrate (31.3 mg, 31 ꢁL, 625 ꢁmol) in ethanol (2.3 mL) and HCl 25%
(23 ꢀL) were used to synthesize 28 following the procedure described for the preparation of 10.
After chromatography, compound 28 (195.2 mg, 99% yield) was obtained as a light yellow solid.
Rf: 0.12 (EtOAc/heptane 1:1), Rf: 0.385 (EtOAc); 1H NMR (600 MHz, DMSOꢀd6): δ 12.86 (br s,
1H, NH), 7.51 (d, J=8.26 Hz, 2H), 7.24 (br d, J=8.4 Hz, 2H), 6.27 (br s, 1H), 3.97 (s, 2H, CH2),
2.56 (s, 3H), 2.43 (s, 3H); NOESY (600 MHz, DMSOꢀd6): Interaction between δ 12.86 (br s,
1H, NH), and 3.97 (s, 2H, CH2) indicating the imidazole tautomer is the one indicated. LCꢀ
HRMS: m/z = 348.0168 [(M+H)+ calculated for C15H14BrN3S = 348.0165 ; Diff = 0.30mD].
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
2,4ꢀDimethylꢀ5ꢀ{5ꢀ[4ꢀ(1ꢀmethylꢀ1Hꢀpyrazolꢀ4ꢀyl)benzyl]ꢀ1Hꢀpyrazolꢀ3ꢀyl}ꢀ1,3ꢀ
thiazole (17). 5ꢀ[5ꢀ(4ꢀBromobenzyl)ꢀ1Hꢀpyrazolꢀ3ꢀyl]ꢀ2,4ꢀdimethylꢀ1,3ꢀthiazole 28 (180 mg,
517 ꢁmol), 1ꢀmethylꢀ4ꢀ(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)ꢀ1Hꢀpyrazole (129 mg, 620
µmol), Cs2CO3 (674 mg, 2.07 mmol, Eq: 4) and tetrakis (triphenylphosphine) palladium (0)
(59.7 mg, 51.7 µmol) were combined in dioxane (2.2 mL) and water (550 µL). The vial was
capped and the mixture was heated in a microwave oven at 165 °C. After 20 min reaction was
poured into a saturated sodium bicarbonate solution (50 mL) and extracted with dichloromethane
(3x 50 mL). The organic layers were dried over MgSO4 and concentrated. The residue was
purified by column chromatography (silica gel, 0% to 100% EtOAc in heptane) to afford 92.3
mg of 17 (51% yield) as a light yellow solid. Rf: 0.17 (EtOAc); 1H NMR (600 MHz, CDCl3): δ
7.74 (s, 1H), 7.60 (s, 1H), 7.45ꢀ7.42 (m, 2H), 7.26 ꢀ 7.22 (m, 2H), 6.26 (s, 1H), 4.05 (s, 2H), 3.95
ACS Paragon Plus Environment