Circumventing seizure activity in a series of G protein coupled receptor 119 (GPR119) agonists

Article abstract of DOI:10.1021/jm5011012

Agonism of GPR119 is viewed as a potential therapeutic approach for the treatment of type II diabetes and other elements of metabolic syndrome. During progression of a previously disclosed candidate 1 through mice toxicity studies, we observed tonic-cloni

Full text of DOI:10.1021/jm5011012

Subscriber access provided by KING MONGKUT'S UNIVERSITY OF TECHNOLOGY THONBURI (UniNet)
Article
Circumventing Seizure Activity in a Series of G
Protein Coupled Receptor 119 (GPR119) Agonists.
James S. Scott, Suzanne S Bowker, Katy J Brocklehurst, Hayley S Brown, David S Clarke, Alison Easter,
Anne Ertan, Kristin Goldberg, Julian A Hudson, Stefan Kavanagh, David Laber, Andrew G. Leach, Philip A.
MacFaul, Elizabeth A. Martin, Darren McKerrecher, Paul Schofield, Per H Svensson, and Joanne Teague
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm5011012 • Publication Date (Web): 06 Oct 2014
Downloaded from http://pubs.acs.org on October 17, 2014
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 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Circumventing Seizure Activity in a Series of G Protein
Coupled Receptor 119 (GPR119) Agonists.
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
James S. Scott,*a Suzanne S. Bowker,a Katy J. Brocklehurst,a Hayley S. Brown,a David S. Clarke,a Alison
Easter,a Anne Ertan,c Kristin Goldberg,a Julian A. Hudson,a Stefan Kavanagh,a David Laber,a Andrew G.
Leach,a Philip A. MacFaul,a Elizabeth A. Martin,a Darren McKerrecher,a Paul Schofield,a Per H.
Svensson,b Joanne Teague.a
Innovative Medicines Unit, AstraZeneca
aMereside, Alderley Park, Macclesfield, Cheshire, SK10 4TG, UK
bPharmaceutical Development, AstraZeneca R&D, Sꢀ151 85 Södertälje, Sweden
cCurrent Address: Arrhenius Laboratory, Stockholm University, Department of Materials and
Environmental Chemistry, Svante Arrheniusvägen 16C, SEꢀ106 91 Stockholm, Sweden
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 43
1
2
3
4
5
Abstract
6
7
8
9
Agonism of GPR119 is viewed as a potential therapeutic approach for the treatment of type II diabetes
and other elements of the metabolic syndrome. During progression of a previously disclosed candidate 1
through mice toxicity studies, we observed tonicꢀclonic convulsions in several mice at high doses. An in
vitro hippocampal brain slice assay was used to assess the seizure liability of subsequent compounds
leading to the identification of an aryl sulfone as a replacement for the 3ꢀcyano pyridyl group. Subsequent
optimization to improve the overall profile, specifically with regard to hERG activity, led to alkyl sulfone
16. This compound did not cause tonicꢀclonic convulsions in mice, had a good pharmacokinetic profile
and displayed in vivo efficacy in murine models. Importantly, it was shown to be effective in wildꢀtype
(WT) but not GPR119 knockꢀout (KO) animals, consistent with the pharmacology observed being due to
agonism of GPR119.
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
ACS Paragon Plus Environment

Page 3 of 43
Journal of Medicinal Chemistry
1
2
Introduction
3
4
5
6
7
8
9
G Protein coupled receptor 119 (GPR119) is a class A type receptor that is predominantly expressed in
the islets of Langerhans in the pancreas and the enteroendocrine cells of the intestine.1 Endogenous
natural agonists of this receptor, such as oleoylethanolamide (OEA)2 and Nꢀoleoyldopamine (OLDA),3
have been identified and their biological effects studied. Agonism of GPR119 has been shown to increase
the secretion of incretins from Lꢀcells (e.g. glucagonꢀlike peptideꢀ1 (GLPꢀ1) and glucoseꢀdependent
insulinotropic peptide (GIP),4 and to stimulate the release of insulin from pancreatic βꢀcells.5
Consequently, agonism of GPR119 has been viewed as having considerable potential in the treatment of
diabetes.6,7
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
Following an initial disclosure by researchers at Arena that a small molecule agonist was capable of
controlling glucose excursions in preclinical animal models,8 a large volume of patent applications and
journal articles have been published disclosing various chemotypes that act as GPR119 agonists. To date,
structures of GPR119 agonists have been published by groups including Amgen,9 Astellas,10 Boehringer
Ingelheim,11 GlaxoSmithKline,12 HoffmanꢀLa Roche,13 Kangwon National University,14 Pfizer,15
Merck,16 Sanwa Kagaku Kenkyusho,17 Shanghai Institute of Materia Medica18 and Takeda.19 Several
GPR119 agonists have now been progressed to the clinic (Figure 1) with published clinical data from
APDꢀ597 (JNJꢀ38431055)8c (Arena / Johnson & Johnson) demonstrating encouraging increases in GLPꢀ1,
GIP and peptide YY levels. However, reductions in glucose excursion and increased insulin secretion
were not significant.20 Clinical data from GSKꢀ1292263 (GlaxoSmithKline) showed a trend towards
improvement in insulin sensitivity in healthy subjects, but no reductions in plasma glucose concentrations
relative to placebo when tested in type 2 diabetics.21 An increase in fasting plasma highꢀdensity
lipoprotein cholesterol was observed, together with a reduction in lowꢀdensity lipoprotein cholesterol and
triglycerides.
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 43
1
2
3
O
H
O
N
O
N
O
N
N
N
O
N
O
S
4
5
6
O
N
N
N
S
O
O
O
GSK-1292263
APD-597
7
8
Figure 1. Structures of selected GPR119 agonists with reported clinical data.
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
Our own efforts to identify novel GPR119 agonists were initially complicated by offꢀtarget effects
observed in mouse in vivo sudies.22a We have subsequently reported on a structurally distinct chemical
series that displayed no activity in GPR119 knockꢀout mice,22b our efforts to improve aqueous
solubility22c and our observations on species differences between mouse and human pharmacology.22d We
herein report our findings related to the development of compounds from this series and our resulting
optimization efforts.
Results & Discussion
Upon progression of a previously disclosed lead compound (1)22b into mouse toxicity studies, we
observed tonicꢀclonic convulsions in two out of six mice which were subsequently terminated for welfare
reasons. The convulsions were observed on the tenth day of the study dosing at 300 mg/kg and exposure
levels obtained from one of the premature decedents was 1.0 µM (Cmaxfree) at the time of termination on
day 10. This alerted us to the fact that our compounds potentially carried a seizure liability, a fact we felt
was unacceptable for a high quality compound intended for the treatment of diabetes. We therefore
sought to identify a method for identifying this liability without recourse to large scale toxicity studies
and identified an in vitro hippocampal brain slice model as a potential way to assess compounds.23 This
involved recording the electrophysiology of evoked synaptic activity of a mouse hippocampal brain slice
using extracellular electrodes positioned in the CA1 cell body layer in the presence of compound.
Increases in population spikes, quantified as areas above and below the 0 mV baseline, were associated
with an increased risk of seizure activity. Results from the screening of an initial set of four compounds
are shown in Figure 2. Compound 1, which had observations of tonicꢀclonic convulsions in the mouse
toxicity study, showed changes in peak shape and quantifiable increases in peak area upon increasing
dose, characteristic of a seizureogenic compound and was classified as positive in this assay. Likewise
4
ACS Paragon Plus Environment

Page 5 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
compounds 2 and 3 were flagged as positive, seizureogenic compounds in this assay. In contrast,
compound 4 showed no discernable activity and was classified as unlikely to be seizureogenic (negative)
in this assay. The common structural motif for compounds 1ꢀ3 was the 3ꢀcyano pyridyl moiety with
compounds 3 and 4 representing a molecular matched pair of compounds where the 3ꢀcyano pyridyl had
been replaced with a 2ꢀfluoro arylsulfone.
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. Results from brainꢀslice assay showing the profile for compound 1, the mean dose response for
compound 1 and the structures of compounds tested (1ꢀ4). Statistical significance is denoted as: * p <
0.05, ** p < 0.01, *** p < 0.001.
N
N
N
N
N
N
O
NC
NC
O
N
N
N
N
O
O
N
N
F3C
N
N
1 (+ve seizurogenic)
2 (+ve seizurogenic)
N
N
N
F
N
N
N
NC
N
N
N
O
O
S
N
O
O
F3C
O
O
F3C
N
N
N
4 (-ve seizurogenic)
3 (+ve seizurogenic)
We concluded that a pragmatic way to mimimise the seizureogenic risk would be to avoid
compounds with the 3ꢀcyano pyridyl motif and seek to reꢀoptimise this portion of the molecule and screen
promising candidates in the in vitro brain slice model. To this end, we embarked on an optimisation
campaign with the syntheses of the compounds described below.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 43
1
2
Synthesis.
3
4
5
6
7
8
9
For variation of the left hand side of the molecule (Scheme 1), routes were developed starting from the
previously described22b Boc protected piperazine pyrimidinol 25. Alkylation under basic conditions
afforded the benzyl ethers 26. Removal of the Boc group with hydrochloric acid gave the piperazines 27
which could be converted to the 3ꢀ(trifluoromethyl)oxetane carbamates 20 – 22 through treatment with
the corresponding perfluorophenyl carbonates. Alternatively, treatment of 27 with cyanogen bromide
gave the versatile intermediate cyanamides 28. Treatment with the appropriate amidoximes in the
presence of zinc chloride, followed by ring closure under acidic conditions afforded the 1,2,4ꢀoxadiazoles
with 3ꢀsubstitution of either CF3 (15, 16) or iPr (23, 24). Alternatively, the cyanamides 28 could be
treated with sodium azide to form the tetrazoles 29. Treatment with trifluoroacetic anhydride following by
heating effected a conversion to the 1,3,4ꢀoxadiazoles 4 and 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
Scheme 1a
N
O
N
N
N
R1
R1
R1
N
BocN
N
23 (R1=F, R2=CH2SO2Me)
24 (R1=F, R2=SO2Me)
OH
N
N
O
N
R2
R2
R2
25
f
a
N
R1
b
N
N
N
N
R1
R1
N
N
N
N
g
R1
d
BocN
N
O
HN
N
N
N
N
N
N
N
h
O
O
28
O
N
N
R2
N
R2
R2
R2
H
27
29
26
c
e
O
CF3
O
N
N
N
N
N
N
N
R1
N
N
N
N
N
N
O
O
N
O
O
O
O
F3C
N
F3C
R2
20 (R1=F, R2=CH2SO2Me)
21 (R1=F, R2=OSO2Me)
22 (R1=F, R2=SO2Me)
4
(R1=F, R2=SO2Me)
11 (R1=F, R2=CH2SO2Me)
15 (R1=F, R2=SO2Me)
16 (R1=F, R2=CH2SO2Me)
o
a Reagents and Conditions: (a) ArCH2Br, Cs2CO3, CH3CN, 20 C, 3 d, 29ꢀ81%; (b) HCl, CH2Cl2, 25
o
oC, 16 h 76ꢀ100%; (c) (3ꢀ(CF3)oxetanꢀ3ꢀyl)OCOOC6F5, NEt3, CHCl3, 25 C, 18 h, 67ꢀ94%; (d) CNBr,
NaHCO3, CH2Cl2/H2O, 25 oC, 2 h, 87ꢀ89%; (e) (i) CF3C(=NOH)NH2, ZnCl2, EtOAc/THF, 20 oC, 4 h; (ii)
o
i
o
conc. HCl, EtOH, 100 C, 18 h, 34ꢀ55%; (f) (i) PrC(=NOH)NH2, ZnCl2, EtOAc/THF, 20 C, 4 h; (ii)
o
o
conc. HCl, EtOH, 100 C, 16 h, 19ꢀ85%; (g) NaN3, NEt3.HCl, toluene, 80 C, 3 d, 87%ꢀ100%; (h)
(CF3CO)2O, (iPr)2NEt, chlorobenzene, 130 oC, 2 h, 61ꢀ88%.
6
ACS Paragon Plus Environment

Page 7 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
For exploration of the right hand side (Scheme 2), complementary routes were established that allowed
construction of the heterocycles prior to alkylation. Boc protected piperazine pyrimidinol 25 was treated
with acid to give the piperazine 30 and then with cyanogen bromide to give cyanamide 31. Both the 1,3,4ꢀ
oxadazole 32 and the 1,2,4ꢀoxadazole 33 could be constructed using analogous chemistry to that
described in Scheme 1, notably without requirement for protection of the phenol. Subsequent alkylations
with the requisite benzyl bromides (7 – 9), mesylates (10ꢀ14, 18) or chlorides (19) afforded the final
compounds. Alkylation with 2ꢀcyanoꢀ4ꢀbromoꢀbenzylbromide afforded intermediate 34. The aryl
bromide could be transformed into either a sulfone or cyano group to afford compounds 5 and 6
respectively.
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
Scheme 2a
d,e
N
O
N
N
N
O
c
R1
N
N
R1
N
N
N
N
N
OH
N
+
X
O
F3C
R2
F3C
R2
32
e (X=OMs):
d (X=Br):
R2=Br
R2=CN
NC
R2=Br
R2=SO2Me
g
34 (R1=CN, R2=Br)
f
10 (R1=F, R2=1-tetrazole)
12 (R1=F, R2=OSO2Me)
N
N
7 (R1=H, R2=SO2Me)
R2
X
N
N
OH
8 (R1=H, R2=CH2SO2Me)
9 (R1=F, R2=CN)
13 (R1=F, R2=CH2SOMe)
14 (R1=F, R2=CH2SO(NH)Me)
5 (R1=CN, Y=SO2Me)
6 (R1=CN, Y=CN)
X= Boc 25
X= H 30
X= CN 31
a
b
O
N
O
N
N
R1
N
R1
h
N
j or k
N
N
N
N
N
+
O
OH
Y
N
N
F3C
R2
F3C
R2
33
18 (R1=F, R2=CH2SO(=NH)Me)
19 (R1=F, R2=N=S(=O)Me2)
k (Y=Cl)
j (Y=OMs)
a
o
Reagents and Conditions: (a) HCl, 1,4ꢀdioxane/CH2Cl2, 20 C, 90 min 98%; (b) CNBr, NaHCO3,
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
CH2Cl2/H2O, 25 oC, 1 h, 59%; (c) (i) NaN3, NEt3.HCl, toluene, 80 C, 18 h, 87%ꢀ100%; (ii) (CF3CO)2O,
(iPr)2NEt, toluene, 250 C, 4 d, 19%; (d) ArCH2Br, K2CO3, DMF, 25 C, 16 h, 47ꢀ89%; (e) ArCH2OMs,
o
o
K2CO3, DMF, 25 oC, 16 h, 12ꢀ32%; (f) CF3SO3Cu, (MeNHCH2)2, DMSO, CH3SO2Na, 120 oC, 2 h, 61%;
o
(g) Zn(CN)2, Pd(dba)2, Xantphos, DMF, 130 C, 2 h, ꢀW, 78%; (h) CF3C(=NOH)NH2, ZnCl2,
o
o
EtOAc/THF, 20 C, 24 h; (ii) conc. HCl, EtOH, 110 C, 18 h, 58%; (j) (i) ArCH2OMs, K2CO3, CH3CN,
75 oC, 4 h; (ii) K2CO3, MeOH, 50 oC, 2 h; 19%; (k) ArCH2Cl, K2CO3, DMF, 100 oC, 35 h, 16%.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 43
1
2
3
4
5
6
7
8
Other examples, such as sulfoxide 17, were synthesised by functional group manipulation of the benzyl
substituent whilst attached to the core (Scheme 3). Previously described bromide 3522b was lithiated and
quenched with DMF to form the aldehyde 36 then reduced using sodium borohydride to the primary
alcohol 37. Formation of the mesylate allowed introduction of a thiomethyl group under ambient
conditions to give sulfide 38. Monoꢀoxidation to the sulfoxide 39 was achieved using Oxone® at 0 oC and
a short reaction time of 15 minutes. The Boc group was then removed under acidic conditions and the
piperazine converted to cyanamide 40 in good yield. Subsequent treatment with trifluoromethyl
amidoxime in the presence of zinc chloride and heating under acid conditions afforded the sulfoxide 17
in low yield.
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
Scheme 3a
N
N
F
c
N
N
N
N
F
F
e
f
O
N
N
N
F
BocN
N
N
O
BocN
N
N
N
N
O
O
N
N
O
R2
F3C
X
S
O
S
O
R2 = Br 35
R2 = C(=O)H 36
R2 = CH2OH 37
40
17
X = SMe 38
X = S(=O)Me 39
a
b
d
n
a Reagents and Conditions: (a) BuLi, THF, ꢀ78 °C, DMF, 1 h, 83%; (b) NaBH4, EtOH, rt, 3 h, 91%; (c)
(i) MsCl, CH2Cl2, rt, 18 h; (ii) MeSNa, DMF, rt, 18 h, 42%; (d) Oxone®, MeOH/MeCN/H2O, 0 °C 15
o
min, 100%; (e) (i) TFA, CH2Cl2, 20 °C, 3 h; (ii) CNBr, NaHCO3, CH2Cl2/H2O, 0 C, rt, 2 h, 88%; (f) (i)
CF3C(=NOH)NH2, ZnCl2, EtOAc/THF, rt, 2 h; (ii) conc. HCl, EtOH, reflux, 16 h, 16%.
Medicinal Chemistry
The results of our strategy of replacing the 3ꢀcyanopyridyl group in 3 is shown in Table 1. Replacement
with a 3ꢀfluoro sulfone group gave compound 4 which had broadly similar potency when tested on HEK
cells transfected to overexpress human GPR119 receptors, albeit with reduced intrinsic activity (IA). In
the analogous mouse GPR119 HEK system, an improvement in potency was observed (EC50 = 144 nM,
IA 93%).24 In line with our previous observations, this compound was much less soluble and somewhat
surprisingly showed increased inhibition of the hERG ion channel tail current using a plateꢀbased planar
patch clamp system (IonWorksTM; IC50 = 6.9 ꢀM).25 The 3ꢀcyano analogue 5 lost potency against the
8
ACS Paragon Plus Environment

Page 9 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
mouse receptor and showed no benefit in terms of hERG inhibition. The corresponding diꢀcyano
compound 6, where the sulfone was replaced with a cyano group, showed enhanced potency in human
and mouse however the solubility remained low and this compound suffered from high hERG inhibition
(IC50 = 3.6 ꢀM). The desꢀfluoro sulfone 7 had reduced potency relative to the 3ꢀfluoro molecular matched
pair 4 in line with previously reported SAR. Extending the sulfone out by insertion of a methylene group
gave compound 8 which had a similar profile to compound 7 in terms of potency, lipophilicity and
solubility but showed no detectable inhibition of the hERG ion channel.26 Compound 9, the cyano
analogue of sulfone 4, showed similar potency but higher lipophilicity (logD7.4 = 3.8) and poor aqueous
solubility.27 The Nꢀtetrazole 10 was more potent in both human and mouse but showed no improvement
in terms of solubility or hERG. The methylene extended sulfone 11 was of comparable potency to the aryl
sulfone 4 and again showed no discernable inhibition of the hERG ion channel (c.f. matched pair 7 and
8). This observation prompted further exploration around this area with replacement of the methylene
with an oxygen atom leading to aryl mesylate 12. This compound showed increased potency, albeit at the
expense of increased lipophilicity,28 and some weak hERG activity (IC50 = 17 ꢀM). The sulfoxide 13 was
less potent, but less lipophilic and showed no hERG inhibition. Notably the aqueous solubility improved
significantly relative to the sulfone although, perhaps unsurprisingly, metabolic instability increased as
measured in rat hepatocytes (Clint = 53 µl/min/106 cells for 13, Clint = 9 µl/min/106 cells for 11). The
corresponding sulfoximine 14 was less lipophilic and showed a further increase in aqueous solubility and
no hERG inhibition. Human potency was reduced, broadly in line with the lipophilicity reduction,
however, in mouse a loss of intrinsic activity at the receptor precluded further progression of this
compound.
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
Table 1. Variation in the 1,3,4ꢀoxadiazole series.
N
N
N
N
N
N
O
O
F3C
R
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 43
1
2
3
4
5
6
7
8
human
human
mouse
GPR119
EC50
mouse
hERG
IC50
Solubility
(ꢁM)d
GPR119 EC50
(ꢁM)a
GPR119
IA (%)b
GPR119
IA (%)b
c
Cpd
R
logD7.4
(ꢁM)e
(ꢁM) a
NC
9
0.028
220
0.436
94
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
4
5
2.7
3.0
2.6
120
1.4
9.3
18
N
(0.027ꢀ0.029)
(209ꢀ230) (0.383ꢀ0.497) (90ꢀ99)
F
NC
NC
0.021
121
0.144
93
O
O
O
6.9
6.4
S
(0.019ꢀ0.023)
(114ꢀ127) (0.128ꢀ0.163) (88ꢀ89)
0.055
182
1.975
252
O
S
(0.049ꢀ0.061)
(171ꢀ192) (1.291ꢀ3.022) (143ꢀ362)
0.010
166
0.072
84
6
7
2.8
2.8
1.9
2.4
3.6
9.4
CN
O
(0.008ꢀ0.011)
(155ꢀ176) (0.056ꢀ0.091) (72ꢀ96)
0.133
119
0.297
71
O
S
(0.110ꢀ0.160)
(110ꢀ127) (0.229ꢀ0.386) (68ꢀ74)
0.150
89
0.319
95
8
S
2.6
2.3
>30
O
O
(0.131ꢀ0.172)
(86ꢀ92)
(0.269ꢀ0.377) (90ꢀ100)
F
F
F
0.015
156
0.216
90
9
3.8
3.6
<0.5
0.8
11
CN
(0.014ꢀ0.017)
(141ꢀ171) (0.154ꢀ0.303) (80ꢀ101)
0.007
170
0.058
91
N
N
N
10
5.7
N
(0.006ꢀ0.007)
(149ꢀ191) (0.050ꢀ0.067) (85ꢀ97)
0.027
100
0.249
100
11
2.9
3.8
>30
S
O
(0.025ꢀ0.029)
(95ꢀ105)
(0.227ꢀ0.274) (95ꢀ106)
O
10
ACS Paragon Plus Environment

Page 11 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
F
F
F
0.004
156
0.042
82
O
S
12
3.7
2.6
2.1
2.3
84
17
(0.003ꢀ0.004)
(143ꢀ169) (0.038ꢀ0.046) (75ꢀ88)
O
O
O
O
0.058
139
0.976
85
13
14
>30
>30
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
S
(0.051ꢀ0.066)
(132ꢀ137) (0.861ꢀ1.107) (79ꢀ92)
0.158
101
47
>5
370
S
(0.141ꢀ0.178)
(90ꢀ113)
(43ꢀ52)
NH
aEC50 data are the geometric mean of at least three independent measurements, with 68% confidence
limits in parentheses as calculated from SEM. bIntrinsic activity expressed as the percent effect compared
c
to that of the control, 50 µM oleoylethanolamide, defined as 100%. Distribution coefficient between 1ꢀ
octonol and aqueous phosphate buffer at pH 7.4. dSolubility of compounds in aqueous phospahte buffer at
o
e
pH7.4 after 24 hours at 25 C. Inhibition of the hERG ion channel using an electrophysiology
(IonWorksTM) patch clamp system.
Based on previous investigations in this area, we believed that the 1,2,4ꢀoxadiazole isomers should be
more lipophilic than their 1,3,4ꢀisomers and this potentially offered an opportunity to increase potency
(Table 2).29 Sulfone 15 was indeed more potent against human and mouse than 4, but remained potent
against hERG. By contrast, the methylene extended sulfone 16 again showed no discernable hERG
activity and had improved potency against the human and mouse isoforms relative to 11. The SAR/SPR
observed in this subꢀseries showed similar trends to those described in Table 1; the sulfoxide 17 having
similar potency and improved solubility and the sulfoximine 18 showing a reduction in potency, broadly
in line with lipophilicity, and a large increase in solubility and little or no hERG activity. An isomeric
sulfoximine 19 was less active, despite being much more lipophilic, and had poor solubility although it
too remained devoid of hERG activity.
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 43
1
2
Table 2. Variation in the 1,2,4ꢀoxadiazole series.
3
4
5
6
7
O
N
N
N
N
N
O
N
F3C
R
8
9
human
human
mouse
IA GPR119
EC50
mouse
hERG
IC50
Solubility
(ꢁM)d
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
GPR119 EC50 GPR119
(ꢁM)a (%)b
GPR119
IA (%)b
c
Cpd
R
logD7.4
(ꢁM)e
(ꢁM) a
F
0.004
89
0.030
78
O
15
16
O
3.7
<1.4
1.8
3.6
S
(0.003ꢀ0.004)
(85ꢀ93)
(0.023ꢀ0.039) (71ꢀ85)
F
F
F
F
0.011
84
0.068
71
3.6
3.4
2.8
3.4
>30
S
S
S
S
(0.010ꢀ0.012)
(79ꢀ89)
(0.059ꢀ0.079) (65ꢀ76)
O
O
0.008
108
0.106
58
17
18
19
49
28
(0.007ꢀ0.010)
(101ꢀ114) (0.094ꢀ0.120) (57ꢀ59)
O
0.091
57
0.492
62
140
2.6
>30
>30
(0.082ꢀ0.102)
(53ꢀ61)
(0.333ꢀ0.726) (45ꢀ78)
O
N
NH
1.102
27
0.389
27
(0.943ꢀ1.287)
(26ꢀ29)
(0.351ꢀ0.430) (19ꢀ34)
O
aEC50 data are the geometric mean of at least three independent measurements, with 68% confidence
limits in parentheses as calculated from SEM with the exception of the human value for 18 which is n=2.
bIntrinsic activity expressed as the percent effect compared to that of the control, 50 µM
oleoylethanolamide, defined as 100%. cDistribution coefficient between 1ꢀoctonol and aqueous phosphate
buffer at pH 7.4. dSolubility of compounds in aqueous phospahte buffer at pH7.4 after 24 hours at 25 oC.
eInhibition of the hERG ion channel using an electrophysiology (IonWorksTM) patch clamp system.
12
ACS Paragon Plus Environment

Page 13 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
A subsequent round of optimization investigated optimal combinations of groups around the central
scaffold (Table 3). Replacement of the oxadiazole with a 2ꢀtrifluoromethyloxetane carbamate identified
previously,22c gave compound 20 with similar potency to 16 but reduced lipophilicity. Unfortunately, this
did not translate to improved aqueous solubility. Aryl mesylate 21 was more potent, however the
lipophilicity increase associated with the switch from carbon to oxygen further eroded the aqueous
solubility of this compound. It is noteworthy that both compounds showed no hERG activity in contrast
to the direct linked sulfone equivalent 22, which had a similar overall profile but did show inhibition of
the hERG ion channel (IC50 = 11 ꢀM). Replacing the trifluroromethyl group of 16 with an isopropyl
substituent resulted in 23 which was less potent, in line with lower lipophilicity. No inhibition of hERG
was observed whereas the molecular matched pair 24, with the direct linked sulfone, did show activity
(IC50 = 7.4 ꢀM), again emphasizing the lack of hERG activity in the CH2SO2Me subꢀseries.
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
Table 3. Combination compounds.
N
R1
N
N
O
N
R2
human
GPR119
EC50
human
mouse
GPR119
EC50
mouse
GPR119
IA
hERG
IC50
GPR119
IA (%)b
Solubility
(ꢁM)d
c
Cpd
R1
R2
logD7.4
(ꢁM)e
(ꢁM)a
(ꢁM) a
(%)b
F
O
CF3
0.020
91
0.140
108
20
21
2.9
3.7
2.5
>30
>30
O
O
O
S
O
(0.018ꢀ0.022) (88ꢀ95)
(0.123ꢀ0.159) (99ꢀ117)
O
O
F
O
CF3
0.004
141
0.028
98
O
S
O
<0.6
O
(0.004ꢀ0.005) (133ꢀ150) (0.024ꢀ0.031) (89ꢀ107)
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 43
1
2
3
4
5
6
7
8
F
F
O
CF3
0.012
99
0.080
100
O
22
23
O
3.0
3.1
<1.8
2.4
11
O
O
S
(0.012ꢀ0.013) (93ꢀ105)
(0.073ꢀ0.087) (94ꢀ106)
O
N
0.018
93
0.160
66
N
>30
S
9
(0.015ꢀ0.022) (75ꢀ111)
(0.120ꢀ0.214) (57ꢀ75)
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
0.006
O
F
N
101
0.060
83
O
24
O
N
(0.005ꢀ0.007)
(97ꢀ105)
3.4
1.0
7.4
S
(0.050ꢀ0.072) (78ꢀ87)
aEC50 data are the geometric mean of at least three independent measurements, with 68% confidence
limits in parentheses as calculated from SEM. bIntrinsic activity expressed as the percent effect compared
c
to that of the control, 50 µM oleoylethanolamide, defined as 100%. Distribution coefficient between 1ꢀ
octonol and aqueous phosphate buffer at pH 7.4. dSolubility of compounds in aqueous phospahte buffer at
o
e
pH7.4 after 24 hours at 25 C. Inhibition of the hERG ion channel using an electrophysiology
(IonWorksTM) patch clamp system.
A plot of the human potency (pEC50 sized by % intrinsic activity) against measured logD7.4 is shown in
Figure 3A with selected compounds from the manuscript labeled.30 A well defined leading edge of
potency/lipophilicicity is apparent from this dataset. Figure 3B shows the potency increase associated
with switching from matched pairs in the 1,3,4ꢀoxadiazole isomers (Table 1) to the 1,2,4ꢀoxadiazole
isomers (Table 2) and Figure 3C shows that this is primarily driven by a consistent increase in
lipophilicity associated with this change.
14
ACS Paragon Plus Environment

Page 15 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Figure 3. A) Plot of human GPR119 pEC50 sized by intrinsic activity (%) against measured logD7.4; B)
Plot of human GPR119 pEC50 for 1,2,4ꢀ and 1,3,4ꢀoxadiazole matched pairs; Plot of measured logD7.4 for
1,2,4ꢀ and 1,3,4ꢀoxadiazole matched pairs. Selected compounds are annotated with their corresponding
manuscript numbers.
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 selected set of compounds (11, 16, 20) were selected for profiling in the in vitro brain slice model. All
of these compounds showed no discernable change in peak shape upon increasing concentration and were
therefore classified as low for seizureogenic risk. The data for compound 16 is shown in Figure 4 and, in
contrast to Figure 2, shows no increases in population spikes or peak area upon increasing compound
concentration. On the overall balance of properties, compound 16 was selected for further evaluation in a
12 day mouse toxicity study specifically designed and powered (n=8) to assess seizure liability. In
contrast to the tonicꢀclonic convulsions observed with 1, no evidence for seizures in any of the animals
was observed during the course of this study with mean free exposures of 16 achieved during the study of
0.7 µM Cmaxfree. This data, taken together with the lack of signal from the brain slice assay and
adequate safety margins achieved with this study, provided confidence to continue to progress this
compound.
Figure 4. Results from brainꢀslice assay showing the profile & mean dose response for compound 16
with aminophylline (4ꢀAP) used as positive control
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 43
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
Further profiling of 16 showed no inhibition (IC50 values > 10 µM) of five cytochrome P450 enzymes
(CYP1A2, CYP2C9, CYP2C19, CYP2D6 and CYP3A4) in a high throughput fluorescence assay. Plasma
protein binding showed less than three fold variation across species (mouse 1.2% free; rat 1.5% free; dog
o
1.6% free; human 0.6% free). The aqueous solubility at pH 7.4 as measured at 25 C on crystalline
material was low (1.8 µM), however, it was shown to be higher (38 µM) in biologically relevant media at
o
37 C.31 Permeability, as measured in an in vitro Cacoꢀ2 assay, was high with no evidence of efflux
(apical to basolateral Papp 10x10ꢀ6 cm/s at a compound concentration of 10 µM; efflux ratio 0.6).
Compound 16 showed no photolytic instability or hydrolytic instability across the pH range 1ꢀ8, although
some degredation was noted at pH 10.32 The secondary pharmacology profile was clean against a panel of
100 targets with the only hit being weak activity against the epidermal growth factor receptor (IC50 = 2.1
µM). The compound showed no discernable activity (IC50 > 30 µM) against a selection of ion channels
(hERG, NaV1.5, Ito, CaV1.2) and only very weak activity against the cardiac K+ channel IKs (IC50 = 27
µM) which was not considered as an issue. Compound 16 was nonꢀmutagenic in a twoꢀstrain Ames test
and negative in a mouse lymphoma assay.
The small molecule crystal structure of 16 shows that the compound adopts a structure broadly similar to
that of the equivalent aryl sulfones: a layer of adjacent molecules (Figure 5a) is arranged into stacks. A
16
ACS Paragon Plus Environment

Page 17 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
key difference is that the introduction of the extra methylene group has caused the molecules shape to be
stepped at one end and this in turn translates into a stepped shape of the layers that the molecules are
arranged in (Figure 5b). The layers are held together by interactions between the polarized CHs of the
pyrimidine interacting with one of the sulfone oxygens while the sulfone methyl interacts with a
pyrimidine nitrogen. All of these weak donor and acceptor features are also present in the structure of the
aryl sulfone but interact in different combinations.22b By contrast to the flatter aryl sulfone, the molecules
do not arrange themselves in a head to head fashion that allows the sulfone of one molecule to interact
with that of an adjacent one. As shown in Figure 5c, the layers are held together by ladders of interactions
involving the sulfones with the benzylic methylene and the sulfone methyl CHs both acting as weak
donors and interacting with one of the sulfone oxygens. Finally, the molecules do not stack perfectly on
top of one another (Figure 5d), this offset has been observed previously in this series22b when the methyl
group is added to the piperazine. Although the aqueous solubility of 16 remains low, these differerences
potentially contribute to improvements seen in respect to the original sulphone lead in this series.22b
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
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 43
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 5. Four views of the crystal structure of compound 16. a) shows the plane of adjacent molecules
from above. b) shows the stepped arrangement of adjacent molecules in the plane shown in a). c) shows
how the molecules stack on top of one another viewed from the side of the molecules and d) shows how
the molecules stack viewed from the end of the molecules. Distances shown are in Å.
Pharmacokinetics
Plasma pharmacokinetic parameters (IV and PO) for compound 16 were determined in C57BL6 mice
(male), Han Wistar rats (male) and beagle dogs (male and female). The compound was administered
18
ACS Paragon Plus Environment

Page 19 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
orally (PO) as a suspension in aqueous 0.1% (w/v) Pluronic F127 and intravenously (IV) as a solution in
5% DMSO:95% hydroxylpropyl βꢀcyclodextrin, at relevant pharmacological doses (2ꢀ5 mg/kg) as shown
in Table 4. The compound was characterized by low clearance (Clp), especially in dog, and moderate
volume of distribution (Vss), consistent across the three species. The in vivo clearance results were in
good accordance with the in vitro predictions based on hepatocyte data (Clint 7 and 10 µl/min/106 cells,
respectively) and plasma protein binding free levels (1.2% and 1.5%, respectively) for mouse and rat.
There was no discernible turnover in either dog or human hepatocytes (Clint <2 µl/min/106 cells) which
fitted well with the low clearance observed in dog and gave confidence that the compound would be low
clearance in humans. Bioavailability (F) and fraction absorbed (Fabs) were high for all three species which
was somewhat surprising given the low aqueous solubility, and may reflect the improved solubility
observed in biologically relevant media (vide supra).
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
Table 4. Pharmacokinetic parameters for compound 16a
Species
Clp
Vss
PO halfꢀ IV halfꢀ Fabs
F (%)
(mL/min/kg)
(L/kg)
life (h)
6.9
life (h)
6.3
Mouse
Rat
9.8
11
5.1
5.2
3.3
0.8
1.0
1.0
76
85
98
6.7
6.1
Dog
1.6
30
26
aCompounds were dosed intravenously at either 2 mg/kg (rat and dog) or 3 mg/kg (mouse) in 5%
DMSO:95% hydroxylpropyl βꢀcyclodextrin and orally as either 5 mg/kg (mouse and rat) or 3.4 mg/kg
(dog) using a 0.1% HMPC / tween suspension (mouse and rat) or 0.1% Pluronic F127 suspension (dog),
respectively at volumes of 4 mL/kg (mouse and rat) and 2 mL/kg (dog).
In order to address concerns that exposure at higher doses would be limited by solubility, oral dose
escalation studies were carried out as shown in Table 5. These results showed significant increases in
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 43
1
2
3
4
5
6
7
exposure of compound upon increasing compound dose, both in terms of maximum plasma concentration
observed (Cmax) and area under the curve (AUC), alleviating these concerns and providing confidence that
compound 16 could be evaluated in preꢀclinical toxicity studies.
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
Table 5. Exposure from rising dose pharmacokinetic studies for compound 16
b
b
Species
Dosea
Cmax
AUCc
Species
Dosea
Cmax
AUCc
Rat
Rat
Rat
Rat
Rat
Rat
4.5
7.1
12
0.8
1.1
3.4
7.5
8.0
13.7
10
Dog
5
1.5
20
11
Dog
50
98
5.7
11
31
57
3.2
42
33
Dog
5.3
78
19
106
123
247
Mouse
Mouse
Mouse
Mouse
2.5
26
100
154
3.3
35
11.8
29.0
151
456
aDose is in mg/kg from a suspension of 0.1% Pluronic F127 in fed animals. Cmax is measured in ꢀM.
cAUC = AUC0ꢀ24hr (ꢀM.h) in all studies. All data is the mean of the results from two animals; in the case
of the dog, one male and one female.
b
In vivo Efficacy
Compound 16 was profiled in vivo for its ability to control the glucose excursion in a mouse
(C57BL6/JAX) oral glucose tolerance test (OGTT).33 As shown in Figure 6a, 16 enhances glucose
disposal in a doseꢀdependant (1.5 ꢀ 15 mg/kg) manner with a minimum efficacious dose of 1.5 mg/kg. In
a separate study, the ability of compound 16 to increase the levels of total GLPꢀ1 in systemic circulation
20
ACS Paragon Plus Environment

Page 21 of 43
Journal of Medicinal Chemistry
1
2
3
4
was investigated. Total GLPꢀ1 levels were measured 30 minutes after an oral dose of the compound and
showed a doseꢀdependent increase with significant elevation at doses of 5 mg/kg (Figure 6b).
5
6
7
8
9
Figure 6. Oral administration of compound 16 dose dependently enhances glucose disposal and increases
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
GLPꢀ1 secretion in lean mice.a
aA) Doseꢀeffect of cpd 16 on blood glucose in male C57BL6/j mice. Group sizes (n) ꢀ Vehicle control
(24); Cpd 16 ꢀ 0.5mg/kg (18), 1.5mg/kg (17), 5mg/kg (18) & 15mg/kg (18). Cpd 16 was administered 30
minutes prior to an oral glucose load of 2 g/kg and glucose blood levels were monitored out to 90 minutes
Variables & covariates (Time 0 & Start) have not been logged. Percentage reduction in blood glucose
AUC calculated relative to window between Vehicle control and zero. Oneꢀsided Student t test for a
decreasing effect using pooled interꢀanimal variability (all groups included in the analysis). B) Doseꢀ
effect of cpd 16 on blood total GLPꢀ1 in male C57BL6/j mice. Group sizes (n) ꢀ Vehicle control (19);
Cpd 16 ꢀ 1.5mg/kg (20), 5mg/kg (20) & 15mg/kg (20). Cpd 16 was administered to fasted animals 30
minutes prior to terminal blood sampling for total GLPꢀ1 measurements & Fold increase in GLPꢀ1
calculated relative to vehicle control. Variables have been logged & all groups included in an ANCOVA
analysis (Start as a covariate) using Oneꢀsided Student t test for an increasing effect. Statistical
significance for glucose and GLP effects is denoted as: * p < 0.05, ** p < 0.01, *** p < 0.001.
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 43
1
2
3
4
5
6
7
8
Confirmation that agonism of GPR119 was responsible for the observed pharmacology was established
by assessing compound 16 in an OGTT with both wildꢀtype (WT) and GPR119 knockꢀout (KO) mice. At
a dose of 15 mg/kg of 16, the wildꢀtype animals showed clear evidence of enhanced glucose disposal
(22%) whereas in the knockꢀout mice, no significant effect was observed (Figure 7).
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. Compound 16 enhances glucose disposal (OGTT) in WT but not GPR119 KO mice.a
a
Cpd 16 blood glucose reduction showing GPR119 dependency in male wildꢀtype (WT) and knockꢀout
(KO) mice. Group sizes (n) ꢀ Vehicle control (WT=16, KO=16); Cpd 16 ꢀ 15mg/kg (WT=10, KO=10).
Cpd 16 was administered 30 minutes prior to an oral glucose load of 2 g/kg and glucose blood levels were
monitored out to 90 minutes Variables & covariates (Time 0 & Start) have not been logged. Percentage
reduction in blood glucose AUC calculated relative to window between Vehicle control and zero. Oneꢀ
sided Student t test for a decreasing effect using pooled interꢀanimal variability (all groups included in the
analysis but results of the other test groups have been excluded for simplicity). Statistical significance is
denoted as: * p < 0.05, ** p < 0.01, *** p < 0.001.
With evidence that compound 16 was displaying robust, onꢀtarget efficacy at 5 mg/kg, we examined the
pharmocology of compound 42 in an OGTT in combination with the dipeptidyl peptidaseꢀ4 (DPPIV)
inhibitor sitagliptin at a dose of 0.3 mg/kg. DPPIV inhibitors achieve their efficacy through preventing
the inactivation of GLPꢀ1 thus prolonging the effect of this incretin hormone.34 The combination of a
22
ACS Paragon Plus Environment

Page 23 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
GPR119 agonist with a DPPIV inhibitor is therefore attractive as a therapeutic paradigm as it should both
elevate and prolong the levels of GLPꢀ1. In an OGTT experiment, the combination of 16 with sitagliptin
showed a statistically significant benefit over each monotherapy (Figure 8).
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 8. Combination of compound 16 and sitagliptin shows benefit in glucose disposal over
monotherapy.a
a Glucoseꢀlowering effect of GPR119 agonist (cpd 16) coꢀadministered with DPPIV inhibitor (sitagliptin)
in an Oral Glucose Tolerance Test (OGTT) in male C57BL6/Jax mice. Group sizes (n) ꢀ Vehicle control
(8), Cpd 16 ꢀ 5mg/kg (11), sitagliptin 0.3 mg/kg (11) & combination (11). Control or test compound was
administered 30 minutes prior to a glucose load of 2 g/kg and glucose levels monitored out to 90 minutes.
Glucose response AUC’s (variable baseline) for test groups were compared to vehicle control by
ANCOVA analysis (Time 0 covariate). Glucose AUC in the combination therapy group was compared to
each monotherapy AUC. Statistical significance is denoted as: * p < 0.05, ** p < 0.01, *** p < 0.001.
On the basis of the attractive profile and pharmacokinetics together with the in vivo results showing clear,
on target efficacy at low doses and an improved safety profile relative to 1, compound 16 was selected for
further development, the results of which will be reported in due course.
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 43
1
2
Conclusion
3
4
5
6
7
8
9
Upon observing tonicꢀclonic convulsions in toxicity studies in mice at high doses with a previously
disclosed candidate 1, we established an in vitro brain slice assay to assess the seizure liability of
subsequent compounds. This led to the identification of an aryl sulfone 4 as a replacement for the 3ꢀcyano
pyridyl group that avoided this liability. Subsequent optimisation to improve the overall profile,
specifically with regard to hERG activity, led to alkyl sulfone 16. This compound did not cause tonicꢀ
clonic convulsions in mice, had a good pharmacokinetic profile and displayed in vivo efficacy in murine
models. Importantly, similar efficacy was not seen in GPR119 knockꢀout mice confirming that the
pharmacology observed was through agonism of GPR119.
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
General Procedures. All solvents and chemicals used were reagent grade. Flash column chromatography
was carried out using prepacked silica cartridges (from 4 g up to 330 g) from Redisep, Biotage, or
1
Crawford and eluted using an Isco Companion system. H NMR were recorded on a Bruker Avance
DPX400 (400 MHz) and were determined in CDCl3 or DMSOꢀd6. 13C NMR spectra were recorded at 101
or 175 MHz. Chemical shifts are reported in ppm relative to tetramethylsilane (TMS) (0.00 ppm) or
solvent peaks as the internal reference and coupling constant (J) values are reported in Hertz (Hz).
Splitting patterns are indicated as follows: s, singlet; d, doublet; t, triplet; m, multiplet. Merck precoated
thin layer chromatography (TLC) plates (silica gel 60 F254, 0.25 mm, art. 5715) were used for TLC
analysis. The purity of compounds submitted for screening was >95% as determined by UV analysis of
liquid chromatographyꢀmass spectroscopy (LCꢀMS) chromatograms at 254 nM and substantiated using
the TAC (Total Absorption Chromatogram). Further support for the purity statement was provided using
the MS TIC (Total Ion Current) trace in ESI +ve and –ve ion modes, HRMS and NMR analysis.
Solutions were dried over anhydrous magnesium sulfate, and solvent was removed by rotary evaporation
under reduced pressure. Melting points were determined on a Mettler FP62 automatic melting point
24
ACS Paragon Plus Environment

Page 25 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
apparatus and are uncorrected. The synthesis of selected examples are described below, for complete
details on all intermediates and final compounds please refer to the supplementary material. All
experimental activities involving animals were carried out in accordance with AstraZeneca animal
welfare protocols which are consistent with The American Chemical Society Publications rules and
ethical guidelines.
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
5-{[2-fluoro-4-(methylsulfonyl)benzyl]oxy}-2-{(2R)-2-methyl-4-[5-(trifluoromethyl)-1,3,4-
oxadiazol-2-yl]piperazin-1-yl}pyrimidine (4). To a stirred suspension of (R)ꢀ5ꢀ(2ꢀfluoroꢀ4ꢀ
(methylsulfonyl)benzyloxy)ꢀ2ꢀ(2ꢀmethylꢀ4ꢀ(1Hꢀtetrazolꢀ5ꢀyl)piperazinꢀ1ꢀyl)pyrimidine (3.68 g, 8.21
mmol) and Nꢀethyldiisopropylamine (4.3 mL, 24.6 mmol) in chlorobenzene (75 mL) under an atmosphere
of nitrogen was added trifluoroacetic anhydride (2.3 mL, 16.4 mmol) at 0 °C. The mixture was heated at
130 °C for 16 hours, cooled to ambient temperature, partitioned between ethyl acetate (425 mL) and
water (175 mL), the ethyl acetate layer washed with saturated sodium hydrogen carbonate solution, brine,
dried (MgSO4) and evaporated in vacuo to a residue which was chromatographed on silica with 50%
ethyl acetate in isohexane as eluant, then on neutral alumina with 25% ethyl acetate in isohexane as eluant
to give a solid which was crystallised from ethyl acetate / isohexane to give (R)ꢀ2ꢀ(4ꢀ(5ꢀ(2ꢀfluoroꢀ4ꢀ
(methylsulfonyl)benzyloxy)pyrimidinꢀ2ꢀyl)ꢀ3ꢀmethylpiperazinꢀ1ꢀyl)ꢀ5ꢀ(trifluoromethyl)ꢀ1,3,4ꢀoxadiazole
1
(2.60 g, 61%). m.p. 145 – 146 °C; H NMR (400 MHz, CDCl3) 1.25 (3H, d, J = 6.7 Hz), 3.08 (3H, s),
3.22 ꢀ 3.38 (2H, m), 3.46 (1H, dd, J = 4.0, 12.9 Hz), 3.89 (1H, dt, J = 1.7, 12.9 Hz), 4.02 ꢀ 4.15 (1H, m),
4.50 ꢀ 4.63 (1H, m), 4.95 ꢀ 5.03 (1H, m), 5.17 (2H, s), 7.70 (1H, dd, J = 1.5, 8.9 Hz), 7.73 ꢀ 7.84 (2H, m),
8.18 (2H, s); 13C NMR (176 MHz, DMSO) 13.7, 37.3, 43.1, 45.3, 45.6, 49.6, 65.0, 114.3 (d, J = 24.4 Hz),
116.1 (q, J = 269 Hz), 123.1, 129.5 (d, J = 14.7 Hz), 131.2 (d, J = 3.8 Hz), 142.5 (d, J = 6.7 Hz), 145.0,
146.4, 147.4 (q, J = 43.0 Hz), 156.7, 159.6 (d, J = 251 Hz), 165.2; HRMS (EI) for C20H21O4N6SF4
(MH+); calcd, 517.1276; found, 517.1274.
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 43
1
2
5-(methylsulfonyl)-2-{[(2-{(2R)-2-methyl-4-[5-(trifluoromethyl)-1,3,4-oxadiazol-2-yl]piperazin-
3
4
5
6
7
8
9
1-yl}pyrimidin-5-yl)oxy]methyl}benzonitrile (5). To a stirred solution of (R)ꢀ5ꢀbromoꢀ2ꢀ((2ꢀ(2ꢀmethylꢀ
4ꢀ(5ꢀ(trifluoromethyl)ꢀ1,3,4ꢀoxadiazolꢀ2ꢀyl)piperazinꢀ1ꢀyl)pyrimidinꢀ5ꢀyloxy)methyl)benzonitrile (230
mg, 0.44 mmol), (trifluoromethylsulfonyloxy)copper (18.65 mg, 0.09 mmol) and N,Nꢀdimethylethaneꢀ
1,2ꢀdiamine (15.47 mg, 0.18 mmol) in DMSO (2.9 mL) at ambient temperature under an atmosphere of
nitrogen was added sodium methanesulfinate (263 mg, 2.19 mmol). The mixture was heated at 120 °C for
2 hours, cooled to ambient temperature, partitioned between ethyl acetate (60 mL) and water (45 mL),
washed with water (1 x 20 mL), brine (2 x 20 mL), dried (MgSO4) and evaporated in vacuo to a residue
which was chromatographed on silica with 50% ꢀ 100% ethyl acetate in isohexane as eluant to give (R)ꢀ2ꢀ
((2ꢀ(2ꢀmethylꢀ4ꢀ(5ꢀ(trifluoromethyl)ꢀ1,3,4ꢀoxadiazolꢀ2ꢀyl)piperazinꢀ1ꢀyl)pyrimidinꢀ5ꢀyloxy)methyl)ꢀ5ꢀ
(methylsulfonyl)benzonitrile (140 mg, 61%). m.p. 65ꢀ66 °C; 1H NMR (400 MHz, CDCl3) 1.19 (3H, d, J =
6.8 Hz), 3.05 (3H, s), 3.17 ꢀ 3.32 (2H, m), 3.47 (1H, dd, J = 4.0, 12.9 Hz), 3.91 (1H, dt, J = 1.7, 12.9 Hz),
4.04 ꢀ 4.17 (1H, m), 4.53 ꢀ4.64 (1H, m), 4.96 ꢀ 5.06 (1H, m), 5.31 (2H, s), 7.95 (1H, d, J = 8.1 Hz), 8.20 ꢀ
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
13
8.25 (3H, m), 8.30 (1H, d, J = 1.8 Hz); C NMR (176 MHz, DMSO) 13.7, 37.3, 43.0, 45.3, 45.6, 49.6,
69.0, 112.2, 115.8, 116.1 (q, J = 267 Hz), 130.2, 131.5, 132.0, 141.4, 145.0, 146.5, 147.4 (q, J = 43.2 Hz),
156.8, 165.2; HRMS (EI) for C21H21O4N7SF3 (MH+); calcd, 524.1322; found, 524.1324.
4-{[(2-{(2R)-2-methyl-4-[5-(trifluoromethyl)-1,3,4-oxadiazol-2-yl]piperazin-1-yl}pyrimidin-5-
yl)oxy]methyl}benzene-1,3-dicarbonitrile (6). To a mixture of (R)ꢀ5ꢀbromoꢀ2ꢀ((2ꢀ(2ꢀmethylꢀ4ꢀ(5ꢀ
(trifluoromethyl)ꢀ1,3,4ꢀoxadiazolꢀ2ꢀyl)piperazinꢀ1ꢀyl)pyrimidinꢀ5ꢀyloxy)methyl)benzonitrile (230 mg,
0.44 mmol), zinc cyanide (41.2 mg, 0.35 mmol), tris(dibenzylideneacetone)dipalladium(0) (16.1 mg, 0.02
mmol) and 4,5ꢀbis(diphenylphosphino)ꢀ9,9ꢀdimethylxanthene (20.3 mg, 0.04 mmol) in a microwave vial
under an atmosphere of nitrogen was added degassed DMF (1.83 mL). The stirred mixture was heated at
130 °C in a Biotage Initiator Microwave oven for 2 hours, cooled to ambient temperature, poured onto
water (28 mL) and extracted with ethyl acetate (3 x 25 mL). The combined ethyl acetate extracts washed
with brine, dried (MgSO4) and evaporated in vacuo to a residue which was chromatographed on silica
with 30% ꢀ 50% ethyl acetate in isohexane as eluant to give (R)ꢀ4ꢀ((2ꢀ(2ꢀmethylꢀ4ꢀ(5ꢀ(trifluoromethyl)ꢀ
26
ACS Paragon Plus Environment

Page 27 of 43
Journal of Medicinal Chemistry
1
2
1,3,4ꢀoxadiazolꢀ2ꢀyl)piperazinꢀ1ꢀyl)pyrimidinꢀ5ꢀyloxy)methyl)isophthalonitrile (160 mg, 78%). m.p. 123
3
4
5
1
– 124 °C; H NMR (400 MHz, CDCl3) 1.26 (3H, d, J = 6.7 Hz), 3.25 ꢀ 3.40 (2H, m), 3.40 (1H, dd, J =
6
7
8
9
4.0, 12.9 Hz), 3.91 (1H, dt, J = 1.7, 12.9 Hz), 4.04 ꢀ 4.13 (1H, m), 4.53 ꢀ 4.64 (1H, m), 4.97 ꢀ 5.06 (1H,
m), 5.27 (2H, s), 7.87 (1H, d, J = 8.1 Hz), 7.90 (1H, dd, J = 1.5, 8.1 Hz), 7.95 (1H, d, J = 1.5 Hz), 8.22
(2H, s); 13C NMR (176 MHz, DMSO) 13.7, 37.3, 45.3, 45.6, 49.6, 69.1, 112.2, 112.4, 115.4, 116.4 (q, J =
266 Hz), 130.1, 136.9, 137.1, 144.7, 145.0, 146.6, 147.4 (q, J = 43.0 Hz), 156.8, 165.2; HRMS (EI) for
C21H18O2N8F4 (MH+); calcd, 471.1499; found, 471.1498.
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-{[4-(methylsulfonyl)benzyl]oxy}-2-{(2R)-2-methyl-4-[5-(trifluoromethyl)-1,3,4-oxadiazol-2-
yl]piperazin-1-yl}pyrimidine (7). To a stirred suspension of (R)ꢀ2ꢀ(2ꢀmethylꢀ4ꢀ(5ꢀ(trifluoromethyl)ꢀ
1,3,4ꢀoxadiazolꢀ2ꢀyl)piperazinꢀ1ꢀyl)pyrimidinꢀ5ꢀol (260 mg, 0.79 mmol) and 1ꢀ(bromomethyl)ꢀ4ꢀ
(methylsulfonyl)benzene (196 mg, 0.79 mmol) in DMF (2.45 mL) at ambient temperature was added
potassium carbonate (163 mg, 1.18 mmol). The mixture was stirred at ambient temperature for 16 hours,
the mixture partitioned between water (40 mL) and ethyl acetate (50 mL), the ethyl acetate layer washed
with brine, dried (MgSO4) and evaporated in vacuo to a residue which was chromatographed on silica
with 50% ethyl acetate in isohexane as eluant to give a solid which was crystallised from ethyl acetate /
isohexane to give (R)ꢀ2ꢀ(3ꢀmethylꢀ4ꢀ(5ꢀ(4ꢀ(methylsulfonyl)benzyloxy)pyrimidinꢀ2ꢀyl)piperazinꢀ1ꢀyl)ꢀ5ꢀ
1
(trifluoromethyl)ꢀ1,3,4ꢀoxadiazole (300 mg, 76%). m.p. 152 – 153 °C; H NMR (400 MHz, CDCl3) 1.26
(3H, d, J = 6.7 Hz), 3.07 (3H, s), 3.25 ꢀ 3.39 (2H, m), 3.47 (1H, dd, J = 4.0, 12.9 Hz), 3.90 (1H, dt, J =
1.7, 12.9 Hz), 4.04 ꢀ 4.15 (1H, m), 4.52 ꢀ 4.62 (1H, m), 4.96 ꢀ 5.04 (1H, m), 5.14 (2H, s), 7.63 (2H, d, J =
7.3 Hz), 7.99 (2H, d, J = 7.3 Hz), 8.16 (2H, s); 13C NMR (176 MHz, DMSO) 13.6, 37.2, 43.4, 45.3, 45.6,
49.6, 70.0, 116.1 (q, J = 269 Hz), 127.1, 128.1, 140.3, 142.5, 145.1, 146.2, 147.3 (q J = 43.0 Hz), 156.6,
165.2; HRMS (EI) for C20H22O4N6SF3 (MH+); calcd, 499.1370; found, 499.1368.
5-{[2-fluoro-4-(methylsulfonyl)benzyl]oxy}-2-{(2R)-2-methyl-4-[3-(trifluoromethyl)-1,2,4-
oxadiazol-5-yl]piperazin-1-yl}pyrimidine (15). Zinc chloride (1M in diethyl ether) (0.987 mL, 0.99
mmol)
was
added
to
(R)ꢀ4ꢀ(5ꢀ(2ꢀfluoroꢀ4ꢀ(methylsulfonyl)benzyloxy)pyrimidinꢀ2ꢀyl)ꢀ3ꢀ
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 43
1
2
3
4
5
6
7
8
methylpiperazineꢀ1ꢀcarbonitrile (400 mg, 0.99 mmol) and 2,2,2ꢀtrifluoroꢀN'ꢀhydroxyacetimidamide (171
mg, 1.33 mmol) in ethyl acetate (7 mL) and THF (6 mL) over a period of 2 minutes under nitrogen. The
resulting solution was stirred at 20 °C for 3 days. The solvents were removed in vacuo and the residue
triturated with diethyl ether to give a white solid. This was dissolved in ethanol (25 mL) then
concentrated HCl (2.0 mL) was added over a period of 2 minutes under nitrogen. The resulting solution
was stirred at 70 °C for 18 hours. It was cooled to room temperature, concentrated in vacuo and
azeotroped once with toluene to give a pale yellow solid. The material was taken up in CH2Cl2,
concentrated in vacuo and adsorbed onto silica. The crude product was purified by flash silica
chromatography, elution gradient 0 to 100% EtOAc in CH2Cl2. Pure fractions were evaporated to dryness
to afford (R)ꢀ5ꢀ(4ꢀ(5ꢀ(2ꢀfluoroꢀ4ꢀ(methylsulfonyl)benzyloxy)pyrimidinꢀ2ꢀyl)ꢀ3ꢀmethylpiperazinꢀ1ꢀyl)ꢀ3ꢀ
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
1
(trifluoromethyl)ꢀ1,2,4ꢀoxadiazole (278 mg, 55%) as a white solid. m.p. 118 – 119 °C; H NMR (400
MHz, DMSO) 1.13 (3H, d, J = 6.7 Hz), 3.21 ꢀ 3.56 (6H, m), 3.83 ꢀ 3.94 (1H, m), 3.98 ꢀ 4.08 (1H, m),
13
4.36 ꢀ 4.46 (1H, m), 4.80 ꢀ 4.94 (1H, m), 5.27 (2H, s), 7.77 ꢀ 7.88 (3H, m), 8.34 (2H, s); C NMR (176
MHz, DMSO) 13.6, 37.3, 43.2, 45.4, 45.8, 49.8, 65.0, 114.3 (d, J = 24.4 Hz), 118.1 (q, J = 273 Hz), 123.1
(d, J = 3.2 Hz), 129.5 (d, J = 14.7 Hz), 131.3 (d, J = 3.7 Hz), 142.5 (d, J = 6.6 Hz), 145.1, 146.4, 156.7,
159.6 (d, J = 252 Hz), 160.6 (q, J = 24.4 Hz), 172.1; HRMS (EI) for C20H21O4N6SF4 (MH+); calcd,
517.1276; found, 517.1274.
5-({2-fluoro-4-[(methylsulfonyl)methyl]benzyl}oxy)-2-{(2R)-2-methyl-4-[3-(trifluoromethyl)-
1,2,4-oxadiazol-5-yl]piperazin-1-yl}pyrimidine (16). Zinc chloride (1M in diethyl ether) (0.715 mL,
0.72 mmol) was added to (R)ꢀ4ꢀ(5ꢀ(2ꢀfluoroꢀ4ꢀ(methylsulfonylmethyl)benzyloxy)pyrimidinꢀ2ꢀyl)ꢀ3ꢀ
methylpiperazineꢀ1ꢀcarbonitrile (300 mg, 0.72 mmol) and 2,2,2ꢀtrifluoroꢀN'ꢀhydroxyacetimidamide (124
mg, 0.97 mmol) in ethyl acetate (4.5 mL) and THF (4.0 mL) over a period of 2 minutes under nitrogen.
The resulting solution was stirred at 20 °C for 4 hours. The solvents were removed in vacuo, leaving a
white foam that was triturated with diethyl ether (10 mL) This was dissolved in ethanol (15 mL) and
fuming HCl (2 mL) was added. The resulting suspension was stirred at 100 °C for 18 hours. It was cooled
to room temperature and the solvent removed in vacuo. The residue was azeotroped once with toluene to
28
ACS Paragon Plus Environment

Page 29 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
leave a yellow gum (846 mg). It was taken up in CH2Cl2 and adsorbed onto silica. The crude product was
purified by flash silica chromatography, elution gradient 0 to 50% EtOAc in CH2Cl2. Pure fractions were
evaporated to dryness to afford (R)ꢀ5ꢀ(4ꢀ(5ꢀ(2ꢀfluoroꢀ4ꢀ(methylsulfonylmethyl)benzyloxy)pyrimidinꢀ2ꢀ
yl)ꢀ3ꢀmethylpiperazinꢀ1ꢀyl)ꢀ3ꢀ(trifluoromethyl)ꢀ1,2,4ꢀoxadiazole (129 mg, 34%) as a white solid. m.p.
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
1
124 – 125 °C; H NMR (400 MHz, CDCl3) 1.23 (3H, d, J = 6.7 Hz), 2.82 (3H, s), 3.27 – 3.38 (2H, m),
3.50 (1H, dd, J = 4.0, 13.1 Hz), 4.00 (1H, dt, J = 1.6, 13.1 Hz), 4.13 – 4.21 (1H, m), 4.24 (2H, s), 4.50 ꢀ
4.61 (1H, m), 4.95 ꢀ 5.04 (1H, m), 5.11 (2H, s), 7.18 ꢀ 7.28 (3H, m), 7.54 (1H, t, J = 7.6 Hz), 8.17 (2H, s);
13C NMR (176 MHz, CDCl3) 13.6, 37.3, 45.4, 45.7, 49.7, 58.6, 65.3, 117.7, 118.1 (q, J = 268 Hz), 123.6,
127.0, 130.8, 131.9, 145.3, 146.3, 156.6, 159.9 (d, J = 246 Hz), 160.6 (q, J = 12.3 Hz), 172.1; HRMS
(EI) for C21H23O4N6SF4 (MH+); calcd, 531.1432; found, 531.1429.
5-({2-fluoro-4-[(methylsulfinyl)methyl]benzyl}oxy)-2-{(2R)-2-methyl-4-[3-(trifluoromethyl)-
1,2,4-oxadiazol-5-yl]piperazin-1-yl}pyrimidine (17). To a stirred solution of (3R)ꢀ4ꢀ(5ꢀ(2ꢀfluoroꢀ4ꢀ
(methylsulfinylmethyl)benzyloxy)pyrimidinꢀ2ꢀyl)ꢀ3ꢀmethylpiperazineꢀ1ꢀcarbonitrile (290 mg, 0.72
mmol) and 2,2,2ꢀtrifluoroꢀN'ꢀhydroxyacetimidamide (124 mg, 0.97 mmol) in a mixture of ethyl acetate
(4.31 mL) and tetrahydrofuran (2.87 mL) under an atmosphere of nitrogen was added a 1.0M solution of
zinc chloride (1.581 mL, 1.58 mmol) in diethyl ether at ambient temperature over 10 minutes. The
mixture was stirred at ambient temperature for 2 hours, the solvents evaporated in vacuo to a residue
which was taken up in ethanol (4.31 mL) and treated with hydrochloric acid (3.59 mL, 7.19 mmol) and
the mixture heated under reflux for 16 hours. The mixture was cooled to ambient temperature, the ethanol
evaporated in vacuo to a residue which was partitioned between ethyl acetate (75 mL) and saturated
sodium hydrogen carbonate solution (25 mL), the ethyl acetate layer washed with brine, dried (MgSO4)
and evaporated in vacuo to a residue which was chromatographed on silica with ethyl acetate as eluant to
give a solid which was crystallised from ethyl acetate / isohexane to give 5ꢀ((3R)ꢀ4ꢀ(5ꢀ(2ꢀfluoroꢀ4ꢀ
(methylsulfinylmethyl)benzyloxy)pyrimidinꢀ2ꢀyl)ꢀ3ꢀmethylpiperazinꢀ1ꢀyl)ꢀ3ꢀ(trifluoromethyl)ꢀ1,2,4ꢀ
1
oxadiazole (61 mg, 16%). m.p. 128 – 129 °C; H NMR (400 MHz, CDCl3) 1.24 (3H, d, J = 6.7 Hz), 2.51
(3H, s), 3.27 ꢀ 3.39 (2H, m), 3.51 (1H, dd, J = 4.0, 13.1 Hz), 3.96 (2H, s), 4.01 (1H, dt, J = 1.6, 13.1 Hz),
29
ACS Paragon Plus Environment