Ago-allosteric modulators of human glucagon-like peptide 2 receptor

Article abstract of DOI:10.1016/j.bmcl.2012.08.026

Glucagon-like peptide 2 (GLP-2) is an intestinotropic peptide that binds to GLP-2 receptor (GLP-2R), a class-B G protein-coupled receptor (GPCR). Few synthetic agonists have been reported so far for class-B GPCRs. Here, we report the first scaffold compounds of ago-allosteric modulators for human GLP-2R, derived from methyl 2-{[(2Z)-2-(2,5-dichlorothiophen-3-yl)-2-(hydroxyimino) ethyl]sulfanyl}benzoate (compound 1).

Full text of DOI:10.1016/j.bmcl.2012.08.026

Bioorganic & Medicinal Chemistry Letters 22 (2012) 6126–6135
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Ago-allosteric modulators of human glucagon-like peptide 2 receptor
⇑
Kazuto Yamazaki , Hiroki Terauchi, Daisuke Iida, Hironori Fukumoto, Shuichi Suzuki, Takaki Kagaya,
Mika Aoki, Koichiro Koyama, Takashi Seiki, Kazuma Takase, Misako Watanabe, Tohru Arai,
Kappei Tsukahara, Junichi Nagakawa
Eisai Product Creation Systems, Eisai Co., Ltd., 5-1-3 Tokodai, Tsukuba, Ibaraki 300-2635, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Glucagon-like peptide 2 (GLP-2) is an intestinotropic peptide that binds to GLP-2 receptor (GLP-2R), a
class-B G protein-coupled receptor (GPCR). Few synthetic agonists have been reported so far for class-
B GPCRs. Here, we report the first scaffold compounds of ago-allosteric modulators for human GLP-2R,
derived from methyl 2-{[(2Z)-2-(2,5-dichlorothiophen-3-yl)-2-(hydroxyimino)ethyl]sulfanyl}benzoate
(compound 1).
Received 20 May 2012
Revised 18 July 2012
Accepted 7 August 2012
Available online 14 August 2012
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Glucagon-like peptide 2
Agonist
Positive allosteric modulator
Glucagon-like peptide 1 (GLP-1) and GLP-2 are peptides that are
encoded, together with glucagon, by a proglucagon gene.1 They are
produced in the intestine by tissue-specific posttranslational regu-
lation. GLP-1 and GLP-2 exert their effects through GLP-1 receptor
(GLP-1R) and GLP-2 receptor (GLP-2R), respectively, both of which
Boc5 acted on the orthosteric site of GLP-1R. In the other paper,
Knudsen et al. reported that quinoxaline derivatives acted on the
allosteric site of GLP-1R, because exendin(9–39) did not antagonize
their effects.11 Intriguingly, these compounds worked as both allo-
steric activators and agonists for GLP-1R. Therefore, the quinoxa-
line derivatives were categorized as ago-allosteric modulators.
Unlike the case with GLP-1R, there have been no reports of
small molecules with agonistic activity for GLP-2R. However, we
are Gas G protein-coupled receptors (GPCRs). Both peptides are lib-
erated by stimulation of nutrient intake, and secreted GLP-1 and
GLP-2 are rapidly degraded by dipeptidyl peptidase IV (DPPIV)
cleaving His-Pro dipeptides from the N-terminus. GLP-1 potenti-
ates glucose-dependent insulin secretion,2 and GLP-2 promotes
intestinal mucosal growth.3 Therefore, GLP-1 and GLP-2 are thera-
peutic targets for diabetes and intestinal damage, respectively. In
fact, DPPIV-resistant peptide GLP-1R agonists (e.g., exenatide, lira-
glutide)4 and DPPIV inhibitors (e.g., vildagliptin, sitagliptin) are
employed clinically for the treatment of diabetes.5 A DPPIV-resis-
tant peptide GLP-2R agonist, teduglutide, is under clinical develop-
ment for short bowel syndrome6 and Crohn’s disease.7
GLP-1R and GLP-2R have ꢀ40% identity, and both receptors be-
long to the glucagon-secretin class B of the GPCRs.8 Although sev-
eral small-molecule GPCR agonists have been found in classes A
and C,9 there had been no reports regarding agonists in class B.
However, in 2007, two reports on non-peptide agonists for GLP-
1R were published. Chen et al. showed that a substituted cyclobu-
tane compound, Boc5, increased intracellular cAMP levels in
HEK293 cells stably expressing rat GLP-1R and cAMP response ele-
ment (CRE)-driven luciferase; the response to this compound was
blocked by exendin(9–39), a peptide antagonist of GLP-1R.10 Thus,
Figure 1. Placental alkaline phosphatase (PLAP) activity induced by human
glucagon-like peptide 2 (GLP-2), or compound 1 or 2 in HEK293 cells stably
overexpressing human GLP-2 receptor and PLAP driven by cAMP response element.
Note that the concentration–activity curves are bell-shaped. Values are expressed
as means. n = 3.
⇑
Corresponding author. Tel.: +81 29 847 7097; fax: +81 29 847 7614.
E-mail address: k5-yamazaki@hhc.eisai.co.jp (K. Yamazaki).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.08.026

K. Yamazaki et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6126–6135
6127
recently discovered compounds with interesting activity toward
GLP-2R. This is the first report on structure–activity relationships
(SARs) of the compounds having both agonist and allosteric posi-
tive modulator activity toward human GLP-2R.
cells (Fig. 2C). These results showed that compound 2 was a GLP-
2R agonist.
We performed chemical development based on the structure of
compounds 1 and 2 to enhance GLP-2R agonistic activity. In the
SAR tables described below, we summarize the maximum PLAP re-
porter activities and concentrations at which the maximum PLAP
activities were observed, as representing the activities of the com-
pounds, because the concentration–activity curves were bell-
shaped as a result of cellular toxicity (Fig. 1).14 The PLAP activities
of the compounds were comparatively weak, particularly at the
beginning of chemical development; the maximum PLAP activities
were then expressed as values relative to those of 10 pM GLP-2, or
of 100 pM GLP-2 when the PLAP activities became stronger.
By synthesizing derivatives of compound 1 bearing an oxime
moiety, we obtained regioisomers with high or low polarity in gen-
eral. Table 1 shows the SARs of the polarities for compound 1, and
phenyl (compounds 2 and 3) and methoxyphenyl (compounds
4–8) derivatives. These results showed a clear-cut correlation
In preliminary screening, we found that methyl 2-{[(2Z)-2-(2,5-
dichlorothiophen-3-yl)-2-(hydroxyimino)ethyl]sulfanyl}benzoate
(compound 1) showed agonistic activity toward human GLP-2R in
cell reporter assays using CRE-driven placental alkaline phospha-
tase (PLAP) and human GLP-2R-overexpressing cells (hGLP-2R/
SE302).12,13 Compound 2, a de-esterified derivative of compound
1, had stronger activity than compound 1 (Fig. 1). Their concentra-
tion–activity curves were bell-shaped owing to their cell toxicity.14
PLAP reporter activity was not observed in CRE-driven PLAP-over-
expressing cells (SE302) or CRE-driven PLAP and human GLP-1R-
overexpressing cells (hGLP-1R/SE302). The same results were ob-
tained for intracellular cAMP levels15: compound 2 did not elevate
cAMP levels in SE302 or hGLP-1R/SE302 cells (Fig. 2A and B). How-
ever, this compound increased cAMP contents in hGLP-2R/SE302
Figure 2. Changes in intracellular cAMP levels in HEK293 cells stably overexpressing placental alkaline phosphatase (PLAP) regulated by cAMP response element (CRE)
(SE302 line) (A); in HEK293 cells overexpressing CRE-driven PLAP and human glucagon-like peptide 1 (GLP-1) receptor (hGLP-1R/SE302 line) (B); and in HEK293 cells
overexpressing CRE-driven PLAP and human glucagon-like peptide 2 (GLP-2) receptor (hGLP-2R/SE302 line) (C) treated with human GLP-1, GLP-2, or compound 2. Values are
expressed as means for GLP-1 and GLP-2 treatment, and as means S.E.M. for vehicle and compound 2 treatment. n = 6, 2, 2, and 6 for vehicle, GLP-1, GLP-2, and compound 2
treatment, respectively. ⁄p <0.05 compared with the vehicle-treated group.18

6128
K. Yamazaki et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6126–6135
Table 1
Structure–activity relationships with low and high polarities of compound 1 and its derivatives
OH
N
OH
O
O
N
S
S
Cl
Cl
R
S
S
Cl
Cl
1
Compound
R
Low/high polaritya
Active/inactive
Concentration at maximum
Maximum activity relative to GLP-2 (%)
activity (lM)
10 pM GLP-2
100 pM GLP-2
1
2
3
4
5
6
7
8
2-CO2Me
/
/
2-OMe
2-OMe
3-OMe
3-OMe
4-OMe
Low
Low
High
Low
High
Low
High
Low
Active
Active
Inactive
Active
Inactive
Active
25
25
284
701
39
55
50
25
50
78
—
—
—
144
217
Inactive
Active
Maximum activity values are expressed as means of P3 samples.
À, Not determined.
a
Products were separated by using silica gel column chromatography.
between their human GLP-2R reporter activity and polarity: low-
polarity compounds had activity (compounds 1, 2, 4, 6 and 8),
whereas high-polarity ones did not (compounds 3, 5 and 7). Our
E/Z analysis clarified that compound 1 was a Z-stereoisomer,16
which has low polarity.
When we replaced the 2-methoxycarbonyl-benzene in com-
pound 1, we observed activity in compounds having phenyl (com-
pound 2), 1-naphthyl (compound 14), 3- and 4-pyridyl
(compounds 16 and 17, respectively), and 2-thienyl (compound
18) groups (Table 2). It seemed that a planar aromatic structure
was needed in this portion, because alkyl or cycloalkyl group sub-
stitution did not give activity (compounds 9, 10 and 11). Addition
of one or three methylenes between the sulfanyl group and phenyl
group resulted in loss of potency (compounds 12 and 13), suggest-
ing that the distance between the two groups was important for
exhibition of activity. Contrary to 3- and 4-pyridyl substitution,
2-pyridyl substitution (compound 15) did not hold the activity.
4-Pyridine gave about double the maximum activity provided by
3-pyridine.
Phenyl SARs are shown in Table 3. Single substitution was
divided into two groups: 2-substitution showed higher activity
than 4-substitution for a methyl (compounds 20–22), chloro (com-
pounds 29–31), and trifluoromethyl (compounds 32 and 33) moi-
ety, and vice versa for a methoxy (compounds 4, 6 and 8), and
Table 2
Structure–activity relationships with substitutions of methylcarbonyl-benzene in compound 1
OH
O
O
OH
N
N
S
S
Cl
X
Cl
S
S
Cl
Cl
1
Compound
X
Low/high polaritya
Active/inactive
Active
Concentration at
maximum activity (lM)
Maximum activity relative to GLP-2 (%)
10 pM GLP-2
100 pM GLP-2
1
Low
25
25
284
701
39
55
2
9
Ph
Et
t-Bu
c-Hexyl
Bn
(CH2)3Ph
1-Naphthyl
2-Pyridyl
3-Pyridyl
4-Pyridyl
2-Thienyl
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Active
Inactive
Inactive
Inactive
Inactive
Inactive
Active
Inactive
Active
Active
10
11
12
13
14
15
16
17
18
25
—
23
50
50
25
—
—
—
27
55
40
Active
Maximum activity values are expressed as means of P3 samples.
À, Not determined.
a
Products were separated by using silica gel column chromatography.

K. Yamazaki et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6126–6135
6129
Table 3
Structure–activity relationships for the phenyl part of compound 1
OH
OH
O
O
N
N
S
S
Cl
Cl
R
S
S
Cl
Cl
1
Compound
R
Low/high polaritya
Active/inactive
Concentration at
Maximum activity relative to GLP-2 (%)
maximum activity (lM)
10 pM GLP-2
100 pM GLP-2
1
2
4
6
2-CO2Me
/
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Active
Active
Active
Active
Active
Inactive
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
25
25
50
25
50
284
701
78
144
217
39
55
—
—
—
2-OMe
3-OMe
4-OMe
2-CO2H
2-Me
3-Me
4-Me
2-Et
2-t-Bu
2-Ph
2-F
3-F
4-F
2-Cl
3-Cl
4-Cl
2-CF3
4-CF3
8
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
25
25
50
25
12.5
12.5
25
25
25
25
25
25
12.5
25
—
—
65
90
30
3
63
40
84
80
73
134
60
53
17
58
16
54
58
40
10
124
31
—
—
945
—
—
293
—
—
—
—
2,3-di-Me
2,4-di-Me
2,5-di-Me
2,6-di-Me
2-Me,4-F
2-Me,3-OMe
25
12.5
12.5
25
25
25
—
—
—
—
Maximum activity values are expressed as means of P3 samples.
À, Not determined.
a
Products were separated by using silica gel column chromatography.
Table 4
Structure–activity relationships of position 3 of pyridine of compound 17
OH
S
N
R
Cl
N
S
Cl
Compound
R
Low/high polaritya
Active/inactive
Concentration at maximum
activity (lM)
Maximum activity relative
to 100 pM GLP-2 (%)
17
40
41
42
43
H
Low
Active
Active
Active
Active
Active
50
50
25
25
55
44
143
59
102
Me
CF3
Et
Low/high mixture
Low
Low/high mixture
Low
Ph
12.5
Maximum activity values are expressed as means of P3 samples.
a
Products were separated by using silica gel column chromatography.
fluoro (compounds 26–28) moiety. Especially, 2-methyl (com-
pound 20), 2-trifluoromethyl (compound 32), 2-phenyl (compound
25), 2-ter-butyl (compound 24), and 4-fluoro (compound 28)
groups displayed relatively high activity. On the other hand, 2-sub-
stitution with a carboxy group caused a complete loss of activity
(compound 19), suggesting that polar groups were not tolerated
at this position. It appeared that no clear additive or synergic in-
crease in the activity was generated by double substitution, com-
paring compounds 20, 22 and 35, compounds 20, 28 and 38, and
compounds 6, 20 and 39.
As 4-pyridyl substitution maintained the potency, the SAR of
position 3 of the 4-pyridine ring was examined (Table 4). Incorpo-
ration of a trifluoromethyl (compound 41) or phenyl (compound
43) group resulted in higher reporter activity than that of a methyl
(compound 40) or ethyl (compound 49) group. This SAR seemed to
be similar to that of position 2 of the benzene ring.
We found that it was possible to replace the thiophene in com-
pound 1 with benzene (compound 44), so we then investigated the
SARs of this benzene (Table 5). Compound 44 had a very weak
activity, but single substitution at position 3 increased the activity

6130
K. Yamazaki et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6126–6135
Table 5
Structure–activity relationships with substitution of benzene for thiophene in compound 2 (Part I)
OH
N
R1
S
R2
Compound
R1
R2
Low/high polaritya
Active/inactive
Concentration
Maximum activity relative to GLP-2 (%)
at maximum activity (lM)
10 pM GLP-2
4
100 pM GLP-2
—
44
45
46
47
48
49
50
51
52
53
54
55
56
H
OH
CN
OMe
Propoxy
i-Propoxy
F
Cl
Br
CF3
CF3
t-Bu
t-Bu
H
H
H
H
H
H
H
H
H
H
CF3
H
t-Bu
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Active
Inactive
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
50
50
25
25
25
25
25
25
25
25
25
25
8
—
—
—
54
46
42
95
517
244
—
—
8
9
14
—
—
—
—
29
18
31
Maximum activity values are expressed as means of P3 samples.
À, Not determined.
a
Products were separated by using silica gel column chromatography.
Table 6
Structure–activity relationships with substitution of benzene for thiophene in compound 2 (Part II)
OH
N
S
R1
R2
Compound
R1
R2
Low/high polaritya
Active/inactive
Concentration at
maximum activity (lM)
Maximum activity relative
to 100 pM GLP-2 (%)
55
56
57
58
59
60
61
H
H
OH
OMe
OMe
OMe
OMe
H
Low
Low
Low
Low
Low
Low
Low
Active
Active
Active
Inactive
Active
Active
Active
25
25
25
18
31
4
t-Bu
t-Boc
CO2H
t-Boc
NMe2
Br
12.5
12.5
12.5
12
43
43
Maximum activity values are expressed as means of P3 samples.
a
Products were separated by using silica gel column chromatography.
Table 7
Structure–activity relationships with substitution of pyrazole for thiophene in compound 2
OH
N
S
R2
Low/high polaritya
N
N
R1
Compound
R1
R2
Active/inactive
Concentration at
Maximum activity
maximum activity (
l
M)
relative to 100 pM GLP-2 (%)
62
63
64
65
66
67
68
69
70
71
72
73
74
Me
Me
Me
Me
Me
Me
n-Pr
i-Pr
s-Bu
1-Et-Pr
Bn
Me
c-Pr
t-Bu
Ph
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
Low
Low
Low
Low
Low
High
Low
Low
Low
Low
Low
Low
Low
Inactive
Inactive
Inactive
Active
Active
Inactive
Active
Active
Active
Active
Active
Active
Active
25
25
14
13
25
25
12.5
12.5
12.5
25
71
87
110
31
59
c-Bu
c-Pentyl
99
127
25
Maximum activity values are expressed as means of P3 samples.
a
Products were separated by using silica gel column chromatography.

K. Yamazaki et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6126–6135
6131
(compounds 46–53, and 55), except for a hydroxyl group (com-
pound 45). Particularly, high activity was caused by substitution
with a trifluoromethyl (compound 53) or ter-butyl (compound
55) group at position 3. Double substitution at positions 3 and 5
made the reporter activity higher than with 3-substitution alone,
as shown by comparing compound 53 with compound 54, and
compound 55 with compound 56.
Additionally, we replaced positions 4 and 5 of 3-ter-butyl-ben-
zene (Table 6). Comparison between compounds 57 and 59 sug-
gested that 4-substitution of a methoxy group resulted in higher
activity than that of a hydroxy group. When we fixed the methoxy
group at position 4, derivatives having a dimethylamino (com-
pound 60), or bromo (compound 61) group at position 5 had com-
paratively high potency. Incorporation of a carboxy group at
position 5 (compound 58) led to no activity, suggesting that this
position did not tolerate polar groups.
compounds had 2-trifluoromethyl-phenyl in place of the 2-
methoxycarbonyl-phenyl group. Comparison with compound 1 re-
vealed that the concentrations at maximum activity were de-
creased by about fourfold. Thus, we were able to potentiate GLP-
2R agonistic activity by suitable combination of the two rings.
Figure 3 shows concentration–activity curves of compounds 1, 75
and 76. Outline of the synthesis of these potent compounds is indi-
cated in Scheme 1.
To investigate the mode of action of our GLP-2R agonists, we
examined whether GLP-2(11–33) antagonized the PLAP reporter
activity of compound 1. GLP-2(11–33) was a novel partial agonist
for human GLP-2R, having 88% binding activity at 1
at 10
M,17 and 11% of intrinsic PLAP reporter activity (EC50
1.3 M) compared with GLP-2. GLP-2(11–33) was able to antago-
lM and 100%
l
,
l
Thiophene was also able to be replaced by pyrazole, and we
examined its 1,3-substitution (Table 7). Fixing a methyl group at
position 1, no activity was observed in compounds having a methyl
(compound 62), cyclopropyl (compound 63), or ter-butyl (com-
pound 64) moiety at position 3. However, incorporation of a phenyl
(compound 65), or trifluoromethyl (compound 66) group gener-
ated the PLAP activity. Then, we incorporated an alkyl, cycloalkyl,
or benzyl group at position 1, with the trifluoromethyl group at po-
sition 3 fixed (compounds 68–74). Among these compounds, sec-
butyl (compound 70) and cyclopentyl (compound 74) substituents
produced higher activity. Compound 67, which was the high-polar-
ity version of compound 66, showed a loss of potency. Therefore,
the correlation between polarity and activity also seemed true
for pyrazole derivatives.
On the basis of the above-mentioned substitution SARs of two
rings in compound 1, we prepared compounds combined with
rings showing higher potency (Table 8), namely, 3-ter-butyl-4-
methoxy-5-bromo-phenyl (compound 75) or 1-cyclopentyl-3-tri-
fluoromethyl-pyrazol-5-yl (compound 76) in place of the 2,5-di-
chloro-thiophen-3-yl group of compound 1; both of these
Figure 3. Placental alkaline phosphatase (PLAP) activity induced by human
glucagon-like peptide 2 (GLP-2), and by compounds 1, 75 and 76 in HEK293 cells
stably overexpressing human GLP-2 receptor and PLAP driven by cAMP response
element. Concentration–activity curves are drawn up until the maximum activity of
the compounds. Values are expressed as means. n = 3.
Table 8
Combination of Ar1 and Ar2 rings
CH3
OH
O
O
OH
N
N
S
S
Ar1
Ar2
Cl
S
Cl
1
Compound
Ar1
Ar2
Low/high polaritya
Active/inactive
Active
Concentration at
maximum activity (lM)
Maximum activity relative to GLP-2 (%)
10 pM GLP-2
100 pM GLP-2
O
O
Cl
1
S
Low
25
284
39
Cl
CF3
75
76
Low
Low
Active
Active
6.3
6.3
714
610
97
83
O
Br
CF3
F3C
N
N
Maximum activity values are expressed as means of P3 samples.
a
Products were separated by using silica gel column chromatography.

6132
K. Yamazaki et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6126–6135
79
Outline of the synthesis of
Br
Br
Br
H
N
Ar2-SH or Ar2-SNa
CuBr2 or
O
OH
O
O
N
78
NH2OH HCl
Br
O
Ar1
S
Ar1
S
Ar1
Ar2
Ar1
Ar2
77
79
77
Synthesis of
O
HCl
R2-Br
K2CO3, DMF
H
O
O
N
O
MeMgBr
O
R1
R1
O
R1
OH
N
R1
N
NH
WSC, Et3N
N
N
N
N
N
NH
N
R2
R2
78
S
N
Synthesis of
R3
Cl
1) Δ
2) NaOH
R3
R3
SNa or SH
OH
NaH, DMF
O
N
S
R4
R4
N
Cl
SH
NaSH
N
Scheme 1. General synthesis of human GLP-2 receptor ago-allosteric modulators.
Figure 4. Effects of a truncated human glucagon-like peptide 2 (GLP-2), GLP-2(11-33), on intracellular cAMP produced by GLP-2 (A), and on placental alkaline phosphatase
(PLAP) activity induced by compound 1 (B) in HEK293 cells stably overexpressing human GLP-2 receptor and PLAP driven by cAMP response element. Values are expressed as
means S.E.M. n = 3.
nize GLP-2 activity, resulting in a concentration-dependent de-
crease in the intracellular cAMP concentration generated by GLP-
2 (Fig. 4A). This examination gave the unexpected result that the
PLAP activity of compound 1 was dramatically increased by the
addition of GLP-2(11–33), in a concentration-dependent manner
(Fig. 4B).
Next, we investigated whether our compounds enhanced the
reporter activity of GLP-2, GLP-2(11–33), and GLP-2(3–33), a trun-

K. Yamazaki et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6126–6135
6133
cated GLP-2 peptide produced by DPPIV in vivo (its intrinsic PLAP
reporter activity was 58% of that of GLP-2 and its EC50 was 31 nM).
Addition of compound 2 shifted the concentration–reporter re-
sponse curves for GLP-2 and its truncated peptides to the left or
upward, or both, in a concentration-dependent fashion: this addi-
tion increased the PLAP activity of GLP-2 additively (Fig. 5A). EC50
values were not affected by compound 2, but maximum PLAP
activity was increased by ꢀ10% with the addition of compound 2
at 10 lM. For GLP-2(3–33), compound 2 at 1 and 10 lM decreased
EC50 values by ꢀ1.2-fold and ꢀ3.3-fold, and increased maximum
activity by ꢀ110% and ꢀ150%, respectively (Fig. 5B). More notable
was the effect of compound 2 on the PLAP activity of GLP-2(11–33)
Figure 5. Effects of compound 2 on placental alkaline phosphatase (PLAP) activity induced by human glucagon-like peptide 2 (GLP-2) (A), GLP-2(3-33) (B), and GLP-2(11-33)
(C) in HEK293 cells stably overexpressing human GLP-2 receptor and PLAP driven by cAMP response element.19 Each value is the mean of three samples.
Table 9
Effects of compound 2 on placental alkaline phosphatase (PLAP) activity of GLP-2 and its truncated peptides in HEK293 cells overexpressing human GLP-2 receptor and PLAP
driven by the cAMP response element
GLP-2 or truncated
GLP-2 peptide
GLP-2
Compound 2
EC50 (nM)
Maximum PLAP activity (%)
Relative to each control
Relative to GLP-2
—
1 lM
0.097
0.084
0.087
31
27
9.4
1300
380
75
100
103
114
100
114
145
100
494
932
100
103
114
59
67
97
11
56
107
10
lM
GLP-2(3–33)
—
1
lM
10
lM
GLP-2(11–33)
—
1
lM
10
lM

6134
K. Yamazaki et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6126–6135
Figure 6. Changes in intracellular cAMP levels induced by human glucagon-like peptide 2 (GLP-2) (A), GLP-2(3–33) (B), and GLP-2(11–33) (C) in HEK293 cells stably
overexpressing human GLP-2 receptor and placental alkaline phosphatase regulated by cAMP response element, in the present or absence of compound 2. Values are
expressed as means S.E.M. n = 3.
(Fig. 5C). Compound 2 treatment at 1 and 10
l
M decreased EC50
allosteric modulators of GLP-2R. These findings, together with the
agonistic actions of the compounds alone, indicated that our com-
pounds functioned as both agonists and positive allosteric modula-
tors—that is, positive ago-allosteric modulators—of human GLP-2R.
Our observations would be useful for developing small-molecule
agonists and/or ago-allosteric modulators of human GLP-2R.
values by ꢀ3.4-fold and ꢀ17-fold, and increased maximum activity
by ꢀ490% and ꢀ930%, respectively. The PLAP activity of GLP-2(3–
33) and GLP-2(11–33) was raised to the maximum intrinsic activ-
ity of GLP-2 by the addition of compound 2 at 10 lM. These effects
are summarized in Table 9. Similar effects were also observed for
compound 56 (data not shown). We also tested the effects of com-
pound 2 on changes in intracellular cAMP levels in the presence of
References and notes
GLP-2, GLP-2(3–33), or GLP-2(11–33). Compound 2 at 10 lM in-
1. Drucker, D. J. Endocrinology 2001, 142, 521.
creased intracellular cAMP levels in the presence of 1 nM GLP-2
(Fig. 6A). In the cases of GLP-2(3–33) and GLP-2(11–33), compound
2 elevated intracellular cAMP levels (Fig. 6B and C, respectively);
this effect was particularly prominent with GLP-2(11–33).
We examined the binding activity of compound 17 to human
GLP-2R,17 because among the active compounds examined this
compound had relatively good solubility. There was no binding
activity until 100 lM. We therefore speculated that the com-
pounds did not interact with the GLP-2-binding site of GLP-2R
(i.e. the orthostatic site), and that these compounds were positive
2. Schmidt, W. E.; Siegel, E. G.; Creutzfeldt, W. Diabetologia 1985, 28, 704.
3. Drucker, D. J.; Ehrlich, P.; Asa, S. L.; Brubaker, P. L. Proc. Natl. Acad. Sci. U.S.A.
1996, 93, 7911.
4. Drab, S. R. Pharmacotherapy 2009, 29, 43S.
5. Ahrén, B. Diabetes Care 2007, 30, 1344.
6. Jeppesen, P. B.; Gilroy, R.; Pertkiewicz, M.; Allard, J. P.; Messing, B.; O’Keefe, S. J.
Gut 2011, 60, 902.
7. Buchman, A. L.; Katz, S.; Fang, J. C.; Bernstein, C. N.; Abou-Assi, S. G. Inflamm.
Bowel Dis. 2010, 16, 962.
8. Munroe, D. G.; Gupta, A. K.; Kooshesh, F.; Vyas, T. B.; Rizkalla, G.; Wang, H.;
Demchyshyn, L.; Yang, Z. J.; Kamboj, R. K.; Chen, H.; McCallum, K.; Sumner-
Smith, M.; Drucker, D. J.; Crivici, A. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 1569.

K. Yamazaki et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6126–6135
6135
9. (a) Aquino, C. J.; Armour, D. R.; Berman, J. M.; Birkemo, L. S.; Carr, R. A. E.;
Croom, D. K.; Dezube, M.; Dougherty, R. W., Jr.; Ervin, G. N.; Grizzle, M. K.;
Head, J. E.; Hirst, G. C.; James, M. K.; Johnson, M. F.; Miller, L. J.; Queen, K. L.;
Rimele, T. J.; Smith, D. N.; Sugg, E. E. J. Med. Chem. 1996, 39, 562; (b) Ankersen,
M.; Crider, M.; Liu, S.; Ho, B.; Andersen, H. S.; Stidsen, C. J. Am. Chem. Soc. 1998,
120, 1368; (c) Mitsukawa, K.; Yamamoto, R.; Ofner, S.; Nozulak, J.; Pescott, O.;
Lukic, S.; Stoehr, N.; Mombereau, C.; Kuhn, R.; McAllister, K. H.; van der Putten,
H.; Cryan, J. F.; Flor, P. J. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 18712; (d) Monn, J.
A.; Massey, S. M.; Valli, M. J.; Henry, S. S.; Stephenson, G. A.; Bures, M.; Hérin,
M.; Catlow, J.; Giera, D.; Wright, R. A.; Johnson, B. G.; Andis, S. L.; Kingston, A.;
Schoepp, D. D. J. Med. Chem. 2007, 50, 233; (e) Westaway, S. M.; Brown, S. L.;
Fell, S. C. M.; Johnson, C. N.; MacPherson, D. T.; Mitchell, D. J.; Myatt, J. W.;
Stanway, S. J.; Seal, J. T.; Stemp, G.; Thompson, M.; Lawless, K.; McKay, F.; Muir,
A. I.; Barford, J. M.; Cluff, C.; Mahmood, S. R.; Matthews, K. L.; Mohamed, S.;
Smith, B.; Stevens, A. J.; Bolton, V. J.; Jarvie, E. M.; Sanger, G. J. J. Med. Chem.
2009, 52, 1180.
(day 2), the compound or peptide, or both, was added to the culture, which was
then incubated at 37 °C overnight. The supernatants were collected and mixed
with substrate/enhancer solution (Applied Biosystems) in the wells of 96-well
plates; the plates were incubated for 1 h at room temperature on day 3, and
PLAP activity was determined. In screening, both hGLP-2R/SE302 and SE302
cells were utilized. Evaluation using hGLP-1R/SE302 cells was added if needed.
Cell viability was always checked using Invitrogen’s alamarBlueÒ
.
14. Decreases in the PLAP reporter activity of a compound were accompanied with
decreases in fluorescence intensity in the alamarBlue assay.
15. Cells were treated with 2 mM of 3-isobutyl-1-methylxanthine for 30 min at
room temperature. The cells (5 Â 104) were then treated with the compound or
peptide, or both, on 96-well half-area plates at 37 °C for 30 min. Intracellular
cAMP was then quantified by using Cisbio’s cAMP measurement kit.
16. The E/Z-stereochemistry of compound
1 was confirmed from its NOESY
spectrum by using Avance 600 MHz NMR (Bruker BioSpin AG).
17. Membranes for binding experiments were prepared from confluent hGLP-2R/
SE302 cells. Receptor binding studies were conducted using 100 pM [125I]-
Bolton Hunter labeled GLP-2 as a ligand. On the day of the experiments, the
10. Chen, D.; Liao, J.; Li, N.; Zhou, C.; Liu, Q.; Wang, G.; Zhang, R.; Zhang, S.; Lin, L.;
Chen, K.; Xie, X.; Nan, F.; Young, A. A.; Wang, M. W. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 943.
membranes were reconstituted at 100 lg protein in 0.5 mL of binding buffer.
11. Knudsen, L. B.; Kiel, D.; Teng, M.; Behrens, C.; Bhumralkar, D.; Kodra, J. T.; Holst,
J. J.; Jeppesen, C. B.; Johnson, M. D.; de Jong, J. C.; Jorgensen, A. S.; Kercher, T.;
Kostrowicki, J.; Madsen, P.; Olesen, P. H.; Petersen, J. S.; Poulsen, F.; Sidelmann,
U. G.; Sturis, J.; Truesdale, L.; May, J.; Lau, J. Proc. Natl. Acad. Sci. U.S.A. 2007, 104,
937.
The binding was equilibrated by incubation at room temperature for 2 h. The
ligand–receptor complexes were separated on a glass filter by rapid filtration
at 4 °C. Nonspecific binding was assessed in the presence of 100 nM GLP-2.
Each filter was collected, and the radioactivity remaining was quantified with a
gamma counter.
12. By transduction using murine leukemia-based retrovirus into the HEK293 cell
18. In the comparison between vehicle and treated groups, Bartlett’s test for equal
variance was conducted, followed by one-way analysis of variance (when p
P0.05 in Bartlett’s test) or the Kruskal-Wallis test (when p <0.05 in Bartlett’s
line, we generated
a reporter cell line stably expressing cAMP response
element-driven human placental alkaline phosphatase (PLAP). The cell line was
named SE302. By the same method, human GLP-1 receptor (GLP-1R) or GLP-2
receptor (GLP-2R) was transduced into SE302 cells to produce cell lines stably
expressing human GLP-1R and GLP-2R. These lines were named hGLP-1R/
SE302 and hGLP-2R/SE302, respectively.
test) using GraphPad PrismÒ
performed as a post-hoc test.
. Dunnett’s multiple comparison test was
19. Regression curves were generated by using GraphPad Prism, and peak and EC50
values were calculated.
13. SE302, hGLP-1R/SE302, and hGLP-2R/SE302 cells were seeded at 2 Â 104/well
on to gelatin-coated 96-well plates and incubated at 37 °C overnight. Next day