3,5-Dihydro-imidazo[4,5-d]pyridazin-4-ones: A class of potent DPP-4 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2008.04.075

Systematic variations of the xanthine scaffold in close analogs of development compound BI 1356 led to the class of 3,5-dihydro-imidazo[4,5-d]pyridazin-4-ones which provided, after substituent screening, a series of highly potent DPP-4 inhibitors.

Full text of DOI:10.1016/j.bmcl.2008.04.075

Bioorganic & Medicinal Chemistry Letters 18 (2008) 3158–3162
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: elsevier.com/locate/bmcl
3,5-Dihydro-imidazo[4,5-d]pyridazin-4-ones: A class of potent DPP-4 inhibitors
a
a
a
b
c
Matthias Eckhardt a, , Norbert Hauel , Frank Himmelsbach , Elke Langkopf , Herbert Nar , Michael Mark ,
*
Moh Tadayyon c, Leo Thomas c, Brian Guth d, Ralf Lotz d
a Department of Chemical Research, Boehringer Ingelheim Pharma GmbH & Co. KG, 88400 Biberach, Germany
b Department of Lead Discovery, Boehringer Ingelheim Pharma GmbH & Co. KG, 88400 Biberach, Germany
c Department of Metabolic Diseases Research, Boehringer Ingelheim Pharma GmbH & Co. KG, 88400 Biberach, Germany
d Department of Drug Discovery Support, Boehringer Ingelheim Pharma GmbH & Co. KG, 88400 Biberach, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Systematic variations of the xanthine scaffold in close analogs of development compound BI 1356 led to
the class of 3,5-dihydro-imidazo[4,5-d]pyridazin-4-ones which provided, after substituent screening, a
series of highly potent DPP-4 inhibitors.
Received 9 April 2008
Revised 28 April 2008
Accepted 29 April 2008
Available online 1 May 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Dipeptidyl peptidase 4
DPP-4
Imidazopyridazinones
Inhibition of the serine protease dipeptidyl peptidase 4 (DPP-4)
has proven to be an effective treatment for improving glycemic
control in type 2 diabetic patients.1 Recovery of glucose homeosta-
sis by DPP-4 inhibition results from the prolonged action of the
incretins glucagon-like peptide 1 (GLP-1) and glucose-dependent
insulinotropic peptide (GIP) that are normally rapidly inactivated
by DPP-4 through the cleavage of a dipeptide from the N-terminus
of these oligopeptides.2 GLP-1 and GIP are released from the gut in
response to food intake, and exert a potent glucose-dependent
insulinotropic action and thereby contribute to the maintenance
of post-prandial glycemic control.3 Further contributing factors
are the inhibition of glucagon release from pancreatic a-cells,
reduction of food intake and retardation of gastric emptying, which
are mediated by GLP-1.4 The observed beneficial effects on pancre-
atic b-cells have also been attributed to an extended action of
GLP-1.5 Consequently, inhibition of DPP-4 with small molecules
has become an attractive approach to treat type 2 diabetes that
has been pursued by several research groups.6
O
7
N
1
N
N
N
DPP-4: IC50 = 1 nM
N
8
N
O
N3
NH2
Figure 1. DPP-4 inhibitor BI 1356.
xanthine scaffold proved to be an excellent platform leading to a
very promising DPP-4 inhibitor, we decided to further explore vari-
ations within the xanthine framework with respect to the overall
profile. Hence, we started evaluating different scaffolds based on
the most favorable residue ensembles discovered for xanthine,
with a particular emphasis on scaffolds that place the various res-
idues in roughly the same spatial locations as that achieved by the
xanthine core.
Our experience with substituents on the xanthine scaffold sug-
gested an arrangement including 3-amino-piperidinyl at C-8 and
but-2-ynyl at N-7 as particularly effective. We discovered that this
arrangement is best complemented with methyl at N-3 and an
arylmethyl group attached to N-1, among which inter alia naph-
th-1-ylmethyl and 3-methyl-isoquinolin-1-ylmethyl proved supe-
rior. An X-ray crystal structure of BI 1356 complexed with
human DPP-4 demonstrated the manner in which these residues
provided a good fit: the primary amine of the piperidine moiety
forms a highly attractive network of charge reinforced hydrogen
bonds with the enzyme, while the butynyl group fits well into a
hydrophobic pocket and the quinazoline interacts favorably by p-
stacking.7 A hydrogen bond between the C-6 carbonyl function of
Recently, we reported the development of BI 1356 (proposed
trade name ONDERO), a highly potent and selective DPP-4 inhibi-
tor that is currently undergoing clinical phase III trials (Fig. 1).7
BI 1356 is a xanthine-based molecule that was identified after
broad exploration of the residue decoration around the xanthine
core with regard to their effect on pharmacodynamic and pharma-
cokinetic properties as well as adverse side-effects. Although the
* Corresponding author. Tel.: +49 7351 545539; fax: +49 7351 545181.
E-mail address: matthias.eckhardt@boehringer-ingelheim.com (M. Eckhardt).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.04.075

M. Eckhardt et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3158–3162
3159
the xanthine and the enzyme completes the attractive interactions.
R
O
O
Accordingly, we designed and synthesized a series of compounds
with different scaffolds satisfying the basic requirements of orient-
ing the residues comparably to xanthine and bearing one of the
two residue ensembles mentioned, including the C-6 carbonyl
group of xanthine.
N
N
N
N
HN
X
a
N
X
Br
Br
X = N, CH
Me/H
Me/H
b
b
R
The initial structural alterations varied the uracil substructure
of xanthine but retained the integrity of the cyclic nature and the
attachment site of the N-1 residue such as in pyridazone 2, meth-
ylpyridazone 3, pyridone 4, and methoxypyridazone 5 (Table 1).
The syntheses of these aromatics have been described before and
are sketched in Schemes 1 and 2.8,9
O
O
a
N
N
N
N
HN
N
N
N
N
X
NHBoc
for R see Table 1
NHPG
Me/H
Me/H
PG = Boc
PG = H 2/3/4
c
More substantial alterations of the uracil moiety included dis-
banding the ring structure leading to monocyclic imidazoles 6–9.
The synthetic access to all these imidazoles via the advanced inter-
mediate I-1 could not be accomplished as originally planned since
the replacement of the 5-NH2 group with H, Br, and CN after diaz-
Scheme 1. Reagents: (a) RCH2Cl/Br, K2CO3, DMF; (b) 3-BocHN-piperidine, Na2CO3,
DMSO; (c) TFA, DCM.
R
R
O
O
BocHN
O
R1OC
N
c
d
N
N
N
N
N
Table 1
Br
N
N
Br
HN
DPP-4 inhibitory activity of xanthine related core structures
Br
MeO2C
b
MeO2C
R1 = OMe
R1 = OH
R1 = Cl
Compound
Structure
DPP-4 IC50 (nM)
a
R
R
O
O
Y
e
N
N
f
N
N
N
X
R =
Br
N
N
N
N
X=CH,Y=H X=N,Y=Me
NHPG
g
OMe
OMe
PG = Boc
PG = H 5
R
O
for R see Table 1
NH2
N
N
N
1
2
3
4
15
13
2
5
Scheme 2. Reagents: (a) NaOH, H2O/THF/MeOH; (b) SOCl2, DCM; (c) RCH2NHNH-
Boc, NEt3, DMF; (d) TFA, DCM; (e) MeI, K2CO3, DMF; (f) 3-BocHN-piperidine,
Na2CO3, DMSO; (g) TFA, DCM.
N
O
R
N
O
NH2
NH2
5
N
3
N
N
2
otization worked only well for H (Scheme 3)10; the latter two imi-
dazoles had to be synthesized using alternative routes. Moreover,
we prepared the triazole analog 1011 of imidazole 6 to investigate
the effect on substituent conformation and, in turn, on activity of a
C–H ? N exchange at the 5-position of imidazole.
N
N
N
7
R
R
O
N
N
N
N
—
7
The DPP-4 inhibitory activity was determined using human
DPP-4,12 and the results obtained for the various scaffolds substi-
tuted with the ensemble comprising either 1-naphthyl or 3-
methyl-isoquinolin-1-yl are compiled in Table 1, including the data
for the analogous xanthine compounds. Replacement of the uracil
moiety of the original xanthine with a pyridazone (? 2 and 3),
either with or without an additional methyl group, had no consid-
erable effect on the inhibitory activity, while a methoxy substitu-
ent on pyridazone (? 5) or switching to a pyridone (? 4) led to
a significant loss of potency. More pronounced scaffold alterations
resulted in compounds of even lower activity. Although compound
6 still exhibited respectable DPP-4 inhibition, the C-5 derivatized
imidazoles 8 and 9 were more than 30-fold less potent than xan-
thine 1. Repulsion of the amide group and the residue at C-5 of
the imidazole causing a different orientation of the arylmethyl
group and the amide carbonyl would explain the considerably infe-
rior potency of the monocyclic aromatics. Support for this rational-
ization is seen with imidazole 7 which has a bulkier secondary
amide, most likely resulting in an even more distorted residue
and amide group positioning and thus a further drop in potency.
Additional evidence is provided by triazole 10 which should be
capable of forming an intramolecular hydrogen bond between
the amide N–H and the 5-N of the triazole by which the initial cyc-
lic form would be reestablished: triazole 10 is significantly more
active than its imidazole congener 6 exhibiting activity comparable
to pyridone 4 and pyridazone 5. These observations, combined
O
NH2
NH2
3N
N
5
N
89
52
21
2
2
N
N
7
R
R
O
3N
N
5
N
5
19
N
7
OMe
O
NH2
3
6
7
R0 = H
107
3059
99
—
N
N
2
R0 = Me
N
R'
5
N
R
R
O
8
9
R0 = CN
R0 = Br
582
559
—
—
NH2
NH2
N
N
N
H
N
N
R'
O
3
N
10
—
24
N
H
2
N
5
N

3160
M. Eckhardt et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3158–3162
R
O
EtO2C
PhO
NC
EtO2C
H2N
R1O2C
EtO2C
a
b
c
d
NHBoc
R
f
N
N
N
N
N
N
N
N
N
OPh
N
N
N
N
N
N
R'
N
OPh
NC
NC
R1=Et
PG=Boc
PG=H 6/7
NHBoc
NHBoc
NHPG
g
e
R1=H
I-1
O
R1O2C
NC
h
i
f
N
N
N
N
EtO2C
NC
EtO2C
O
N
H
N
N
N
N
Br
Br
N
NC
N
R1=Et
NHBoc
i
NHPG
PG=Boc
PG=H 8
g
e
R1=H
R
O
R
O
H
EtO2C
Br
R1O2C
Br
j
f
N
N
N
N
N
N
N
Br
H
N
H
Br
Br
N
N
Br
N
Br
NHPG
R1=Et
PG=Boc
PG=H 9
g
e
R1=H
Scheme 3. Reagents: (a) i—EtO2CCH2NH2, K2CO3, acetone; ii—MeCCCH2Br, K2CO3, DMF; (b) 3-BocHN-piperidine, K2CO3, DMF; (c) NaOEt, EtOH; (d) NaNO2, aq HCl; H3PO2; (e)
aq KOH, THF; (f) RCH2NHR0 or RCH2NH2, TBTU, Eti-Pr2N, DMF; (g) TFA, DCM; (h) H2NOSO3H, pyridine, EtOH; (i) 3-BocHN-piperidine, Na2CO3, DMSO; (j) MeCCCH2 Br, K2CO3,
DMF; for R and R’ see Table 1.
with the X-ray data of BI 1356, allow for the assumption that a co-
planar relationship of the imidazole ring and the amide group, with
the carbonyl oxygen positioned next to the butynyl group, is a pre-
requisite for optimal arylmethyl and carbonyl group alignment.
Of the different scaffolds prepared, only the promising data of
the pyridazones warranted further studies.13 In addition to
in vitro DPP-4 inhibitory activity we included activity on the mus-
carinic receptor M1 and inhibition of DPP-4 in rats14 into our pro-
Table 2
Activities of 5-derivatized 3,5-dihydro-imidazo[4,5-d]pyridazin-4-ones
Compounda
DPP-4 IC50 (nM)
M1 IC50 (nM)
DPP-4 inhib. in ratb (%)
R
O
N
N
N
N
N
NH2
R'
R
R0
H
N
N
11-(R)
12-(R)
1
1
1190
93
80
76
N
N
Me
13
13-(S)
13-(R)
H
14
25
3
5607
—
>5000
73
9
63
O
N
14
15
PhCO
H
H
26
4
—
8
862
84
NC
O
16
H
4
—
64
N
a
(R) or (S) is appended to the compound number when the pure enantiomer was used.
Determined ex vivo after 7 h post-compound administration (po, 10 mg/kg) versus control group.
b

M. Eckhardt et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3158–3162
3161
Table 3
Activities with variations at C-2 of a 3,5-dihydro-imidazo[4,5-d]pyridazin-4-one
Compounda
DPP-4 IC50 (nM)
M1 IC50 (nM)
DPP-4 inhib. in ratb (%)
O
N
N
N
R
N
N
N
N
N
11-(R)
1
5
1190
2804
80
45
NH2
NH
17
18
19
3
1
954
20
29
N
N
N
NH
>5000
NH2
NH2
20
2
>5000
50
a
(R) is appended to the compound number when the pure enantiomer was used.
Determined ex vivo after 7 h post-compound administration (po, 10 mg/kg) versus control group.
b
filing scheme for the subsequent optimization process; significant
M1 receptor inhibition was a characteristic of some early xanth-
ines. We set out to confirm the substituent SAR established for
the xanthines with the 3,5-dihydro-imidazo[4,5-d]pyridazin-4-
ones. A set of residues known to be well suited from the xanthine
series were attached to C-2 and N-5 of the imidazopyridazone
core; substitution of the butynyl group was not considered as it
proved to be an important feature of xanthines showing no un-
wanted side-effects, such as hERG or receptor M1 inhibition.
Attaching 4-methyl-quinazolin-2-ylmethyl, the residue of BI
1356, to the two pyridazones [? 11-(R) and 12-(R)] led to very po-
tent DPP-4 inhibitors in vitro as well as in vivo (Table 2). Unfortu-
nately, methyl derivatized imidazopyridazone 12-(R) inhibited the
M1 receptor to a considerable extent forcing us to discontinue the
examination of this scaffold. Conversely, the methyl free congener
11-(R) showed only poor affinity for the M1 receptor. Additional N-
5 derivatives confirmed the close relationship to the xanthines,
giving similarly potent DPP-4 inhibitors. Compounds 13-(R) and
13-(S) exemplify this further by the former enantiomer being
about 8-fold more potent than the latter, a trend generally seen
within the xanthine series.
Based on the structure of compound 11-(R), the impact of differ-
ent amino groups at C-2 on the inhibitory potency was studied (Ta-
ble 3). Compounds 11-(R) and 17–20, bearing the most potent
amino residues discovered in the xanthine project, showed potent
inhibition of DPP-4 throughout and low affinity for the M1 recep-
tor. The DPP-4 inhibition results in rats confirmed the superiority
of 3-amino-piperidinyl over the other amino substituents: the
residual inhibition of DPP-4 7 h after oral administration with all
compounds but 11-(R) had declined considerably from earlier time
points.
Table 4
Preliminary pharmacological and kinetic data of 11-(R)
hERG, current remaining at 1 lM
Cyp450, 3A4/2D6/2C9/2C19/1A2
Human liver cytosol t1/2
Human liver microsome t1/2
CL (rat)
88%
>50 lM
>90 min
>90 min
56 mL/min/kg
Vss (rat)
40 L/kg
MRT (rat)
8.8 h
Foral (rat)
15%
DPP-4 inhibition in rats, ex vivo
71% after 24 h (10 mg/kg po)
the long-lasting strong DPP-4 inhibition (>70% after 24 h after
compound administration) is particularly noteworthy. These find-
ings support a more detailed study of the 3,5-dihydro-imidazo[4,5-
d]pyridazin-4-one class and of compound 11-(R) in particular, as
potential candidates for development in the treatment of type 2
diabetes.
In conclusion, we have shown that variations of the xanthine
scaffold used in the clinical development compound BI 1356,
which retain the spatial substituent alignment, lead to alternative,
potent DPP-4 inhibitor classes such as the 3,5-dihydro-imi-
dazo[4,5-d]pyridazin-4-ones. One representative of the latter class,
compound 11-(R), has been further advanced and shows a very
promising activity profile, demonstrating that this series is an
appealing gateway to new, highly potent DPP-4 inhibitors.
References and notes
1. McIntosh, C. H. S. Front. Biosci. 2008, 13, 1753.
2. Kieffer, T. J.; McIntosh, C. H. S.; Pederson, R. A. Endocrinology 1995, 136, 3585.
3. (a) Thorens, B. Diabetes Metab. 1995, 21, 311; (b) Meier, J. J.; Nauck, M. A.;
Schmidt, W. E.; Gallwitz, B. Regul. Pept. 2002, 107, 1.
Examination of the residues of the imidazopyridazones con-
firmed BI 1356’s highly favorable arrangement for closely related
scaffolds yielding compound 11-(R) as one of the most promising
ones. 11-(R) was submitted to advanced testing including preli-
minary experiments on pharmacokinetic parameters in rats (Table
4). Additionally, compound 11-(R) was investigated in various in
vitro receptor binding and enzyme assays that indicated no known
liabilities at test concentrations of 3 lM.15 Among the data listed,
4. Murphy, K. G.; Dhillo, W. S.; Bloom, S. R. Endocrine Rev. 2006, 27, 719.
5. Egan, J. M.; Bulotta, A.; Hui, H.; Perfetti, R. Diabetes Metab. Res. Rev. 2003, 19,
115.
6. Augustyns, K.;Van derVeken, P.;Haemers, A. ExpertOpin. Ther. Pat. 2005, 15, 1387.
7. (a) Eckhardt, M.; Langkopf, E.; Mark, M.; Tadayyon, M.; Thomas, L.; Nar, H.;
Pfrengle, W.; Guth, B.; Lotz, R.; Sieger, P.; Fuchs, H.; Himmelsbach, F. J. Med.
Chem. 2007, 50, 6450; (b) Thomas, L.; Eckhardt, M.; Langkopf, E.; Tadayyon, M.;
Himmelsbach, F.; Mark, M. J. Pharmacol. Exp. Ther. 2008, 325, 175.
8. Eckhardt, M.; Hauel, N.; Langkopf, E.; Himmelsbach, F. Tetrahedron Lett. 2008,
49, 1931.

3162
M. Eckhardt et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3158–3162
9. Hauel, N.; Himmelsbach, F.; Langkopf, E.; Eckhardt, M.; Maier, R.; Mark, M.;
Tadayyon, M.; Kauffmann-Hefner, I. WO 2004050658, 2004.
10. Eckhardt, M.; Himmelsbach, F.; Langkopf, E.; Thomas, L.; Tadayyon, M. WO
2006000354, 2006.
H.; Kira, K.; Yasuda, N.; Nagakura, T.; Yamazaki, K. WO 2003104229, 2003;
(b) Kurukulasuriya, R.; Rohde, J. J.; Szczepankiewicz, B. G.; Basha, F.; Lai, C.;
Jae, H.-S.; Winn, M.; Stewart, K. D.; Longenecker, K. L.; Lubben, T. W.;
Ballaron, S. J.; Sham, H. L.; von Geldern, T. W. Bioorg. Med. Chem. Lett. 2006,
16, 6226.
11. For synthesis of compound 10 see Ref. 10.
12. For experimental DPP-4 and M1 activity determination see supporting
information of Ref. 7a.
14. In vitro inhibition data of rat DPP-4 were comparable to the results obtained
for the human enzyme.
13. Imidazopyridazones with deviating structure have been reported as DPP-4
inhibitors in: (a) Yoshikawa, S.; Emori, E.; Matsuura, F.; Richard, C.; Ikuta,
15. Enzyme set Diversity profile and receptor set ExpresSProfile offered by Cerep
(Poitiers), France.