Discovery of dipeptidyl peptidase IV (DPP4) inhibitors based on a novel indole scaffold

Article abstract of DOI:10.1016/j.cclet.2014.03.047

Dipeptidyl peptidase IV (DPP4) inhibitors are proven in the treatment of type 2 diabetes. We designed and synthesized a series of novel indole compounds that selectively inhibit the activity of DPP4 over dipeptidyl peptidase 9 (DPP9) (>200 fold). We further co-crystallized DPP4 with indole sulfonamide (compound 1) to confirm a proposed binding mode. Good metabolic stability of the indole compounds represents another positive attribute for further development.

Full text of DOI:10.1016/j.cclet.2014.03.047

Accepted Manuscript
Title: Discovery of dipeptidyl peptidase IV (DPP4) inhibitors
based on a novel indole scaffold
Author: Peng-Fei Xiao Rui Guo Shao-Qiang Huang
Heng-Jun Cui Sheng Ye Zhi-Yuan Zhang
PII:
S1001-8417(14)00149-1
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.cclet.2014.03.047
CCLET 2933
To appear in:
Chinese Chemical Letters
Received date:
Revised date:
Accepted date:
11-2-2014
14-3-2014
17-3-2014
Please cite this article as: P.-F. Xiao, R. Guo, S.-Q. Huang, H.-J. Cui, S. Ye, Z.-Y. Zhang,
Discovery of dipeptidyl peptidase IV (DPP4) inhibitors based on a novel indole scaffold,
Chinese Chemical Letters (2014), http://dx.doi.org/10.1016/j.cclet.2014.03.047
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Graphical Abstract
Discovery of dipeptidyl peptidase IV (DPP4) inhibitors based on a novel
indole scaffold
†
Peng-Fei Xiao a,b, , Rui Guo c,†, Shao-Qiang Huang b, Heng-Jun Cui c, Sheng Ye c,∗ Zhi-Yuan
Zhang a,b,
*
a Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing 100730, China
b National Institute of Biological Sciences (NIBS), Beijing 102206, China
c Life Sciences Institute, Zhejiang University, Hangzhou 310058, China
H2
N
Cl
Cl
N
O
S
O
Indole sulfonamide 1
DPP4 inhibitor IC50 = 232.8nm/L
The story of our effort in the de novo design and development of a series of potent and selective DPP4 inhibitors based on an indole
scaffold utilizing structure-based drug design (SBDD) technology.
† These authors contributed equally to this work.
∗ Corresponding author.
E-mail addresses: zhangzhiyuan@nibs.ac.cn, sye@zju.edu.cn
Page 1 of 7

Original article
Discovery of dipeptidyl peptidase IV (DPP4) inhibitors based on a novel
indole scaffold
†
Peng-Fei Xiao a,b, , Rui Guo c,†, Shao-Qiang Huang b, Heng-Jun Cui c, Sheng Ye c,∗ Zhi-Yuan
Zhang a,b,
*
a Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing 100730, China
b National Institute of Biological Sciences (NIBS), Beijing 102206, China
c Life Sciences Institute, Zhejiang University, Hangzhou 310058, China
ARTICLE INFO
ABSTRACT
Article history:
Dipeptidyl peptidase IV (DPP4) inhibitors are proven in the treatment of type 2 diabetes.
We designed and synthesized a series of novel indole compounds that selectively inhibit the
activity of DPP4 over dipeptidyl peptidase 9 (DPP9) (>200 fold). We further co-crystallized
DPP4 with indole sulfonamide (compound 1) to confirm a proposed binding mode. Good
metabolic stability of the indole compounds represents another positive attribute for further
development.
Received 11 February 2014
Received in revised form 14 March 2014
Accepted 17 March 2014
Available online
Keywords:
DPP4 inhibitor
Type 2 diabetes
De novo design
Indole scaffold
Crystal structure
1. Introduction
Incretin hormones, including glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like-peptide-1 (GLP-1), stimulate
insulin secretion, inhibit glucagon secretion, delay gastric emptying, and reduce food intake [1]. The rapid increasing plasma levels of
both hormones in response to elevated serum glucose levels are negatively regulated by dipeptidyl peptidase IV (DPP4) [2], a serine
protease expressed in most tissues [3]. DPP4 regulates insulin secretion by the inactivation of GIP and GLP-1 through removing two
amino acids from the N terminus of both hormones [4]. Enhancing the duration of endogenous incretin hormone by inhibiting DPP4
function is now a validated approach in treatment of type 2 diabetes [1]. Several DPP4 inhibitors have been approved by the FDA, such
as sitagliptin [5], saxagliptin [6], linagliptin [7] and alogliptin [8a] while several others are in late stage clinical trials [9, 10].
Structure-based drug design (SBDD) has been extensively used in modern medicinal chemistry for the discovery of many drugs. In
this paper we describe our effort in the de novo design and development of a series of potent and selective DPP4 inhibitors based on an
indole scaffold utilizing this technology.
2. Experimental
The synthesis of compound 1, 2, 8a-8f and 11a-f (Scheme 1) began with 1-(1H-indol-3-yl)-N,N-dimethyl methylamine (3) [11].
Compound 3 was treated with MeI in THF, followed by heating with potassium phthalimide in anhydrous DMF at 150 oC to generate 4
[12]. Compound 5 were prepared from 4 using pyridinium tribromide at -10 oC. The intermediate compound 6 was synthesized through
Suzuki coupling reaction of compound 5 with 2,4-dichlorophenyl boronic acid in a mixture of toluene and EtOH at 105 oC.
Compounds 8a-8f and 1 were obtained by reacting 5 with R1-sulfonyl or acyl chloride, followed by the removal of the phthalimide
group using NH2NH2 in EtOH at room temperature. Similarly, 11a-11f were readily synthesized in three steps including Suzuki
coupling reaction between compound 5 and different phenyl boronic acids, sulfonamide formation using different alkyl sulfonyl
chloride, and the removal of the phthalimide group.
Detailed synthesis procedure and characterization data of the synthesized compounds can be found in Supporting information.
† These authors contributed equally to this work.
∗ Corresponding author.
E-mail addresses: zhangzhiyuan@nibs.ac.cn, sye@zju.edu.cn
Page 2 of 7

3. Results and discussion
Before the initiation of the de novo design, we first studied the structure of several known DPP4-inhibitor complexes [8a-d] and
identified several key interactions between the protein and the inhibitors. As shown in Fig. 1, alogliptin tightly binds in the active site
of DPP4 through several important interactions. The cyanobenzyl group fits in the P1 hydrophobic pocket formed by Val656, Tyr631,
Try662, Trp659, Tyr666 and Val711. The cyano group also forms a hydrogen bond with Arg125. Another key interaction is a salt bridge
interaction between the aminopiperidine in alogliptin and residues Glu205/Glu206 in DPP4. The 4-carbonyl oxygen in the
primidindione interacts with the backbone NH of Tyr631 through a hydrogen bond, while the pyrimidindione ring itself forms π-stack
interaction with the phenyl ring on the Tyr547 residue. From DPP4-alogliptin and other co-crystal structures, we recapitulated the
following important interactions: (a) hydrophobic filling in the P1 pocket, (b) salt bridge to Glu205/Glu206, (c) π-stacking and
hydrogen bond with NH of Tyr631or OH of Ser630. Rather than starting from a screening hit, we decided to use the information
described above to de novo design new chemical structures that could capture all or most important interactions with DPP4. An indole
scaffold having certain functional groups was one of our designs. As schematically shown in Fig. 2, a methylamine motif at C-3
provides a salt bridge to E205/E206, the phenyl group fills the P1 pocket, while an acyl group or alkyl sulfonyl group at N-1 could
form one or two potential hydrogen bonds to Ser630, Tyr547 and Tyr361.
To evaluate the design, we first synthesized methylsulfonyl derivative 1 and acetyl derivative 2, which showed inhibitory activity
against DPP4 at an IC50 concentration of 232 nmol/L and 38% inhibition at 10 umol/L, respectively (Table 1). Furthermore, compound
1 also showed more than 200 fold of selectivity over DPP9. According to our modeling results, the sulfonyl group in compound 1
could form more favorable hydrogen bonds with Ser630 and Tyr547 residues of DPP4, thus explained the better activity of compound
1. To verify our modeling results and our hypotheses, we co-crystallized compound 1 and DPP4 (Table 2). The detail structural
information (Fig. 3) revealed the mode of binding of compound 1 with DPP4 in the active site. The indole ring forms a cation-π
interaction with the guanidinium moiety of Arg125, the methylamine motif forms salt bridges to Glu205/Glu206, and the
2,4-dichlorobenzene group effectively fits in the P1 pocket. Comparing the proposed binding mode and the real crystal structure, the
inhibitor conformation was as predicted except the amine orientation. The proposed hydrogen bond between the sulfonyl group and Tyr
547 and the orientation of the methylsulfonyl group are in good agreement with the crystal structure. However, the residue Tyr547 of
DPP4 rotates 90 degrees in the co-crystal structure thus differed from the structure of DPP4-Alogliptin and other DPP4 complexes,
which causes the loss of the hydrophobic interactions.
The encouraging results motivated us to develop structure activity relationship (SAR) both on the alkyl sulfonyl and Ar groups.
Based on the analysis of our crystal structure, we decided to modify the R1 group on the sulfonyl group to generate more interactions
with the phenyl ring of Tyr547 in order to improve the potency. According to the SAR shown in Table 1, similar to compound 1
ethylsulfonyl (8a), isopropylsulfonyl (8c) and isobutylsulfonyl (8d) analogs demonstrate good potency against DPP4, while the
propylsulfonyl (8c) and cyclopropylsulfonyl (8d) analogs are relatively weak. Our interpretation for the SAR is that the hydrogen bond
between the indole compound and the Tyr547 is not as strong comparing with the hydrogen bond between Alogliptin and Tyr631, while
the phenol group in Tyr547 is more flexible compared with the backbone NH in the Tyr631. Therefore increasing the π-stack
interactions with the Tyr547 might not be beneficial.
The P1 pocket of DPP4 is the most important area utilized by others in their inhibitor design; therefore our second optimization
study was focused on modifying the Ar group on the indole core. With knowledge learned from others [8a-d], the 2,4-di-Cl phenyl
group was used in our first inhibitor design (compound 1). To understand SAR at this position, compounds with mono or di-
substituted phenyl group (11a-11f) were synthesized. The inhibition testing results (Table 3) showed that only 2,4-disubstituted phenyl
compounds (8c and 11d) were tolerated and all other modifications caused significant activity loss, which indicated that the P1 pocket
might be rigid and sensitive to the substituents on the phenyl ring.
In general, the indole sulfonamide analogues are stable in the liver microsome metabolic stability test (see Table 1), indicating that
indole sulfonamide is a suitable scaffold for further development. DPP9 belongs to the same family of DPP4 and blocking its function
could cause some undesired side effects [13]. Therefore the most potent compounds (8a-8e and 1) were also tested against DPP9
enzymatic activity, and more than 200 fold selectivity has been achieved for DPP4 over DPP9.
4. Conclusion
In conclusion, we have de novo designed and developed a series of indole DPP4 inhibitors using computer modeling and known
DPP4-inhibitor complex structures. The proposed binding mode of the indole compound in the active pocket overlapped well with that
observed from the co-crystal structure of DPP4 and compound 1. Preliminary SAR around indole core has been developed and a series
of potent and selective DPP4 inhibitors with favorable physical properties were obtained. Further optimization around the core is in
progress.
Acknowledgments
We thank Yongfen Ma and Xiao Liu for experimental assistances; and Sheng Huang and Jianhua He at the Shanghai Synchrotron
Page 3 of 7

Radiation Facility (SSRF) for on-site assistance. This work was supported in part by funds from the Ministry of Science and
Technology (No. 2014CB910300), the Natural Science Foundation of Zhejiang Province (No. R2100439) and the Specialized Research
Fund for the Doctoral Program of Higher Education (No. 20110101110122).
References
[1] D.J. Drucker, M.A. Nauck, The incretin system: glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4 inhibitors in type 2 diabetes, The
Lancet 368 (2006) 1696–1705.
[2] G.A. Herman, P.P. Stein, N.A. Thornberry, et al., Dipeptidyl peptidase-4 inhibitors for the treatment of type 2 diabetes: focus on sitagliptin practice, Clin.
Pharmacol. Ther. 81 (2007) 761-767.
[3] Y.L. He, A Horowitz, C.E Watson, et al., Vildagliptin does not affect C-peptide clearance in patients with type 2 diabetes, J. Clin. Pharmacol. 47 (2007)
127-131.
[4] E.M. Migoya, R. Bergeron, J.L. Miller, et al., Dipeptidyl peptidase-4 inhibitors administered in combination with metformin result in an additive increase in
the plasma concentration of active GLP-1, Clin. Pharmacol. Ther. 88 (2010) 801-808.
[5] (a) D. Kim, L.P. Wang, M. Beconi, et al., (2R)-4-Oxo-4-[3-(trifluoromethyl)-5,6-dihydro[1,2,4]triazolo[4,3-a]pyrazin-7(8H)-yl]-1-
(2,4,5-trifluorophenyl)butan-2-amine: a potent, orally active dipeptidyl peptidase IV inhibitor for the treatment of type 2 diabetes, J. Med. Chem. 48 (2005)
141−151;
(b) N.A. Thornberry, A.E. Weber, Discovery of JANUVIA (Sitagliptin), a selective dipeptidyl peptidase IV inhibitor for the treatment of type 2 diabetes,
Curr. Top. Med. Chem. 7 (2007) 557–568;
(c) K.A. Lyseng-Williamson, Sitagliptin, Drugs 67 (2007) 587–597.
[6] (a) D.J. Augeri, J.A. Robl, D.A. Betebenner, et al., Discovery and preclinical profile of saxagliptin (BMS-477118): a highly potent, long-acting, orally active
dipeptidyl peptidase IV inhibitor for the treatment of type 2 diabetes, J. Med. Chem. 48 (2005) 5025−5037;
(b) A.A. Tahrani, M.K. Piya, A.H. Barnett, Saxagliptin: a new DPP-4 inhibitor for the treatment of type 2 diabetes mellitus, Adv. Ther. 26 (2009) 249–262;
(c) C.F. Deacon, J.J. Holst, Saxagliptin: a new dipeptidyl peptidase-4 inhibitor for the treatment of type 2 diabetes, Adv. Ther. 26 (2009) 488–499.
[7]M. Eckhardt, E. Langkop, M. Mark, M. Tadayyon, et al. 8-[(3R)-3-aminopiperidin-1-yl]-7-(but-2-yn-1-yl)-3-methyl-1-[(4-methylquinazolin-2-yl)methyl]-3,
7-dihydro-1H-purine-2,6-dione a highly potent, selective, long-acting, and orally bioavailable DPP-4 inhibitor for the treatment of type-2 diabetes. J.
Med.Chem. 50 (2007) 6450−6453.
[8] (a) J. Feng, Z.Y. Zhang, M.B. Wallace, et al., Discovery of alogliptin: a potent, selective, bioavailable, and efficacious inhibitor of dipeptidyl peptidase IV, J.
Med. Chem. 50 (2007) 2297-2300;
(b) Z.Y. Zhang, M.B. Wallace, J. Feng, et al., Design and synthesis of pyrimidinone and pyrimidinedione inhibitors of dipeptidyl peptidase IV, J. Med. Chem. 54
(2011) 510-524;
(c) M.B. Wallace, J. Feng, Z.Y. Zhang, et al., Structure-based design and synthesis of benzimidazole derivatives as dipeptidyl peptidase IV inhibitors, Bioorg.
Med. Chem. Lett. 18 (2008) 2362-2367;
(d) B. Lam, Z.Y. Zhang, J.A. Stafford, et al., Structure-based design of pyridopyrimidinediones as dipeptidyl peptidase IV inhibitors, Bioorg. Med. Chem. Lett.
22 (2012) 6628-6631.
[9] (a) N. Ervinna, T. Mita, E. Yasunari, et al., Anagliptin, a DPP-4 inhibitor, suppresses proliferation of vascular smooth muscles and monocyte inflammatory
reaction and attenuates atherosclerosis in male apo E-deficient mice, Endocrinology 154 (2013) 1260−1270;
(b) M.K. Kim, Y.N. Chae, H.D. Kim, et al., DA-1229, a novel and potent DPP4 inhibitor, improves insulin resistance and delays the onset of diabetes, Life Sci.
90 (2012) 21−29.
[10] M. Brunavs, P. Cowley, S.E Ward, P. Weber, Recent disclosures of clinical candidates, Drugs Future 38 (2013) 127−133.
[11] H.G. Dai, J.T. Li, T.S. Li, Efficient and practical synthesis of mannich bases related to gramine mediated by zinc chloride, Synthetic Communications 36
(2006) 1829–1835.
[12] T. Haruko, M. Minoru, K. Yuichi, Intramolecular photoreactions of phthalimide-alkene systems. Oxetane formation of N-(ω-indol-3-ylalkyl)phthalimides,
Chem. Pharm. Bull. 36 (1988) 2853-2863.
[13] G.R. Lankas, B. Leiting, R.S. Roy, et al., Dipeptidyl peptidase IV inhibition for the treatment of type 2 diabetes, Diabetes 54 (2005) 2988–2994.
Page 4 of 7

Fig. 1. Cocrystal structure of Alogliptin in the DPP4 active site.
Fig. 2. A model of compound 1 in the DPP4 active site showing structure based design.
Page 5 of 7

Fig. 3. Cocrystal structure of 1 in the DPP4 active site.
O
O
H2N
N
Cl
e
N
Cl
d
O
Cl
O
Cl
Cl
N
R1
Cl
N
R1
N
H
6
7a-h
1, 2 and 8a-f
c
O
N
O
N
N
a
b
N
H
O
O
Br
N
H
N
3
4
5
H
f
O
H2N
O
N
R2
O
e
N
g
R2
O
N
O
R2
S
N
O
O
R1
S
O
N
R1
10a-f
H
11a-f
9a-f
Scheme 1. Synthesis of compounds 1, 2 and 8a-f and 11a-f. Reagents and conditions: (a) MeI, THF, potassium pthalimide, DMF, 150 oC for 5 h, 64%; (b) THF,
CHCl3, pyridinium tribromide, -10 oC for 3 h, 56%; (c) 2,4-dichlorophenylboronic acid, Pd(PPh3)4, LiCl, Na2CO3, PhMe, EtOH, 105 oC for 4 h, 30%; (d)
R1-sulfonyl or acyl chloride, NaH, DMF, 0 oC for 16 h, 31%-50%; (e) NH2-NH2, EtOH r.t. for 1 h, 84%-91%; (f) R2-phenyl boronic acid, Pd(PPh3)4, LiCl,
Na2CO3, PhMe, EtOH,105 oC for 4 h, 31%-57%; (g) R1-sulfonyl chloride, NaH, DMF, 0 oC for 16 h, 31%-50%.
Page 6 of 7

Table 1 Selected data for indole scaffold analogues (R1 substitution).
H2N
Cl
Cl
N
R1
Compd. R1
DPP4
IC50 (nmol/L)
232.8
38%@10 umol/L
219.63
375.93
216.39
374.96
214.97
DPP9
IC50 (umol/L)
53
NT
73
104
85
86
110
NT
RLM a
t1/2 (min)
99
HLM b
t1/2 (min)
182
NT
347
99.0
693
347
24.2
NT
MLM c
t1/2 (min)
84.5
NT
46.8
26.5
36.9
50.2
9.33
methylsulfonyl
acetyl
ethylsulfonyl
1
2
8a
8b
8c
8d
8e
8f
NT
73.0
70.7
85.6
66.6
35.5
NT
propylsulfonyl
isopropylsulfonyl
cyclopropylsulfonyl
isobutylsulfonyl
cyclohexylsulfonyl
66%@10 umol/L
NT
a RLM: incubation with rat liver microsomes.
b HLM: incubation with human liver microsomes.
c MLM: incubation with mouse liver microsomes.
NT: not tested.
Table 2 X-ray data and refinement statistics.
Data set
Refinement
Space Group
P32
Resolution (Å)
3.25
Unit Cell (a,b) (Å)
Resolution (Å)
79.8, 286.8
3.25
167,140
32,631
5.1
99.9 (100.0)
29.0 (2.0)
11.0 (100)
31,241
22.7/26.6
11,914
2
No. of reflections |F|>0 σF
R-factor/R-free (%)
Measured reflections
Unique reflections
Redundancy
Completeness (%, highest shell)
Mean I/σI (highest shell)
Rsym (%, highest shell)
No. of protein atoms
No. of compound molecules
No. of water molecules
rmsd bond lengths (Å)
rmsd bond angles (°)
0
0.008
1.17
Table 3 Selected data for indole scaffold analogues (R2 substitution).
NH2
N
R2
O
S
R1
O
Compd.
R1
R2
DPP4
DPP9
IC50 (nmol/L)
IC50 (umol/L)
isopropyl
isopropyl
isopropyl
isopropyl
Me
2,4-diCl
2,3-diCl
2,5-diCl
2-Cl,4-F
2-Cl,4-Me
2-Cl
216.39
85
8c
27%@10umol/L
45%@10umol/L
70%@10umol/L
284.58
34%@10umol/L
>10000
NT
NT
NT
75
NT
NT
11a
11b
11c
11d
11e
11f
isopropyl
isopropyl
4-Cl
NT: not tested.
Page 7 of 7