Rational design of a novel, potent, and orally bioavailable cyclohexylamine DPP-4 inhibitor by application of molecular modeling and X-ray crystallography of sitagliptin

Article abstract of DOI:10.1016/j.bmcl.2007.03.095

Molecular modeling was used to design a rigid analog of sitagliptin 1. The X-ray crystal structure of sitagliptin bound to DPP-4 suggested that the central β-amino butyl amide moiety could be replaced with a cyclohexylamine group. This was confirmed by structural analysis and the resulting analog 2a was synthesized and found to be a potent DPP-4 inhibitor (IC₅₀ = 21 nM) with excellent in vivo activity and pharmacokinetic profile.

Full text of DOI:10.1016/j.bmcl.2007.03.095

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3384–3387
Rational design of a novel, potent, and orally
bioavailable cyclohexylamine DPP-4 inhibitor by application
of molecular modeling and X-ray crystallography of sitagliptin
Tesfaye Biftu,a,* Giovanna Scapin,a Suresh Singh,a Dennis Feng,a Joe W. Becker,a
George Eiermann,b Huaibing He,a Kathy Lyons,a Sangita Patel,a Aleksandr Petrov,b
Ranabir Sinha-Roy,c Bei Zhang,c Joseph Wu,c Xiaoping Zhang,c George A. Doss,d
Nancy A. Thornberryc and Ann E. Webera
aDepartment of Medicinal Chemistry, Merck Research Laboratories, Merck & Co., Inc., PO Box 2000, Rahway, NJ 07065, USA
bDepartment of Pharmacology, Merck Research Laboratories, Merck & Co., Inc., PO Box 2000, Rahway, NJ 07065, USA
cDepartment of Metabolic Disorders, Merck Research Laboratories, Merck & Co., Inc., PO Box 2000, Rahway, NJ 07065, USA
dDepartment of Drug Metabolism, Merck Research Laboratories, Merck & Co., Inc., PO Box 2000, Rahway, NJ 07065, USA
Received 27 February 2007; revised 23 March 2007; accepted 29 March 2007
Available online 2 April 2007
Abstract—Molecular modeling was used to design a rigid analog of sitagliptin 1. The X-ray crystal structure of sitagliptin bound to
DPP-4 suggested that the central b-amino butyl amide moiety could be replaced with a cyclohexylamine group. This was confirmed
by structural analysis and the resulting analog 2a was synthesized and found to be a potent DPP-4 inhibitor (IC50 = 21 nM) with
excellent in vivo activity and pharmacokinetic profile.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Diabetes is a major health problem that is growing
rapidly. It is estimated that there are over 170 million
patients with type 2 diabetes worldwide.1 The incretin
hormones glucagon-like peptide 1 (GLP-1) and glucose
dependent insulinotropic polypeptide (GIP) play impor-
tant roles in glucose homeostasis.2,3 GLP-1 and GIP
have been evaluated as potential antidiabetic agents.4,5
However, both hormones are inactivated rapidly
in vivo through the action of dipeptidyl peptidase IV
(DPP-4), a serine exopeptidase which cleaves a dipeptide
from the N-terminus.6,7 Development of DPP-4 resis-
tant GLP-1 agonists, such as exenatide,8 circumvents
this problem. An alternative approach is to use small-
molecule inhibitors of DPP-4 to prolong the beneficial
effects of endogenous GLP-1,9,10 as well as to stabilize
GIP.11 Human clinical trials of several small molecule
DPP-4 inhibitors, including sitagliptin12–14 and vildag-
liptin,15 have shown improved glucose tolerance in
patients with diabetes. In addition, DPP-4 inhibitors
are not likely to cause hypoglycemia or body-weight
gain, and could potentially alter the progression of dia-
betes by restoring b-cell function in the pancreas.
Sitagliptin (1), the first DPP-4 inhibitor approved by the
FDA for the treatment of diabetes, has a trifluorophenyl
group linked to a b-amino butanoyl moiety coupled to a
triazolopiperazine. We were interested in replacing the
F
F
F
NH2
F
F
F
NH2
N
N
N
N
O
N
N
N
N
CF3
CF3
sitagliptin 1
2a
central b-amino butanoyl group with a rigid analog.
However, in the absence of structural information as
to the mode of binding of sitagliptin to the DPP-4
binding site, choice of an appropriate ring system
Keywords: DPP-4; Sitagliptin; Januvia; Cyclohexylamine.
*
Corresponding author. Tel.: +1 732 594 6514; e-mail: Tesfaye_Biftu
@merck.com
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.03.095

T. Biftu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3384–3387
3385
was not obvious. As soon as the crystal structure of
sitagliptin in the DPP-4 binding site was solved (PDB
entry 1 · 70), it appeared that cyclohexylamine could
be an appropriate replacement of the middle b-amino
butanoyl portion of the molecule, providing ring con-
strained analogs such as 2a. This was later confirmed
by molecular modeling and by solving the structure
of 2a bound to DPP-4 (PDB entry 2P8S; Fig. 1A).
Overlaying of 2a (yellow) onto sitagliptin (magenta)
in the DPP-4 binding site shows that the cyclohexyl-
amine group was very well accommodated. The inter-
actions made by sitagliptin with the protein are
retained in the new class of inhibitors: the cyclohexyl-
amine nitrogen hydrogen bonds to the side chains of
Glu205, Glu206, and Tyr662, and the hydrophobic
interactions with the side chain of Phe357 are also
retained. Interestingly, the structure also showed that
the triazolopiperazine moiety in 2a is disordered, that
is, has multiple conformations in the binding site,
although only two conformers were modeled in the
current structure (Fig. 1B). In one of the two conform-
ers (yellow, also shown in panel A), the triazole is
stacked over the side chain of Phe357, and the trifluo-
romethyl group is loosely interacting with the side
chain of Arg358 and Ser209. In the second conformer
(orange), the piperazine is making a side-to-face hydro-
phobic interaction with Phe357, and one of the triazole
nitrogens is part of a hydrogen bonding network that
includes a water molecule, the side chain of His126,
and the main chain oxygen of Glu205 (red lines in fig-
ure). In this paper, we will describe the synthesis and
activity of the novel cyclohexylamine lead 2a.
Figure 1. X-ray co-crystal structure of compound 2a bound to DPP-4
overlaid with sitagliptin.
As shown in Schemes 1 and 2, compounds 2a and 2b
were prepared by reductive amination of ketone 11 and
trifluoromethyltriazolopiperazine used in sitagliptin.14
lithiated with n-butyl lithium and then treated with
1,4-dioxaspiro[4.5]decan-8-one to form the alcohol 3.
Alcohol 3 was dehydrated by treatment with phospho-
rus oxychloride or by treatment with p-toluenesulfonic
acid in toluene with azeotropic removal of water, to pro-
vide styrene 4. Reduction by treatment with hydrogen in
1-Halo-2,4,5-trifluorobenzene was treated with magne-
sium to form the corresponding Grignard reagent or
OH
Ar
1) n-Bu Li or Mg
Ar
POCl3
H2, Pd/C
Ar - Br
Ar -I
O
O
2)
O
or TsOH/toluene
O
O
O
O
4
3
Ar
Ar
Ar
H+
PhIO / TMSN3
OTIPS
base / TIPSOTf
O
O
O
6
7
5
NHBoc
Ar
NHBoc
N3
NH2
1. F-
Ar
Ar
LAH
Ar
protection
OTIPS
2. Chiral Chromatography
OTIPS
O
OTIPS
10
9
8
11a
NH2
NH2
Ar
Ar
1. Decaborane / amine / MeOH
N
N
N
N
N
N
2. HCl/Dioxane
N
N
2a
2b
CF3
CF3
Ar = 2,4,5-trifluorophenyl
Scheme 1. Synthesis of cyclohexylamine DPP-4 inhibitors.

3386
T. Biftu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3384–3387
O
O
O
O
O
with magnesium in methanol and provided the ester 18
as a mixture of cis and trans isomers. Conversion to
the thermodynamically more stable trans isomer 19
was effected by treatment with a base such as sodium
methoxide in methanol. Hydrolysis of the ester with lith-
ium hydroxide to form the acid 20 followed by Curtius
rearrangement in the presence of benzyl alcohol gave
the amine 21, as its benzylcarbamate derivative. Depro-
tection of the ketal by treatment with sulfuric acid in
dioxane provided Intermediate 11b. Compounds 2a
and 2b were prepared by reductive amination of ketone
11b and trifluoromethyltriazolopiperazine used in sitag-
liptin14 and deprotection with hydrogen in the presence
of palladium on carbon.
dimethyl
O
Tf2O
carbonate
O
TfO
O
Hunig's base
O
O
NaH
O
O
15
14
O
O
O
O
Mg
MeOH
Ar
16
B(OH)2
Ar
Ar
Ar
Ar
PdCl2(dppf)
O
O
O
O
DMF, Na2CO3
18
17
O
O
O
OH
LiOH
THF/MeOH
Ar
NaOMe
MeOH
O
O
O
O
19 trans racemic)
20
Compounds 2a and 2b were evaluated in vitro for their
inhibition of DPP-4 activity17 and selectivity against
DASH family members.18 Among them, selectivity
against DPP-8 and DPP-9 was of particular concern
since safety studies using a selective DPP-8/9 dual inhib-
itor suggested that inhibition of DPP-8 and/or DPP-9 is
associated with profound toxicity in preclinical spe-
cies.19 DPP-8, DPP-9, FAP, and quiescent cell proline
dipeptidase (QPP, also known as DPP-7)17,20 inhibition
data are also given in Table 1.
NHCbz
H2SO4
NHCbz
Ar
Et3N, DPPA
O
dioxane
BnOH, toluene
O
O
21
11b
1. Decaborane / Amine / MeOH
2a + 2b
2. Pd-C / H2
Scheme 2. Synthesis of cyclohexylamine DPP-4 inhibitors.
the presence of palladium on carbon as a catalyst gave
the 4-aryl substituted cyclohexane 5. Deprotection un-
der acidic condition gave the cyclohexanone 6, which
was then converted to triisopropylsilyl enol ether 7.
The enol ether 7 upon treatment with iodosobenzene
and trimethylsilyl azide formed the azido cyclohexene
8, which upon reduction to amine with lithium alumi-
num hydride yielded amine 9, as a mixture of cis and
trans isomers. The desired trans isomer was separated
by chromatography on silica. Protection of the resulting
amine as its BOC derivative by treatment with di-tert-
butyl dicarbonate gave 10. Treatment of 10 with a
source of fluoride anion removed the silyl group and
gave the racemic trans-11a which was resolved on chiral
column to yield the desired 1S,2R isomer. The disposi-
tion of substituents around the cyclohexyl ring (axial
vs equatorial) was deduced from the vicinal 1H–1H split-
ting and was confirmed by NOE data. The NMR signal
assignments were obtained by 2D NOESY, COSY, and
HSQC experiments.16
As shown in Table 1, isomer 2a where all substituents
are in equatorial positions has similar potency
(IC50 = 21 nM) to sitagliptin (IC50 = 18 nM) and was
also selective in the counter-screening assays. However,
when the triazolopiperazine group is in the axial posi-
tion, 2b was about an order of magnitude
(IC50 = 140 nM) less potent than sitagliptin. The two
left-hand side rings are similar in both molecules, and
that explains the partial activity of 2b.
The pharmacokinetic properties of 2a in male Sprague–
Dawley rats, dogs, and rhesus monkeys are summarized
in Table 2. Cyclohexylamine 2a has low to moderate
plasma clearance, excellent oral bioavailability, and long
half-life in rats, dogs, and monkeys.
Table 2. Pharmacokinetic parameters of compound 2a
Species
Clp t1/2
(mL/min/kg) (h)
Foral Cmax AUC
(%) (lM) (lM h kg mg)
An alternative method to prepare Intermediate 11b is
shown in Scheme 2. The commercially available 1,4-
dioxaspiro[4.5]decan-8-one was treated with dimethyl
carbonate to form the keto ester 14, which was then
transformed into the enol triflate 15 by treatment with
trifluoromethanesulfonic anhydride. Treatment of 15
with 3,4,5-trifluorophenylboronic acid 16 gave the aryl
cyclohexene 17. Reduction of 17 was readily achieved
Rat
Dog
7.6
3.9
6.6 100
1.4
6.8
10.1
3.9
18
14
95
94
0.93
0.56
Monkey 9.5
Clp, plasma clearance; Foral, oral bioavailability; Cmax, peak plasma
concentration after indicated oral dosing. Pharmacokinetic parameters
were obtained following an IV (1 mg/kg) or po (2 mg/kg) dose
(amorphous dihydrochloride salt) in water.
Table 1. Activities of cyclohexylamine DPP-4 inhibitors
Compound
DPP-4 IC50 (nM)
DPP-8 IC50 (lM)
DPP-9 IC50 (lM)
QPP IC50 (lM)
FAP IC50 (lM)
1
18
21
48
>100
>100
>100
>100
>100
>100
>100
>100
>100
60
>100
2a
2b
140

T. Biftu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3384–3387
3387
10. Pederson, R. A.; White, H. A.; Schlenzig, D.; Pauly, R.
P.; McIntosh, C. H.; Demuth, H.-U. Diabetes 1998, 47,
1235.
11. Gautier, J. F.; Fetita, S.; Sobngwi, E.; Salaun-Martin, C.
Diabetes Metab. 2005, 31, 233.
12. Herman, G. A.; Bergman, A.; Stevens, C.; Kotey, P.; Yi,
B.; Zhao, P. L.; Dietrich, B.; Golor, G.; Schrodter, A.;
Keymeulen, B.; Lasseter, K. C.; Kipnes, M. S.; Snyder, K.;
Hilliard, D.; Tanen, M.; Cilissen, C.; De Smet, M.; de
Lepeleire, I.; Van Dyck, K.; Cilissen, C.; Wang, A. Q.;
Zeng, W.; Davies, M. J.; Tanaka, W.; Holst, J. J.; Deacon,
C. F.; Gottesdiener, K. M.; Wagner, J. A. J. Clin.
Endocrin. Metab. 2006, 91, 4612.
13. Aschner, P.; Kipnes, M.; Lunceford, J.; Sanchez, M.;
Mickel, C.; Williams-Herman, D. Diabetes Care 2006, 29,
2632.
14. Kim, D.; Wang, L.; Beconi, M.; Eiermann, G. J.; Fisher,
M. H.; He, K.; Hickey, G. J.; Kowalchick, J. E.; Leiting,
B.; Lyons, K.; Marsilio, F.; McCann, M. E.; Patel, R. A.;
Petrov, A.; Scapin, G.; Patel, S. B.; Roy, R.; Wu, J.;
Wyvratt, M. J.; Zhang, B. B.; Zhu, L.; Thornberry, N. A.;
Weber, A. E. J. Med. Chem. 2005, 48, 141.
In pharmacodynamic studies, oral glucose tolerance
tests (OGTT) in lean male mice were conducted to deter-
mine the efficacy of 2a. The glucose AUC was deter-
mined from 0 to 120 min. Percent inhibition values for
each treatment were generated from the AUC data nor-
malized to the water-challenged controls. Compound 2a
significantly reduced blood glucose excursions in a
dose-dependent manner as follows: 0.03 mg/kg (26%
inhibition), 0.1 mg/kg (48% inhibition), 0.3 mg/kg
(53% inhibition), 1 mg/kg (60% inhibition), 3 mg/kg
(54% inhibition). The mouse DPP-4 IC50 is 8 nM. In a
separate OGTT of 2a, plasma DPP-4 inhibition, com-
pound concentration, and active GLP-1 level were mea-
sured 20 min after dextrose challenge. At a dose of
0.1 mg/kg, the corresponding plasma concentration of
2a reached 27 nM, and 94% inhibition of plasma DPP-
4 activity was observed, resulting in a 3.5-fold increase
in active GLP-1 levels, and full efficacy in glucose
reduction.
15. Ahren, B.; Landin-Olsson, M.; Jansson, P. A.; Svensson,
M.; Holmes, D.; Schweizer, A. J. Clin. Endocrinol. Metab.
2004, 89, 2078.
In summary, compound 2a, designed by molecular mod-
eling of the X-ray crystal structure of sitagliptin, has
similar potency to sitagliptin, excellent pharmacokinetic
properties across species, and an excellent profile in an
OGTT assay. Further investigation of this lead is
continuing.
1
16. (a) All new compounds were characterized by H NMR
and LC–MS prior to submission for biological evaluation.
(b) All multiple determinations of the IC50 values were
within 1.5-fold of the reported average. Compound 12: 1H
NMR (600 MHz, CD3OD) d 7.41 (m, 1H), 7.22 (m, 1H),
4.22 (t, 2H, J = 5.5 Hz), 4.11 (AB, 2H, J = 15.6 Hz), 3.59
(m, 1H), 3.19 (m, 2H), 3.04 (tt, 1H, J = 11.9, 3.4 Hz), 2.97
(br m, 1H), 2.38 (dm, 1H, J = 11.8 Hz), 2.09 (dm, 1H,
J = 12.3 Hz), 2.00 (dq, 1H, J = 13.8, 3.6 Hz), 1.76 (m, 1H),
1.67 (q, 1H, J = 11.8 Hz), 1.58 (dq, 1H, J = 3.4, 12.3 Hz).
Acknowledgments
We thank Dr. Phil Eskola, Regina Black, Mark Levorse,
Joe Leone, Bob Frankshun, Amanda Makarewicz, Dan-
iel Kim, and Dr. Derek Von Langen of Synthetic
Services Group for large scale synthetic support, and
Dr. Bernard Choi and Eric Streckfuss of Analytical Sup-
port Group for open access LC–MS service.
1
Compound 13: H NMR (600 MHz, CD3OD) d 7.35 (m,
1H), 7.16 (m, 1H), 4.28 (m, 2H), 4.04 (d, 1H, J = 15.7 Hz),
3.96 (d, 1H, J = 15.7 Hz), 3.77 (dt, 1H, J = 2.7, 11.8 Hz),
3.16 (m, 1H), 3.09 (br, 1H), 3.06 (m, 1H), 2.88 (quintet,
1H, J = 3.0 Hz), 2.54 (dq, 1H, J = 14.0, 2.9 Hz), 2.28 (dm,
1H, J = 14.1 Hz), 1.98 (qm, 1H, J = 12.4 Hz), 1.78 (ddd,
1H, J = 2.6, 13.5, 14.4 Hz), 1.74 (m, 1H), 1.70 (m, 1H).
17. Leiting, B.; Pryor, K. D.; Wu, J. K.; Marsilio, F.; Patel, R.
A.; Craik, C. S.; Ellman, J. A.; Cummings, R. T.;
Thornberry, N. A. Biochem. J. 2003, 371, 525.
References and notes
1. WHO, 2006, <http://www.who.int/diabetes/en>.
2. (a) Holst, J. J. Gastroenterology 1994, 107, 1048; (b)
Orsakov, C. Diabetologia 1992, 35, 701.
18. Rosenblum, J. S.; Kozarich, J. W. Curr. Opin. Chem. Biol.
2003, 7, 1.
3. Drucker, D. J. Diabetes 1998, 47, 159.
19. Lankas, G. R.; Leiting, B.; Sinha Roy, R.; Eiermann, G.
J.; Beconi, M. G.; Biftu, T.; Chan, C.-C.; Edmondson, S.;
Feeney, W. P.; He, H.; Ippolito, D. E.; Kim, D.; Lyons, K.
A.; Ok, H. O.; Patel, R. A.; Petrov, A. N.; Pryor, K. A.;
Qian, X.; Reigle, L.; Woods, A.; Wu, J.; Zaller, D.; Zhang,
X.; Zhu, L.; Weber, A. E.; Thornberry, N. A. Diabetes
2005, 54, 2988.
20. Maes, M.-B.; Lambier, A.-M.; Gilany, K.; Senten,
K.; Van der veken, P.; Leiting, B.; Augustyns, K.;
Scharpe, S.; De Meester, I. Biochem. J. 2005, 386,
315.
4. Kieffer, T. J.; Habener, J. F. Endocr. Rev. 1999, 20, 876.
5. Vilsboll, T.; Holst, J. J. Diabetologia 2004, 47, 357.
6. Kieffer, T. J.; McIntosh, C. H. S.; Pederson, T. A.
Endocrine 1995, 136, 3585.
7. Deacon, C. F.; Nauck, M. A.; Toft-Nielson, M.; Pridal,
L.; Willms, B.; Holst, J. J. Diabetes 1995, 44, 1126.
8. Heine, R. J.; Van Gaal, L. F.; Johns, D.; Mihm, M. J.;
Widel, M. H.; Brodows, R. G. Ann. Intern. Med. 2005,
143, 559.
9. Deacon, C. F.; Hughes, T. E.; Holst, J. J. Diabetes 1998,
47, 764.