indole-based glucagon receptor antagonists

Bioorganic & Medicinal Chemistry Letters xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Bioorganic & Medicinal Chemistry Letters journal homepage: www.elsevier.com/locate/bmcl

A novel series of indazole-/indole-based glucagon receptor antagonists Songnian Lin ⇑, Fengqi Zhang, Guoqiang Jiang, Sajjad A. Qureshi, Xiaodong Yang, Gary G. Chicchi, Laurie Tota, Alka Bansal, Edward Brady, Maria Trujillo, Gino Salituro, Corey Miller, James R. Tata, Bei B. Zhang, Emma R. Parmee Early Development and Discovery Science, and Preclinical Development, Merck Research Laboratories, 2015 Galloping Hill Rd, Kenilworth, NJ 07033, United States

a r t i c l e

i n f o

Article history: Received 1 June 2015 Revised 29 July 2015 Accepted 6 August 2015 Available online xxxx This publication is dedicated to Professor Iwao Ojima on the occasion of his 70th birthday

a b s t r a c t A novel, potent series of glucagon receptor antagonists (GRAs) was discovered. These indazole- and indole-based compounds were designed on an earlier pyrazole-based GRA lead MK-0893. Structure– activity relationship (SAR) studies were focused on the C3 and C6 positions of the indazole core, as well as the benzylic position on the N-1 of indazole. Multiple potent GRAs were identified with excellent in vitro profiles and good pharmacokinetics in rat. Among them, GRA 16d was found to be orally active in blunting glucagon induced glucose excursion in an acute glucagon challenge model in glucagon receptor humanized (hGCGR) mice at 1, 3 and 10 mg/kg (mpk), and significantly lowered acute glucose levels in hGCGR ob/ob mice at 3 mpk dose. Ó 2015 Elsevier Ltd. All rights reserved.

Keywords: Glucagon receptor antagonist Indazole Indole Diabetes Glucose lowering

Type 2 Diabetes Mellitus (T2DM) is a growing worldwide epidemic currently affecting an estimated 300 million people.1 Despite a number of diabetic therapies available nowadays, there remains a significant unmet medical need for additional therapies.2 T2DM is partially characterized by elevated and dysregulated hepatic glucose production (HGP). Glucagon is a 29-amino acid peptide that acts as a major counter-regulatory hormone to insulin, and stimulates gluconeogenesis and glycogenolysis in liver and thus increasing HGP.3 Studies in type 2 diabetic subjects have demonstrated a causal role for glucagon in promoting excessive glucose production.4 An inappropriately high rate of HGP is the predominant cause of fasting hyperglycemia and a major contributor to the postprandial hyperglycemia characteristic of T2DM.5 It is therefore postulated that blocking the action of the glucagon receptor would lead to improved glycemic control in T2DM patients.6 With this, there have been many Letters detailing efforts to identify both small and large molecule glucagon receptor

⇑ Corresponding author. Tel.: +1 908 740 0585; fax: +1 908 740 3126.

antagonists (GRAs) as potential treatments for T2DM over the last two decades.6,7 Clinical data on one of the early GRAs, Bay 27-9955 (1) was reported and an acute effect was observed with the compound in blocking glucagon-induced glucose increase in healthy volunteers.8 A series of b-alanine acid containing urea based GRAs such as NNC 25-0926 (2) was later reported by NovoNordisk.9–11 Various scaffold changes replacing the urea core while keeping the b-alanine moiety by many groups have since been reported in patent applications7 and literature (e.g., GRA 3 reported by Pfizer12). Prominent examples of this include several GRA clinical candidates, such as MK-0893 (4, Merck)13 and likely LY2409021 (Eli Lilly).14 With the continued interest in the area of developing novel GRAs as a potential treatment for T2DM, we designed a novel series of GRAs possessing an indazole or an indole/azaindole core (5, Fig. 1) as intrigued by the presence of a unique 5-naphthylpyrozole moiety in the core structure of MK-0893. In this Letter, we would like to discuss the synthesis and the structure–activity relationship (SAR) studies of this series of GRAs, leading to the identification of several potent compounds which demonstrated excellent in vitro and in vivo profiles.

E-mail address: [email protected] (S. Lin). http://dx.doi.org/10.1016/j.bmcl.2015.08.015 0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Lin, S.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.08.015

2

S. Lin et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx F

HO Cl

O

OH

N

N H

Cl

OH

H N

O

O

(+) 1 Bay 27-9955

2 NNC 25-0926

F3C OH

H N

N N

N H

O

O

3 (Pfizer) Cl Cl

Ar1 Ar

N N

X N

Y

OH

H N

MeO

2

R

O

O

4 MK-0893

5

Figure 1. Structures of several key glucagon receptor antagonists.

The synthesis of the indazole-based GRAs is shown in Scheme 1. Treatment of 5-bromo-, 4-bromo-, or 3-bromo-2-fluorobenzaldehydes 6 with methoxyamine in the presence of K2CO3, followed by refluxing with hydrazine in DME afforded corresponding 5-, 6-, or 7-bromoindazoles 7 in good yields. Iodination of 7 with iodine/KOH in DMF provided 3-Iodoindazoles 8 in excellent yields.

MeONH2.HCl, K2CO3, DME, 40 oC; H2N-NH2, reflux

O H

Br F

I I2, KOH

N

Br

60-85%

6

N H

N

Br N H

DMF, rt 76-90%

7

F3C

H 2N

O OH

R

NBS BzOOBz

Cl

9

N

R

17

OEt

83%

N H

R

F3C

Br

11

DMF, 60 oC 2h 71-87%

19 F3C

OEt

H N

R

O

N F3C

O

12

N

Pd(Ph 3P)2Cl2, Ar2-B(OH)2 NaHCO 3, DME/H2O, 90 oC 35-78% (2 steps)

Ar2

OMe

B(OH)2 Pd(dppf)Cl2 CH2Cl2 NaHCO3, DME/H2O, 90 oC 90%

20

N

F3C H N

* R

O

i) Chiral HPLC; ii) NaOH

O

O

N

TMSCHN2 THF/MeOH (1/1)

OtBu

H N

O

Ar1 Pd(Ph 3P)2Cl2, Ar1-B(OH)2 NaHCO 3, DME/H2O; 90 oC

Cs2CO3, DMF, 60 oC, 4 h

I

N

OtBu

11a R = n-Pr 82%

O

N

Br

O

Br

N H

N

I OEt 8, Cs2CO3

18

N H

R

I2, KOH DMF, rt

O

O O

91-100%

10

N H

N F3C

I

CH2Cl2, rt Br

84-95%

O

NaHCO3, DME/H2O, 90 oC quantitiative

N H

EDC, HOBt DIEA

OH

CCl4, reflux 20 min

B(OH) 2

O

Pd(dppf)Cl2 CH2Cl2,

8 O

O

Bromination at the benzylic position of 4-alkylbenzoic acids 9 was achieved with NBS and benzoyl peroxide, and the resulting 10 was coupled with b-alanine ethyl ester to afford amides 11. Indazoles 8 were then alkylated at the N-1 position with benzylbromides 11 in the presence of Cs2CO3 in DMF to provide 12. Suzuki couplings of 12 with Ar1B(OH)2 catalyzed by Pd(Ph3P)2Cl2 preferentially at the 3-postion of the indazole (resulted in the partial hydrolysis of ethyl ester), followed by a second Suzuki coupling with Ar2B(OH)2 at the 5-, 6-, or 7-position of the indazole core under similar conditions were then performed. The sequential Suzuki couplings provided desired bis-substituted indazoles 13–16 in moderate to good yields over two steps. To facilitate the chiral resolution of the racemic compounds, the acids were converted to methyl esters by simple treatment with TMSCHN2 in MeOH/THF, and re-hydrolyzed back to acids after chiral separation.15 To synthesize the 7-azaindole based compound 22, Suzuki coupling of 6-chloro-7-azaindole 17 with 4-CF3O-phenylboronic acid was carried out in the presence of Pd(dppf)Cl2
CH2Cl2 to supply biaryl 18 in quantitative yield (Scheme 2). Iodination at the C-3 position resulted 3-iodo-7-azaindazole 19. Alkylation of the N-1 position of 19 with bromide 11a, prepared similarly to 11 as shown in Scheme 1, provide iodide 20 in 82% yield. Suzuki coupling of 20 with 5-trifluoromethyl-2-methoxyphenylboronic acid afforded tert-butyl ester 21 in high yield, which was hydrolyzed to provide desired acid 22 after chiral resolution.16 Similarly, synthesis of indole-based compound 30 is shown in Scheme 3. Briefly, 6-bromoindole 23 was alkylated at the N-1 position with bromide 24, and the resulting indole 25 was iodinated at the C-3 position to afford 26. Suzuki coupling of 26 with 5-trifluoromethyl-2-methoxyphenylboronic acid afforded methyl eater 27, which was hydrolyzed to acid 28, and coupled with b-alanine ethyl ester to provide 29. The second Suzuki coupling of 29 with substituted phenylboronic acids followed by ester saponification afforded desired compounds 30. The compounds thus synthesized were first evaluated in vitro using both binding and functional assay formats.13 The binding

R' = H R' = Me R' = H

Scheme 1. Synthesis of indazole GRAs 13–16.

OR' O

OMe

13-16 N F3C

O

N *

H N

OR' 21 R' = t-Bu O

22 R' = H

O

Scheme 2. Synthesis of azaindole GRA 22.

Please cite this article in press as: Lin, S.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.08.015

i) Chiral HPLC; ii) TFA, CH2Cl2

3

S. Lin et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx Table 1 SAR on the position of Ar2 on indazole

Br O

Br

N

Br

OMe

24

NIS/DMF

O

N H

68%

23

OMe

25

F3C

60 oC, 4 h

NaH/DMF, rt, 2 h

60%

Ar2

F3C F3C

I

7

B(OH)2 Na 2CO3, DME,

O

60 oC, 12 h

OMe

26

N

Br

O

aq. NaOH

28 R' = H

quantitative

OMe

β-Ala-OEt

78%

N

Position

Ar2

hGCGR binding (nM)

hGCGR cAMP (nM)

13a 13b 13c 13d 13e 13f

C-5 C-6 C-7 C-5 C-6 C-7

4-MeO-Ph 4-MeO-Ph 4-MeO-Ph 4-Me-Ph 4-Me-Ph 4-Me-Ph

10 4.0 1700 7.0 2.0 780

26 11 2700 17 5.0 3800

OEt

H N 29

O

Compound

40%

F3C

OH

H N O

OR'

Br

N

OMe

27 R' = Me

EDC, HOBt DIEA, CH 2Cl2

N

6

OMe

Pd(dppf)Cl2

N

Br

OMe

5

Table 2 hGCGR activity of indole-, 7-azaindole-, and indazole-based GRAs

O

O F3C

1) Ar2-B(OH)2

OMe

Pd(dppf)Cl2 Na 2CO3, DME, 60 oC, 16 h

Ar2

N

H N

2) aq. NaOH, MeOH 72-83%

OH O

O 30

Compound

X/Y

Ar2

hGCGR binding (nM)

hGCGR cAMP (nM)

13b 30aa 13e 30ba 16a 22

N/CH CH/CH N/CH CH/CH N/CH CH/N

4-MeO-Ph 4-MeO-Ph 4-Me-Ph 4-Me-Ph 4-CF3O-Ph 4-CF3O-Ph

4.0 8.0 2.0 4.0 1.0 12

11 130 5.0 48 5.6 360

Scheme 3. Synthesis of indole GRA 30.

affinity of compounds for the human glucagon receptor (hGCGR) was determined by their ability to displace [125I] Glucagon from hGCGR using a membrane preparation from a CHO cell line expressing the human GCGR (CHO hGCGR). The functional activity was assessed by the ability of these compounds to inhibit glucagon-induced cAMP production in a cell line expressing hGCGR. Both binding and functional activities are expressed as values of IC50. All data reported are from the more active enantiomers tested unless otherwise noted. Based on the structure homology of MK-0893 and indazole GRAs designed, it is obvious that the C-3 aryl (Ar1) of 5 is critical for the activity against the hGCGR. However, it is not as clear as to where the second aryl substituent (Ar2) should be placed on the indazole ring. We therefore set out to compare all C-5, C-6, and C-7 positions of indazole. As shown in Table 1, we were pleased to find that all indazole-based GRAs were active in vitro against the hGCGR. However, it is clear from the SAR that C-7 is the least preferred position for the second aryl substitution (13c and 13f). While both C-5 and C-6 afforded GRAs with potent activity, C-6 position seems to be more preferred with an edge over C-5 position (13b vs 13a, and 13e vs 13d). With the identification of the C-6 position being the preferred position for the 2nd aryl substitution, we set out to find out how the indazole core compared to other 6/5 ring systems such as indole or azaindole (Table 2). Interestingly, while head to head comparison of C-3/C-6 substituted indazoles with corresponding indoles showed very similar activity in binding to the hGCGR, indoles exhibited ca. 5-fold weaker potency in the cAMP functional assay (13b vs 30a, and 13e vs 30b). Indazole 16a was found to be a very potent GRA in both binding assay (IC50 = 1 nM) and cAMP functional assay (IC50 = 5.6 nM). However, moving the nitrogen at

a

Compounds were tested as a racemic mixture.

indazole 2 position to 7 position, in another word, replacement of indazole core of 16a with a 7-azaindole core, resulted in significant loss of hGCGR activity (binding IC50 = 12 nM, cAMP IC50 = 360 nM for 7-azaindole indole 22). Obviously, the indazole core is the most preferred scaffold for hGCGR activity among the three heterocycles studied. Aryl substitution on the C-3 position of indazole was then examined for their hGCGR activity (Table 3). In general, most of the aryl groups, including substituted phenyl, pyridinyl, or pyrimidinyl groups, are tolerated. Evidently, phenyl groups substituted with simple halides, a methoxyl group, or a trifluoromethyl group is more likely to provide GRAs with good potency (14a–f, and 14l–m). A polar substitution on the phenyl group is usually less preferred (14g, 14p, and 14q), and often resulted in weaker functional activity (14g and 14q). Similarly, placement of substituted pyridines (14h–k) and pyrimidine (14r) generally afforded compounds with weaker hGCGR activity. Not surprisingly, 14a, an analog of 16a with a 1-MeO-4-CF3O-phen-2-yl group at C-3 position, exhibited best functional activity (cAMP IC50 = 11 nM) among all aryl groups tested, and excellent binding activity (IC50 = 7.0 nM). The 1-MeO-4-CF3O-phen-2-yl group was then selected for the C-3 position when we investigated the aryl substituents at the C-6 position of indazole. As shown in Table 4, C-6 position seems

Please cite this article in press as: Lin, S.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.08.015

4

S. Lin et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx

Table 3 SAR on the C-3 position of indazole

Table 5 SAR on the N-1 benzylic position of indazole F3C

Ar1 N N

OMe

OH

H N

F3CO

N

O

N

O Me

Compound

Ar1

hGCGR binding (nM)

hGCGR cAMP (nM)

14a 14b 14c 14d 14e 14f 14g 14h 14i 14j 14k 14l 14m 14n 14o 14p 14q 14r

1-MeO-4-CF3O-Ph-2-yl 1-MeO-4-Cl-Ph-2-yl 1-MeO-4,5-diF-Ph-2-yl 1-F-4-CF3-Ph-2-yl 1-Cl-4-F-Ph-2-yl 1,2,5-TriF-Ph-3-yl 1-NH2CO-4-Cl-Ph-2-yl 4-MeO-pyridine-3-yl 3-MeO-pyridine-5-yl 2-MeO-pyridine-5-yl 2-MeO-pyridine-3-yl 2,4-DiCl-Ph 3-Cl-5-CF3-Ph 3-F-4-Cl-Ph 3,4,5-TriF-Ph 4-CN-Ph 4-(Me2N)CO-Ph 2-MeO-pyrimidine-4-yl

7.0 3.0 4.0 5.0 10 7.0 2.0 110 50 26 72 8.0 3.0 4.0 6.0 15 40 35

11 36 37 120 140 52 220 760 100 65 77 41 56 44 20 43 210 84

Table 4 SAR on the C-6 position of indazole

OH O

O

Compound

R

hGCGR binding (nM)

hGCGR cAMP (nM)

15b 16b 13e 16c 16d 16e

n-Butyl i-Propyl n-Propyl Ethyl Methyl H

2.6 4.0 2.0 4.0 2.6 39

6.6 69 5.0 21 7.6 140

Table 6 hGIP cAMP activity of selected compounds Compound

hGCGR cAMP (nM)

hGIPR cAMP (nM)

hGIPR/hGCGR

15a 15b 15g 16a 16d

14 6.6 8.0 5.6 7.6

1100 1000 750 1400 1300

79 150 94 250 170

Table 7 Rat pharmacokinetic profiles of selected compoundsa

F 3C

OMe Ar2

H N

R

N N

H N

OH O

O

Compound

Ar2

hGCGR binding (nM)

hGCGR cAMP (nM)

15a 15b 15c 15d 15e 15f 15g 15h 15i 15j 15k 15l

4-MeO-Ph 4-Me-Ph 4-MeS-Ph phenyl 2-Cl-4-MeO-Ph 3-F-4-MeO-Ph 4-CF3O-Ph 4-CF3-Ph 2-MeO-pyridine-5-yl 3-F-5-Cl-Ph 5-Me-thiophen-2-yl 3,4,5-TriF-Ph

1.9 2.6 1.5 3.0 4.0 1.5 0.8 1.6 3.4 7.0 4.0 3.5

14 6.6 9.0 18 18 18 8.0 14 24 46 18 26

to be more tolerant than the C-3 position, as all the aryl groups examined exhibited potent hGCGR activity. Interestingly, even methoxypyridinyl group (15i, binding IC50 = 3.4 nM), which is not well tolerated in C-3 position (14j, binding IC50 = 26 nM), is well accepted. 4-Methoxyphenyl (15a, binding IC50 = 1.9 nM), 4methylphenyl (15b, binding IC50 = 2.6 nM), and 4-trifluoromethylphenyl (15g, binding IC50 = 0.8 nM) groups, are among the most preferred aryl groups identified at the C-6 position. Our attention was then focused on the N-1 benzylic position of indazole. As Table 5 shows, a small alkyl group at this position is clearly preferred, as no substitution resulted in much weaker

Compound

Clp (mL/min/Kg)

Vdss (L/Kg)

t1/2 (h)

AUCN (lM
h
Kg/mg)

F (%)

15a 15b 15g 16a 16d

1.7 3.9 7.5 2.2 6.3

0.67 1.0 0.72 1.3 0.65

6.4 4.0 4.7 9.7 1.5

4.0 1.2 0.30 2.0 2.0

27 17 9.0 19 44

a Wistar rats dosed at 0.5 mg/kg iv and 2 mg/kg po. Clp, plasma clearance; Vdss, volume of distribution at steady state; T1/2, half-life; AUCN, dose normalized area under the plasma concentration versus time curve; F, oral bioavailability.

hGCGR activity (16e, binding IC50 = 39 nM, cAMP IC50 = 140 nM). This is consistent with the findings on the pyrazole-based GRAs exemplified by MK-0893.13 Although there is no clear difference among small alkyl groups investigated for the hGCGR binding activity (binding IC50 = 2-4 nM), their cAMP functional activity suggested that a n-butyl, n-propyl, and methyl group (15b, 13e, and 16d, cAMP IC50 = 6.6, 5.0, and 8.0 nM, respectively), is more preferred than an isopropyl or ethyl group (16b and 16c, cAMP IC50 = 69 and 21 nM, respectively). As the pyrazole-based GRAs were often found not to be highly selective against human glucose dependent insulinotropic peptide receptor (hGIPR), one of other class B GPCR’s,13 selected compounds were screened in hGIPR cAMP assay to verify selectivity of this series of GRAs. A compound’s hGIPR cAMP potency was determined in CHO cells expressing human GIPR, measuring cAMP production. As shown in Table 6, these indazole based GRAs were found to be generally very selective against hGIPR, with selectivity ranging from 79- to 250-fold. Based on their favorable in vitro profiles, several compounds were chosen for in vivo pharmacokinetic characterization in rat. The pharmacokinetic parameters are summarized in Table 7. All compounds studied exhibited relatively low clearance, and moderate to long half-lives (1.5–9.7 h). Oral dosing of these compounds

Please cite this article in press as: Lin, S.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.08.015

S. Lin et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx Vehicle Glucagon Glucagon + 16d @ 10 mpk Glucagon + 16d @ 3 mpk Glucagon + 16d @ 1 mpk

90

Blood Glucose Delta From 0 time

80 70 60 50

5

GRA 16d was found to be orally active in blunting glucagon induced glucose excursion in an acute glucagon challenge model in hGCGR mice at doses as low as 1 mpk, and significantly lowered acute glucose levels in hGCGR ob/ob mice at 3 mpk dose. Further evaluations of these exciting new GRAs are underway and will be reported in due course.

40

Acknowledgments

30 20

We thank Drs. Yusheng Xiong, James Mu, Ester Carballo-Jane, and Ravi Nargund for their assistance and helpful discussion during the preparation of the manuscript.

10 0

0

12

24

36

48

Time (min)

Figure 2. Effect of compound 16d in acute glucagon-induced glucose levels in hGCGR mice. Data represents mean ± SEM (n = 8 10).

Δ Glucose [m g/dL]

50 veh 16d @ 3.0 mpk

0

Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.bmcl.2015.08. 015. These data include MOL files and InChiKeys of the most important compounds described in this article. References and notes

-50

** -100

***

***

-150 -200 0

1

2

3 Tim e [h]

4

5

6

cpd dosed Figure 3. Acute effect of compound 16d in ambient glucose in the hGCGR ob/ob mice (N = 10–12 mice in each group). Compound was dosed orally at 0 h and food was removed at 0 h. **p <0.01 and ***p <0.001 relative to control group at each time point.

in rat also found them to possess good exposure and low/moderate to good bioavailability (9–44%), consistent with their low clearance. Encouraged by these desired in vitro profile and pharmacokinetic properties, we studied compound 16d in an acute glucagon challenge model in glucagon receptor humanized (hGCGR) mice.17 As shown in Fig. 2, compound 16d was found to be active in blunting glucagon-induced glucose excursion when dosed orally at 1, 3, and 10 mg/kg (mpk) one hour prior to a glucagon challenge (IP, 15 lg/kg). The drug levels at 60 min were found to be 0.18, 0.50, and 2.54 lM for 1, 3, and 10 mpk dose groups, respectively. Compound 16d was further evaluated for its efficacy in glucose lowering in a hGCGR ob/ob mice13 in an acute model as shown in Fig. 3. Compared to vehicle control group, compound 16d at 3 mpk oral doses lowered blood glucose level significantly at all the time points checked during the 6 h post treatment monitoring, highlighting the excellent efficacy of compound 16d in this ob/ob mice model. The drug levels at 1, 3, and 6 h were found to be 0.40, 0.28, and 0.12 lM, respectively. In summary, a novel series of indazole- and indole-based compounds was designed based on an earlier pyrazole-based GRA lead MK-0893 and synthesized. Structure–activity relationship studies of this series of compounds focused on the C3 and C6 positions of the indazole core, as well as the benzylic position on the N-1 of indazole. Multiple potent GRAs were identified with excellent in vitro profiles and good pharmacokinetics in rat. Among them,

1. Global Status Report on Noncommunicable Diseases 2014; World Health Organization: Geneva, 2014. 2. Mittermayer, F.; Caveney, E.; De Oliveira, C.; Gourgiotis, L.; Puri, M.; Li-Jung Tai, L.-J.; Turner, J. R. Curr. Diab. Rev. 2015, 11, 17. 3. (a) Zhang, B. B.; Moller, D. B. Curr. Opin. Chem. Biol. 2000, 4, 461; (b) Jiang, G.; Zhang, B. B. Am. J. Physiol. 2003, 284, E671; (c) Sloop, K. W.; Michael, D. D. Drugs Future 2004, 29, 835. 4. DeFronzo, R. A. Diabetes 1988, 37, 667. 5. (a) Stumvoll, M.; Goldstein, B. J.; van Haeften, T. W. Lancet 2005, 365, 1333; (b) Staehr, P.; Hother-Nielsen, O.; Beck-Nielsen, H. Diab. Obes. Metab. 2000, 4, 215; (c) Shah, P.; Vella, A.; Basu, A.; Basu, R.; Schwenk, W. F.; Rizza, B. A. J. Clin. Endocrinol. Metab. 2000, 85, 4053. 6. (a) DeMong, D. E.; Miller, M. W.; Lachowicz, J. E. Ann. Rep. Med. Chem. 2008, 43, 119; For an earlier review see (b) Ling, A. Drugs Future 2002, 27, 987. 7. (a) Shen, D.; Lin, S.; Parmee, E. R. Expert Opin. Ther. Patent 2011, 21, 1211; (b) Kurukulasuriya, R.; Link, J. T. Expert Opin. Ther Patent 2005, 15, 1739. 8. Petersen, K. F.; Sullivan, J. T. Diabetologia 2001, 44, 2018. 9. (a) Ling, A.; Plewe, M. B.; Truesdale, L. K.; Lau, J.; Madsen, P.; Sams, C. J.; Dehrens, C.; Vagner, J.; Christensen, I. T.; Lundt, B. F.; Sidelmann, U. G.; Thøgersen, H. WO2000/069810; (b) Lau, J.; Behrens, C.; Sidelmann, U. G.; Knudsen, L. B.; Lundt, B.; Sams, C.; Ynddal, L.; Brand, C. L.; Pridal, L.; Ling, A.; Kiel, D.; Plewe, M.; Shi, S.; Madsen, P. J. Med. Chem. 2007, 50, 113; (c) Kodra, J. T.; Jørgensen, A. S.; Andersen, B.; Behrens, C.; Brand, C. L.; Christensen, I. T.; Guldbrandt, M.; Jeppesen, C. B.; Knudsen, L. B.; Madsen, P.; Nishimura, E.; Sams, C.; Sidelmann, U. G.; Pedersen, R. A.; Lynn, F. C.; Lau, J. J. Med. Chem. 2008, 51, 5387. 10. Madsen, P.; Kodra, J. T.; Behrens, C.; Nishimura, E.; Jeppesen, C. B.; Pridal, L.; Andersen, B.; Knudsen, L. B.; Valcace-Aspegren, C.; Guldbrandt, M.; Christensen, I. T.; Jorgenson, A. S.; Ynddal, L.; brand, C. L.; gagger, M. A.; Lau, J. J. Med. Chem. 2009, 52, 2989. 11. Rivera, N.; Everett-Grueter, C. A.; Edgerton, C. S.; Rodewald, T.; Neal, D. W.; Nishimura, E.; Larsen, M. O.; Jacobsen, L. O.; Kristensen, K.; Brand, C. L.; Cherrington, A. D. J. Pharm. Exp. Ther. 2007, 321, 743. 12. Filipski, K. J.; Bian, J.; Ebner, D. C.; Lee, E. C. Y.; Li, J.-C.; Sammons, M. F.; Wright, S. W.; Stevens, B. D.; Didiuk, M. T.; Tu, M.; Perreault, C.; Brown, J.; Atkinson, K.; Tan, B.; Salatto, C. T.; Litchfield, J.; Pfefferkorn, J. A.; Guzman-Perez, A. Bioorg. Med. Chem. Lett. 2012, 22, 415. 13. Xiong, Y.; Guo, J.; Candelore, M. R.; Liang, R.; Miller, C.; Dallas-Yang, Q.; Jiang, G.; McCann, P. E.; Qureshi, S. A.; Tong, X.; Xu, S. S.; Shang, J.; Vincent, S. H.; Tota, L. M.; Wright, M. J.; Yang, X.; Zhang, B. B.; Tata, J. R.; Parmee, E. R. J. Med. Chem. 2012, 55, 6137. 14. (a) Kelly, R. P.; Garhyan, P.; Abu-Raddad, E. J.; Fu, H.; Lim, G. N.; Prince, M. J.; Pinare, J. A.; Loh, M. T.; Deeg, M. A. Diabetes 2011, 60, A84; (b) Conner, S. E.; Zhu, G. US Patent 7816557, 2010; (c) Conner, S. E.; Zhu, G.; Li, J. US Patent 0124648, 2011; (d) Conner, S. E.; Zhu, G.; Li, J. US Patent 0324140, 2010. 15. Enantiomers were separated by ChiralPack Column IA or IB, using heptane/IPA (8–15%) as the eluents. 16. Enantiomers of racemic 21 were separated by chiralPak OD-H column, using 7% IPA in heptane as eluent. 17. Qureshi, S. A.; Candelore, M. R.; Xie, D.; Yang, X.; Tota, L. M.; Ding, V. D.; Li, Z.; Bansal, A.; Miller, C.; Cohen, S. M.; Jiang, G.; Brady, E.; Saperstein, R.; Duffy, J. L.; Tata, J. R.; Chapman, K. T.; Moller, E. D.; Zhang, B. B. Diabetes 2004, 53, 3267.

Please cite this article in press as: Lin, S.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.08.015