N-substituted sultam carboxylic acids as novel glycogen synthase activators

Article abstract of DOI:10.1039/c3md00053b

Decreased glycogen synthesis and turnover is a common defect in type 2 diabetic patients. Activating glycogen synthase, the enzyme that catalyses the transfer of glucose from UDP-glucose to a glycogen polymer chain, could be a potential therapeutic target for the treatment of diabetes. We discovered a series of N-substituted sultam carboxylic acids as potent glycogen synthase activators. Treatment of human skeletal muscle cells with these compounds resulted in an increase in glycogen synthesis. Compound 4 displayed good oral bioavailability and therefore may be a useful tool molecule to study GS as a potential anti-diabetic target. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3md00053b

MedChemComm
View Article Online
View Journal
CONCISE ARTICLE
N-substituted sultam carboxylic acids as novel glycogen
synthase activators
Cite this: DOI: 10.1039/c3md00053b
Yimin Qian, David R. Bolin,a Karin Conde-Knape,b Paul Gillespie,a Stuart Hayden,a
a
*
c
d
b
b
b
´
Kuo-Sen Huang, Mei Liu, Andree R. Olivier, Yonglin Ren, Joseph Sergi,
Qing Xiang,c Lin Yi,a Weiya Yuna and Xiaolei Zhangc
Decreased glycogen synthesis and turnover is a common defect in type 2 diabetic patients. Activating
glycogen synthase, the enzyme that catalyses the transfer of glucose from UDP-glucose to a glycogen
polymer chain, could be a potential therapeutic target for the treatment of diabetes. We discovered a
series of N-substituted sultam carboxylic acids as potent glycogen synthase activators. Treatment of
human skeletal muscle cells with these compounds resulted in an increase in glycogen synthesis.
Compound 4 displayed good oral bioavailability and therefore may be a useful tool molecule to study
GS as a potential anti-diabetic target.
Received 15th February 2013
Accepted 12th March 2013
DOI: 10.1039/c3md00053b
www.rsc.org/medchemcomm
Type 2 diabetes (T2D) is a common and serious disorder 13C-NMR spectroscopy in hyperglycemic–hyperinsulinemic
affecting a large population around the globe. The disease clamp studies with 13C-glucose tracer, a greater than 50%
pathology is characterized by an abnormal increase in hepatic decrease in muscle glycogen synthesis in T2D patients
glucose production and a decrease in glucose utilization in compared to normal subjects was found.8 Therefore, activating
peripheral tissues. Although several classes of anti-diabetic GS by GS activators may provide a potential therapeutic target
agents are available, there is still a strong need to identify effi- for T2D treatment.
cacious treatments without serious side effects.1
The success of our discovery of glucokinase (GK) activators
The two major pathways of glucose utilization in the liver and as potential anti-diabetic agents prompted us to investigate
skeletal muscles are glycolysis, or oxidation of glucose to pyru- other targets in the glucose homeostasis pathway.9,10 Since GS is
vate, and glycogenesis, or storage of glucose as glycogen. One of the key regulatory enzyme in glucose storage in peripheral
the key enzymes in glycogen synthesis is glycogen synthase (GS). tissues, we were interested in identifying small molecule GS
GS catalyses the transfer of glucose from UDP-glucose to the activators. Screening of our compound library provided us a
growing glycogen chain through a 1,4-a linkage, and this step is single hit 1 shown in Fig. 1 (GYS1 EC50 ¼ 15.1 mM). Our initial
the rate-limiting step under most circumstances.2,3
hit expansion was focused on the le-hand bi-phenyl and the
Two forms of GS exist in mammals, encoded by the genes central furan ring, and a signicant increase in GS activation
GYS1 and GYS2. The gene products share 69% homology, where potency was achieved as exemplied by 2 (GYS1 EC50 ¼ 0.14
GYS2 encodes the liver specic form and GYS1 encodes the mM).11 Subsequently, the modication of the proline moiety on
form expressed in all other tissues including muscle.4,5 Both the right-hand side led to the discovery of N-alkylglycine
genetic and clinical data suggest the involvement of GS in T2D. analogue 3 (GYS1 EC50 ¼ 0.1 mM).12 To further explore the
It has been reported that a polymorphism of the muscle gene chemical diversity of GS activators from this biphenyl ether
GYS1 is more prevalent in T2D patients than in non-diabetic chemical class, we took a bioisostere approach to replace the
controls.6 Furthermore, basal and insulin-stimulated GS activ- proline or glycine fragments in 2 and 3 and identied N-alkyl-
ities in muscle cells from diabetic subjects are signicantly sultam carboxylic acid 4, a novel GS activator with oral
decreased relative to cells from non-diabetic subjects.7 Using bioavailability. Herein we report the discovery and biological
evaluation of these sultam-derived GS activators.
To further explore the structure–activity relationship of 3, we
aDiscovery Chemistry, Hoffmann-La Roche, 340 Kingsland Street, Nutley, NJ 07110,
USA. E-mail: yiminqian81@gmail.com; Tel: +1 973-641-9072
bMetabolic and Vascular Diseases, Hoffmann-La Roche, 340 Kingsland Street, Nutley,
NJ 07110, USA
replaced the amide carbonyl group in 3 with a bioisosteric
methylene group and prepared the corresponding amide,
carbamate and sulfonamide analogues (5–12, in Table 1). In
order to reduce the number of rotatable bonds in 7 and 12, we
rigidied the molecule and prepared the corresponding lactam
13 and sultam 14. Compounds 5–14 were synthesized according
to Schemes 1–3.19
cDiscovery Technologies, Hoffmann-La Roche, 340 Kingsland Street, Nutley, NJ 07110,
USA
dPharmacokinetics and Drug Metabolism, Hoffmann-La Roche, 340 Kingsland Street,
Nutley, NJ 07110, USA
This journal is ª The Royal Society of Chemistry 2013
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
Table 1 Activation potency of GS activators 5–14 in GS enzymatic and cellular assaysa
GS enzyme
GS enzyme
GS enzyme fold
GS cellular
GS cellular fold
Compound
R
EC50 (mM)
SC200 (mM)
increase @ 75 mM
SC250 (mM)
increase @ 75 mM
2
3
5
6
0.14 ꢀ 0.04
0.10 ꢀ 0.01
4.9
0.02 ꢀ 0.01
0.02 ꢀ 0.01
0.38
5.4 ꢀ 1.6
6.8 ꢀ 1.5
5.5
0.20 ꢀ 0.10
0.16 ꢀ 0.10
6.1
32.5 ꢀ 10.5
33.2 ꢀ 5.0
3.9
H
1.93
0.50
7.6
3.7
6.5
7
0.37
0.10
5.4
10.3
7.9
8
0.72
0.18
6.6
2.4
8.6
9
1.34
0.21
6.4
5.5
9.1
10
11
12
1.82
0.42
3.5
0.71
14.9
0.47
0.03
7.9
0.37 ꢀ 0.05
0.28 ꢀ 0.07
15.6 ꢀ 0.2
28.2 ꢀ 6.0
0.31 ꢀ 0.10
0.04 ꢀ 0.01
5.7 ꢀ 0.4
13
0.12
0.02
6.0
0.70
0.44
25.7
30.7
14
0.13 ꢀ 0.05
0.02 ꢀ 0.01
5.2 ꢀ 1.2
a
Numbers with ꢀSEM are from at least three independent biological assays and numbers without ꢀSEM are the average from dose-dependent
enzymatic or cellular assays in triplicate with the overall CV% less than 30%.
Scheme 2 Reagents and conditions: (a) K2CO3, acetone, reflux, 15 h, 51%; (b) L-
glutamic acid diethyl ester hydrochloride, MeOH/THF/HOAc, NaBH(OAc)3, 69%;
(c) LiOH (aq.)/THF, 97%.
Scheme 1 Reagents and conditions: (a) K2CO3, PdCl2(dppf) cat., DMF/H2O, 60
ꢁC, 15 h, 92%; (b) LiAlH4, ether/THF, rt, 2 h, 98%; (c) BrCH2COOEt, K2CO3, CH3CN,
reflux, 15 h, 69%; (d) LiOH (aq.)/THF, 63%; (e) EtI, TEA, CH2Cl2, then (d), 54%; (f)
acyl chloride, TEA, CH2Cl2, then (d), 95%; (g) alkyl chloroformate, TEA, CH2Cl2,
Scheme 3 Reagents and conditions: (a) NaBH4, 100%; (b) PBr3, CH2Cl2, 0 ꢁC to rt,
then (d), 90%; (h) EtSO2Cl, TEA, CH2Cl2, then (d), 45%.
42%; (c) 1,1-dioxo-isothiazolidine-3-carboxylic acid methyl ester, K2CO3, DMF, rt,
3 h, 73%; (d) LiOH (aq.)/THF, 98%.
Compounds 5–14 were studied in the biochemical assay for
their activities as activators of human muscle glycogen syn-
thase (GYS1). The recombinant human muscle GYS1 expressed activators) of GYS1 was observed and compounds were char-
and partially puried from sf9 cells was used as glycosyl- acterized with three parameters as listed in Table 1: fold-
transferase and the assay was carried out by coupling with increase in enzyme activity at the highest compound concen-
pyruvate kinase and lactate dehydrogenase.13 Dose-dependent tration tested (75 mM), EC50 (compound concentration
activation (in comparison with the same assay without required to achieve half of the maximum activation) and SC200
Med. Chem. Commun.
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Concise Article
MedChemComm
(compound concentration required to achieve a 2-fold enzyme
activity). We next tested these activators in activating GYS1 in
human muscle cells. Primary human myoblasts were fully
differentiated into polynucleated myotubes and cells were
treated with D-glucose with 14C-labelled tracer. Following
insulin stimulation (50 nM), the incorporation of 14C-glucose
into glycogen was measured.14 Dose-dependent cellular GS
activation was observed by comparing the same assay without
activator addition. Cellular GS activation was characterized by
two parameters as listed in Table 1: fold-increase in GS activity
at the highest compound concentration tested (75 mM) and
SC250 (compound concentration required to achieve a 2.5-fold
cellular GS activity). As shown in Table 1, compounds 2, 3, and
10–14 displayed signicantly higher levels of enzyme activation
Scheme 4 (a) (1) Isobutyl chloroformate, N-methylmorpholine, THF, ꢂ15 ꢁC, 15
min; (2) NaBH4, THF, MeOH, 81%; (b) thioacetic acid, PPh3, DIAD, 0 ꢁC to rt, 62%;
(c) HOAc, H2O, NaOAc, Cl2, 0 ꢁC, 100%; (d) CH2Cl2, TFA, 0 ꢁC to rt, 50%; (e) K2CO3,
DMF, rt, 3 h, 73%; (f) CH2Cl2, TFA, 0 ꢁC to rt, 80%.
in the cellular assay than in the corresponding biochemical and each enantiomer with dened stereochemistry was
assay. There is a clear synergistic effect between insulin and GS matched with components separated from chiral SFC.
activators (augmentation of the activator's capability of acti-
By using the chiral synthesis of the sultam and the chiral SFC
vating GYS1 in the presence of insulin stimulation). In separation of the racemate, we obtained 4 pairs of enantiomers (4
comparison with 2 and 3, compounds 5–9 displayed a signi- and 15–21). These 8 compounds were studied in both biochem-
cant loss of cellular GS activation potency (SC250 and fold- ical and cellular assays for their GS activation potency and results
increase), while the two carbamates (10 and 11) showed are listed in Table 2. When comparing each pair of enantiomers,
medium cellular GS activation (fold-increase). The sulfon- the (S)-isomer is signicantly more potent than the correspond-
amide 12, lactam 13 and sultam 14 showed comparable acti- ing (R)-isomer. The stereochemical effect on the GS enzyme
vation potency when compared with
2 and 3 in both activation potency is most signicant for uorine substitution
biochemical and cellular GS activation assays. When lactam 13 (4), while the methoxy substitution displayed only a 3-fold
was orally dosed to C57 mice, poor oral exposure was observed difference in the EC50 value in the enzyme assay. It is also
at both 2 h and 4 h time points, while the sulfonamide 12 and important to note that the more potent (S)-isomer demonstrated
sultam 14 displayed higher oral exposure (data not shown).
higher potency in the GS cellular assay as measured by both the
Since 14 is a mixture of enantiomers, we used super-critical SC250 value and the fold-increase of glycogen synthesis at the
uid chromatography (SFC) with a chiral column to separate the highest concentration. Interestingly, the sultam analogue with
racemate. As listed in Table 2, the (S)-enantiomer is signicantly methoxy substitution (16) demonstrated the highest cellular
more active than the corresponding (R)-enantiomer. To conrm potency, which is consistent with our observation for compounds
the absolute stereochemistry of two enantiomers from SFC in the proline and N-alkylglycine chemical series (2 and 3). Data
separation, pure enantiomers were synthesized as described in from Table 2 clearly suggest a good correlation between our GS
Scheme 4 using the literature route with our modications,15 biochemical assay and the human skeletal muscle cellular assay.
Table 2 Activation potency of GS activators 4, 15–21 in GS enzymatic and cellular assaysa
GS enzyme
GS enzyme
GS enzyme fold
GS cellular
GS cellular fold
Compd
R
(R) or (S)
EC50 (mM)
SC200 (mM)
increase @ 75 mM
SC250 (mM)
increase @ 75 mM
4
F
F
(S)
(R)
(S)
(R)
(S)
(R)
(S)
(R)
0.09 ꢀ 0.03
1.82 ꢀ 0.20
0.12 ꢀ 0.02
0.39 ꢀ 0.06
0.10
0.84
0.11
0.90 ꢀ 0.10
0.01 ꢀ 0.0
0.65 ꢀ 0.27
0.01 ꢀ 0.0
0.10 ꢀ 0.01
0.03
0.42
0.01
0.32 ꢀ 0.11
5.7 ꢀ 0.4
5.4 ꢀ 0.7
5.9 ꢀ 0.5
5.8 ꢀ 0.5
5.3
5.0
6.1
5.2 ꢀ 1.1
2.2 ꢀ 0.5
30.8 ꢀ 10.0
0.30
11.2
1.0
12.1
1.4
8.2 ꢀ 0.1
34.1 ꢀ 4.0
4.5 ꢀ 0.5
46.5
10.9
24.7
11.2
39.1
11.1 ꢀ 1.1
15
16
17
18
19
20
21
OCH3
OCH3
Cl
Cl
CH3
CH3
a
Numbers with ꢀSEM are from at least three independent biological assays and numbers without ꢀSEM are the average from dose-dependent
enzymatic or cellular assays in triplicate with the overall CV% less than 30%.
This journal is ª The Royal Society of Chemistry 2013
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
compounds demonstrated activation potency in both
biochemical and cellular assays. The stereochemistry of these
sultam analogues signicantly affected their GS activation
potency. Compound 4 further showed good oral bioavailability.
Apart from glucose-6-phosphate, an endogenous GS activator,
currently there are no small molecule GS activators reported in
the literature except for our biphenyl ether chemical class.16,17
The sultam carboxylic acids described here may provide tool
compounds to evaluate GS activation as a potential anti-diabetic
target.
Fig. 1 Structures of GS activators 1–4.
Acknowledgements
The authors gratefully acknowledge the following people who
contributed to our GS activator project: Mushtaq Ahmad, Kshitij
Thakkar, Bruce Banner, Matthew Hamilton, Lee McDermott
and Xilin Liu. We thank Ted Lambros for chiral SFC analyses
and separation, Chandra Pamulapati for pharmacokinetic
analyses, Jefferson Tilley and Cristina Rondinone for project
guidance, and Karen Lackey for management support of
manuscript preparation.
Table 3 Pharmacokinetic properties of 4 in rats (n ¼ 3)a
Route
iv
po
10
Dose (mg kgꢂ1
)
5
Cl (mL minꢂ1 kgꢂ1
)
4.26
0.98
19 855
Vdss (L kgꢂ1
)
AUC0-inf (ng h mLꢂ1
)
20 640
3327
52
Cmax (ng mLꢂ1
F (%)
)
a
Notes and references
The observed CV% for the iv route was less than 20%.
1 C. S. Mizuno, A. G. Chittiboyina, T. W. Kurtz,
H. A. Pershadsingh and M. A. Avery, Curr. Med. Chem.,
2008, 15, 61.
2 E. Cid, R. R. Gomis, R. A. Geremia, J. J. Guinovart and
J. C. Ferrer, J. Biol. Chem., 2000, 275, 33614.
3 P. J. Roach, A. A. DePaoli-Roach, T. D. Hurley and
V. S. Tagliabracci, Biochem. J., 2012, 441, 763.
4 G. Bai, Z. Zhang, R. Werner, F. Q. Nuttall, A. W. H. Tan and
E. Y. C. Lee, J. Biol. Chem., 1990, 265, 7843.
5 M. F. Browner, K. Nakano, A. G. Bang and R. J. Fletterick,
Proc. Natl. Acad. Sci. U. S. A., 1989, 86, 1443.
6 L. C. Groop, M. Kankuri, C. Schalin-Jantti, A. Ekstrand,
P. Nikula-Ijas, E. Widen, E. Kuismanen, J. Eriksson,
A. Franssila-Kallunki, C. Saloranta and S. Koskimies, N.
Engl. J. Med., 1993, 328, 10.
7 R. R. Henry, T. P. Ciaraldi, L. Abrams-Carter, S. Mudaliar,
K. S. Park and S. E. Nikoulina, J. Clin. Invest., 1996, 98, 1231.
8 G. I. Schulman, D. L. Rothman, T. Jue, P. Stein,
R. A. DeFronzo and R. G. Shulman, N. Engl. J. Med., 1990,
322, 223.
9 J. Grimsby, R. Sarabu, W. L. Corbett, N. E. Haynes,
F. T. Bizzarro, J. W. Coffey, K. R. Guertin, D. W. Hilliard,
R. F. Kester, P. E. Mahaney, L. Marcus, L. Qi, C. L. Spence,
J. Tenji, M. A. Magnuson, C. A. Chu, M. T. Dvorozniak,
F. M. Matschinsky and J. F. Grippo, Science, 2003, 301, 370.
10 Y. Qian, W. L. Corbett, S. J. Berthel, D. S. Choi,
M. T. Dvorozniak, W. Geng, P. Gillespie, K. R. Guertin,
N. E. Haynes, R. F. Kester, F. A. Mennona, D. Moore,
J. Racha, R. Radinov, R. Sarabu, N. R. Scott, J. Grimsby and
N. L. Mallalieu. ACS Med. Chem. Lett. DOI: 10.1021/
ml400027y.
Comparing the structural features of the sultam analogues
described in this report with proline derivatives (Fig. 1), it is
clear that they share the common fragment of the le hand
biphenyl ether and the carboxylic acid functional group. We
carried out extensive enzyme activation kinetic studies for both
HTS hit (compound 1) and highly potent GS activators in the
proline chemical series (chemical class of 2). Our data revealed
that they are non-competitive against the substrate UDP-glucose
and the physiological activator glucose-6-phosphate (G6P).
These compounds appear to bind GS at the allosteric site and
synergize with G6P. Although we did not further characterize
the kinetic proles of sultam analogues, it is likely that these
sultam carboxylic acids share the same allosteric binding mode
as the proline derivatives based on the SAR and the common
structural fragments.
To further evaluate the potential of our novel sultam-derived
GS activators as tool compounds, we investigated the pharma-
cokinetic properties of 4 in male Wistar rats.18 As listed in
Table 3, following intravenous dosing at 5 mg kgꢂ1, the mean
clearance was low (4.26 mL minꢂ1 kgꢂ1) and a moderate volume
of distribution was observed (967 mL kgꢂ1). When 4 was orally
dosed to rats at 10 mg kgꢂ1 (aqueous suspension containing 10%
PEG400 in phosphate buffer), a good oral bioavailability (52%)
was observed. To answer the question whether these chiral sul-
tam carboxylic acids can undergo in vivo racemization, we ana-
lysed plasma samples (using a chiral column) from mice orally
dosed with 4 and 15, and no in vivo racemization was observed.
Conclusions
In summary, a novel class of sultam-derived carboxylic acids 11 D. R. Bolin, M. Ahmad, B. Banner, J. Cai, K. Conde-Knape,
were identied as potent glycogen synthase activators. These
P. Gillespie, R. Goodnow, M. L. Gubler, M. M. Hamilton,
Med. Chem. Commun.
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Concise Article
MedChemComm
S. Hayden, K. S. Huang, X. Liu, J. P. Lou, L. McDermott,
A. R. Olivier, A. Perrotta, Y. Qian, C. M. Rondinone,
K. Rowan, J. A. Sergi, J. W. Tilley, K. Thakkar, L. Yi, W. Yun
and Q. Xiang, 239th ACS National Meeting, San Francisco,
CA, March 21–25, 2010, MEDI-529.
without the GS activator. Aer 5 h, monolayers were
washed twice with PBS and residual liquid in each well
was aspirated. Cells in 24 well plates were then frozen
overnight at ꢂ80 ꢁC. The cells were lysed and the 14C-
labeled glycogen was extracted as described below.
Radioisotope-labelled glycogen extraction: Frozen myotubes
were lysed with addition of 100 mL of 1 M NaOH and
placed on a shaker for 15 min at room temp. The cell
lysates were rst transferred individually to a 96-well plate,
this allowed rapid transfer (using a multichannel pipette)
12 D. R. Bolin, Y. Qian, K. C. Thakkar, L. Yi and W. Yun, US Pat.,
8039495, 2011.
13 The enzymatic activity of GS was measured in 384-well
polystyrene plates (BD Biosciences) using
a coupled
enzyme assay in a total volume of 32 mL per well. An 8 mL
aliquot of activator solution (diluted compounds in 30 mM
glycylglycine, pH 7.3, 40 mM KCl, 20 mM MgCl2, 9.2%
DMSO, with or without 20 mM glucose-6-phosphate) was
added to 12 mL aliquot of substrate solution containing
glycogen (4.32 mg mLꢂ1), 2.67 mM UDP-glucose, 21.6 mM
phospho(enol)pyruvate and 2.7 mM NADH in 30 mM
glycylglycine, pH 7.3 buffer. The reaction was then started
by adding 12 mL of enzyme solution containing glycogen
synthase (16.88 mg mLꢂ1), pyruvate kinase (0.27 mg mLꢂ1),
and lactate dehydrogenase (0.27 mg mLꢂ1) in 50 mM Tris–
HCl, pH 8.0, 27 mM DTT and bovine serum albumin (BSA,
to 96-well glass-bre lter plates (Millipore Cat.
#
MAFBNOB50 – Multiscreen FB) pre-loaded with cold 100%
ethanol (100 mL per well). The plates were incubated at
ꢁ
4 C for 2 h to ensure the glycogen was precipitated. Aer
precipitation the free radio ligand was separated from
glycogen on a vacuum manifold. The plate was dried and
snapped into the lter plate adapter (Packard # PPN
6005178) of Microscint-20 and cocktail (50 mL) was added
to each well. Aer covering with a sealing lm, the plate
was loaded and counted directly in the TopCount and the
incorporated glucose into glycogen was calculated.
0.2 mg mLꢂ1), mixed and incubated at room temperature. 15 R. J. Cherney, R. Mo, D. T. Meyer, K. D. Hardman, R. Q. Liu,
The conversion rate of NADH to NAD was measured every
3 min over a period of 15 minutes at abs. 340 nm on an
Envision reader (Perkin Elmer). The enzyme activity (with
M. B. Covington, M. Qian, Z. R. Wasserman, D. D. Christ,
J. M. Trzaskos, R. C. Newton and C. P. Decicco, J. Med.
Chem., 2004, 47, 2981.
or without compound) was calculated by the reaction rate 16 D. R. Bolin, S. Hayden, Y. Qian and K. C. Thakkar, US Patent
per minute. Application, 2011/0136792, 2011.
14 Human Skeletal Muscle Cell Culture and Differentiation 17 (a) Y. Uto, Expert Opin. Ther. Pat., 2012, 22, 89; (b)
Conditions: Normal human skeletal muscle cells were
A. L. Skaltsounis and N. Gaboriaud-Kolar, Expert Opin.
Ther. Pat., 2011, 21, 1925.
purchased from Lonza. Myoblasts were kept in culture at a
concentration of 4 ꢃ 105 mLꢂ1 in Lonza SkBM medium 18 Experiments in the pharmacokinetic studies were performed
(without the addition of insulin) and 2.0% FBS, 1%
glutamine in two T225 Flasks (65 mL per Flask). Aer 3
days they reached 70–75% conuency and were sub-
in compliance with the relevant laws and institutional
guidelines, and approvals were obtained prior to the
experiments.
cultured using 0.025% Trypsin/EDTA 0.01% (Lonza) and 19 Compounds 4–21 were characterized by 1H-NMR and
neutralization solution (Lonza). Myoblasts were plated in
24-well plates (pre-coated the same day with Matrigel
diluted 1 : 60 in high glucose DMEM, 25 mM Hepes) at a
concentration of 1.2 ꢃ 103 cells per well in Lonza medium
(without the addition of insulin) and 2% FBS, 1%
glutamine. Aer 4 days cells reached 70–75% conuency
and were differentiated by the addition of differentiation
medium (DMEM (high glucose), 25 mM Hepes and 2%
FBS). Myoblasts reached full differentiation into myotubes
6 days later. Biological assays were run between 6–8 days
aer differentiation. Myotubes responded to insulin and
were used up to and including passage 5.
analyzed by LC/MS with observed purity greater than 95%.
All chiral compounds were analyzed by super-critical uid
chromatography on a chiral column (diacel AD column,
35% MeOH in CO2) in comparison with the corresponding
racemate and the enantiomeric purity was greater than
95%. The spectral data of the selected compounds are
listed below:
1
Compound 11: H NMR (300 MHz, DMSO-d6) d ppm 1.11–
1.21 (m, 6H), 3.75 (s, 3H), 3.84 (d, J ¼ 18.1 Hz, 2H), 4.45
(d, J ¼ 4.5 Hz, 2H), 4.71–4.85 (m, 1H), 5.12 (s, 2H), 7.03 (d,
J ¼ 8.1 Hz, 2H), 7.18–7.28 (m, 2H), 7.30–7.46 (m, 6H); LC/
MS calcd for C27H27NF2O6 (m/z) 499, obsd 500.1 (M + H, ES+).
Compound 12: 1H NMR (300 MHz, DMSO-d6) d ppm 1.25 (t, J
¼ 7.2 Hz, 3H), 3.20 (q, J ¼ 7.2 Hz, 2H), 3.75 (s, 3H), 3.81 (s,
2H), 4.48 (s, 2H), 5.15 (s, 2H), 7.04 (d, J ¼ 8.7 Hz, 2H),
7.17–7.48 (m, 8H), 12.92 (br, s, 1H); LC/MS calcd for
Glycogen Synthesis in human skeletal muscle cells:
Differentiated human myotubes were fed with fresh
differentiation medium (as described above) the night
before the assay. Prior to compound or hormone
stimulation the medium was changed to DMEM (low
glucose), 5 mg mLꢂ1 low endotoxin BSA for 2 hours at
C
25H25NF2O6S (m/z) 505, obsd 504.1 (M ꢂ H, ESꢂ).
1
Compound 13: H NMR (300 MHz, DMSO-d6) d ppm 1.87–
2.04 (m, 1H), 2.15–2.43 (m, 3H), 3.75 (s, 3H), 3.82–3.97 (m,
2H), 4.91 (d, J ¼ 15.4 Hz, 1H), 5.13 (s, 2H), 7.03 (d, J ¼ 8.7
Hz, 2H), 7.11–7.46 (m, 8H), 13.05 (br, s, 1H); LC/MS calcd
for C26H23NF2O5 (m/z) 467, obsd 468.0 (M + H, ES+).
ꢁ
37 C. For stimulation the medium was then replaced with
DMEM (low glucose), 25 mM Hepes medium containing
0.8 mCi per well of D-glucose/D-[14C] with 50 nM insulin
(Gibco Human Insulin Catalogue # 12585-014) with or
This journal is ª The Royal Society of Chemistry 2013
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
1
Compound 14: H NMR (300 MHz, CDCl3) d ppm 2.40–2.69
(m, 2H), 3.02–3.16 (m, 1H), 3.32 (dt, J ¼ 12.4, 7.8 Hz, 1H),
3.78–3.87 (m, 1H), 3.77 (s, 3H), 4.28 (d, J ¼ 15.1 Hz, 1H),
4.64 (d, J ¼ 15.1 Hz, 1H), 5.12 (s, 2H), 6.78 (dd, J ¼ 12.1,
6.6 Hz, 1H), 6.99 (d, J ¼ 8.4 Hz, 2H), 7.04–7.17 (m, 1H),
7.30–7.48 (m, 6H); LC/MS calcd for C25H23NF2O6S (m/z)
503, obsd 502.0 (M ꢂ H, ESꢂ).
3.78–3.87 (m, 1H), 3.77 (s, 3H), 4.28 (d, J ¼ 15.1 Hz, 1H),
4.64 (d, J ¼ 15.1 Hz,1H), 5.12 (s, 2H), 6.78 (dd, J ¼ 12.1,
6.6 Hz, 1H), 6.99 (d, J ¼ 8.4 Hz, 2H), 7.04–7.17 (m, 1H),
7.30–7.48 (m, 6H); LC/MS calcd for C25H23NF2O6S (m/z)
503, obsd 502.0 (M ꢂ H, ESꢂ); micro analysis calcd C
59.64%, H 4.60%, N 2.78%, F 7.55%, S 6.37%; obsd C
58.81%, H 4.67%, N 2.70%, F 7.58%, S 6.25%.
1
1
Compound 4: H NMR (400 MHz, CDCl3) d ppm 2.45–2.55
Compound 18: H NMR (300 MHz, CDCl3) d ppm 2.45–2.70
(m, 1H), 2.55–2.67 (m, 1H), 3.05–3.16 (m, 1H), 3.33 (dt, J ¼
12.5, 7.9 Hz, 1H), 3.84 (dd, J ¼ 8.3, 4.5 Hz, 1H), 4.29 (d, J ¼
15.1 Hz, 1H), 4.65 (d, J ¼ 15.1 Hz, 1H), 5.12 (s, 2H), 6.96–
7.06 (m, 3H), 7.17–7.25 (m, 1H), 7.32–7.49 (m, 6H); LC/MS
calcd for C24H20NF3O5S (m/z) 491, obsd 490.0 (M ꢂ H,
ESꢂ); micro analysis calcd C 58.65%, H 4.10%, N 2.85%, F
11.60%, S 6.52%; obsd C 58.87%, H 3.98%, N 2.78%, F
11.43%, S 6.55%; [a]D ¼ ꢂ26.2 (4 mg mLꢂ1, in EtOAc).
(m, 2H), 3.06–3.20 (m, 1H), 3.34 (dt, J ¼ 12.4, 7.9 Hz, 1H),
3.85 (dd, J ¼ 8.1, 4.5 Hz, 1H), 4.31 (d, J ¼ 15.1 Hz, 1H), 4.62
(d, J ¼ 15.1 Hz, 1H), 5.11 (s, 2H), 7.03 (d, J ¼ 8.7 Hz, 2H),
7.16 (dd, J ¼ 10.4, 8.6 Hz, 1H), 7.29–7.39 (m, 4H), 7.39–7.49
(m, 3H); LC/MS calcd for C24H20ClNF2O5S (m/z) 507, obsd
506.0 (M ꢂ H, ESꢂ).
1
Compound 20: H NMR (400 MHz, CDCl3) d ppm 2.20 (s,
3H), 2.47–2.58 (m, 1H), 2.58–2.70 (m, 1H), 3.09–3.19 (m,
1H), 3.35 (dt, J ¼ 12.3, 7.9 Hz, 1H), 3.87 (dd, J ¼ 8.4,
4.5 Hz, 1H), 4.30 (d, J ¼ 14.9 Hz, 1H), 4.65 (d, J ¼ 14.9
Hz, 1H), 5.12 (s, 2H), 6.96–7.09 (m, 4H), 7.19 (d, J ¼ 8.5
Hz, 2H), 7.33–7.50 (m, 4H); LC/MS calcd for
1
Compound 15: H NMR (400 MHz, CDCl3) d ppm 2.44–2.55
(m, 1H), 2.55–2.68 (m, 1H), 3.05–3.16 (m, 1H), 3.28–3.39
(m, 1H), 3.84 (dd, J ¼ 7.9, 4.3 Hz, 1H), 4.29 (d, J ¼ 15.1 Hz,
1H), 4.65 (d, J ¼ 14.9 Hz, 1H), 5.12 (s, 2H), 6.95–7.08 (m,
3H), 7.16–7.26 (m, 1H), 7.31–7.50 (m, 6H); LC/MS calcd for
C
25H23NF2O5S (m/z) 487, obsd 486.0 (M ꢂ H, ESꢂ);
C
24H20NF3O5S (m/z) 491, obsd 490.0 (M ꢂ H, ESꢂ).
micro analysis calcd C 61.59%, H 4.76%, N 2.87%, F
7.79%, S 6.58%; obsd C 61.33%, H 4.78%, N 2.74%, F
8.03%, S 6.52%.
1
Compound 16: H NMR (300 MHz, CDCl3) d ppm 2.40–2.69
(m, 2H), 3.02–3.16 (m, 1H), 3.32 (dt, J ¼ 12.4, 7.8 Hz, 1H),
Med. Chem. Commun.
This journal is ª The Royal Society of Chemistry 2013