Discovery of 5-Hydroxyalkyl-4-phenylpyridines as a new class of glucagon receptor antagonists

Article abstract of DOI:10.1016/S0960-894X(01)00766-1

5-Hydroxyalkyl-4-phenylpyridines have been identified as a novel class of glucagon antagonists with potential utility for the treatment of diabetes. A lead structure with moderate activity was discovered through a high throughput screening assay. Structure-activity relationships led to the discovery of a potent antagonist, IC₅₀=0.11 μM, more than 60-fold improvement over the lead structure.

Full text of DOI:10.1016/S0960-894X(01)00766-1

Bioorganic & Medicinal Chemistry Letters 12 (2002) 461–464
Discovery of 5-Hydroxyalkyl-4-phenylpyridines as a New Class of
Glucagon Receptor Antagonists
Gaetan H. Ladouceur,a,* James H. Cook,a Elizabeth M. Doherty,a William R. Schoen,a
Margit L. MacDougallb and James N. Livingstonb
aDepartment of Chemistry Research, Bayer Research Center, 400 Morgan Lane, West Haven, CT 06516, USA
bDepartment of Diabetes and Obesity Research, Bayer Research Center, 400 Morgan Lane, West Haven, CT 06516, USA
Received 27 September 2001; accepted 16 November 2001
Abstract—5-Hydroxyalkyl-4-phenylpyridines have been identified as a novel class of glucagon antagonists with potential utility for
the treatment of diabetes. A lead structure with moderate activity was discovered through a high throughput screening assay.
Structure–activity relationships led to the discovery of a potent antagonist, IC50=0.11 mM, more than 60-fold improvement over
the lead structure. # 2002 Elsevier Science Ltd. All rights reserved.
Glucagon is a 29 amino acid peptidic hormone secreted
by a-cells in pancreatic islets. The glucagon receptor is a
member of the G-protein coupled receptor family which
signals through the adenylate cyclase pathway by
increasing cAMP levels in hepatocytes.1,2 Elevated
intracellular cAMP levels stimulate liver glycogenolysis
and gluconeogenesis, resulting in an increase of hepatic
glucose production.3 A sustained high concentration of
glucagon in the plasma is believed to be an important
cause of excessive hepatic glucose production in diabetic
subjects.4,5 This alteration contributes to the elevation
in blood glucose, the underlying cause of morbidity and
mortality in diabetes.
Our research efforts were thereby aimed at the identifi-
cation of a lead structure capable of competitively
altering glucagon binding to the human glucagon
receptor (hGR). A high throughput screening assay
resulted in the discovery of compound 1 (Fig. 1), which
exhibited modest activities (binding affinity, IC50=7
mM and inhibition of cAMP production, IC50=2 mM).
In this article, we describe the synthesis and the SAR of
this new class of glucagon receptor antagonists. Com-
pounds were tested in a binding assay using the human
glucagon receptor22 and selected compounds were tested
in a functional assay.23
Most analogues were synthesized by following the gen-
eral procedure described in Scheme 1, exemplified by the
synthesis of compound 6.11,24 Several other analogues
were subsequently made using compounds similar to 6
as a precursor. The synthesis of 6 started with the
Hantzsch pyridine synthesis.25 Condensation of
enamine 3, obtained from ethyl isobutyrylacetate 2,
with 4-fluorobenzaldehyde afforded pyridine 4 after
oxidative treatment with DDQ (14%, three steps).
Numerous modifications in the amino acid sequence of
glucagon led to the identification of several potent pep-
tide antagonists and is the subject of recent reviews.6,7
Due to the poor oral bioavailability associated with
peptides, their potential medical use has been limited.
However, it was shown that some of the most potent
peptide antagonists significantly decrease blood glucose
levels in diabetic animal models when administered
intravenously.8,9 Thus, this implied that a selective glu-
cagon receptor antagonist could be useful for the treat-
ment of diabetics, and consequently, prompted a strong
interest from several research groups.10ꢀ19 Reviews cov-
ering the discovery of non-peptidic small molecule
antagonists have been published.20,21
*Corresponding author. Tel.: +1-203-812-2055; fax: +1-203-812-
2650; e-mail: gaetan.ladouceur.b@bayer.com
Figure 1. Lead structure found in a high throughput screening assay.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00766-1

462
G. H. Ladouceur et al. / Bioorg. Med. Chem. Lett. 12 (2002) 461–464
Selective reduction of one of the ester groups of pyridine
4 with Red-Al, followed by treatment with PCC,
afforded aldehyde 5 (66%, two steps). A Wittig reaction
of aldehyde 5 with the ylide of propyltriphenylphos-
phonium bromide provided the corresponding alkene
ester, which was then successively reduced with LAH
and catalytic hydrogenation, to afford alcohol 6. Var-
ious ylides in the Wittig reaction could be used, leading
to wide variation of alkyl groups at position 5 of the
pyridine moiety (Tables 1 and 2).
SAR at position 3 of the pyridine moiety was then
explored using compound 6 as an intermediate to gen-
erate new analogues (Scheme 2). The lack of significant
activity found with the methyl ether 28, the acetate 27,
the aldehyde 25, and the fluoromethyl derivative 30
suggested that a hydrogen donor group is required at
that position. Attempts to replace the hydroxy group by
other hydrogen donors also failed. Indeed, the thiol 29
and the carboxylic acid 27 exhibited no significant
activities up to a concentration of 20 mM.
A series of analogues with variation at position 5 of the
pyridine moiety is listed in Table 1. We discovered that
straight alkyl chains provided potent inhibitors with
IC50 values in the sub-micromolar range. The optimum
potency was obtained with alkyl chains from 3 to 7
carbon atoms in length. Addition of steric bulk by
branching the alkyl group resulted in similar or
decreased activity (e.g., 1 and 14). Likewise, less lipo-
philic groups attached to an alkyl chain, such as
hydroxy (16) or ether (17), gave either no improvement
or substantially decreased the activity.
The hydroxymethyl group at position 3 of the pyridine
moiety was therefore the key pharmacophoric group for
activity and became the focus of further investigation
(Scheme 3). A breakthrough result was obtained with
the discovery of compound 31, which exhibited an IC50
of 0.3 mM, or a 2-fold increase of activity over its cor-
Table 1. SAR for position 5 of the pyridine moiety
We then investigated the SAR at positions 2 and 6 of
the pyridine moiety, as seen in Table 2. To avoid the
formation of regioisomers and to keep the synthesis
relatively simple, we limited our study to symmetrical
analogues. Smaller substituents were not favorable as
compounds 19 to 22, where R1 is methyl or ethyl,
exhibited only weak activity. The diisopropyl-pyridines
23 and 24 were significantly more active with IC50
values just above 1 mM.
Compd
R
hGR
IC50 (mM)
1
7
CH¼C(CH3)2
7.0
7.5
Me
8
9
Et
Pr
1.0
0.4
Steric bulk at this region of the molecule clearly
increased the activity. It may also be important for the
two isopropyl groups to shield the nitrogen atom of the
pyridine moiety from a lipophilic area of the receptor.
We were also interested in evaluating if the fluoro sub-
stituent found in the original lead structure (1) and our
initial SAR (7 to 13) had any effect on binding affinity.
This was addressed by comparison of 8 and 6, which are
the corresponding para-fluorophenyl analogues of com-
pounds 23 and 24. The compounds, 8 and 6, exhibited
approximately a two-fold increase in activity over their
des-fluoro counterparts.
10
6
Bu
Pent
Hex
Hept
Oct
0.75
0.7
0.35
0.7
1.8
1.0
6.0
11
12
13
14
15
16
17
18
CH2CH2CH(CH3)2
CH¼CH2
CH2OH
CH2OCH3
CO2CH3
>20
9.0
9.5
Table 2. SAR for positions 2,6 of the pyridine moiety
Compd
R1
R2
X
hGR
IC50 (mM)
19
20
21
22
23
24
8
Me
Me
Et
Et
Pent
Et
Pent
Et
Pent
Et
Pent
H
H
H
H
H
H
F
>20
>20
>20
20
Et
Scheme 1. Synthesis of phenylpyridine 6. Reagents and conditions:
(a) AcO–NH+4 , cyclohexane, –H2O, 92%; (b) ethyl isobutyrylacetate,
4-fluorobenzaldehyde, 20%; (c) DDQ, CH2Cl2, 78%; (d) Red-Al,
THF, 67%; (e) PCC, CH2Cl2, 99%; (f) nBuPPh3Br, KOtBu, THF,
90%; (g) LiAlH4, THF, 80%; (h) H2, Pd/C, EtOH, 98%.
iPr
iPr
iPr
iPr
1.8
1.4
1.0
0.7
6
F

G. H. Ladouceur et al. / Bioorg. Med. Chem. Lett. 12 (2002) 461–464
463
Table 3. Racemates versus most active enantiomers of secondary
alcohols
Compd
(racemate)
hGR
IC50 (mM)
R
Compd
(More potent
enantiomer)
hGR
IC50 (mM)
Scheme 2. Synthesis of phenylpyridine analogues. Reagents and con-
ditions: (a) PCC, CH2Cl2, 74%; (b) Jones reagent, acetone, 58%; (c)
AcCl, Et3N, CH2Cl2, 41%; (d) NaH, MeI, THF, 75%; (e) Lawesson’s
reagent, toluene, 26%; (f) DAST, CH2Cl2, 76%.
36
37
31
38
0.22
0.25
0.30
0.24
Pr
Bu
Pent
Hex
36a
–
31a
38a
0.11
–
0.15
0.12
This interesting result prompted us to evaluate a series
of hydroxyethyl analogues bearing different alkyl chains
at position 5 of the pyridine moiety (Table 3). In the
listed examples, a similar enantiomeric preference was
observed in which compound 36a exhibited the best
activity (IC50=0.11 mM).
Finally, the absolute configuration of the secondary
alcohol group was assigned by X-ray analysis of the
Mosher ester derivative.28 Thus, the secondary alcohol
of compound 36a was assigned the R configuration
(Fig. 2) and by analogy, 31a and 38a were assumed to
possess the same configuration. Compound 36a was
also evaluated for functional activity. In that assay, 36a
exhibited approximately the same activity (IC50=0.065
mM) as measured in the receptor binding assay.
Scheme 3. Synthesis of hydroxyalkylpyridine analogues. Reagents and
conditions: (a) MeLi, THF, 70%; (b) EtMgBr, THF, 70%; (c)
CH2¼CHMgBr, THF, 70%; (d) CH3OCH2PPh3Br, KOtBu, THF,
60%; (e) HCl, THF, 50%; (f) NaBH4, EtOH, 42%; (g) TMSCF3,
TBAF, THF, 70%.
In conclusion, SAR optimization of a lead structure
found in a high throughput screening led to the dis-
covery of new small molecule glucagon receptor
antagonists with an approximately 70-fold increase of
potency. Further optimization of this novel series will
be reported in due course.
Figure 2. Absolute configuration of 36a.
responding primary alcohol 6. This analogue was syn-
thesized by direct alkylation of aldehyde 25 with
methyllithium to produce a racemic mixture of second-
ary alcohols. Larger secondary alcohols were thereafter
explored (e.g., 32 and 33), but resulted in a loss of
activity, suggesting that the corresponding binding site
of the receptor is relatively compact in that area. A
transposition of the hydroxy group to the adjacent car-
bon atom (34) also resulted in a loss of activity. In an
attempt to modify the acidity of the hydroxy group, the
trifluoro isostere 35 was prepared.26 It, however, exhib-
ited a much weaker activity (IC50=13.0 mM) than its
methyl isostere.
Acknowledgements
We are grateful to A. Paiva and L. Huang for efficient
LC/MSanalyses, D. Eustice for contributions to the
initial screening work and A. Goehrt for X-ray analysis.
References and Notes
1. Dohlman, H. G.; Thorner, J.; Caron, M. G.; Lefkowitz,
R. J. Annu. Rev. Biochem. 1991, 60, 653.
2. Christophe, J. Biochim. Biophys. Acta 1995, 1241, 45.
3. Unger, R. H.; Orci, L. New Engl. J. Med. 1981, 1518.
4. Gerich, J. E.; Lorenzi, M.; Bier, D.; Schneider, V.; Tsalikian,
E.; Karam, J.; Forsham, P. H. New Engl. J. Med. 1975, 292,
985.
5. Unger, R. H. Metabolism 1978, 27, 1691.
6. Johnson, D. G.; Goebel, C. U.; Hruby, V. J.; Bergman,
M. D.; Trivedi, D. Science 1982, 215, 1115.
Since the racemic secondary alcohol 31 was the most
active analogue at that point, it was resolved by chiral
chromatography27 to evaluate the activity of each
enantiomer. We discovered that the faster eluting enan-
tiomer was significantly more active (31a, IC50=0.15
mM) than the slower enantiomer (31b, IC50=3.0 mM).

464
G. H. Ladouceur et al. / Bioorg. Med. Chem. Lett. 12 (2002) 461–464
7. Unson, C. G.; Andreau, D.; Gurzenda, E. M.; Merrifield,
R. B. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 4083.
8. Unson, C. G.; Gurzenda, E. M.; Merrifield, R. B. Peptides
1989, 10, 1171.
9. Salzig, T. M.; Unson, C.; Merrifield, B. Diabetes 1996, 45,
220.
10. Collins, J. L.; Dambek, P. J.; Goldstein, S. W.; Faraci,
W. S. Bioorg. Med. Chem. Lett. 1992, 2, 915.
11. Schoen, W. R.; Ladouceur, G. H.; Cook, J. H.; Lease,
T. G.; Wolanin, D. J.; Kramss, R. H.; Hertzog, D. L.;
Osterhout, M. H. USPatent 6,218,431, 2001. Chem. Abstr.
2001, 134, 311111.
12. De Laszlo, S. E.; Chang, L. L.; Kim, D.; Mantlo, N. B.
PCT Int. Appl. WO 97/16442, 1997. Chem. Abstr. 1997, 127,
50543.
13. Cascieri, M. A.; Koch, G. E.; Ber, E.; Sadowski, S. J.;
Louizides, D.; de Laszlo, S. E.; Hacker, C.; Hagmann, W. K.;
MacCoss, M.; Chicchi, G. G.; Vicario, P. P. J. Biol. Chem.
1999, 274, 8694.
14. Azizeh, B. Y.; van Tine, B. A.; Sturm, N. S.; Hutzler,
A. M.; David, C.; Trivedi, D.; Hruby, V. J. Bioorg. Med.
Chem. Lett. 1995, 5, 1849.
15. Azizeh, B. Y.; Shenderovich, M. D.; Trivedi, D.; Li, G.;
Sturm, N. S.; Hruby, V. J. J. Med. Chem. 1996, 39, 2449.
16. Ling, A.; Kuki, A.; Shi, S.; Plewe, M. B.; Feng, J.; Trues-
dale, L. K.; May, J.; Kiel, D.; Madsen, P.; Sams, C.; Lau, J.
PCT Int. Appl. WO 00/39088, 2000. Chem. Abstr. 2000, 133,
73951.
genization in a 4 ꢁC hypotonic lysis buffer that contained 10
mM Tris, pH 7.4, 2 mM EDTA, 5 mM MgCl2, and 1 mM
PMSF. Nuclei were removed by centrifugation at 800g for 15
min at 4 ꢁC. Membranes were collected by centrifugation of
the supernate at 15,000g for 15 min. The membranes were
washed, repelleted and suspended in 0.25 M sucrose, 10 mM
Tris, pH 7.4, 5 mM EDTA and stored at ꢀ80 ꢁC. The gluca-
gon binding assay was carried out in 96-well filtration plates
(1.2, glass fiber type C, Millipore) pretreated for 1 h with 0.3%
polyethyleneimine. Each well contained 5 mg of membrane
protein in 150 mL of binding buffer (20 mM Tris, pH 7.4, 1
mM EDTA, 1 mg/mL BSA, 1 mg/mL bacitracin). Five L of
compound were added in duplicate at concentrations of 20, 2,
0.2, and 0.02 M final, followed by the addition of 9 fmol 125I-
glucagon (New England Nuclear) contained in 50 mL binding
buffer. Control wells contained membranes, 0.2% DMSO,
radiolabeled glucagon without and with excess native gluca-
gon (1 mM) to establish nonspecific binding. The plates were
incubated for 60 min at 21 ꢁC, and filtered on the Millipore
vacuum apparatus. Plates were washed three times with cold
PBS/0.1% BSA, pH 7.4. The filters were punched into 12 ꢂ 75
tubes, and the membrane-associated radioactivity determined.
23. hGR-CHO cells were grown to confluency in 24-well
plates. The cells were washed twice with PBSand preincubated
for 10 min at 37 ꢁC in assay buffer as described (Feth, F.;
Rascher, W.; Michel, M. C. Arch. Pharmacol. 1991, 344, 1).
The cells were washed once and incubated for an additional 10
min in 250 L of assay buffer supplemented with 100 M iso-
butylmethylxanthine. Control DMSO (0.1%) or test com-
pound (10 nM to 10 mM final concentration) was added to the
wells followed by 50 L of glucagon to give a 10 pM final con-
centration. The incubation was continued for 15 min, then
stopped by removal of the assay buffer and the immediate
addition of 250 L ethanol (65%, v/v). After 5 min the super-
natant was transferred to a microfuge tube and the ethanol was
evaporated by Speed Vac. cAMP content was measured using
the Amersham SPA cAMP assay kit (RPA 556) as described by
the manufacturer. Results were reported as pmol per assay.
24. All data reported herein reflect purified and characterized
(1H NMR, MS) samples.
17. Madsen, P.; Knudsen, L. B.; Wiberg, F. C.; Carr, R. D. J.
Med. Chem. 1998, 41, 5150.
18. Gonzales, J.; Sams, C.; Teng, M.; Ling, A.; Gregor, V.;
Hong, Y.; Kiel, D.; Kuki, A; Shi, S.; Naerum, L.; Madsen, P.;
Lau, J.; Plewe, M. B.; Feng, J.; Jonhson, M. D.; Teston, K. A.;
Siedelmann, U. G.; Knudsen, L. B. PCT Int. Appl. WO 99/
01423, 1999. Chem. Abstr. 1999, 130, 110061.
19. Ling, A.; Yufeng, H.; Gonzalez, J.; Gregor, V.; Polinsky,
A.; Kuki, A.; Shi, S.; Teston, K.; Murphy, D.; Porter, J.; Kiel,
D.; Lakis, J.; Anderes, K.; May, J. J. Med. Chem. 2001, 44,
3141.
20. Livingston, J. N.; Schoen, W. R. Annu. Rep. Med. Chem.
1999, 34, 189.
21. Zhang, B. B.; Moller, D. E. Curr. Opin. Chem. Biol. 2000,
4, 461.
25. Hantzsch, A. Justus Liebigs Ann. Chem. 1882, 215, 1.
26. Sura Prakash, G. K.; Krishnamurty, R.; Olah, G. J. Am.
Chem. Soc. 1989, 111, 393.
22. Stable lines of hGR-CHO (Chinese hamster ovary) cells
were prepared as previously described (Heurich, R.; Buggy, J.;
Vandenberg, M.; Rossomando, A. Biochem. Biophys. Res.
Comm. 1996, 220, 905). Cells were grown in roller bottles then
pelleted. The cell pellet was subjected to Polytron homo-
27. Chiralpak AD column; isocratic elution 1.5 mL/min (99%
hexane/methyl t-butyl ether), retention times (33a, 7.3 min;
33b, 9.9 min).
28. Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95,
512.