Synthesis and structure-activity relationship of 4-quinolone-3-carboxylic acid based inhibitors of glycogen synthase kinase-3β

Article abstract of DOI:10.1016/j.bmcl.2011.07.073

The synthesis, GSK-3β inhibitory activity, and anti-microbial activity of bicyclic and tricyclic derivatives of the 5,7-diamino-6-fluoro-4-quinolone- 3-carboxylic acid scaffold were studied. Kinase selectivity profiling indicated that members of this class were potent and highly selective GSK-3 inhibitors.

Full text of DOI:10.1016/j.bmcl.2011.07.073

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5948–5951
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and structure–activity relationship of 4-quinolone-3-carboxylic
acid based inhibitors of glycogen synthase kinase-3b
Oana M. Cociorva a, , Bei Li a, , Tyzoon Nomanbhoy a, Qiang Li a, Ayako Nakamura b,
Kai Nakamura a, Masahiro Nomura b, Kyoko Okada b, Shigeki Seto b, Kazuhiro Yumoto b, Marek Liyanage a,
Melissa C. Zhang a, Arwin Aban a, Brandon Leen a, Anna Katrin Szardenings a, Jonathan S. Rosenblum a,
John W. Kozarich a, Yasushi Kohno b, Kevin R. Shreder a,
⇑
a ActivX Biosciences, Inc., 11025 N. Torrey Pines Road, La Jolla, CA 92037, USA
b Discovery Research Laboratories, Kyorin Pharmaceutical Co. Ltd, 2399-1, Nogi, Nogi-machi, Shimotsuga-gun, Tochigi, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis, GSK-3b inhibitory activity, and anti-microbial activity of bicyclic and tricyclic derivatives
of the 5,7-diamino-6-fluoro-4-quinolone-3-carboxylic acid scaffold were studied. Kinase selectivity pro-
filing indicated that members of this class were potent and highly selective GSK-3 inhibitors.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 27 May 2011
Revised 19 July 2011
Accepted 20 July 2011
Available online 5 August 2011
Keywords:
GSK-3b inhibitor
4-Quinolone
3-Carboxylic acid
Anti-bacterial
Kinase profiling
Glycogen synthase kinase-3 (GSK-3) is a serine/threonine pro-
tein kinase having and b isoforms that are encoded by distinct
as attenuate any anti-microbial activity that might be present in
this class of inhibitors.
a
genes.1 GSK-3 has been implicated in various diseases including
diabetes,2 Alzheimer’s disease, CNS disorders,3 and cardiomyocete
hypertrophy.4 These diseases may be caused by, or may result in,
the abnormal operation of certain cell signaling pathways in which
GSK-3 plays a role. For example, GSK-3 has been found to phos-
phorylate and modulate the activity of a number of regulatory pro-
teins, including glycogen synthase, which is the rate-limiting
enzyme required for glycogen synthesis. Even though small mole-
cule inhibitors of GSK-3 have been reported in literature5 there is,
however, a continued need to find more effective therapeutic
agents to treat GSK-3 mediated diseases.
The synthesis of R7 analogs of compound 1 began with nucleo-
philic displacement of the 5-fluoro group of quinolone 27 with
benzylamine in toluene followed by palladium catalyzed hydrogen-
olysis to yield amine 4. Acid hydrolysis of the ester of compound 4
yielded acid 5. Nucleophilic displacement of the 7-fluoro group of
quinolone 5 with a variety of primary amines was accomplished in
DMSO at 85 °C to yield the inhibitors shown in Table 1.
Replacement of the R7 side chain with selected amines had a sig-
nificant impact on potency. Incorporation of the branched isopropyl
or cyclohexyl amines resulted in derivatives (compounds 6 and 7,
respectively) with similar potency as the original HTS hit. Likewise,
use of 3-(2-ethylamino)indole- (8) or morpholine (9) -containing
As part of an effort to find novel GSK-3b inhibitors, we con-
ducted a high throughput screen of an internal compound library
which led to the identification of compound 1 ( Fig. 1) as a
900 nM inhibitor of GSK-3b.6 Interestingly, this compound was
originally synthesized as an intermediate during a medicinal
chemistry effort to identify 4-quinolone derived anti-microbial
agents. As part of a hit-to-lead optimization effort, we set out to
improve the potency of this hit by modifying the 7-position as well
⇑
Corresponding author. Tel.: +1 858 526 2576.
E-mail address: kevins@activx.com (K.R. Shreder).
These authors contributed equally to this work.
Figure 1. The structure of GSK-3b HTS hit
numbering scheme for the bicyclic 4-quinolone-3-carboxylic acid based inhibitors
in this study.
1 (GSK-3b IC50 = 900 nM) and the
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.07.073

O. M. Cociorva et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5948–5951
5949
Table 1
GSK-3b IC50 values for R7 analogs of the HTS hit 1
Compound
R7
GSK-3b IC50 (nM)
6
1200
7
8
9
600
660
2100
440
10
Scheme 1. Synthesis of R7 analogs of the HTS hit 1. Reagents and conditions: (a)
benzylamine, Et3N, toluene, 90 °C, 2 h, 82%; (b) H2, 10% Pd/C, AcOH, 50 °C, 2 h, 89%;
(c) AcOH/H2SO4/H2O, reflux, 2 h, 77%; (d) RNH2, Et3N, DMSO, 85 °C, 1 h.
11⁄
12
45
290
13
14
3400
45
15
16
47
100
17
18
190
110
Scheme 2. Synthesis of R7 amine derivatives of 5-amino and 5-hydro derivatives of
1-cyclopropyl-6-fluoro-8-methoxy-4-quinolone-3-carboxylic acid. Reagents and
conditions: (a) AcOH/H2SO4/H2O, reflux (b) RNH2, Et3N, DMSO, 85 °C.
*
Profiled against a panel of human kinases (vide infra).
Having investigated the 7-position, we turned our exploratory
SAR efforts to the primary amine present in the 5-position. Specif-
ically, we synthesized (see Scheme 2) 8-methoxy analogs of 11 and
15 with and without the C5 amine present to assess the impact of
this functional group on the inhibitory activity of GSK-3b (Table 2).
The replacement of the 8-F group present in earlier series with a 8-
methoxy group was made to minimize the phototoxic potential
generally associated with 8-F quinolones.8 After the acidic hydroly-
sis of the 4-quinolone ethyl esters 19a and 19b,9 nucleophilic dis-
placement of the 7-fluoro group adjacent to the methoxy groups of
carboxylic acids 20a and 20b with 3-phenyl-propylamine or N-(1-
naphthyl)ethylenediamine yielded the compounds in Table 2.
Comparison of the analogous 5-hydro versus 5-amino derivatives
(21 vs 22; 23 vs 24) indicated the presence of the 5-amino group
yielded significantly more potent GSK-3b inhibitors (260- and
840-fold, respectively).
side chains offered no improvement in potency. To examine the
influence of chain length, a series of 7-phenalkyl amino derivatives
(10–12) were examined and it was found that a chain length of
three methylenes (11) versus two (10) or four (12) was found to
be optimal. Consistent with this observation was the comparative
potency of the 1-imidazole-ethyl-amine derivatives 13 and 14
where compound 14, which contained linker of three methylenes
was a factor of 75 more potent than compound 13, with a linker
length of two methylenes. Other compounds containing selected
aryl-containing R7 side chains (15–17) showed improved potency
relative to the HTS hit 1 and exhibited GSK-3b IC50 values ranging
between 40 nM and 200 nM. Interestingly, despite the lack of an
aromatic group, derivative 18 was a relatively potent inhibitor with
GSK-3b IC50 value of 110 nM.

5950
O. M. Cociorva et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5948–5951
Table 2
Table 4
GSK-3b IC50 values for two 8-methoxy-5-amino-4-quinolone-3-carboxylic acid
GSK-3b IC50 values and anti-bacterial activity for bicyclic 4-quinolone-6-carboxylic
derivatives and their 5-hydro analogs
acid derivatives
Compound
R7
R5
GSK-3b IC50 (nM)
Compound
X
Y
GSK-3b IC50 (nM)
Anti-bacterial activity
MIC ( M)
21
22
–H
–NH2
17000
65
l
E. coli
>243
4
S. aureus
23
24
–H
–NH2
26000
31
35
560
44
243
2
36*
37*
31
>2
>2
38
39
40
75
>29
59
14
Table 3
GSK-3b IC50 values and anti-bacterial activity for bicyclic 4-quinolone-3-carboxylic
acid derivatives
49
59
440
>37
>37
41
160
>7
>7
Compound R8
R1
GSK-3b IC50 (nM)
Anti-bacterial activity
MIC ( M)
l
42
57
12
>36
>71
>36
71
E. coli
S. aureus
43*
25
26
–H
13
0.8
0.2
*
–OCH3
160
9
1
7
Profiled against a panel of human kinases (vide infra).
l
M MIC values, all of these compounds had GSK-3b IC50 values sig-
27
–OCH3
–F
1500
>28
nificantly higher than 100 nM. It was clear from this study that the
interplay between the 8 and 1-positions was critical in identifying
the subset of compounds we were targeting.
28*
22
>29
>29
To further explore this interplay, tricyclic 4-quinoline deriva-
tives (Table 4) were synthesized as shown in Scheme 1, but from
a tricylic starting material, an exemplary synthesis of which is
shown in Scheme 3. Treatment of the ketoester 32 with triethylor-
thoformate followed by (R)-valinol yielded the intermediate imine
33. The cyclization of this material was accomplished using potas-
sium fluoride as a base and microwave irradiation12 to yield the
trifluoroquinolone 34. This compound was further elaborated as
shown in Scheme 1 to yield the inhibitor 37 (Table 4).
29
30
–OCH3 –H
–OCH3 –CH3
1000
270
25
>13
25
13
31
–OCH3
16000
>28
>28
*
Profiled against a panel of human kinases (vide infra).
Selected examples of these tricyclic quinolones yielded sub-
100 nM GSK-3b inhibitors and minimal anti-bacterial activity. The
3-(R)-methyl derivative 35 was found to be a relatively poor GSK-
3b inhibitor (GSK-3b IC50 = 560 nM) but have essentially no anti-
bacterial activity. By contrast, its enantiomer 36 was 13 times more
potent against GSK-3b (GSK-3b IC50 = 44 nM) with single digit
micromolar MIC values. In the case of the analogous chiral isopro-
pyl derivatives 37 and 38, the difference in potency was not as pro-
nounced with GSK-3b IC50 values of 31 nM and 75 nM, respectively.
Substituted achiral derivatives were also investigated including the
3,30-gem-dimethyl derivative 39 which was found to have a GSK-3b
IC50 value of 49 nM. Moving this gem-dimethyl group to the 2-posi-
tion (40) or expanding the saturated ring of compound 39 (i.e., com-
pound 41) resulted in derivatives that were significantly less potent
Because for certain GSK-3b-related indications, anti-microbial
activity may not be desirable, we set out to identify a subset of
compounds with potent GSK-3b inhibitory activity but minimal
anti-bacterial activity.10 A variety of bicyclic quinolones with a
fixed R7 N-(2-pyridyl)ethylenediamine side chain and varied
groups in the 8 and 1-positions were synthesized and evaluated
(Table 3). Among these, compounds with a N1-cyclopropyl group
(25 and 26) gave the lowest MIC values, a result consistent with
the presence of this functionality in quinolone based antibiotics
with potent anti-bacterial activity.11 By contrast, two N1-cyclopen-
tyl derivatives (27 and 28) were also investigated and were found
to have significantly less anti-bacterial activity. While other 8-
methoxy derivatives with N1-hydro (29), N1-methyl (30), and N1-
isopentyl (31) groups were shown to have double digit or greater

O. M. Cociorva et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5948–5951
5951
lowed by trypsinization, kinase active site peptides were identified
and quantified using LC–MS/MS. Percent inhibition was calculated
by the decrease in the signal from each peptide relative to the
intensity seen in the absence of inhibitor. Over 90 kinases were
profiled including both GSK-3 isoforms, and selected data is shown
in the form of a heat map in Table 5.14 Notably, all 5 GSK-3b inhib-
itors were found to be highly selective for GSK-3, even when
screened against kinases from the CMGC family, in which both
GSK-3 isoforms are members. The only off-targets observed had
relatively weak inhibitory activity compared to GSK-3, with the
predicted affinity (based on the percent inhibitions observed at
the screening concentration) expected to be in the
In conclusion, the hit-to-lead modification of the HTS hit 1
yielded a series of sub- M, tricyclic GSK-3b inhibitors based on
lM range.
l
the 5,7-diamino-6-fluoro-4-quinolone-3-carboxylic acid scaffold.
Several potent GSK-3b inhibitors were determined to be highly
selective for GSK-3 when profiled against over 90 kinases. Addi-
tional SAR and modifications to improve the potency of this series
will be presented in due course.
Scheme 3. Synthesis of the tricyclic trifluoroquinolone 36, the starting material
used to produce compound 37 according to Scheme 1.
References and notes
Table 5
1. Kim, L.; Kimmel, A. R. Curr. Opin. Genet. Dev. 2000, 10, 508.
Kinase selectivity profiling (KiNativ™) of selected GSK-3b inhibitors in HL60 lysate
showing % inhibition of the indicated kinase(s) at a screening concentration of 10 lM
2. Coghlan, M. P.; Culbert, A. A.; Cross, D. A.; Corcoran, S. L.; Yates, J. W.; Pearce, N.
J.; Rausch, O. L.; Murphy, G. J.; Carter, P. S.; Roxbee Cox, L.; Mills, D.; Brown, M.
J.; Haigh, D.; Ward, R. W.; Smith, D. G.; Murray, K. J.; Reith, A. D.; Holder, J. C.
Chem. Biol. 2000, 7, 793.
3. Bhat, R. V.; Budd Haeberlein, S. L.; Avila, J. J. Neurochem. 2004, 89, 1313.
4. Haq, S.; Choukroun, G.; Kang, Z. B.; Ranu, H.; Matsui, T.; Rosenzweig, A.;
Molkentin, J. D.; Alessandrini, A.; Woodgett, J.; Hajjar, R.; Michael, A.; Force, T. J.
Cell Biol. 2000, 151, 117.
5. Gentles, R. G.; Hu, S.; Dubowchik, G. Annu. Rep. Med. Chem. 2009, 44, 736.
6. GSK-3b IC50 assay: 10–25 ng of recombinant full-length human GSK-3b
(Upstate) was incubated in the presence or absence of compound at varying
concentrations for 1 h at 30 °C in 20 mM MOPS, pH 7.0, 10 mM magnesium
acetate, 0.2 mM EDTA, 2 mM EGTA, 30 mM magnesium chloride, 62.5
lM
phospho-glycogen synthase peptide-2, ATP, 10 mM -glycerol
5
lM
a
phosphate, 1 mM sodium orthovanadate, and 1 mM dithiothreitol. Following
the completion of the kinase reaction an equal volume of KinaseGlo or
KinaseGlo Plus luciferase reagent (Promega) was added and the luminescence
read using a luminescence plate reader within 5–10 min. Compound activity
was expressed as % inhibition relative to maximal inhibition observed at the
maximal dose. IC50 values were calculated from the resulting curve using curve
fitting software (GraphPad Prism).
7. Moran, D. B.; Ziegler, C. B.; Dunne, T. S.; Kuck, A. K.; Lin, Y. J. Med. Chem. 1989,
32, 1313.
8. De Sarro, A.; De Sarro, G. Curr. Med. Chem. 2001, 8, 371.
9. Sanchez, J. P.; Gogliotti, R. D.; Domagala, J. M.; Gracheck, S. J.; Huband, M. D.;
Sesnie, J. A.; Cohen, M. A.; Shapiro, M. A. J. Med. Chem. 1995, 38, 4478.
10. Anti-bacterial activity was measured by determining MIC values using
standard microdilution methods recommended by the Clinical and
Laboratory Standards Institute with Mueller–Hinton Broth (Becton
Dickinson; Cockeysville, MD). The MIC was defined as the lowest
concentration of the compound that inhibited visible growth after incubation
for 18 h at 35 °C. See: Wikler, M. A.; Cockerill, F. R., III; Bush, K.; Dudley, M. N.;
Eliopoulos, G. M.; Hardy, D. J.; Hindler, J. F.; Patel, J. B.; Powell, M.; Turnidge, J.
D.; Weinstein, M. P.; Zimmer, B. L.; Ferraro, M. J.; Swenson, J.M. in Methods for
Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically;
Approved Standard-Eighth Edition. Clinical and Laboratory Standards Institute,
2009, M07-A8.
11. Bouzard, D.; Di Cesare, P.; Essiz, M.; Jacquet, J. P.; Ledoussal, B.; Remuzon, P.;
Kessler, R. E.; Fung-Tomc, J. J. Med. Chem. 1992, 35, 518.
12. We found the use of microwave chemistry was preferred to an oil bath for
sterically hindered ring closings.
13. (a) Patricelli, M. P.; Szardenings, A. K.; Liyanage, M.; Nomanbhoy, T. K.; Wu, M.;
Weissig, H.; Aban, A.; Chun, D.; Tanner, S.; Kozarich, J. W. Biochemistry 2007, 46,
350; (b) Patricelli, M. P.; Nomanbhoy, T. K.; Wu, J.; Brown, H.; Zhou, D.; Zhang,
J.; Jagannathan, S.; Aban, A.; Okerberg, E.; Herring, C.; Nordin, B.; Weissig, H.;
Yang, Q.; Lee, J.-D.; Gray, N. S.; Kozarich, J. W. Chem. Biol. 2011, 18, 699.
14. Other kinases that gave <35% inhibition but are not shown in Table 5: AKT1,
GSK-3b inhibitors. Replacement of the gem-dimethyl group of com-
pound 39 with cyclobutyl (42, GSK-3b IC50 = 57 nM) was well toler-
ated. When replaced with cyclopentyl (43) the resulting derivative
was found to be not only the most potent GSK-3b inhibitor in this
study (GSK-3b IC50 = 12 nM), but also have minimal anti-microbial
AKT2 , DMPK1, MSK1, p70S6K, PKAC
RPS6KC1, SGK3, EEF2K, CaMK2 , CHK1, DCAMKL1, MARK3, PHK
PKD2, RSK1/RSK2/RSK3, RSK4, CK1 , FRAP, PIK4CA, PIP5K2 , SMG1, AurA,
AurA/AurB/AurC, IKK , IKKb, IRE1, MPSK1, NEK6/NEK7, NEK9, NEK9, PEK, PKR,
a/PKAC
c
, PKC
a
/PKCb, PKN2, ROCK1/2,
c
c
2, PKD1/2,
a
a
activity (MIC P71 lM).
a
Selected potent GSK-3b inhibitors (compounds 11, 28, 36, 37,
and 43) were profiled against a panel of kinases using a desthiobi-
otin, acyl-phosphate ATP probe as previously described.13 Post
PLK1, Wee1, Wnk1/Wnk2/Wnk4, HSER, HPK1, KHS1/KHS2, LOK, MAP2K1,
MAP2K4, MAP3K2, MAP3K5, MST1, MST1, MST2, MST4/YSK1, OSR1, SLK,
STLK5, TAO1/3, TAO2, ABL1/ABL2, CSK, FER, FES, FGR, JAK1, LYN, PYK2, SYK,
ARAF, BRAF, IRAK4, RIPK3, TAK1, ZAK.
probe-labeling of HL60 lysate in presence of inhibitors (10 lM) fol-