Synthesis of selective SRPK-1 inhibitors: Novel tricyclic quinoxaline derivatives

Article abstract of DOI:10.1016/j.bmcl.2005.04.064

SR protein-specific kinase-1 (SRPK-1) has been identified as a validated target for hepatitis B virus (HBV). A series of novel tricyclic quinoxaline derivatives was designed and synthesised as potential kinase inhibitory antiviral agents and was found to be active and selective for SRPK-1 kinase. Most of these novel compounds have drug-like properties according to experimentally determined Log P and Log S values.

Full text of DOI:10.1016/j.bmcl.2005.04.064

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3241–3246
Synthesis of selective SRPK-1 inhibitors:
Novel tricyclic quinoxaline derivatives
a
Zsolt Szekelyhidi, Janos Pato, Frigyes Waczek, Peter Banhegyi,
b
a
a
´
Balint Hegymegi-Barakonyi, Daniel Eros, Gyo¨rgy Meszaros, Ferenc Hollosy,
´
´
´
´
´
a
b
a
a
}
Doris Hafenbradl, Sabine Obert, Bert Klebl, Gyo¨rgy Keri
´
´
´ ´
´
´
d
d
d
a,b,
a,b,c
}
´
and Laszlo Orfi
*
´
aPeptide Biochemistry Res. Group of Hung. Acad. Sci. and Cooperative Research Centre, Semmelweis University,
´
Rippl-Ronai u. 37., Budapest 1062, Hungary
bVichem Chemie Research Ltd., Herman Otto u. 15., Budapest 1022, Hungary
c
}
Department of Pharmaceutical Chemistry, Semmelweis University, Hogyes Endre u. 9., Budapest 1092, Hungary
dAxxima Pharmaceuticals AG, Max-Lebsche-Platz 32, Munich 81377, Germany
Received 21 March 2005; accepted 28 April 2005
Available online 31 May 2005
Abstract—SR protein-specific kinase-1 (SRPK-1) has been identified as a validated target for hepatitis B virus (HBV). A series of
novel tricyclic quinoxaline derivatives was designed and synthesised as potential kinase inhibitory antiviral agents and was found to
be active and selective for SRPK-1 kinase. Most of these novel compounds have drug-like properties according to experimentally
determined LogP and LogS values.
Ó 2005 Elsevier Ltd. All rights reserved.
1. Introduction
quinoxaline derivatives, which act as inhibitors of plate-
let derived growth factor receptor kinase (PDGF). The
most active compound was a tricyclic benzo[g]quinoxa-
line (1, IC50 = 0.9 lM);7 therefore we synthesised a series
of its derivatives. The compounds were tested on differ-
ent targets and it was found that 2,3-di-thiophen-20-yl-
benzo[g]quinoxaline 2 had significant and selective
inhibitory activity against SRPK-1 kinase (Fig. 1).
SR protein kinases (SRPKs) and their substrates, the SR
family of serine/arginine-rich pre-mRNA splicing fac-
tors, appear to be key regulators of alternative splicing.
Their splicing activity is regulated by reversible serine
phosphorylation of their carboxyl-terminal serine/threo-
nine motifs by a number of interesting kinases, including
the serine/threonine protein kinases SRPK-1 and
SRPK-2, and the scleroderma autoantigen topoisomer-
ase I. Both SRPK-1 and SRPK-2 are most likely the cel-
lular protein kinases mediating hepatitis B virus (HBV)
core protein phosphorylation during viral infection and
therefore represent important host cell targets for thera-
peutic intervention in HBV infection.1
Even though compound 2 was very active in the enzyme
assay, it suffered from high lipophilicity and poor water
solubility. These properties make it difficult to develop
this compound into a pharmaceutical agent. Therefore,
we modified the structure with hydrophilic moieties8
and tested the resulting compounds on SRPK-1 kinase
assay (Table 3) and on a selectivity panel of 19 kinase
assays (Table 4).
The quinoxaline scaffold can be found in a number of
biologically active compounds especially antibacterial,2
anti-inflammatory3 and antitumour4–6 drug candidates.
Their kinase inhibitory activity might be responsible
for their therapeutic effects. Gazit et al. prepared several
Keywords: SRPK-1; Quinoxaline; Kinase inhibitor.
*
Corresponding author. Tel.: +36 1 487 2080; fax: +36 1 487 2081;
e-mail: keri@vichem.hu
Figure 1. Tricyclic benzo[g]quinoxaline leads.
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.04.064

´
Z. Szekelyhidi et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3241–3246
3242
2. Physicochemical properties
(Table 2). The symmetric key intermediate compounds
4a–d were obtained via commercially available 2- or 3-
thiophene-alkyl-carboxylic acids 3a–d which were re-
acted with thionyl chloride and esterified with methanol,
then acylated with oxalyl chloride in the presence of alu-
minium chloride in Friedel–Crafts reaction. These
disubstituted thenil-esters 4a–d were reacted with naph-
thalene-2,3-diamine in ethanol, yielding 5a–d. Esters
were hydrolysed with aq NaOH–THF, producing the
appropriate alkyl-carboxylic acids 6a–d (Scheme 1).
The dicarboxylic acid derivative without alkyl linker 7
was prepared from lithiated 2 and carbon dioxide
(Scheme 2).
About one-half of the potential drug candidates fail be-
cause of poor ADME properties, which may be partially
related to the LogP and water solubility (S, [mol/l]) val-
ues of the compound. According to generally accepted
criteria of drug-likeliness, the LogP value of a drug can-
didate must be below 5. The concentration range where
the compounds are active requires that LogS be at least
6. We have experimentally determined the LogS and
LogP values of the compounds by a dedicated HPLC
method.9 We found that all modified derivatives of 2
had lower lipophilicity than the parent compound. The
LogP value of 2 is 5.98, much higher than those of com-
pounds 6a–c, 15 and 16, which had LogP values of <4.5.
Replacement of the first aromatic ring with alkylene-di-
oxo derivatives (17–20) should also improve the physi-
cochemical properties of this class of compounds.
Fifteen out of eighteen newly synthesised compounds
had higher water solubility than LogS = ꢀ2 and ten
compounds had lower lipophilicity than LogP = 5. Vi-
chem Chemie Research Ltd has developed an automatic
QS(P)AR program called 3DNET4W.10–12 By using
3DNET4W software we have also calculated the LogP
values of selected quinoxalines and benzo[g]quinoxa-
lines to compare the experimental and predicted values.
We found that our software had good LogP predicting
capability (Table 1).13,14
Carboxyl group was also inserted onto the position 7 of
2, producing 2,3-di-thiophen-20-yl-benzo[g]quinoxaline-
7-carboxylic acid 13. This synthesis started from com-
mercially available 2,3-dimethoxy-naphthalene 8 as
shown in Scheme 3. Its Friedel–Crafts acylation with
acetic anhydride resulted in 9, which was then converted
to 6,7-dimethoxy-naphthalene-2-carboxylic acid 10 with
aq NaOCl at 65 °C. The cleavage of methoxy groups
with 48% aq HI afforded 11, which was aminated in cc
NH4OH on 120 bar (12). This compound was con-
densed with 2,20-thenil in EtOH, yielding the desired
compound 13.
We modified the first aromatic ring of 2 with alkylene-
dioxo derivatives (17–20) and substituted the middle
phenyl ring with pyrazine, obtaining 23. Compounds
17–20 and 23 were prepared from 4,5-dimethoxy-benz-
ene-1,2-diamine 14 (Scheme 4). Condensation of 14
with 2,20-thenil in ethanol followed by methoxy cleav-
age with HBr in glacial acetic acid at 100 °C provided
2,3-di-thiophen-2-yl-quinoxaline-6,7-diol 16. Compounds
17–19 were afforded by cyclization of 16 with
corresponding alkyl-dihalogenides. The carboxyl-
derivative 20 was obtained by treatment of 16 with
3. Chemistry
Since the carboxyl function is very suitable for enhanc-
ing the water solubility of the molecules, they were in-
serted into parent compound 2, obtaining 6a–d, 7 and
13. A general synthetic route for the preparation of
benzo[g]quinoxalines disubstituted with thenil-alkyl-car-
boxylic acids was developed as outlined in Scheme 1
Table 1. Physicochemical properties of selected quinoxaline derivatives
Compound
Log P
C Log Pa
Log S
Compound
Log P
C Log Pa
Log S
2
5.98
4.95
5.00
5.00
5.50
3.30
4.11
4.28
4.77
6.33
5.02
5.04
5.15
5.90
4.00
3.99
4.16
4.89
ꢀ3.05
ꢀ1.02
ꢀ1.07
ꢀ2.29
ꢀ2.12
ꢀ0.13
ꢀ0.24
ꢀ0.19
ꢀ0.10
7
5.03
5.40
4.00
4.20
4.80
5.00
5.00
4.70
4.79
4.78
5.64
4.92
4.14
5.06
4.94
4.98
4.01
5.08
ꢀ0.11
ꢀ0.60
ꢀ1.08
ꢀ1.10
ꢀ1.98
ꢀ1.08
ꢀ1.19
ꢀ0.07
ꢀ0.78
5a
5b
5c
5d
6a
6b
6c
6d
13
15
16
17
18
19
20
23
a 3DNET4W software; C Log P = calculated Log P.
Scheme 1. Reagents and conditions: (i) thionyl chloride/MeOH, rt, 3 h; (ii) oxalyl chloride/AlCl3, abs DCM, rt, 3 h; (iii) naphthalene-2,3-diamine,
EtOH, 79 °C, 2 h; (iv) aq NaOH/THF, rt, 3 h, then 1 N HCl.

´
Z. Szekelyhidi et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3241–3246
3243
Table 2. Side-chain modification on thiophene rings
4. Biological results
Compounds
Position of modification
n
The compounds were tested on SRPK-1 assay and 10
compounds were found to inhibit SRPK-1 activity to low-
er than 10 lM IC50 (Table 3).15 The most active com-
pound on the SRPK-1 kinase assay was the compound
13 (IC50 = 0.04 lM). The dose–activity curve of 13 is pre-
sented in Figure 2. For statistical studies, we calculated
the Hill parameter and linear regression of the dose–activ-
ity curve. We found good correlation between observed
and predicted values of doses and activity (r2 = 0.9734),
and the shape of the dose–activity curve was ideal as well
(Hill = 0.62). Inhibition specificity of selected quinoxa-
lines and benzo[g]quinoxalines was investigated on a
selectivity panel of 19 kinases (Table 4).15 All compounds
were measured at 10 lM concentration in triplicate. It
seems squarely that carboxylic acid derivatives 6a–d, 13,
20 and the diaza derivative 23 exhibited high inhibitory
activity against SRPK-1 kinase. Alkyl-carboxylic acid
derivatives 6a–d showed higher inhibition of SRPK-1 ki-
nase than ester analogues 5b–d. The compounds 6a and b,
which have an acetic acid group at 2 or 3 position on the
thiophene moiety, exhibited medium inhibitory activity
towards EGFR tyrosine kinase andMAPK-ERK1 kinase
even at 10 lM concentration. When carboxyl group was
inserted into the position 7 of 2 (compound 13), we found
that besides the excellent SRPK-1 inhibitory activity, c-
Src inhibitory activity was detected. Replacement of the
first phenyl ring with alkylene-dioxo derivatives (17–19)
3–6 a
3–6 b
3–6 c
3–6 d
2
3
2
2
1
1
2
3
Scheme 2. Reagents and conditions: (i) LDA/THF, ꢀ78 °C, 2 h, then
CO2, rt, 3 h.
2,3-dibromo-propionic acid ethyl ester, and triethyl-
amine in EtOH, followed by removing the ester func-
tion with aq NaOH. Compound 23 was also
prepared in three steps from 14. The first one was cycli-
zation of 14 with oxalic acid, and then quinoxaline-2,3-
dione was converted to dichloro-quinoxaline 21 with
phosphorus oxychloride. Compound 21 was substituted
with ammonia to give diamino-quinoxaline 22, which
was reacted with 2,20-thenil in ethanol to produce the
desired diaza derivative 23.
Scheme 3. Reagents and conditions: (i) Ac2O, AlCl3/nitrobenzene, 0 °C, 2 h; (ii) aq NaOCl, 65 °C, 2 h; (iii) 48% HI, reflux, 5 h; (iv) cc NH4OH,
120 atm, 180 °C, 48 h; (v) 2,20-thenil/EtOH, reflux, 2 h.
Scheme 4. Reagents and conditions: (i) 2,20-thenil/EtOH, reflux, 2 h; (ii) HBr/glacial acetic acid, 100 °C, 3 h; (iii) Br–X–Br/10% aq NaOH/
dichloromethane, 20–100 °C; (iv) (1) 2,3-dibromo-propionic acid ethyl ester, TEA/EtOH, rt, 3 h, then (2) aq NaOH; (v) (1) oxalic acid/HCl/water,
reflux, 2 h, (2) POCl3/reflux, 2 h; (vi) cc NH4OH, 120 atm, 180 °C, 48 h; (vii) 2,20-thenil/EtOH, reflux, 2 h.

´
Z. Szekelyhidi et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3241–3246
3244
Table 3. Inhibitory activity of selected quinoxalines on SRPK-1 kinase
assay
decreased SRPK-1 activity and resulted in high c-Src
inhibitory potency. Compounds 17 and 18 exhibited med-
ium activity against PDGFR-beta tyrosine kinase. The
dimethoxy-quinoxaline 15, which is the open form of
18, did not show inhibitory activity against any kinase.
Very interestingly, the addition of a carboxyl group at
the position 2 of 18 (compound 20) resulted in a drastic
change of activity. Compound 20 did not show the inhibi-
tion of c-Src kinase, but specific and high inhibitory activ-
ity was observed in the case of SRPK-1 kinase. The diaza
derivative 23 did not have significantly modified activity,
but it showed higher selectivity and lower lipophilicity
than compound 2.
Compound
SRPK-1 IC50, lMa
2
1.36 ( 0.05)
11.75 ( 0.09)
6.13 ( 1.67)
na
5b
5c
5d
6a
6b
6c
6d
7
1.28 ( 0.05)
0.64 ( 0.03)
0.87 ( 0.01)
0.82 ( 0.10)
10 ( 0.03)
0.04 ( 0.02)
16.31 ( 0.25)
9.61 ( 0.28)
28.49 ( 0.72)
23.36 ( 0.80)
na
13
15
16
17
18
19
20
23
According to the selectivity data, the most potent and
selective compounds against SRPK-1 kinase were the
carboxyl derivatives 6a–d, 20 and the diaza derivative
23. These compounds had higher selectivity than the
parent compound 2 and besides the kinase inhibitory
activity, they had better physicochemical properties as
well. The most potent compound on SRPK-1 kinase
assay was the compound 13 (IC50 = 0.04 lM) but it
showed high activity on c-Src kinase assay as well.
0.95 ( 0.04)
1.14 ( 0.01)
a Values are means of three experiments; standard deviation is given in
parentheses (na = not active).
5. Conclusion
In summary, we have synthesised a series of novel
tricyclic quinoxaline analogues as SRPK-1 inhibitors
with advantageous physicochemical properties. We
found that carboxylic acid derivatives 6a–d, 20 and
the diaza compound 23 had high and more selective
inhibitory activity against SRPK-1 kinase than the
parent compound 2. The most active compound against
SRPK-1 was 13 but it showed lower selectivity. Correla-
tion was found between the structure and the SRPK-1
inhibitory activity. According to the structure–activity
relationships, we found that position of carboxyl
group and replacement of the first phenyl ring of 2 to
alkylene-dioxo derivatives can cause a drastic change
Figure 2. Dose–activity curve of compound 13 on SRPK-1 kinase
assay.
Table 4. Inhibitory activity of selected quinoxalines in a selectivity panel of 19 kinase assays (%)
Compounds a
c-Src
2
5b
5c
19
5d
6a 6b
6c 6d
70 35
7
13
83 ( 3)
15
10
16
17
18
19
20
19
23
20
8
ꢀ12 14
29
17
ꢀ7
20
ꢀ3 71 ( 3) 84 ( 1) 75 ( 6)
ꢀ7 ꢀ6
9
AKT1/PKBa
Abl
12
1
14
ꢀ6 15
0
ꢀ13
12 ꢀ10
ꢀ6
ꢀ2
ꢀ5
ꢀ1
10
ꢀ12
11
1
ꢀ13
10
4
ꢀ4
ꢀ1
27
2
4
3
7
6
6
6
ꢀ10
ꢀ4
3
ꢀ8
12
14
32
7
15 10
3
ꢀ11 ꢀ13
11
ꢀ11
10
40
4
1
7
9
14
ꢀ4
9
ꢀ6
10
CDK2/CycA(h)
CK1-alpha
EGFR
25
30 ꢀ2
52
8
ꢀ5 ꢀ4
18 40
54 40
22 23
37 54
7
24
32
13
43
3
7
ꢀ7
ꢀ5
26
8
15
39
11
2
8
49
51
29
25 35
34
ꢀ15
38
48
24
46
31
61
41
19
4
20
4
GSK-3beta
IKK-beta
InsR
24 47
16 25
ꢀ13
10
5
3
ꢀ1
ꢀ12
22
33
14
14
31
42
42
20
56
18
68
15
3
28
10
ꢀ7
55 ꢀ16 ꢀ12 ꢀ11
5
ꢀ14 ꢀ5 ꢀ11 ꢀ19
ꢀ7
ꢀ13
ꢀ6
10
ꢀ4 29
ꢀ3 18
ꢀ7 38
25 14
34
15
ꢀ12
Jnk1a1
Kit
6
8
ꢀ10 ꢀ10
9
5
8
ꢀ3 ꢀ23 ꢀ5 ꢀ12 ꢀ25
ꢀ15 ꢀ15
1
11
9
12
13
13
5
15
52
17 52
54 11
22 49
18 18
27 30
40 18
21 32
33
6
12
49
22
21
4
39
5
ꢀ4
20
5
MAPK-ERK1
PDGFR-beta
ROCK2
RSK1
11
15
19
6
47 10
5
37
0
23
19
7
ꢀ11 54
ꢀ18 31
24
32
4
5
2
15
ꢀ1
14
12
ꢀ2
ꢀ2
ꢀ1
ꢀ1
48
8
13
17
ꢀ2
4
31
18
39
5
38 23
42 39
12 27
20
12
7
31
16
6
3
1
21
34
41
6
17
c-Raf
cMet
16
43
6
11
ꢀ5
27
ꢀ4
ꢀ19
14
49
47
5
ꢀ3
9
0
42
53
ꢀ11
12
ꢀ7
ꢀ2
P56Lck
P70S6K
ꢀ19
2
ꢀ9
8
10 29
ꢀ3
9
18
1
39
2
8
ꢀ5
11
ꢀ7 17
ꢀ9
ꢀ8
a Values were determined at 10 lM concentration in triplicate; standard deviation is given in parentheses.

´
Z. Szekelyhidi et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3241–3246
3245
Buffered Saline; pH 7.4). Five microlitres of each sample
with 95 lL PBS was dispensed in the SAMPLE plate
(96-well NUNC plate). After 1 h of equilibration time,
the SAMPLE plate was filtered onto the FILTER plate
(Millipore hydrophilic PTFE filter plate) by centrifuga-
tion (10 min at 1500 rpm). And then 10 lL of each well
of STANDARD and FILTER plates (including the
blanks) was injected into the HPLC device. The same
procedure was carried out with our internal standards
using compounds of the chromatography hydrophobic-
in activity. These new compounds can be starting points
for the development of novel antiviral agents.
Acknowledgments
This work was supported by a research grant of NKFP
´
1A/0020/2002 and OTKA T047478. We thank Mihaly
Hegedus, Chemical Research Centre of Hungarian
}
Academy of Science, for the preparation of diamino-
´
´
ity index test mixture according to Valko et al. (1997).
derivatives and Istvan Szabadkai for her helpful advice.
We are thankful to Birgit Flicke and Sieglinde Schinzel
for excellent technical assistance.
Stock solutions of internal standards such as benzimid-
azole, 5-phenyl-1H-tetrazole, theophyllin, 8-phenylteo-
phylin, indole, acetofenone, propiofenone, butirofenone,
valerofenone (all from Sigma) were prepared with
DMSO (5 mg/ml solution) and 5 lL of each standard
was mixed in 95 lL PBS and then subjected to
chromatography measurements. Solubility of the com-
pounds was calculated from the gradient retention times
corrected with the slope and intercept values obtained
from the test curve.
References and notes
1. Daub, H.; Blencke, S.; Habenberger, P.; Kurtenbach, A.;
Dennenmoser, J.; Wissing, J.; Ullrich, A.; Cotten, M.
J. Virol. 2002, 8124.
peak area ‘sample’
peak area ‘std’=10 ꢁ sample injðlLÞ=std concðlg=mlÞ=1000
Solubilityðmg=mlÞ ¼
2. Carta, A.; Loriga, M.; Zanetti, S.; Sechi, L. A. Farmaco
2003, 58, 1251.
Log S [M] = Log10 (concentration (mg/ml)/molecular weight).
14. Measurement of Log P and Log S by HPLC: During high
performance liquid chromatography measurements, the
following system was applied. RP-C18 column (LiChro-
spher), 10 cm, 4.6 I.D. with a particle size of 10 lm.
3. Brown, M. F.; Avery, M.; Brissette, W. H.; Chang, J.
H.; Colizza, K.; Conklyn, M.; DiRico, A. P.; Gladue,
R. P.; Krueger, S. S.; Lira, P. D.; Lillie, B. M.;
Lundquist, G. D.; Mairs, E. N.; McElroy, E. B.;
McGlynn, M. A.; Paradis, T. J.; Poss, C. S.; Rossulek,
M. I.; Shepard, R. M.; Smiths, J.; Strelevitz, T. J.;
Truesdell, S.; Tylaska, L. A.; Yoon, K.; Zheng, D.
Bioorg. Med. Chem. Lett. 2004, 14, 2175.
4. Hazeldine, S. T.; Polin, L.; Kushner, J.; Paluch, J.; White,
K.; Edelstein, M.; Palomino, E.; Corbett, T. H.; Horwitz,
J. P. J. Med. Chem. 2001, 44, 1758.
5. Myers, M.; He, W.; Hanney, B.; Setzer, N.; Maguire,
M.; Zulli, A.; Bilder, G.; Galzcinski, H.; Amin, D.;
Needle, S.; Spada, A. Bioorg. Med. Chem. Lett. 2003,
13, 3091.
Eluents:
A eluent: 50 mM ammonium acetate, pH 7.4
B eluent: 100% acetonitrile
Linear gradient system:
Time (min)
0–1
1–8
B (%)
0
100
100
0
8–11
11–12
12–16
0
6. He, W.; Myers, M.; Hanney, B.; Spada, A.; Bilder, G.;
Galzcinski, H.; Amin, D.; Needle, S.; Page, K.; Jayyosi,
Z.; Perrone, M. Bioorg. Med. Chem. Lett. 2003, 13, 3097.
7. Gazit, A.; App, H.; McMahon, G.; Chen, J.; Levitzki, A.;
Bohmer, F. J. Med. Chem. 1996, 39, 2170.
The flow rate was 1 mL/min.
Temperature: 25 °C
Detection: k = 230 and 254 nm.
For Log P calibration, the following test compounds were
used.
}
´
´
´
8. Pato, J.; Keri, G. Y.; Orfi, L.; Waczek, F.; Horvath, Z.;
Banhegyi, P.; Szabadkai, I.; Marosfalvi, J.; Hegymegi-
´
´
No.
Name
Log P
´
Barakonyi, B.; Szekelyhidi, Z. S.; Greff, Z.; Choidas, A.;
Bacher, G.; Daub, H.; Obert, S.; Kurtenbach, A.;
Habenberger, P. WO Patent 02/094796 A2, 2002.
1
2
Theophylline
Colchicine
ꢀ0.05
0.92
1.42
1.55
1.66
2.05
2.14
2.2
´
9. Valko et al. Anal. Chem. 1997, 69, 2022.
´
3
5-Phenyl-1H-tetrazole
Benzimidazole
Acetophenone
8-Phenyl-theophylline
Indole
´
´
10. Ero}s, D.; Ko¨vesdi, I.; O}rfi, L.; Takacs-Novak, K.; Acsady,
4
´
Gy.; Keri, Gy. Curr. Med. Chem. 2002, 9, 1819.
´
5
´
11. Ero}s, D.; Keri, G. Y.; Ko¨vesdi, I.; Szantai-Kis, C. S.;
6
´
´
Meszaros, G. Y.; O}rfi, L. Mini-Rev. Med. Chem. 2004, 4,
167.
7
}
12. Ko¨vesdi, I.; Keri, G. Y.; Orfi L. WO Patent 02/082329,
´
8
Propiophenone
Butyrophenone
Valerophenone
2002.
9
10
2.73
3.26
13. Determination of solubility: Compounds were submitted
for the solubility assay as 10 mM solutions in DMSO in
a mother (96-well NUNC plate) plate. From this plate
two daughter plates were derived: a STANDARD plate
containing compounds in DMSO solution and a sample
plate of the compounds in PBS (Fluka Phosphate-
15. General description of the kinase assays: the kinase
activity was assayed in 96-well microtitre plates at a final
compound concentration of 10 lM in a total volume of

´
Z. Szekelyhidi et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3241–3246
3246
50 ll. Compounds were dissolved in 100% DMSO to pre-
strate concentration was used at 5-fold excess over the
Km for the substrate. The reaction was incubated for 1 h
at room temperature and stopped by addition of 10 ll of
50 mM EDTA. The assay was transferred to a 96-well
MAPH filter plate (Millipore). The radioactivity was
counted on a Microbeta microplate counter.
pare a 10 mM stock solution and then diluted with buffer
to reach a 10 lM (or the required) final concentration. The
kinase concentration was used to yield 10% ATP turnover.
The ATP concentration was used at the Km ATP and
12.5 lCi/ml adenosine 50-[c-33P]triphosphate. The sub-