Kinase inhibitions in pyrido[4,3-h] and [3,4-g]quinazolines: Synthesis, SAR and molecular modeling studies

Article abstract of DOI:10.1016/j.bmc.2019.04.005

New pyrido[3,4-g]quinazoline derivatives were prepared and evaluated for their inhibitory potency toward 5 protein kinases (CLK1, DYRK1A, GSK3, CDK5, CK1). A related pyrido[4,3-h]quinazoline scaffold with an angular structure was also synthesized and its

Full text of DOI:10.1016/j.bmc.2019.04.005

Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Kinase inhibitions in pyrido[4,3-h] and [3,4-g]quinazolines: Synthesis, SAR
and molecular modeling studies
Wael Zeinyeha, Yannick J. Esvana, Béatrice Josselinb, Blandine Baratteb, Stéphane Bachb,
⁎
Lionel Nautona, Vincent Thérya, Sandrine Ruchaudb, Fabrice Anizona, Francis Girauda,
,
⁎
Pascale Moreaua,
a Université Clermont Auvergne, CNRS, SIGMA Clermont, ICCF, F-63000 Clermont-Ferrand, France
b Sorbonne Université, CNRS, Plateforme de criblage KISSf (Kinase Inhibitor Specialized Screening facility), Protein Phosphorylation and Human Diseases Unit, Station
Biologique, Place Georges Teissier, F-29688 Roscoff, France
A R T I C L E I N F O
A B S T R A C T
Keywords:
New pyrido[3,4-g]quinazoline derivatives were prepared and evaluated for their inhibitory potency toward 5
protein kinases (CLK1, DYRK1A, GSK3, CDK5, CK1). A related pyrido[4,3-h]quinazoline scaffold with an angular
structure was also synthesized and its potency against the same protein kinase panel was compared to the
analogous pyrido[3,4-g]quinazoline. Best results were obtained for 10-nitropyrido[3,4-g]quinazoline 4 toward
CLK1 with nanomolar activities.
Pyrido[3,4-g]quinazolines
Pyrido[4,3-h]quinazoline
Protein kinase inhibition
CLK1
1. Introduction
leading to improved potencies toward CDK5/GSK3.5 In addition, in this
series, the presence of a nitro group at the 10-position could lead to a
CLK1 and DYRK1A are serine-threonine protein kinase members of
the CMGC family. CLK1 and DYRK1A are implicated in pre-mRNA
splicing regulation and, in this context, phosphorylate various serine/
arginine-rich proteins. Moreover, DYRK1A and CLK1 are over-ex-
pressed in certain types of cancer. Therefore both protein kinases are
involved in numerous signalization pathways implicated in various
diseases such as cancer or neurodegenerative disorders.1–3 These two
protein kinases constitute relevant targets for the development of new
cellular tools and/or therapies. Therefore, the development of new
small molecules able to modulate their activity is of high interest. As
part of our research program dedicated to the discovery of new het-
eroaromatic compounds exhibiting protein kinase inhibitory potencies,
we recently reported the synthesis of pyrido[3,4-g]quinazolines with
interesting activities toward CLK1 (cdc2-like kinase 1)/DYRK1A (dual
specificity tyrosine-phosphorylation-regulated kinase 1A) or CDK5
(cyclin dependent kinase 5)/GSK3 (glycogen synthase kinase 3), de-
pending on the heteroaromatic scaffold substitution (Fig. 1). Previous
structure-activity relationship studies (SAR) undertaken on this scaffold
demonstrated that the aminopyrimidine and the pyridine parts were
essential to the interactions within the ATP-binding site of the targeted
kinases.4–6 Moreover, we reported that no substitution at the 5-position
led to potent CLK1/DYRK1A inhibitions. In contrast, the substitution at
the 5-position by alkyl/aryl groups impaired CLK1/DYRK1A inhibition
better selectivity profile for CLK1 over DYRK1A while a 10-amino
substitution could achieve a better activity toward DYRK1A compared
to CLK1.4,6 Recently, we demonstrated that the incorporation at the 2-
position of aminoethylamino or aminopropylamino substituents did not
cause major steric hindrance issues and therefore could lead to potent
CLK1 and/or DYRK1A inhibitors (Fig. 1). To enlarge the SAR studies
performed around this scaffold, we describe herein the preparation of
new aminoalkylamino/aminoalkylimino derivatives bearing shortened
and/or rigidified groups at the 2-position (Fig. 1). In addition, we
continued to explore the chemical space in the pyridoquinazoline series
by preparing an angular analogue of the linear pyrido[3,4-g]quinazo-
line scaffold. Thus, pyrido[4,3-h]quinazolin-2-amine 11 (Fig. 1) was
prepared to measure the impact of this structural change on kinase
inhibition.
The first 10-nitropyrido[3,4-g]quinazoline series was prepared in 11
steps from commercially available 2-chloro-4-nitrobenzoic acid.4,5 The
formation of the aminopyrimidine part is based on the condensation of
a guanidinium salt with key 6-chloro-5-nitroisoquinoline-7-carbalde-
hyde intermediate A (Scheme 1). As recently reported, this strategy was
employed to generate N-2 substituted derivatives, bearing NO2 or NH2
groups at the 10-position (compounds B) by reaction of aminoethyl- or
aminopropylguanidinium salts with A (Scheme 1).6 In this work we
report the preparation of new N-2 substituted analogues (compounds C)
⁎ Corresponding authors.
E-mail addresses: francis.giraud@uca.fr (F. Giraud), pascale.moreau@uca.fr (P. Moreau).
https://doi.org/10.1016/j.bmc.2019.04.005
Received 25 February 2019; Received in revised form 29 March 2019; Accepted 3 April 2019
0968-0896/©2019ElsevierLtd.Allrightsreserved.
Pleasecitethisarticleas:WaelZeinyeh,etal.,Bioorganic&MedicinalChemistry,https://doi.org/10.1016/j.bmc.2019.04.005

W. Zeinyeh, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Fig. 1. Chemical structures of pyrido[3,4-g]quinazolines and pyrido[4,3-h]quinazolin-2-amine. Previous SAR studies and this work.
Scheme 1. Synthetic pathways leading to new pyrido[3,4-g]quinazolines C series.
directly from corresponding aminopyrimidine 1 (Scheme 1).
In addition to this SAR study around the pyrido[3,4-g]quinazoline
scaffold, we also prepared a regioisomeric analogue with an angular
pyrido[4,3-h]quinazoline structure. This was carried out from com-
pound D (Scheme 2) which was obtained as a minor iodination side-
product during the synthesis of compound A4 (Scheme 1) from 2-
chloro-4-nitrobenzoic acid (Scheme 2).
Isopropyl analogue 3 was prepared similarly by reaction between 1
and oxalyl chloride/diisopropylformamide in dichloromethane
(Scheme 3).11 Our first attempts to reduce both amidine moiety and
nitro group of compounds 2 and 3 using catalytic hydrogenation (H2,
Pd/C, CH2Cl2, MeOH) failed. Therefore, we performed the reduction of
the amidine part in the presence of NaBH4 in MeOH leading to com-
pounds 4 and 5 in moderate yields.12 Despite numerous efforts, we
never managed to get the corresponding 10-amino analogues of com-
pounds 2–5. Indeed, due to rapid degradation processes, all reduction
attempts led to the formation of complex reaction mixtures.
Finally, the activity of new synthesized compounds was evaluated
toward a panel of 5 protein kinases (CLK1, DYRK1A, GSK3, CDK5,
CK1). This panel was chosen on the basis of previous results (CLK1,
DYRK1A, GSK3, CDK5)4–6 and of known kinase cross-inhibitions due to
the ATP-binding site kinase homology (CK1).7
The preparation of pyrido[4,3-h]quinazolin-2-amine 11 started
from ethyl 4-amino-2-chlorobenzoate 6.4 The product of the first iodi-
nation step (I2, Ag2SO4, EtOH), containing minor regioisomer
D
(Scheme 2) was engaged in a 3-steps sequence starting by a CuCN-
mediated cyanation before iodination using a Sandmeyer type reaction
and reduction (Scheme 4). Minor regioisomer 7 was isolated in 7%
overall yield. After Sonogashira coupling of 7 with TMS-acetylene to
give 8, isoquinoline 9 was formed by reaction with ammonia in me-
thanol under microwave irradiation. Oxidation of 9 led to chlor-
oaldehyde 10 that was reacted with guanidine carbonate under mi-
crowave irradiation to give product 11.
2. Results and discussion
2.1. Chemistry
Diversely substituted pyrido[3,4-g]quinazolines were prepared from
known compound 1.4 Compound 2 was obtained in 97% yield under
Vilsmeier conditions using phosphorous oxychloride and DMF (Scheme
3).8–10
Scheme 2. Retrosynthetic pathway leading to angular pyrido[4,3-h]quinazoline analogue 11.
2

W. Zeinyeh, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Scheme 3. Preparation of 10-nitropyrido[3,4-g]quinazolines 2–5.
2.2. Kinase inhibition evaluation
R = H, R′ = NO2, R″ = H, Fig. 1). However, the tertiary amino group at
the 2-position could advantageously be converted into the corre-
sponding ammonium salt, thus achieving an improved solubility and a
better pharmacokinetic profile.
Compounds 2–5 and 11 were evaluated toward a panel of 5 protein
kinases CLK1, DYRK1A, CDK5/p25, GSK-3α/β and CK1δ/ε. The results
are expressed as % of residual kinase activities at 10 μM or 1 μM com-
pound concentration. Except for 11 presenting a new scaffold, IC50
values were determined only when % of residual kinase activity was
≤25% at 1 μM. As shown in Table 1, all compounds, including the
angular analogue 11, inhibited CLK1 more than, or around 50% at
1 μM, indicating IC50 values in the micro/submicromolar range. Com-
pared to CLK1, compounds 2–4 were less active toward CDK5, GSK3
and CK1 with an important decrease of inhibition at 1 μM. In summary,
most inhibited kinases were CLK1 > DYRK1A/GSK3 > CDK5/CK1
with 2-dimethylaminomethylamino derivative 4 being the most potent
of the series with IC50 value of 87 nM against CLK1. With the exception
of CK1, compound 4 exhibited similar in vitro kinase inhibitory po-
tencies compared to the reference of the nitro series (Compound 1,
Compound 11, first analogue of the angular series, exhibited mi-
cromolar potencies toward CLK1 and DYRK1A. In comparison, the
linear pyrido[3,4-g]quinazolin-2-amine analogue (R = R′ = R″ = H,
Fig. 1) was more active toward CLK1 and DYRK1A with submicromolar
activities (CLK1 IC50 value = 0.19 μM, and DYRK1A IC50
value = 0.18 μM).4 Therefore, in order to get insights for further opti-
mization of the kinase inhibitory potential in this new series, we per-
formed molecular modeling studies to determine its plausible binding
mode within CLK1 ATP-binding pocket.
2.3. Molecular modeling experiments
The docking experiments were performed using a CLK1 model
Scheme 4. Synthesis of pyrido[4,3-h]quinazolin-2-amine 11.
3

W. Zeinyeh, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Table 1
Kinase inhibitory potencies of reference compound 1 in the nitro series (R = H, R′ = NO2, R″ = H, Fig. 1) and new analogues 2–5, 11.
Cpds
Kinase inhibitory potencies
CLK1
DYRK1A
CDK5
GSK3
CK1
10 μM
2
1 μM
6
10 μM
5
1 μM
36
10 μM
56
1 μM
88
10 μM
25
1 μM
74
10 μM
9
1 μM
41
1
(0.069 μM)4
(0.62 μM)4
(> 10 μM)4
(2.2 μM)4
(0.78 μM)4
2
3
4
3
1
6
33
53
13
31
46
4
100
95
59
45
41
94
30
15
15
86
73
73
53
60
38
100
96
100
99
39
100
(0.087 μM)
(1.48 μM)
5
5
9
44
48
46
25
69
71
61
63
99
1
53
98
38
22
74
64
11
100
75
(2.18 μM)
% of Residual kinase activities and IC50 values in brackets. Assays performed using a 32P radioassay (method A), except for the determination of compound 11 IC50
values that was carried out using the ADP-Glo assay (method B). Compound 1 IC50 values against CLK1, DYRK1A, CDK5, GSK3 and CK1 were previously reported.4
Kinase activities were assayed in triplicate. Typically, the standard deviation of single data points was below 10%.
Fig. 2. A) Plausible binding mode of 11 within CLK1 ATP-binding site. B) Binding mode of linear pyrido[3,4-g]quinazolin-2-amine analogue solved by X-ray
crystallography (PDB code 5J1W).4 The images were produced using UCSF Chimera.16
generated from 6FT8 structure available in the Protein Data Bank13
using AutoDock Vina.14,15 As shown in Fig. 2A, compound 11 does not
establish H-bond within the hinge, but is inserted in CLK1 ATP-binding
pocket with 2 hydrogen bonds established between the aminopyr-
imidine part and Asp325 and Lys191 residues. In addition, hydrophobic
interactions involving Leu295, Val324 and Phe241 residues partici-
pated in the binding of 11 to CLK1.
The biological potential of compound 4, particularly its cellular effects,
will be evaluated in more details. Furthermore, we described the first
synthesis of pyrido[4,3-h]quinazolin-2-amine 11. Its kinase inhibitory
potential was inferior to the one of the 10-nitropyrido[3,4-g]quinazo-
lines. However, the study of 11 binding mode within CLK1 binding site
gave valuable information for future structure-activity relationship
studies in this angular series, in order to reach better kinase inhibitory
potencies.
Contrarily to what was observed for the linear pyrido[3,4-g]quina-
zolin-2-amine (Fig. 2B) in which both aminopyrimidine and pyridine
parts were H-bonded to the ATP-binding pocket (hinge Leu244 and
Lys191 residues) the angular structure of 11 does not allow H-bonding
of the pyridine part within CLK1 active site. This could explain the
lower activity of 11 in comparison to best active pyrido[3,4-g]quina-
zolines from the linear series.4,6 These results indicated that the bio-
logical profile of this new series could be improved by modification of
the pyridine part in order to reinforce the interaction with targeted
kinase.
4. Experimental
4.1. Chemistry
4.1.1. General
Starting materials were obtained from commercial suppliers and
used without further purification. Solvents were distilled before use. IR
spectra were recorded on a Shimadzu FTIR-8400S spectrometer (v¯ in
cm−1). NMR spectra, performed on a Bruker AVANCE 400 (1H:
400 MHz, 13C: 100 MHz), are reported in ppm using the solvent residual
peak as an internal standard; the following abbreviations are used:
singlet (s), doublet (d), triplet (t), heptuplet (hept), doublet of doublet
(dd), multiplet (m), broad signal (br s). High resolution mass spectra
(ESI+) were determined on a high-resolution Waters Micro Q-Tof ap-
paratus (UCA-Partner, Université Clermont Auvergne, Clermont-
Ferrand, France). Chromatographic purifications were performed by
column chromatography using 40–63 μm silica gel. Experiments under
3. Conclusion
The synthesis of new 2-aminoalkylamino substituted 10-nitropyrido
[3,4-g]quinazolines was achieved from the corresponding 10-ni-
tropyrido[3,4-g]quinazolin-2-amine 1. Compound 4 was the most ac-
tive of the series with nanomolar activities toward CLK1 which is an
attractive target for the development of new therapies in various
therapeutical areas such as cancer and neurodegenerative disorders.
4

W. Zeinyeh, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
microwave irradiation were performed using
a
CEM Discover
organic layer was dried over MgSO4 and concentrated under reduced
pressure. Residue was purified by preparative TLC using DCM/Acetone
2:3 as eluant, yielding the pure compound 4 (12 mg, 0.040 mmol, 60%)
Benchmate apparatus. Reactions were monitored by TLC using fluor-
escent silica gel plates (60 F254 from Merck). Melting points were
measured on a Stuart SMP3 apparatus and are uncorrected. The purity
of tested compounds 2–5 and 11 was determined by HPLC analysis
using a Hitachi liquid chromatograph (Oven 5310, 30 °C; Pump 5160;
DAD detector 5430) and a C18 Acclaim column (4.6 mm × 250 mm,
5 μm, 120 Å). Detection wavelength was 240 nm (compound 11) or
280 nm (compounds 2–5), and flow rate 0.5 mL/min. Gradient elution
used (A) water/0.1% TFA; (B) acetonitrile: 95:5 A/B for 5 min then 95:5
A/B to 5:95 A/B in 25 min and then 5:95 A/B for 10 min.
as
a yellow powder. TLC Rf = 0.20 (CH2Cl2/MeOH 96:4). mp:
190–191 °C. IR (ATR): 3314–2315, 1630, 1517, 1415, 1335,
1270 cm−1
.
1H NMR (400 MHz, DMSO‑d6): 2.99 (3H, s), 3.11 (3H, s),
4.72 (2H, s), 7.22 (1H, d, J = 5.6 Hz), 7.73 (1H, br s), 7.75 (1H, s), 8.36
(1H, d, J = 6.0 Hz), 8.48 (1H, s), 9.03 (1H, s). 13C NMR (100 MHz,
DMSO‑d6): 34.7, 40.9 (CH3), 42.4 (CH2), 112.1, 125.4, 144.9, 151.0,
158.3 (CHarom), 122.6, 125.5, 128.1, 138.1, 141.8, 161.1 (Carom).
HRMS (ESI+) calcd for
C
14H15N6O2 (M+H)+ 299.1256, found
299.1271. HPLC: purity > 95%, tR = 18.6 min.
4.1.2. N,N-Dimethyl-N′-(10-nitropyrido[3,4-g]quinazolin-2-yl)
formimidamide (2)
4.1.5. N,N-Diisopropyl-N′-(10-nitropyrido[3,4-g]quinazolin-2-yl)
methanediamine (5)
Phosphorus oxychloride (116 µL, 1.244 mmol, 3.0 eq.) was added
dropwise to 3.5 mL of anhydrous DMF at 0 °C under argon. Stirring at
0 °C was continued for 45 min before dropwise addition of a solution of
compound 1 (100 mg, 0.415 mmol) in 6 mL of DMF. After 1 h at 0 °C,
the reaction was hydrolysed by addition of 4 mL of saturated NaHCO3
solution at this temperature. Water was added and the solution was
extracted with DCM. The organic layer was dried over MgSO4 and
concentrated under reduced pressure. Residue was purified by flash
chromatography using DCM/Acetone 1:1 and finally 1:2 as eluant,
yielding the pure compound 2 (120 mg, 0.405 mmol, 97%) as a yellow
powder. TLC Rf = 0.30 (CH2Cl2/acetone 1:1). mp: 184–185 °C. IR
Following the same procedure as described for compound 4, starting
from 25 mg (0.071 mmol) of compound 3 dissolved in 1 mL MeOH
under argon at 0 °C. NaBH4 (6 mg, 0.159 mmol, 2.2 eq.) was added and
the solution stirred at room temperature for 2 h. Additional 6 mg of
NaBH4 were added and stirring continued for 2 h. The solution was
extracted with DCM. The organic layer was dried over MgSO4 and
concentrated under reduced pressure. Residue was purified by column
chromatography using DCM/MeOH 97:3 as eluant, yielding residual
starting material 3 (8 mg, 0.023 mmol). Changing to DCM/MeOH 95:5
afforded pure compound 5 as an oily residue which was subsequently
triturated in cyclohexane to give a yellow powder (13 mg, 0.036 mmol,
51%). TLC Rf = 0.65 (CH2Cl2/MeOH 96:4). mp: 131–132 °C. IR (ATR):
(ATR): 2922, 1624, 1563, 1465, 1324 cm−1
.
1H NMR (400 MHz,
DMSO‑d6): 3.16 (3H, s), 3.26 (3H, s), 7.65 (1H, d, J = 6.4 Hz), 8.63
(1H, d, J = 6.0 Hz), 8.95 (1H, s), 9.17 (1H, d, J = 0.8 Hz), 9.65 (1H, d,
J = 0.8 Hz), 9.78 (1H, s). 13C NMR (100 MHz, DMSO‑d6): 35.2, 41.2
(CH3), 112.5, 134.0, 146.1, 155.3, 160.2, 164.9 (CHarom), 120.6, 124.0,
128.0, 138.1, 141.1, 161.3 (Carom). HRMS (ESI+) calcd for C14H13N6O2
(M+H)+ 297.1100, found 297.1098. HPLC: purity > 95%,
tR = 18.8 min.
3045–2853, 1592, 1561, 1504, 1275, 1108 cm−1
.
1H NMR (400 MHz,
DMSO‑d6): 1.22 (6H, d, J = 6.8 Hz), 1.24 (6H, d, J = 6.8 Hz), 3.79 (1H,
hept, J = 6.8 Hz), 4.62 (1H, hept, J = 6.8 Hz), 4.72 (2H, br s), 7.24
(1H, d, J = 6.0 Hz), 7.75 (2H, br s), 8.36 (1H, d, J = 6.0 Hz), 8.81 (1H,
s,), 9.02 (1H, s). 13C NMR (100 MHz, DMSO‑d6): 19.4, 23.6 (2 × 2CH3),
42.4 (CH2), 46.5 (2CH), 112.2, 125.5, 144.9, 151.0, 156.0 (CHarom),
122.6, 125.7 (2C), 128.1, 141.9, 161.4 (Carom). HRMS (ESI+) calcd for
C
18H23N6O2 (M+H)+ 355.1882, found 355.1859. HPLC: purity >
4.1.3. N,N-Diisopropyl-N′-(10-nitropyrido[3,4-g]quinazolin-2-yl)
formimidamide (3)
90%, tR = 21.6 min.
To a solution of N,N-diisopropylformamide (150 µL, 1.03 mmol,
24.9 eq.) in 1 mL of anhydrous DCM at 0 °C under argon was added
dropwise oxalyl chloride (90 µL, 1.03 mmol, 24.9 eq.). The solution was
allowed to reach room temperature for 30 min, the obtained yellow
4.1.6. 2-Chloro-3-(hydroxymethyl)-6-iodobenzaldehyde (7)
Compound 7 was prepared using a 4-step procedure, similar to the
one described in the literature, without separating the mixture of re-
gioisomers produced step A until step D.4 Only chemical quantities used
and protocol changes are described below.
suspension was cooled again to 0 °C. Compound
1 (10.0 mg,
0.0415 mmol) was added in one portion as a solid. Reaction mixture
was maintained for 20 min at room temperature; an orange solution
was observed. DCM was added and the solution poured on 10 mL of
aqueous K2CO3 solution (10%). The organic layer was washed with
water, dried over MgSO4 and concentrated under reduced pressure.
Residue was purified by flash chromatography using DCM/acetone 9:1
and finally 8:2 as eluant, yielding the pure compound 3 (10 mg,
0.028 mmol, 71%) as an orange powder. TLC Rf = 0.20 (CH2Cl2/
Acetone 8:2). mp: 218–219 °C. IR (ATR): 3027–2767, 1617, 1587,
4.1.6.1. Step A. Compound 6 (5.00 g, 25.05 mmol), anhydrous ethanol
(250 mL), silver sulfate (7.81 g, 25.05 mmol), iodine (6.36 g,
25.05 mmol). 45 min (0 °C) and additional 90 min (room temperature).
Mixture (7.28 g) of ethyl 4-amino-2-chloro-5-iodobenzoate (major)
and ethyl 4-amino-2-chloro-3-iodobenzoate (minor) was directly en-
gaged in the next step.
1561, 1527, 1291 cm−1
.
1H NMR (400 MHz, DMSO‑d6): 1.29 (6H, d,
4.1.6.2. Step B. Mixture from step A (7.28 g), N,N-dimethylformamide
(21 mL), copper(I) cyanide (4.00 g, 44.66 mmol), L-proline (2.57 g,
22.37 mmol). 60 h (80 °C).
J = 7.2 Hz), 1.34 (6H, d, J = 6.8 Hz), 3.86–4.01 (1H, m), 4.84–4.97
(1H, m), 7.66 (1H, d, J = 6.4 Hz), 8.63 (1H, d, J = 6.0 Hz), 9.10 (1H, s),
9.18 (1H, s), 9.66 (1H, s), 9.78 (1H, s). 13C NMR (100 MHz, DMSO‑d6):
19.5, 23.7 (2 × 2 CH3), 47.1, 47.4 (CH), 112.5, 134.1, 146.1, 155.3,
157.6, 164.9 (CHarom), 120.6, 123.1, 128.0, 138.1, 141.1, 164.7
(Carom). HRMS (ESI+) calcd for C18H21N6O2 (M+H)+ 353.1726, found
353.1715. HPLC: purity > 96%, tR = 21.5 min.
Mixture (4.67 g) of ethyl 4-amino-2-chloro-5-cyanobenzoate
(major) and ethyl 4-amino-2-chloro-3-cyanobenzoate (minor) was ob-
tained and directly engaged in the next step.
4.1.6.3. Step C. Mixture from step B (4.67 g), diiodomethane (15 mL),
isoamyl nitrite (5.23 mL, 39.10 mmol). 45 min (room temperature),
80 °C (3 h).
4.1.4. N,N-Dimethyl-N′-(10-nitropyrido[3,4-g]quinazolin-2-yl)
methanediamine (4)
To a solution of compound 2 (20 mg, 0.067 mmol) in 1 mL of an-
hydrous MeOH under argon was added sodium borohydride (5 mg,
0.132 mmol, 2.0 eq.) at room temperature. After 1 h of reaction, addi-
tional 2.5 mg of NaBH4 were added and reaction was left to continue for
30 min. Water was added and the solution was extracted with DCM. The
Residue was purified by column chromatography (EtOAc/cyclo-
hexane 1:99 to 3:97). Combined fractions were triturated in a minimal
amount of pentane to give a mixture (5.58 g) of ethyl 2-chloro-5-cyano-
4-iodobenzoate (major) and ethyl 2-chloro-3-cyano-4-iodobenzoate
(minor).
5

W. Zeinyeh, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
4.1.6.4. Step D. Mixture from step C (5.58 g), toluene (95 mL), 1.2 M
solution of DIBAL-H in toluene (44.4 mL, 53.25 mmol). 1 h 30 min
(−78 °C) and additional 1 h 30 (room temperature). The solution was
extracted with EtOAc, the organic layers were washed with brine, dried
over MgSO4 and concentrated under reduced pressure. The crude
product was purified by successive recrystallizations in CH2Cl2.
4-Chloro-5-hydroxymethyl-2-iodobenzaldehyde (major isomer) was
obtained as a pale yellowish powder (2.377 g, 8.02 mmol, 32% over 4
steps). 2-Chloro-3-(hydroxymethyl)-6-iodobenzaldehyde (minor) 7 was
obtained as a yellow powder (520 mg, 1.75 mmol, 7% over 4 steps)
after filtrate concentration. TLC Rf = 0.15 (EtOAc/cyclohexane 1:4).
mp: 82–83 °C. IR (ATR): 3600–2345, 1699, 1550, 1432, 1336 cm−1. 1H
NMR (400 MHz, DMSO‑d6): 4.56 (2H, d, J = 6.0 Hz), 5.60 (1H, t,
J = 5.6 Hz), 7.46 (1H, d, J = 8.4 Hz), 8.03 (1H, d, J = 8.4 Hz), 10.07
(1H, s). 13C NMR (100 MHz, DMSO‑d6): 59.8 (CH2), 132.3, 139.1
(CHarom), 94.3, 131.9, 133.9, 141.7 (Carom), 193.4 (CO). HRMS (ESI
+) calcd for C8H7O235Cl127I (M+H)+ 296.9174, found 296.9166.
4.1.10. Pyrido[4,3-h]quinazolin-2-amine (11)
A solution of compound 10 (50.3 mg, 0.26 mmol) and guanidine
carbonate (236.3 mg, 1.31 mmol, 5.2 equiv) in N,N-dimethylformamide
(10 mL) was purged with argon for 30 min, sealed and irradiated for
45 s (Discover Mode, Control Type = Standard, P = 300 W,
T = 166 °C). After cooling, the reaction was diluted with EtOAc and
filtered through a pad of Celite. The organic layer was concentrated
under reduced pressure. Residue was purified by flash chromatography
using EtOAc, yielding the compound 11 (15.0 mg, 0.076 mmol, 30%) as
a yellow powder. TLC Rf = 0.22 (EtOAc). mp: degradation. IR (ATR):
3403–2984, 2923, 1652, 1611, 1582, 1474 cm−1
.
1H NMR (400 MHz,
DMSO‑d6): 7.26 (2H, br s), 7.56 (1H, d, J = 8.8 Hz), 7.85 (1H, d,
J = 5.6 Hz), 7.93 (1H, d, J = 8.8 Hz), 8.74 (1H, d, J = 5.6 Hz), 9.13
(1H, s), 10.08 (1H, s). 13C NMR (100 MHz, DMSO‑d6): 120.1, 120.8,
129.2, 147.5, 147.8, 161.4 (CHarom), 116.5, 123.1, 139.4, 151.5, 162.1
(Carom). HRMS (ESI+) calcd for C11H9N4 (M+H)+ 197.0822, found
197.0825. HPLC: purity > 95%, tR = 18.3 min.
4.1.7. 2-Chloro-3-(hydroxymethyl)-6-(2-(trimethysilyl)ethynyl)
benzaldehyde (8)
4.2. In vitro kinase inhibition assays
4.2.1. Method A (33P radioassay)
Under argon, PdCl2(PPh3)2 (94.6 mg, 0.135 mmol, 2%), CuI
(51.3 mg, 0.270 mmol, 4%) and trimethylsilylacetylene (2.37 mL,
11.6 mmol, 2.5 eq) were added to a solution of 7 (2.00 g, 6.74 mmol) in
Et3N (30 mL). After heating at 50 °C for 10 h, evaporation and pur-
ification by column chromatography (EtOAc/cyclohexane 3:7), com-
pound 8 (1.04 g, 3.90 mmol, 58%) was obtained as a beige powder. TLC
Rf = 0.50 (EtOAc/cyclohexane 3:7). mp: 106–108 °C. IR (ATR):
Kinase activities were assayed as previously described.17,18 Only
substrates used for each kinase evaluated are mentioned below.
HsCDK5/p25 (human, recombinant, expressed in bacteria), sub-
strate: histone H1.
SscCK1δ/ε (Sus scrofa domesticus, casein kinase 1δ/ε, affinity pur-
ified from porcine brain), peptide substrate: RRKHAAIGSpAYSITA.
SscGSK-3 α/β (Sus scrofa domesticus, glycogen synthase kinase-3,
affinity purified from porcine brain), GS-1 peptide substrate: YRRAA-
VPPSPSLSRHSSPHQSpEDEEE (“Sp” stands for phosphorylated serine).
RnDYRK1A-kd (Rattus norvegicus, kinase domain aa 1–499, ex-
pressed in bacteria, DNA vector kindly provided by Dr. W. Becker,
Aachen, Germany), peptide substrate: KKISGRLSPIMTEQ.
MmCLK1 (from Mus musculus, recombinant, expressed in bacteria),
peptide substrate: GRSRSRSRSRSR.
3597–3113, 1707, 1585, 1541, 1463, 1248, 1068 cm−1
.
1H NMR
(400 MHz, DMSO‑d6): 0.25 (9H, s), 4.61 (2H, d, J = 5.6 Hz), 5.64 (1H,
t, J = 5.6 Hz), 7.63 (1H, d, J = 8.0 Hz), 7.77 (1H, d, J = 8.0 Hz), 10.47
(1H, s). 13C NMR (100 MHz, DMSO‑d6): 0.4 (3 CH3), 59.9 (CH2), 100.9,
101.6 (Calkyne), 131.3, 132.4 (CHarom), 123.3, 131.6, 133.4, 142.5
(Carom), 189.9 (CO). HRMS (ESI+) calcd for C13H16ClO2Si (M+H)+
267.0603, found 267.0595.
4.1.8. 8-Chloro-7-(hydroxymethyl)isoquinoline (9)
A solution of compound 8 (100 mg, 0.349 mmol) in ammonia in
methanol (7 M, 1 mL) was irradiated for 3 × 5 min (Discover mode,
Control Type Standard, P = 75 W, T = 100 °C). After solvent evapora-
tion and purification by flash chromatography (EtOAc/cyclohexane
7:3), isoquinoline 9 (20.2 mg, 0.104 mmol, 30%) was obtained as a
dark red powder. TLC Rf = 0.48 (EtOAc/cyclohexane 7:3). mp: de-
4.2.2. Method B (ADP-glo)
Kinase assays were carried out at 10 µM ATP in a final volume of
5 µL, following ADP-Glo protocol (Promega). The emitted luminescent
signal measured using an Envision luminometer (PerkinElmer,
Waltham, MA) was expressed in Relative Light Unit (RLU).
gradation. IR (ATR): 3500–2995, 1594, 1546, 1352, 1226, 1079 cm−1
.
1H NMR (400 MHz, DMSO‑d6): 4.81 (2H, d, J = 5.6 Hz), 5.65 (1H, t,
J = 5.6 Hz), 7.92 (1H, dd, J1 = 5.6 Hz, J2 = 1.2 Hz), 7.98 (1H, d,
J = 8.4 Hz), 8.02 (1H, d, J = 8.4 Hz), 8.61 (1H, d, J = 5.6 Hz), 9.58
(1H, s). 13C NMR (100 MHz, DMSO‑d6): 60.3 (CH2), 120.3, 125.8,
130.1, 143.4, 148.1 (CHarom), 124.9, 135.8 (2 Carom), 139.2 (Carom).
HRMS (ESI+) calcd for C10H9ClNO (M+H)+ 194.0367, found
194.0370.
4.3. Molecular modeling experiments
After geometric optimization of compound 11 structure using
Gaussian 09,19 the docking studies were performed with AutoDock
Vina.14,15 Files for the docking were prepared from 6FT8 CLK1 struc-
ture13 after removing of water molecules and 11 pdbqt files prepared
with AutoDockTools (ADT).20 Apolar hydrogen atoms were removed
and Gasteiger charges were added. Docking experiments were per-
formed using the default AutoDock Vina parameters.
4.1.9. 8-Chloroisoquinoline-7-carbaldehyde (10)
A suspension of 9 (359.5 mg, 1.862 mmol) and MnO2 (485.7 mg,
5.587 mmol) in CHCl3 (17 mL) was refluxed for 21 h. After filtration
through a pad of Celite, washing with EtOAc and filtrate evaporation,
the crude product was purified by flash chromatography (EtOAc/cy-
clohexane 1:1) Compound 10 (245.3 mg, 1.283 mmol, 69%) was ob-
tained as a brown powder. TLC Rf = 0.70 (EtOAc/cyclohexane 7:3).
Acknowledgements
Aurélie Job is acknowledged for HPLC analysis. W. Z., F. A., F. G.
and P. M are grateful to the Auvergne Region (Jeune Chercheur
Program) for funding. The French Ministry of Higher Education and
Research is acknowledged for a doctoral fellowship (Y. J. E). Authors
(BJ, BB, SB, SR) thank the Cancéropôle Grand Ouest (axis: natural sea
products in cancer treatment), IBiSA (French Infrastructures in Life
Science) and Biogenouest (Western France Life Science and
Environment Core Facility Network) for supporting KISSf screening
facility of USR3151 in Roscoff.
mp: 135–137 °C. IR (ATR): 1686, 1614, 1544, 1440, 1347, 1224 cm−1
.
1H NMR (400 MHz, DMSO‑d6): 8.03 (1H, dd, J1 = 5.6 Hz, J2 = 1.2 Hz),
8.11 (2H, s), 8.80 (1H, d, J = 5.6 Hz), 9.80 (1H, s), 10.62 (1H, s). 13C
NMR (100 MHz, DMSO‑d6): 120.6, 126.9, 127.8, 146.6, 149.4 (CHarom),
125.0, 130.5, 137.5, 139.2 (Carom), 189.5 (CO). HRMS (ESI+) calcd for
C
10H7ClNO (M+H)+ 192.0211, found 192.0213.
6

W. Zeinyeh, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
References
protection of anilines in metalation reactions. Synlett. 2010;20:3039–3044. https://
doi.org/10.1055/s-0030-1259058.
12. Billimoria AD, Cava MP. Chemistry of indolo[1,2-c]quinazoline: an approach to the
marine alkaloid hinckdentine A. J Org Chem. 1994;59:6777–6782. https://doi.org/
10.1021/jo00101a043.
1. Jain P, Karthikeyan C, Moorthy NS, Waiker DK, Jain AK, Trivedi P. Human CDC2-like
kinase 1 (CLK1): a novel target for Alzheimer’s disease. Curr Drug Targets.
2014;15:539–550. https://doi.org/10.2174/1389450115666140226112321.
2. Elhady AK, Abdel-Halim M, Abadi AH, Engel M. Development of selective Clk1 and -4
inhibitors for cellular depletion of cancer-relevant proteins. J Med Chem.
2017;60:5377–5391. https://doi.org/10.1021/acs.jmedchem.6b01915.
3. Abbassi R, Johns TG, Kassiou M, Munoz L. DYRK1A in neurodegeneration and
cancer: molecular basis and clinical implications. Pharmacol Ther. 2015;151:87–98.
https://doi.org/10.1016/j.pharmthera.2015.03.004.
13. Walter A, Chaikuad A, Helmer R, et al. Molecular structures of cdc2-like kinases in
complex with new inhibitor chemotype. PLoS One. 2018;13:e0196761https://doi.
org/10.1371/journal.pone.0196761.
14. Morris GM, Huey R, Lindstrom W, et al. Autodock4 and autodocktools4: automated
docking with selective receptor flexibility. J Comput Chem. 2009;16:2785–2791.
https://doi.org/10.1002/jcc.21256.
15. Trott O, Olson AJ. Autodock Vina: improving the speed and accuracy of docking with
a new scoring function, efficient optimization and multithreading. J Comput Chem.
2010;31:455–461. https://doi.org/10.1002/jcc.21334.
4. Esvan YJ, Zeinyeh W, Boibessot T, et al. Discovery of pyrido[3,4-g]quinazoline de-
rivatives as CMCG family protein kinase inhibitors: design, synthesis, inhibitory
potency and X-ray co-crystal structure. Eur J Med Chem. 2016;118:170–177. https://
doi.org/10.1016/j.ejmech.2016.04.004.
16. Pettersen EF, Goddard TD, Huang CC, et al. UCSF chimera-a visualization system for
exploratory research and analysis. J Comput Chem. 2004;25:1605–1612. https://doi.
org/10.1002/jcc.20084.
5. Zeinyeh W, Esvan YJ, Nauton L, et al. Synthesis and preliminary in vitro kinase in-
hibition evaluation of new diversely substituted pyrido[3,4-g]quinazoline deriva-
tives. Bioorg Med Chem Lett. 2016;26:4327–4329. https://doi.org/10.1016/j.bmcl.
2016.07.032.
17. Bach S, Knockaert M, Reinhardt J, et al. Roscovitine targets, protein kinases and
pyridoxal kinase. J Biol Chem. 2005;280:31208–31219. https://doi.org/10.1074/jbc.
M500806200.
6. Tazarki H, Zeinyeh W, Esvan YJ, et al. New pyrido[3,4-g]quinazoline derivatives as
CLK1 and DYRK1A inhibitors: synthesis, biological evaluation and binding mode
analysis. Eur J Med Chem. 2019;166:304–317. https://doi.org/10.1016/j.ejmech.
2019.01.052.
18. Winfield HJ, Cahill MM, O’Shea KD, et al. Synthesis and anticancer activity of novel
bisindolylhydrowymaleimide derivatives with potent GSK-3 kinase inhibition. Bioorg
Med Chem. 2018;26:4209–4224. https://doi.org/10.1016/j.bmc.2018.07.012.
19. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Scalmani
G, Barone V, Mennucci B, Petersson GA, Nakatsuji H, Caricato M, Li X, Hratchian HP,
Izmaylov AF, Bloino J, Zheng G, Sonnenberg JL, Hada M, Ehara M, Toyota K, Fukuda
R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Vreven T,
Montgomery Jr JA, Peralta JE, Ogliaro F, Bearpark M, Heyd JJ, Brothers E, Kudin
KN, Staroverov VN, Kobayashi R, Normand J, Raghavachari K, Rendell A, Burant JC,
Iyengar SS, Tomasi J, Cossi M, Rega N, Millam MJ, Klene M, Knox JE, Cross JB,
Bakken V, Adamo C, Jaramillo J, Gomperts R, Stratmann RE, Yazyev O, Austin AJ,
Cammi R, Pomelli C, Ochterski JW, Martin RL, Morokuma K, Zakrzewski VG, Voth
GA, Salvador P, Dannenberg JJ, Dapprich S, Daniels AD, Farkas Ö, Foresman JB,
Ortiz JV, Cioslowski J, Fox DJ. Gaussian 09, Revision D.01. Wallingford CT: Gaussian
Inc.; 2009.
7. Echalier A, Bettayeb K, Ferandin Y, et al. Meriolins (3-(pyrimidin-4-yl)-7-azaindoles):
synthesis, kinase inhibitory activity, cellular effects, and structure of a CDK2/cyclin
A/meriolin complex. J Med Chem. 2008;51:737–751. https://doi.org/10.1021/
jm700940h.
8. Wang Z, Wang C, Sun Y, Zhang N, Liu Z, Liu J. A novel strategy to the synthesis of 4-
anilinoquinazoline derivatives. Tetrahedron. 2014;70:906–913. https://doi.org/10.
1016/j.tet.2013.12.028.
9. Enthaler S, Schröder K, Inoue S, et al. Formamidines-versatile ligands for zinc-cata-
lyzed hydrosilylation and iron-catalyzed epoxidation reactions. Eur J Org Chem.
2010;2010:4893–4901. https://doi.org/10.1002/ejoc.201000648.
10. Meng G, Sha YW, Zhang R, Bai N. Mild synthesis of N′-aryl-N,N-dimethylformami-
dinium chloride by Vilsmeier-Haack reagent. Chin Chem Lett. 2011;22:1043–1046.
https://doi.org/10.1016/j.cclet.2011.03.009.
20. Sanner MF. Python: a programming language for software integration and devel-
opment. J Mol Graph Mod. 1999;17:57–61.
11. Zhichkin PE, Krasutsky SG, Krishnamoorthy R. N, N-Diisopropylformamidine (DIFA)
7