4-Anilino-2-pyridylquinazolines and -pyrimidines as Highly Potent and Nontoxic Inhibitors of Breast Cancer Resistance Protein (ABCG2)

Article abstract of DOI:10.1021/acs.jmedchem.7b00441

Multidrug resistance (MDR) mediated by ATP-binding cassette (ABC) transport proteins remains a major problem in the chemotherapeutic treatment of cancer and might be overcome by inhibition of the transporter. Because of the lack of understanding, the complex mechanisms involved in the transport process, in particular for breast cancer resistance protein (BCRP/ABCG2), there is a persistent need for studies of inhibitors of ABCG2. In this study, we investigated a systematic series of 4-substituted-2-pyridylquinazolines in terms of their inhibitory potency as well as selectivity toward ABCG2. For comparison, the quinazoline scaffold was reduced to the significantly smaller 4-methylpyrimidine basic structure. Furthermore, the cytotoxicity and the ability to reverse MDR was tested with the chemotherapeutic agents SN-38 and mitoxantrone (MX). Interaction of the compounds with ABCG2 was investigated by a colorimetric ATPase assay. Enzyme kinetic studies were carried out with Hoechst 33342 as fluorescent dye and substrate of ABCG2 to elucidate the compounds binding modes.

Full text of DOI:10.1021/acs.jmedchem.7b00441

Subscriber access provided by HKU Libraries
Article
4-Anilino-2-pyridylquinazolines and -pyrimidines as highly potent and
non-toxic Inhibitors of Breast Cancer Resistance Protein (ABCG2)
Michael K Krapf, Jennifer Gallus, and Michael Wiese
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 04 May 2017
Downloaded from http://pubs.acs.org on May 5, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4ꢀAnilinoꢀ2ꢀpyridylquinazolines and ꢀpyrimidines as Highly Potent and Nontoxic
Inhibitors of Breast Cancer Resistance Protein (ABCG2)
Michael K. Krapf, Jennifer Gallus, Michael Wiese*
Pharmaceutical Institute
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
University of Bonn
An der Immenburg 4
53121 Bonn, Germany
Keywords
ABC transporter; inhibitor; ABCG2; quinazolines
ABSTRACT
Multidrug resistance (MDR) mediated by ATPꢀbinding cassette (ABC) transport proteins
remains a major problem in the chemotherapeutic treatment of cancer and might be overcome
by inhibition of the transporter. Due to the lack of understanding the complex mechanisms
involved in the transport process, in particular for breast cancer resistance protein
(BCRP/ABCG2), there is a persistent need for studies of inhibitors of ABCG2.
In this study we investigated a systematic series of 4ꢀsubstitutedꢀ2ꢀpyridylquinazolines in
terms of their inhibitory potency as well as selectivity toward ABCG2. For comparison, the
quinazoline scaffold was reduced to the significantly smaller 4ꢀmethylpyrimidine basic
structure. Furthermore, the cytotoxicity and the ability to reverse MDR was tested with the
chemotherapeutic agents SNꢀ38 and mitoxantrone (MX). Interaction of the compounds with
ABCG2 was investigated by a colorimetric ATPase assay. Enzyme kinetic studies were
carried out with Hoechst 33342 as fluorescent dye and substrate of ABCG2 to elucidate the
compounds binding modes.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Introduction
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ATPꢀbinding cassette (ABC) transporter form a large superfamily of which members are
found in all organisms from bacteria over plants to men. In humans 49 ABC transporters and
were identified and classified according to their phylogenetic similarities in seven subfamilies,
1,2,3
ABC A to ABC G.
While most ABC transporters have well defined substrates, some are
multispecific playing an important role in the protection of the human body against
4
xenobiotics to enhance survival. Naturally these are expressed in organs with barrier function,
as intestine or the blood brain barrier. Unfortunately, they are frequently overexpressed in
some tumor tissues which may lead to resistance against anticancer drugs and failure of a
5
chemotherapeutic treatment due to insufficient drug concentration in the cancer cell.
Regarding dissemination of tumors, chemotherapy remains the most promising therapy
although its prognosis of curing is rather limited due to acquired or intrinsic drug resistance
6
which accounts for roughly 90% of treatment failures. This so called multidrug resistance
(MDR) is often associated with the three major MDR related ABC transport proteins, Pꢀ
glycoprotein (Pꢀgp, ABCB1), multidrug resistanceꢀassociated protein 1 (MRP1, ABCC1) and
7
,8
breast cancer resistance protein (BCRP, ABCG2). Among these, ABCG2 was most recently
discovered and found to be mainly overexpressed in drugꢀresistant solid tumors as well as
9,10
hematopoietic tumors.
Unlike ABCB1 and ABCC1, the breast cancer resistance protein ABCG2 is a halfꢀtransporter
consisting of 655 amino acids with only one cytosolic nucleotideꢀbinding domain (NBD) and
8
one transmembrane domain (TMD). For its function it is assumed to form dimers or
1
1, 12, 13
tetramers as several studies suggest.
Also, the details of the transport process of the
14, 15
drugs remain unclear and there is need for further investigations.
ACS Paragon Plus Environment

Page 3 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Recently, a crystal structure of the human sterol transporter ABCG5/ABCG8 was published
1
6
that could be used for homology modeling of ABCG2. Although the Xꢀray crystal structure
of the protein possessed a relative low resolution of 3.9 Å only it is the first structure with an
inverse organization of the NBD and TMD domains. Shortly afterwards a homology model of
ABCG2 was created and employed to investigate substrate binding, and effects of point
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
7
mutations in the transporter comparing the in silico results to experimental data.
Since its discovery in 1998 by Doyle et al. in MCFꢀ7/AdrVp human breast cell carcinoma
cells, many chemotherapeutics like mitoxantrone (MX), flavopiridol, methotrexate, irinotecan
1
8 ,1
and its active metabolite SNꢀ38 were identified as substrates of ABCG2.
Therefore,
inhibition of the target with potent, selective and nontoxic inhibitors in coꢀadministration
1
9, 20
could be a possible approach to overcome ABCG2 caused MDR.
Recent studies suggest,
that the expression of ABCG2 in colorectal cancer is associated with shortened patient
survival and less response to chemotherapeutic treatment with oxaliplatin and 5ꢀfluorouracil
2
1, 22
(FOLFOX).
Although ABCG2 expression and clinical outcomes sometimes found to be
contradictory, inhibition of ABCG2 remains a promising tool to increase the oral
1
9
bioavailability of chemotherapeutics or to overcome MDR.
Relating to inhibitors of ABCG2, only a few potent compounds active at a submicromolar
2
3
concentration range have been reported, revealing high selectivity and low cytotoxicity. As
one of the first potent inhibitors, fumitremorgin C (FTC) was isolated from Aspergillus
2
4,19
fumigates and found to restore sensitivity toward MX in the resistant S1ꢀM1ꢀ3.2 cell line.
25
Due to neurotoxic effects of FTC, an analog named Ko143 (55) was developed. This second
generation inhibitor exhibited moderate toxicity and a high inhibitory potency as well as good
selectivity toward ABCG2. However, recent studies suggest that selectivity at concentrations
≥ 1 µM is restricted, since effects on the transport activity of ABCB1 and ABCC1 were
2
6
observed. Recently the synthesis of highly potent and nontoxic inhibitors with a phenolic
27
indeno[1,2ꢀb]indole core has been reported. And more recently highly potent and nontoxic
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
inhibitors containing the chromone scaffold with potency comparable to compound 55 have
2
8
been developed by the same working group. Also tyrosine kinase inhibitors like gefitinib
that found clinical application in the treatment of cancer have been shown to be potent
2
9
inhibitors of ABCG2.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Based on the quinazoline scaffold of gefitinib, we designed a focused library of different 4ꢀ
substitutedꢀ2ꢀphenylquinazolines. The studies revealed distinct trends regarding different
substitutions and yielded new compounds, some with higher inhibitory potencies than
30, 31,
compound 55 as well as high selectivity toward ABCG2 and considerable low toxicities.
3
2
In the present study, a pyridyl substituent was introduced in position 2 with the nitrogen in
ortho, meta and para position. Additionally, a systematic variation at the anilino linker in
position 4 was carried out in meta and para position to elucidate the structureꢀactivity
32
relationship (SAR) providing a comparison to the 2ꢀphenyl analogs from our previous study.
Moreover, compounds based on a 4ꢀmethylpyrimidine scaffold were synthesized to
investigate the importance of the condensed aromatic ring of the quinazoline scaffold with
regard to its inhibitory potency toward ABCG2.
On the basis of the new findings, insights in the development of new potent inhibitors based
on both, a quinazoline and a 4ꢀmethylpyrimidine scaffold is provided, yielding new inhibitors
of ABCG2 with promising characteristics for further in vitro and in vivo applications.
Results and Discussion
Chemistry. A brief description of the synthetic route for all compounds is provided in
Scheme 1. Precursors of the quinazoline derivatives were prepared by a cyclic condensation
reaction of anthranilamide with the corresponding pyridine carboxaldehyde to yield the 2ꢀ
substituted quinazolinone derivatives 1ꢀ3. Chlorination of position 4 of the quinazolinone
3
derivative was achieved by refluxing with POCl to obtain the 4ꢀchloro derivatives 4ꢀ6. Final
ACS Paragon Plus Environment

Page 5 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds 11ꢀ41 were synthesized by aromatic nucleophilic substitution of 4ꢀ6 with a meta
and/or para substituted aniline derivative using microwave irradiation.
The 2ꢀsubstituted 6ꢀmethylꢀpyrimidine derivatives 42ꢀ48 were synthesized starting from a
cyclic condensation reaction of methyl 3ꢀoxobutanoate with isonicotinimidamide or
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
benzimidamide to obtain 7 and 8, respectively. Refluxing with POCl yielded the 4ꢀchloro
derivatives 9 and 10. In the last step the chloroꢀsubstituted compounds were reacted with a
meta and/or para substituted aniline derivatives using microwave irradiation to yield 42ꢀ48 by
a nucleophilic aromatic substitution reaction.
Since orthoꢀsubstitution of the 4ꢀanilino linker resulted in poor inhibitory activities in
previous studies, only meta and para substitutions were considered in this work.
1
13
The identity of all compounds was confirmed by H and C NMR spectra and the purity by
elemental analysis. Moreover, the melting points of all test compounds were determined.
Biological Evaluation.
Hoechst 33342 Accumulation Assay. The Hoechst 33342 accumulation assay was used to
determine the inhibitory activity of each test compound against ABCG2 using the MDCK II
BCRP cell line. The fluorescent dye Hoechst 33342 is a substrate of ABCG2 and significantly
increases its fluorescence when bound to DNA or embedded in lipophilic environment like
3
3
the cell membrane. This is utilized to determine the intracellular accumulation of the
substrate that correlates with the degree of inhibition of the transport protein by a compound.
An overview of the inhibitory activities obtained in the Hoechst 33342 accumulation assay
along with the structural features of the compounds is depicted in Table 1. A summary of the
SAR is given in the end of this section providing a quick overview of the most important
results (Figure 2).
At first, the interrelation of the ortho, meta and para substituted pyridyl residue on position 2
1
2
(
R ) with differently substituted 4ꢀanilino residues (R ) was investigated for the quinazoline
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
1
scaffold A: compounds 11, 12 and 13 all contain a 3ꢀnitro substitution at R whereas R is
equipped with an ortho, meta or para substituted pyridyl residue, respectively. Inhibitory
activities against ABCG2 expressed by the IC50 values of 376 nM, 105 nM and 117 nM were
obtained in the given order of the compounds. Interestingly, the ortho pyridyl substitution on
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
R yields approximately a threeꢀfold decreased inhibitory activity in comparison to the meta
and para analogs. Similar observations were made for the next series with a 3ꢀcyano residue
2
at R : again the ortho pyridyl species 15 showed a significantly lower activity (IC : 507 nM)
50
than its meta and para analogs 16 (108 nM) and 17 (134 nM), respectively. For derivatives
2
with a 3ꢀmethoxy group at R the same trend was observed with regard to the pyridyl group at
1
R : compound 19 with the 2ꢀpyridyl substitution was the least potent (1630 nM), while meta
and para pyridyl derivatives yielded lower IC50 values of 1020 nM (20) and 558 nM (21),
1
respectively. This clearly illustrates that an ortho pyridyl residue at R decreased the
inhibitory activity against ABCG2 in every derivative tested so far. Therefore we
concentrated our efforts on compounds with meta and para pyridyl groups.
1
2
Compound 14 with a 3ꢀpyridyl residue at R and a 4ꢀnitro substituent at R yielded the lowest
IC50 value in the test set (64.1 nM) and confirmed the results of previous studies that meta and
2
para substitutions at the anilino moiety (R ) with nitro, cyano, trifluoromethyl and other
3
0, 32
residues is beneficial for high inhibitory activities.
In comparison to the standard inhibitor
55 (IC50: 227 nM in the Hoechst 33342 accumulation assay), currently one of the most potent
inhibitors of ABCG2, compound 14 is roughly fourꢀfold more potent.
2
In most cases a meta substitution at R resulted in higher inhibitory potency than the
corresponding para substituted derivative. The differences varied from slight 3ꢀmethoxy (21,
IC50: 558 nM) and 4ꢀmethoxy (22, IC50: 742 nM) to more than twoꢀfold (compounds 26, IC50:
216nM; 27, IC50: 488 nM). But there were significant exceptions where the para substituted
derivative was more potent. Compare compounds 12 and 14, or 49 and 50. Due to the fact that
further enhancement of the inhibitory activity was often achieved by disubstitution of the
ACS Paragon Plus Environment

Page 7 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
30, 32
anilino moiety R at positions 3 and 4 of a 2ꢀphenylquinazoline scaffold,
several
combinations were investigated.
The 3,4ꢀdimethoxy substituted compound was investigated with different pyridyl residues at
1
R . Again the ortho pyridyl derivative 23 was least potent with an IC50 of 1060 nM, while the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
meta and para derivatives yielded lower IC50 values of 753 nM and 545 nM, respectively.
Disubstitution led to a slight potency increase compared to the monoꢀmethoxy derivatives.
2
The combination of a 3ꢀtrifluoromethyl and 4ꢀmethoxy group at R along with a 4ꢀpyridyl
1
residue at R as present in compound 28 resulted in an IC50 of 146 nM which exceeded the
inhibitory potency of both single substitutions found in compounds 22 and 26. In the
following, the effect of substituents with strong influence on the electron density of the
2
aromatic residue R was investigated using a strong electron withdrawing group and a strong
2
electron donating group. Hence, compound 30 with a 3ꢀfluorosulfonyl residue at R and 4ꢀ
1
2
pyridyl at R was compared with 31, comprising a 3ꢀdimethylamino group at R and 3ꢀpyridyl
1
at R . Both compounds yielded increased IC50 values of 1380 nM and 1200 nM, respectively.
This is in agreement with earlier results confirming that the electron density at the anilino
moiety does not play a major role for the inhibitory activity of a compound. Furthermore a
more bulky ester was synthesized to elucidate some more structure activity relations. Hence, a
1
tertꢀbutyl 3ꢀaminobenzoate residue at position 4 was combined with a 4ꢀpyridyl residue at R
(39). The obtained IC50 of 299 nM is negligibly lower than the value found for the methyl
ester analog 38. Compounds containing an ionized carboxy group possessed the lowest
inhibitory activities in the test set with IC values in the micromolar range.
50
Although the IC50 values of the meta and para substituted 2ꢀpyridylquinazoline derivatives
were so far in good agreement with the IC50 values in the 2ꢀphenylquinazoline series, a 3ꢀ
2
1
fluoro residue at R with a 3ꢀpyridyl moiety at R (29) resulted in a much lower IC50 of 156
nM than its phenyl analog 53 (363 nM). With regard to substitution of phenyl or para pyridyl
1
2
at R , the largest differences were observed for a hydroxy substitutent at R . Compound 32
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
1
with a 3ꢀhydroxy group at R and 4ꢀpyridyl at R yielded an IC50 of 1070 nM and the
corresponding 4ꢀhydroxy derivative 33 yielded an IC50 of only 4260 nM. While for the 2ꢀ
2
phenylquinazoline analog 52 with 4ꢀhydroxy at R an excellent IC50 of 211 nM had been
2
found. Furthermore, a 3ꢀhydroxymethyl group at R was either combined with 3ꢀpyridyl (34:
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
8
10 nM) or 4ꢀpyridyl (35: 921 nM) at R . The alteration from 3ꢀhydroxy to 3ꢀhydroxymethyl
had only slight effects on the inhibitory activity. Again, no significant distinction between
meta and para pyridyl was observed regarding the inhibitory activity. Following, a
2
1
disubstitution of 3ꢀhydroxy and 4ꢀnitro at R was investigated with a 3ꢀpyridyl moiety on R .
The resulting IC50 value of 245 nM for compound 36 was about threeꢀfold higher than for the
2ꢀphenylquinazoline analog 51 (IC50: 81.1 nM). This is probably due to unfavourable effects
1
of the 3ꢀpyridyl group at R which results in lower inhibitory activities when combined with
2
meta or para hydroxy residues at R as observed in case of compounds 32 and 33.
In the next series, the condensed aromatic ring next to the heterocycle of the quinazoline
scaffold was replaced by a methyl group (Table 1, scaffold B) to evaluate the impact of the
aromatic moiety on the inhibitory activity. For better comparison a 4ꢀpyridyl as well as a
1
phenyl residue was investigated at R to give insights in the SAR of these small molecules.
1
2
First, a 4ꢀpyridyl residue at R and a 3ꢀcyano group at R were introduced in compound 42.
The resulting IC50 of 132 nM is in excellent agreement with its quinazoline analog 17 (IC50
:
1
134 nM). Furthermore, compounds 43 and 44 with a 4ꢀpyridyl residue at R together with a 3ꢀ
2
methoxy or 4ꢀmethoxy group at R yielded IC50 values of 271 nM and 874 nM. In accordance
to the meta and para methoxy substituted quinazoline analogs 21 (IC : 558 nM) and 22 (IC :
50
50
7
42 nM) the meta substitution resulted in higher inhibitory potency. Moreover, some 4ꢀ
methylpyrimidines bearing a 2ꢀphenyl group were investigated. Two derivatives with meta or
2
para cyano substituents at R yielded IC50 values of 122 nM (45) and 128 nM (46),
respectively. Both values are in good agreement with their 2ꢀphenylquinazoline analogs with
2
a 3ꢀcyano (49: 140 nM) or a 4ꢀcyano (50: 71.4 nM) residue at R . Also, the disubstitution with
ACS Paragon Plus Environment

Page 9 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
3ꢀnitro and 4ꢀhydroxy at R exhibited only negligible differences between the 2ꢀphenyl
derivatives of scaffold B (47: 98.8 nM) and scaffold A (51: 81.1 nM). For the basic structures
1
2
of scaffold A and scaffold B with R = phenyl and R = H, very similar IC50 values of 648 nM
48) and 882 nM (54) were found. Concentrationꢀresponse curves of the two most potent
compounds 14 and 47 in comparison to the standard inhibitor 55 are depicted in Figure 1.
(
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
From this study it was found that the substitution pattern of the pyridyl residue at R of
scaffold A has a major impact on the inhibitory activity: substitution with ortho pyridyl
yielded less potent compounds while meta and para substitutions both resulted in increased
2
activities. Regarding the substituents at R , nitro, cyano, trifluoromethyl or fluoro substituents
in meta and para yielded very potent inhibitors. No clear preference was noticeable for meta
1
and para substitution of compounds containing a pyridyl or phenyl moiety at R . In contrast
to earlier studies, hydroxyl residues turned out to decrease the inhibitory activity when
1
32
combined with pyridyl at R instead of phenyl. Apart from that, very similar potencies were
obtained with both moieties. Again, the electron density of the aromatic ring influenced by
2
substituents at R exhibits no clear correlation with the IC50 values. Introducing esters in meta
position also resulted in high activities independent of the bulkiness of the alkyl residue.
According to calculation, the pyridyl residue decreases the logP in comparison to a phenyl
substitution by about 1.18 units to an average of 3.00. No correlation of the calculated logP
values with the pIC50 values was observed.
Interestingly, the potency of the 4ꢀmethylpyrimidine analogs, lacking the condensed aromatic
ring of the quinazoline scaffold, was similar to the quinazoline based derivatives. Earlier
1
studies have emphasized the need of aromatic substitution at R as well as the importance of
30, 32
the nitrogen in position 3 of the heterocycle for potent compounds.
Simplification of the
quinazoline scaffold to a 4ꢀmethylpyrimidine heterocycle leads to an extraordinary high
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ligand efficacy, reducing relevant components for a potent inhibitor to three aromatic rings
substituted with few small residues.
Screening for ABCB1 and ABCC1 Inhibition. The selectivity of selected compounds
toward ABCG2 was investigated by the calcein AM assay. For this purpose, the ABCC1
overexpressing cell line H69AR and the ABCB1 overexpressing cell line A2780adr were used
as described in the experimental section. A screening of the inhibitory effect of the substances
toward the polyspecific transporters ABCB1 and ABCC1 was carried out at a concentration
of 10 µM. Cyclosporine A (CsA) was used as positive control, leading to total inhibition of
both transporters. Compounds which exhibit an inhibitory potency of more than 25% relative
to the standard were additionally tested with more dilutions to generate a concentrationꢀ
response curve to calculate an IC50 value. The screening results for the inhibitory activity
toward ABCB1 and ABCC1 are depicted in the bar charts Figure 3 (a) and (b), respectively.
A summary of the IC50 values of the compounds showing more than 25% of inhibition in the
screening is given in Table 2.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In the calcein AM assay with ABCB1 overexpressing cells, the following compounds showed
a relative inhibitory activity of more than 25% in relation to the positive control CsA. Bearing
2
a 3ꢀmethoxy residue at R , compounds 19, 20 and 21 show relatively high inhibitory activity
toward ABCB1. They were synthesized in the given order with a ortho, meta and para
1
pyridyl moiety at R and have IC50 values of 12.7 µM, 4.78 µM and 5.43 µM, respectively.
As expected, the inhibitory activity toward ABCB1 is increased for the 3,4ꢀdimethoxy
3
2
derivatives, since methoxy groups are favourable for ABCB1 inhibition. The ortho, meta
1
and para pyridyl (R ) substituted compounds 23, 24 and 25 show inhibitory activities of 6.30
µM, 3.11 µM and 2.08 µM, respectively. As observed for inhibition of ABCG2, the
substitution pattern of the pyridyl residue is also crucial for the inhibitory potency against
ABCB1. Substitution with ortho pyridyl decreases the activity in comparison to the meta and
ACS Paragon Plus Environment

Page 11 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
para derivatives. In comparison to the monoꢀsubstituted compound 22, 28 revealed a good
1
IC value of 2.78 µM with 4ꢀpyridyl at R and a disubstitution with 3ꢀtrifluoromethyl and 4ꢀ
5
0
2
1
methoxy at R . Furthermore, a 4ꢀpyridyl residue at R combined with a methyl 3ꢀ
aminobenzoate moiety (38: IC50 = 6.37 µM) or a tertꢀbutyl 3ꢀaminobenzoate residue (39: IC50
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
=
0.334 µM) at position 4, led to noticeable inhibition of ABCB1. In particular, the tertꢀbutyl
ester obtained a considerable potency superior to the standard CsA (IC50: 1.21 µM). A
concentration response curve of 39 and CsA is provided in Figure 4. Then, a substitution of 4ꢀ
1
pyridyl at R each with a 3ꢀaminobenzoic acid (40) and a 4ꢀaminobenzoic acid moiety (41:
IC50 = 3.67 µM) was carried out. The meta substituted compound exhibited no activity
whereas the para substituted one possessed a considerable potency.
All of the investigated nitro and cyano substituted compounds 11ꢀ18 showed high activities
toward ABCG2 and low activities toward ABCB1. Compounds bearing at least one methoxy
group generally showed an increased inhibitory activity toward ABCB1. Adding more
methoxy groups to the molecule resulted in a further increased activity. Additionally, a
1
correlation of the substitution pattern of the pyridyl residue at R was observed, showing
increased activities from ortho to meta to para. Some ester moieties also exhibit significant
potencies in the screening: compound 39 with a tertꢀbutyl ester had the lowest IC value of
50
0.334 µM.
To investigate the inhibitory effect of the compounds toward ABCC1, they were screened in a
calcein AM assay with the ABCC1 overexpressing cell line H69 AR. The results of the
screening are depicted in Figure 3 (b) as a bar chart. Only compounds 22 and 47 slightly
exceeded the 25% range of inhibitory activity relative to the positive control CsA (IC : 2.97
50
µM) at a concentration of 10 µM. It was found in earlier studies that compounds with a
quinazoline scaffold, like the substituted 4ꢀanilinoꢀ2ꢀphenylquinazolines, exhibit no inhibitory
32
activity toward ABCC1 unless they form charged species under physiological conditions.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Compound 22 (IC50: 15.9 µM) was the only one from the quinazoline series that slightly
exceeded the 25% mark.
For the substances with a 4ꢀmethyl pyrimidine structure (scaffold B), only a negligible
activity was found. Mere compound 47 (IC50: 17.8 µM) with a disubstitution of 3ꢀnitro and 4ꢀ
hydroxy at the aniline moiety showed barely more than 25% of inhibition compared to the
positive control CsA. Regarding quinazoline derivatives as inhibitors of ABCC1, this study is
consistent with earlier findings, that inhibitory activity toward ABCC1 plays only a negligible
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
30, 32
role.
Determination of the Intrinsic Cytotoxicity via MTT Assay. The intrinsic cytotoxicity of
selected compounds was determined in the MTT assay. The results for selected compounds
are depicted in Table 3 and given as cell growth inhibition leading to a cell survival of 50% of
the control (GI50). Small amounts of methanol (1.8%) and DMSO (1%) from the stock
solutions of the compounds were present at the highest concentration of 100 µM leading to
apparent slight toxicity of otherwise nontoxic compounds. Therefore, additionally, the toxic
effect of the used dilution without modulator was determined at GI50 = 96.1 µM (MDCK II
ABCG2 overexpressing cells). For comparison, the GI50 of the established inhibitor of
ABCG2 compound 55 (12.6 µM) was obtained in analogy to the other test substances.
Four potent compounds (14, 16, 17 and 47) with low intrinsic cytotoxicity were additionally
investigated in a modified MTT viability assay without DMSO and with a constant amount of
1% methanol. This modification was possible due to the good water solubility of the
compounds. Either no effect or stimulation of the cell proliferation was observed at low
concentrations which is probably explained by a defensive cellular mechanism. Further
increase of the concentration eventually reversed this effect leading to a decreasing cell
viability.
ACS Paragon Plus Environment

Page 13 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
All tested compounds showed no significant cytotoxic effects up to 100 µM. Due to restricted
solubility, higher concentrations could not be used in order to create sigmoidal doseꢀviability
curves obtaining a GI50 value. Thus, the GI50 of the nontoxic compounds is very likely higher
than 96 µM, resulting from the cytotoxic effects of the solvents only.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The following section discusses GI50 values determined with the MTT assay using MDCK II
ABCG2 overexpressing cells.
2
For the 3ꢀnitro (12) and 4ꢀnitro (14) substituted compounds (R ), both with a 3ꢀpyridyl moiety
1
at R , GI50 values of 70.8 µM and 73.3 µM were calculated, respectively, revealing almost no
cytotoxic effect caused by the compounds. Moreover, a compound containing 3ꢀcyano residue
2
1
at R was investigated with a meta pyridyl (16) and para pyridyl (17) residue at R . Here,
both compounds show very low cytotoxicity of 75.3 µM and 88.1 µM, respectively.
1
2
Furthermore, the 3ꢀpyridyl (R ), 4ꢀcyano (R ) derivative 18 was investigated. Again, only
negligible cytotoxicity was found giving a GI50 of 79.4 µM. By the first results some trends
2
among the nitro and cyano substituents at R combined with meta and para pyridyl moieties
1
at R can be perceived: substitution with cyano resulted in slightly decreased toxic effects in
comparison to nitro, although both yielded considerably high GI50 values. Moreover,
substitution with meta pyridyl seems to be beneficial for low cytotoxicity over a para
substitution based on compounds 16 and 17. Although the differences among the pyridyl
derivatives are rather small, the advantage of the pyridyl over a phenyl moiety becomes
clearly evident by comparison of compounds 18 and 49. Both contain a 4ꢀcyano substitution
2
1
at R along with 3ꢀpyridyl or a phenyl residue at R , respectively. The phenyl derivative 49
has approximately an 8ꢀfold lower GI (10.4 µM) than the 3ꢀpyridyl derivative 18 (GI : 79.4
50
50
µM). This observation was also made for several more compounds presented below. Another
1
effect was noticed regarding the substitution in meta and para position: the 4ꢀpyridyl (R )
derivatives substituted with trifluoromethyl groups in meta (26) or para (27) position resulted
in GI50 values of 12.6 µM and 117 µM what reflects the great impact of this modification
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(meta vs. para), which is also noticeable by the significant difference of the compounds IC50
values. Similar results were also found in earlier studies with the 2ꢀphenyl analogs regarding
32
the IC50 values and particularly the GI50 values. A possible explanation for this observation
was suggested by studies of the enzyme kinetics and ATPase assays described in the
corresponding chapters. The lowest GI50 value (2.88 µM) in the test set was obtained for
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
compound 28 with a 4ꢀpyridyl residue at R and a disubstitution of 3ꢀfluoromethyl and 4ꢀ
methoxy at the aniline moiety in position 4. In comparison to compound 26, the additional
1
methoxy group lowers the GI50 about fourꢀfold. Compound 29 with a 3ꢀpyridyl moiety at R
2
and 3ꢀfluoro at R has a GI50 of 46.8 µM. The comparison to its 2ꢀphenyl analog 53 (GI50
:
18.0 µM) once more demonstrates the beneficial effect of the pyridyl moieties for cytotoxicity.
1
2
Compound 31 with a 3ꢀpyridyl moiety (R ) and a 3ꢀdimethylamino group (R ) shows an
1
increased cytotoxicity (GI : 10.0 µM). Furthermore, a 4ꢀpyridyl substitution at R was
5
0
2
combined with 3ꢀhydroxy (32) and 4ꢀhydroxy residue (33) at R , respectively. The meta
substitution was found to be considerably more cytotoxic (GI50: 9.22 µM) than the para
derivative (GI50: 55.0 µM). Similar tendencies regarding IC50 and GI50 values as found for
compounds 26 and 27 apply for the meta and para hydroxy derivatives. As expected, the 2ꢀ
phenyl analog 52 exhibits a significantly higher cytotoxicity than compound 33 with a pyridyl
moiety.
Additionally, some derivatives with the pyrimidine scaffold B were investigated. First,
1
2
compound 42 containing a 4ꢀpyridyl substituent at R and a 3ꢀcyano group at R . The
determined GI of 6.92 µM discloses a considerably higher cytotoxicity than its quinazoline
50
1
analog 17 (GI : 88.1 µM). Moreover, compound 46 with a phenyl substitution at R along
50
2
with a 4ꢀcyano group at R was investigated yielding a GI50 value of 17.8 µM which is
slightly less toxic than the corresponding quinazoline analog 49 (GI50: 10.4 µM). Further,
1
2
compound 47 with a phenyl residue at R and a disubstitution of 3ꢀnitro and 4ꢀhydroxy at R
had a GI of 49.0 µM which is identical to the quinazoline analog 51 (GI : 52.4). Due to the
5
0
50
ACS Paragon Plus Environment

Page 15 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
low cytotoxicity along with the observed high inhibitory activity, high therapeutic ratios were
obtained for such combinations (see Table 3). Finally, the basic scaffold B bearing phenyl at
1
2
R and hydrogen at R was investigated: the calculated GI50 of 107 µM for compound 48 was
significantly higher than for its analog derivative 54 based on scaffold A (GI50 = 26.6 µM).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
From the data it can be concluded, that meta and para pyridyl residues (R ) on a quinazoline
scaffold A reduce the cytotoxicity in comparison to a phenyl moiety considerably. For the
2
meta and para substitutions at R distinct differences were observed concerning the GI50 as
well as IC50 values. In general the 2ꢀphenyl derivatives based on a 4ꢀmethylpyrimidine
scaffold B exhibit a lower cytotoxicity in comparison with their quinazoline analogs based on
scaffold A. However, scaffold B was less beneficial in comparison to scaffold A with a 4ꢀ
1
2
pyridyl residue at R and 3ꢀcyano at R (see 17 and 42). Due to the limited amount of
investigated compounds containing scaffold B in the MTT toxicity assay, further compounds
have to be synthesized and investigated to confirm these findings.
Determination of the Ability to Reverse MDR via MTT Assay. A MDR reversal assay was
performed for some of the most active compounds (14, 16 and 47). The assay was carried out
with MDCK II ABCG2 overexpressing cells. As a cytostatic drug, SNꢀ38, which is a wellꢀ
known substrate of ABCG2, was used. Different dilutions of the cytostatic were combined
with different modulator concentrations. Intracellular accumulation of the cytostatic is
induced by the inhibitory effect of the modulator toward ABCG2, which restrains the efflux
of SNꢀ38 out of the cell and leads to a decreased cell viability. As a positive control the
MDCK II parental cells were used to represent total inhibition due to the lack of ABCG2
transport proteins. The evaluation of the assay was carried out analogously to the MTT
viability assay described above.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
For this investigation, the most potent 2ꢀpyridylquinazoline derivatives containing a nitro (14)
and a cyano (16) group were selected. Additionally, compound 47 was chosen from the
derivatives based on the 4ꢀmethylpyrimidine scaffold B due to its highest inhibitory activity
among them. All compounds displayed the ability to fully reverse the MDR of the ABCG2
overexpressing cells, as shown in Figure 7 (a, c and e). Reversal of the resistance is illustrated
by the shift of the IC50 values of the ABCG2 overexpressing cells toward the parental cells
with increasing modulator concentration (grey arrow). At a modulator concentration of ≥ 1
µM full sensitization of the ABCG2 overexpressing cells was achieved, indicated by a similar
IC50 value as the parental cells. Additionally, the MTT assay was repeated using parental
MDCK II cells to elucidate possible interactions of the modulator and SNꢀ38 which could
affect the IC50 values obtained from the concentrationꢀresponse curves. The obtained
concentrationꢀresponse curves are shown in Figure 7 (b, d and f). Clearly, there is no effect of
the compounds on the wild type cells. Furthermore, a quantification of a compounds efficacy
in reversing MDR was calculated by plotting each GI50 value from the reversal assay against
the logarithm of the compound concentration and calculating an IC50 value: the state “no
inhibition” is defined with ABCG2 overexpressing cells without modulator, which show
maximal resistance toward SNꢀ38, and “total inhibition” of ABCG2 is defined by the pIC ꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
50
value of the parental cells. Corresponding concentrationꢀresponse curves were fitted with the
logistic equation yielding pEC50 values that can be compared with the inhibitory activities
from the Hoechst 33342 accumulation assay (Figure 8).
1
2
Compound 14 with a 3ꢀpyridyl residue at R and a 4ꢀnitro group at R (scaffold A) yielded an
EC of 21.2 nM determined in the MDR reversal assay. This suggests an about 3ꢀfold higher
50
inhibitory potency for inhibiting the transport of SNꢀ38 than found in the Hoechst 33342
accumulation assay (64.1 nM). As expected, the compound obtained a sigmoidal curve with
the ABCG2 overexpressing cells (●) and reaches its top on a similar level to the parental cells
(○) with increasing modulator concentration (Figure 8 a). Moreover, the IC50 values of the
ACS Paragon Plus Environment

Page 17 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
parental cells stay the same throughout the different modulator concentrations. Following, for
1
2
compound 16 with a 3ꢀpyridyl moiety at R and 3ꢀcyano at R (scaffold A) an IC value of 85
50
nM was calculated with this method (Figure 8 b). This is again slightly less than the IC50
value of 108 nM found in the Hoechst 33342 accumulation assay. Also compound 47 with a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2
phenyl moiety at R and a disubstitution of 3ꢀnitro and 4ꢀhydroxy at R (scaffold B) exhibits a
good correlation of the IC50 value derived from the MDR reversal assay (62.3 nM) and the
Hoechst 33342 accumulation assay (98.8 nM). The differences in the IC50 values can be
explained by different affinities of Hoechst 33342 and SNꢀ38 for ABCG2, suggesting a higher
affinity of Hoechst 33342. The concentrationꢀresponse curves are depicted in Figure 8 c.
Additionally, the antineoplastic drug mitoxantrone (MX) was used to demonstrate the reversal
of resistance of the ABCG2 overexpressing MDCK II cells with compounds 14 and 47. For
this purpose, the cell viability was measured as described above after 72 h incubation with
different modulator concentrations in the presence of 5 µM cytostatic. Decrease of cell
viability caused by the cytotoxicity of the compound was taken into account by performing a
dilution of the substance without MX. Wells with only cells and medium in presence and
absence of MX were used as negative control. For both compounds, the reversal of resistance
was demonstrated as shown in Figure 9. The halfꢀmaximal growth inhibition was achieved for
compound 14 at 12.6 nM and for 47 at 22.0 nM, respectively. The same tendencies in
reversing MDR as observed for SNꢀ38 could be reproduced using MX as cytostatic as well as
confirming the ability of the compounds to effectively reverse ABCG2 mediated MDR.
Investigation of ATPase Activity. ATPase measurements were carried out using High Five
insect cell membranes, which were infected with baculovirus containing ABCG2 cDNA.
Therefore, selected compounds based on scaffold A (4ꢀsubstitutedꢀ2ꢀphenylquinazolines)
with high inhibitory potencies were investigated together with compounds comprising
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
scaffold B (4ꢀmethylpyrimidine). The determined EC50 values and the maximal stimulation
caused by the compounds including quercetin as the standard activator are summarized in
table 4. Three compounds (13, 14, 17) exhibited about the same activating effect as the
standard quercetin. While most substances produced only low activation of less than 40%,
having only a low effect on the ATPase activity. One compound (48) was inactive producing
a stimulation of less than 10%. To confirm this unexpected result, the measurement was
repeated six times. Compound 39 was the only compound in the data set exhibiting an
inhibitory ATPase activity down to 40% of control. Interestingly there are two compounds,
namely compound 26 and 27, which demonstrate an increase of activation at lower
concentrations on the one hand and a decrease of activation at higher concentrations on the
other hand. About oneꢀthird of all investigated compounds showed an activation level
between 140% to 160%.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Inspection of the EC50 values shows that in case of the methylpyrimidines with scaffold B a
large difference in magnitude exists depending on the substituent at position two. Derivatives
containing a pyridyl group (42, 43, 44) have EC50 values in the low nanomolar range, like the
corresponding quinazolines with scaffold A (17, 21, 22). However, all investigated
compounds with scaffold B and a 2ꢀphenyl substituent (45, 46, 47) were much less potent
with EC50 values in the micromolar range. The ATPase activation is only noticeable at
concentrations far higher than the IC50 for ABCG2 inhibition. Therefore we investigated
additionally some quinazolines from our previous publication containing a 2ꢀphenyl
substituent for their effect on ATPase activity (49, 51, 53). Interestingly for the quinazolines
this difference between a 2ꢀphenyl substituent and a pyridyl group in position 2, in activating
efficacy is not apparent. Both groups of compounds are potent activators of ATPase activity
with comparable EC50 values. But like most of the quinazolines (scaffold A) with a pyridyl
substituent, the investigated 2ꢀphenyl derivatives were found to exert only low stimulating
effect of less than 50%.
ACS Paragon Plus Environment

Page 19 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
While comparing the substitution pattern of scaffold A in relation to the position of the
nitrogen atom in the pyridyl group, it becomes obvious that compounds comprising a 2ꢀ
pyridyl moiety are less potent in ABCG2 inhibition assay as well as less effective in ATPase
activation than their corresponding 3ꢀpyridyl and 4ꢀpyridyl analogs (11, 12, 13 and 15, 16, 17).
Representative concentrationꢀresponse curves of compounds 14, 26, 39 and quercetin are
illustrated in Figure 10. Compound 14, being the most active with the lowest IC50 value in the
Hoechst 33342 accumulation assay (64 nM), exhibited solely activation of ATPase activity
over the whole range of concentrations similar to quercetin. Compound 26, one of the two
derivatives showing a bellꢀshaped curve, which is typical for substrates having different
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
4, 35, 36
affinities for two distinct binding sites.
It is assumed that the transporter has two
distinct binding sites for substrates, where the high affinity site is activating/transporting and
the other one is inhibiting and has a low affinity for substrates. If both affinities are
sufficiently different, a bellꢀshaped curve results. However, if the affinity to the activating
binding site is several orders of magnitude higher than for the inhibitory site, a Michaelisꢀ
Menten type curve will be observed as only the activating site is targeted. The maximum of
ATPase activity is reached at a concentration of about 0.5 µM with an activation of 55%.
Further increase of concentration results in a decrease of ATPase activity going down to 25%
at 35 µM. Fitting the data with a combination of an increasing and a decreasing logistic
M i
equation with a Hill coefficient fixed to one, an activating K of 9 nM and an inactivating K
of 28000 nM was calculated. Although stimulation of ATPase activity is often considered as
being typical for substrates, several direct and indirect evidence exists, that challenges this
simple view. In the case of high ATPase activity stimulation, recently a fluorescent
heterodimeric ABCG2 inhibitor comprised of a quinazoline and chalcone moiety was shown
to accumulate in transfected MDCK II BCRP cells to the same extent as in parental cells not
3
7
expressing ABCG2 using confocal fluorescence microscopy. Further indirect evidence that
the compounds are no substrates comes from the MTT data of their intrinsic cytotoxicity.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Here no differences between the GI50 values of the ABCG2 expressing and the parental cell
line were observed, pointing to no or insufficient transport of the compounds. The same
identical toxicity was observed in case of the quinazolineꢀchalcones. For further
investigations of the binding mode of these compounds enzyme kinetic studies with the
substrate Hoechst 33342 were undertaken.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Investigation of Inhibitory Enzyme Kinetics. Investigation of the mode of interaction was
undertaken for compounds 14, 21 and 26, which were selected based on their different
ATPase activity profiles. Studies were carried out with ABCG2 overexpressing MDCK II
cells and Hoechst 33342. The LineweaverꢀBurk linearization of the data for compound 14
proposes a nonꢀcompetitive interaction with Hoechst 33342, as indicated by the intersection
of the lines at the Xꢀaxis (Figure 11 a). Although linearization according to LineweaverꢀBurk
gives direct information on the type of interaction, this method has been criticized as being
notoriously susceptible to experimental errors, due to the double reciprocal plot. Therefore we
additionally analyzed the data using the method according to HanesꢀWoolf. The halfꢀ
reciprocal method according to HanesꢀWoolf is claimed to be more robust to experimental
errors. Compound 14 yielded an intersection of all lines at the Xꢀaxis, meaning that KM
remained constant but Vmax decreased with increasing substrate concentrations that is
characteristic for nonꢀcompetitive inhibitors (Figure 11 d). Thus, both methods indicate a nonꢀ
competitive interaction with Hoechst 33342 suggesting a different binding mode and/or
binding pocket for compound 14.
For compound 21, on the other hand, a competitive interaction with Hoechst 33342 was
derived from the LineweaverꢀBurk linearization visualized by the intersection of the lines at
the Yꢀaxis (Figure 11 b). This result is substantiated by the results of the analysis according to
M
HanesꢀWoolf (Figure 11 e). Here an increase of K with higher inhibition concentrations is
noticeable while Vmax remains constant. According to both methods a competitive interaction
ACS Paragon Plus Environment

Page 21 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
is proposed for compound 21, suggesting the same binding mode and/or pocket as Hoechst
33342. Furthermore, the data of compound 26 were analyzed with the LineweaverꢀBurk
linearization obtaining straight lines with an intersection in the second quadrant of the
coordinate system (Figure 11 c). This phenomenon is observed for “mixedꢀtype” inhibitors
with both, competitive and nonꢀcompetitive characteristics. For instance, conformational
change of the protein epitope upon binding of the “mixedꢀtype” inhibitor can still affect the
active site. Such inhibitors bind differently from the substrate but still influence substrate
binding and turnover rate.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Applying the HanesꢀWoolf method equivalent results were obtained, as indicated by
M
decreasing Vmax values und increasing K values with higher substrate concentrations in
agreement with a nonꢀcompetitive interaction of the “mixed type” (Figure 11 f).
As a third method the direct linear plot according to CornishꢀBowden was used to analyze the
data. This method has been claimed to be insensitive to outliers. For compound 14 analysis of
the data according to CornishꢀBowden reveals a decline of the maximal velocity Vmax with
increasing compound concentration while the MichaelisꢀMenten constant K
same, indicative of a nonꢀcompetitive interaction with Hoechst 33342 (Figure 11 g). In
compound 21 the plot of K (increasing) and V (constant) calculated from the analysis
M
remains the
M
max
according to CornishꢀBowden (Fig. 11 h) nicely confirms previous results, exhibiting a
competitive character. In Figure 11 i) a decreasing K and a slightly increasing Vmax with
M
increasing compound concentration is seen for compound 26 indicative of a mixed type of
inhibition.
Conclusions
The purpose of this study was to extent the database of inhibitors based on a quinazoline
scaffold. Notably, the comparison between 4ꢀsubstituted 2ꢀphenylquinazolines of our
previous study and 4ꢀsubstituted 2ꢀpyridylquinazolines and 4ꢀmethylpyrimidine analogs with
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
2
lower molecular weight was targeted. The investigation of the inhibitory activities against
ABCG2 showed a distinct correlation between the position of the nitrogen atom in the pyridyl
residue and the inhibitory potency: a nitrogen atom in ortho position resulted in less potent
compounds while meta and para pyridyl substituents were found to be beneficial. Regarding
the substitution at the aniline linker in position 4, nitro, cyano, trifluoromethyl and fluoro
substituents yielded considerably high inhibitory activities toward ABCG2. The most potent
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
compound in the test set contained a 3ꢀpyridyl substitution at R and a para nitroꢀgroup at R
14) with an IC50 value of 64.1 nM, which is 3.5ꢀfold lower than for compound 55. Among
(
the 4ꢀmethylpyrimidine derivatives based on scaffold B, several compounds yielded IC50
values between 100 and 150 nM. This demonstrates the extraordinary high ligand efficacy of
this scaffold. For both scaffolds A and B, significant conformity regarding the IC50 value was
observed between derivatives bearing pyridyl and phenyl, respectively. However, substitution
2
with hydroxy on R in meta and para was found to be significantly more favourable using a
1
phenyl substitution at R . In general, the determined differences between the potencies of a
meta or para substitution with the same residue is mentionable. Another advantage of the
pyridyl derivatives compared to their phenyl analogs is the considerably lower calculated logP
value that could be beneficial for bioavailability. For the whole test set however, no
correlation between logP and pIC50 was found. Although sufficient lipophilicity is needed for
membrane permeability, several compounds exhibit similar logP values but different
inhibitory potencies. Hence, other determinants are probably influencing the potency of a
compound to a higher degree. Furthermore, substitution with pyridyl yielded considerably
lower cytotoxicities than with phenyl, resulting in several compounds with extraordinary high
therapeutic ratios. Like for the IC50, the GI50 of the compounds was often significantly
2
influenced by the choice of substitution at R , as well as by the position in meta or para.
Regarding selectivity toward ABCG2, only compounds with one or more methoxy groups
showed activity toward ABCB1, while none of the selected compounds obtained a significant
ACS Paragon Plus Environment

Page 23 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
activity toward ABCC1. The ability to reverse MDR was confirmed for the most potent
compounds carrying out MDR reversal assays with the anticancer drugs SNꢀ38 and MX. The
IC50 values derived from the MTT assay often exceeded the inhibitory potency found in the
Hoechst 33342 accumulation assays suggesting a higher affinity for Hoechst 33342 toward
ABCG2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Investigation of the effect on ATPase activity for selected compounds revealed four different
types of ATPase activity curves. A small group of compounds showed solely an activating
curve with high stimulating effect up to 85% in comparison to the basal activity. This group
of compounds is comparable to quercetin, which is a wellꢀknown substrate of ABCG2 and
shows high ATPase stimulating effects. Only one compound exhibited an inhibiting ATPase
activity curve with 60% under the basal activity in the highest concentration. The third larger
group of compounds comprises a half maximal ATPase activation of less than 50%. The last
group illustrates a bellꢀshaped curve with a stimulation of ATPase activity by around 55% at
the top of the curve and an inhibiting character at higher concentrations as also suggested in
31, 32
former studies
. Interestingly, in the enzyme kinetic studies the same observations become
obvious. Enzyme kinetic interaction studies with Hoechst 33342 as substrate with selected
compounds exhibited three different modes of interaction: competitive, nonꢀcompetitive and
“
mixedꢀtype” inhibitors were identified using the doubleꢀreciprocal LineweaverꢀBurk
linearization, the halfꢀreciprocal method according to HanesꢀWoolf and the direct linear plot
according to CornishꢀBowden. The results of all three methods were in accordance. The
current study of 4ꢀsubstituted 2ꢀpyridylquinazolines showed to be a worthwhile continuation
of the earlier studies of 4ꢀsubstituted 2ꢀphenylquinazolines: extended insights into the SAR of
this new class of inhibitors regarding the inhibitory activity and selectivity toward ABCG2
together with the predominantly low cytotoxic effects observed in the MTT assay are some of
the findings. Potential applicability for in vivo studies was confirmed by the high efficacy in
reversing MDR when coꢀadministered with the cytostatic agents SNꢀ38 and MX. ATPase
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
results reflect the diversity of the compound interactions with the ABCG2 transport protein.
This was also illustrated by the results of enzyme kinetic studies yielding different modes of
interaction with the substrate Hoechst 33342.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Experimental Section
Chemistry. Materials. Chemicals were purchased from Acros Organics (Geel, Belgium),
Alfa Aesar (Karlsruhe, Germany), Sigma Aldrich (Steinheim, Germany) or Merck (Darmstadt,
Germany) and used without further purification. Microwave reactions were carried out in 50
ml vials with a CEM Discover SP (CEM GmbH, KampꢀLintfort, Germany). Progress of the
reaction was monitored by thin layer chromatography (TLC) using an aluminium plate coated
with silica gel 60 F254 (Merck Millipore, Billerica, MA, USA). A mixture of dichloromethane
and methanol (9:1) was used as eluent and the TLCs analyzed in a UV cabinet with an
1
excitation wavelength of 254 nm. Identity of the compounds was confirmed by NMR. Hꢀ and
1
3
C spectra were either obtained on a Bruker Advance 500 MHz (500/126 MHz) or Bruker
Advance 600 MHz (600/151 MHz). DMSOꢀd6 was used as solvent for the compounds as well
1
3
as internal standard to give chemical shifts (δ) expressed in ppm. Assignment of the
C
signals was carried out by distortionless enhancement by polarization transfer (DEPT) and
attached proton test (APT). The signals multiplicity is indicated as singulet (s), doublet (d),
doublet of doublets (dd), triplet of doublets (td), triplet (t), doublet of triplets (dt), quartet (q)
and multiplet (m). Coupling constants J are given in Hz. The purity of all biologically
evaluated compounds was determined to be > 95% by elemental analysis recorded on a Vario
EL V24 CHN Elemental Analyzer (Elementar Analysensysteme GmbH, Hanau, Germany).
Found values were in the range of ± 0.4% of the theoretical values, unless indicated.
Calculation of the logP values was performed with ACD/ChemSketch, file version C10E41,
Advanced Chemistry Development, Inc., Toronto, ON, Canada, www.acdlabs.com, 2015.
ACS Paragon Plus Environment

Page 25 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
General Procedure for the Preparation of the 2ꢀPyridylquinazolinꢀ4(3H)ꢀone Derivatives
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
ꢀ3.
pyridinecarboxaldehyde (20 mmol), iodine (3.17 g, 25 mmol), anhydrous potassium carbonate
2.76 g, 20 mmol) and 20 ml DMF was stirred at 70ꢀ90 °C for 4ꢀ8 h. The end of the reaction
A
mixture of anthranilamide (2.72 g, 20 mmol), the corresponding
(
was monitored by TLC and the mixture poured on crushed ice to form a precipitate.
Incomplete precipitation was prevented by adjusting the pH with concentrated HCl solution to
about 7. After filtering off the precipitate, it was thoroughly washed with 100 mL of a 20%
sodium thiosulfate solution followed by 100 mL of hot distilled water. Purification was
performed by recrystallization from ethanol.
2
ꢀ(Pyridinꢀ2ꢀyl)quinazolinꢀ4(3H)ꢀone (1). The compound was synthesized from
picolinaldehyde (2.14 g, 20 mmol) as described in the general procedure for 1ꢀ3 to yield 1 as
1
a white solid (3,84 g, 86%). H NMR (500 MHz, DMSOꢀd
6
) δ 11.77 (s, 1H), 8.76 (ddd, J =
4
=
=
.7, 1.7, 0.9 Hz, 1H), 8.46 (dt, J = 7.9, 1.1 Hz, 1H), 8.18 (dd, J = 7.9, 1.5 Hz, 1H), 8.07 (td, J
7.7, 1.7 Hz, 1H), 7.87 (ddd, J = 8.5, 7.0, 1.6 Hz, 1H), 7.80 (d, J = 8.0 Hz, 1H), 7.66 (ddd, J
13
7.6, 4.8, 1.2 Hz, 1H), 7.57 (ddd, J = 8.1, 7.0, 1.2 Hz, 1H). C NMR (126 MHz, DMSO) δ
149.1, 148.8, 138.1, 134.8, 127.8, 127.4, 126.7, 126.2, 122.3, 122.1.
2
ꢀ(Pyridinꢀ3ꢀyl)quinazolinꢀ4(3H)ꢀone (2). The compound was synthesized from
nicotinaldehyde (2.14 g, 20 mmol) as described in the general procedure for 1ꢀ3 to yield 2 as
1
a white solid (3,34 g, 75%). H NMR (500 MHz, DMSOꢀd
6
) δ 12.70 (s, 1H), 9.30 (d, J = 2.3
Hz, 1H), 8.76 (dd, J = 4.8, 1.6 Hz, 1H), 8.51 (dt, J = 8.1, 1.9 Hz, 1H), 8.17 (dd, J = 7.9, 1.5
Hz, 1H), 7.85 (ddd, J = 8.5, 7.0, 1.6 Hz, 1H), 7.76 (dd, J = 8.2, 1.1 Hz, 1H), 7.60 (ddd, J = 8.0,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
3
4.9, 0.8 Hz, 1H), 7.55 (ddd, J = 8.1, 7.1, 1.2 Hz, 1H). C NMR (126 MHz, DMSO) δ 162.2,
1
51.7, 150.8, 148.7, 148.6, 135.8, 134.8, 129.0, 127.6, 127.1, 126.0, 123.7, 121.3.
2
ꢀ(Pyridinꢀ4ꢀyl)quinazolinꢀ4(3H)ꢀone (3). The compound was synthesized from
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
isonicotinaldehyde (2.14 g, 20 mmol) as described in the general procedure for 1ꢀ3 to yield 3
1
as a white solid (3,04 g, 68%). H NMR (500 MHz, DMSOꢀd
6
) δ 12.73 (s, 1H), 8.83 – 8.74
(
m, 2H), 8.18 (dd, J = 7.9, 1.5 Hz, 1H), 8.14 – 8.08 (m, 2H), 7.87 (ddd, J = 8.5, 7.1, 1.6 Hz,
13
1
H), 7.79 (dd, J = 8.2, 1.1 Hz, 1H), 7.58 (ddd, J = 8.1, 7.1, 1.2 Hz, 1H). C NMR (126 MHz,
DMSO) δ 162.1, 150.7, 150.3, 148.4, 140.1, 134.9, 127.9, 127.5, 126.1, 121.7, 121.6.
General Procedure for the Preparation of the 4ꢀChloroꢀ2ꢀpyridylquinazoline Derivatives
4
ꢀ6.
The corresponding 2ꢀpyridylquinazolinꢀ4(3H)ꢀone derivative (10 mmol) was added to
phosphorus oxychloride (30 mL, 0.32 mol) and stirred for 10 min at room temperature. The
mixture was then refluxed for 4ꢀ8 h and the reaction monitored by TLC. After completion of
3
the reaction, excess POCl was removed under reduced pressure and the residue poured in 50
mL ice water. Subsequently, 50 mL DCM was added while stirring and the pH of the mixture
slowly adjusted to 7 with 25% ammonium solution. With a separatory funnel, the organic
phase was collected, washed with 50 mL brine and dried over MgSO
4
. The solvent was
removed under reduced pressure and the obtained solid recrystallized from isopropanol.
4
ꢀChloroꢀ2ꢀ(pyridinꢀ2ꢀyl)quinazoline (4). The compound was synthesized from 1 (2.23 g,
1
0 mmol) as described in the general procedure for 4ꢀ6 to yield 4 as a white solid (2.15 g,
1
89%). H NMR (500 MHz, DMSOꢀd
6
) δ 8.77 (ddd, J = 4.8, 1.8, 1.0 Hz, 1H), 8.47 (dt, J = 7.9,
1
.1 Hz, 1H), 8.21 – 8.14 (m, 1H), 8.09 (td, J = 7.8, 1.7 Hz, 1H), 7.87 (ddd, J = 8.5, 7.0, 1.6 Hz,
1
H), 7.82 (dd, J = 7.9, 1.5 Hz, 1H), 7.67 (ddd, J = 7.6, 4.8, 1.2 Hz, 1H), 7.57 (ddd, J = 8.2, 7.0,
ACS Paragon Plus Environment

Page 27 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
3
1.3 Hz, 1H). C NMR (126 MHz, DMSO) δ 160.9, 150.1, 149.0, 148.4, 147.8, 138.1, 134.7,
1
27.3, 127.2, 126.7, 126.1, 122.3, 121.9.
4
ꢀChloroꢀ2ꢀ(pyridinꢀ3ꢀyl)quinazoline (5). The compound was synthesized from 2 (2.23 g,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
0 mmol) as described in the general procedure for 4ꢀ6 to yield 5 as a white solid (2.27 g,
1
94%). H NMR (500 MHz, DMSOꢀd
6
) δ 9.29 (d, J = 2.2 Hz, 1H), 8.74 (dd, J = 4.8, 1.6 Hz,
1
H), 8.49 (dt, J = 8.0, 2.0 Hz, 1H), 8.16 (dd, J = 7.9, 1.5 Hz, 1H), 7.84 (ddd, J = 8.5, 7.0, 1.6
13
Hz, 1H), 7.78 – 7.72 (m, 1H), 7.61 – 7.49 (m, 2H). C NMR (126 MHz, DMSO) δ 162.2,
1
51.9, 150.9, 148.9, 148.6, 135.5, 134.8, 128.9, 127.7, 127.1, 126.0, 123.6, 121.3.
4
1
9
ꢀChloroꢀ2ꢀ(pyridinꢀ4ꢀyl)quinazoline (6). The compound was synthesized from 3 (2.23 g,
0 mmol) as described in the general procedure for 4ꢀ6 to yield 6 as a white solid (2.18 g,
1
0%). H NMR (500 MHz, DMSOꢀd
6
) δ 9.05 – 8.96 (m, 2H), 8.54 – 8.46 (m, 2H), 8.20 (dd, J
=
8.0, 1.4 Hz, 1H), 7.90 (ddd, J = 8.5, 7.1, 1.6 Hz, 1H), 7.83 (dd, J = 8.3, 1.2 Hz, 1H), 7.62
13
(ddd, J = 8.1, 7.1, 1.2 Hz, 1H). C NMR (126 MHz, DMSO) δ 162.0, 149.4, 148.0, 145.5,
1
45.0, 135.1, 128.2, 128.1, 126.1, 124.0, 121.8.
6
ꢀMethylꢀ2ꢀ(pyridinꢀ4ꢀyl)pyrimidinꢀ4(3H)ꢀone (7). To a solution of isonicotinimidamide
hydrochloride (158 mg, 1 mmol) in methanol (5 mL) was added sodium methoxide (54.0 mg,
mmol) and stirred for 10 min at room temperature. After adding methyl acetoacetate (116
mg, 1 mmol) the mixture was transferred into a microwave tube, sealed and heated to 70 °C at
0 watt microwave irradiation for 4 h. After cooling to room temperature, the pH of the
1
6
mixture was adjusted with 1 M hydrochloric acid to 7 and the formed precipitate filtered with
suction. The solid was washed with water and dried in vacuo to give 7 as a white solid (76.8
1
mg, 41%). H NMR (500 MHz, DMSOꢀd
6
) δ 12.59 (s, 1H), 8.80 – 8.66 (m, 2H), 8.09 – 7.99
13
(m, 2H), 6.37 (s, 1H), 2.39 – 2.24 (m, 3H). C NMR (126 MHz, DMSO) δ 150.4, 121.6, 23.5.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
6
ꢀMethylꢀ2ꢀphenylpyrimidinꢀ4(3H)ꢀone (8). This compound was sythesized according to
the procedure described for 7. Instead of isonicotinimidamide hydrochloride, benzimidamide
hydrochloride (157 mg, 1 mmol) was used in the first step to yield 8 as a white solid (89.4 mg,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
4
8%). H NMR (500 MHz, DMSOꢀd
6
) δ 12.47 (s, 1H), 8.22 – 7.99 (m, 2H), 7.60 – 7.42 (m,
1
3
3H), 6.19 (s, 1H), 2.26 (s, 3H). C NMR (126 MHz, DMSO) δ 164.7, 163.9, 157.2, 132.9,
131.6, 128.7, 127.9, 110.0, 23.6.
General Procedure for the Preparation of the 2ꢀSubstituted 4ꢀChloroꢀ6ꢀ
methylpyrimidine Derivatives 9ꢀ10.
The compounds were synthesized according to the method described in the general methods
for compounds 4ꢀ6 from the corresponding 2ꢀsubstituted 6ꢀmethylꢀpyrimidinꢀ4(3H)ꢀones 7
and 8. Complete chlorination was achieved within 4 h.
4ꢀChloroꢀ6ꢀmethylꢀ2ꢀ(pyridinꢀ4ꢀyl)pyrimidine (9). The compound was synthesized from 7
(1.87 g, 10 mmol) as described in the general procedure for 9ꢀ10 to yield 9 as a white solid
1
(1.79 g, 87%). H NMR (600 MHz, DMSOꢀd ) δ 8.81 – 8.74 (m, 2H), 8.21 – 8.16 (m, 2H),
6
1
3
7
.69 (d, J = 0.7 Hz, 1H), 2.58 (s, 3H). C NMR (151 MHz, DMSO) δ 170.6, 162.1, 160.9,
150.8, 143.0, 121.8, 120.8, 23.7.
4
ꢀChloroꢀ6ꢀmethylꢀ2ꢀphenylpyrimidine (10). The compound was synthesized from 8 (1.86
g, 10 mmol) as described in the general procedure for 9ꢀ10 to yield 10 as a white solid (1.90 g,
1
9
3%). H NMR (500 MHz, DMSOꢀd
6
) δ 8.39 – 8.27 (m, 2H), 7.60 – 7.45 (m, 4H), 2.55 (d, J
1
3
= 0.6 Hz, 3H). C NMR (126 MHz, DMSO) δ 170.1, 163.8, 160.6, 135.9, 131.6, 128.9, 128.1,
119.0, 23.8.
ACS Paragon Plus Environment

Page 29 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
General Procedure for the Preparation of the Substituted 4ꢀAnilinoquinazolines 11ꢀ41.
The corresponding 4ꢀchloroquinazoline derivative 4ꢀ6 (1 mmol) was added to isopropanol (5
mL) with the corresponding substituted aniline derivative (1 mmol) and sealed in a
microwave tube. The mixture was heated by 100 watt microwave irradiation to 110 °C for a
period of 15 – 30 min until completion of the reaction as indicated by TLC. The formed
precipitate was filtered off, washed with 10 mL isopropanol and dried in vacuo. If no
precipitate is formed, the solvent was removed under reduced pressure and the remaining
solid recrystallized from ethanol.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Nꢀ(3ꢀNitrophenyl)ꢀ2ꢀ(pyridinꢀ2ꢀyl)quinazolinꢀ4ꢀamine (11).
The compound was synthesized from 4 (242 mg, 1 mmol) and 3ꢀnitroaniline (138 mg, 1
mmol) as described in the general procedure for compounds 11ꢀ41 to yield 11 as a yellow
1
solid (292 mg, 85%), mp 274ꢀ275 °C. H NMR (500 MHz, DMSOꢀd
6
) δ 12.09 (s, 1H), 9.09
(dd, J = 8.4, 1.3 Hz, 1H), 9.02 (t, J = 2.2 Hz, 1H), 8.93 (ddd, J = 4.8, 1.7, 0.9 Hz, 1H), 8.43
(dt, J = 7.9, 1.1 Hz, 1H), 8.39 (ddd, J = 8.1, 2.1, 0.9 Hz, 1H), 8.33 (dd, J = 8.5, 1.1 Hz, 1H),
8
7
1
.24 – 8.17 (m, 2H), 8.14 (ddd, J = 8.5, 7.1, 1.2 Hz, 1H), 7.89 (ddd, J = 8.3, 7.1, 1.2 Hz, 1H),
13
.86 – 7.78 (m, 2H). C NMR (126 MHz, DMSO) δ 159.8, 154.9, 149.4, 148.3, 148.0, 141.4,
39.4, 138.5, 136.4, 130.2, 128.8, 128.2, 125.0, 124.7, 122.1, 120.6, 118.8, 113.8. Anal. Calcd.
13 5 2
for C19H N O : C, 66.47; H, 3.82; N, 20.40. Found: C, 66.38 ; H, 4.19; N, 20.16.
Nꢀ(3ꢀNitrophenyl)ꢀ2ꢀ(pyridinꢀ3ꢀyl)quinazolinꢀ4ꢀamine (12).
The compound was synthesized from 5 (242 mg, 1 mmol) and 3ꢀnitroaniline (138 mg, 1
mmol) as described in the general procedure for compounds 11ꢀ41 to yield 12 as a yellowꢀ
1
beige solid (278 mg, 81%), mp >300 °C. H NMR (500 MHz, DMSOꢀd
6
) δ 10.51 (s, 1H),
9.61 (s, 1H), 9.17 (t, J = 2.2 Hz, 1H), 8.99 (dt, J = 8.1, 1.9 Hz, 1H), 8.84 (d, J = 4.9 Hz, 1H),
8.68 (dt, J = 8.4, 1.0 Hz, 1H), 8.36 (ddd, J = 8.2, 2.2, 0.9 Hz, 1H), 8.04 (ddd, J = 8.2, 2.3, 0.9
ACS Paragon Plus Environment