C2-substituted quinazolinone derivatives exhibit A1 and/or A2A adenosine receptor affinities in the low micromolar range

Article abstract of DOI:10.1016/j.bmcl.2020.127274

Antagonists of the adenosine receptors (A₁ and A_2A subtypes) are widely researched as potential drug candidates for their role in Parkinson's disease-related cognitive deficits (A₁ subtype), motor dysfunction (A_2A subtype) and to exhibit neuroprotective properties (A_2A subtype). Previously the benzo-α-pyrone based derivative, 3-phenyl-1H-2-benzopyran-1-one, was found to display both A₁ and A_2A adenosine receptor affinity in the low micromolar range. Prompted by this, the α-pyrone core was structurally modified to explore related benzoxazinone and quinazolinone homologues previously unknown as adenosine receptor antagonists. Overall, the C2-substituted quinazolinone analogues displayed superior A₁ and A_2A adenosine receptor affinity over their C2-substituted benzoxazinone homologues. The benzoxazinones were devoid of A_2A adenosine receptor binding, with only two compounds displaying A₁ adenosine receptor affinity. In turn, the quinazolinones displayed varying degrees of affinity (low micromolar range) towards the A₁ and A_2A adenosine receptor subtypes. The highest A₁ adenosine receptor affinity and selectivity were favoured by methyl para-substitution of phenyl ring B (A₁K_i = 2.50 μM). On the other hand, 3,4-dimethoxy substitution of phenyl ring B afforded the best A_2A adenosine receptor binding (A_2AK_i = 2.81 μM) among the quinazolinones investigated. In conclusion, the quinazolinones are ideal lead compounds for further structural optimization to gain improved adenosine receptor affinity, which may find therapeutic relevance in Parkinson's disease-associated cognitive deficits and motor dysfunctions as well as exerting neuroprotective properties.

Full text of DOI:10.1016/j.bmcl.2020.127274

Journal Pre-proofs
C2-substituted quinazolinone derivatives exhibit A1 and/or A2A adenosine re‐
ceptor affinities in the low micromolar range
Lianie Pieterse, Mietha M. van der Walt, Gisella Terre'Blanche
PII:
DOI:
Reference:
S0960-894X(20)30384-X
https://doi.org/10.1016/j.bmcl.2020.127274
BMCL 127274
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
6 February 2020
11 May 2020
16 May 2020
Please cite this article as: Pieterse, L., van der Walt, M.M., Terre'Blanche, G., C2-substituted quinazolinone
derivatives exhibit A1 and/or A2A adenosine receptor affinities in the low micromolar range, Bioorganic &
Medicinal Chemistry Letters (2020), doi: https://doi.org/10.1016/j.bmcl.2020.127274
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Ltd.

C2-substituted quinazolinone derivatives
exhibit A1 and/or A2A adenosine receptor
affinities in the low micromolar range
Lianie Pieterse a; Mietha M. van der Walt a,b*and Gisella Terre’Blanche a,c
a
Centre of Excellence for Pharmaceutical Sciences, School of Pharmacy, North-West University, Private Bag X6001,
Potchefstroom 2520, South Africa;
b
Human Metabolomics, North-West University, Private Bag X6001, Box 269, Potchefstroom, 2531, South Africa;
c
Pharmaceutical Chemistry, School of Pharmacy, North-West University, Private Bag X6001, Potchefstroom 2520, South Africa;
Abstract
Antagonists of the adenosine receptors (A1 and A2A subtypes) are widely researched as
potential drug candidates for their role in Parkinson’s disease-related cognitive deficits (A1
subtype), motor dysfunction (A2A subtype) and to exhibit neuroprotective properties (A2A
subtype). Previously the benzo--pyrone based derivative, 3-phenyl-1H-2-benzopyran-1-one,
was found to display both A1 and A2A adenosine receptor affinity in the low micromolar range.
Prompted by this, the -pyrone core was structurally modified to explore related
benzoxazinone and quinazolinone homologues previously unknown as adenosine receptor
antagonists. Overall, the C2-substituted quinazolinone analogues displayed superior A1 and
A2A adenosine receptor affinity over their C2-substituted benzoxazinone homologues. The
benzoxazinones were devoid of A2A adenosine receptor binding, with only two compounds
displaying A1 adenosine receptor affinity. In turn, the quinazolinones displayed varying
degrees of affinity (low micromolar range) towards the A1 and A2A adenosine receptor
subtypes. The highest A1 adenosine receptor affinity and selectivity were favoured by methyl
para-substitution of phenyl ring B (A1Ki = 2.50 μM). On the other hand, 3,4-dimethoxy
substitution of phenyl ring B afforded the best A2A adenosine receptor binding (A2AKi = 2.81
μM) among the quinazolinones investigated. In conclusion, the quinazolinones are ideal lead
compounds for further structural optimization to gain improved adenosine receptor affinity,
which may find therapeutic relevance in Parkinson’s disease-associated cognitive deficits and
motor dysfunctions as well as exerting neuroprotective properties.
Keywords
Keywords: benzoxazinone; quinazolinone; benzo--pyrone; Adenosine A2A receptor; Adenosine A1 receptor; antagonist; GTP
shift; radioligand binding assay; Parkinson's disease.
*Corresponding author. Human Metabolomics, North-West University, Private Bag X6001, Box 269, Potchefstroom, 2531, South
Africa. E-mail address: dalenevanderwalt@gmail.com and 13035134@nwu.ac.za (M.M van der Walt).
Abbreviations: PD; Parkinson’s Disease; AR, adenosine receptor; GTP, guanosine triphosphate; [3H]DPCPX, [3H]-dipropyl-8-
cyclopentylxanthine; [3H]NECA, N-[3H]-ethyladenosin-5’-uronamide; CPA, N6-cyclopentyladenosine.

Parkinson’s disease (PD) is a complex neurodegenerative disorder and is characterized by
the formation of Lewy bodies and the death of dopaminergic neurons along the nigrostriatal
pathway [1, 2]. The loss of dopaminergic neurons results in a diminished amount of dopamine
present in the corpus striatum [3]. In 2005 the prevalence of PD among individuals living in
the United States, aged above 65 to 70 years, was documented as 9.5 per 1000 individuals
[4] and it is estimated that between 2005 and 2030 the prevalence will have doubled [5].
Individuals suffering from PD have been found to present with a range of mainly motor features
as first described by James Parkinson in 1817 [6, 7]. Four principal motor features were
identified, namely tremors, rigidity, bradykinesia and postural instability [6]. Over time some
non-motor symptoms may develop which are typically defined by cognitive impairment [8].
At present, the treatment of PD still follows a more traditional approach in the sense that
current treatment is more symptomatic rather than curative [9]. Symptomatic treatment seeks
to replenish dopamine stores or to disrupt the degradation of endogenous dopamine by
inhibition of dopamine degrading enzymes [10]. The current treatment with regards to the
motor symptoms of PD entails oral preparations of levodopa (L-3,4-dihydroxyphenylalanine)
and dopamine receptor agonists as well as deep brain stimulation and apomorphine in severe
cases [11]. Although levodopa remains the cornerstone treatment of PD-related motor
symptoms [9], the development of dyskinesia with long-term use [12], along with the alarming
prevalence and the need for neuroprotective and disease-modifying agents, demonstrate the
necessity for novel drug development [1].
In the brain, adenosine functions as a neuromodulator and fulfils a physiological role opposite
to that of dopamine [13, 14]. Four adenosine receptor (AR) subtypes (A1, A2A, A2B and A3) exist
and of the four subtypes, A1 and A2A ARs have the highest density in the brain [15]. The A1
ARs are found predominantly in the cortex, cerebellum and hippocampus of the brain [16, 17],
and in turn, the A2A ARs are generally more restricted to the dorsal striatum, nucleus
accumbens, and olfactory tubercle [18]. Both the A1 and A2A ARs have been identified as
possible targets for drug development in neurological disorders, such as PD [19], and are
currently being examined for their potential benefit in PD [20]. The A2A AR antagonists have
been found to enhance motor activity with a reduced risk of developing dyskinesia and other
associated side effects, making them viable candidates for the treatment of PD [21]. Preclinical
data exists demonstrating that A2A AR antagonists may possess neuroprotective potential [22],
where neuroprotective qualities have been displayed by a selective A2A AR antagonist,
Istradefylline, in
a
1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) induced
dopaminergic toxicity study [23]. More importantly, Istradefylline has recently been approved
by the FDA as add-on therapy in PD under the trade name Nourianz [24].

A1 ARs are important for cognitive function and antagonism of these receptors may improve
cognition [25]. A selective A1 AR antagonist, such as FR-194921, has been examined in
animal models and appears to positively modulate memory processes while having a
stimulating effect on the central nervous system [26]. Therefore, A1 AR antagonists are being
examined for use as cognitive enhancers in neurological disorders such as dementia,
Alzheimer’s disease and PD [27]. It was also observed that the blockade of both the A1 and
A2A AR subtypes lead to a synergistic positive motor response, which could be ascribed to the
release of dopamine triggered by the antagonism of the A1 AR with the simultaneous
enhancement of the postsynaptic response to dopamine potentiated by antagonism of the A2A
AR [28].
Thus, antagonists of the A1 and/or A2A AR are thought to exert neuroprotective properties and
enhance motor function via A2A AR antagonism, while A1 AR antagonism improves PD-
associated cognitive dysfunction. To date, an ideal drug candidate has not been identified for
PD treatment. Therefore, considering the aforementioned potential of AR antagonists as
prospective PD treatment and the lack of an ideal drug candidate, the need exists to discover
high affinity A1 and/or A2A AR antagonists.
Recently, a series of benzopyrone derivatives were evaluated as AR antagonists [29]. The
benzo--pyrone compound 3-phenyl-1H-2-benzopyran-1-one (1), possessed both A1 and A2A
AR affinity (A1Ki = 7.41 μM; A2AKi = 3.35 μM) with a selectivity index (SI) of 2 towards the A2A
AR subtype (Figure 1; Table 1). Compound 1 consists of a basic benzo--pyrone skeleton
(ring A and C is fused) with ring C bearing a C3-phenyl substituted side-chain (ring B) (Figure
1, A). The double bond between position C3 and C4 of ring C was found to be imperative for
AR binding and the arrangement of the ketone and hetero oxygen in ring C was optimal for
A2A AR binding. In analogy to the structure of 3-phenyl-1H-2-benzopyran-1-one (1) the present
study investigates the A1 and A2A AR binding properties of structurally related C2-substituted
benzoxazinones (2a–j, 3), 3-phenyl-isoquinolinone (4) and C2-substituted quinazolinones
(5a–j) (Figure 1; Table 1).
The investigated C2-substituted benzoxazinones (2a–j) possess the basic scaffold of
compound 1 with the addition of hetero nitrogen to ring C (Figure 1, B). This structural
modification allows the structure-activity relationship (SAR) exploration of additional nitrogen
to ring C, while retaining compound 1’s aforementioned double bond (ring C) deemed
essential for AR affinity. The benzoxazinones are further assessed by various substitutions on
phenyl ring B (2b–g). The inclusion of compound 3 will provide insight into the necessity of
the double bond in ring C of the benzoxazinone scaffold to govern AR binding (Figure 1, B).
The comparison of the isoquinolinone derivative (4) to compound 1 will highlight the

importance of the hetero oxygen in ring C of the benzo--pyrone backbone to favour AR
affinity (Figure 1, C). Furthermore, the C2-substituted quinazolinones (5a–j) is included to
afford further SAR exploration of the AR affinity if ring C bears two hetero nitrogen atoms
(Figure 1, D). The effect of other heterocyclic ring systems (e.g. furyl and thiophene) replacing
phenyl ring B of the benzoxazinones (2h and i) and quinazolinones (5h and i), as well as the
insertion of a styryl side-chain between ring C and B of the benzoxazinones (2j) and
quinazolinones (5j) are additionally evaluated in vitro. Lastly, the rearrangement of the ketone
and hetero oxygen of compound 1 is investigated by including compounds 6 and 7, where this
rearrangement was switched/flipped (Figure 2). Table 1 documents the A1 and A2A AR
dissociation constant (Ki) values determined via radioligand binding assays.
A
benzo--pyrone
O
O
C
A
(1)
B
B
benzoxazinone
O
D quinazolinone
A1Ki = 7.41 M
2AKi = 3.35 M
Add N
O
Replace O
Add N
A
O
NH
C
A
C
A
N
R
N
R
(5a-j)
(2a-j)
C isoquinolinone
O
NH
A
C
(4)
B
Figure 1: General structures of the various scaffolds explored in the current investigative
study: (A) benzo--pyrone (1), (B) benzoxazinone (2a-j), (C) isoquinolinone (4) and (D)
quinazolinones (5a-j).

O
O
O
O
O
A
C
C
A
B
(1)
B
(6)
O
O
O
A
C
C
A
N
(7)
N
(2a)
B
B
Figure 2: Depicting the rearrangement of the ketone and hetero oxygen of ring C by the benzo--
pyrone (1 vs 6) and the benzoxazinones (2a vs 7).
The test compounds were either obtained commercially (1, 2b, 2c, 2e, 2g, 2j, 3, 6 and 7) or
synthesized (2a, 2d, 2f, 2h, 2i, 4 and 5a–j) according to literature procedures. The test
compounds obtained from standard commercial sources (Sigma Aldrich) were used without
further purification.
The benzoxazinone analogues (2a, 2d, 2f, 2h and 2i) that were not commercially obtained,
were synthesized in an alkaline environment with commercially available acyl chlorides and
anthranillic acid as key starting materials (Scheme 1) [30].
The synthesis of the proposed quinazolinone derivatives (5a–j) was performed by either
method A and/or B. Method A involved a one-pot synthesis with the appropriate aldehyde,
formamide and isatoic anhydride (Scheme 2) [31], while method B included treatment of the
corresponding benzoxazinone with ammonia (Scheme 3) [32]. Isoquinolinone derivative 4 was
synthesized by treating 3-phenyl-1H-2-benzopyran-1-one (1) with ammonia — a similar
synthetic approach than outlined in method B (Scheme 3) [32]. In each instance, the structures
of the proposed synthesized compounds (2a, 2d, 2f, 2h, 2i, 4 and 5a–j) were verified by 1H-
NMR, 13C-NMR and mass spectrometry. The reaction conditions and compound
characterizations are detailed in the supplementary information.
O
O
O
OH
2
O
0°C to r.t
Pyridine
+
R
Cl
NH
N
R
2a, 2d, 2f, 2h & 2i
Anthranilic acid
Acyl chloride
Scheme 1: A synthetic pathway for C2-substituted benzoxazinone derivatives (2a, 2d, 2f, 2h and 2i).

O
N
O
O
NH
O
120-125°C
+
R
H
NH2CHO, PEG-400
R
N
H
O
Isatoic anhydride
5a-g, 5i and 5j
Aldehyde
Scheme 2: A synthetic pathway to the synthesized C2-substituted quinazolinone derivatives (5a–g, 5i
and 5j), method A.
O
O
NH
O
100°C
NH3
X
X
4: X=CH
5a, 5d, 5f, 5h & 5i: X=N
1: X=CH
2a, 2d, 2f, 2h & 2i: X=N
Scheme 3: A synthetic pathway for the isoquinolinone (4, X=CH) and C2-substituted quinazolinone
derivatives (5a, 5d, 5f, 5h and 5i, X=N), method B.
Assays were completed using rat whole brain (A1) and rat striatal membranes (A2A).
Noteworthy, a recently study investigated potential differences between human, rat and
mouse ARs. Alnouri and co-workers calculated correlation coefficients and found that rat and
humans share a 0.80 correlation of pKi for the A1 AR and 0.83 for the A2A AR [33]. Furthermore,
Maemoto and colleagues found that selected pyrazolopyridine derivatives exhibited a similar
affinity in rat, human and guinea pig [34]. The A1 and A2A AR binding affinities of the test
compounds (1, 2a–j, 3, 4, 5a–j, 6 and 7) were evaluated in vitro with radioligand binding
assays described previously [35, 36]. The Research Ethics Committee of North-West
University approved the collection of tissue samples for the A1 and A2A AR radioligand binding
assays (application number NWU-0035-10-A5). The AR binding affinities are expressed as
the mean Ki values of triplicate determinations and are documented in Table 1. Additional
information with regards to the radioligand binding assays are detailed in the supplementary
information.
Introducing hetero nitrogen (ring C) to the -pyrone core of compound 1 afforded the
benzoxazinone derivative 2a (Figure 1, B). The latter compound exhibited a similar A1 AR
affinity as documented for compound 1 (1 A1Ki = 7.41 μM vs 2a A1Ki = 6.38 μM), whereas the
A2A AR binding affinity was diminished (1 A2AKi = 3.35 μM vs 2a A2AKi = > 100 μM).

The SARs of the benzoxazinone scaffold was further explored at ring B by investigating
several para-substituents on the phenyl ring (2b–g) and other heterocyclic ring systems which
included a furyl (2h) and a thiophene (2i) ring. In addition, the insertion of a styryl side-chain

Table 1 Dissociation constant values (Ki values) for the binding of the test compounds to rat
A1 and A2A adenosine receptors.
O
O
O
X
X
X
Y
3
N
N
R
4
(2a-g & 5a-g)
(1 & 4)
(2h-i & 5h-i)
O
O
N
O
O
X
X
N
X
H
(3)
(2j & 5j)
(6 & 7)
Ki ± SEM (µM)a
(% displacement)b
c
c
A1 vs
A2A vs
A1c + GTPd vs
GTP
SIf
(A2A/A)
Compd
X
R
-
Y
[3H]DPCPX
[3H]NECA
[3H]DPCPX
6.49 ± 0.82a,g
6.92 ± 1.16a
shifte
Benzo--pyrone scaffold
1
O
-
7.41 ± 0.4a,g
3.35 ± 0.80a,g
0.9g
0.5g
Benzoxazinone scaffold
2a
2b
2c
2d
2e
2f
2g
2h
2i
O
O
O
O
O
O
O
O
O
O
O
H
4-F
4-Cl
4-Br
-
-
-
-
-
-
-
O
S
-
6.38 ± 0.40a
> 100 (100%)b
> 100 (86%)b
> 100 (95%)b
> 100 (84%)b
> 100 (54%)b
> 100 (99%)b
> 100 (42%)b
32.4 ± 0.58a
> 100 (56%)b
5.06 ± 0.46a
> 100 (49%) b
> 100 (70%)b
> 100 (81%)b
> 100 (92%)b
> 100 (77%)b
> 100 (37%)b
> 100 (77%)b
> 100 (66%)b
> 100 (59%)b
> 100 (85%)b
> 100 (69%)b
1.09
-
-
-
-
-
-
-
-
-
-
-
4-CH3
4-OCH3
4-OCH2CH3
-
-
-
-
2j
3
-
5.72 ± 0.56a
3.77 ± 0.28a
1.12
1.02
Isoquinolinone scaffold
NH
Quinazolinone scaffold
4
-
-
>100 (100%)b
> 100 (100%)b
-
5a
5b
5c
5d
5e
5f
5g
5h
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
H
4-F
4-Cl
4-Br
-
-
-
-
-
-
3.67 ± 0.02a
> 100 (50%)b
> 100 (54%)b
> 100 (98%)b
2.50 ± 0.47a
> 100 (44%)b
> 100 (67%)b
4.62 ± 0.63a
> 100 (39%)b
> 100 (100%)b
18.7 ± 1.99a
> 100 (62%)b
> 100 (82%)b
> 100 (84%)b
> 100 (55%)b
> 100 (79%)b
2.81 ± 0.40a
8.11 ± 0.03a
> 100 (30%)b
> 100 (56%)b
5.09
-
-
-
-
-
-
1.76
-
-
4-CH3
4-OCH3
3,4-OCH3
2.95 ± 0.18a
5.89 ± 0.42
1.18
1.23
-
-
-
O
S
-
5i
5j
Coumarin scaffold
6
CH
-
-
-
-
> 100 (51%)b,g
17.3 ± 1.9a
> 100 (77%)b,g
> 100 (91%)b
-
-
Benzoxazinone scaffold
7
N
Reference compounds
CPA (A1 agonist)
0.0051 ± 0.0002 0.557 ± 0.024
29.5 ± 1.14
(35.2)i
5.8
(6)i
5784
(0.0079)h;
(0.460) h
(0.0059)i
ZM-241385 (A2A antagonist)
0.420 ± 0.01
0.0013 ± 0.00002
(0.002)j
0.510 ± 0.03
1.2
0.003
(0.225)j
a
b
All Ki values determined in triplicate and expressed as mean ± SEM; Percentage displacement of the radioligand at a maximum tested
concentration (100 µM); c Rat receptors were used (A1: rat whole brain membranes; A2A: rat striatal membranes); d GTP shift assay, where the 100
µM GTP was added to the A1 AR radioligand binding assay; e GTP shifts calculated by dividing the Ki in the presence of GTP by the Ki in the absence
of GTP; f Selectivity index (SI) for the A1 AR isoform calculated as the ratio of Ki (A2A)/Ki (A1); g Literature value obtained from reference [35];
h
Literature value obtained from reference[37]; i Literature value obtained from reference[38]; j Literature value obtained from reference [39].

(2j) between ring C and phenyl ring B was also investigated. Compared to 2a, para-
substitution (2b–g) of phenyl ring B with F (2b), Cl (2c), Br (2d), CH3 (2e), OCH3 (2f) and
OCH2CH3 (2g) diminished A1 AR affinity, while A2A AR activity remained elusive. The furyl
compound (2h) and insertion of a styryl side-chain (2j) showed no A1 or A2A AR affinity. Thus,
the C2-substituted benzoxazinones (2a–j) were generally found to be devoid of A1 and A2A AR
affinity. However, the unsubstituted phenyl ring B (2a) and the thiophene ring B (2i) were the
only benzoxazinones investigated to exhibited A1 AR affinity with Ki values of 6.38 µM and
32.4 µM, respectively.
Upon further investigation, ring C of compound 2a was saturated (absence of the double bond)
to yield compound 3. This structural modification led to a slight improvement of A1 AR affinity
compared to 2a (2a A1Ki = 6.38 μM vs 3 A1Ki = 5.06 μM), but affinity at the A2A AR subtype
remained absent (Figure 1, B; Table 1).
Further structural modification to ring C included the replacement of the hetero oxygen of
compound 1 with hetero nitrogen to yield compound 4 (Figure 1, C). Both the A1 and A2A AR
affinity were diminished, emphasizing the preference of the benzo--pyrone scaffold (1) over
the isoquinolinone moiety (4) for enhanced A1 and A2A AR activity.
Based on the aforementioned finding that the hetero oxygen (ring C) of the benzo--pyrone
derivative 1 is essential for AR activity, the structurally related benzoxazinone (2a) and
quinazolinone (5a) homologues were compared. It was found that replacing the hetero oxygen
(ring C) of 2a with hetero nitrogen resulted in 5a (A1Ki = 3.67 μM; A2AKi = 18.7 μM) which
exhibited superior A1 and A2A AR affinity over its benzoxazinone counterpart 2a (A1Ki = 6.38
μM; A2AKi >100 μM). The latter observation (2a vs 5a) is contrary to the above finding (1 vs 4).
Thus, replacement of compound 1’s hetero oxygen of ring C (benzo--pyrone) with hetero
nitrogen (5a) in addition to a second hetero nitrogen led to improved A1 AR affinity and reduced
A2A AR binding (Figure 1, D). The improved AR affinity of 5a compared to 2a may be ascribed,
in part, to ring C containing two hetero nitrogens (5a) compared to one (2a). The latter
observation is supported by the research of Gillespie and colleagues [40, 41]. In the study by
Gillespie and co-workers, they compared the binding affinities of pyridine, pyrimidine and
triazine to each other and found that two nitrogens in the heterocyclic ring are optimum to
enhance both the A1 and A2A AR binding [40, 41]. The latter statement was based on their
finding that the binding affinity of the pyridine scaffold was seven-fold less potent than the
triazine and 45-fold less potent than the corresponding aminopyrimidine [40, 41].
Furthermore, unlike compounds 5a and 1, the benzoxazinone derivative 2a only possessed
affinity toward the A1 AR. Noteworthy, an approximate two-fold A1 AR affinity improvement

was documented for 5a compared to compounds 1 and 2a. On the other hand, compound 1
displayed a six-fold higher A2A AR binding than 5a.
Since compound 5a favoured both A1 and A2A AR affinity (A1Ki = 3.67 μM; A2AKi = 18.7 μM),
further SAR of the quinazolinones was explored (5a–j) at ring B. Several para-substituents on
the phenyl ring (5b–g) and other heterocyclic ring systems which included a furyl (5h) and a
thiophene (5i) ring were investigated. In addition, the insertion of a styryl side-chain (5j)
between ring C and phenyl ring B was also explored. Halogen substitution (5b–d) exhibited
no affinity towards either the A1 or A2A AR subtypes. In contrast, para-substitution with a CH3
(5e) group exerted a favourable effect on A1 AR affinity with a Ki value of 2.50 μM. This
structural modification resulted in the highest A1 AR activity documented for the current study.
Interestingly, simultaneous meta- and para-substitution of ring B with a methoxy group led to
compound 5g with gained A2A AR affinity (A2AKi = 2.81 μM), whereas mono-substitution with
an OCH3 in the para-position (5f) led to diminished A2A AR affinity. Noteworthy, compound 5g
displayed the best A2A AR affinity amongst the test compounds. Surprisingly the C2-
substituted furyl ring (5h) exhibited both A1 and A2A AR affinity in the low micromolar range
(A1Ki = 4.62 μM; A2AKi = 8.11 μM).
In a previous study rearrangement of the ketone and hetero oxygen in ring C of compound 1
afforded compound 6 with diminished AR affinity of both AR subtypes [29]. Based on the latter
finding, compound 2a was directly compared to compound 7. This structural modification
resulted in a three-fold decrease in A1 AR activity (2a A1Ki = 6.38 μM vs 7 A1Ki = 17.3 μM),
while A2A AR affinity remained elusive. Generally, the optimal arrangement of the ketone and
hetero oxygen of ring C to favour A1 AR affinity is exhibited by the benzo--pyrone (1) and
benzoxazinone (2a) backbones.
Previously, the benzo--pyrone based derivative 1 was found to act as an A1 AR antagonist
[29]. In analogy to compound 1, the agonistic or antagonistic activity at the A1 AR subtype for
compounds 2a, 3, 5a, 5e and 5h were explored by performing guanosine triphosphate (GTP)
shift experiments as described in the literature [42]. In the case of an A1 AR agonist, a
rightward shift of the binding curve in the presence of GTP (due to an uncoupling of the A1 AR
from its Gi protein) is expected. On the other hand, an A1 AR antagonist is anticipated not to
shift the binding curve significantly in the presence of GTP [35, 36, 43]. Generally, a GTP shift
of approximately 6 is indicative of a full agonist and a GTP shift of 1, a full antagonist [38]. The
GTP shift binding curve of N6-cyclopentyladenosine (CPA) (Table 1, Figure 3) exhibited a GTP
shift value of 5.8, thus, indicative of a full agonist. In contrast, in the presence of GTP,
compounds 2a, 3, 5a, 5e and 5h was documented with no significant shift of the binding curve
and they exhibited shifts of 1.09 (2a), 1.12 (3), 1.02 (5a), 1.18 (5e) and 1.23 (5h), respectively.

Thus, compounds 2a, 3, 5a, 5e and 5h may be considered antagonists of the A1 AR (Figure
3).
A.
B.
100
100
+ GTP (100 M)
+ GTP (100 M)
- GTP
- GTP
50
0
50
0
-2 -1
0
1
2
3
4
-2 -1
0
1
2
3
4
Log [CPA]
Log [5e]
Figure 3: The binding curves of the reference compound CPA and 5e are examples of A1 AR agonistic
and antagonistic action, respectively. The functionality was determined via GTP shift assays (with and
without 100 µM GTP) in rat whole brain membranes expressing A1 ARs with [3H]DPCPX as radioligand.
(A) A GTP shift of 5.8 was calculated for CPA and (B) a GTP shift of 1.18 was calculated for compound
5e.
A1 AR affinity
-pyrone core
(lead compound)
O
O
C
A
(1)
B
± 2-fold improvement
of A1 AR affinity
slight improvement
of A1 AR affinity
A1Ki = 7.41 M
pyrimidone core
O
oxazinone core
O
NH
O
A
C
A
C
N
(5a)
N
(2a)
B
B
± 2-fold improvement
of A1 AR affinity
A1Ki = 3.67 M
A1Ki = 6.38 M
Figure 4: Structural changes to the -pyrone core resulted in the pyrimidone core and
oxazinone core. The pyrimidone core (5a) exhibited improved A1 AR affinity in analogy to the
oxazinone core (2a).

A
2A AR affinity
-pyrone core
(lead compound)
O
O
C
A
(1)
B
6-fold decline in
2A AR affinity
A2AKi = 3.35 M
loss of A2A AR affinity
A
pyrimidone core
O
oxazinone core
O
NH
O
A
C
A
C
N
(5a)
N
(2a)
B
B
improvement of
2A AR affinity
A1Ki = 18.7 M
A
2AKi > 100 M
A
Figure 5: Structural changes to the -pyrone core resulted in the pyrimidone core and
oxazinone core. The pyrimidone core (5a) exhibited improved A2A AR affinity in analogy to the
oxazinone core (2a).
In conclusion, the newly proposed structural modifications of the benzo--pyrone based
derivative 3-phenyl-1H-2-benzopyran-1-one (1) resulted in varying degrees of affinity and
selectivity towards the A1 and A2A AR subtypes. The A1 AR binding was improved two-fold
after replacing the -pyrone core (ring C) of compound 1 with a pyrimidone ring to afford the
quinazolinone derivative 5a (1 A1Ki = 7.41 μM vs 5a A1Ki = 3.67 μM) (Figure 4). However,
replacement of the -pyrone core (ring C) with an oxazinone afforded the benzoxazinone
derivative 2a which only showed a slight A1 AR affinity improvement (1 A1Ki = 7.41 μM vs 2a
A1Ki = 6.38 μM). Generally, the A2A AR binding was best governed by ring C as the -pyrone
(1 A2AKi = 3.35 μM) compared to oxazinone (2a A2AKi = > 100 μM) and pyrimidone (5a A2AKi
= 18.7 μM) (Figure 5).
Noteworthy, the benzoxazinone and quinazolinone based scaffolds were previously unknown
to exhibit AR affinity and that analogues of these moieties may act as A1 AR antagonists.
Generally, the C2-substituted quinazolinone derivatives displayed superior A1 and A2A AR
affinity over the C2-substituted benzoxazinone derivatives. Among the test compounds, the
quinazolinone analogue 5a exhibited the second-highest A1 AR binding (A1Ki = 3.67 μM). The
introduction of a CH3 group (para-position) to ring B of 5a afforded compound 5e (A1Ki = 2.50
μM)) with the highest A1 AR affinity among the compounds investigated. Furthermore, 5e was
found to be selective for the A1 AR subtype. The 3,4-dimethoxy substituted quinazolinone 5g
exhibited the highest A2A AR affinity and selectivity (A2AKi = 2.81 μM). Thus, 5g is an example

of a selective A2A AR drug and may find therapeutic relevance to enhance PD-related motor
dysfunction and exert neuroprotective properties, while 5e may enhance cognitive dysfunction
associated with PD. The identified drug candidates (5a, 5e and 5g) are ideal for future drug
optimization and in vivo examinations as selective A1 (5a, 5e) and A2A (5g) AR drugs in the
treatment of neurological disorders.

Acknowledgements
The research project was designed by MMvdW. The manuscript was drafted, written and
revised by MMvdW, LP and GT. The radioligand binding assays were performed by MMvdW
and the biological data interpreted by MMvdW, LP and GT. The organic synthesis, where
applicable, was performed by LP and the chemical characterizations were interpreted by
MMvdW, LP and GT. All authors give their approval of this version of the manuscript. We are
grateful to Dr. J. Jordaan of the SASOL Centre for Chemistry, North-West University, for
recording the NMR and MS spectra of the synthesized compounds. Financial support for this
work was provided by the North-West University and the National Research Foundation of
South Africa.
References
[1] L.V. Kalia, A.E. Lang, Parkinson disease in 2015: evolving basic, pathological and clinical concepts in
PD, Nature reviews Neurology 12(2) (2016) 65.
[2] M. Lelos, R. Morgan, C. Kelly, E.M. Torres, A.E. Rosser, S.B. Dunnett, Amelioration of non-motor
dysfunctions after transplantation of human dopamine neurons in a model of Parkinson's disease, Exp.
Neurol. 278 (2016) 54-61.
[3] H. Ehringer, O. Hornykiewicz, Distribution of noradrenaline and dopamine (3-hydroxytyramine) in
the human brain and their behavior in diseases of the extrapyramidal system, Parkinsonism Relat.
Disord. 4 (1998) 53-57.
[4] D. Hirtz, D. Thurman, K. Gwinn-Hardy, M. Mohamed, A. Chaudhuri, R. Zalutsky, How common are
the “common” neurologic disorders?, Neurology 68(5) (2007) 326-337.
[5] E. Dorsey, R. Constantinescu, J. Thompson, K. Biglan, R. Holloway, K. Kieburtz, F. Marshall, B.
Ravina, G. Schifitto, A. Siderowf, Projected number of people with Parkinson disease in the most
populous nations, 2005 through 2030, Neurology 68 (2007) 384-386.
[6] J. Jankovic, Parkinson’s disease: clinical features and diagnosis, J. Neurol. Neurosurg. Psychiatry
79(4) (2008) 368-376.
[7] J. Parkinson, An essay on the shaking palsy, The Journal of neuropsychiatry and clinical
neurosciences 14(2) (2002) 223-236.
[8] A. Lees, E. Smith, Cognitive deficits in the early stages of Parkinson's disease, Brain 106 (1983) 257-
270.
[9] D.B. Calne, Treatment of Parkinson's disease, N. Engl. J. Med. 329(14) (1993) 1021-1027.
[10] P. Goodarzi, H.R. Aghayan, B. Larijani, M. Soleimani, A.-R. Dehpour, M. Sahebjam, F. Ghaderi, B.
Arjmand, Stem cell-based approach for the treatment of Parkinson's disease, Med. J. Islam. Repub.
Iran 29 (2015) 168.
[11] M. Politis, O. Lindvall, Clinical application of stem cell therapy in Parkinson's disease, BMC Med.
10(1) (2012) 1.
[12] G.C. Cotzias, P.S. Papavasiliou, R. Gellene, Modification of Parkinsonism—chronic treatment with
L-dopa, N. Engl. J. Med. 280(7) (1969) 337-345.
[13] S.H. Snyder, Adenosine as a neuromodulator, Annu. Rev. Neurosci. 8(1) (1985) 103-124.
[14] S. Ferré, P. Popoli, L. Gimenez-Llort, R. Rimondini, C. Müller, I. Strömberg, S.O. Ögren, K. Fuxe,
Adenosine/dopamine interaction: implications for the treatment of Parkinson's disease, Parkinsonism
Relat. Disord. 7(3) (2001) 235-241.

[15] C.V. Gomes, M.P. Kaster, A.R. Tomé, P.M. Agostinho, R.A. Cunha, Adenosine receptors and brain
diseases: neuroprotection and neurodegeneration, Biochimica et Biophysica Acta (BBA)-
Biomembranes 1808(5) (2011) 1380-1399.
[16] R.A. Cunha, Neuroprotection by adenosine in the brain: from A 1 receptor activation to A 2A
receptor blockade, Purinergic signalling 1(2) (2005) 111-134.
[17] B.B. Fredholm, A.P. IJzerman, K.A. Jacobson, K.-N. Klotz, J. Linden, International Union of
Pharmacology. XXV. Nomenclature and classification of adenosine receptors, Pharmacol. Rev. 53(4)
(2001) 527-552.
[18] S. Sachdeva, M. Gupta, Adenosine and its receptors as therapeutic targets: an overview, Saudi
Pharmaceutical Journal 21(3) (2013) 245-253.
[19] B.B. Fredholm, Adenosine receptors as drug targets, Exp. Cell Res. 316(8) (2010) 1284-1288.
[20] K. Xu, E. Bastia, M. Schwarzschild, Therapeutic potential of adenosine A2A receptor antagonists
in Parkinson's disease, Pharmacol. Ther. 105(3) (2005) 267-310.
[21] W. Bara-Jimenez, A. Sherzai, T. Dimitrova, A. Favit, F. Bibbiani, M. Gillespie, M. Morris, M.M.
Mouradian, T.N. Chase, Adenosine A2A receptor antagonist treatment of Parkinson’s disease,
Neurology 61(3) (2003) 293-296.
[22] M.T. Armentero, A. Pinna, S. Ferré, J.L. Lanciego, C.E. Müller, R. Franco, Past, present and future
of A2A adenosine receptor antagonists in the therapy of Parkinson's disease, Pharmacol. Ther. 132(3)
(2011) 280-299.
[23] M. Pierri, E. Vaudano, T. Sager, U. Englund, KW-6002 protects from MPTP induced dopaminergic
toxicity in the mouse, Neuropharmacology 48(4) (2005) 517-524.
[24] M. Hoffman, Istradefylline Approved for Parkinson Add-On Therapy, 2019.
https://www.neurologylive.com/clinical-focus/istradefylline-approved-as-nourianz-as-parkinson-
addon-therapy. (Accessed 3 December, 2019).
[25] T. Mihara, K. Mihara, J. Yarimizu, Y. Mitani, R. Matsuda, H. Yamamoto, S. Aoki, A. Akahane, A.
Iwashita, N. Matsuoka, Pharmacological characterization of a novel, potent adenosine A1 and A2A
receptor dual antagonist, 5-[5-amino-3-(4-fluorophenyl) pyrazin-2-yl]-1-isopropylpyridine-2 (1H)-one
(ASP5854), in models of Parkinson's disease and cognition, J. Pharmacol. Exp. Ther. 323(2) (2007) 708-
719.
[26] T. Maemoto, M. Tada, T. Mihara, N. Ueyama, H. Matsuoka, K. Harada, T. Yamaji, K. Shirakawa, S.
Kuroda, A. Akahane, Pharmacological characterization of FR194921, a new potent, selective, and
orally active antagonist for central adenosine A1 receptors, J. Pharmacol. Sci. 96(1) (2004) 42-52.
[27] O. Yuzlenko, K. Kiec-Kononowicz, Potent adenosine A1 and A2A receptors antagonists: recent
developments, Curr. Med. Chem. 13(30) (2006) 3609-3625.
[28] B.C. Shook, P.F. Jackson, Adenosine A2A receptor antagonists and Parkinson’s disease, ACS Chem.
Neurosci. 2(10) (2011) 555-567.
[29] M.M. Van der Walt, G. Terre'Blanche, Benzopyrone represents a privilege scaffold to identify
novel adenosine A1/A2A receptor antagonists, Bioorg. Chem. 77 (2018) 136-143.
[30] Z.A. Khan, N. Afzal, Z. Hussain, S.A.R. Naqvi, A. Bari, S.A. Shahzad, M. Yar, N. Mahmood, S.A.
Bukhari, A. Mansha, Synthesis of 2-Aryl-4H-3, 1-Benzoxazin-4-ones: A Class of [alpha]-Chymotrypsin
Inhibitors, Asian J. Chem. 26(15) (2014) 4561.
[31] K.R. Rao, R. Mekala, A. Raghunadh, S.B. Meruva, S.P. Kumar, D. Kalita, E. Laxminarayana, B.
Prasad, M. Pal, A catalyst-free rapid, practical and general synthesis of 2-substituted quinazolin-4 (3
H)-ones leading to luotonin B and E, bouchardatine and 8-norrutaecarpine, RSC Advances 5(76) (2015)
61575-61579.
[32] S.T. Asundaria, N.S. Patel, K.C. Patel, Synthesis, characterization, and antimicrobial studies of
novel 1, 3, 4-thiadiazolium-5-thiolates, Med. Chem. Res. 21(7) (2012) 1199-1206.
[33] M.W. Alnouri, S. Jepards, A. Casari, A.C. Schiedel, S. Hinz, C.E. Müller, Selectivity is species-
dependent: characterization of standard agonists and antagonists at human, rat, and mouse
adenosine receptors, Purinergic signalling 11(3) (2015) 389-407.

[34] T. Maemoto, K. Finlayson, H.J. Olverman, A. Akahane, R.W. Horton, S.P. Butcher, Species
differences in brain adenosine A1 receptor pharmacology revealed by use of xanthine and
pyrazolopyridine based antagonists, Br. J. Pharmacol. 122(6) (1997) 1202-1208.
[35] M.M. Van der Walt, G. Terre’Blanche, 1, 3, 7-Triethyl-substituted xanthines—possess nanomolar
affinity for the adenosine A1 receptor, Biorg. Med. Chem. 23(20) (2015) 6641-6649.
[36] M.M. Van der Walt, G. Terre'Blanche, Selected C8 two-chain linkers enhance the adenosine
A1/A2A receptor affinity and selectivity of caffeine, Eur. J. Med. Chem. 125 (2017) 652-656.
[37] R.F. Bruns, J.H. Fergus, E.W. Badger, J.A. Bristol, L.A. Santay, J.D. Hartman, S.J. Hays, C.C. Huang,
Binding of the A 1-selective adenosine antagonist 8-cyclopentyl-1, 3-dipropylxanthine to rat brain
membranes, Naunyn-Schmiedeberg's Arch. Pharmacol. 335(1) (1987) 59-63.
[38] E.M. Van der Wenden, H.R. Hartog-Witte, H.C. Roelen, J.K.v.F.D. Künzel, I.M. Pirovano, R.A.
Mathôt, M. Danhof, A. Van Aerschot, M.J. Lidaks, A.P. IJzerman, 8-Substituted adenosine and
theophylline-7-riboside analogues as potential partial agonists for the adenosine A1 receptor,
European Journal of Pharmacology: Molecular Pharmacology 290(3) (1995) 189-199.
[39] C.E. Muller, S. Ferré, Blocking striatal adenosine A2A receptors: a new strategy for basal ganglia
disorders, Recent patents on CNS drug discovery 2(1) (2007) 1-21.
[40] R.J. Gillespie, S.J. Bamford, S. Gaur, A.M. Jordan, J. Lerpiniere, H.L. Mansell, G.C. Stratton,
Antagonists of the human A2A receptor. Part 5: Highly bio-available pyrimidine-4-carboxamides,
Bioorg. Med. Chem. Lett. 19(10) (2009) 2664-2667.
[41] R.J. Gillespie, S.J. Bamford, A. Clay, S. Gaur, T. Haymes, P.S. Jackson, A.M. Jordan, B. Klenke, S.
Leonardi, J. Liu, Antagonists of the human A2A receptor. Part 6: Further optimization of pyrimidine-4-
carboxamides, Biorg. Med. Chem. 17(18) (2009) 6590-6605.
[42] M.M. Van der Walt, G. Terre’Blanche, A. Petzer, J.P. Petzer, The adenosine receptor affinities and
monoamine oxidase B inhibitory properties of sulfanylphthalimide analogues, Bioorg. Chem. 59 (2015)
117-123.
[43] M. Gütschow, M. Schlenk, J.r. Gäb, M. Paskaleva, M.W. Alnouri, S. Scolari, J. Iqbal, C.E. Müller,
Benzothiazinones: A novel class of adenosine receptor antagonists structurally unrelated to xanthine
and adenine derivatives, J. Med. Chem. 55(7) (2012) 3331-3341.
Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be
considered as potential competing interests:
We acknowledge the following person and institutions for their contributions. We are grateful to Dr.
J. Jordaan of the SASOL Centre for Chemistry, North-West University, for recording the NMR and
MS spectra of the synthesized compounds (The SASOL Centre for Chemistry was paid for its
service). Financial support for this work was provided by the North-West University and the National
Research Foundation of South Africa. Thus, no competing financial interests or personal
relationships to declare that could have influenced the work reported in this article.

benzo--pyrone
O
O
C
A
(1)
B
quinazolinone
O
quinazolinone
O
Replace O
Add N
A1Ki = 7.41 M
2AKi = 3.35 M
Replace O
Add N
A
NH
NH
C
A
C
A
OCH3
OCH3
N
(5e)
N
(5g)
B
CH3
A1Ki = 2.50 M
2AKi > 100 M
A1Ki > 100 M
2AKi = 2.81 M
A
A
Highlights:






The C2-substituted quinazolinones are ideal to identify novel AR antagonists.
3,4-OCH3-substitution on ring B of quinazolinone modulate A2A AR affinity.
4-CH3-substitution on ring B of quinazolinone modulate A1 AR affinity.
Quinazolinones display superior AR affinity over benzoxazinones.
A2A AR binding is best governed by ring C as the -pyrone.
Further structural modifications are necessary.