Monocyclic Quinone Structure-Activity Patterns: Synthesis of Catalytic Inhibitors of Topoisomerase II with Potent Antiproliferative Activity

Article abstract of DOI:10.1002/cmdc.201900548

The monocyclic 1,4-benzoquinone, HU-331, the direct oxidation product of cannabidiol, inhibits the catalytic activity of topoisomerase II but without inducing DNA strand breaks or generating free radicals, and unlike many fused-ring quinones exhibits minimal cardiotoxicity. Thus, monocyclic quinones have potential as anticancer agents, and investigation of the structural origins of their biological activity is warranted. New syntheses of cannabidiol and (±)-HU-331 are here reported. Integrated synthetic protocols afforded a wide range of polysubstituted resorcinol derivatives; many of the corresponding novel 2-hydroxy-1,4-benzoquinone derivatives are potent inhibitors of the catalytic activity of topoisomerase II, some more so than HU-331, whose monoterpene unit replaced by a 3-cycloalkyl unit conferred increased antiproliferative properties in cell lines with IC₅₀ values extending below 1 mM, and greater stability in solution than HU-331. The principal pharmacophore of quinones related to HU-331 was identified. Selected monocyclic quinones show potential for the development of new anticancer agents.

Full text of DOI:10.1002/cmdc.201900548

Full Papers
DOI: 10.1002/cmdc.201900548
1
2
3
4
5
6
7
8
9
Monocyclic Quinone Structure-Activity Patterns: Synthesis
of Catalytic Inhibitors of Topoisomerase II with Potent
Antiproliferative Activity
Thomas M. Waugh,[a] John Masters,[b] Abil E. Aliev,[a] and Charles M. Marson*[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
The monocyclic 1,4-benzoquinone, HU-331, the direct oxidation
product of cannabidiol, inhibits the catalytic activity of top-
oisomerase II but without inducing DNA strand breaks or
generating free radicals, and unlike many fused-ring quinones
exhibits minimal cardiotoxicity. Thus, monocyclic quinones have
potential as anticancer agents, and investigation of the
structural origins of their biological activity is warranted. New
syntheses of cannabidiol and (�)-HU-331 are here reported.
ing novel 2-hydroxy-1,4-benzoquinone derivatives are potent
inhibitors of the catalytic activity of topoisomerase II, some
more so than HU-331, whose monoterpene unit replaced by a
3-cycloalkyl unit conferred increased antiproliferative properties
in cell lines with IC50 values extending below 1 mM, and greater
stability in solution than HU-331. The principal pharmacophore
of quinones related to HU-331 was identified. Selected mono-
cyclic quinones show potential for the development of new
anticancer agents.
Integrated synthetic protocols afforded
a wide range of
polysubstituted resorcinol derivatives; many of the correspond-
Introduction
eviction from open chromatin structures, leading to activation
of the DNA damage response or apoptosis of cancer cells.[13]
A major therapeutic goal is the advancement of new
quinonoid compounds that display anti-neoplastic activity as
well as reduced toxicity. However, anthracycline usage is often
limited by cardiotoxicity,[4] which also applies to a number of
other classes of fused-ring quinones. In particular, chemo-
therapy using anthracyclines causes cardiomyocyte injury, left
ventricular dysfunction, and at the upper levels of dosage
required for cancer therapy carries a high risk of heart failure
and death.[1] Since such cardiotoxicity might arise through
undesired intercalation into normal DNA, or through non-
selective damage via quinonoid oxy radicals, we considered
whether monocyclic quinones (whose lack of ring fusion would
be expected to show little DNA intercalation) might confer
reduced toxicity compared to their fused-ring congeners, while
retaining anticancer potency. Indeed, a number of monocyclic
1,4-benzoquinone derivatives (Figure 1) show promise including
ardisianone (inducing antiproliferation and apoptosis of several
cancer cell lines, including PC-3, DU-145 and human hormone-
refractory prostate cancer),[14] irisoquin,[15] primin,[16] and HU-395,
a dihydro derivative of HU-331.[17] Embelin, another important
Quinones, especially derivatives of 1,4-benzoquinone, represent
an important group of potent therapeutic agents,[1–4] especially
as antibiotics and anticancer agents, including anthracyclines
such as doxorubicin and daunorubicin,[5] as well as other
streptomyces-derived compounds such as streptonigrin and
mitomycin C,[6] and synthetic quinonoids such as mitoxantrone
and epirubicin.[7,8] The mechanisms by which anthracyclines and
many fused-ring quinones act are complex, multiple pathways
usually being involved, including intercalation between base
pairs, resulting in the inhibition of RNA and DNA synthesis in
cancer cells.[9,10] Secondly, quinones may act as inhibitors of the
DNA topoisomerase II enzyme, preventing the relaxation of
supercoiled DNA, thereby blocking transcription and
replication.[11] Thirdly, they may generate haeme-mediated
oxygen radicals that damage DNA, proteins and cell
membranes,[12] although not necessarily selectively. Fourthly,
anthracyclines have recently been shown to cause histone
[a] Dr. T. M. Waugh, Dr. A. E. Aliev, Prof. C. M. Marson
Department of Chemistry
University College London
Christopher Ingold Laboratories
20 Gordon Street
London WC1H OAJ (UK)
E-mail: c.m.marson@ucl.ac.uk
[b] Prof. J. Masters
Prostate Cancer Research Centre
Research Department of Urology
Charles Bell House
University College London
67-75 Riding House Street
London W1 W 7EJ (UK)
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/cmdc.201900548
Figure 1. Some monocyclic 1,4-benzoquinone anticancer agents.
ChemMedChem 2019, 14, 1–12
1
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
monocyclic quinone natural product, is an analgesic which also
possesses antifertility, anti-inflammatory and antitumour
properties.[18]
potent derivatives than HU-331, preferably with improved
stability and other drug-like properties.
1
2
3
4
5
6
7
8
9
The monocyclic quinone HU-331 is of particular interest,[19–21]
since it showed no impairment of heart function,[20] in contrast Results and Discussion
to doxorubicin cardiotoxicity that was severe in a mouse model,
even at only one-third of the concentration of HU-331
administered. Quinone cardiotoxicity is generally thought to
arise by generation of reactive oxygen species (ROS), however,
in that respect, as well as others, HU-331 shows patterns of
biological activity that are atypical of many quinones. The lack
of cardiac toxicity upon administration of HU-331 is consistent
with findings that observed no production of ROS,[22] as well as
other studies which showed that ROS were produced, but were
not the cause of cell death.[23] Additionally, in various studies,
cell death induced by HU-331 could not be attributed to DNA
strand breaks, caspase activation, apoptosis, or cell cycle
arrest.[24] Although the main mode of action of HU-331 is
thought to be catalytic inhibition of DNA topoisomerase IIα,[20,24]
the unusual properties of HU-331 have not been fully
accounted for, especially in terms of its molecular structure;
despite being of cannabinoid origin, HU-331 does not target
cannabinoid receptors.[20] Consequently, of central importance
is to gain a detailed understanding of the biological and
potentially therapeutic effects as a function of substitution
patterns around the quinone ring; as shown in the present
study, the extent and nature of substitution can have profound
effects.
The close structural relationship (Scheme 3) of HU-331 to
cannabidiol (CBD), from which HU-331 can be obtained solely
by oxidation,[21] invites the question as to whether cannabinoids
can possess anticancer activity. Recently, cannabinoids have
been shown to inhibit tumour growth and angiogenesis, and to
induce apoptosis,[25–27] notably in hormone-sensitive prostate
cancers.[7] Cannabidiol can prevent metastasis of breast cancer,
possibly by suppressing the activity of the gene Id-1.[28] A recent
study concluded that 5-hydroxycannabidiol and HU-331 inhibit
both the α- and β-forms of topoisomerase II.[29] Δ9-Tetrahydro-
cannabinol inhibits the growth and metastasis of lung cancer
cell lines and prevents the growth of Lewis lung adenocarcino-
ma by inhibiting DNA synthesis.[25,30] Cannabinoids comprise a
diverse class of compounds that interact with a number of
molecular targets including the endocannabinoid-1 and À 2
receptors and the vanilloid receptor,[31] although their anti-
cancer mechanisms are not fully understood.
Chemistry
The initial objective was to devise a set of integrated synthetic
protocols to cover a wide range of polysubstituted resorcinol
derivatives, thereby affording by oxidation novel 2-hydroxy-1,4-
benzoquinone derivatives encompassing maximised chemical
space. Access to alkyl, cycloalkyl, cycloalkenyl and aryl sub-
stituents, among others, was required. The main synthetic
strategies to substituted 1,4-benzoquinones are depicted in
Scheme 1, in which the last step has in common the oxidation
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
of
a substituted resorcinol derivative 8 by Frémy’s salt
(dipotassium nitrosodisulfonate),[33] in an aqueous buffer of
KH2PO4,[34] giving the desired corresponding quinone 9, a means
of accessing the quinones which was found in the present
study to be both convenient and widely applicable.
Where feasible, Friedel-Crafts alkylation at the 2-position of
resorcinols 2 was the most succinct and hence the preferred route,
and was successful with various allylic cycloalkenols. To explore
the viability of the Friedel-Crafts route as a general strategy,
olivetol was reacted with 1-methylcyclohex-2-en-1-ol,[35] in the
presence of BF3 ·Et2O as previously described;[36] in our hands, only
a very low yield of 8p was obtained (Table 1, entry 1). Although a
freshly supplied sample of alumina (entry 2) afforded an improved
result, 8p was accompanied by its cyclised derivative (Table 1), as
an inseparable mixture. Increasing the equivalents of allylic alcohol
(1:1 reactants, entry 3) and lowering the amount of Lewis acid
gave a high yield of the undesired cyclised compound A which
was formed in lower yield when the reaction was carried out at
°
20 C (entry 4) but still as a 1:1 mixture of products. Evidently, the
basic alumina found in other cannabinoid studies to suppress
acidity,[36] and hence minimise subsequent cyclisation, did not
prove effective in this pilot reaction.
In the search for new anticancer agents containing a
quinone ring, cannabinoid quinones were considered to be a
suitable class since the structural motif affords in vitro and
in vivo anticancer potency, drug-likeness and, especially, as is
the case for HU-331, the potential for little or no cardiac toxicity
or myelotoxicity, but with inhibition of angiogenesis.[32] How-
ever, HU-331 is too chemically unstable, especially in solution,
to be used for cancer therapy. Consequently, aims of the
present studies were to identify the anticancer pharmacophore
latent in HU-331 through the study of structurally related
monocyclic quinones, and to synthesise and identify more
Scheme 1. Synthetic routes to monocyclic 2-hydroxy-1,4-benzoquinones.
°
°
Reagents and conditions: (a) BBr3 (3 equiv.), CH2Cl2, À 78 C, 10 min then 20 C,
°
°
1 h; (b) n-BuLi (1.1 equiv.), THF, 0 C then cyclic ketone, 0 C, 30 min then
°
20 C, 18 h; (c) TFA, CH2Cl2; (d) TFA, CH2Cl2 then Et3SiH; (e) Frémy’s salt (3.5–
10 equiv.), KH2PO4, aq. acetone; (f) cycloalkenol and CSA (0.1 equiv.), CH2Cl2
or BF3 ·Et2O, CH2Cl2; (g) n-BuLi, THF then alkyl iodide.
ChemMedChem 2019, 14, 1–12
www.chemmedchem.org
2
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
Table 1. Condensation of 1-methylcyclohex-2-en-1-ol with olivetol (1 equiv.) in CH2Cl2.[a]
1
2
3
4
5
6
7
°
T [ C]
Entry
Allylic
alcohol (eq.)
Reagent
Additive
Time[b]
8p [%]:
A [%]
8
9
1
2
3
4
5
0.8
0.8
1.0
1.0
1.5
BF3 ·Et2O
(2.5 eq.)
BF3 ·Et2O
(2.5 eq.)
BF3 ·Et2O
(1.6 eq.)
BF3 ·Et2O
(1.6 eq.)
CSA
Al2O3
(20 eq.)
reflux
reflux
reflux
20
10 s
<5:0
25:25
0:89
Al2O3
10 s
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
(20 eq.)[c]
Al2O3
10 s
(20 eq.)[c]
Al2O3
5 min
6h[d]
31:31
45:0
(20 eq.)
–
20
(0.1eq.)
1
[a] Percentage of conversion was determined from the H NMR spectra of the crude products. [b] For entries 1–3, a suspension of alumina in a solution of
BF3 ·Et2O in dry CH2Cl2 was stirred for 15 min, then heated at reflux for 1 min (entries 1 and 2) or 5 min (entry 3), and then the reactants added. After the
reaction time stated the mixture was quenched with saturated aqueous NaHCO3. [c] A new bottle of alumina was used. [d] 1-Methylcyclohex-2-en-1-ol was
added dropwise over 3 h via a syringe pump; then the mixture was stirred for an additional 3 h.
The overall inconsistency of these results using BF3 ·Et2O to
perform related condensations to give cannabinoids has also
been reported by Kassiou’s group,[37] although in their work
generally satisfactory results were obtained using only a 0.1%
solution of that reagent. In the event, a Brønsted acid approach
using 10-camphorsulfonic acid, (CSA),[38] proved much more
successful in the present work. In our parallel study of the
condensation of the isopropyl cyclohexenol 14 (1.2 equiv.) with
olivetol in the presence of CSA (0.1 equiv.) reactant stoichiom-
etry greatly influenced the outcome, an appreciable quantity of
olivetol remaining from an initial 1:1 stoichiometry, whereas 8a
was obtained in 35% yield using 1.2 equivalents of 14, and in
59% yield using 1.2 equivalents of 14, no olivetol being
detected in the latter case. When a 1:1.5 stoichiometry of 1-
methylcyclohex-2-en-1-ol: olivetol was used in the presence of
dimethoxy-5-methylbenzene with n-BuLi and alkylation with
MeI to give 1,3-dimethoxy-2,5-dimethylbenzene in 78%
yield,[39,40] followed by demethylation of the latter in 88% yield
using BBr3.
In order to obtain a full range of diversely substituted
resorcinols, several synthetic methods were required
(Scheme 2). Suzuki reaction of the boronic acid 10 with a
cycloalkenyl trifluoromethanesulfonate (‘triflate’) afforded the 5-
substituted cycloalkenyl derivatives 11 and 12 which were
reduced to the corresponding 5-substituted cycloalkyl deriva-
tives 1d and 1f using ammonium formate and 5% PdÀ C in
ethanol, and TFA-Et3SiH in CH2Cl2, respectively. Demethylation
of those and other substituted resorcinols was found to be
generally satisfactory using BBr3, using a procedure of addition
at À 78 C (stirred for 10 min) followed by warming to 20 C over
1 h.[41] The 5-(2-indenyl) resorcinol ether 16, prepared in 59%
yield by Suzuki coupling of boronic acid 10 with 2-bromo-1H-
indene, was reduced to the indanyl derivative 1e using
ammonium formate and 5% PdÀ C in ethanol.
°
°
°
CSA (0.1 equiv.) at 20 C no olivetol remained, and the required
cannabinoid adduct 8p was isolated in 45% yield (Table 1,
entry 5). This condensation protocol (General Procedure D) was
found to be effective in fourteen cases of this study, isolation
being usually straightforward, and cyclised by-products being
absent for reaction periods of 3–6 h. The resulting resorcinol
derivatives were oxidised with Frémy’s salt to give a focused set
of closely related congeners of HU-331 (Table 2).
In cases where an aromatic alkylation route (i.e. 2 to 8)
using a Brønsted acid or a Lewis acid was not efficient,
metalation of resorcinol derivatives 1 was generally used,
leading to the carbinols 4 which were dehydrated to the
cycloalkenyl derivatives 5 which were then reduced with
CF3CO2H-Et3SiH. More usually, and preferably, the sequential
dehydration-reduction of alcohols 4 was accomplished in one
overall step by addition of TFA followed by Et3SiH. The resulting
methyl ethers 7 were demethylated using BBr3 in dichloro-
methane to give the corresponding resorcinol derivatives 8
which were oxidised to the target quinones 9. The 2,5-dimeth-
ylresorcinol derivative 3 was prepared by metalation of 1,3-
5-Substituted resorcinols usually underwent Friedel-Crafts
alkylation mainly or exclusively at the 2-position, as might be
expected on grounds of a combination of electronic effects of
the two ortho-phenolic OH groups, together with steric
hindrance of approach to the 4- and 6-positions provided by
the 5-substituent. However, the only isolated product of the
condensation of resorcinol with 14 was the 4-substituted
derivative 8j (18%). Also, in the condensation of 13 with 14, an
appreciable quantity of the 4-substituted product 8l was
isolated. Those results are consistent with observations on the
condensation of methylcyclohex-2-en-1-ol with orcinol (5-meth-
ylresorcinol) in CH2Cl2 in the presence of BF3 ·Et2O, in which the
main product was that of alkylation at the 4-position.[42]
Conversely, from olivetol (larger C-5 substituent) the main
product was the result of solely 2-alkylation (e.g. Scheme 2).
ChemMedChem 2019, 14, 1–12
www.chemmedchem.org
3
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
1
2
3
4
5
6
7
8
9
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
Scheme 2. Regioselective synthesis of substituted resorcinol derivatives: Reagents and conditions: (a) For 1d: cycloalkenyl triflate (0.9 equiv.), Ph3P (1 mol%), Pd
(OAc)2 (1 mol%), K2CO3 (3 equiv.), 1:1 dimethoxyethane: water; (b) HCO2NH4 (10 equiv.), 5% PdÀ C, EtOH; (c) TFA (1.0 equiv.), CH2Cl2 then Et3SiH; (d) BBr3,
°
°
3 equiv., CH2Cl2, À 78 C, 10 min then 20 C,1 h; (e) 14 (1.5 equiv.), CSA (0.1 equiv.), CH2Cl2; (f) trimethylsilylacetylene, (Ph3P)2PdCl2 then TBAF, THF, 26% (2
steps); (g) H2, PdÀ C, MeOH, 59%; (h) PhB(OH)2, (Ph3P)4Pd, Na2CO3, aq., 1,2-dimethoxyethane, 69%.
With the principal reactions for the synthesis of HU-331
analogs now in hand, consideration was given to a total
synthesis of the cannabinoid HU-331. The natural product HU-
331 has usually been prepared by aerial oxidation of cannabi-
diol (CBD), obtained from plant sources.[21] A pure source of HU-
331 was required as a benchmark against which to measure
biological activity of the novel quinones here reported.
Accordingly, a synthetic route was sought, preferably using the
optimised Friedel-Crafts alkylation route described above. In our
hands, a direct approach to (�)-CBD by the condensation of
olivetol with isopiperitenol in the presence of CSA afforded an
inseparable mixture of (�)-cannabidiol (CBD) and (�)-Δ9-
Scheme 3. A convergent synthesis of (�)-cannabidiol and (�)-HU-331 via the
triol 20. Reagents and conditions: (a) 19 (1.5 equiv.), CSA (0.1 equiv.), CH2Cl2,
tetrahydrocannabidiol (THC), the latter formed by acid-catalysed
°
°
°
20 C, 3 h; (b) MsCl (6.0 equiv.), Et3N (10 equiv.), CH2Cl2; 0 C, 1 h, then 20 C,
cyclisation of the former. Instead, olivetol was condensed with
the diol 19 using the Friedel-Crafts conditions optimised as
above; pleasingly, the desired triol 20 was obtained in 73%
yield (Scheme 3). Although this triol has been reported to
cyclise to THC and its derivatives,[43] to the best of our
knowledge it has not been used to synthesise CBD. Initially,
protection of the phenolic OH groups by tosylation and also by
triflation, with concomitant elimination of the carbinol was
°
16 h; (c) MeLi (23 equiv.), THF, 0 C, 1 h; (d) Frémy’s salt (5.5 equiv.), KH2PO4
°
(2.3 equiv.), aq. acetone, 20 C, 16 h.
attempted, but in neither case could efficient O-deprotection
be achieved. However, mesylation of 20 in the presence of Et3N
also induced the desired elimination to the isopropenyl group,
ChemMedChem 2019, 14, 1–12
www.chemmedchem.org
4
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
and subsequent removal of the aromatic mesylate groups using
MeLi afforded (�)-CBD in satisfactory yield.[44] Notably, oxidation
of (�)-CBD using Frémy’s salt afforded (�)-HU-331 in 92% yield.
This use of Frémy’s salt compares very favourably with the yield
of 20% achieved in the early oxygenation of cannabidiol in
aqueous 5% KOH to give HU-331.[21]
1
2
3
4
5
6
7
8
9
Regioselectivity of the Friedel-Crafts condensation reactions
involving resorcinol and its derivatives was established by NMR
spectroscopy. For example, reaction of resorcinol at the 4-
1
position giving 9j was established from the singlet in the H
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
NMR spectrum for the isolated C-3 hydrogen atom at δ 6.09
and the four-bond coupling constant of 0.7 Hz for the C-6
hydrogen atom at δ 6.57. Consideration of the number of lines
in the 13C NMR spectra of the resorcinol derivatives 8 was also
often diagnostic of the position of condensation (i.e. C2- versus
C-4).
Figure 2. Comparison of the effect of commercial versus synthetic samples
of HU-331 pre-incubated with human topoisomerase II for 10 minutes
followed by addition of supercoiled plasmid DNA.
1
In the H NMR spectrum of the quinones 9 the trans-diaxial
bromide fluorescence could be accurately measured using
ImageJ software (Figure 3).
coupling constant of approximately 11.5–12.0 Hz confirmed the
trans configuration of the isopropyl (or isopropenyl) substituent
with the 1,4-benzoquinonyl group. Although the relative
configuration of the methyl group in 8h and 9h could not be
deduced from NMR data owing to overlapping multiplets in the
relevant regions, only one diastereoisomer was detected, and
by analogy with the hydroboration-oxidation of (+)-2-carene to
give (À )-2-isocaranol (92%),[45] hydroboration on the face
opposite to the bulky isopropyl group would be expected,
affording the triequatorially substituted stereoisomer 8h, and
hence 9h.
Quinone Structure-Activity Relationship Based on
Topoisomerase IIα Inhibition
In our hands, synthetic racemic HU-331 and commercial HU-331
were also found to be equipotent in the topoisomerase IIα DNA
relaxation assay (Table 2). Additionally, enantiomers of HU-331
have been reported to be equipotent against a variety of cancer
cell lines.[20] As with a set of bisdioxopiperazine catalytic
inhibitors of topoisomerase II (i.e. inhibitors of one or more
steps in the enzyme-DNA ligation/de-ligation/strand cleavage
cycle) whose enantiomers showed equipotency in respect of
both enzyme inhibition and cellular cytotoxicity,[46] a similar
explanation can be advanced for the quinone HU-331, namely
that the site of binding to topoisomerase IIα is either sufficiently
large or sufficiently flexible to accommodate each enantiomer
equally well.
Of the trisubstituted HU-331 analogs 9a and 9c–9h, in
which the terpenoid unit had been retained (or hydrogenated
in the cases of 9a and 9h) each of this series of compounds
showed some inhibition, but none greater than or equal to HU-
331 (Table 2). Saturation of the isopropenyl group of HU-331
lowered the inhibition (61% for 9a, versus 86% for HU-331).
Complete saturation of the terpenoid unit lowered the inhib-
ition further (53% for 9h). However, removal of the isopropenyl
In vitro Inhibition of Topoisomerase II by Monocyclic
Quinones
Several studies have confirmed that the monocyclic quinone
HU-331 inhibits the relaxation activity of topoisomerase
II,[19,20,24,46] and is highly selective for this isozyme. Both the
commercial HU-331 and our synthetic samples of HU-331
(concentration range tested from 0.3-100 mM) were found not
to increase the amount of nicked DNA in a topoisomerase I
DNA cleavage assay, (data not shown) in accord with previous
studies.[20] Equally, no evidence was found for the intercalation
of HU-331 (concentration range tested from 0.03–100 mM), as
assessed by a DNA intercalation assay using ethidium bromide
as the control. For the topoisomerase II DNA relaxation assay
(Figures 2–4) a reliable system involved preincubation of the
test compounds with human topoisomerase II (Affymetrix) for
10 min, and a relaxation buffer containing ATP (10 mM) and
MgCl2 (0.1 M), giving full inhibition between 27–270 mM and
some inhibition at 0.27 mM. First, under those conditions, the
same concentrations of commercial HU-331 and synthetic (�)-
HU-331 were shown to perform in the same way in the
topoisomerase IIα assay (Figure 2). Secondly, a suitable bench-
mark concentration of 80 mM was established for the evalua-
tion of the series of monocyclic quinones (Table 2). At that
concentration, significant amounts of both supercoiled plasmid
DNA and relaxed topoisomeric DNA were observed in the
histograms and the corresponding pixel intensities of ethidium
Figure 3. Human topoisomerase II-induced relaxation of supercoiled plasmid
DNA in the presence of a range of concentrations of synthetic HU-331 (pre-
incubated with the enzyme for 10 minutes).
ChemMedChem 2019, 14, 1–12
www.chemmedchem.org
5
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
Table 2. Inhibition of Topoisomerase II by Substituted Monocyclic Quinones.[a,b]
1
2
3
4
5
6
7
8
9
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
[a] Inhibition at 80 μM concentration of test compound. [b] Average of triplicate runs�SEM.
group in HU-331 resulted in only a small loss of inhibition (79%
for 9p), suggesting that the isopropenyl group is not essential
for potent inhibition of topoisomerase IIα. Although the fully
reduced terpenoid unit in 9h showed the lowest potency
within the series of terpenoid derivatives studied, this com-
pound illustrated that a 2-cyclohexyl ring can be compatible
with good inhibition. Following this observation, the subset of
saturated cycloalkyl quinones 9m–9o and 9q was synthesised;
not only was good-to-excellent inhibition of topoisomerase IIα
observed, but in the set of 6-n-pentyl quinones synthesised, 9o
ChemMedChem 2019, 14, 1–12
www.chemmedchem.org
6
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
was the only crystalline compound, and also showed much
of the quinone with the N-terminal domain, thereby preventing
the N-terminal clamp from capturing DNA.[24,29,48] The fully
substituted quinone ring of 9b retards or prevents adduct
formation.
1
2
3
4
5
6
7
8
9
°
greater stability when kept as a solution in DMSO-d6 at 20 C,
with little decomposition after 8 days (90% of initial concen-
tration), in contrast to HU-331 which had almost completely
decomposed (10% of initial concentration, Figure 6, Supporting
Information); indeed, after 2 days, more than 50% of HU-331
had decomposed.
The above survey of topoisomerase IIα inhibition as a
function of variations in substitution suggested further inves-
tigation of 3-cycloalkyl derivatives of 2-hydroxy-6-n-pentyl-1,4-
benzoquinone, particularly since 9o had been shown to have
much greater stability than HU-331. Initially, and to identify a
reference for acyclic substituents, quinone 9s was prepared and
evaluated; its considerable potency (58%) was diminished upon
O-methylation (31% for 9t), confirming the need for the
unsubstituted hydroxy group. Introduction of a cyclohexyl
group at C-3 afforded the most potent compound in the
quinone series, showing 97% inhibition at 80 μM, and hence
that an n-pentyl chain is not a requirement for potent inhibition
of topoisomerase IIα. Variation of the ring size of the C-3
cycloalkyl substituent showed that cyclopentyl conferred the
greatest enzyme inhibition (95% for 9n), being more potent
than HU-331 (86%). The cycloheptyl congener 9q was the least
potent (49% inhibition) of the set 5m–9r. The cyclohexenyl
derivative 9p (78% inhibition) was not significantly more
potent than the 3-cycloalkyl quinone 9o (76%).
The cyclohexyl ring present in the corresponding com-
pound 9o confers markedly greater stability as well as
crystallinity, an observation which helped focus the search for
potency a set of quinones 9 that contained a simplified and
saturated 3-cycloalkyl unit, compared to the terpenoid unit
present in HU-331. Since compounds 9n, 9o and 9u were more
potent inhibitors of topoisomerase IIα than HU-331, as well as
being considerably more stable, they were tested against a
number of cancer cell lines to determine whether the potent
enzyme assay results would be accompanied by antiprolifera-
tion in whole cell assays.
Within the terpenoid series described above, variation of
the 6-substituent was examined; conformational restriction by
annulation to give a five-membered ring (9d) or six-membered
ring (9f) removed all significant inhibition, as did replacement
of the 6-n-pentyl chain by a 2-indanyl group (9e) or an n-
butyloxy group (9i). In the latter case, electronic factors may
play a role by lowering the electrophilicity of the quinone ring,
and in the former cases, increased steric hindrance appears to
be the likely contributor. In contrast, a 6-phenyl substituent
retained considerable inhibitory potency (72% for 9g), as also
did an unadorned 3-cyclohexyl substituent (62% for 9r),
perhaps because of the planarity afforded by the extended sp2-
character of the 6-phenyl substituent whose planarity may
permit π-stacking with a nearby aromatic amino acid residue.
Within the C-5 substituted congeners, the tetrasubstituted
quinones 9b and 9c showed no significant inhibition, consis-
tent with a proposed mechanism involving conjugate addition
of a topoisomerase II residue to an unsubstituted position on
the quinone ring;[47,48] residues Cys392 and Cys405 have been
identified as significant for addition to p-benzoquinone.[48] In
contrast, quinones 9k and 9l showed moderate and good
potency, respectively, compared with HU-331. If the novel
topoisomerase IIα inhibitors 9j, 9k and 9l, which all possess a
5-substituent, act in a way similar to the parent compound HU-
331, conjugate addition presumably occurs at C-3. This require-
ment for an unsubstituted (Michael acceptor) site on the
quinone ring was further probed by examining the effect of 9d
and 9b at 200 μM on topoisomerase IIα (Figure 4); at this
higher concentration, 9d showed complete inhibition, but
without the formation of observable topoisomers. Conversely,
the tetrasubstituted quinone 9b showed no inhibition. Those
observations are consistent with covalent adduct formation
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
Cytotoxic Effects of Monocyclic Quinones
Inhibitory activity against proliferation of the prostate cancer
cell line DU-145 was measured using an XTT assay with
continuous exposure for 72 h (Table 3); compared to HU-331,
four compounds were considerably more potent inhibitors. Of
the compounds assayed that contain a terpenoid-like substitu-
ent, the isopropyl derivative 9a and the corresponding system
9h with a saturated terpenoid system were of similar potency,
both being more potent than HU-331, implying that a more
saturated terpenoid skeleton can be an improvement. Com-
pared to 9a, conformational restriction of the 6-substituent to
cyclopentyl, as in 9d, further increased the inhibitory potency
two-fold. Of similar or greater potency against DU-145 cell
proliferation was the 3-cyclohexyl derivative 9o. The other
derivatives showed medium to low inhibitory activity, despite
9p and 9u showing, respectively, similar and greater potency
than HU-331 in the topoisomerase IIα assay. This can be
accounted for by a lack of sufficient lipophilicity in the 6-
substituent (9s–9w), although that is partly compensated for
by the 3-cyclohexyl substituent in 9u. That 9w was by far the
being
a principal mode by which monocyclic quinones
including HU-331 can act on human topoisomerase IIα,[24,48,49]
and which recently has been proposed to involve interference
Figure 4. Human topoisomerase II-induced relaxation of supercoiled plasmid
DNA in the presence of (a) the trisubstituted quinone 9d and (b) the
tetrasubstituted quinone 9b.
ChemMedChem 2019, 14, 1–12
www.chemmedchem.org
7
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
Table 3. Antiproliferation IC50 [μM] data monocyclic quinones.[a]
more potent than HU-331), but which was less potent than HU-
331, 9a, 9d, 9h and 9o in the DU-145 cell line assay. Fourthly,
the results of the topoisomerase IIα assay showed that
decreasing the 3-cycloalkyl ring size from 7 to 6 to 5 increased
potency (9n having the second-highest inhibition of 95% at
80 μM of all compounds tested in the enzyme assay).
Conversely, in the cell line assays, decreasing the ring size was
detrimental to inhibition of proliferation, the 3-cyclopentyl
derivative 9n exhibiting over 6-fold less potency than the 3-
cyclohexyl derivative 9o. Thus, while the greater requirement
for lipophilicity in whole cell assays compared to an in vitro
enzyme assay must be considered, and is expected on account
of cell membrane permeability, overall the relative potency of
whole cell inhibitory activity still shows some significant differ-
ences with the relative potency of the topoisomerase IIα assay,
and may indicate that inhibition of topoisomerase IIα (e.g. for
9o) is not the only relevant molecular target;[50] further studies
are indicated.
1
2
3
4
5
6
7
8
9
Cell line: Compound
DU-145
Jurkat
Raji
(�)-HU-331
9.15�0.57
5.45�0.26
2.4�0.60
5.2�0.20
13.7�0.60
2.05�0.10
20.1�4.4
27.8�4.3
40.6�2.9
12.8�0.52
19.0�0.23
57.0�15.5
1.30�0.20[20]
0.61�0.05[21]
9a
9d
9h
9n
9o
9p
9s
ND[b]
ND
ND
ND
2.37�0.10
0.29�0.02
ND
ND
ND
ND
ND
5.27�0.25
0.68�0.03
ND
ND
ND
9t
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
9u
9v
9w
0.42�0.02
1.45�0.06
ND
ND
ND
ND
[a] Data from cell viability (ATP) assay for compounds 9n, 9o and 9u; all
other data for DU-145 from XTT assay. Average of triplicate runs�SEM. [b]
ND=not determined.
least active compound evaluated against DU-145 cell line
proliferation is consistent with the absence of any lipophilic
group at both C-3 and C-6 (hence very low potency), despite
the presence at C-6 in 9w of the complete terpenoid group Conclusions
that is found in HU-331. The main factor accounting for the
difference in cellular inhibitory potency is probably the difficulty
of the much more polar 9w in traversing the DU-145 cell
membrane, as expected from the markedly different lipophilic-
ities (HU-331 cLogP=6.14 versus 9w cLogP=3.51). However,
the lower potency of 2-methoxy compounds compared to their
2-hydroxy analogs, for both enzyme inhibition and DU-145 cell
line antiproliferation cannot be accounted for in terms of
lipophilicities (9s cLogP=0.99 and 9t cLogP=1.15), and is
ascribed to the hydroxy group acting as a hydrogen bond
donor.
A selection of the more potent compounds was then
evaluated for inhibition of proliferation of Jurkat human acute T
cell leukaemia and Raji human lymphoma cell lines (Table 3).
Comparison of 9n with 9o shows that 3-cyclohexyl ring confers
nearly an eight-fold greater inhibitory activity across the three
cell lines studied than a 3-cyclopentyl ring. The 3-cyclohexyl
derivative 9u also showed good-to-excellent potency. The most
potent compound was identified as 9o, showing significantly
improved potency in two of the three cell lines studied,
including sub-micromolar potency in two of the cell lines.
Simplification of chemical structure by removal of metabol-
ically active unsaturation led to greater overall potency than
HU-331 as well as greatly increased in vitro stability.
Some differences were observed in the relative inhibition of
topoisomerase IIα versus inhibition of cell line proliferation. For
example, a terpenoid isopropenyl group (HU-331, 86%) con-
ferred greater potency than the corresponding isopropyl group
(9a, 61%), whereas in the DU-145 cell line assay the isopropenyl
group in HU-331, 86%) conferred 1.7-fold less potency than the
isopropyl group in 9a. Also, absence of the isopropenyl group
in HU-331 (86% inhibition of topoisomerase IIα) gave only a
small loss in enzyme inhibition (79% for 9p), whereas
compared to HU-331 inhibition of DU-145 cell proliferation by
9p was more than halved. A third difference was seen for 9u,
the most potent inhibitor of all the novel quinones tested (and
New syntheses of cannabidiol and (�)-HU-331 are reported
herein. General synthetic approaches to a wide variety of
substituted derivatives of 1,4-benzoquinone have been de-
scribed, the shortest employing a Friedel-Crafts alkylation of a
resorcinol derivative followed by oxidation to the quinone using
Frémy’s salt. New derivatives of 1,4-benzoquinone have been
identified as more potent than HU-331, a potent natural
product anticancer agent showing no significant cardiotoxicity.
Novel potent inhibitors of human topoisomerase IIα and of
proliferating cancer cell lines were identified, including 9a, 9d,
9h and especially 9o which showed a greater than four-fold
inhibition of the DU-145 prostate cancer cell line than HU-331,
as well as sub-micromolar inhibitory potency against both
Jurkat human acute T cell leukaemia and Raji human lymphoma
cell lines. The simplified analogs, including 9o, showed much
greater in vitro stability compared to HU-331, ascribed to the
absence of chemically reactive groups (which in HU-331 are
likely sites of metabolism).
Anticancer properties here identified were correlated with
structure (Figure 5), which indicated clear directions for further
optimisation, especially variation in 6-alkyl substituents; 6-aryl
and 3-(4-substituted)-cyclohexyl derivatives may also be prom-
Figure 5. Correlation of substituted hydroxy-1,4-benzoquinones with anti-
cancer properties identified.
ChemMedChem 2019, 14, 1–12
www.chemmedchem.org
8
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
evaporated. The residue was purified by column chromatography,
typically using 10–30% by volume of ethyl acetate in hexane as the
eluent to give the resorcinol derivative (details provided for each
compound in Supporting Information).
ising areas of chemical space. This preliminary structure–activity
profile identified for these quinones should facilitate the
discovery of new, drug-like monocyclic quinone anticancer
agents with low cardiotoxicity (partly attributable to minimal
ROS production), unlike many fused quinone systems, especially
the clinically important anthracyclines whose use is constrained
by severe and dose-limiting cardiotoxicity.
Although inhibition of human topoisomerase IIα is a
confirmed and likely major target, several observations indicate
that some monocyclic quinone inhibitors act on at least one
other molecular target, not revealed by the present study.
Based on broad-spectrum activity against multiple cell lines, its
crystallinity and its greatly increased stability compared to HU-
331, compound 9o is suitable as a probe to elucidate uncharted
quinone chemical biology, and also as the basis for next-
generation antiproliferative hydroxyquinones possessing mini-
mal cardiotoxicity compared to existing clinical quinone
anticancer agents.
1
2
3
4
5
6
7
8
9
General procedure
B (metalation of a 1,3-dimethoxyarene,
addition to a ketone and dehydration of the carbinol): To a
stirring solution of the 1,3-dimethoxy-5-substituted benzene (0.2 M,
°
1.0 equiv., approx. 1.5 mmol scale) in dry tetrahydrofuran at 0 C
under nitrogen was added n-butyllithium (1.6 M, 1.1 equiv.) drop-
°
wise over 5 min. Stirring was continued at 0 C for a further 1 h. A
solution of the cyclic ketone (1.8 M, 1.5 mol equiv.) in dry
tetrahydrofuran was then added dropwise over 10 min. The mixture
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
°
°
was stirred for an additional 30 min at 0 C, then at 20 C for 18 h.
Saturated aqueous ammonium chloride (20 mL) was then added.
The mixture was then extracted with diethyl ether (2×20 mL) and
the combined organic layers were washed with water, dried
(Na2SO4), filtered and the solvent was evaporated. The residue was
purified by column chromatography, typically using 10–25% by
volume of ethyl acetate in hexane as the eluent to give the tertiary
alcohol (details provided for each compound in Supporting
Information).
General procedure C (reductive elimination of carbinols, and
reduction of alkenes to the corresponding aryl cycloalkanes): To
a stirring solution of the tertiary alcohol or the alkene (0.36 M,
Experimental Section
°
1.0 equiv.) in dichloromethane at 20 C was added trifluoroacetic
acid (5.5 equiv.), whereupon the colourless solution immediately
Instruments and Methods
°
turned dark red. Stirring was continued at 20 C for 15 min, then
Reactants and reagents were used as received. Commercial HU-331
was purchased from Abcam. Compound homogeneity was moni-
tored by ascending thin-layer chromatography, performed on
Merck 0.2 mm aluminium-backed silica gel 60 F254 plates and
visualised using ultraviolet irradiation (254 nm) or by using aqueous
alkaline potassium permanganate. Flash column chromatography
was performed using Merck silica gel 60 of particle size 33–70 mm.
Evaporation refers to the removal of solvent under reduced
pressure. Melting points were determined using an Optimet
apparatus and are uncorrected. Infrared (IR) spectra were recorded
triethylsilane (2.5 mol equiv.) was added dropwise over 5 min, and
the mixture stirred a further 1.5 h. Saturated aqueous sodium
hydrogen carbonate (20 mL) was then added. The organic layer
was separated, washed with water (10 mL), dried (Na2SO4), filtered
and the solvent was evaporated. The residue was purified by
column chromatography, typically using 0–5% by volume of ethyl
acetate in hexane as the eluent to give the aryl cycloalkane (details
provided for each compound in Supporting Information).
General procedure D (condensation of resorcinol derivatives with
cyclic allylic alcohols): To a stirring solution of the resorcinol
derivative (0.20 M, 1.0 equiv., approx. 0.5 mmol scale) and camphor-
sulfonic acid (0.10 mol equiv.) in dichloromethane was added
dropwise a solution of the cyclic allylic alcohol (0.20 M, 1.5 equiv.)
1
on a Bruker alpha FT-IR spectrometer using neat samples. H and
13C NMR spectra (respective operating frequencies of 500 MHz and
125 MHz) were obtained using a Bruker AM-500 spectrometer in a
variety of solvents and a residual nondeuterated solvent peak as
the internal reference: CDCl3 (δH =7.26 ppm, δC =77.0 ppm), DMSO
(δH =2.50 ppm, δC =39.5 ppm) CD3OD (δH =3.31 ppm, δC =
49.0 ppm). Mass spectra were obtained on a Fisons VG70-SE mass
spectrometer or Thermo Finnigan MAT900xp instrument using
electrospray ionisation (ESI), electron ionisation (EI) or chemical
ionisation (CI).
°
in dichloromethane via a syringe pump at 20 C over 3 h. The
°
solution was stirred for a further 3 h at 20 C then saturated
aqueous sodium hydrogen carbonate (10 mL) was added. The
mixture was extracted with dichloromethane (3×20 mL) and the
combined organic layers were washed with water, then with brine.
The combined organic layers were dried (Na2SO4), filtered and the
solvent was evaporated. The residue was purified by column
chromatography (details provided in Supporting Information:
usually using 5% by volume of ethyl acetate in hexane for
cannabinoids) to give the cycloalkenyl resorcinol derivative.
Supporting Information
Preparation and spectral data of compounds are given in the
Supporting Information.
Representative general procedure E (oxidative dearomatisation
of resorcinol derivatives using Frémy’s salt): The detailed
procedures and quantities given in the Supporting Information for
specific reactions should be used, especially the quantity of Frémy’s
salt. To a stirring solution of the resorcinol derivative (0.10 mmol) in
General procedure A (demethylation of 1,3-dimethoxyarenes): To
a stirring solution of the 1,3-dimethoxyarene (0.27 M, 1.0 equiv.,
approx. 0.4 mmol scale) in dichloromethane under nitrogen and
cooled in a dry ice-acetone bath was added dropwise boron
°
acetone (5.0 mL) at 20 C was added dropwise over 2 min a pre-
°
tribromide (3.0 equiv.) at À 78 C over 2 min. Stirring was continued
mixed buffered aqueous solution containing potassium dihydrogen
orthophosphate (0.24 mmol, 0.060 M, 4.0 mL) and Frémy’s salt
(dipotassium nitrosodisulfonate, 0.268 g, 1.0 mmol; CAUTION!). Only
small quantities of Frémy’s salt were used, and under the conditions
described did not lead to a rapid exotherm. However, the use of
Frémy’s salt represents an explosion hazard, and suitable protection
°
at À 78 C for a further 10 min. The dry ice-acetone bath was then
°
removed and the mixture allowed to warm to 20 C over 1 h.
Saturated aqueous sodium hydrogen carbonate (20 mL) was then
added and the mixture was stirred at 20 C for a further 10 min.
The mixture was extracted with diethyl ether (3×20 mL) and the
combined organic layers were washed with saturated aqueous
sodium hydrogen carbonate then with brine. The combined
organic layers were dried (MgSO4), filtered and the solvent was
[41]
°
°
during handling and use should be taken. After stirring at 20 C
until the reaction was complete by TLC (typically 3–4 h), the
ChemMedChem 2019, 14, 1–12
www.chemmedchem.org
9
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
mixture was extracted with diethyl ether (3×10 mL). The combined
organic layers were dried (MgSO4), filtered and evaporated. In most
cases the hydroxyquinone product was of high purity, as assessed
by 1H and 13C NMR spectroscopy and TLC, but could be further
purified by column chromatography, typically using 2%–10% by
volume of ethyl acetate in hexane.
(pH 7.9), 500 mM KCl, 500 mM NaCl, 50 mM MgCl2, 1 mM EDTA,
0.15 mg/mL BSA and 10 mM ATP) diluted tenfold. To this mixture
was added the diluted inhibitor to give a final concentration of
80 μM. The mixtures were incubated for at room temperature for
10 min then the supercoiled DNA substrate (0.3 μg) was added,
making the final volume up to 30 μL. The mixture was then
1
2
3
4
5
6
7
8
9
°
incubated at 37 C for 30 min prior to reaction termination with the
stopping buffers SDS and proteinase K to give a final concentration
of 0.25% and 250 μg/mL respectively. The mixture was then
analysed on 1% agarose gel in a 10% Tris-borate EDTA buffer. The
different forms of DNA were then separated by electrophoresis at
150 V conducted at room temperature for approximately 2 h. Gels
were visualised by staining with ethidium bromide (0.5 μg/mL) for
30 min then de-staining in water for 15 min. The gels were
photographed under UV irradiation using a UV transilluminator
coupled with a camera. For quantitation, the integrated pixel
intensities of the ethidium bromide fluorescence were determined
using ImageJ software. Background fluorescence (mainly from open
circular ‘nicked’ DNA) was subtracted from all fluorescence values
to give normalised intensities, the percentage inhibition being
given by 100–100 x (normalised intensity for relaxed DNA formed).
Topoisomerase I DNA Cleavage Assay
Human recombinant topoisomerase I (Affymetrix) was supplied in a
storage buffer (15 mM sodium phosphate, pH 7.1, 700 mM NaCl,
0.1 mM EDTA, 0.5 mM DTT, 50% glycerol) at a concentration of
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
°
20 units/μL and was stored at À 20 C. Inhibitor solutions were
prepared by serial dilution of a 10 mM stock solution in DMSO with
relaxation buffer (20 mM Tris-HCl pH 7.5, 200 mM NaCl, 0.25 mM
EDTA, 5% glycerol). To each Eppendorf tube was added supercoiled
plasmid DNA (pBR322, 0.3 μg) followed by different concentrations
of HU-331 (0.3 μM to 1 mM). Human recombinant topoisomerase I
(2 U) was added to each mixture, making the final volume up to
°
30 μL. The mixtures were incubated together at 37 C for 45 min.
The reactions were then terminated by addition of the stopping
buffers SDS and proteinase K to a final concentration of 0.25% and
250 μg/mL respectively. The reaction mixtures were incubated at
Cell Proliferation XTT Assay
°
50 C for 30 min then analysed by electrophoresis on 1% agarose
gel with 0.5 μg/mL ethidium bromide in a 10% Tris-borate EDTA
buffer. The different forms of DNA were then separated by
electrophoresis at 90–140 V conducted at room temperature for
approximately 2 h. Gels were photographed under UV irradiation
using a UV transilluminator coupled with a camera.
A procedure for an MTT assay was used with the following
adaptations.[53] DU-145 cells were suspended in RPMI 1640 medium,
supplemented with fetal calf serum (40 mL) and L-glutamine (5 mL)
°
at 37 C in a humidified atmosphere of 5% CO2 and 95% air.
Aliquots (200 μL) of suspensions of cancer cells were dispensed into
wells of 96-well tissue culture plates at densities of 4000 cells/well.
After incubation for 24 h various concentrations of the 2-hydroxy-
1,4-benzoquinone derivative were added and cell viability deter-
mined after 72 h using 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-
2H-tetrazolium-5-carboxanilide (XTT) in the presence of phenazine
methosulfate (PMS). Absorbance was measured against a back-
ground control as a blank (0.5% DMSO in growth medium) using a
scanning multiwell spectrophotometer (Wallac Victor 1420 multi-
label counter) at 450 nm. IC50 values for various compounds were
then calculated using OriginPro 9 by plotting the Log of the
concentrations against the mean percentage inhibition, and using a
nonlinear curve fit algorithm.
DNA Intercalation Assay
A literature procedure, reported by Peixoto and co-workers,[51] was
used to investigate the intercalation activity of HU-331 compared
to ethidium bromide based on the topoisomerase I relaxation assay.
Inhibitor concentrations were made by serial dilution of stock
solutions (10 mg/mL for ethidium bromide and 10 mM for HU-331)
with topoisomerase I relaxation buffer (20 mM Tris-HCl pH 7.5,
200 mM NaCl, 0.25 mM EDTA, 5% glycerol). To each Eppendorf
tube was added supercoiled plasmid DNA (pBR322, 0.3 μg) followed
by different concentrations of ethidium bromide (0.03–3 μM) and
HU-331 (0.03–100 μM). Human recombinant topoisomerase I (5 U)
was added to each mixture, and the final volume made up to 30 μL
using dilution buffer. The mixtures were incubated together at
Cell Viability Assay
°
37 C for 45 min. Reactions were then terminated by addition of the
Cell lines were purchased from American Type Culture Collection
(Manassas, VA). Cells of the DU-145 prostate cancer cell line were
grown in EMEM medium, whereas those from Raji human
lymphoma and Jurkat human acute T cell leukaemia cell line were
grown in RPMI-1640 medium. All culture media were supplemented
with 10% fetal bovine serum, penicillin (100 μg/mL) and streptomy-
stopping buffers SDS and proteinase K to a final concentration of
0.25% and 250 μg/mL respectively. The reaction mixtures were
°
incubated at 50 C for 30 min then analysed by electrophoresis on
1% agarose gel in a 10% Tris-borate EDTA buffer. The different
forms of DNA were then separated by electrophoresis at 150 V
conducted at room temperature for approximately 2 h. The gels
were visualised by staining with ethidium bromide (0.5 μg/mL) for
30 min then de-staining in water for 15 min. The gels were
photographed under UV irradiation using a UV transilluminator
coupled with a camera.
°
cin (100 μg/mL). Cultures were maintained at 37 C in a humidified
atmosphere of 5% CO2 and 95% air.
Test compounds and the reference compound staurosporine
(Sigma-Aldrich) were prepared in DMSO as 10 mM stock solutions
which then were diluted with 10-dose and three-fold dilutions in a
source plate. A solution of each test compound (0.25 mL) or of
staurosporine (0.025 mL) was delivered from the source plate to
each well of the 384-well cell culture plates by Echo 550. Culture
medium (25 μL) containing one thousand cells from the DU-145
prostate cancer cell line, or the Raji human lymphoma cell line, or
the Jurkat human acute T cell leukaemia cell line was added to the
wells of the cell culture plates. The cells were incubated with the
Topoisomerase II Relaxation Assay
The procedure was similar to the topoisomerase I relaxation assay
but with slight modifications.[52] Each inhibitor was tested at one
concentration (80 μM) by diluting the stock concentrations with
topoisomerase II dilution buffer (10 mM sodium phosphate pH 7.1,
50 mM NaCl, 0.2 mM DTT, 0.1 mM EDTA, 10% glycerol and 0.5 mg/
mL BSA. To each Eppendorf tube was added human recombinant
topoisomerase II (2 U) followed by reaction buffer (100 mM Tris-HCl
°
compounds at 37 C for 72 h under an atmosphere of 5% CO2 and
95% air. To each well was added Cell Titer Glo 2.0 reagent (25 μL)
(Promega, Madison, WI), then the contents were mixed on an
ChemMedChem 2019, 14, 1–12
www.chemmedchem.org
10
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
[18] M. Chitra, E. Sukumar, V. Suja, C. S. Devi, Chemotherapy 1994, 40, 109–
113.
orbital shaker for 2 min prior to incubation at room temperature for
10 min to stabilise the luminescent signal. Luminescence was
recorded by an Envision 2104 Multilabel Reader (PerkinElmer, Santa
Clara, CA). The number of viable cells in culture was determined
based on quantitation of the ATP present in each culture well. IC50
curves were plotted and IC50 values were calculated using the
GraphPad Prism 4 program based on a sigmoidal dose-response
equation.
1
2
3
4
5
6
7
8
9
[19] N. M. Kogan, M. Schlesinger, M. Peters, G. Marincheva, R. Beeri, R.
Mechoulam, J. Pharmacol. Exp. Ther. 2007, 322, 646–653.
[20] N. M. Kogan, M. Schlesinger, E. Priel, R. Rabinowitz, E. Berenstein, M.
Chevion, R. Mechoulam, Mol. Cancer Ther. 2007, 6, 173–183.
[21] N. M. Kogan, R. Rabinowitz, P. Levi, D. Gibson, P. Sandor, M. Schlesinger,
R. Mechoulam, J. Med. Chem. 2004, 47, 3800–3806.
[22] C. Petronzi, M. Festa, A. Peduto, M. Castellano, J. Marinello, A. Massa, A.
Capasso, G. Capranico, A. La Gatta, M. De Rosa, M. Caraglia, R. Filosa, J.
Exp. Clin. Cancer Res. 2013, 32, 24.
[23] N. Usami, I. Yamamoto, K. Watanabe, Life Sci. 2008, 83, 717–724.
[24] K. M. Regal, S. L. Mercer, J. E. Deweese, Chem. Res. Toxicol. 2014, 27,
2044–2051.
Acknowledgements
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
[25] M. Guzmán, Nat. Rev. Cancer 2003, 3, 745–755.
[26] I. Díaz-Laviada, Nat. Rev. Urol. 2011, 8, 553–561.
Financial support from the Engineering and Physical Sciences
Research Council for a PhD studentship to T.D.W. is gratefully
acknowledged. Prof. John A. Hartley (UCL Cancer Institute) is
thanked for helpful discussions, and Dr. Konstantinos Kiakos and
Dr. Luke Pett are thanked for assistance with gel electrophoresis
studies.
[27] C. Grimaldi, A. Capasso, Curr. Med. Chem. 2011, 18, 1571–1583.
[28] S. D. McAllister, R. T. Christian, M. P. Horowitz, A. García, P.-Y. Desprez,
Mol. Cancer Ther. 2007, 6, 2921–2927.
[29] A. Preet, R. K. Ganju, J. E. Groopman, Oncogene 2008, 27, 339–346.
[30] J. T. Wilson, C. A. Fief, K. D. Jackson, S. L. Mercer, J. E. Deweese, Chem.
Res. Toxicol. 2018, 31, 137–144.
[31] N. Kogan, Mini-Rev. Med. Chem. 2005, 5, 941–952.
[32] N. M. Kogan, C. Blázquez, L. Alvarez, R. Gallily, R. M. Schlesinger, M.
Guzmán, R. Mechoulam, Mol. Pharmacol. 2006, 70, 51–59.
[33] E. Frémy, Ann. Chim. Phys. 1845, 5, 408–488.
[34] T. A. Engler, R. Iyengar, J. Org. Chem. 1998, 63, 1929–1934.
[35] V. Pace, L. Castoldi, P. Hoyos, J. V. Sinisterra, M. Pregnolato, J. M.
Sánchez-Montero, Tetrahedron 2011, 67, 2670–2675.
[36] S.-H. Baek, M. Srebnik, R. Mechoulam, Tetrahedron Lett. 1985, 26, 1083–
1086.
[37] S. M. Wilkinson, J. Price, M. Kassiou, Tetrahedron Lett. 2013, 54, 52–54.
[38] F. E. Souza, J. E. Field, M. Pan, N. J. Ramjit, T. Tharmanathan, T. Jende-
Tindall, US Patent No. 074,252, 2006, April 6.
[39] A. Makryannis, S. P. Nikas, S. O. Alapafuja, V. G. Shukla, US Patent No.
306,016, 2009, December 10.
[40] M. S. Malamas, T. Q. Dinh, US Patent No. 207,927, 2003, November 6.
[41] P. J. Crocker, B. Saha, W. J. Ryan, J. L. Wiley, B. R. Martin, R. A. Ross, R. G.
Pertwee, R. K. Razdan, Tetrahedron 1999, 55, 13907–13926.
[42] S.-H. Baek, Bull. Korean Chem. Soc. 1988, 9, 71–77.
[43] D. A. Evans, D. M. Barnes, J. S. Johnson, T. Leckta, P. von Matt, S. J. Miller,
J. A. Murry, R. D. Norcross, E. A. Shaughnessy, K. R. Kampos, J. Am. Chem.
Soc. 1999, 121, 7582–7594.
[44] J. A. Teske, A. Deiters, Org. Lett. 2008, 10, 2195–2198.
[45] S. P. Acharya, H. C. Brown, J. Am. Chem. Soc. 1967, 89, 1925–1932.
[46] B. B. Hasinoff, T. I. Kuschak, J. C. Yalowich, A. M. Creighton, Biochem.
Pharmacol. 1995, 50, 953–958.
[47] M. Peters, N. M. Kogan, Expert Opin. Investig. Drugs 2007, 16, 1405–1413.
[48] R. P. Bender, A.-J. L. Ham, N. Osheroff, Biochemistry 2007, 46, 2856–2864.
[49] H.-Y. Wu, T.-R. Jan, Toxicol. Lett. 2010, 195, 68–74.
[50] D. Iacopetta, C. Rosano, F. Puoci, O. I. Parisi, C. Saturnino, A. Caruso, P.
Longo, J. Ceramella, A. Malzert-Fréon, P. Dallemagne, S. Rault, M. S.
Sinicropi, Eur. J. Pharm Sci. 2017, 96, 263–272.
Conflict of Interest
The authors declare no conflict of interest.
Keywords: anticancer agents · cannabidiol · Frémy’s salt · HU-
331 · monocyclic quinones · structure-activity relationships
[1] P. A. Henriksen, Heart 2018, 104, 971.
[2] S. N. Sunassee, M. T. Davies-Coleman, Nat. Prod. Rep. 2012, 29, 513–535
[3] S. M. Gordaliza, Marine Drugs 2010, 8, 2849–2870.
[4] C. A. Asche, Mini-Rev. Med. Chem. 2005, 5, 449–467.
[5] V. McGowan, R. Chung, A. Maulik, I. Piotrowska, J. M. Walker, D. M.
Yellon, Cardiovasc. Drugs Ther. 2007, 31, 63–75.
[6] M. M. Paz, A. Das, Y. Palom, Q.-Y He, M. Tomasz, J. Med. Chem. 2001, 44,
2834–2842.
[7] P. Morales, D. Vara, M. Goméz-Cañas, M. C. Zúñiga, C. Olea-Azar, P.
Goya, J. Fernández-Ruiz, I. Díaz-Laviada, N. Jagerovic, Eur. J. Med. Chem.
2013, 70, 111–119.
[8] F. Arcamone, G. Cassinelli, Curr. Med. Chem. 1998, 5, 391–419.
[9] G. Barone, C. F. Guerra, N. Gambino, A. Silvestri, A. Lauria, A. M.
Almerico, F. M. Bickelhaupt, J. Biomol. Struct. Dyn. 2008, 26, 115–130.
[10] C. A. Frederick, L. D. Williams, G. Ughetto, G. A. Van der Marel, J. H.
Van Boom, A. Rich, A. H. J. Wang, Biochemistry 1990, 29, 2538–2549.
[11] J. L. Nitiss, Nat. Rev. Cancer 2009, 9, 338–350.
[12] C. H. Takimoto, E. Calvo, ’Principles of Oncologic Pharmacotherapy’ in R.
Padzur, L. D. Wagman, K. A. Camphausen, W. J. Hoskins (eds.) Cancer
Management: a Multidisciplinary Approach 2008, 11th ed.
[13] B. Pang, X. Qiao, L. Janssen, A. Velds, T. Groothuis, R. Kerkhoven, M.
Nieuwland, H. Ovaa, S. Rottenberg, O. van Tellingen, J. Janssen, P.
Huijgens, W. Zwart, J. Neefjes, Nat. Commun. 2013, 4, 1908.
[14] C. C. Yu, P. J. Wu, J. L. Hsu, Y. F. Ho, L. C. Hsu, Y. J. Chang, H. S. Chang,
I. S. Chen, J. H. Guh, Prostate 2013, 73, 133–145.
[51] P. Peixoto, C. Bailly, M.-H. David-Cordonnier, Methods Mol. Biol. 2010,
613, 235–256.
[52] X. Pan, J. M. Hartley, J. A. Hartley, K. N. White, Z. Wang, S. W. A. Bligh,
Phytomedicine 2012, 19, 618–625.
[53] T. Mosmann, J. Immunol. Methods 1983, 65, 55–63.
[15] U. Mahmood, V. K. Kaul, L. Jirovetz, Phytochemistry 2002, 61, 923–926.
[16] A. A. Gunatilaka, J. M. Berger, R. Evans, J. S. Miller, J. H. Wisse, K. M.
Neddermann, I. Bursuker, D. G. Kingston, J. Nat. Prod. 2001, 64, 2–5.
[17] R. Mechoulam, N. M. Kogan, R. Rabinowitz, M. Schlesinger, WO/2005/
067917, 2005, July 28.
Manuscript received: September 26, 2019
Revised manuscript received: October 28, 2019
Version of record online: ■■■, ■■■■
ChemMedChem 2019, 14, 1–12
www.chemmedchem.org
11
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
FULL PAPERS
1
2
3
4
5
6
7
8
9
Heart safe: Many potent anticancer
and antibiotic agents contain a fused
1,4-benzoquinone ring, but also
show dose-limiting cardiotoxicity.
Noting the lack of cardiotoxicity of
the monocyclic quinone HU-331,
new monocyclic quinones were syn-
thesised to determine their anti-
cancer properties and to identify a
provisional pharmacophore. We
describe a crystalline monocyclic
quinone that, like HU-331, is also a
topoisomerase IIα inhibitor, but
showed greater potency against DU-
145 and Jurkat cell lines, as well as
having much greater stability in
solution than HU-331.
Dr. T. M. Waugh, Prof. J. Masters,
Dr. A. E. Aliev, Prof. C. M. Marson*
1 – 12
Monocyclic Quinone Structure-
Activity Patterns: Synthesis of
Catalytic Inhibitors of Topoisomer-
ase II with Potent Antiproliferative
Activity
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