Palladium-catalysed synthesis of arylnaphthoquinones as antiprotozoal and antimycobacterial agents

Article abstract of DOI:10.1016/j.ejmech.2020.112837

Malaria and tuberculosis are still among the leading causes of death in low-income countries. The 1,4-naphthoquinone (NQ) scaffold can be found in a variety of anti-infective agents. Herein, we report an optimised, high yield process for the preparation o

Full text of DOI:10.1016/j.ejmech.2020.112837

Journal Pre-proof
Palladium-Catalysed Synthesis of Arylnaphthoquinones as Antiprotozoal and
Antimycobacterial Agents
Marc-Manuel Kalt, Wolfgang Schuehly, Robert Saf, Sandra Ochensberger, Julia
Solnier, Franz Bucar, Marcel Kaiser, Armin Presser
PII:
S0223-5234(20)30809-6
DOI:
https://doi.org/10.1016/j.ejmech.2020.112837
EJMECH 112837
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 24 July 2020
Revised Date: 26 August 2020
Accepted Date: 7 September 2020
Please cite this article as: M.-M. Kalt, W. Schuehly, R. Saf, S. Ochensberger, J. Solnier, F. Bucar,
M. Kaiser, A. Presser, Palladium-Catalysed Synthesis of Arylnaphthoquinones as Antiprotozoal
and Antimycobacterial Agents, European Journal of Medicinal Chemistry, https://doi.org/10.1016/
j.ejmech.2020.112837.
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 Masson SAS.

trypanocidal activity
antiplasmodial activity
antimycobacterial activity

1
2
Palladium-Catalysed Synthesis of Arylnaphthoquinones as
Antiprotozoal and Antimycobacterial Agents
3
4
5
Marc-Manuel Kalt1 ● Wolfgang Schuehly2 ● Robert Saf3● Sandra
Ochensberger2 ● Julia Solnier2 ● Franz Bucar2 ● Marcel Kaiser4,5 ●
Armin Presser1
6
7
Abstract
8 Malaria and tuberculosis are still among the leading causes of death in low-
9 income countries. The 1,4-naphthoquinone (NQ) scaffold can be found in a
10 variety of anti-infective agents. Herein, we report an optimised, high yield
11 process for the preparation of various 2-arylnaphthoquinones by a
12 palladium-catalysed Suzuki reaction. All synthesised compounds were
13 evaluated for their in-vitro antiprotozoal and antimycobacterial activity.
14 Antiprotozoal activity was assessed against Plasmodium falciparum (P.f.)
15 NF54 and Trypanosoma brucei rhodesiense (T.b.r.) STIB900, and
16 antimycobacterial activity against Mycobacterium smegmatis (M.s.) mc2
17 155. Substitution with pyridine and pyrimidine rings significantly increased
18 antiplasmodial potency of our compounds. The 2-aryl-NQs exhibited
19 trypanocidal activity in the nM range with a very favourable selectivity
20 profile. (Pseudo)halogenated aryl-NQs were found to have a pronounced
21 effect indicating inhibition of mycobacterial efflux pumps. Cytotoxicity of
22 all compounds towards L6 cells was evaluated and the respective
1

1 selectivity indices (SI) were calculated. In addition, the physicochemical
2 parameters of the synthesised compounds were discussed.
3
4
5
Keywords Antiprotozoal ● Tuberculosis ● Suzuki cross-coupling●
6 Naphthoquinone
7
8
9
10 ____
11 * Armin Presser
12
13
armin.presser@uni-graz.at
orcid.org/0000-0003-0845-750X
14
15
16
17
18
19
20
21
22
23
1 Institute of Pharmaceutical Sciences, Pharmaceutical Chemistry,
University of Graz, Schubertstrasse 1, A-8010 Graz, Austria
2 Institute of Pharmaceutical Sciences, Pharmacognosy, University of
Graz, Universitaetsplatz 4, A-8010 Graz, Austria
3 Institute for Chemistry and Technology of Materials (ICTM),
University of Technology, Stremayrgasse 9, A-8010 Graz, Austria
4 Swiss Tropical and Public Health Institute, Socinstrasse 57, CH-
4002 Basel, Switzerland
5 University of Basel, Petersplatz 1, 4003 Basel, Switzerland
2

1
1. Introduction
2 Infectious diseases (IDs) are caused by a diverse group of pathogenic
3 microorganisms, such as bacteria, viruses, fungi and parasites. At least 25%
4 of about 60 million deaths occurring worldwide annually are estimated to
5 be induced by IDs [1]. Especially mortality caused by lower respiratory
6 tract infections has risen sharply over the last decade, which is mainly
7 attributed to the spread of drug-resistance.
8 Vector-borne parasitic diseases like malaria, human African
9 trypanosomiasis (HAT, also known as sleeping sickness), leishmaniasis
10 and Chagas disease are still among the leading causes of morbidity and
11 mortality in countries of the developing world without access to adequate
12 health care systems [2]. Even though huge efforts are put forth in
13 eradicating these maladies, they are still affecting more than 1 billion
14 people, or about 15% of the world population [3,4]. With no vaccines
15 available, vector control and drug therapy remain the cornerstones of
16 disease control [5–8]. Caused by the apicomplexan genus Plasmodium (P.
17 falciparum, P. vivax, P. malariae, P. ovale and P. knowlesi), malaria is the
18 most common and clinically relevant vector-borne parasitic disease.
19 Transmitted by infected Anopheles mosquitos, this ID is claiming the lives
20 of more than 435 000 people each year [5].
3

1 Efforts in malaria control often show poor outcome due to the highly
2 resilient and adaptive nature of the Plasmodium parasite and the
3 consequential emergence of drug-resistance. In some areas the number of
4 drug-resistant strains is continuously rising, which limits the efficacy of
5 currently available therapies [5,6].
6 Another pathogen posing an enormous threat to human life is
7
Mycobacterium tuberculosis, a gram-positive mycobacterium that causes
8 tuberculosis (TB). This ID which mainly affects the lungs has been
9 declared a global health emergency by the WHO due to its latent
10 occurrence in nearly one quarter of the global population [9].
11 With the emergence of drug-resistant strains, TB control efforts have been
12 further complicated in recent decades. Multidrug-resistant (MDR),
13 extensively drug-resistant (XDR) and totally drug-resistant (TDR) strains
14 are a great reason of concern [10]. Extrusion of antibiotics from the
15 bacterial cells by efflux pumps, i.e. bacterial transporters which are able to
16 confer resistance to groups of chemically diverse compounds, is a
17 frequently occurring mechanism of resistance [11]. In this study, selected
18 compounds were tested for potential antimycobacterial effects using
19 Mycobacterium smegmatis, a fast growing mycobacterial strain, in a
20 microbroth dilution assay. In addition, subinhibitory concentrations of the
21 test compounds were evaluated for their potency to lower the minimal
4

1 inhibitory concentration (MIC) of the efflux pump substrate ethidium
2 bromide (EtBr). Reduction of the MIC of EtBr in combination with a test
3 compound could be indicative for a potential inhibitor of bacterial efflux,
4 and could be useful in combination with existing antituberculotics in MDR
5 TB [12].
6 Despite the effective use of chemotherapeutic approaches to treat and
7 prevent parasitic infections, satisfactory medications are often still lacking.
8 Many of the currently used drugs suffer from major drawbacks such as
9 significant toxicity, variable efficacy, or lack of oral bioavailability [13,14].
10 Special efforts have to be placed on the development of new trypanocidal
11 drugs, as these therapeutic agents must be able to cross the blood brain
12 barrier (BBB) in order to effectively combat late-stage T. b. rhodesiense
13 infections [7,15].
14 The naphthoquinone (NQ) scaffold can be found in a variety of secondary
15 plant metabolites, e.g. shikonin, with a broad spectrum of biological
16 activities [16–19]. Some mechanisms of action have been proposed for
17 these substances, mainly related to excessive generation of reactive oxygen
18 species (ROS) [20] and inhibition of DNA gyrase [21]. Previous studies
19 demonstrated that NQs represent an interesting pharmacophore for further
20 drug development due to their activity against several apicomplexan
21 parasites and their antimicrobial properties in general [22,23]. For example,
5

1 the 2-hydroxynaphthoquinones lapachol and atovaquone (ATV) act as
2 potent antimalarial redox-active agents. ATV has been used for
3 chemoprophylaxis against malaria infections since its introduction to
4 market, as it acts both on liver and blood stages of the parasite. Used in a
5 fixed-dose combination with proguanil, ATV inhibits the mitochondrial
6 electron transport chain (mETC) at the level of the cytochrome bc1
7 complex, which results in a loss of mitochondrial function [24]. The
8 emergence of ATV-resistant P. falciparum strains highlights limitations to
9 its clinical efficacy in areas with endemic malaria and provides motivation
10 for the synthesis of second-generation bc1-targeted inhibitors. Fortunately,
11 many of these research compounds used as leads (including ATV
12 analogues) demonstrate little cross resistance with atovaquone-resistant
13 parasite strains [22,25].
14 In the present study, we modified the alkyl side chain of naturally
15 occurring shikonin to create naphthoquinone derivatives that bear various
16 aryl moieties instead. In contrast to atovaquone, shikonin has no hydroxyl
17 group adjacent to its lipophilic attachment. Moreover, it shows numerous
18 beneficial pharmacological properties [18].
19 Herein, we report a highly efficient method for the introduction of aromatic
20 substituents in position 2 of the 1,4-naphthoquinone scaffold and our
6

1 findings on the antiprotozoal and antimycobacterial activity of 18
2 naphthoquinone derivatives.
3
4
5
6
7
8
Fig. 1 Naphthoquinones with remarkable biological activity
2. Results and discussion
2.1 Synthetic chemistry
9 2-Arylnaphthoquinones have been obtained only through a few general
10 methods that depend on the substitution pattern and only in rare cases by
11 direct arylation [26,27]. The palladium-catalysed cross-coupling of aryl
12 halides with arylboronic acids (Suzuki-Miyaura cross-coupling) has
13 become a reliable technique in organic synthesis for the introduction of aryl
14 groups due to its versatility and efficiency under mild reaction conditions
15 [28]. Recent developments focused especially on the reduction of the
16 amount of transition metal needed by introduction of highly specialised
17 directing ligands, as well as the use of non-organic solvents [29,30].
18 However, for the arylation of haloquinones, these methods have received
7

1 limited attention because of the poor stability of bromo- and iodoquinones
2 and the limited susceptibility of chloronaphthoquinones to Suzuki
3 couplings.
4 Nevertheless, we were encouraged by published protocols [31] to attempt
5 such couplings on 2-chloro-1,4-naphthoquinones, but the preparation of the
6 chlorinated precursor was tedious and the yields of the following coupling
7 reaction were unsatisfactory.
8 Therefore, we focused on the utilisation of bromonaphthoquinone 2 as
9 reagent for the Suzuki reaction. Oxidative bromination of 1-naphthol (1)
10 using N-bromosuccinimide (NBS) in acetic acid [32] delivered 2 in
11 excellent yield. Unfortunately, our attempts to couple intermediate 2 with
12 phenylboronic acid via a published Pd-catalysed Suzuki coupling [33] gave
13 only moderate yields, which was perhaps due to the fact that the reaction
14 conditions in the original paper were not specified in detail. Therefore, we
15 evaluated the effects of different Pd catalysts, bases and solvents that are
16 frequently used for this type of reactions with 2 and phenylboronic acid as
17 model compounds (see Supplementary Table S1 for details). Optimal
18 conversion of the starting material was achieved using Pd(PPh3)4 (2.5 mol
19 %), aromatic boronic acid (1.5 equiv.) and Cs2CO3 (1.5 equiv.) in THF-
20 water (20:3) as a solvent. A further decrease in the equivalents of boronic
21 acids and Pd catalyst reduced the respective yields. Optimal solvent
8

1 composition not only led to high yields, but also to a dramatic reduction of
2 reaction time. Under these improved conditions, 3a was obtained in 94%
3 yield (the highest described for this type of Suzuki coupling so far) in 1.5 h
4 without the usage of specially designed ligands and minimised Pd load.
5 Use of the corresponding boronic acid pinacol ester is also possible,
6 resulting in a similar overall yield.
7 Next, arylnaphthoquinones 3a-r were synthesised from 2 with
8 commercially available aromatic boronic acids under optimised conditions
9 in good to excellent yields with the exception of the furyl (3g, 3h) and the
10 quinolyl (3i) derivatives (Scheme 1). Fortunately, the starting material is
11 completely consumed even in the low yield examples 3g-i and thus
12 simplifying subsequent chromatographical purification. The preparation of
13 3s and 3t failed, which suggests that the described method is, as of yet,
14 limited to the use of aromatic boronic acids.
15
9

O
O
O
OH
Br
R
i
ii
O
1
2
3a-t
CH3
CH3
CH3
a R =
b R =
c R =
O
O
N
e R =
C N
S
O
N
d R =
f R =
H
i R =
g R =
j R =
h R =
O
O
N
N
N
F
F
k R =
n R =
Cl
Cl
l R =
Cl
N
N
F
o R =
m R =
F
OCF3
p R =
CF3
q R =
OCF3
r R =
s R =
t R =
1
2
Scheme 1 Reagents and conditions: (i) NBS, AcOH; H2O, 45 °C, 1 h
3 (94%); (ii) boronic acid, Pd(PPh3)4, Cs2CO3, THF/H2O (20:3), 65 °C, 1.5–
4 20 h (3a: 94%, 3b: 83%, 3c: 81%, 3d: 92%, 3e: 82%, 3f: 88%, 3g: 22%,
10

1
3h: 12%, 3i: 12%, 3j: 75%, 3k: 82%, 3l: 56%, 3m: 84%, 3n: 74%, 3o:
2 68%, 3p: 95%, 3q: 91%, 3r: 98%, 3s: 0%, 3t: 0%).
3
4
5
2.2 Biological evaluation
2.2.1 Antiprotozoal activity
6 The synthesised 2-arylnaphthoquinones (3a-r) were evaluated in vitro for
7 their antiprotozoal activity against P. falciparum (NF54) and T. brucei
8
rhodesiense (STIB900). Cytotoxicity was measured using L6 rat skeletal
9 myoblasts to calculate a selectivity index for each parasite (SI =
10 IC50(L6)/IC50(parasite)).
11 According to TDR (Special Program for Research and Training in Tropical
12 Diseases, WHO) criteria [34], all derivatives showed moderate (IC50 = 1–
13 10 µM) or high (IC50 < 1 µM) antiplasmodial activity towards NF54 (Table
14 1), except the inactive phenylnaphthoquinone 3a, and the benzofuranyl
15 derivative 3h, with IC50 values > 10 µM. Strongest activity in the NF54
16 assay with excellent selectivity was exhibited by the 2-chloro-5-pyridyl
17 naphthoquinone 3k (IC50 = 0.07 µM, SI = 462.6). Except for 3k, the
18 selectivity indices ranged between 1.8 (3a) and 26.1 (3r) and have to be
19 improved to target the Plasmodium parasites effectively.
20 High trypanocidal activity could also be observed. According to the criteria
21 set above, most of the tested derivatives showed high activity against T. b.
11

1
rhodesiense with IC50 values as low as 0.06 µM and favourable selectivity,
2 especially in the case of 3a, 3b, 3g, 3p (SI = 57.2–322.8). Good selectivity
3 was also observed for the benzofuranyl derivative 3h (SI = 153.9), but
4 biological activity was among the weakest overall.
5
12

1
2
Table 1: In vitro antiparasitic activity, host toxicity and lipophilic ligand efficiency of the tested compounds.
compd R
P. f.a
SIb
T.b.r.c
SIb Cyt. L6d P. f.a T.b.r.c P. f. T.b.r.
e
e
IC50 μM
IC50 μM
IC50 μM pIC50 pIC50 LLE f LLE f
Chl.
Mel.
Pod.
0.003
8.52
8.22
0.006
0.024
phenyl
4-methylphenyl
4-isopropylphenyl
3-(methylsulfonylamino)phenyl
4-cyanophenyl
10.95
3.86
2.96
2.56
1.11
1.8
1.9
6.9
3.6
6.5
0.06
0.08
1.45
0.20
0.28
322.8 19.37
4.96
5.41
5.53
5.59
5.96
7.26
7.12
5.84
6.70
3a
3b
3c
3d
3e
2.27 4.56
2.37 4.08
1.65 1.96
3.85 4.95
90.8
14.1
46.3
25.8
7.26
20.40
9.25
7.21
6.56 3.43 4.03
4-(piperidin-1-
ylcarbonyl)phenyl
2-furyl
2-benzo[b]furanyl
3-quinolyl
6-fluoro-3-pyridyl
6-chloro-3-pyridyl
2-chloro-5-pyrimidinyl
2.37
4.1
0.75
12.9
9.71
5.63
6.13 2.09 2.59
3f
2.07
24.96
0.45
0.11
0.07
0.47
3.6
8.4
12.5
19.1
462.6
2.2
0.13
1.36
0.22
0.14
2.20
0.13
57.2
153.9 209.37
25.6
15.0
14.7
7.8
7.44
5.68
4.60
6.35
6.94
7.18
6.33
6.89 3.80 5.01
5.87 1.40 2.67
6.65 3.34 3.64
6.86 4.80 4.71
5.66 4.78 3.26
6.89 4.35 4.91
3g
3h
3i
3j
3k
3l
5.64
2.10
32.38
1.02
13

4-fluorophenyl
4-chlorophenyl
3,5-difluorophenyl
4-(trifluoromethyl)phenyl
4-(trifluoromethoxy)phenyl
3-(trifluoromethoxy)phenyl
1.81
2.64
0.59
1.08
3.87
0.89
5.0
3.4
16.4
14.0
4.3
0.22
0.22
0.24
0.08
2.36
1.80
41.2
40.7
40.3
189.0 15.12
7.0
9.06
8.96
9.67
5.74
5.58
6.23
5.97
5.41
6.05
6.66 2.95 3.87
6.65 2.28 3.35
6.62 3.33 3.72
7.12 2.42 3.58
5.63 1.62 1.83
5.75 2.26 1.95
3m
3n
3o
3p
3q
3r
16.62
23.23
26.1
12.9
b
c
1
2
3
4
5
6
aP. falciparum, strain NF54, erythrocytic stages; SI is defined as the ratio: IC50 in L6 cells/IC50 in each parasite; T.
brucei rhodesiense, strain STIB900 trypomastigote forms; dcytotoxicity L6 cells rat skeletal myoblasts; pIC50 = -log10
e
IC50 (M). fLLE was calculated using the OSIRIS Datawarrior 5.0.0 (http://www.openmolecules.org/). Reference drugs:
chloroquine (chl.), melarsoprol (mel.), podophyllotoxin (pod.). The IC50 value of each reference drug is the mean from
multiple measurements in parallel with the compounds of interest. IC50 values of the tested compounds are means of 2
to 3 measurements. The SD was < 5%.
14

1
2.2.2 Antimycobacterial and resistance-modulatory activity
2 The synthesised compounds were also tested against the strain
Mycobacterium smegmatis mc² 155 (ATCC 700084) to further gain insight
3
4 into the antimycobacterial potential of these substances. Each compound
5 was screened for their minimal inhibitory concentration (MIC) and the
6 modulation factor (MF) tested against ethidium bromide (EtBr). The MIC
7 value indicates the concentration (in µg/mL) required to inhibit visible
8 growth of the bacteria. The MF shows the x-fold reduction of the MIC of
9 ethidium bromide when incubated using a constant subinhibitory (½ MIC)
10 concentration of modulator (3a-r). An MF of 8 states that an 8 times lower
11 amount is needed to exert the same antibiotic combined effect. In contrast,
12 an MF of 1 indicates no effect of the modulator on the MIC of the tested
13 substance.
14 The most active compound according to the obtained MIC is 3l with an
15 MIC of 4 µg/mL. Similarly pronounced growth inhibitory effects were seen
16 for 3a and 3i with MIC of 8 µg/mL and 3j, 3n and 3q with MIC of 16
17 µg/mL, respectively. However, all MIC values were above the respective
18 IC50 values of cytotoxicity in L6 cells.
19 An MF of 4 can be considered as cut-off for a promising MF. Nearly half
20 (8 of 18) of the synthesised arylnaphthoquinones showed an MF of at least
15

1 128. Highly promising candidates with an MF of > 128 are compounds 3e,
2
3
4
3j, 3m and 3n.
Table 2: Antimycobacterial and resistance-modulatory activity of the
5 tested compounds against Mycobacterium smegmatis mc² 155.
compd MICa µg/mL MFb (EtBr)
INH
EtBr
3a
3b
3c
4-8
8
8
32
64
-
-
4
2
4
> 128
64
32
128
> 128
8
3d
3e
3f
128
> 128
8
2
2
8
3g
3h
3i
16
32
4
> 128
128
2
3j
3k
3l
128
16
> 128
128
16
> 128
> 128
4
4
128
128
3m
3n
3o
3p
3q
3r
64
6
aMinimal inhibitory concentration for the Mycobacterium smegmatis mc²
b
7 155 strain; modulation factor tested against ethidium bromide. Reference
8 drugs: isonicotinic acid hydrazide (INH), ethidium bromide (EtBr).
9
10 2.3. Physicochemical parameters
16

1 To find the ideal balance of potency and physicochemical properties is of
2 utmost relevance in the drug development process [35,36]. For this reason,
3 an assessment of drug-likeness was made for all tested compounds, and
4 various physicochemical properties were calculated (see Suppl. Material).
5 All compounds had relatively low molecular weights within the range of
6 224–345 g·mol-1. Except the methylsulfonylamino derivative 3d, all
7 compounds were predicted to possess good central nervous system (CNS)
8 penetration [37,38], a key feature for the treatment of second stage HAT.
9 All synthesised arylnaphthoquinones fulfil the Lipinsky rule of five [39],
10 Veber’s rule [40] and the drug-likeness classifier defined by Ghose et al.
11 [41].
12 Ligand efficiency, especially lipophilic ligand efficiency (LLE) has gained
13 traction in recent years as a helpful metric in lead discovery and
14 optimisation [42–45]. LLE normalises a molecule’s biological potency in
15 relation to its lipophilicity. The use of LLE as a metric can greatly simplify
16 multi-parameter optimisation in drug development by focussing on the
17 design of potent compounds with preferably low lipophilicity i.e. high LLE
18 [46].
19 An LLE above 3 indicates favourable compounds for further development.
20 Several of the newly developed compounds were found well above this
21 threshold. For the P. falciparum parasite, the heteroaryls 3j, 3k and 3l
17

1 showed drug-like behaviour in this regard (LLE = 4.35 – 4.80). For T.b.
2
3
4
5
rhodesiense, good LLE values were observed for compounds 3a, 3b, 3d,
3g, 3j and 3l (LLE = 4.08–5.01).
2.4 Structure-Activity Relationships (SAR) of the Antiparasitic Activity
6 By taking a closer look at the LLE and pIC50 values of the
7 arylnaphthoquinone derivatives, we observed a clear clustering of
8 compounds with similar structural elements and trypanocidal activities
9 (Fig. 2). Sterically hindered or strongly polarised compounds (3c, 3f, 3h,
10 3q, 3r; pIC50 = 5.63–6.13) turned out to be inactive. (Pseudo)-halogenated
11 compounds (3e, 3m, 3n, 3o; pIC50 = 6.56–6.66) also showed a distinct
12 clustering with an increase of 0.5–1 log unit for this set. Small heteroaryls
13 (3g, 3j, 3l; pIC50 = 6.86–6.89) represent the third group with a further
14 increase in activity of around 0.3 log units. Small lipophilic compounds
15 (3a, 3b, 3p) were the most potent (pIC50 = 7.12–7.26) with SI values in the
16 range of 100–300. Only the 3-(methylsulfonylamino)phenyl derivative 3d,
17 the quinolylnaphthoquinone 3i, and the 2-chloro-4-pyridyl naphthoquinone
18 3k, do not seem to follow any clear trend from an SAR point of view.
18

1
2
Fig. 2 Plot of pIC50 (T.b. rhodesiense) data obtained in this study (x) vs.
3 calculated lipophilic ligand efficiency (y). Substances are labelled in the
4 following way: green (sterically hindered or strongly polarised: 3c, 3f, 3h,
5
3q, 3r), red (halogenated: 3e, 3m, 3n, 3o), yellow (heteroaryls: 3g, 3j, 3l),
6 blue (small lipophilic residues: 3a, 3b, 3p).
7
8 A relationship between trypanocidal potency and SI can also be found in
9 this dataset (Fig. 3). This hints at target-specific effects of the tested
10 arylnaphthoquinones, opposed to non-target specificity often exhibited by
11 the quinone structure class. Compounds with SI > ~100 and pIC50 values
12 greater than 6.7 (corresponding to activity < 0.2 µM) combine high activity
13 with acceptable selectivity over an L6 cell line and deemed to be promising
14 drug candidates for a further study based on this assessment [34].
19

1
2
Fig. 3 Plot of pIC50 (T.b. rhodesiense) data obtained in this study (x) vs.
3 calculated log SI (T.b. rhodesiense) (y). Substances appearing in the yellow
4 shaded area combined high activity with acceptable selectivity over an L6
5 cell line.
6
7 Regarding antiplasmodial activity, the pyridyl and pyrimidinyl derivatives
8
3j-l, showed the strongest biological activity. In particular, the 2-chloro-5-
9 pyridyl-1,4-naphthoquinone 3k (IC50 = 0.07 µM; SI = 462.6) was the most
10 active compound with excellent selectivity tested in the P.f. NF54.
11 Compared to the fluoro analogon 3j (IC50 = 0.11 µM; SI = 19.1), the
12 introduced chlorine group significantly reduces the cytotoxicity while
13 simultaneously increasing its antiplasmodial potency.
20

1 Still, in comparison to their trypanocidal effects, the antiplasmodial
2 activities of the tested arylnaphthoquinones were rather sparse.
3 Interestingly, cytotoxicity and antiplasmodial activity are strongly
4 correlated (R2 = 0.909). The benzofuranyl derivative 3h showed the
5 generally lowest cytotoxicity, but it is still strongly correlated with its
6 antiplasmodial activity. Remarkably, a strong influence of the position of
7 the trifluoromethoxy group of compounds 3q and 3r was observed: the
8
meta-position is more favourable, as antiplasmodial activity is raised by
9 one log unit and SI increases fivefold.
10 In view of antimycobacterial activity, compound 3j, 3n and 3q appear to be
11 the most promising, since they combine high activity in both the MIC (16
12 µg/mL) and MF assay (≥ 128). Chlorinated compounds often showed
13 considerably higher activity in the MIC assay (3l, 3n) compared to
14 derivatives with fluorine substitution (3m, 3o) (Fig. 4). The pyrimidine
15 moiety (3l) resulted in the lowest observed MIC of all tested compounds at
16 4 µg/mL, but this modification leads to nearly a complete loss of efflux
17 pump modulatory activity.
18 Another general observation is that the para (pseudo)-halogenation of aryl
19 derivatives is crucial for high efflux inhibitory effects. With exception of
20 the pyrimidine derivative 3l, all tested compounds with this structural
21 feature (3e, 3j, 3k, 3m, 3n, 3q) showed an MF ≥ 128.
21

1
2
MIC
64
MF
MIC
16
MF
> 128
128
2
X = CN (3e)
X = F (3m)
X = Cl (3n)
> 128
> 128
> 128
X = F, Y = C (3j)
X = Cl, Y = C (3k)
X = Cl, Y = N (3l)
128
16
32
3
4
4
Fig. 4 Comparison between the (pseudo)-halogenated phenyl and pyridine
5 NQs with regards to their antimycobacterial properties.
6
7
3. Conclusions
8 We present an efficient, high-yield synthetic procedure for the preparation
9 of 2-arylnaphthoquinones by a Suzuki type cross-coupling, which gave
10 good to excellent yields for a variety of aromatic boronic acids. Many of
11 the new synthesised compounds showed encouraging biological activities
12 against P.falciparum, T.b. rhodesiense and M. smegmatis (Fig. 5), as well
13 as favourable physicochemical properties. Trypanocidal effects were
14 extraordinary and remarkable selectivity could be achieved, especially in
15 the case of 3a (IC50 = 0.06 µM; SI = 322.8) and 3p (IC50 = 0.08 µM; SI =
16 189.0). The pyridyl derivative 3k exhibits the lowest IC50 of all prepared
17 compounds against the P.falciparum NF54 strain at 0.07 µM with very
18 good selectivity (SI = 462.6), demonstrating that these heteroaromatics
19 represent promising chemical leads for further derivatisations.
22

1 Remarkable antimycobacterial properties have also been observed: the
2 most active compound against M. smegmatis proved to be 3l with an MIC
3 of 4 µg/mL. In addition, efflux pump modulation of M. smegmatis was
4 outstanding, with 3e, 3j, 3m and 3n showing an MF > 128 in the ethidium
5 bromide efflux assay. The findings presented in this work demonstrate the
6 potential for the 2-aryl-1,4-naphthoquinones in the development of new
7 anti-infective drugs.
Cl
N
Cl
O
O
O
N
N
O
O
O
3a
T.b.r. IC50 (µM) 0.06
3k
P.f. NF54 IC50 (µM) 0.07
3l
M.s. MIC (mg/L) 4.0
(SI = 323)
(SI = 463)
8
9
Fig. 5 Overview of the most active compounds in their respective assays.
10
11 4. Experimental section
12 4.1. General
13 All reagents and solvents were purchased from Sigma-Aldrich and
14 Fluorochem Ltd. The moisture-sensitive reactions were performed under an
15 inert atmosphere of argon. Each reaction was monitored by TLC on Merck
16 TLC plates (silica gel 60 F254 0.2 mm, 200 × 200 mm) and detected at
17 254ꢀnm. All the compounds were purified by flash column chromatography
18 using silica gel 60 (Merck, 70 - 230 mesh, pore-diameter 60 Å), unless
23

1 otherwise specified. Purity and homogeneity of the final compounds were
2 assessed by TLC and high-resolution mass spectrometry. The melting
3 points were determined with a digital melting point apparatus
1
4 (Electrothermal IA 9200). The structure elucidation was confirmed by H
13
5 and C NMR (1D and 2D) on a Varian Unity Inova 400 MHz instrument
6 (at 298 K) using 5 mm tubes. The chemical shifts are expressed in δ (ppm)
13
7 using tetramethylsilane (TMS) as internal standard or the C signal of the
8 solvent (CDCl3 δ 77.04), coupling constants (J) were reported in Hertz
1
13
9 (Hz). H and C resonances are numbered as given in the formulae, the
10 signals marked with an asterisk are interchangeable. High-resolution EI
11 mass spectra (70 eV, source temperature 220 °C) were recorded on an
12 orthogonal TOF spectrometer (Waters GCT Premier) equipped with a
13 direct insertion (DI) probe. Typically, 0.2 µl of a solution of the sample (c
14 = 0.1 mg/mL) were placed in the glass cup used for DI, dried under
15 atmospheric pressure, and transferred into the vacuum. Mass spectra (50–
16 800 Da; 1 spectrum/s; resolution appr. 7500 FWHM) were continuously
17 acquired while the sample was evaporated rapidly. ESI mass spectra were
18 acquired on an Exactive Orbitrap mass spectrometer equipped with a
19 heated ESI II source.
20
21 4.2. Synthesis of 2-bromo-1,4-naphthoquinone (2)
24

1 To a vigorously stirred solution of NBS (5.34 g, 30 mmol) in AcOH (15
2 mL) and H2O (30 mL), a mixture of 1-naphthol (1.08 g, 7.5 mmol) in
3 AcOH (15 mL) was added dropwise at 45 °C within 20 min. The clear
4 solution slowly turned orange and a yellow precipitate was formed. Stirring
5 at 45°C was continued until TLC showed complete consumption of the
6 starting material (1 h). Then, additional 40 mL of H2O were added and the
7 reaction mixture was cooled to 0°C. The yellow precipitate was filtered,
8 washed two times with ice-cold H2O and dried under reduced pressure to
9 obtain 1.67 g (94 %) of 2 as a bright yellow solid. Rf = 0.37 (CHCl3:CH =
10 2:1), mp 127–128 °C (lit. [32] mp 132–133 °C). The spectroscopic data
11 were in accordance with those described in Ref. [32].
12
13 4.3. General synthetic procedures for arylnaphthoquinones 3a-r
14 Under exclusion of air or humidity, bromonaphthoquinone 2 (237 mg, 1
15 mmol) and Cs2CO3 (489 mg, 1.5 mmol) were added successively to a
16 stirred solution of Pd(PPh3)4 (29 mg, 0.025 mmol) in anhydrous THF (4
17 mL). Then, 0.6 mL of H2O and 1.5 mmol of the corresponding boronic acid
18 were added and the reaction mixture was heated to 65°C until TLC showed
19 complete consumption of the starting material (1.5–20 h). After cooling to
20 ambient temperature, the mixture was diluted with water (15 mL) and
21 extracted several times with EtOAc. The combined organic layers were
25

1 dried over Na2SO4 and concentrated in vacuo to give a residue, which was
2 purified by flash chromatography.
3
4 4.3.1. 2-Phenyl-1,4-naphthoquinone (3a)
5 Compound 3a was obtained after heating for 1.5 h and purified by flash
6 chromatography using toluene. Yellow solid; yield: 94%; Rf = 0.32
7 (toluene); mp: 107–108 °C (lit. [47] mp 106–108 °C). The spectroscopic
8 data were found to be identical to the ones described in Ref. [47,48].
9 Although the structure of 3a is already known, to the best of our
10 knowledge, a complete assignment of the NMR signals has never been
11 published: 1H-NMR (400 MHz, CDCl3) δ 8.23 – 8.17 (m, 1H, H-8), 8.17 –
12 8.09 (m, 1H, H-5), 7.85 – 7.77 (m, 2H, H-6, H-7), 7.74 (d, J = 8.3 Hz, 2H,
13
13 H-3’, H-5’), 7.11 (s, 1H, H-3), 7.69 (d, 2H, H-2’, H-6’) ppm; C-NMR
14 (100 MHz, CDCl3) δ 185.1 (C-4), 184.3 (C-1), 148.1 (C-2), 135.2 (C-3),
15 133.9 (C-7), 133.8 (C-6), 133.4 (C-1’), 132.4 (C-8a), 132.1 (C-4a), 130.0
16 (C-4’), 129.4 (C-2’, C-6’), 128.5 (C-3’, C-5’), 127.0 (C-8), 126.0 (C-5)
17 ppm.
18
19 4.3.2. 2-(4-Methylphenyl)-1,4-naphthoquinone (3b)
20 Compound 3b was obtained after heating for 4 h and purified by flash
21 chromatography using toluene. Yellow solid; yield: 83%; Rf = 0.44
26

1 (toluene); mp: 109–110 °C (lit. [47] mp 100–104 °C). Although the
2 structure of 3b is already known [47], to the best of our knowledge, a
1
3 complete assignment of the NMR signals has never been published: H-
4 NMR (400 MHz, CDCl3) δ 8.21 – 8.16 (m, 1H, H-8), 8.14 – 8.08 (m, 1H,
5 H-5), 7.80 – 7.74 (m, 2H, H-6, H-7), 7.49 (d, J = 7.9 Hz, 2H, H-2’, H-6’),
6 7.28 (d, J = 7.9 Hz, 2H, H-3’, H-5’), 7.07 (s, 1H, H-3), 2.42 (s, 3H, CH3)
7 ppm; 13C-NMR (100 MHz, CDCl3) δ 185.2 (C-4), 184.6 (C-1), 148.0 (C-2),
8 140.4 (C-4’), 134.6 (C-3), 133.8 (C-6, C-7), 132.5 (C-8a), 132.1 (C-4a),
9 130.5 (C-1’), 129.4 (C-2’, C-6’), 129.2 (C-3’, C-5’), 127.0 (C-8), 125.9 (C-
10 5), 21.4 (CH3) ppm; HRMS (EI) calcd. for C17H12O2 [M]+ = 248.0837;
11 found: 248.0837.
12
13 4.3.3. 2-(4-Isopropylphenyl)-1,4-naphthoquinone (3c)
14 Compound 3c was obtained after heating for 4 h and purified by flash
15 chromatography using toluene. Yellow crystallising oil; yield: 81%; Rf =
16 0.31 (toluene); mp: 85–86 °C. Although the structure of 3c is already
17 known [49], to the best of our knowledge, a complete assignment of the
1
18 NMR signals has never been published: H-NMR (400 MHz, CDCl3) δ
19 8.21 – 8.16 (m, 1H, H-8), 8.15 – 8.09 (m, 1H, H-5), 7.81 – 7.73 (m, 2H, H-
20 6, H-7), 7.53 (d, J = 8.0 Hz, 2H, H-2’, H-6’), 7.34 (d, J = 8.0 Hz, 2H, H-3’,
21 H-5’), 7.07 (s, 1H, H-3), 2.97 (hept, J = 7.0 Hz, 1H, CH(CH3)2), 1.29 (d, J
27

13
1 = 7.0 Hz, 6H, (CH3)2-CH) ppm; C-NMR (100 MHz, CDCl3) δ 185.2 (C-
2 4), 184.6 (C-1), 151.3 (C-4’), 148.0 (C-2), 134.6 (C-3), 133.8 (C-6, C-7),
3 132.5 (C-8a), 132.1 (C-4a), 130.8 (C-1’), 129.5 (C-2’, C-6’), 127.0 (C-8),
4 126.7 (C-3’, C-5’), 125.9 (C-5), 34.1 (CH-(CH3)2), 23.8 ((CH3)2-CH) ppm;
5 HRMS (EI) calcd. for C19H16O2 [M]+ = 276.1150; found: 276.1151.
6
7 4.3.4. 2-(3-(Methylsulfonylamino)phenyl)-1,4-naphthoquinone (3d)
8 Compound 3d was obtained after heating for 20 h and purified by flash
9 chromatography using toluene/EtOAc (5:1). Orange oil; yield: 92%; Rf =
10 0.22 (toluene:EtOAc = 5:1); 1H-NMR (400 MHz, CDCl3) δ 8.20 – 8.17 (m,
11 1H, H-8), 8.15 – 8.12 (m, 1H, H-5), 7.83 – 7.77 (m, 2H, H-6, H-7), 7.48 (t,
12 J = 8.0 Hz, 1H, H-5’), 7.46 – 7.44 (m, 1H, H-2’), 7.41 – 7.36 (m, 1H, H-
13 6’), 7.38 – 7.34 (m, 1H, H-4’), 7.10 (s, 1H, H-3), 6.61 (s, 1H, NH), 3.10 (s,
13
14 3H, CH3-(SO2)N) ppm; C-NMR (100 MHz, CDCl3) δ 184.9 (C-4), 184.1
15 (C-1), 147.0 (C-2), 136.8 (C-3’), 135.7 (C-3), 135.0 (C-1’), 134.1 (C-6, C-
16 7), 132.3 (C-8a), 132.0 (C-4a), 129.9 (C-5’), 127.1 (C-8), 126.2 (C-6’),
17 126.1 (C-5), 121.9 (C-4’), 121.5 (C-2’), 39.7 (CH3-(SO2)N) ppm; HRMS
18 (EI) calcd. for C17H13NO4S [M]+ = 327.0565; found: 327.0557.
19
20 4.3.5. 2-(4-Cyanophenyl)-1,4-naphthoquinone (3e)
28