Oxadiazole-substituted naphtho[2,3-b]thiophene-4,9-diones as potent inhibitors of keratinocyte hyperproliferation. Structure?activity relationships of the tricyclic quinone skeleton and the oxadiazole substituent

Article abstract of DOI:10.1016/j.ejmech.2017.03.084

Novel analogues of oxadiazole-substituted naphtho[2,3-b]thiophene-4,9-diones were synthesized in which the tricyclic quinone skeleton was systematically replaced with simpler moieties, such as structures with fewer rings and open-chain forms, while the oxadiazole ring was maintained. In addition, variants of the original 1,2,4-oxadiazole ring were explored. Overall, the complete three-ring quinone was essential for potent suppression of human keratinocyte hyperproliferation, whereas analogous anthraquinones were inactive. Also, the oxadiazole ring per se was not sufficient to elicit activity. However, rearrangement of the heteroatom positions in the oxadiazole ring resulted in highly potent inhibitors with compound 24b being the most potent analogue of this series showing an IC₅₀ in the nanomolar range. Furthermore, experiments in isolated enzymatic assays as well as in the keratinocyte-based hyperproliferation assay did not support a major role of redox cycling in the mode of action of the compounds.

Full text of DOI:10.1016/j.ejmech.2017.03.084

Accepted Manuscript
Oxadiazole-substituted naphtho[2,3-b]thiophene-4,9-diones as potent inhibitors of
keratinocyte hyperproliferation. Structure−activity relationships of the tricyclic quinone
skeleton and the oxadiazole substituent
Atila Basoglu, Simone Dirkmann, Nader Zahedi Golpayegani, Silke Vortherms, Jan
Tentrop, Dominica Nowottnik, Helge Prinz, Roland Fröhlich, Klaus Müller
PII:
S0223-5234(17)30252-0
10.1016/j.ejmech.2017.03.084
EJMECH 9344
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 11 January 2017
Revised Date: 1 March 2017
Accepted Date: 31 March 2017
Please cite this article as: A. Basoglu, S. Dirkmann, N.Z. Golpayegani, S. Vortherms, J. Tentrop, D.
Nowottnik, H. Prinz, R. Fröhlich, K. Müller, Oxadiazole-substituted naphtho[2,3-b]thiophene-4,9-diones
as potent inhibitors of keratinocyte hyperproliferation. Structure−activity relationships of the tricyclic
quinone skeleton and the oxadiazole substituent, European Journal of Medicinal Chemistry (2017), doi:
10.1016/j.ejmech.2017.03.084.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.

ACCEPTED MANUSCRIPT
Oxadiazole-substituted naphtho[2,3-b]thiophene-4,9-diones as potent inhibitors of
keratinocyte hyperproliferation. Structure−activity relationships of the tricyclic
quinone skeleton and the oxadiazole substituent
†
†
†
Atila Basoglu, Simone Dirkmann, Nader Zahedi Golpayegani, Silke
†
†
†
†
‡
Vortherms, Jan Tentrop, Dominica Nowottnik, Helge Prinz, Roland Fröhlich,
,
†
and Klaus Müller*
†
Institute of Pharmaceutical and Medicinal Chemistry, PharmaCampus,
Westphalian Wilhelms University, Corrensstraße 48, D-48149 Münster, Germany
‡
Organic Chemistry Institute, Westphalian Wilhelms University, Corrensstraße
40, D-48149 Münster, Germany

1
ACCEPTED MANUSCRIPT
Oxadiazole-substituted naphtho[2,3-b]thiophene-4,9-diones as potent inhibitors
of keratinocyte hyperproliferation. Structure−activity relationships of the
tricyclic quinone skeleton and the oxadiazole substituent
†
†
†
†
†
Atila Basoglu, Simone Dirkmann, Nader Zahedi Golpayegani, Silke Vortherms, Jan Tentrop,
†
†
‡
,†
Dominica Nowottnik, Helge Prinz, Roland Fröhlich, and Klaus Müller*
†
Institute of Pharmaceutical and Medicinal Chemistry, PharmaCampus, Westphalian Wilhelms
University, Corrensstraße 48, D-48149 Münster, Germany
‡
Organic Chemistry Institute, Westphalian Wilhelms University, Corrensstraße 40, D-48149
Münster, Germany
*
Corresponding author: Tel.: +49 251 833 3324
E-mail address: kmuller@uni-muenster.de

2
ACCEPTED MANUSCRIPT
Abstract
Novel analogues of oxadiazole-substituted naphtho[2,3-b]thiophene-4,9-diones were synthesized in
which the tricyclic quinone skeleton was systematically replaced with simpler moieties, such as
structures with fewer rings and open-chain forms, while the oxadiazole ring was maintained. In
addition, variants of the original 1,2,4-oxadiazole ring were explored. Overall, the complete three-
ring quinone was essential for potent suppression of human keratinocyte hyperproliferation, whereas
analogous anthraquinones were inactive. Also, the oxadiazole ring per se was not sufficient to elicit
activity. However, rearrangement of the heteroatom positions in the oxadiazole ring resulted in
highly potent inhibitors with compound 24b being the most potent analogue of this series showing
an IC in the nanomolar range. Furthermore, experiments in isolated enzymatic assays as well as in
5
0
the keratinocyte-based hyperproliferation assay did not support a major role of redox cycling in the
mode of action of the compounds.
Keywords:

3
ACCEPTED MANUSCRIPT
Antiproliferative Activity
HaCaT Cells
Quinone
Redox Activation
Superoxide Radical
Abbreviations:
7-AAD, 7-aminoactinomycin D; CPR, NADPH-cytochrome P450 oxidoreductase; DCC, N,N’-
dicyclohexylcarbodiimide; DHE, dihydroethidium; DMAP, 4-(dimethylamino)pyridine; DTPA,
diethylenetriaminepentaacetic acid; NQO-1, NAD(P)H:quinone oxidoreductase 1; ROS, reactive
oxygen species; rt, room temperature; SAR, structure−activity relationship; SOD, superoxide
dismutase.
1. Introduction

4
ACCEPTED MANUSCRIPT
Naphthoquinones constitute a class of naturally occurring compounds found in animals, plants and
microorganisms [1]. Due to their biological and structural properties they are regarded as privileged
structures in Medicinal Chemistry [2], and because of their demonstrated activity particularly
against cancer cells these agents have been the subject of intensive research [3,4]. The exact
molecular mechanism by which naphthoquinones exert their biological action still needs to be
elucidated. At least in part, the leading theory holds that the quinone moiety reacts with two major
enzymes such as NADPH-cytochrome P-450 oxidoreductase (CPR) and NAD(P)H:quinone
oxidoreductase 1 (NQO-1), leading to redox cycling between the quinone and the semiquinone
radical or hydroquinone forms (Fig. 1), thereby generating superoxide and even more reactive
oxygen species (ROS) derived therefrom that could react with biological targets [5-11].
Prominent examples of bioactive naphthoquinones (Fig. 2) include ortho-quinoid β-lapachone (1)
and the linearly anellated naphtho[2,3-b]furan-4,9-diones 2 (napabucasin) and 3, which were
isolated from the heartwood of the lapacho (Tabebuia) tree of the Bignoniaceae family together with
a variety of other naphthoquinones and anthraquinones [12-14]. The inner bark, commonly known
as “pau-d’arco”, is used as an analgesic, an antiinflammatory, and an antineoplastic by the native
people in South America [15,16]. The biological significance of these quinones in cancer therapy
has stimulated enormous research interest in this class of compounds [17-22]. Furthermore, we
have previously reported that the lapacho tree provided some naphthoquinones with comparable
activity against the growth of human keratinocytes relative to the antipsoriatic drug anthralin (4, Fig.
2) [23].
In our continuous interest in tricyclic antiproliferative agents such as anthracenones [24]
naphtho[2,3-b]thiophen-4(9H)-ones [25], acridones [26], phenoxazines and phenothiazines [27] we
prepared novel analogues of napabucasin (2) and studied the influence of side chain modification on

5
ACCEPTED MANUSCRIPT
the potency of these agents to suppress keratinocyte hyperproliferation [28]. Such a feature may be
useful in the treatment of hyperproliferative skin disorders such as psoriasis, which is mainly
characterized by excessive growth of keratinocytes [29]. In a subsequent study, we reported the
synthesis and structure−activity relationships (SAR) for a series of compounds in which the furano-
oxygen of the parent naphtho[2,3-b]furan-4,9-dione core was replaced by other heteroatoms [30].
Of these heterocyclic ring variants, the 8-hydroxynaphtho[2,3-b]thiophene-4,9-dione analogue
bearing an oxadiazole ring on the 2-position of the tricyclic skeleton (5, Fig. 2) was the most potent
inhibitor of keratinocyte hyperproliferation [30]. In this context, it is of interest to note that several
compounds possessing an oxadiazole moiety have recently been reported as antiproliferative agents
[
31-35].
To further explore structure–activity relationships for the class of linearly anellated tricyclic
quinones, we have now focused on identifying the minimm structural requirements for inhibitory
action against keratinocyte hyperproliferation on the one hand, and, on the other, elucidating the
role of the 2-oxadiazole substituent. Toward this goal, we compared the inhibitory potential of
certain structural analogues in which the tricyclic quinone skeleton of naphtho[2,3-b]thiophene-4,9-
dione 6 (Fig. 2) has been systematically replaced with simpler moieties while the 1,2,4-oxadiazole
fragment was kept unchanged. We also report on the modification of the 1,2,4-oxadiazole ring of 6,
which resulted in a highly potent inhibitor of keratinocyte hyperproliferation. The biological testing
procedures applied in this study were similar to those in our earlier work [28,30].
2. Chemistry

6
ACCEPTED MANUSCRIPT
The syntheses of the desired substitution patterns and oxadiazole ring variants are shown in
Schemes 1-6. Scheme 1 shows the preparation of 6 and its modified analogues 13a–13c and 13l.
The required electron-rich 4,9-dimethoxynaphtho[2,3-b]thiophene (8) was prepared from
unsubstituted quinone skeleton 7 by reductive methylation with sodium dithionite and dimethyl
sulfate in the presence of tetrabutylammonium bromide. Regioselective 2-lithiation of the
thiophene ring followed by carboxylation with dry ice yielded the activated carboxylic acid 9a,
which was then converted to 1,2,4-oxadiazole 13a with N-hydroxypropionamidine in a one-pot
reaction [36]. Oxidative deprotection of 13a with diammonium cerium(IV) nitrate afforded quinone
6
. In a similar fashion, naphthothiophens 13b and 13c, which lack the methoxy group at the 9-
position or both methoxy groups at the 4,9-positions, respectively, were prepared from precursors
0 and 12. 4-Hydroxy analogue 13l was obtained from its methoxy precursor 13b by ether cleavage
1
with boron tribromide.
The preparation of analogous compounds 13d–13f and 13m–13u (Charts 1, 3 and 4), in which the
tricyclic quinoid system was downsized or replaced by other moieties, was also accomplished by
reaction of the activated carboxylic acids and N-hydroxypropionamidine (Scheme 2). The requisite
carboxylic acids were prepared as described in the Supplementary content or according to literature
methods. Analogues comparable to 6 in which the 1,2,4-oxadiazole was appended to a quinone
moiety (13g–13k, Chart 2) were obtained from their corresponding para-dimethoxy aromatic
precursors by oxidative demethylation as described above.
Scheme 3 outlines the synthesis of 5-ethyl-1,2,4-oxadiazole 19, which is a substitution isomer of 6
where the 3,5-substituents at the oxadiazole ring have been exchanged with each other.
Carbaldehyde 14 was transformed into the corresponding oxime 15 and dehydrated to nitrile 16,
which upon addition of hydroxylamine furnished amidoxime 17. Ring closure with activated
propionic acid in a one-pot reaction in pyridine following the method of Borg et al. [36] provided

7
ACCEPTED MANUSCRIPT
oxadiazole 18, which was then converted to the quinoid structure 19. In a similar fashion,
anthraquinone-substitued oxadiazole 19a (Table 1), a substitution isomer of 13g, was prepared from
9,10-dimethoxyanthracene-2-carbaldehyde (see Supplementary content).
A series of related 1,3,4-oxadiazoles and 1,3,4-thiadiazoles was synthesized as shown in Scheme 4.
Steglich esterification [37] of carboxylic acid 9a, followed by reaction of methyl ester 20 with
hydrazine, provided hydrazide 21. Condensation of 21 and appropriate commercially available
®
orthoesters under microwave irradiation, which was catalyzed by solid supported Nafion NR50
according to the method of Polshettiwar [38], afforded 1,3,4-oxadiazoles 22a–22e. Attempts to
extend this method for the preparation of 1,3,4-thiadiazole analogues 23a–23c using phosphorus
pentasulfide in aluminum oxide as thionating agent were met with difficulty. While trimethyl
orthoformate yielded 23a as the main product, analysis of the reaction products showed that other
orthoesters led predominantly to the formation of oxadiazoles, indicative of incomplete thionation
of the hydrazides. To circumvent this problem, hydrazide 21 was refluxed with the thionating agent
in acetonitrile. To our surprise, the expected thiohydrazide was already cyclized to 23b under these
conditions, suggesting that acetonitrile reacted with the thiohydrazide followed by cyclization to
form the 5-methyl-1,3,4-thiadiazole ring. When acetonitrile was replaced by dioxane and triethyl
orthopropionate, moderate yields of the 5-ethyl-1,3,4-thiadiazole 23c were obtained. In the final
step, deprotection of the hydroquinone ethers 22a–22e and 23a–23c gave the corresponding target
quinones 24a–24e and 25a–25c, respectively.
The syntheses of the 2-substituted open-chain analogues of oxadiazole 6 are schown in Scheme 5.
The Tanaka procedure [39] was used to prepare the key intermediates 9a and 14, which involved
metalation of 8 with n-butyllithium and quenching the resulting anion with dry ice or N,N-
dimethylformamide, respectively. Nitrile 16 was obtained from 14, whereas ester 20 and hydrazide
21 were obtained from 9a as described in Schemes 3 and 4, respectively. Reaction of nitrile 16 with

8
ACCEPTED MANUSCRIPT
sodium azide in DMF resulted in tetrazole 26. The α,β-unsaturated carbonyl compounds 27 and 28
were prepared from carbaldehyde 14 in an aldol reaction. The final deprotection step yielded
analogues 29–36. Hydrazones 37 and 38 were directly prepared from deprotected aldehyde 32 and
the appropriate hydrazines.
Finally, nitration of naphtho[2,3-b]thiophene-4,9-dione (7) according to a literature procedure [40]
gave rise to an isomeric mixture of the 2- and 3-nitro analogues (Scheme 6), which were separable
by chromatography. NMR data did not allow unambiguous spectral assignment of the position of
the nitro group. However, X-ray crystal structures of 39a and 39b (see Supplementary content for
details) unequivocally confirmed the position of the nitro substituent for both isomers. The 2-nitro
isomer 39a was then reduced with sodium dithionite to the 2-amino analogue 40. Schotten-
Baumann reaction [41] with aliphatic acid chlorides in pyridine or aromatic acid chlorides in
dichloromethane provided the aliphatic amides 41a–41c and aromatic amides 41d and 41e,
respectively (Scheme 6).
3. Biological evaluation and discussion
3.1. Suppression of keratinocyte hyperproliferation and SAR for the three-ring skeleton and the
oxadiazole ring
The ability of the compounds to suppress keratinocyte hyperproliferation was measured using a
spontaneously transformed human keratinocyte (HaCaT) cell line [42] which displays a
keratinization pattern typical of that seen in the psoriatic epidermis. These adherent cells grow as a
layer and are very good mimics for hyperproliferative skin, making them a higly relevant model for
psoriasis and for the preclinical evaluation of potential antipsoriatic agents [43]. The data for the

9
ACCEPTED MANUSCRIPT
antipsoriatic agent anthralin (4) and the naturally occurring β-lapachone (1) obtained in this assay
are shown for comparison (Table 1). IC values for antihyperproliferative activity of the novel
5
0
compounds are recorded in Charts 1–4 and Tables 1 and 2.
The compounds prepared were designed to answer two questions concerning SAR for inhibitory
activity against keratinocyte hyperproliferation among the 2-substituted naphtho[2,3-b]thiophene-
4,9-diones. For this purpose, three series of compounds were studied. In order to more closely
delineate the minimum requirements for activity, we have prepared analogues of the most active
representative, 2-oxadiazole-substituted 5, in which drastic alterations of the original tricyclic
quinone structure were made to see whether compounds with fewer rings and open-chain forms are
still active (Charts 1–4). As the tricyclic skeleton contains the quinone motif, it may lead to redox
reaction and nonspecific interaction with sulfur nucleophiles such as cysteine present in proteins
[
44]. The latter feature, however, can be ignored in the case of the heterocycle-fused
naphthoquinones. As compared to menadione, which has a nonfused quinone ring, these agents do
not have an unsubstituted β-carbon open to nucleophilic attack and cannot form adducts as a
Michael acceptor. Nonetheless, they are redox active by virtue of their quinone moiety [28,30]. To
shed some more light on a possible nonspecific contribution of the quinone moiety in the context of
the compounds under study, we also included two anthraquinone-analogous and three nonfused
naphthoquinone-analogous structures of naphtho[2,3-b]thiophene-4,9-dione 6 (13g–13k, Chart 2,
19a, Table 1).
Compounds depicted in Tables 1 and 2, however, explore further modifications of the 2-oxadiazole
substituent of 5 and are either oxadiazole ring variants (Table 1) or open-chain analogues (Table 2).
The decision to remove the 8-hydroxy group in 5 is based on an observation from our earlier study
that this group is associated with release of larger amounts of lactate dehydrogenase into the culture
medium and responsible for membrane-damaging effects [30]. As recently reported, this drawback

1
0
ACCEPTED MANUSCRIPT
can be eliminated by substitution with a chloro or an hydrogen atom, even though the 8-chloro-
substituted compound was somewhat less potent [45]. As a compromise, 8-unsubstituted 6, a
closely related congener of 5, was chosen for further SAR studies.
3.1.1. Simplification of the tricyclic quinone skeleton
Some of the para-dimethoxy substituted precursors of the quinones were tested for inhibition of
keratinocyte hyperproliferation, along with unsubstituted three or two ring-fused analogues (13a–
1
3f, Chart 1). None of these are active except for the 4,9-dimethoxynaphthothiophene derivative
3a, which, however, is almost 20-fold less potent than 6. Chart 2 reveals that a quinone structure
1
alone is not sufficient for potency. Those analogues in which either the fused phenyl (13k) or
thiophene ring (13i) of 6 was omitted display only moderate activity, whereas an anthraquinone-
based analogue (13g) of 6 as well as naphthoquinone 13h, where the oxadiazole ring is directly
attached to the quinone ring, are both inactive (defined as an IC of > 30 µmol/L). In the tricyclic
5
0
compounds of Chart 3, either one of the carbonyl oxygen atoms was removed (13l, 13p) or one
carbonyl group of the quinone was replaced with an oxygen (13m, 13o) or a sulfur atom (13n).
While in the latter cases, as documented by xanthones 13m and 13o as well as thioxanthone 13n,
activity is lost, anthrone-type agents 13l and 13p are about 20-fold less potent than lapacho
analogue 6. In addition, the compounds of Chart 4 were prepared to study the further dissection of
6
into simpler analogues such as 13q, which completely lacks the 4-carbonyl group, 13s, which
retains the thiophene ring together with the original 4-carbonyl group, 13u with only the thiophene
ring left at the oxadiazole, or the corresponding phenyl analogues of the last-mentioned compounds,
i. e. 13r and 13t, respectively. Taken together, all these modifications of the tricyclic quinone
structure of 6 are accompanied by loss of activity or reduced potency against keratinocyte
hyperproliferation. Accordingly, the results from these downsized or modified structures lead to the

1
1
ACCEPTED MANUSCRIPT
conclusion that the intact tricyclic molecular skeleton is the essential moiety required for
antihyperproliferative activity. Our results also question whether an oxadiazole moiety per se is
sufficient to elicit an antiproliferative action. Initially, the oxadiazole ring was considered to be a
bioisostere for esters or amides [46]. More recently, a body of evidence has emerged that suggests a
role for a 1,2,4- or a 1,3,4-oxadiazole ring system in antiproliferative active agents [31-34], which is
not supported by this study, at least with respect to the action against keratinocyte
hyperproliferation.
Wth respect to the quinone motif, the results obtained with naphthoquinones 13h–13k (Chart 2),
menadione (Table 1), anthraquinones 13g (Chart 2) and 19a (Table 1) clearly show that the potent
action of the linearly anellated naphtho[2,3-b]thiophene-4,9-diones against keratinocyte
hyperproliferation is not simply the consequence of their quinone structure, since in particular both
anthraquinones in which the anellated thiophene ring is replaced by a phenyl ring are inactive.
3.1.2. Oxadiazole ring variants
In the following compound series, variants of the original 3-ethyl-1,2,4-oxadiazole ring have been
explored, and data for suppression of keratinocyte hyperproliferation are collected in Table 1.
When the tricyclic skeleton in position 5 of the 1,2,4-oxadiazole of 6 has been switched with the
ethyl group of the 3-position, as exemplified by its isomer 19, potency is still in the low micromolar
range, but somewhat reduced as compared to 6 (IC of 0.58 µmol/L vs. 0.36 µmol/L). However,
5
0
when the naphtho[2,3-b]thiophene-4,9-dione moiety of 19 has been exchanged for an anthraquinone
core (19a), activity is lost, confirming the results obtained with anthraquinone 13g (Chart 2), which
is an isomer of 19a. By contrast, modification of the heteroatom positions in the oxadiazole ring
from a 1,2,4- to a 1,3,4-arrangement resulted in highly potent inhibitors with IC values in the
5
0

1
2
ACCEPTED MANUSCRIPT
submicromolar range. The corresponding 1,3,4-isomer 24c was more potent than 1,2,4-arranged 6
(
IC = 0.15 µmol/L vs. IC = 0.36 µmol/L).
50 50
Compounds 24a–24e explore the consequences of varying the length of the 5-alkyl chain on the
,3,4-oxadiazole ring. Extending the chain of the original ethyl group of the 1,3,4-isomer 24c by
1
two carbon atoms to give butyl substituted 24d reduced potency. On the other hand, when the chain
was shortened by one carbon atom to provide the methyl substituted 24b, potency was improved,
but did not show further improvement when the methyl group was removed as in 24a. Replacement
of the alkyl group with a phenyl ring gave an inactive compound (24e). By comparing the
homologous series of 5-alkyl-1,3,4-oxadiazoles 24a–24e, peak activity against keratinocyte
hyperproliferation (IC = 0.08 µmol/L) is shown by 24b, which is the most potent compound seen
5
0
so far under our assay conditions. Exchanging the 1,3,4-oxadiazole for a 1,3,4-thiadiazole ring
resulted in analogues 25a–25c. The inhibitory potency determined for these compounds clearly
suggests that the oxadiazole is preferred relative to the thiadiazole, i. e. 24b, IC of 0.08 µmol/L
5
0
compared to 25b, IC of 3.70 µmol/L.
5
0
3.1.3. Open-chain analogues
A third series of compounds was prepared for comparison with 6 and to evaluate SAR for open
side-chain substituents. A number of carbonyl-containing groups such as esters and amides that
would be mimicked by the oxadiazole ring have been explored at the 2-position of the naphtho[2,3-
b]thiophene-4,9-dione (Table 2). There was a clear distinction in hyperproliferation inhibitory
properties between the free acid 29 and its corresponding methyl ester 30. While the free acid is not
active and may not be able to cross the keratinocyte membrane, ester 30 is a potent inhibitor of
keratinocyte hyperproliferation, with an IC of 0.71 µM in the same order of magnitude to those of
5
0
anthralin and β-lapachone.

1
3
ACCEPTED MANUSCRIPT
Also, when the carboxylic acid group of 29 was replaced by a tetrazole-5-yl group (34), which was
used as a nonclassical bioisosteric replacement, as it has comparable acidity to that observed for the
carboxylic acids while being about ten times more lipophilic [47], activity was not apparent in this
assay. As with ester 30, hydrazide 31 and aldehyde 32 display activity, though slightly reduced but
3–5-fold better than the 2-methyl analogue 43. Comparing chalcones 35 and 36 as well as
hydrazones 37 and 38 with 6 shows that opening the oxadiazole ring results in the loss of inhibitory
activity for each analogue. This is also the case for compound 40 having a 2-amino group. Activity
is regained by converting the amine into amides 41a–41e. Within this homologous series of small
and branched alkyl groups (41a–41c), these compounds are active in the micromolar range, but 9–
27-fold less potent than oxadiazole 6. Benzamides 41d and 41e are somewhat more potent than the
aliphatic amides, and their potency is similar to that of the two nitro derivatives 39a and 39b.
The most potent compounds of the open-chain analogues are nitrile 33 and the 2-acetylated 42 (IC50
of 0.44 µmol/L), which is the sulfur analogue of the original lapacho constituent napabucasin (2),
and these compounds are essentially equipotent. Altogether, smaller groups at the 2-position may
be advantageous for potency, and there also seems to be a requirement for an electron-withdrawing
group.
3.2. Enzymatic redox activation
As previously reported [28], selected compounds were subjected to redox cycling experiments
where superoxide generation was evaluated both by the CPR- and NQO-1-catalyzed one- and two-
electron reduction, and this was also confirmed in keratinocyte-based assays. Typically, the
quinones are reduced in two sequential reactions to generate hydroquinones via semiquinone
radicals that can react back to the parent compounds, thus perpetuating a redox cycle (Fig. 1). As a
result, these reduced forms of the quinones rapidly produce superoxide as the initial species by

1
4
ACCEPTED MANUSCRIPT
electron transfer to molecular oxygen. Dismutation of superoxide yields hydrogen peroxide, which
can react with trace amounts of ferrous ions to form even more reactive species, such as hydroxyl
radicals. If accumulation of reactive oxygen species overwhelms the antioxidant capacity of the
cell, this can lead to oxidative stress [48], failure of normal cellular functions and even cell death
[
49]. The ability of the naphtho[2,3-b]thiophene-4,9-diones to stimulate superoxide generation was
determined as described [28], and the results are presented in Table 3. The known redox cycler
menadione [50,51] (2-methylnaphthalene-1,4-dione) served as a positive control.
All the tested compounds were significant generators of superoxide in the one-electron reduction
assay catalyzed by human recombinant CPR, and except for the basic structure 7, also in the two-
electron reduction by human recombinant NQO-1, though the contribution of the latter enzyme was
somewhat less pronounced. The rates of superoxide generated by the lapacho compounds reflect an
efficient interaction of their corresponding semiquinone radicals with molecular oxygen resulting in
redox cycling and concomitant superoxide production. In contrast to the other agents tested, 2-nitro
substituted 39a was also an excellent substrate for NQO-1 and generated the highest amount of
superoxide compared to other functional groups such as amino (40), acetyl (42), or methyl (43) at
C-2 of the tricyclic quinone structure. This is not surprising, as in addition to quinones, NQO-1 is
capable of catalyzing two-electron reduction of a wide variety of functional groups such as aromatic
nitro compounds and can even catalyze four-electron reduction of the nitro group [52]. 2-Nitro-
substituted 39a was actually synthesized to study the effect of an additional structural feature in the
molecule capable of redox cycling. Such a strategy has very recently been applied successfully to
the design and synthesis of quinone-based triazoles possessing a second redox center [53].
3.3. Superoxide generation in keratinocytes
The amount of superoxide production was also directly determined in a keratinocyte-based assay
under the conditions comparable to that of the hyperproliferation model in HaCaT cells to confirm

1
5
ACCEPTED MANUSCRIPT
that an excessive amount of superoxide by redox cycling of the quinones is generated within whole
cells. The fluorescence intensity arising from the product of the reaction of superoxide with the
fluorescent probe dihydroethidium (DHE) was monitored by flow cytometry [54]. The oxidized
+
form, 2-hydroxyethidium (2-OH-E ), is a measure of intracellular superoxide generation and a
specific marker for superoxide [55,56]. As a positive control in these experiments we have also
employed menadione, for which DHE was used to detect intracellular superoxide generation after
treatment of cells with this quinone [55,57].
In most cases, significantly increased superoxide generation following both acute and chronic
treatment of HaCaT keratinocytes with the selected lapacho compounds was observed, as quantified
by the mean fluorescence intensities (Table 4). When keratinocytes were pretreated with
dicoumarol, a commonly used inhibitor of NQO-1 [50,58] that competes with NADH for binding
and prevents reduction of quinones, superoxide generation by the lapacho analogues was still
significant as compared to controls. Accordingly, inhibition of the NQO-1-promoted two-electron
metabolism of the quinones in keratinocytes by dicoumarol may increase the accessibility of these
agents for one-eletron reduction by CPR, and nevertheless allow efficient redox cycling with
+
concomitant superoxide release via this alternative pathway (Fig. 1). As a consequence, 2-OH-E
fluorescence is still enhanced and points towards a predominant role of the CPR pathway under
these conditions. This is in line with the observation that all quinones were also activated by CPR
in the isolated enzyme assay (Table 3).
As expected, also in the cellular assay we observed that the nitro aromatic analogue 39a produced
higher amounts of superoxide than compounds with other substituents in the 2-position of the
tricyclic quinone skeleton. This is consistent with the well-known concept that reduction of nitro
aromatic compounds by NADH-enzymes is initiated by means of the addition of an electron to
generate the corresponding nitro radical anions. In the presence of oxygen, the nitro radical anions
can rapidly interfere with oxygen to regenerate the original nitro compounds and produce

1
6
ACCEPTED MANUSCRIPT
superoxide [59]. Moreover, the situation with a nitro group is further complicated by the fact that
its consecutive reduction involves a nitroso derivative, a nitroxyl radical and the hydroxylamino
functionality which in turn may also undergo redox cycling via the nitroxyl radical to produce
superoxide [60]. The possible consequences of oxidative stress [48] in human keratinocytes with
respect to the potential therapeutic action of the naphtho[2,3-b]thiophene-4,9-diones in the
treatment of hyperproliferative skin disorders such as psoriasis have been discussed in detail
previously [28].
However, while there are clear SARs for the structural requirements of the compounds to inhibit
keratinocyte hyperproliferation, their potential to redox cycle does not correlate with their ability to
arrest or slow down cell proliferation events. Thus, the nitro aromatic compound 39a was prepared
to check the contribution of an additional redox center at the tricyclic skeleton, which is reflected by
the highest amounts of superoxide generated in both enzymatic and cell-based assays, but this
analogue shows only mediocre activity in the keratinocyte model. On the other hand, amino
analogue 40, which is completely inactive against keratinocyte hyperproliferation, produces rates of
superoxide comparable to those of potent analogues such as 6 or 42. Accordingly, our present data
do not support a major role of redox cycling in the mode of action of the lapacho analogues.
3.4. Detection of apoptotic cells using 7-aminoactinomycin D (7-AAD)
Then, we examined apoptosis from FACS analysis of HaCaT cells using 7-AAD staining for
quantification following treatment with the compounds for 18 h. Cells in the early phase of
apoptosis stain positive for 7-AAD in flow cytometry, depending on the concentration of the
fluorescent intercalator [61]. When HaCaT keratinocytes were examined after chronic treatment
with representative lapacho analogues (5 µmol/L), induction of cell death was enhanced as
compared to controls but not significant (Table 4). Interestingly, oxadiazole analogue 6 produced
an exceptionally high level of apoptotic cells. Furthermore, cells could be substantially protected by

1
7
ACCEPTED MANUSCRIPT
pretreatment of HaCaT keratinocytes with the NQO-1 inhibitor dicoumarol suggesting the
involvement of this enzyme in the activation of this compound. Finally, relatively high levels of
superoxide produced by the most potent superoxide generator of this series, analogue 39a, did not
correlate with the potency of this agent to trigger cell death after chronic treatment. Therefore, the
crucial species responsible for the biological action of the compounds that can interact with cellular
macromolecules such as DNA, lipids or proteins and ultimately cause apoptosis of the cell needs
still to be identified.
4. Conclusions
In this SAR study of oxadiazole-substituted naphtho[2,3-b]thiophene-4,9-diones, the importance of
the tricyclic quinone skeleton was explored through the stepwise degradation of this moiety while
the oxadiazole ring was kept unchanged. However, removal of one of the anellated rings or
modifications within the quinone moiety were poorly tolerated, leading to decreases or loss in
potency, suggesting the need for an intact heterocycle-fused naphthoquinone, whereas an oxadiazole
ring per se was not sufficient for activity.
On the basis of their inhibitory effects on HcCaT cell hyperproliferation, it was further revealed that
the replacement of the 1,2,4-oxadiazole by a 1,3,4-oxadiazole ring in the structure of 6 improved its
antihyperproliferative ability. Moreover, within this particular class, further modifications of the 5-
alkyl side chain have marked effects on potency. As a result, a methyl substituent was highly
beneficial and led to the identification of 24b, which with an IC of 0.08 µmol/L was the most
5
0
potent inhibitor of keratinocyte hyperproliferation we have obtained under the assay conditions.
These findings support further development of this class of compounds for psoriasis or epidermal
hyperplasia resulting from other causes.

1
8
ACCEPTED MANUSCRIPT
5. Experimental section
5.1.
Chemistry
1
Melting points were determined with a Kofler melting point apparatus and are uncorrected.
H
1
3
NMR (400 MHz, 600 MHz) and C NMR (151 MHz, 100 MHz) spectra were recorded on a Varian
Mercury 400 plus or an Agilent DD2 600 spectrometer, using tetramethylsilane as an internal
standard. Fourier transform IR spectra were recorded on a Shimadzu spectrometer type Miracle 10
(attenuated total reflection, ATR) by applying ATR correction. Mass spectra (ESI or APCI) were
recorded on a Finnigan MAT GCQ instrument (70 eV) or on a Bruker-Daltonics MicroTOF QII
instrument. Chromatography refers to column chromatography, which was performed on silica gel
(
0.063–0.200 mm, 6 nm, Macherey-Nagel) with CH Cl as the eluant unless otherwise stated.
2 2
Yields have not been optimized. The purity of all tested compounds was ≥95% as determined by
analytical reversed-phase HPLC (Ultimate 3000, Thermo-Fisher) or elemental analyses.
5
5
.1.1. General procedure for the oxidative demethylation of hydroquinone dimethyl ethers
.1.1.1. 2-(3-Ethyl-1,2,4-oxadiazol-5-yl)naphtho[2,3-b]thiophene-4,9-dione (6)
To a suspension of 13a (0.30 g, 0.88 mmol) in MeCN (8 mL) and water (1.5 mL) at 0 °C was added
dropwise within 20 min (NH ) [Ce(NO ) ] (1.50 g, 2.74 mmol) in MeCN (2.5 mL) and water (2.5
4
2
3 6
mL) under vigorous stirring. The mixture was allowed to react for 30 min under the same
conditions. Then it was poured onto ice-water (50 mL) and extracted with CH Cl (3 × 50 mL).
2
2
The combined organic phase was washed with a saturated solution of NaCl (2 × 50 mL), dried over

1
9
ACCEPTED MANUSCRIPT
Na SO , then concentrated, and purified by chromatography to afford light yellow, felted needles;
2
4
-
1 1
8
7
9% yield; mp 182 °C; FTIR 1681, 1645 cm ; H NMR (CDCl ) δ 8.38 (s, 1H), 8.32–8.24 (m, 2H),
3
1
3
.88–7.78 (m, 2H), 2.86 (q, J = 7.6 Hz, 2H), 1.40 (t, J = 7.6 Hz, 3H); C NMR (CDCl , 101 MHz)
3
δ 178.70, 178.20, 172.93, 169.79, 148.61, 143.18, 134.72, 134.38, 133.63, 133.39, 133.16, 129.89,
+
1
27.87, 127.44, 19.98, 11.59; MS (APCI) m/z 311.0418, calcd for C H N O S [M + H] ,
1
6
10
2
3
311.0485.
Analogously, target compounds 13g–13k, 19, 19a, 24a–24e, 25a–25c, 29–36, were prepared from
the corresponding hydroquinone dimethyl ethers. See Supplementary content for details.
5
5
.1.2 General procedure for the reductive methylation of quinones
.1.2.1. 4,9-Dimethoxynaphtho[2,3-b]thiophene (8)
To a solution of 7 [40] (5.00 g, 23.34 mmol) and tetrabutylammonium bromide (77 mg, 0.24 mmol)
in dry THF (220 mL) under N sodium dithionite (24.39 g, 140 mmol) in water (110 mL) was added
2
dropwise. The mixture was vigorously stirred for 30 min and then treated with sodium hydroxide
(28.04 g, 500 mmol) in water (110 mL). After 30 min it was cooled to 0 °C on an ice-bath,
dimethyl sulfate (26.68 mL, 280 mmol) was added, and the mixture was then stirred for 2 h at 0 °C.
Then it was allowed to warm to rt, stirred for an additional 2 h, and extracted with CH Cl (3 × 200
2
2
mL). The combined organic phase was washed with water (3 × 200 mL), then with a saturated
solution of NaCl (200 mL), dried over Na SO and evaporated. The residue was purified by
2
4
-1 1
chromatography to afford yellow crystals; 83% yield; mp 156 °C; FTIR 1620, 1543 cm ; H NMR
(
CDCl ) δ 8.28 (dd, J = 6.2, 3.4 Hz, 1H), 8.22 (dd, J = 6.3, 3.4 Hz, 1H), 7.58 (d, J = 5.6 Hz, 1H),
3
7.55–7.48 (m, 2H), 7.43 (d, J = 5.6 Hz, 1H), 4.15 (s, 3H), 4.12 (s, 3H); MS (APCI) m/z 245.0639,
+
calcd for C H O S [M + H] , 245.0631.
1
4
12
2

2
0
ACCEPTED MANUSCRIPT
5.1.3. 4,9-Dimethoxynaphtho[2,3-b]thiophene-2-carboxylic acid (9a)
To a mixture of n-butyllithium (1.6 mol/L, 5.40 mL, 8.60 mmol) in dry THF (8.5 mL) at -30 °C
under N was added dropwise a solution of 8 (1.07 g, 4.37 mmol) in dry THF (52 mL). The mixture
2
was stirred and cooled to -78 °C, and then dry ice (22.00 g, 500 mmol) was added in portions under
N . The mixture was stirred at -78 °C for 10 min, then allowed to warm to -20 °C within 30 min,
2
and then poured into HCl (2 mol/L, 50 mL). The organic phase was separated and the aqueous
phase extracted with ether (3 × 50 mL). The combined organic phase was extracted with NaOH (1
mol/L, 3 × 50 mL) and the aqueous phase acidified with HCl (2 mol/L) to pH 2. The carboxylic
acid was extracted CH Cl (3 × 200 mL), the organic phase dried over Na SO and then the solvent
2
2
2
4
-1 1
removed in vacuo to afford an orange powder; 76% yield; mp 258 °C; FTIR 1681 (CO H) cm ; H
2
NMR (DMSO-d ) δ 8.28–8.24 (m, 1H), 8.23 (s, 1H), 8.19–8.15 (m, 1H), 7.70–7.58 (m, 2H), 4.07
6
+
(
s, 3H), 3.95 (s, 3H); MS (APCI) m/z 289.0535, calcd for C H O S [M + H] , 289.0529.
1
5
12
4
Analogously, compounds 9b and 9c were prepared from 11 and 12 [62], respectively. See
Supplementary content for details.
5.1.4. 4-Methoxynaphtho[2,3-b]thiophene (11)
To a mixture of anhydrous K CO (1.87 g, 13.54 mmol) and methyl tosylate (4.52 g, 24.26 mmol,
2
3
3
.66 mL) was added dropwise a suspension of 10 [63] (2.47 g, 12.31 mmol) in CH CN (150 mL).
3
The mixture was stirred at 50 °C for 15 h. Then the solvent was evaporated, the residue treated
with CH Cl (100 mL), washed with HCl (2 mol/L, 2 × 50 mL), and then with a saturated solution
2
2
of NaCl (2 × 50 mL). The organic phase was dried over Na SO , then the solvent was evaporated
2
4
and the product purified by chromatography (cyclohexane/CH Cl , 3/1), to give a dark white
2
2
1
powder; 73% yield; mp 71 °C; H NMR (CDCl ) δ 8.31–8.23 (m, 1H), 8.14 (s, 1H), 7.95–7.84 (m,
3

2
1
ACCEPTED MANUSCRIPT
1H), 7.56 (d, J = 5.7 Hz, 1H), 7.49–7.47 (m, 1H), 7.47–7.44 (m, 1H), 7.43 (d, J = 5.7 Hz, 1H), 4.14
+
(
s, 3H); MS (APCI) m/z 215.0528, calcd for C H OS [M + H] , 215.0525.
1
3
10
5.1.5. 3-Ethyl-5-(4,9-dimethoxynaphtho[2,3-b]thiophen-2-yl)-1,2,4-oxadiazole (13a)
To a solution of 9a (1.50 g, 5.72 mmol) in dry CH Cl (30 mL) was added DCC (0.54 g, 2.60
2
2
mmol) under N , and the mixture was stirred at 0 °C for 1 h. Then it was filtered and the solvent
2
evaporated.
The residue was treated with pyridine (30 mL), and a solution of N-
hydroxypropionamidine (0.22 g, 2.51 mmol) in pyridine (6 mL) was added dropwise at rt. Then the
mixture was refluxed for 3 h, cooled to rt, poured onto ice-water (200 mL), acidified with HCl (2
mol/L), and extracted with CH Cl (3 × 50 mL). The combined organic phase was washed with
2
2
water (2 × 200 mL) and dried over Na SO . The solution was evaporated, and the product was
2
4
-
purified by chromatography to afford yellow needles; 43% yield; mp 135 °C; FTIR 2843, 1597 cm
1
1
;
H NMR (CDCl ) δ 8.39 (s, 1H), 8.31–8.20 (m, 2H), 7.61–7.51 (m, 2H), 4.18 (s, 3H), 4.16 (s,
3
3
H), 2.88 (q, J = 7.6 Hz, 2H), 1.42 (t, J = 7.6 Hz, 3H); MS m/z 341.0888, calcd for C H N O S
18 16 2 3
+
[
M + H] , 341.0954.
Analogously, compounds 13b–13f, 13m–13o, and 13q–13u were prepared from the appropriate
carboxylic acids. See Supplementary content for details.
5.1.6. 4,9-Dimethoxynaphtho[2,3-b]thiophene-2-carbaldehyde (14)
To a solution of 8 (3.68 g, 14.95 mmol) in dry THF (150 mL) at -30 °C under N was added
2
dropwise n-butyllithium (1.6 mol/L in THF, 19.63 mL, 31.40 mmol), and the mixture was stirred
for 15 min. Then N,N-dimethylformamide (2.75 g, 2.89 mL, 37.64 mmol) in THF (30 mL) was
added, and the mixture was stirred for an additional 3 h at rt. Then it was poured into ice-water
(100 mL) and extracted with ether (3 × 50 mL). The combined organic phase was washed with a

2
2
ACCEPTED MANUSCRIPT
saturated solution of NaCl (3 × 50 mL), dried over Na SO , concentrated and purified by
2
4
chromatography to afford an orange-yellow powder; 70% yield; mp 169–170 °C; FTIR 1658
-1 1
(
CHO), 1581 cm ; H NMR (CDCl ) δ 10.09 (s, 1H), 8.25–8.13 (m, 3H), 7.56–7.44 (m, 2H), 4.12
3
+
(
s, 3H), 4.07 (s, 3H); MS (APCI) m/z 273.0579, calcd for C H O S [M + H] , 273.0580.
1
5
12
3
5.1.7. 4,9-Dimethoxynaphtho[2,3-b]thiophene-2-carbaldehyde oxime (15)
A solution of 14 (2.56 g, 9.39 mmol) in dry methanol (100 mL) was heated to reflux, and a solution
of hydroxylamine hydrochloride (0.65 g, 9.39 mmol) and Na CO (0.99 g, 9.39 mmol) in water (15
2
3
mL) was added. The mixture was refluxed for 3 h and then diluted with water (100 mL). To
complete precipitation, it was kept at -20 °C for 12 h. Then the precipitate was collected by
filtration and dried under vacuum to afford a yellow powder; 79% yield; mp 132 °C; FTIR 1651,
-
1 1
1
4
603 cm ; H NMR (CDCl ) δ 8.35 (s, 1H), 8.20–8.10 (m, 2H), 7.60 (s, 1H), 7.50–7.41 (m, 2H),
3
+
.06 (s, 3H), 4.05 (s, 3H); MS (APCI) m/z 288.0626, calcd for C H NO S [M + H] , 288.0689.
1
5
13
3
5.1.8. 4,9-Dimethoxynaphtho[2,3-b]thiophene-2-carbonitrile (16)
A solution of oxime 15 (2.00 g, 7.0 mmol) in acetic anhydride (50 mL) was refluxed for 4 h. It was
allowed to cool to rt and then poured onto ice-water (200 mL). The precipitate was filtered by
suction and purified by chromatography to afford a yellow powder; 65% yield; mp 167–168 °C;
-1 1
FTIR 2210 (CN), 1585 cm ; H NMR (CDCl ) δ 8.21 (dd, J = 8.0, 1.3 Hz, 1H), 8.14 (dd, J = 7.5,
3
0.8 Hz, 1H), 8.07 (s, 1H), 7.57–7.46 (m, 2H), 4.07 (s, 3H), 4.05 (s, 3H); MS (APCI) m/z 270.0513,
+
calcd for C H NO S [M + H] , 270.0583.
1
5
11
2
5.1.9. N’-Hydroxy-4,9-dimethoxynaphtho[2,3-b]thiophene-2-carboximidamide (17)
Hydroxylamine hydrochloride (0.26 g, 3.71 mmol) in methanol (20 mL) was added to a solution of
sodium (85 mg, 3.71 mmol) in methanol (10 mL), and the mixture was stirred for 40 min at rt. A

2
3
ACCEPTED MANUSCRIPT
precipitate was filtered off, and the filtrate was treated with 16 (1.00 g, 3.71 mmol) and refluxed for
18 h. Then it was cooled and the solvent evaporated under vacuum. The residue was purified by
-1 1
chromatography to afford a yellow powder; 34% yield; mp 215–216 °C; FTIR 1643, 1589 cm ; H
NMR (CDCl ) δ 8.19–8.11 (m, 2H), 7.65 (s, 1H), 7.49–7.42 (m, 2H), 4.06 (s, 3H), 4.05 (s, 3H); MS
3
+
(
APCI) m/z 303.0705, calcd for C H N O S [M + H] , 303.0798.
1
5
14
2
3
5.1.10. 5-Ethyl-3-(4,9-dimethoxynaphtho[2,3-b]thiophen-2-yl)-1,2,4-oxadiazole (18)
Propionic acid (0.18 g, 2.47 mmol) in dry CH Cl (20 mL) at 0 °C under N was treated with DCC
2
2
2
(0.26 g, 1.24 mmol) and stirred for 1 h. A precipitate was filtered off, and the solvent was
evaporated. The residue was treated with pyridine (20 mL), and a solution of 17 (0.36 g, 1.19
mmol) in pyridine (4 mL) was added. Then the mixture was refluxed for 3 h, cooled to rt, poured
onto ice-water (200 mL), acidified with HCl (2 mol/L), and extracted with CH Cl (3 × 50 mL).
2
2
The combined organic phase was successively washed with water (2 × 200 mL), a saturated solution
of NaHCO (2 × 200 mL), HCl (2 mol/L, 2 × 200 mL), and a saturated solution of NaCl (2 × 200
3
mL). Then it was dried over Na SO , concentrated, and the residue was purified by chromatography
2
4
-1 1
to provide yellow crystals; 50% yield; mp 129–131 °C; FTIR 1558, 1358 cm ; H NMR (CDCl ) δ
3
8.23 (s, 1H), 8.23–8.12 (m, 2H), 7.47 (m, 2H), 4.09 (s, 3H), 4.09 (s, 3H), 2.96 (q, J = 7.6 Hz, 2H),
+
1
.42 (t, J = 7.6 Hz, 3H); MS m/z 341.0860, calcd for C H N O S [M + H] , 341.0954.
1
8
16
2
3
5.1.11. 2-(5-Ethyl-1,2,4-oxadiazol-3-yl)naphtho[2,3-b]thiophene-4,9-dione (19)
This was prepared from 18 as described for 6 to afford a yellow powder; 63% yield; mp 156–158
-1 1
°
C; FTIR 1662, 1573 cm ; H NMR (CDCl ) δ 8.25 (s, 1H), 8.19 (m, 2H), 7.73 (m, 2H), 2.95 (q, J
3
1
3
=
7.6 Hz, 2H), 1.41 (t, J = 7.6 Hz, 3H); C NMR (CDCl ) δ 181.73, 178.83, 178.13, 163.44,
3,
146.88, 143.00, 136.56, 134.23, 133.95, 133.56, 133.27, 127.84, 127.51, 127.08, 20.31, 10.71; MS
+
m/z 311.0417, calcd for C H N O S [M + H] , 311.0485.
1
6
10
2
3

2
4
ACCEPTED MANUSCRIPT
An anthraquinone-based analogue 19a and its precursors 15a–18a were also prepared, as described
above for 19 and 15–18, respectively. See Supplementary content for details.
5.1.12. Methyl 4,9-Dimethoxynaphtho[2,3-b]thiophene-2-carboxylate (20)
To the carboxylic acid 9a (1.50 g, 5.20 mmol) in absolute methanol (0.63 mL, 15.60 mmol) was
added DMAP (0.05 g, 0.41 mmol) under stirring. Then DCC (1.18 g, 5.72 mmol) was added at 0
°
C, and the mixture was stirred for 10 min, then for an additional 3 h at rt. The mixture was
filtered, the filtrate evaporated and dissolved in CH Cl (100 mL). The solution was washed with
2
2
HCl (0.5 mol/L, 2 × 50 mL), then with a saturated solution of NaHCO (2 × 50 mL), and the organic
3
phase was dried over Na SO . Then the solvent was evaporated and the product purified by
2
4
-1 1
chromatography to give yellow crystals; 89% yield; mp 162 °C; FTIR 1710 (CO) cm ; H NMR
(
CDCl ) δ 8.20 (m, 2H), 8.18 (s, 1H), 7.84–7.71 (m, 2H), 4.28 (s, 3H), 4.05 (s, 3H), 4.01 (s, 3H);
3
+
MS (APCI) m/z 303.0678, calcd for C H O S [M + H] , 303.0686.
1
6
14
4
5
.1.13. 4,9-Dimethoxynaphtho[2,3-b]thiophene-2-carbohydrazide (21)
To a solution of hydrazine hydrate (1.72 g, 1.67 mL, 34.44 mmol) in ethanol (25 mL) was added 20
2.60 g, 8.61 mmol) in small portions. The mixture was refluxed for 5 h, and the residue was
filtered and washed with hexane to afford a light yellow powder; 66% yield; mp 210 °C; FTIR
(
-1 1
1
620, 1589 cm ; H NMR (DMSO-d ) δ 10.21 (s, 1H), 8.33 (s, 1H), 8.26–8.19 (m, 1H), 8.16–8.11
6
1
3
(
m, 1H), 7.65–7.54 (m, 2H), 4.11 (s, 3H), 4.05 (s, 3H); C NMR (DMSO-d ) δ 160.88, 148.00,
6
1
6
45.25, 138.78, 130.95, 128.54, 126.40, 125.37, 125.33, 125.07, 122.33, 121.16, 120.99, 63.13,
+
0.68; MS (APCI) m/z 303.0808, calcd for C H N O S [M + H] , 303.0798.
1
5
14
2
3

2
5
ACCEPTED MANUSCRIPT
5.1.14. 2-(4,9-Dimethoxynaphtho[2,3-b]thiophen-2-yl)-1,3,4-oxadiazole (22a)
Hydrazide 21 (0.20 g, 0.66 mmol), trimethyl orthoformate (84 mg, 0.087 mL, 0.79 mmol) and the
®
catalyst Nafion NR50 (40 mg) were placed in a 10 mL microwave reaction vessel equipped with a
magnetic stirrer. The mixture was heated to 130 °C and irradiated at 40–140 W for 25 min. The
product was purified by chromatography to afford a yellow powder; 51% yield; mp 147–149 °C;
-1 1
FTIR 1597 cm ; H NMR (CDCl ) δ 8.45 (s, 1H), 8.23–8.19 (m, 2H), 8.15 (s, 1H), 7.54–7.45 (m,
3
1
3
2
H), 4.10 (s, 3H), 4.09 (s, 3H); C NMR (CDCl ) δ 161.05, 152.62, 148.93, 146.25, 130.81,
3
129.08, 126.82, 126.58, 126.07, 125.55, 124.62, 124.44, 122.67, 121.44, 63.64, 60.96; MS (APCI)
+
m/z 313.0650, calcd for C H N O S [M + H] , 313.0641.
1
6
12
2
3
In a similar fashion, compounds 22b–22e and 23a–23c were prepared from the appropriate
orthoesters. See Supplementary content for details.
5.1.15. 5-(4,9-Dimethoxynaphtho[2,3-b]thiophen-2-yl)-1H-tetrazole (26)
To a solution of 16 (0.23 g, 0.86 mmol) in DMF (100 mL) was added NaN (0.62 g, 9.50 mmol)
3
and NH Cl (0.51 g, 9.50 mmol), and the mixture was stirred at 90 °C for 5 h. Then it was poured
4
onto ice-water (200 mL), and the precipitate was filtered off. The filtrate was acidified with 36%
HCl to afford an additional precipitate. Purification by chromatography provided a yellow powder;
-1 1
7
8% yield; mp 251–252 °C; FTIR 1660, 1620 cm ; H NMR (DMSO-d ) δ 8.26 (s; 1H), 8.20–8.15
6
•+
(
m; 2H), 7.90–7.79 (m; 2H), 4.20 (s; 3H), 4.05 (s; 3H); MS m/z 312 (55, M ), 268 (100).
5.1.16. (E)-4-(4,9-Dimethoxynaphtho[2,3-b]thiophen-2-yl)but-3-en-2-one (27)
To a solution of 14 (0.32 g, 1.18 mmol) in acetone (30 mL) and MeOH (30 mL) under N was
2
added KOH (15 %, 0.60 mL), and the mixture was stirred for 6 h under N at rt. Then it was poured
2

2
6
ACCEPTED MANUSCRIPT
onto ice-water (30 mL) and acidified with glacial acetic acid (pH 4). The precipitate was washed
with water, dried under vacuum and purified by chromatography (petroleum ether/EtOAc, 8/2) to
-1 1
afford bright orange crystals: 73 % yield; mp 214 °C; FTIR: 1667, 1566 cm ; H NMR (CDCl ) δ
3
8
2
1
1
3
.24 (dd, J = 1.4, 7.5 Hz, 1H), 8.18 (dd, J = 1.5, 7.6 Hz, 1H), 7.79–7.73 (m, 2H), 7.58–7.47 (m,
1
3
H), 6.60 (d, J = 15.8 Hz, 1H), 4.13 (s, 6H), 2.40 (s, 3H); C NMR (101 MHz, CDCl ) δ 197.46,
3
48.39, 146.07, 140.35, 135.98, 131.69, 129.13, 128.64, 126.95, 126.92, 126.43, 126.15, 125.48,
+
22.79, 121.56, 63.52, 60.92, 28.27; MS (APCI) m/z 313.0895, calcd for C H O S [M + H] ,
1
8
16
3
13.0893.
In a similar fashion, compound 28 was prepared from acetophenone. See Supplementary content
for details.
5.1.17. 2-((2-Methylhydrazono)methyl)naphtho[2,3-b]thiophene-4,9-dione (37)
To a solution of 32 (0.23 g, 0.95 mmol) in benzene (40 mL) was added methylhydrazine (44 mg, 50
µL, 0.95 mmol), and the mixture was stirred at rt for 12 h. The solvent was evaporated and the
residue was purified by chromatography to afford a dark red powder; 21% yield; mp 289 °C; FTIR
-1 1
3
313, 1666, 1643, 1577 cm ; H NMR (DMSO-d ) δ 8.39 (d, J = 4.4 Hz, 1H), 8.12–8.04 (m, 2H),
6
1
3
7
1
1
.88–7.83 (m, 2H), 7.62 (s, 1H), 7.51 (s, 1H), 2.86 (d, J = 3.6 Hz, 3H); C NMR (DMSO-d ) δ
6
79.49, 177.50, 153.64, 143.69, 140.51, 134.53, 134.37, 133.90, 133.23, 127.16, 126.61, 123.52,
+
21.46, 33.43; MS (APCI) m/z 271.0566, calcd for C H N O S [M + H] , 271.0536.
1
4
10
2
2
Analogously, compound 38 was prepared from phenylhydrazine. See Supplementary content for
details.

2
7
ACCEPTED MANUSCRIPT
5.1.18. 2-Nitronaphtho[2,3-b]thiophene-4,9-dione (39a)
This was obtained in a similar fashion as described [40], together with 39b. For details and
crystallographic data of 39a and 39b, see the Supplementary content.
5.1.19. 2-Aminonaphtho[2,3-b]thiophene-4,9-dione (40)
To a solution of 39a (1.5 g, 5.8 mmol) and tetrabutylammonium bromide (16 mg, 0.05 mmol) in
THF (100 mL) under N at 0 °C was added dropwise with vigorous stirring a solution of sodium
2
dithionite (3.33 g, 19.14 mmol) in water (50 mL). The solution was allowed to warm to rt and
stirred for an additional 30 min. Then the solution was extracted with CH Cl (3 × 100 mL), the
2
2
solvent was evaporated and the residue was purified by chromatography (CH Cl /EtOAc, 1/1) to
2
2
-1 1
afford dark violet crystals; 44% yield; mp 292 °C; FTIR 3326, 1656, 1610, 1590 cm ; H NMR
(
DMSO-d ) δ 8.07–8.03 (m, 2H), 7.86–7.78 (m, 2H), 7.67 (s, 2H), 6,34 (s, 1H); MS (APCI) m/z
6
+
2
30.0270, calcd for C H NO S [M + H] , 230.02703.
1
2
7
2
5.1.20. N-(4,9-Dioxo-4,9-dihydronaphtho[2,3-b]thiophen-2yl)propionamide (41a)
To a solution of 40 (58.1 mg, 0.26 mmol) in pyridine (5 mL) at 0 °C was added dropwise propionyl
chloride (70.5 mg, 0.76 mmol). The solution was allowed to warm to rt and stirred for an
additional 2 h. Then CH Cl (20 mL) was added, and the solution was extracted with HCl (2 mol/L,
2
2
3
× 20 mL). The combined aqueous phase was extracted with CH Cl (3 × 20 mL), and the
2 2
combined organic phase was dried over Na SO . Then it was concentrated and the residue was
2
4
purified by chromatography (CH Cl /EtOAc, 9/1) to afford orange-brown crystals; 56% yield; mp
2
2
-
1 1
2
7
2
94–296 °C; FTIR 3325, 1670, 1632, 1589 cm ; H NMR (DMSO-d ) δ 8.09–8.06 (m, 2H), 7.87–
6
.82 (m, 2H), 7.06 (s, 1H), 2.47 (q, J = 7.5 Hz, 2H), 1.11 (t, J = 7.5 Hz, 3H); MS (APCI) m/z
+
86.0532, calcd for C H NO S [M + H] , 286.05324.
1
5
11
3

2
8
ACCEPTED MANUSCRIPT
In a similar fashion, compounds 41b–41e were prepared from the appropriate acyl chlorides. See
Supplementary content for details.
Compounds 7 [30], 42 [64], and 43 [30] were prepared as described.
5.2. Biological assay methods
5.2.1. Keratinocyte culture and determination of cell hyperproliferation
HaCaT keratinocytes [42] were obtained from CLS cell lines service (Eppelheim, Germany). The
cell line was confirmed as mycoplasma negative using MycoAlert™ (Lonza, Basel) mycoplasma
detection kit, and then the assay was performed as described previously [65]. After 48 h of
incubation with the test compounds in a 96-well microtiter plate, MTT [3-(4,5-dimethylthiazol-2-
yl)-2,5-diphenyltetrazolium bromide] (20 µL of a 5 mg/mL solution in PBS) was added to each well
and the plate was incubated for 2 h at 37 °C. Then the plate was centrifuged (1500g for 10 min)
and unreduced MTT removed. DMSO (100 µL
) was added to dissolve the blue formazan. The
absorbance at 570 nm (with background substraction at 630 nm) was read using a Spectra Max 384
Plus microplate reader (Molecular Devices). Appropriate blanks were performed with MTT. Cell
numbers were calculated from blank-subtracted absorbance, and data were analyzed with the
software Softmax Pro. Inhibition was calculated by the comparison of the mean values of the test
compound (
N
= 3, SD < 10%) with the control (
00. Inhibition was statistically significant compared to that of the control (Student’s t-test;
.05). IC values were obtained by nonlinear regression (GraphPad Prism).
N
= 6–8) activity: (1 - test compound/control)
×
1
0
P
<
5
0