Development of novel 2,4-bispyridyl thiophene–based compounds as highly potent and selective Dyrk1A inhibitors. Part I: Benzamide and benzylamide derivatives

Article abstract of DOI:10.1016/j.ejmech.2018.07.050

The protein kinase Dyrk1A modulates several processes relevant to the development or progression of Alzheimer's disease (AD), e. g. through phosphorylation of tau protein, amyloid precursor protein (APP) as well as proteins involved in the regulation of alternative splicing of tau pre-mRNA. Therefore, Dyrk1A has been proposed as a potential target for the treatment of AD. However, the co-inhibition of other closely related kinases of the same family of protein kinases (e.g. Dyrk1B and Dyrk2) or kinases from other families such as Clk1 limits the use of Dyrk1A inhibitors, as this may cause unpredictable side effects especially over long treatment periods. Herein, we describe the design and synthesis of a series of amide functionalized 2,4-bispyridyl thiophene compounds, of which the 4-fluorobenzyl amide derivative (31b) displayed the highest potency against Dyrk1A and remarkable selectivity over closely related kinases (IC₅₀: Dyrk1A = 14.3 nM; Dyrk1B = 383 nM, Clk1 > 2 μM). This degree of selectivity over the frequently hit off-targets has rarely been achieved to date. Additionally, 31b inhibited Dyrk1A in intact cells with high efficacy (IC₅₀ = 79 nM). Furthermore, 31b displayed a high metabolic stability in vitro with a half-life of 2 h. Altogether, the benzamide and benzylamide extension at the 2,4-bispyridyl thiophene core improved several key properties, giving access to compound suitable for future in vivo studies.

Full text of DOI:10.1016/j.ejmech.2018.07.050

Accepted Manuscript
Development of novel 2,4-bispyridyl thiophene–based compounds as highly potent
and Selective Dyrk1A inhibitors. Part I: Benzamide and benzylamide derivatives
Sarah S. Darwish, Mohammad Abdel-Halim, Mohamed Salah, Ashraf H. Abadi,
Walter Becker, Matthias Engel
PII:
S0223-5234(18)30606-8
10.1016/j.ejmech.2018.07.050
EJMECH 10585
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 14 June 2018
Revised Date: 13 July 2018
Accepted Date: 18 July 2018
Please cite this article as: S.S. Darwish, M. Abdel-Halim, M. Salah, A.H. Abadi, W. Becker, M. Engel,
Development of novel 2,4-bispyridyl thiophene–based compounds as highly potent and Selective
Dyrk1A inhibitors. Part I: Benzamide and benzylamide derivatives, European Journal of Medicinal
Chemistry (2018), doi: 10.1016/j.ejmech.2018.07.050.
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
Graphical Abstract

ACCEPTED MANUSCRIPT
Development of Novel 2,4-Bispyridyl Thiophene–
based Compounds as Highly Potent and
Selective Dyrk1A Inhibitors. Part I: Benzamide
and Benzylamide Derivatives
AUTHOR NAMES
1
1
2
1
Sarah S. Darwish , Mohammad Abdel-Halim , Mohamed Salah , Ashraf H. Abadi , Walter
3
2
Becker and Matthias Engel
AUTHOR ADDRESS
1
Department of Pharmaceutical Chemistry, Faculty of Pharmacy and Biotechnology, German
University in Cairo, Cairo 11835, Egypt
2
Pharmaceutical and Medicinal Chemistry, Saarland University, Campus C2.3, D-66123
Saarbrücken, Germany
3
Institute of Pharmacology and Toxicology, Medical Faculty of the RWTH Aachen University,
Wendlingweg 2, 52074 Aachen, Germany
1

ACCEPTED MANUSCRIPT
Abstract.
The protein kinase Dyrk1A modulates several processes relevant to the development or
progression of Alzheimer’s disease (AD), e. g. through phosphorylation of tau protein, amyloid
precursor protein (APP) as well as proteins involved in the regulation of alternative splicing of
tau pre-mRNA. Therefore, Dyrk1A has been proposed as a potential target for the treatment of
AD. However, the co-inhibition of other closely related kinases of the same family of protein
kinases (e.g. Dyrk1B and Dyrk2) or kinases from other families such as Clk1 limits the use of
Dyrk1A inhibitors, as this may cause unpredictable side effects especially over long treatment
periods. Herein, we describe the design and synthesis of a series of amide functionalized 2,4-
bispyridyl thiophene compounds, of which the 4-fluorobenzyl amide derivative (31b) displayed
the highest potency against Dyrk1A and remarkable selectivity over closely related kinases (IC :
5
0
Dyrk1A = 14.3 nM; Dyrk1B = 383 nM, Clk1 > 2µM). This degree of selectivity over the
frequently hit off-targets has rarely been achieved to date. Additionally, 31b inhibited Dyrk1A in
intact cells with high efficacy (IC50 = 79 nM). Furthermore, 31b displayed a high metabolic
stability in vitro with a half-life of 2h. Altogether, the benzamide and benzylamide extension at
the 2,4-bispyridyl thiophene core improved several key properties, giving access to compound
suitable for future in vivo studies.
Keywords: Dyrk1A, Alzheimer’s disease, bispyridyl thiophenes amides, selectivity, metabolic
stability
2

ACCEPTED MANUSCRIPT
1
. Introduction
Dual-specificity tyrosine-regulated kinases (Dyrks) are a family of eukaryotic kinases that
belong to a larger super family known as the CMGC group. [1] The Dyrk family consists of five
mammalian subtypes including Dyrk1A, 1B, 2, 3, and 4. [2] Dyrk1B is highly expressed in some
types of cancers, where it is claimed to exert an anti-apoptotic role in tumour cells by mediating
some survival signals. [3] For Dyrk2, several controversial roles were reported in cancer biology:
this isoenzyme was found to be amplified and overexpressed in lung and esophageal
adenocarcinoma as well as gastrointestinal stromal tumours. [4, 5] However, a previous study
proposed that a Dyrk2 inactivation contributes to the invasiveness of human tumours due to the
decreased levels of phosphorylated c-Jun and c-Myc transcription factors as well as the lowered
activation of the p53 tumour suppressor protein. [6, 7] Comparably little information exists on
the remaining Dyrk isoforms. Dyrk3, which is most closely related to Dyrk2 among the Dyrk
family, was shown to attenuate erythropoiesis by efficiently inhibiting NFAT transcriptional
activation. [8] Dyrk4, for which few reports exist, was found expressed in human brain during
neuronal differentiation. [9] Dyrk1A, the most studied isoform of this family, also has a well-
established neurodevelopmental role, as confirmed by the examination of Dyrk1A expression in
the developing mouse brain. [10] In humans, Dyrk1A was identified in the Down syndrome (DS)
critical region (DSCR) of chromosome 21 [11, 12] as a potential candidate gene causing the DS
phenotype. DS individuals are characterized by mental retardation, cognitive as well as cranio-
facial characteristics, congenital heart disease, an increased risk of childhood leukemia, and an
early onset of Alzheimer´s Disease (AD). [13] In DS individuals, the 1.5-fold overexpression of
Dyrk1A is believed to cause an imbalance in the neurodevelopmental processes such as
neurogenesis and neuronal differentiation, [14-18] eventually resulting in reduced brain size,
3

ACCEPTED MANUSCRIPT
neuronal deficits and altered neuronal morphology. Especially the early onset of AD was
attributed to an increased phosphorylation of the tau protein and presenilin-1 by Dyrk1A. [19]
Importantly, Dyrk1A was found to be involved in the regulation of alternative splicing of pre-
mRNA, which is crucial for controlling embryonic development, cell growth and apoptosis. [20]
The splicing process is catalyzed by a macromolecular machine called spliceosome which is
controlled by protein phosphorylation. [21] Splice site recognition is determined mainly through
the interaction of serine-arginine-rich (SR) proteins and heterogeneous nuclear
ribonucleoproteins with specific exonic and intronic pre-mRNAsequences. [22] Dyrk1A was
reported to phosphorylate several SR proteins, thus modulating their activity. [23, 24] In
addition, Dyrk1A is known to phosphorylate the non-SR protein SF3b1 which plays an important
role in the basic splicing reaction since it is detected only in functional spliceosomes. [25]
Among the important SR proteins phosphorylated by Dyrk1A are SRp55, SC35 and ASF, which
regulate the inclusion of exon 10 in the tau mRNA in the brain. [23, 24, 26] Indeed,
overexpression of Dyrk1A led to the increase of 3R-tau with incorporated exon 10 in the
neonatal brains of Ts65Dn mice, a model of Down syndrome. [27] In the same study, inhibition
of Dyrk1A restored the balance of 3R/4R tau protein by preventing the inclusion of exon 10 in
the tau mRNA. The relatively increased amount of 3R-tau relative to 4R-tau facilitates tau
aggregation, one of the main hallmarks of AD. [28] Not surprisingly, increased Dyrk1A activity
in DS brains and Dyrk1A overdosage models was found to change further transcript
compositions, e. g. in the mRNAs of neuroligin and of acetylcholinesterase variants, thus
producing key synaptic proteins with unusual features. [29]
In addition, Dyrk1A was reported to phosphorylate tau protein and APP directly [1, 13, 30], thus
promoting tau aggregation and formation of amyloid plaques, the main molecular hallmarks of
4

ACCEPTED MANUSCRIPT
sporadic as well as DS-linked AD neurodegeneration. [31, 32] A recent study proved that
normalization of the gene dosage of Dyrk1A could reduce several phenotypes of AD in a mouse
model of DS. [33] Another group showed that Dyrk1A inhibition could improve AD pathologies
in mouse model. [34] Altogether, there is ample evidence that Dyrk1A is a promising target in
AD [27, 33, 34] and possibly in splicing–related disorders, [35] although the latter indication has
scarcely been explored up to now.
Any therapeutic use of Dyrk1A inhibitors requires high selectivity over closely related kinases in
the Dyrk family, such as Dyrk1B and Dyrk2, as well as frequently co-inhibited kinases from
other families, including Clks, to prevent unwanted adverse effects. However, most of the
previously published Dyrk1A inhibitors affected other kinases known to maintain or promote
cell proliferation, mainly Clk1/4, haspin, and CK2. [31, 36-39] Indeed, several Dyrk1A
inhibitors, including harmine, [40] meridianin derivatives [41] and pyrido[2,3-d]pyrimidines,
[42] were reported to affect cell growth at low µM concentrations, which could well be
explained by off-target activities. With respect to cancer, multi-targeted inhibitors might even
have advantages over specific ones; for instance, in a study by Zhou et al., 7-azaindole–based
compounds that simultaneously targeted Dyrk1A, Dyrk1B and Clk1, efficiently blocked cell
proliferation and migration of glioblastoma cells. [43] In contrast, potentially cytotoxic effects of
Dyrk inhibitors are expected to limit the therapeutic applicability in neurodegenerative diseases,
where the cell viability is already affected by harmful peptide and protein aggregates.
In a previous report, we introduced bispyridyl thiophenes as a novel scaffold for the inhibition of
Dyrk and Clk kinases. [38] While the overall selectivity of these inhibitors was promising, they
did not show a pronounced preference among the close homologues Dyrk1A/B, Dyrk2 or
5

ACCEPTED MANUSCRIPT
Clk1/4; only slight selectivity shifts could be achieved by the introduction of different small
substituents; the most potent compound C29 (compound 29 in ref. [38]) is depicted in Fig. 1.
In the current study, we present new derivatives of the bispyridyl thiophene scaffold with a
crucial amide extension that greatly enhanced the potency and selectivity of the compounds for
Dyrk1A as well as the metabolic stability in vitro.
2
. Results and Discussion
Compound design. We previously described the development of a series of ATP-competitive,
bisheterocyclic substituted thiophenes based on the original hit compound C4 (compound 4 in
ref. [38]), depicted in Fig. 1 (IC s: Dyrk1A = 300 nM; Dyrk1B = 300 nM; Dyrk2 = 300 nM).
5
0
The optimized compound C29 displayed higher potency against Dyrk1A, [38] however, still
lacked selectivity for Dyrk1A over the closely related kinases Dyrk1B and Dyrk2 as well as
Clk1.
Small substituents at one of the pyridine rings of C4, such as methyl, methoxy, cyano or
halogens effected only a slight modulation of the Dyrk subtype selectivity; however, it was
found that particularly at the meta position, diverse substituents were tolerated without affecting
the potency. Hence we surmised that an amino function, which would give access to a focused
library of amides, could also be installed on the pyridine ring. In light of the good
physicochemical properties and the low molecular weight (238.31 g/mol) of the core structure
C4, [38] the attachment of benzamide, sulfonamide and benzylamide moieties was deemed
acceptable to enhance binding affinity and selectivity whilst keeping molecular weight and
lipophilicity in a drug-like range.
6

ACCEPTED MANUSCRIPT
Based on our docking simulation (Fig. 2), the two important hydrogen bonds with the conserved
Lys188 and the hinge region residue Leu241 as well as the CH–π interactions involving Lys188,
Leu241 and the Phe238 benzene ring were not expected to be compromised by the probe amino
function that was added to ring A of C4. In accordance, the potency of the resulting compound 2
was even somewhat higher than that of C4 (Table 1).
Firstly, further aminopyridine derivatives of C4 were synthesized as probe compounds to
identify the optimum position with respect to potency and selectivity. Similar to what was
observed with other substituents, the amino function was best tolerated in the 3-position of ring
A (Compound 2, Table 1), however, it did not improve the selectivity over Dyrk1B.
Interestingly, the amino function could be introduced at both pyridine rings (ring A and B) with
basically the same outcome (cf. compound 35, Table 1), indicating that it was equally tolerated at
each end of the ATP binding pocket or, as also suggested by our docking simulation (Fig. 2), that
the inhibitor could flip in the binding site by 180 degrees without loss of affinity due to the rather
symmetric shape of the 2,4-bispyridyl scaffold Shifting the pyridine nitrogen of ring B by
.
incorporating 4-pyridyl (compound 3) almost abolished the potency, indicating that the added 3-
amino did not induce a new binding mode that would favour or tolerate another position of the
acceptor nitrogen in pyridine ring B (cf. Fig. 1). Eventually, we selected probe compound 2 as
basis for further modifications.
Since in both possible binding orientations, the amino function pointed to the borders of the ATP
binding pocket, molecule extensions utilizing an amide linkage could potentially address several
side chains and/or backbone peptide functions of unexploited residues located at the borders of
the ATP binding pocket (encircled in yellow, Fig. 2). These included the Ile165 and Tyr243 side
chains at the hinge region side (Fig. 2A) as well as the Phe170 and Asp307 residues located at
7

ACCEPTED MANUSCRIPT
the opposite edge of the pocket (Fig. 2B). Using a diversification strategy involving an extension
of probe compound 2 through amide and sulfonamide couplings, we aimed at enhancing both the
potency towards Dyrk1A and the selectivity over other kinases, in particular Dyrk1B, Dyrk2,
Clk1 and haspin.
Chemistry. The synthesis of the planned amide functionalized bispyridyl thiophene compounds
was accomplished through a four-step synthesis as summarized in Scheme 1. Firstly, a Miyaura
reaction was carried out on 2/3-amino-5-bromopyridine and bis(pinacolato)diboron in presence
of potassium acetate and Pd(dppf)Cl as a catalyst to afford the corresponding boronic acid
2
pinacol ester. Next, the produced boronic acid derivative became accessible by a consecutive
Suzuki cross coupling reaction with 2,4-dibromothiophene in the presence of Cs CO and
2
3
palladium-tetrakis(triphenylphosphine) to give 4-bromothiophen-2-yl pyridine amines
compounds C-D). In the third step and starting from compound D, a coupling reaction took
(
place with diverse acid chlorides or benzenesulfonyl chlorides to yield a series of amides in
addition to one sulfonamide derivative, respectively. The final reaction was a second Suzuki
reaction to synthesize the bispyridyl thiophene functionalized amides (4b-34b). This was done
using 3-pyridine boronic acid and Pd(dppf)Cl in presence of Na CO as a base. It is worth
2
2
3
mentioning that the poor yield of the later Suzuki reaction hindered diversification via amide
coupling as a last step. Few compounds (1-3) were synthesized by direct coupling of the 4-
bromothiophen-2-yl pyridine amine intermediates (C-D) with 3/4-pyridine boronic acids
(Scheme 1). To synthesize the inverted positional isomer of compound 2, compound E was
synthesized by coupling 2,4-dibromothiophene with 3-pyridine boronic acid, followed by
coupling with compound B to yield compound 35 (Scheme 2).
8

ACCEPTED MANUSCRIPT
Biological evaluation.
The inhibitory activity of the synthesized compounds was evaluated against recombinant
Dyrk1A by screening at 250 nM. For compounds showing more than 60% inhibition, the IC50
values were determined (Tables 1 and 2). In order to obtain an early indication of the selectivity,
all new compounds were tested in parallel against Dyrk1B, the most closely related isoenzyme.
In the present study, we generally focused on the installation of an aromatic system, that could
engage in several of the potential interactions with the residues indicated in Fig. 2.
Structure-Activity-Relationships (SAR). Evaluation of the carboxamide vs. sulfonamide
extension. We started with the attachment of a plain phenyl ring through an amide linkage to
obtain the benzamide derivative 4b. When compared to the probe compound 2, 4b gave a
slightly reduced inhibition without enhancement of selectivity over Dyrk1B (Tables 1 and 2).
However, this did not argue for a poor steric fit of the benzoyl moiety, because the amide
coupling strongly decreased the mesomeric effect of the amino function on the pyridyl, thus
weakening the acceptor strength of the ring nitrogen (cf. Fig. 3A). Considering that this caused a
partial loss of affinity, we concluded that the benzoyl moiety was overall tolerated.
In contrast, the substitution of the amide function in 4b with a sulfonamide moiety yielded a
totally inactive analogue (5b), clearly indicating that the carboxamide group was much more
preferred by the binding site (Table 2).
Evaluation of electron–donating substituents. In an attempt to improve the potency of
compound 4b, different electron–donating as well as electron–withdrawing groups were tested as
substituents on the phenyl ring. Firstly, mono-substitution by electron–donating groups was
investigated, starting with the addition of the lipophilic methyl group in the ortho, meta and para
9

ACCEPTED MANUSCRIPT
positions. The o- and m-toluoyl derivatives (compounds 6b and 7b) showed slightly lower
inhibitory activity than compound 4b while the p-methyl substitution in 8b proved to be more
favourable (IC = 51.4 nM), raising the potency over that of the probe compound 2 (IC = 101
5
0
50
nM, Table 2). In addition, 8b retained the selectivity over Dyrk1B. To confirm the optimum
substitution position, the more polar and stronger electron–donating methoxy group was
explored in the same manner. However, the methoxy group effected leveling of potency between
the ortho-, meta- and para-regioisomers, as all of the compounds showed similar degrees of
inhibition (compounds 9b-11b, inhibition between 53.5 and 59.5%, Table 2). There was only a
marginal preference for the para-position. Altogether, it could be concluded that the para-
position is superior in case of electron–donating groups, but that the potential activity–enhancing
effect of the methoxy substituent might be counteracted by steric clashes. To probe the available
space surrounding the para-position, the bulky, lipophilic and electron donating tert-butyl group
was introduced (12b); this modification abolished the inhibitory activity against Dyrk1A,
suggesting that substituents which are essentially larger than methyl – such as methoxy and
particularly t-butyl – are likely to produce a steric clash with the ATP binding pocket. Of note,
the methoxy–substituted congeners 9b and 10b exhibited diminished activity toward Dyrk1B
(Table 2), indicating that the variation of substituents at the benzoyl moiety permits to modulate
the selectivity.
Evaluation of electron–withdrawing substituents. The preference of the benzoyl para
position, as found with the electron–donating substituents, was not observed when electron–
withdrawing groups were explored. In light of the finding described above, we restricted to
substituents that did not largely exceed the size of a methyl group. First, we analyzed the effect
of a chloro substituent; among the three positional isomers, the o-chlorobenzamide analogue 13b
10

ACCEPTED MANUSCRIPT
(
IC50= 90.5 nM, Table 2) was clearly more active than the m- or p-chloro isomers (14b and 15b,
respectively) and represented the most potent mono-substituted benzamide analogue. By
examining Dyrk1B inhibition, 13b showed considerable preference for Dyrk1A over Dyrk1B.
Similarly, the smaller fluoro substituent was also tested; on average, the three positional isomers
showed a higher activity than the chlorinated analogues (cf. 16b–18b, Table 2), which might be
attributable to the lower steric demand of the fluorine, in particular at the meta- and para-
positions. Additionally, all fluoro derivatives possessed superior selectivity for Dyrk1A over
Dyrk1B. However, the most active o–fluorobenzamide derivative 16b (IC50= 106.7 nM) did not
fully reach the potency level of the chlorinated congener 13b. Finally, we investigated whether
the trifluoromethyl group could boost the activity stronger than fluorine, which was, however,
not the case (compounds 19b-21b, Table 2); all three regioisomers displayed lower inhibitory
activity than compound 4b. Altogether, there was no clear correlation between the inhibitory
activity and the modulation of the electron density by the substituents in the benzamide class.
Most probably, the influence of the substituents on potency is governed by more than one
parameter, e. g. by direct interaction with the binding site, modulation of the electron density,
conformational stabilizations, and potential steric interference. Various substituents, in particular
methyl, chlorine and fluorine, favourably enhanced the selectivity for Dyrk1A over Dyrk1B,
which had not been observed with the unsubstituted benzamide 4b.
Effect of multiple substitutions. Assuming that some of the favourable effects observed with
different substituents at several ring positions could be additive, disubstitution patterns were
investigated using combinations of the substituents described above. Initially, disubstitution
using identical groups was investigated, starting with the 2,3-dichlorobenzamide analogue
(compound 22b) which led to a decrease in the inhibitory activity compared with the benzamide
11

ACCEPTED MANUSCRIPT
derivative 4b and with the o-chloro derivative 13b, confirming the deleterious effect of the m-
chloro on potency. On the other hand, the 2,4-difluoro congener 23b showed a great
improvement in the inhibitory activity with an IC50 of 49 nM, indicating that the effect of
combining two fluorine atoms is indeed additive. However, changing the nature of substituents
to 2-chloro-4-fluoro (compound 24b) decreased the inhibitory activity below that of the
unsubstituted analogue 4b, again confirming that the electron withdrawing properties of the
halogens do not play a major role, as they would have been additive. Similarly, the 3-chloro-4-
fluoro substitution was detrimental to the activity (compound 25b). Consequently, the compound
design was extended to include a combination of the favourable 2-fluorine with an electron–
donating 5-methoxy group (26b). 26b showed a further improved activity with an IC of 37.3
5
0
nM, indicating that the 2-fluoro substitution can favourably be combined with other substituents
to enhance the potency. Fortunately, the resulting disubstituted compounds 23b and 26b also
exhibited a good selectivity towards Dyrk1A over its closely related isoform Dyrk1B. Prompted
by the favourable effects especially by the difluorination, we tested whether a trifluorinated
analogue, bearing another fluorine in ortho, would even show higher activity. Indeed, compound
2
7b with the 2,4,6-trifluoro substitution proved to be the most potent inhibitor among the
benzamide series (IC50 = 23.5 nM). Moreover, this improvement of potency of 27b was
accompanied by a remarkable 13-fold increase in selectivity over Dyrk1B. Somewhat
unexpectedly, the 3-fluoro in the 2,3,4-trifluoro–substituted isomer (28b), completely neutralized
the activity–enhancing effects of the other fluorine substituents.
Changing the nature of the benzamide ring. To decouple effects of electron withdrawing
effects of substituents from potential secondary effects such as steric clashes, we replaced the
benzamide phenyl by the electron deficient 4-pyridyl (29b), which was supposed to mimic the 4-
12

ACCEPTED MANUSCRIPT
fluoro substitution in 18b. Indeed, the inhibitory activity of 29b followed the same trend as 18b:
Reaching an IC of 107 nM, 29b was more active than the unsubstituted benzamide 4b, but less
5
0
active than the 4-methyl–substituted analogue 8b (IC = 51.4 nM). However, since the absence
5
0
of a p-substituent in 29b diminished the selectivity toward Dyrk1B (compare with the 4-
fluorobenzamide derivative 18b) no further 4-pyridyl derivatives were synthesized. In many
cases, placing substituents at the benzamide favourably reduced the inhibitory activity against
Dyrk1B, in particular methoxy (9b-11b), 2-chloro (13b) and fluoro (16b-19b).
Insertion of a spacer in some benzamide derivatives (homologation). In our last set of
modifications, alkyl spacers were introduced between the amide group and the phenyl ring.
Compared with its direct benzamide homologue 4b, the benzyl amide derivative 30b displayed a
marked improvement in the inhibitory activity to reach an IC of 14.9 nM, which was even
5
0
slightly superior to the most potent benzamide derivative 27b. However, the introduction of a 4-
fluoro substituent (analogous to 18b) did not further increase the potency toward Dyrk1A (31b,
IC50 = 14.3 nM), and substitution of the benzyl by the electron–donating 3-methoxy strongly
diminished the activity.
The main advantage of the benzylamide vs. the benzamide group was that it did not
concomitantly enhance the activity towards Dyrk1B, thus increasing the selectivity factor to
almost 27 for 31b (IC50 ratio Dyrk1B/Dyrk1A). Further elongation of the spacer by one
methylene unit to afford compounds 33b and 34b did not give any indication that further
homologations could be beneficial for the inhibitory activity.
Prediction of the binding mode of 31b to Dyrk1A. As described above, the addition of a
benzamide and especially of a benzylamide moiety to the basic bispyridyl thiophene scaffold led
to a boost of potency against Dyrk1A. In order to predict a potential binding mode to the ATP
13

ACCEPTED MANUSCRIPT
pocket of Dyrk1A, the most potent compound, 31b, was docked in the coordinates from PDB
entry 5A3X. [44] As depicted in Fig. 3A, our docking study predicted two new interactions that
are brought about by the benzylamide group: an H-bond between the amide NH and the Asp307
carboxylate, and a CH–π interaction (edge-to-face) between the phenyl ring and Phe170 at the
Dyrk1A ATP binding pocket. Moreover, the latter interaction forces the benzyl group to
sterically occlude this part of the pocket, thus shielding the H-bond between Asp307 and the
amide NH from competing water molecules (Fig. 3B). Taken together, this ensemble of effects
might explain the 21-fold increase in potency resulting from the addition of the benzylamide
group to the basic scaffold (cf. Table 1).
14

ACCEPTED MANUSCRIPT
Extended selectivity profiling of 31b.
As mentioned earlier, one of our major objectives was to improve the selectivity of the previous
bispyridyl thiophene compound class for Dyrk1A. After confirming its superior selectivity over
Dyrk1B, our most potent inhibitor 31b was further evaluated for its selectivity against an
extended panel of kinases that were frequently reported as being co-inhibited by chemically
diverse classes of Dyrk1A inhibitors (Table 3). Of note, 31b showed only negligible inhibition of
Clk1, which was frequently hit by previous Dyrk1A inhibitors published by us and others. [37-
3
9, 45-47] In addition, compound 31b showed no or only moderate inhibitory activity towards
the other frequently reported off-target kinases such as CDK5/p25, CK1 delta, Dyrk2, HIPK1,
TRKB, PIM1, MLCK2, SRPK1 and STK17A (Table 3). Also considering the remarkable
selectivity towards the closest homologue, Dyrk1B, 31b was found to be one the most selective
Dyrk1A inhibitors described to date. However, it still inhibited the known off-target haspin, with
about 2.5-fold lower potency than Dyrk1A. Haspin is a histone H3 kinase and required for cell
proliferation; hence chemical inhibition of intracellular haspin leads to cell growth arrest. [48]
However, we did not observe strong effects on HeLa cell proliferation with 31b (see below for
details).
How was selectivity over Clk1 achieved? Of note, while the core compound C4 had shown the
same affinity to both Dyrk1A and Clk1, [38] the benzylamide extension decreased the potency of
3
1b against Clk1 below that of the parent compound C4. Although all residues except Leu167
(corresponds to Ile165 in Dyrk1A) are conserved in Clk1, small spatial shifts of the backbone
and the side chains are detectable in the published X-ray structures of the two kinases and may
account for the affinity difference. A docking simulation revealed that the small topology
differences translated into a less efficient binding of 31b to Clk1 than to Dyrk1A, as reflected by
15

ACCEPTED MANUSCRIPT
the reduction in number of CH–π bonds (6 vs. 8) and H-bonds (2 vs. 3) (cf. Fig. S1,
Supplementary Information). In accordance with the observed structure–selectivity relationships,
the benzyl extension was predicted to have a key role in selectivity tuning towards Dyrk1A: in
contrast to that of Dyrk1A, the ATP pocket of Clk1 did not allow favourable interaction of the 4-
fluorobenzyl ring with Phe172 (corresponding to Phe170 in Dyrk1A, cf. Fig. 3A) inside the
pocket; instead, the CH–π interaction is only possible with Gly170 from the glycine–rich loop,
which can only occur at the expense of some interactions inside the pocket (Fig. S1,
Supplementary Information).
Inhibition of Dyrk1A by 31b in HeLa cells.
To assess the ability of our most potent analogue 31b to inhibit Dyrk1A in intact cells, we tested
its effect on the Dyrk1A-catalysed phosphorylation of splicing factor 3b1 (SF3b1).
Phosphorylation of SF3b1 on Thr434 correlated with the cellular activity of endogenous
Dyrk1A. [25] HeLa cells were transiently transfected with a GFP-SF3b1 expression vector and
treated with 31b. As can be seen in Fig. 4, 31b inhibited the Dyrk1A–catalyzed phosphorylation
of SF3b1 at Thr434 with high efficacy; a significant reduction of phosphorylated SF3b1 was
already noted at 30 nM (IC50 = 79 nM, Fig. 4).
Evaluation of cytotoxicity in HeLa cells.
Selective Dyrk1A inhibitors are not expected to induce cell growth arrest, since Dyrk1A activity
was even reported to diminish cell proliferation in several cell types. [49-51] Therefore, we
evaluated our most potent compound, 31b, for cytotoxicity, which would indicate potential off-
target effects or non-specific toxicity. The cell viability assay was performed using HeLa cells, in
which the inhibition of intracellular Dyrk1A had been quantified for the same compound (see
16

ACCEPTED MANUSCRIPT
above). As shown in Table 4, 31b exerted minimal cytotoxicity up to a concentration of 3 µM, at
which almost full inhibition of cellular Dyrk1A was observed before (cf. Fig. 4, IC = 79 nM).
5
0
A weak inhibition of cell proliferation was noted at 10 µM; this could be due to the inhibition of
haspin, which might become effective in the cells at this higher concentration. In accordance,
chemical inhibition of haspin in HeLa cells was previously shown to induce cell cycle arrest.
[48] In the former study, compound HSD972, possessing an IC50 of 12 nM against purified
haspin, inhibited HeLa cell proliferation with an IC50 of 4.8 µM.
Altogether, our results corroborated that – at least in HeLa cells – inhibition of Dyrk1A does not
affect cell proliferation. It is therefore likely that the effects of previously published Dyrk1A
inhibitors on cell growth of HeLa or other cell lines (e. g. harmine, IC50 with HeLa cells: 8 µM
[40]) were rather caused by co-inhibition of Dyrk1B, haspin or other kinases.
Evaluation of the metabolic stability.
We further assessed the suitability of two of the most potent analogues (27b and 31b) for a
potential application in vivo. To this end, we tested their phase I and phase II metabolic stability
against S9 fraction from human liver homogenate. Different samples were taken at various time
points, and the remaining fraction of parent compound was determined by LC-MS/MS. As
shown in Table 5, both 27b and 31b exhibited a metabolic half-life of about 2h. Even
considering that the metabolic stability assays with our previous series were performed using rat
liver microsomes, this was a substantial improvement to the non-amide extended precursor
series; for instance, C29 (Fig. 1) had shown a half-life of only 27 min. [38] Thus, both the
benzamide and the benzylamide modification yielded compounds, 27b and 31b, respectively
with promising metabolic stability, permitting to test them in future in vivo studies.
17

ACCEPTED MANUSCRIPT
Calculation of key physicochemical properties. In order to assess whether the most potent
compound 31b might be able to enter the CNS, we calculated some physicochemical parameters
that were found to have a predictive value in retrospective studies. [52, 53] In our previous study,
the calculated values were found in good agreement with the corresponding experimental values
for this compound class. [38] It can be concluded that the values for 31b (Table 6) are still in a
good range with respect to a potential application in the CNS, although the clogP is above the
average value found for CNS drugs. However, in general, the lipophilicity reported for CNS
drugs tends to be higher than that of non-CNS drugs; for instance, in the survey from Mahar
Doan et al. the respective clogP median values were 3.43 for CNS drugs (range 0.16–6.59), and
2
.78 for non-CNS drugs (range: -2.81–6.09).
3
. Conclusions
The current work presented a series of novel Dyrk1A inhibitors bearing a crucial amide
extension. Various benzamide derivatives were synthesized by probing mono-, di- and
trisubstitutions with distinct electron–modulating properties. Finally, the insertion of a spacer
was tested, which resulted in a dramatic improvement in potency as observed with compounds
3
0b and 31b (IC50 = 14.9 nM and 14.3 nM, respectively). Importantly, 31b also showed a
remarkable selectivity over the closely related kinases Dyrk1B and Clk1, which has rarely been
reported for previous Dyrk1A inhibitors (IC ; Dyrk1B = 383 nM, Clk1 > 2µM). Since Dyrk1A
5
0
and Clk1 share several splicing factors as substrates, including AF2/ASF, SC35, and SRp55, [23,
4, 26] our selective inhibitor 31b will allow to dissect Dyrk1A– from Clk1–mediated splicing
2
regulations; moreover, potential side effects due to a synergistic modulation of SR protein
activities through Dyrk1A/Clk1 co-inhibition will be avoided in potential therapeutic trials.
18

ACCEPTED MANUSCRIPT
Furthermore, 31b successfully inhibited Dyrk1A–mediated phosphorylation of SF3b1 in HeLa
cells in the two-digit nanomolar range (IC = 79 nM) while no cytotoxic effects were noted at 3
5
0
µM, and only marginal effects at 10 µM, thus providing an appropriate toxicity window for in
vivo applications. The absence of cytotoxic side effects is especially important when a prolonged
treatment of neurodegenerative diseases is envisaged, where the cell viability is already
compromised by harmful peptide and protein aggregates.
The metabolic stability was also found to be significantly improved by the new benzamide and
benzylamide modification of the previous scaffold. Altogether, the new modification led to an
optimization of the key parameters potency, selectivity and metabolic stability, thus allowing to
envisage in vivo studies in the next step.
4
. Experimental Section
Chemistry. Solvents and reagents were obtained from commercial suppliers and used as
received. Melting points were determined on a Stuart SMP3 melting point apparatus. All final
compounds had a percentage purity of at least 95%, and this could be verified by means of
HPLC coupled with mass spectrometry. Mass spectra (HPLC−ESIMS) were obtained using a
TSQ quantum (Thermo Electron Corp.) instrument prepared with a triple quadrupole mass
detector (Thermo Finnigan) and an ESI source. All samples were injected using an autosampler
(Surveyor, Thermo Finnigan) by an injection volume of 10 µL. The MS detection was
determined using a source CID of 10 V and carried out at a spray voltage of 4.2 kV, a nitrogen
5
sheath gas pressure of 4.0 × 10 Pa, a capillary temperature of 400 °C, a capillary voltage of 35
5
V, and an auxiliary gas pressure of 1.0 × 10 Pa. The stationary phase used was an RP C18
NUCLEODUR 100-3 (125 mm × 3 mm) column (Macherey & Nagel). The solvent system
19

ACCEPTED MANUSCRIPT
consisted of water containing 0.1% TFA (A) and 0.1% TFA in acetonitrile (B). The HPLC
method used a flow rate of 400 µL/min. The percentage of B started at 5%, was increased up to
1
00% during 7 min, was kept at 100% for 2 min, and was flushed back to 5% in 2 min and was
1
13
kept at 5% for 2 min. A Bruker DRX 500 spectrometer was used to obtain the H NMR and C
NMR spectra. The chemical shifts are referenced to the residual protonated solvent signals.
General synthetic procedures and experimental details for some key compounds.
Procedure A, Procedure for synthesis of Compounds A – D. A mixture of 5 mmol of the
appropriate amino-5-bromo pyridine and 1.96 gm (20 mmol) of potassium acetate and 0.18 gm
(
0.25 mmol) of Pd(dppf)Cl and 5.08 gm (20 mmol) of bis(pinacolato)diboron in dioxane was
2
heated to reflux under argon for 2 hours to yield compounds A and B. The mixture was left to
attain room temperature and then filtered under vacuum. Without further purification, a reaction
2 3
flask containing the filtrate was charged with 6.5 gm (20 mmol) of Cs CO , 0.29 gm (0.25
mmol) of palladium-tetrakis(triphenylphosphine) and 6 mmol of 2,4-dibromothiophene together
with 30% water in a Suzuki coupling reaction. The reaction was left to reflux under argon for 3.5
hours. The mixture was concentrated in vacuo. The residue was partitioned between 150 mL
ethyl acetate and 50 mL brine solution and then the aqueous layer was re-extracted using 3
portions of 100 mL ethyl acetate. The organic layers were collected and the volume was reduced
under reduced pressure. Afterwards the product was purified by CC to yield compounds C-D.
Procedure B, General procedure for amide synthesis (4a, 6a-34a). A solution of 0.18 gm (0.7
mmole) of compound D dissolved in acetone was treated with 0.84 mmol of the appropriate acid
chloride followed by the addition of 0.12 gm (1.05 mmol) of TEA. The reaction mixture was left
20

ACCEPTED MANUSCRIPT
to stir at room temperature for 2 hours which was followed by evaporation of solvent in vacuo
and the product was purified by CC.
Procedure C, General procedure for synthesis of compounds 1-3, 4b-34b, 35. The bromo
derivative was added to a suspension of 4 equiv of Na CO and 5 mmol% of Pd(dppf)Cl in
2
3
2
dioxane/water mixture. This was followed by the addition of the pyridine boronic acid
derivative. The reaction was heated to reflux for 2 hours under argon atmosphere. The solvent
was removed in vacuo. Small amount of brine solution was added and extraction was done using
ethyl acetate (3 x 50 mL). The ethyl acetate portions were collected and the volume was reduced
in vacuo. Afterwards the product was purified by CC.
5
-(4-Bromothiophen-2-yl)pyridin-2-amine (C). The compound was synthesized according to
procedure A using 2-amino-5-bromopyridine: yield: 42%. The product was purified by CC (ethyl
1
acetate/petroleum ether 7:3); H NMR (500 MHz, DMSO) δ 8.23 (d, J = 2.5 Hz, 1H), 7.64 (dd, J
=
8.6, 2.6 Hz, 1H), 7.51 (d, J = 1.4 Hz, 1H), 7.32 (d, J = 1.4 Hz, 1H), 6.47 (d, J = 8.7 Hz, 1H),
1
3
6
1
.27 (s, 2H).; C NMR (126 MHz, DMSO) δ 159.71, 144.87, 143.27, 134.41, 123.36, 120.71,
+
17.24, 109.61, 107.94; MS (ESI) m/z = 254.79 (M + H)
5
-(4-Bromothiophen-2-yl)pyridin-3-amine (D). The compound was synthesized according to
Procedure A using 3-amino-5-bromopyridine: yield 80%. The product was purified by CC (ethyl
1
acetate); H NMR (500 MHz, DMSO) δ 8.08 (d, J = 2.0 Hz, 1H), 7.91 (d, J = 2.5 Hz, 1H), 7.70
1
3
(
d, J = 1.4 Hz, 1H), 7.51 (d, J = 1.5 Hz, 1H), 7.16 – 7.05 (m, 1H), 5.50 (s, 2H); C NMR (126
MHz, DMSO) δ 144.98, 142.31, 136.28, 133.61, 128.34, 126.16, 123.47, 115.68, 109.89; MS
+
(ESI) m/z = 254.82 (M+H)
21

ACCEPTED MANUSCRIPT
3
-(4-Bromothiophen-2-yl)pyridine (E). The title compound was synthesized by reacting 0.5 gm
(
4 mmol) 3-pyridine boronic acid with 1.2 gm (4.8 mmol) 2,4-dibromothiophene in presence of
.2 gm (16 mmol) Cs CO and 0.14 gm (0.2 mmol) palladium-tetrakis(triphenylphosphine) in
5
2
3
dioxane/water. The mixture was left to stir under reflux in inert conditions for 3 hours. The
solvent was concentrated in vacuo, a small amount of brine solution was added to the residue and
extracted with ethyl acetate (3 X 50 mL). The organic layers were collected and the solvent was
evaporated in vacuo: yield 77%. The product was purified by CC (ethyl acetate/petroleum ether
1
6
1
:4); H NMR (500 MHz, DMSO) δ 8.92 (dd, J = 2.4, 0.8 Hz, 1H), 8.54 (dd, J = 4.8, 1.5 Hz,
H), 8.06 (ddd, J = 8.0, 2.4, 1.6 Hz, 1H), 7.78 (d, J = 1.4 Hz, 1H), 7.72 (d, J = 1.5 Hz, 1H), 7.46
1
3
(
ddd, J = 8.0, 4.8, 0.8 Hz, 1H); C NMR (126 MHz, DMSO) δ 149.17, 146.14, 141.14, 132.72,
+
1
28.55, 127.15, 124.28, 124.08, 110.15; MS (ESI) = 239.80 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)benzamide (4a). The compound was synthesized
according to procedure B using benzoyl chloride: yield 79%. The product was purified by CC
1
(
ethyl acetate/petroleum ether 7:3); H NMR (400 MHz, DMSO) δ 10.59 (s, 1H), 8.94 (d, J = 2.0
Hz, 1H), 8.71 (d, J = 1.9 Hz, 1H), 8.48 (t, J = 2.1 Hz, 1H), 8.05 – 7.98 (m, 2H), 7.70 (d, J = 1.2
1
3
Hz, 1H), 7.66 – 7.49 (m, 4H); C NMR (101 MHz, DMSO) δ 166.51, 141.67, 141.58, 136.57,
1
3
34.55, 132.49, 129.50, 129.16, 128.96, 128.20, 127.54, 124.85, 123.73, 110.67; MS (ESI) m/z =
+
58.85 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)benzenesulfonamide (5a). The title compound was
synthesized through adding 1.4 mmol of the benzenesulfonyl chloride to a stirred solution of
0
.18 gm (0.7 mmol) of compound B dissolved in pyridine. The reaction was heated to 60°C and
left overnight. This was followed by the removal of solvent in vacuo: yield 89%. The product
1
was purified by precipitation in ethyl acetate; H NMR (500 MHz, DMSO) δ 10.17 (s, 1H), 8.65
22

ACCEPTED MANUSCRIPT
(d, J = 2.0 Hz, 1H), 8.28 (d, J = 2.3 Hz, 1H), 7.72 (t, J = 2.2 Hz, 1H), 7.65 (d, J = 1.5 Hz, 1H),
1
3
7
1
.64 – 7.56 (m, 6H); C NMR (126 MHz, DMSO) δ 148.22, 141.33, 140.14, 138.86, 134.95,
+
33.40, 129.53, 127.67, 126.75, 125.48, 124.92, 123.51, 110.32; MS (ESI) m/z = 387.94 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-2-methylbenzamide (6a). The compound was
synthesized according to procedure B using o-toluoyl chloride: yield 96%. The product was
1
purified by CC (ethyl acetate/petroleum ether 2:3); H NMR (400 MHz, DMSO) δ 10.63 (s, 1H),
8
.83 (d, J = 2.1 Hz, 1H), 8.70 (d, J = 2.1 Hz, 1H), 8.47 (s, 1H), 7.80 (d, J = 1.4 Hz, 1H), 7.69 (d,
J = 1.3 Hz, 1H), 7.53 (d, J = 7.6 Hz, 1H), 7.47 – 7.38 (m, 1H), 7.33 (t, J = 7.4 Hz, 2H), 2.41 (s,
1
3
3
1
3
H); C NMR (101 MHz, DMSO) δ 168.46, 141.13, 140.99, 140.63, 136.28, 136.19, 135.56,
30.69, 130.09, 128.37, 127.38, 127.12, 125.70, 124.44, 122.55, 110.21, 19.38; MS (ESI) m/z =
+
72.9 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-3-methylbenzamide (7a). The compound was
synthesized according to procedure B using m-toluoyl chloride: yield 87%. The product was
1
purified by CC (ethyl acetate/petroleum ether 2:3); H NMR (500 MHz, DMSO) δ 10.53 (s, 1H),
8
.93 (d, J = 2.3 Hz, 1H), 8.71 (d, J = 2.1 Hz, 1H), 8.46 (t, J = 2.2 Hz, 1H), 7.82 (s, 1H), 7.81 (d,
J = 1.4 Hz, 1H), 7.80 – 7.77 (m, 1H), 7.70 (d, J = 1.4 Hz, 1H), 7.45 (dd, J = 3.9, 1.8 Hz, 2H),
1
3
2
1
.42 (s, 3H); C NMR (126 MHz, DMSO) δ 166.17, 141.22, 141.15, 141.01, 137.86, 136.15,
34.13, 132.62, 128.42, 128.28, 128.19, 127.09, 124.94, 124.41, 123.28, 110.21, 20.96; MS
+
(ESI) m/z = 372.96 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-4-methylbenzamide (8a). The compound was
synthesized according to procedure B using p-toluoyl chloride: yield 90%. The product was
1
purified by CC (ethyl acetate/petroleum ether 1:1); H NMR (500 MHz, DMSO) δ 10.48 (s, 1H),
23

ACCEPTED MANUSCRIPT
8
1
.93 (d, J = 2.3 Hz, 1H), 8.70 (d, J = 2.1 Hz, 1H), 8.47 (t, J = 2.2 Hz, 1H), 7.93 (s, 1H), 7.91 (s,
13
H), 7.81 (d, J = 1.4 Hz, 1H), 7.70 (d, J = 1.4 Hz, 1H), 7.37 (d, J = 7.9 Hz, 2H), 2.40 (s, 3H); C
NMR (126 MHz, DMSO) δ 165.87, 142.19, 141.23, 141.18, 140.95, 136.20, 131.23, 129.05,
+
1
28.27, 127.80, 127.09, 124.40, 123.29, 110.21, 21.05; MS (ESI) m/z = 372.89 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-2-methoxybenzamide (9a). The compound was
synthesized according to procedure B using 2-methoxybenzoyl chloride: yield 60%. The product
1
was purified by CC (ethyl acetate/petroleum ether 3:7); H NMR (500 MHz, DMSO) δ 10.43 (s,
1
H), 8.84 (d, J = 2.2 Hz, 1H), 8.69 (d, J = 2.1 Hz, 1H), 8.47 (t, J = 2.1 Hz, 1H), 7.81 (d, J = 1.4
Hz, 1H), 7.70 (d, J = 1.3 Hz, 1H), 7.68 (dd, J = 7.6, 1.7 Hz, 1H), 7.57 – 7.49 (m, 1H), 7.21 (d, J
1
3
=
8.4 Hz, 1H), 7.09 (dd, J = 7.7, 7.2 Hz, 1H), 3.92 (s, 3H); C NMR (126 MHz, DMSO) δ
1
1
65.26, 156.62, 141.11, 140.97, 140.83, 135.94, 132.54, 129.78, 128.38, 127.16, 124.43, 124.12,
+
22.68, 120.55, 112.08, 110.19, 55.96; MS (ESI) m/z = 388.95 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-3-methoxybenzamide (10a). The compound was
synthesized according to procedure B using 3-methoxybenzoyl chloride: yield 84%. The product
1
was purified by CC (ethyl acetate/petroleum ether 3:2); H NMR (500 MHz, DMSO) δ 10.53 (s,
1
H), 8.92 (d, J = 2.3 Hz, 1H), 8.72 (d, J = 2.1 Hz, 1H), 8.46 (t, J = 2.2 Hz, 1H), 7.81 (d, J = 1.4
Hz, 1H), 7.70 (d, J = 1.4 Hz, 1H), 7.62 – 7.57 (m, 1H), 7.55 – 7.52 (m, 1H), 7.49 (t, J = 7.9 Hz,
1
3
1
1
1
H), 7.24 – 7.14 (m, 1H), 3.85 (s, 3H); C NMR (126 MHz, DMSO) δ 165.79, 159.26, 141.30,
41.12, 136.05, 135.50, 129.72, 128.30, 127.13, 124.44, 123.39, 121.54, 119.93, 117.75, 113.06,
+
10.22, 55.40; MS (ESI) m/z = 388.89 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-4-methoxybenzamide (11a). The compound was
synthesized according to procedure B using 4-methoxybenzoyl chloride: yield 72%. The product
24

ACCEPTED MANUSCRIPT
1
was purified by CC (ethyl acetate/petroleum ether 7:3); H NMR (500 MHz, DMSO) δ 10.46 (s,
1
8
3
1
H), 8.93 (d, J = 2.3 Hz, 1H), 8.68 (d, J = 2.1 Hz, 1H), 8.46 (t, J = 2.2 Hz, 1H), 8.02 (s, 1H),
.00 (s, 1H), 7.80 (d, J = 1.4 Hz, 1H), 7.69 (d, J = 1.4 Hz, 1H), 7.10 (s, 1H), 7.09 (s, 1H), 3.85 (s,
1
3
H); C NMR (126 MHz, DMSO) δ 165.41, 162.26, 141.25, 141.23, 140.78, 136.36, 129.78,
+
28.25, 127.05, 126.12, 124.36, 123.25, 113.74, 110.20, 55.49; MS (ESI) m/z = 388.95 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-4-(tert-butyl)benzamide (12a). The compound was
synthesized according to procedure B using 4-tert-butylbenzoyl chloride: yield 79%. The product
1
was purified by CC (ethyl acetate/petroleum ether 1:1); H NMR (500 MHz, DMSO) δ 10.49 (s,
1
H), 8.92 (d, J = 2.2 Hz, 1H), 8.70 (d, J = 2.1 Hz, 1H), 8.47 (t, J = 2.2 Hz, 1H), 7.94 (d, J = 8.4
Hz, 2H), 7.81 (d, J = 1.4 Hz, 1H), 7.70 (d, J = 1.4 Hz, 1H), 7.58 (d, J = 8.4 Hz, 2H), 1.33 (s, 9H);
1
3
C NMR (126 MHz, DMSO) δ 165.97, 154.96, 141.15, 140.93, 136.19, 131.38, 129.16, 128.26,
1
4
27.61, 127.07, 126.43, 125.34, 125.28, 124.39, 123.16, 110.19, 34.73, 30.89; MS (ESI) m/z =
+
14.92 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-2-chlorobenzamide (13a). The compound was
synthesized according to procedure B using 2-chlorobenzoyl chloride: yield 59%. The product
1
was purified by CC (ethyl acetate/petroleum ether 3:2); H NMR (500 MHz, DMSO) δ 10.88 (s,
1
H), 8.80 (d, J = 2.2 Hz, 1H), 8.73 (d, J = 2.0 Hz, 1H), 8.44 (t, J = 2.1 Hz, 1H), 7.81 (d, J = 1.4
Hz, 1H), 7.74 – 7.66 (m, 1H), 7.66 (dd, J = 7.5, 1.6 Hz, 1H), 7.62 – 7.58 (m, 1H), 7.55 (td, J =
1
3
7
1
1
.7, 1.7 Hz, 1H), 7.49 (td, J = 7.4, 1.2 Hz, 1H); C NMR (126 MHz, DMSO) δ 165.60, 141.34,
40.97, 140.44, 136.16, 135.85, 131.54, 129.96, 129.79, 129.06, 128.47, 127.35, 127.24, 124.54,
+
22.43, 110.25; MS (ESI) m/z = 392.9 (M+H)
25

ACCEPTED MANUSCRIPT
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-3-chlorobenzamide (14a). The compound was
synthesized according to procedure B using 3-chlorobenzoyl chloride: yield 65%. The product
1
was purified by CC (ethyl acetate/petroleum ether 3:2); H NMR (500 MHz, DMSO) δ 10.65 (s,
1
H), 8.92 (d, J = 2.3 Hz, 1H), 8.73 (d, J = 2.1 Hz, 1H), 8.44 (t, J = 2.2 Hz, 1H), 8.06 (t, J = 1.8
Hz, 1H), 7.98 – 7.92 (m, 1H), 7.81 (d, J = 1.4 Hz, 1H), 7.71 (qd, J = 2.2, 1.0 Hz, 2H), 7.61 (t, J =
1
3
7
1
3
.9 Hz, 1H); C NMR (126 MHz, DMSO) δ 164.61, 141.32, 141.27, 141.04, 136.10, 135.85,
33.32, 131.86, 130.57, 128.33, 127.49, 127.17, 126.62, 124.48, 123.43, 110.24; MS (ESI) m/z =
+
92.81 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-4-chlorobenzamide (15a). The compound was
synthesized according to procedure B using 4-chlorobenzoyl chloride: yield 78%. The product
1
was purified by CC (ethyl acetate/petroleum ether 7:3); H NMR (500 MHz, DMSO) δ 10.62 (s,
1
H), 8.91 (d, J = 2.2 Hz, 1H), 8.72 (d, J = 2.0 Hz, 1H), 8.44 (t, J = 2.2 Hz, 1H), 8.03 (d, J = 1.9
Hz, 1H), 8.02 (s, 1H), 7.81 (d, J = 1.3 Hz, 1H), 7.70 (d, J = 1.4 Hz, 1H), 7.66 (d, J = 1.9 Hz, 1H),
1
3
7
1
.65 (s, 1H); C NMR (126 MHz, DMSO) δ 164.98, 141.28, 141.24, 141.07, 136.92, 135.93,
+
32.82, 129.72, 128.63, 128.32, 127.15, 124.46, 123.42, 110.23; MS (ESI) m/z = 392.89 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-2-fluorobenzamide (16a). The compound was
synthesized according to procedure B using 2-fluorobenzoyl chloride: yield 68%. The product
1
was purified by CC (ethyl acetate/petroleum ether 7:3); H NMR (500 MHz, DMSO) δ 11.16 (s,
1
7
9
1
H), 9.01 (d, J = 2.0 Hz, 1H), 8.91 (d, J = 1.9 Hz, 1H), 8.66 (s, 1H), 7.89 (d, J = 1.4 Hz, 1H),
.83 (d, J = 1.4 Hz, 1H), 7.77 (td, J = 7.5, 1.7 Hz, 1H), 7.69 – 7.62 (m, 1H), 7.39 (dt, J = 12.0,
1
3
.0 Hz, 2H); C NMR (126 MHz, DMSO) δ 163.58, 159.09 (d, JC-F= 250.4 Hz), 139.63, 138.12,
36.94, 136.79, 133.40 (d, J = 8.5 Hz), 130.10 (d, J = 2.2 Hz), 129.87, 128.32, 125.57,
C-F
C-F
26

ACCEPTED MANUSCRIPT
1
25.51, 124.69 (d, JC-F= 3.5 Hz), 123.62 (d, JC-F= 14.1 Hz), 116.38 (d, JC-F= 21.5 Hz), 110.41;
+
MS (ESI) m/z = 376.89 (M+H)
N-(5-(4-bromothiophen-2-yl)pyridin-3-yl)-3-fluorobenzamide (17a). The compound was
synthesized according to procedure B using 3-fluorobenzoyl chloride: yield 79%. The product
1
was purified by CC (ethyl acetate/petroleum ether 1:1); H NMR (500 MHz, DMSO) δ 10.63 (s,
1
7
H), 8.92 (d, J = 2.3 Hz, 1H), 8.73 (d, J = 2.1 Hz, 1H), 8.45 (s, 1H), 7.86 (d, J = 7.8 Hz, 1H),
.83 – 7.80 (m, 2H), 7.71 (d, J = 1.4 Hz, 1H), 7.63 (td, J = 8.0, 5.9 Hz, 1H), 7.50 (td, J = 8.1, 2.2
1
3
Hz, 1H); C NMR (126 MHz, DMSO) δ 164.67, 161.94 (d, J
= 244.5 Hz), 141.29 (d, J C-F =
C-F
4
1
.4 Hz), 141.04, 136.40 (d, J C-F = 7.0 Hz), 135.85, 130.77 (d, J C-F = 8.0 Hz), 128.32, 127.16,
24.47, 124.01, 123.98, 123.43, 118.97 (d, JC-F = 21.2 Hz), 114.60 (d, JC-F = 23.0 Hz), 110.23;
+
MS (ESI) m/z = 377 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-4-fluorobenzamide (18a). The compound was
synthesized according to procedure B using 4-fluorobenzoyl chloride: yield 98%. The product
1
was purified by CC (ethyl acetate/petroleum ether 3:2); H NMR (500 MHz, DMSO) δ 10.57 (s,
1
2
H), 8.91 (d, J = 2.3 Hz, 1H), 8.72 (d, J = 2.1 Hz, 1H), 8.44 (t, J = 2.2 Hz, 1H), 8.12 – 8.05 (m,
1
3
H), 7.81 (d, J = 1.4 Hz, 1H), 7.70 (d, J = 1.4 Hz, 1H), 7.41 (t, J = 8.9 Hz, 2H); C NMR (126
MHz, DMSO) δ 164.96, 164.34 (d, J = 249.7 Hz), 141.26, 141.12 (d, J = 5.1 Hz), 136.03,
C-F
C-F
1
1
32.10 (d, J = 9.5 Hz), 130.60, 127.13, 124.44, 123.37, 115.62 (d, J = 22.0 Hz), 115.61,
C-F C-F
+
15.44, 110.23; MS (ESI) m/z = 376.98 (M+H)
N-(5-(4-bromothiophen-2-yl)pyridin-3-yl)-2-(trifluoromethyl)benzamide fluorobenzamide
(19a). The compound was synthesized according to procedure B using 2-trifluoromethylbenzoyl
1
chloride: yield 86%. The product was purified by CC (ethyl acetate/petroleum ether 3:2); H
27

ACCEPTED MANUSCRIPT
NMR (500 MHz, DMSO) δ 10.96 (s, 1H), 8.78 (d, J = 2.3 Hz, 1H), 8.74 (d, J = 2.1 Hz, 1H),
8
7
1
.40 (s, 1H), 7.89 (d, J = 7.8 Hz, 1H), 7.83 – 7.81 (m, 2H), 7.76 (ddd, J = 8.2, 7.4, 4.9 Hz, 2H),
1
3
.71 (d, J = 1.4 Hz, 1H); C NMR (126 MHz, DMSO) δ 166.24, 141.40, 140.90, 140.45,
35.80, 132.69, 132.62, 131.16, 130.46, 129.64, 128.61, 128.47, 127.23, 126.46 (q, J = 5.0, 2.2
C-F
+
Hz), 124.52, 122.48, 110.23; MS (ESI) m/z = 426.93 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-3-(trifluoromethyl)benzamide fluorobenzamide
(20a). The compound was synthesized according to procedure B using 3-trifluoromethylbenzoyl
1
chloride: yield 84%. The product was purified by CC (ethyl acetate/petroleum ether 3:2); H
NMR (500 MHz, DMSO) δ 10.78 (s, 1H), 8.93 (d, J = 2.3 Hz, 1H), 8.75 (d, J = 2.1 Hz, 1H),
8
.44 (t, J = 2.2 Hz, 1H), 8.30 (d, J = 7.9 Hz, 1H), 8.23 (d, J = 7.8 Hz, 1H), 8.18 (s, 1H), 7.81 (d,
13
J = 1.4 Hz, 1H), 7.77 (s, 1H), 7.72 (d, J = 1.4 Hz, 1H); C NMR (126 MHz, DMSO) δ 166.00,
1
3
1
41.02, 135.78, 135.03, 133.22, 130.10, 129.89, 129.40 (q, J = 3.5 Hz) , 128.58 (q, J = 7.2,
C-F C-F
.6 Hz), 128.36, 127.20, 125.50 (q, J = 3.9 Hz), 124.50, 124.35 (q, J = 7.7, 3.9 Hz), 123.60,
C-F
C-F
+
22.71, 110.25; MS (ESI) m/z = 426.88 (M+H)
N-(5-(4-Bromothiophen-2-yl)pyridin-3-yl)-4-(trifluoromethyl)benzamide fluorobenzamide
(21a). The compound was synthesized according to procedure B using 4-trifluoromethylbenzoyl
1
chloride: yield 80%. The product was purified by CC (ethyl acetate/petroleum ether 7:3); H
NMR (500 MHz, DMSO) δ 10.78 (s, 1H), 8.92 (d, J = 2.3 Hz, 1H), 8.74 (d, J = 2.1 Hz, 1H),
8
1
1
1
.46 (t, J = 2.2 Hz, 1H), 8.20 (s, 1H), 8.19 (s, 1H), 7.97 (s, 1H), 7.95 (s, 1H), 7.81 (d, J = 1.4 Hz,
1
3
H), 7.71 (d, J = 1.4 Hz, 1H); C NMR (126 MHz, DMSO) δ 164.93, 141.44, 141.31, 141.01,
37.93, 135.80, 131.87, 128.71, 128.36, 127.21, 126.95, 125.55 (q, JC-F= 7.4, 3.7 Hz), 124.51,
+
23.49, 110.25; MS (ESI) m/z = 426.93 (M+H)
28