Extended structure-activity study of thienopyrimidine-based EGFR inhibitors with evaluation of drug-like properties

Article abstract of DOI:10.1016/j.ejmech.2015.11.012

Thieno[2,3-d]pyrimidines are attractive derivatives for cancer treatment, among others through regulation of the epidermal growth factor receptor tyrosine kinase (EGFR-TK). In an extended SAR study, 44 new compounds of this class have been evaluated as inhibitors, while simultaneously focussing on ADME properties. Through the application of bioisosters, hybrid structures, solubilizing tails, and a combination approach several successful alterations in terms of activity and physiochemical properties were accomplished. Compounds based on benzylamines were found superior to aniline hybrid structures with respect to activity and ADME profile. Exploration of the former class revealed meta-and para amides as favourable 6-aryl substituents, contributing to an increase in activity and acting as a linker for solubilizing tails. Next, combinations of activity-inducing groups on the same scaffold resulted in new drug candidates. Compounds containing 6-aryls with the (2-(dimethylamino)ethyl)carbamoyl substituent were found equipotent to Erlotinib. Compared to this commercial drug, improved solubility and metabolic stability were observed. However, the thieno[2,3-d]pyrimidines with a solubilizing tail was by Caco-2 experiments found to have permeability issues, making further drug development difficult. Selected compounds were further analysed for toxicity and teratogenicity in zebrafish embryos. Two thienopyrimidines were both found to be less lethal than Erlotinib and to perform as well in terms of teratogenicity. Finally, the most promising thienopyrimidine drug was evaluated in a panel of human cancer cell lines, showing a clear potential for thienopyrimidines as anti-cancer agents.

Full text of DOI:10.1016/j.ejmech.2015.11.012

European Journal of Medicinal Chemistry 107 (2016) 255e274
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Extended structureeactivity study of thienopyrimidine-based EGFR
inhibitors with evaluation of drug-like properties
Steffen Bugge a, Audun Formo Buene a, Nathalie Jurisch-Yaksi b, Ingri Ullestad Moen a,
,
*
Ellen Martine Skjønsfjell a, Eirik Sundby c, Bård Helge Hoff a
a Department of Chemistry, Norwegian University of Science and Technology, Høgskoleringen 5, NO-7491 Trondheim, Norway
b Kavli Institute for Systems Neuroscience and Center for Neural Computation, Norwegian University of Science and Technology, Olav Kyrres gate 9, MTFS,
NO-7491 Trondheim, Norway
c Sør-Trøndelag University College, E.C. Dahls gate 2, NO-7004 Trondheim, Norway
a r t i c l e i n f o
a b s t r a c t
Article history:
Thieno[2,3-d]pyrimidines are attractive derivatives for cancer treatment, among others through regu-
lation of the epidermal growth factor receptor tyrosine kinase (EGFR-TK). In an extended SAR study, 44
new compounds of this class have been evaluated as inhibitors, while simultaneously focussing on ADME
properties. Through the application of bioisosters, hybrid structures, solubilizing tails, and a combination
approach several successful alterations in terms of activity and physiochemical properties were
accomplished. Compounds based on benzylamines were found superior to aniline hybrid structures with
respect to activity and ADME profile. Exploration of the former class revealed meta- and para amides as
favourable 6-aryl substituents, contributing to an increase in activity and acting as a linker for solubi-
lizing tails. Next, combinations of activity-inducing groups on the same scaffold resulted in new drug
candidates. Compounds containing 6-aryls with the (2-(dimethylamino)ethyl)carbamoyl substituent
were found equipotent to Erlotinib. Compared to this commercial drug, improved solubility and meta-
bolic stability were observed. However, the thieno[2,3-d]pyrimidines with a solubilizing tail was by Caco-
2 experiments found to have permeability issues, making further drug development difficult. Selected
compounds were further analysed for toxicity and teratogenicity in zebrafish embryos. Two thienopyr-
imidines were both found to be less lethal than Erlotinib and to perform as well in terms of teratoge-
nicity. Finally, the most promising thienopyrimidine drug was evaluated in a panel of human cancer cell
lines, showing a clear potential for thienopyrimidines as anti-cancer agents.
Received 17 September 2015
Received in revised form
4 November 2015
Accepted 5 November 2015
Available online 11 November 2015
Keywords:
EGFR
Structureeactivity relationship (SAR)
Thieno[2,3-d]pyrimidine
ADME
Erlotinib
Gefitinib
Zebrafish
© 2015 Elsevier Masson SAS. All rights reserved.
1. Introduction
namely potency, selectivity, and ADMET (absorption, distribution,
metabolism, excretion and toxicology) [4]. However, in the devel-
Kinase inhibitors have through the last couple of decades been a
success never before witnessed in the pharmaceutical world [1];
especially the impact on oncology has been tremendous [2]. Ki-
nases that often are associated with tumour development include
members of the epidermal growth factor receptor (EGFR) family
[3]. Through persistent research, this field has brought fundamental
change in cancer treatment, but also to the understanding of ki-
nases and how to develop small molecule kinase inhibitors. In the
latter, it is essential to find a balance between three vital properties,
opment of small molecule kinase modulators, there is still a gap in
figuring out optimal ADME profiles of preclinical compounds [5].
Two fundamental properties affecting the absorption of oral
drugs are solubility and permeability [6]. Insufficient aqueous sol-
ubility of a compound can be particularly critical in drug develop-
ment [7], as a compound must be dissolved before it can pass the
intestinal membrane [8]. Lipophilicity is a key parameter in
determining the solubility of a compound [9]. Consequently, the
somewhat higher mass and more lipophilic nature of orally
bioavailable kinase inhibitors [10] indicate that solubility could be
an issue. Moreover, a study of the 14 marketed kinase inhibitors up
to 2011 revealed more challenges caused by low solubility, rather
than low permeability [5].
* Corresponding author.
E-mail addresses: steffen.bugge@ntnu.no (S. Bugge), audun.f.buene@ntnu.no
(A.F. Buene), nathalie.jurisch-yaksi@ntnu.no (N. Jurisch-Yaksi), ingriumoen@gmail.
com (I.U. Moen), ellenmsk@stud.ntnu.no (E.M. Skjønsfjell), Eirik.Sundby@hist.no
(E. Sundby), bard.helge.hoff@chem.ntnu.no (B.H. Hoff).
As the discovery of novel drugs with new modes of action are
getting increasingly difficult, many follow-up drugs are being
http://dx.doi.org/10.1016/j.ejmech.2015.11.012
0223-5234/© 2015 Elsevier Masson SAS. All rights reserved.

256
S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
developed. Such me-too drugs often imitate the successful drugs on
the market by borrowing structural fragments [11], and can often
have advantages in the form of cost, safety, and efficiency [12].
Hybrid structures using the aniline fragments from the commercial
EGFR-TK inhibitors Gefitinib (3-chloro-4-fluoroaniline) [13,14],
Erlotinib (3-ethynylaniline) [15,16] and Lapatinib (3-chloro-4-((3-
fluorobenzyl)oxy)aniline) [17e19] have been demonstrated to be
potent EGFR-TK antagonists.
Our group has developed thienopyrimidine-based EGFR-TK in-
hibitors [20], including hybrid based structures [21]. The most
active thienopyrimidine compound was equipotent to the estab-
lished drug Erlotinib in enzymatic assays. Herein we have evaluated
how hybrid structures and bioisosteric replacement of the key
substituents on our initial lead affect EGFR-TK inhibitory- and
selectivity profile, as well as ADME properties.
aldehyde to a primary amide [29e31]. The major obstacles were the
low solubility of the substrates and the need for tuning of every
single transformation. This gave low to mediocre yields and diffi-
cult purifications. In consequence, functionalized boronic acid de-
rivatives were prepared, followed by Suzuki cross-coupling.
The majority of the cross-couplings were performed under
standard conditions resulting in complete conversion. On the
contrary, with (2-fluoro-4-formylphenyl)boronic acid and (2-
carbamoylphenyl)boronic acid, there are innate properties of the
organoboron compounds resulting in low conversions. Satisfactory
yields were achieved by increasing catalyst loading from 1 to 5%
Pd(PPh3)4.
2.3. Hybrid structures: anilines vs benzylamines at C-4
We have recently evaluated the suitability of various 4-amino
groups as part of thieno[2,3-d]pyrimidine EGFR-TK inhibitors [21].
To investigate this approach further, the aniline fragments attached
to C-4 of the core structure were combined with previously iden-
tified activity-inducing substitution patterns at the 6-aryl moiety.
These hybrid structures were then tested as EGFR-TK inhibitors,
and the results are presented in Fig. 1.
For the three series of benzylamine based compounds (series a,
b and c), the activity increased upon introduction of an ortho-
methoxy (1) or a para-hydroxymethyl (2) substituent, and a com-
bination of the two gave compounds 3 with improved inhibitory
2. Result and discussion
2.1. Design of the inhibitors
A previous SAR study performed by us on the thieno[2,3-d]py-
rimidine scaffold, resulted in a potent EGFR-TK inhibitor (see
Scheme 1). However, evaluation of key physiochemical parameters
influencing ADME properties of this lead compound had not been
done. Earlier reports on quinazolines have concluded that both
substitution at C-2 [22] and exchanging the pyrimidines with
pyridines or other diazines [23] lowers the activity. Our data on
thienopyrimidines [20,21] and pyrrolopyrimidines [24] have indi-
cated that modification of the 6-aryl group is an excellent way to
fine-tune the potency, while variation at the 4-amino group can
more dramatically affect the activity. With this in mind, we set out
to explore how variation at the 4-amino group and the 6-aryl group
affected potency and ADME properties. For the former, previous
results had indicated that pyridines and other heterocycles strongly
reduced activity, while more promising results were seen with
anilines [21]. For the latter, the benzylic hydroxyl group might be a
metabolic soft spot. This initiated a search for a bioisosteric
replacement to improve the overall performance of the compound
in vivo [25,26].
properties. The 4-formyl-2-methoxy analogues
4 were also
included and tested since they are potential metabolites of the
corresponding hydroxymethyl compounds 3. As seen, compound
4c still is a very potent inhibitor.
The aniline substituted compounds (def) were less active and
less soluble than the benzylamine based analogues; hence, prepa-
ration of the monosubstituted 4-hydroxymethyl compounds where
not prioritised. However, it was clear that including both an ortho-
methoxy and a para-hydroxymethyl group in the same molecule for
the aniline based series d, e, and f improved the potency. The gain
in activity for the less potent inhibitors was larger than for the more
potent ones, as also noted for the benzylamine-based series. This is
in line with our previous conclusion that each series needs specific
fine tuning to maximize its activity [21]. Furthermore, with the
substitution pattern selected herein, the benzylamine-based
structures are indicated to be more potent.
2.2. Synthesis
A challenge in development of new inhibitors is selectivity be-
tween kinases. To determine the selectivity profile of the prepared
drug candidates, a selection of compounds was evaluated in assays
towards 48 additional kinases (data in Supplementary material).
Fig. 2 compares the selectivity profile of Erlotinib, the Erlotinib
hybrid 3e and compound 3c, for kinases showing more than 20%
inhibition when subjected to any of these compounds.
As clearly visible from the data, both compound 3c and the
Erlotinib hybrid 3e are more selective for EGFR-TK than Erlotinib.
For only four out of the 21 kinases the thienopyrimidines were
more active than the reference drug, and chances for off-target
effects are thus reduced. The same general trends were also seen
for eight other compounds evaluated in the same kinase panel.
An additional advantage of using benzylamines as C-4 sub-
stitutes, compared to the anilines, is improved solubility. Intro-
duction of a substituted benzylic carbon have been reported to
disrupt crystal stacking capability [32]. This effect was confirmed
experimentally by comparing turbidimetric solubility of com-
pounds 3c and the Gefitinib hybrid 3f. Here 3c showed more than 18
fold higher solubility than 3f, and almost twice as good as Erlotinib.
Permeability is difficult to estimate based only on physi-
ochemical properties (e.g. pKa, lipophilicity and solubility), among
others because of efflux transporters [33]. These systems render the
The chemical diversity within the series was introduced by
previously developed methodologies, starting with 6-bromo-4-
chlorothieno[2,3-d]pyrimidine (I) [27,28], as shown in Scheme 2.
Variation at C-4 was introduced through an amination of com-
pound I with anilines and benzylamines, giving compounds IIaef
[20,21]. To search for bioisosteric replacements on the 6-aryl moi-
ety a Suzuki coupling with the resulting 6-bromo-4-N-subsituted
thieno[2,3-d]pyrimidines was used.
The initial strategy utilized commercially available boronic
acids. To obtain some of the hydroxymethyl substituted analogues,
reduction of the parent aldehyde, after the Suzuki coupling, was
necessary. However, when more advanced substitution patterns at
the 6-aryl ring were required, the post-modifications were more
challenging. One particular example was the direct conversion of an
Scheme 1. Lead structure and design strategy.

S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
257
Scheme 2. Synthesis of the thieno[2,3-d]pyrimidines investigated in this study.
drugs ineffective by active transport [34]. As evaluated by Caco-2
assay, compound 3c (efflux ratio ¼ 1.48) and Erlotinib (efflux
ratio ¼ 0.80) demonstrated good permeability. This indicates that
there is no major permeability issues with the thieno[2,3-d]py-
rimidine scaffold. The permeability study of the Gefitinib hybrid 3f
was inconclusive, possibly due to solubility issues.
In conclusion, the lead structure 3c demonstrated good solubi-
lity, permeability and metabolic stability. The most active hybrid
structure, 3f, had metabolic stability comparable to 3c, but had low
solubility. Erlotinib, on the other hand, was indicated to be more
metabolically labile as compared to 3c and 3f. Accordingly, the
benzylamine-based structures seemed more promising and this
compound class was therefore evaluated further.
After oral dosing, a drug must be metabolically stable to achieve
an efficient concentration at the target site. The prepared inhibitors
were subjected to metabolizing conditions in human liver micro-
somes (HLM). Again compound 3c performed quite well
2.4. Bioisosteric replacement at 6-aryl
(t1/2 ¼ 154 min, CLint ¼ 9.02
hybrid 3f (t1/2 ¼ 156 min, CLint ¼ 8.90
demonstrated similar values, while Erlotinib (t1/2 ¼ 31.7 min,
L minꢀ1 mg proteinꢀ1) was indicated to be somewhat
m
L minꢀ1 mg proteinꢀ1), the Gefitinib
L minꢀ1 mg proteinꢀ1
As clearly highlighted in the prior section 4-CH2OH, 4-CHO and
2-OMe proved beneficial as 6-aryl substituents to give thieno[2,3-
d]pyrimidines with potent EGFR-TK inhibitory properties. However,
aldehydes, benzylic alcohols and methoxy groups could easily
suffer from metabolic inactivation in vivo, and alternatives were
thus sought. SAR information was gathered using the C-4 (R)-
methylbenzylamine substituted thieno[2,3-d]pyrimidines (series
b) with variation at the 6-aryl moiety. For this purpose series c
could have been used, but series b was selected since these de-
rivatives were expected more metabolically stabile and easier to
prepare. Additionally, the inhibition data could more easily be base
lined with our previous SAR studies.
m
)
CLint ¼ 43.8
m
unstable. This could cause toxicity issues owing to the need for high
daily doses for this reference drug.
ErloƟnib
Compound 3c
Compound 3e
100
90
80
70
60
50
40
30
20
10
0
Fig. 2. Selectivity profiling of Erlotinib (blue), compound 3c (red) and compound 3e
(green) as % inhibition at 500 nM. Data for 3c is taken from Ref. [20]. (For interpre-
tation of the references to colour in this figure legend, the reader is referred to the web
version of this article.)
Fig. 1. Effect of variation at C-4 (aef) and C-6 (1e4) on EGFR-TK inhibitory potency.
Erlotinib used as a positive control had IC50 ¼ 0.4 nM. Data for series b and c is taken
a)
from Ref. [20], while data for series 1 is taken from Ref. [21]. Not determined.

258
S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
At the meta- and para positions of the 6-aryl ring a carboxylic
TK inhibitor (IC50 ¼ 2 nM). Related findings have been reported by
Takahashi et al. [39], with the hypothesis that the increase in
binding affinity was related to an appropriate conformation
enforced by the two ortho-methoxy groups. Through the testing of
the 2,3-di-OMe derivative 18b (IC50 ¼ 13 nM) and 2,4-di-OMe
analogue 20b (IC50 ¼ 5 nM), it was clear that the origin of this
observation was probably not caused by electronic effects.
acid, a primary amide and a methyl sulfone were introduced, see
Fig. 3. Additionally, compound 15b bearing an aldehyde in meta
position was made for exploring SAR. The methyl sulfonyl group
proved least successful (IC50 ¼ 41 nM and 26 nM), while the car-
boxylic acid derivatives 7b and 12b gave mediocre to good activities
(IC50 ¼ 19 nM and 8 nM respectively). Moreover, the amides 8b and
13b (IC50 ¼ 7 nM and 5 nM respectively) were equally potent and
slightly better than the corresponding hydroxymethyl compounds
2b and 6b. This demonstrates a successful bioisosteric replacement.
The 2-methoxy group has been hypothesised to be a weak
hydrogen bond acceptor in EGFR-ligand binding [20]. Stronger
hydrogen bond acceptors were sought, and a primary amide [35]
and 1,2-di-OMe substitution [36] both fulfil this criteria. The 2,3-
di-OMe substituted compound 18b (IC50 ¼ 13 nM) was slightly
less potent than the benchmark 2-methoxy analogue 1b, while the
2-CONH2 compound 17b, were 30 fold less active. In the former
case, it might be that the positioning of the additional methoxy
group was not optimal for binding with EGFR-TK. In the latter case,
both steric and conformational effects might cause the reduction in
EGFR-TK inhibitory properties. Additionally, extra desolvation en-
ergy must be compensated by favourable binding energy to give
improved inhibition. Further studies would be required for a more
thorough understanding of these effects.
The insertion of a fluorine atom can, in addition to altered
physiochemical properties, also influence the potency of the drug
candidate, and thus provide useful SAR information. A series of four
compounds with fluorine substitution in ortho- and para-position
of the 6-aryl moiety (10b, 11b, 16b and 19b), as well as their cor-
responding hydrogen analogues (2b, 5b,15b and 1b), was evaluated
as EGFR-TK inhibitors (see Fig. 3). Upon inspection of the data, it is
clear that a para-fluoro substituent does not influence the activity
significantly. However, in ortho position a clear drop in potency was
seen, especially for the aldehyde analogue 11b (IC50 ¼ 59 nM). As
ortho- and para substitution influenced the activity unequally,
electronic effects could probably be ruled out. A conformational
search around the 6-aryl bond of compounds 4b and 11b identified
different global minima, but the relative difference in energy was
small (see Supplementary material).
Another strategy to increase aqueous solubility is to introduce
polar functional groups, especially those that at a physiological pH
are charged. A noteworthy feature of kinases, as compared to many
other drug targets, are that they frequently are able to tolerate such
ionized groups [8]. The reason for this is the characteristic solvent-
exposed front pocket located outside the ATP binding site, which at
the same time can be utilised to enhance selectivity and activity
[40,41]. The thieno[2,3-d]pyrimidines substituted at 6-aryl with a
primary amide in meta- or para position are some of the most
potent monosubstituted analogues (IC50 ¼ 5 and 7 nM respec-
tively), and also provide an excellent handle for inclusion of a sol-
ubilizing tail. The resulting secondary amides still keeps their
ability to act as both a hydrogen bond donor and -acceptor, while
simultaneously functioning as a linker. A 2-(dimethylamino)ethyl
group was chosen as a representative test substituent, see Fig. 3. To
our delight, this modification in para position (compound 22b) did
not only lead to improved solubility, but also a 3-fold increase in
activity as compared to the corresponding primary amide 8b. Mo-
lecular modelling showed a high docking score and the possibility
for new interactions from the introduced functionality (see Fig. 5).
Furthermore, when a 2-(dimethylamino)ethyl group was intro-
duced, the meta derivative 23b (IC50 ¼ 0.9 nM) turned out to be
twice as potent as the aforementioned para compound 22b
(IC50 ¼ 2 nM).
2.5. Combination of activity-inducing groups
Because of these inspiring data, some of the most potency-
inducing substituents were combined in the same structure. Priv-
ileged 6-aryl fragments were fused with the activity-inducing (S)-
(2-hydroxy-1-phenylethyl)amino moiety. The resulting com-
pounds were mostly more active than their deshydroxy analogues
(see Fig. 4). However, degree of increase in activity varied within
the series, from more than 6-fold for compounds 25 (IC50 from 58 to
9 nM), to no particular gain for compounds 22. This is in consis-
tency with previous observations; the benefit of the additional
hydroxyl group is largest for the least potent inhibitors evaluated.
Docking, followed by molecular dynamics, revealed interactions
that might be important for the binding to EGFR-TK. For compound
Calculations by HF/6e31 G(d) methods have indicated 2-
arylsubstituted thiophenes to be relatively planar structures
[37,38]. An approach to improve the solubility of such compounds
is disruption of molecular planarity by including ortho-sub-
stituents. As previous results had shown the 2,4,6-tri-Me and 2,4,6-
tri-i-Pr substituted analogues to be inactive [20], it was surprising
to find the 2,6-di-OMe compound 21b to be a highly potent EGFR-
Fig. 3. IC50 values of compounds presented in this and previous [20] SAR studies of 4-methylbenzylaminthieno[2,3-d]pyrimidines (series b) as EGFR-TK inhibitors.

S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
259
Fig. 4. Effect of side chain substitution on activity. Data for 6b and 25b is taken from Ref. [20].
22c, especially the two hydrogen bonds from the NeH groups in the
solubilizing tail, through water to Asp800, and an interaction with
Lys745 are worth highlighting (see Fig. 5). As the latter is not so
frequently observed for other compounds having the (S)-(2-
hydroxy-1-phenylethyl)amino fragment, it could be that the
former two interactions enforces a slightly distorted binding mode.
The stronger interaction with Lys745 could thus weaken some of
the other interactions, resulting in the similar activities for com-
pounds 22b and 22c. Therefore, no further compounds combining
the solubilizing tail (2-(dimethylamino)ethyl)carbamoyl and the
(S)-(2-hydroxy-1-phenylethyl)amino fragment c were prepared.
With this limitation in mind, the combination approach was
instead applied within the 6-aryl fragment. The selected groups
were 2-OMe and 2,6-di-OMe, the solubilizing tails 3-, 4- and 5-
CONH(CH2)2NMe2, as well as 3- and 5-CH2OH. Again, the alde-
hydes 4b, 26b and 32b were included as they are potential me-
tabolites of the corresponding hydroxymethyl compounds 3b, 27b
and 33b. The methyl esters 28b, 30b, 34b and 36b are synthetic
precursors to the secondary amides 29b, 31b, 35b and 37b, and
were evaluated as EGFR-TK inhibitors because of their potential
usefulness in a pro-drug approach. Fig. 6 summarises the activity
data.
hydroxymethyl groups; the 2-OMe-3-CH2OH compound 27b was
now the most potent inhibitor of the two with an IC50 value of
1.5 nM. In combination with the 2-methoxy group the potency
increase for the amide derivative 24b (IC50 ¼ 2 nM) was not as
prominent as seen the 4-hydroxymethyl analogue 3b
(IC50 ¼ 0.6 nM). However, when such a primary amide was
extended with a solubilizing tail in compounds 31b and 35b, some
of the most active compounds in series b was obtained. IC50 values
of 0.5 nM for both compounds, demonstrate that these analogues
are equipotent to Erlotinib in this enzymatic assay. For compound
29b the somewhat lower activity could be caused by a less optimal
positioning of substituents or an intramolecular hydrogen bond
between the amide hydrogen and the methoxy oxygen, visible in
1H NMR.
Further, merging the 2,6-di-methoxy groups of fragment 21
with the 4-(2-(dimethylamino)ethyl)carbamoyl tail of fragment 22
in compound 37b, using 36b as a precursor, yielded a highly active
antagonist (IC50 ¼ 0.4 nM).
2.6. SAR model
A summary of the SAR for thieno[2,3-d]pyrimidines as EGFR-TK
inhibitors is illustrated in Fig. 7.
Considering the two aldehydes (compound 26b and 32b) the
most notable result was the decreased activity for the 2-OMe-3-
CHO substituted compound compared to the 5- substituted deriv-
ative, showing IC50 values of 21 and 3 nM respectively. Surprisingly,
the trend was reversed upon reduction of the aldehydes to
Most importantly, the potency is highly dependent on the ste-
reochemistry at the chiral centre and a free NeH group at C-4. Next,
the activity is prominently affected by the electron distribution and
size of the aromatic unit in the benzylamine moiety at C-4,
favouring an unsubstituted phenyl ring over pyridyls, 2-thienyl and
2-furyl. Lastly, the aliphatic side chain at the stereogenic centre can
Fig. 5. Ligand-EGFR contacts for compound 22c. Highlighted amino acids are within
5 Å distance from the docked ligand. The colours indicate residue type: green e
lipophilic residues; red e acidic residues; blue e polar residues; purple e basic resi-
dues. The lines indicate contacts with the enzyme. Ligand atoms that are exposed to
solvent are marked with grey spheres. Only those interactions are shown that occur
more than 30% of the 10 ns simulation time. (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 6. Comparison of IC50 values for 6-(2-methoxyphenyl) and 6-(2,6-
dimethoxyphenyl) substituted thieno[2,3-d]pyrimidin-4-amines. Data for compound
3b and 4b is taken from Ref. [20]. Not determined.
a)

260
S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
Fig. 7. Structureeactivity relationship model for thieno[2,3-d]pyrimidines as EGFR-TK inhibitors based on this and previous work [20,21].
be altered to optimize EGFR-TK binding, with a hydroxymethyl or a
methyl group being the most favourable. When moving over to the
C-6 position, activity cliffs [42] are not as frequently observed. In
combination with the 2-methoxy group, an additional methoxy
group in position 6, or a secondary amide with a solubilizing tail in
4- or 5 position has proven favourable. Whereas, a bulky group in 4-
position or lipophilic groups at position 2, 4 and 6 simultaneously
were detrimental for the activity.
Fig. 8. In-silico determined sites on 21c prone to CYP3A4 oxidations. Radius of the ring
is proportional to the likelihood of metabolism.
2.7. Further biological evaluation
A selection of compounds was tested in ADME assays to evaluate
solubility, metabolic stability and permeability. The results are
summarised in Table 1.
As expected from the introduction of a hydroxyl group, com-
pounds in series c demonstrated an improvement in aqueous sol-
ubility as compared to the analogues in series b. Moreover, a further
increased solubility was seen for the series of compounds con-
taining a solubilizing tail, with compound 31b being at least 50 fold
more soluble than the Gefitinib hybrid 3f.
further stability increase was displayed by compound 31b (t1/
¼ 227 min, CLint ¼ 6.1
m
L minꢀ1 mg proteinꢀ1), having only one
2
ortho methoxy group.
The permeability study showed that both compound 3c (efflux
ratio 1.48) and the 2,6-dimethoxy analogue 21c (efflux
¼
ratio ¼ 0.97) had good permeability. On the other hand, the highly
active amide based derivatives 24c, 31b, 35b and 37b, having the
solubilizing tail, seems to have challenges with permeability.
Further work has to be conducted in order to find suitable bio-
isosteric replacements.
Overall, the human liver microsomal (HLM) stability tests indi-
cated that most of the prepared compounds have good metabolic
stability. However, compound 21c was somewhat metabolically
Selected compounds were also analysed for toxicity and tera-
togenicity in zebrafish embryos. Briefly, wild-type embryos were
exposed to increasing concentrations of 3c, 21c, 24c, 37b, Erlotinib
and DMSO as vehicle control. Embryos were treated from shield
stage (at the onset of gastrulation) and monitored over 3.5 days of
development for survival and morphological abnormalities. LC50
and EC50 were calculated at 3.5 days of development. As indicated
in Fig. 9a, no effects on survival were observed at any time point
following exposure with high doses of 3c, 24c and DMSO
labile (t1/2 ¼ 34 min, CLint ¼ 41
m
L minꢀ1 mg proteinꢀ1), which is
comparable to the stability of Erlotinib. This could be a result of CYP
metabolism on the electron rich position 3 (ortho and para to the
methoxy groups) on the 6-aryl ring and/or demethylation of one of
the two methoxy groups, see Fig. 8. A more than 4 fold improved
stability was observed when introducing a secondary amide at
position
4
m
in
compound
37b
(t1/2
¼
142
min,
CLint ¼ 9.7
L minꢀ1 mg proteinꢀ1), which could be due to steric
(LC50 > 100 mM). In contrast, compounds 21c and 37b were lethal
hindrance [43] or the lower oxidation potential. Additionally, a
Table 1
ADME data for Erlotinib and compounds 3c, 3f, 21c, 22c, 23b, 24c, 31b, 35b and 37b.
Erlotinib
3c
3f
21c
22c
23b
24c
1.5
11.5
308
31
31b
35b
37b
0.4
65
142
40
IC50 EGFR [nM]
Solubility [
M]a
0.4
20
32
0.3
37.5
154
1.48
2
2
0.9
37.5
34
2
65
66
0.9
0.5
0.5
m
>100
95
>100
227
18
>100
116
63
HLM t1/2 [min]
Efflux ratio
156
b
0.80
0.97
>20
8.9
a
Turbidimetric aqueous solubility given as an estimated mid-range precipitation concentration.
The experiment was inconclusive, likely due to low solubility.
b

S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
261
Table 2
during development with LC50 ¼ 18
mM and 15 mM, respectively.
Cell proliferation study (IC50 values in
human cancer cell lines.
m
M) using Erlotinib and compound 3c in
Interestingly, embryos treated with 37b died within 24 h of treat-
ment, while 21c was lethal to the embryos at around 54e70 h post
fertilization, suggesting that the toxic mechanism of 21c and 37b
are different. While results obtained upon treatment with 3c, 21c,
24c and 37b were consistent between experiments; Erlotinib
induced variable toxic responses at high concentrations as indi-
cated by the large standard deviation in Fig. 9a.
Cancer type
Cell line
Erlotinib
Comp. 3c
Lung model
Lung
A-431a
PC-9
0.4 0.0
0.07 0.01
>20
1.1 0.3
0.09 0.01
NCIeH82a
Calu-3
AU-565a
BT-474
Hs578T
SkBr3
11
0
4.7 0.7
3.3 0.6
4.2 0.5
6.3 0.4
Breast
We also determined the morphological defects present in the
surviving larvae at 3.5 days post fertilization. The EC50 determina-
tion was based on occurrence of any morphological defect (tail
curvature, defect in hatching, heart beat and blood circulation) and
are indicated in Fig. 9b. Compounds 24c, 3c and Erlotinib were well
13
>100
>100
3
11
43
17
1
2
0
Ovarian
C-33Aa
OVCAR-3
SW480
H-9
0.9 0.0
4.6 1.1
>100
1.3 0.6
9.9 0.5
1.3 0.6
1.6 0.1
9.7 2.4
Gastric
Leukemia
19
12
18
1
2
1
tolerated by the embryos up to 25
mations of the notochord at higher doses. Indeed, treatment with
100 M and 50 M of 24c, 3c and Erlotinib resulted in kinked tails of
various severities. Representative embryos are shown in the
Supplementary material. EC50 values were found to be 46 M, 49
and 60 M for compounds 24c, 3c and Erlotinib, respectively.
Compound 21c induced cardiac oedema, blood circulation defects
and kinked tails with an EC50 value of 12 M. Compound 37b, which
was lethal at dose ranging from 25 M, did not have any toxic effect
at dose below 12.5 M. Therefore, it was not possible to calculate
mM, but induced some malfor-
U937
Head & neck
CAL-27a
2.8 1.4
m
m
FaDua
18
5.8 1.2
6
43
37
6
6
Normal
MRC-5
m
mM
a
Data is taken from Ref. [20].
m
why inhibitory activity is measurable for the HS578T cell line. To
evaluate toxicity, compound 3c was also tested in an assay of
rapidly growing healthy cells, the MRC-5 lung fibroblast cells. As
compared to Erlotinib, compound 3c proved to be less toxic.
Overall, it can be concluded that compound 3c cannot outcompete
Erlotinib as a single therapeutic agent. However, given its favour-
able ADME properties and selective kinase inhibitory profile,
compound 3c might still be an attractive candidate for combination
therapy in HER driven cancers.
m
m
m
EC50 values. Finally, treatment with the vehicle only (DMSO) even at
high doses (10%) did not result in any teratogenic phenotype. Based
on high LC50 and EC50 values, the compounds 24c and 3c performed
as well as Erlotinib in terms of teratogenicity. Moreover, these two
compounds were also less toxic than Erlotinib for early embryos.
Compound 3c has a good balance between EGFR-TK activity on
one hand and ADME properties on the other. A summary of the cell
proliferation data towards human cancer cell lines and one normal
cell line is compiled in Table 2.
3. Conclusion
The investigated cells showed similar sensitivity towards the
two compounds, indicating that the main cellular toxicity is due to
EGFR inhibition. Both Erlotinib and 3c was particular active towards
the EGFR driven cell lines A-431 and PC-9. Compounds 21c, 23b and
35b were also evaluated in the PC-9 cell line showing IC50 values of
Based on the thienopyrimidine scaffold, an extended structure-
activity investigation focussing on drug-like properties has been
performed. The study, which reports on 44 new chemical entities,
involves evaluation of benzylamine- vs aniline based structures,
bioisosteric replacement of metabolic soft spots and fluorinated
derivatives. Also studied are solubility manipulations by disruption
of planarity of biaryls and incorporation of amine based tails.
Overall, the benzylamine based thienopyrimidines showed higher
solubility and kinase activity as compared to their aniline coun-
terparts. Further focus on these privileged structures revealed that
bioisosteric replacement of a benzyl alcohol function could suc-
cessfully be achieved using amide functionalities. Moreover, this
handle was subsequently used to attach a (2-(dimethylamino)
ethyl) group, boosting potency and giving high solubility and
metabolic stability. The solubilizing tail, however, resulted in
permeability issues for these compounds. Based on a balance be-
tween potency, ADME properties and zebrafish embryo toxicity
compound 3c was regarded as the most suited drug candidate, and
0.29, 0.23 and 0.20 mM, respectively. This correlates well with the
activities seen in the enzymatic inhibition studies. Overall, Erlotinib
was more active than 3c in nine of the cases, while 3c proved more
potent than Erlotinib towards two lung, three breast and one
gastric cancer cell line. As Erlotinib and compound 3c are equi-
potent EGFR-TK inhibitors with similar permeability, the sensitivity
differences seen in the proliferation assay are likely due to off target
effects. Erlotinib is a more broad spectrum kinase inhibitor than 3c
(see Fig. 2 and Supplementary material), and thus show lower
inhibitory potency for the most sensitive cell lines. Off-target ef-
fects for Erlotinib have previously been documented [44]. On the
other hand, 3c has a better solubility profile, allowing for IC50
measurement at higher concentration, which might be the reason
Fig. 9. Toxicity and teratogenicity in zebrafish embryos. A) Relative survival after 3.5 days as response to inhibitor concentration based on four experiments. B) Tetratogeneicity
score after 3.5 days as a function of inhibitor concentration based on four experiments.

262
S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
was thus further profiled in human cancer cell lines. Generally,
somewhat lower potency than Erlotinib was observed. Kinase
profiling indicates this to be caused by Erlotinib being a less se-
lective inhibitor than compound 3c. However, given its selective
kinase inhibitory profile, the identified thienopyrimidine drug
candidate might be attractive for combination therapy.
structures of 2J6M and ligands were subsequently docked using
€
induced-fit docking (IFD) of Schrodinger [46e48]. The resulting
docked poses were analysed using Glide pose viewer tool. All the
graphical pictures were made using Maestro.
4.4. General procedure for thermal amination of 4-chloro-thieno
[2,3-d]pyrimidines
4. Experimental
Compound I (1.00 g, 4.01 mmol) was mixed with the benzyl-
amine (2e3 eq.) and i-PrOH (2e10 mL) and agitated at 80 ꢃC for
1e50 h, under nitrogen atmosphere. Then the mixture was cooled
to rt, concentrated in vacuo, diluted with water (50 mL) and diethyl
ether (100 mL) or EtOAc (100 mL). After phase separation, the water
phase was extracted with more diethyl ether (2 ꢂ 50 mL) or EtOAc
(2 ꢂ 50 mL). The combined organic phases were washed with
saturated aq NaCl solution (25e50 mL), dried over anhydrous
Na2SO4, filtered and concentrated in vacuo. The crude oil was pu-
rified by drying under reduced pressure to constant weight, by
silica-gel column chromatography or crystallized.
4.1. General
K2CO3, Pd(PPh3)4, NaBH4, and nearly all the boronic acid were
from Sigma Aldrich. (4-Formyl-2-methoxyphenyl) boronic acid was
from Combi-Blocks. 6-Bromo-4-chlorothieno[2,3-d]pyrimidine
(compound I), IIa-f, 1ae1f, 2b, 2c, 3b, 3c, 4b, 4c, 5b, 6b and 25b
were prepared and characterized in other studies from our labo-
ratory [20,21,28,45]. Silica-gel column chromatography was per-
formed using silica-gel 60A from Fluka, pore size 40e63 mm. Celite
545 from Fluka was also used.
4.2. Analyses
4.5. General procedure for esterification of benzoic acids
1H and 13C NMR spectra were recorded with Bruker Avance 400
spectrometer operating at 400 MHz and 100 MHz, respectively. 19
The benzoic acid (1e4 g) was dissolved in methanol (25e75 mL),
before conc. H2SO4 (2e6 mL) was added dropwise while stirring.
The reaction mixture was heated to reflux and stirred for 24 h. The
reaction mixture was quenched in a saturated aq NaHCO3 solution
(100e200 mL) ensuring neutral to basic pH, and the aqueous phase
was extracted with CH2Cl2 (4 ꢂ 50 mL). The combined organic
phases were washed with saturated aq NaCl solution (25e50 mL),
dried over anhydrous Na2SO4 and filtered. Concentration yielded
the target compound in high yield and purity.
F
NMR was performed on a Bruker Avance 500 operating at 564 MHz.
For 1H and 13C NMR chemical shifts are in ppm rel. to DMSO-d6,
while for 19F NMR the shift values are relative to hexafluorobenzene.
Coupling constants are in hertz. HPLC (Agilent 110-Series) with a
G1379A degasser, G1311A Quatpump, G1313A ALS autosampler and
a G1315D Agilent detector (230 nm) was used to determine the
purity of the synthesised compounds. All compounds evaluated for
EGFR-TK inhibitory potency had a purity of ꢁ95%. Conditions:
Poroshell C18 (100 ꢂ 4.6 mm) column, flow rate 0.8 mL/min, elution
starting with water/CH3CN (90/10), 5 min isocratic elution, then
linear gradient elution for 35 min ending at CH3CN/water (100/0).
The software used with the HPLC was Agilent ChemStation. Accurate
mass determination (ESI) was performed on an Agilent G1969 TOF
MS instrument equipped with a dual electrospray ion source. Ac-
curate mass determination in positive and negative mode was per-
formed on a “Synapt G2-S” Q-TOF instrument from Waters. Samples
were ionized by the use of an ASAP probe, no chromatography
separation was used before the mass analysis. FTIR spectra were
recorded on a Thermo Nicolet Avatar 330 infrared spectrophotom-
eter. All melting points are uncorrected and measured by a Stuart
automatic melting point SMP40 apparatus.
4.5.1. Methyl 3-bromo-4-methoxybenzoate [49]
Methyl 3-bromo-4-methoxybenzoate was prepared as
described in Section 4.5, starting with 3-bromo-4-methoxybenzoic
acid (2.01 g, 8.66 mmol). The procedure yielded 2.13 g (8.66 mmol,
100%) of methyl 3-bromo-4-methoxybenzoate as a white crystal-
line product, mp. 94e95 ꢃC (lit [49]. 97e98 ꢃC); HPLC-purity: 99%,
tR ¼ 21.9 min; 1H NMR (600 MHz, DMSO-d6)
d: 8.08e8.06 (m, 1H),
7.97e7.94 (m, 1H), 7.25e7.22 (m, 1H), 3.94 (s, 3H), 3.83 (s, 3H); 13
C
4.3. Molecular modelling
NMR (150 MHz, DMSO-d6) d: 164.8, 159.2, 133.6, 130.6, 123.1, 112.5,
110.6, 56.7, 52.1; HRMS (APCI/ASAP, m/z): 244.9812 (calcd.
79
The X-ray crystal structures of the protein 2J6M (Wild-type
EGFR) were prepared using the protein preparation wizard, which
is part of the Maestro software package (Maestro, v8.5;
C9H BrO3, 244.9813, [MþH]þ). 1H NMR data are in accordance
10
with that reported previously [49].
€
4.5.2. Methyl 3-bromo-2-methoxybenzoate [50]
Schrodinger, LLC, New York, NY, USA). Bond orders and formal
charges were added for het-groups, and hydrogens were added to
all atoms in the system. Water molecules beyond 5 Å from het-
groups were removed. To alleviate steric clashes that may exist in
the original PDB structures, an all-atom constrained minimization
was carried out with the Impact Refinement module (Impref)
€
(Impact, v5.0; Schrodinger, LLC) using the OPLS-2005 force field.
The minimization was terminated when the energy converged or
the RMSD reached a maximum cutoff of 0.30 Å. The resulting
protein structures were used in the following docking study. Li-
gands were drawn using ChemBioDraw (ChemBioDraw Ultra 13.0,
CambridgeSoft, PerkinElmer) and were prepared using LigPrep2.2
Methyl 3-bromo-2-methoxybenzoate was prepared as
described in Section 4.5, starting with 3-bromo-2-methoxybenzoic
acid (1.00 g, 4.33 mmol). The procedure yielded 1.07 g (4.33 mmol,
100%) of methyl 3-bromo-2-methoxybenzoate as clear oil, HPLC-
purity: 99%, tR ¼ 21.5 min; 1H NMR (600 MHz, DMSO-d6)
d:
€
(LigPrep, v2.2; Schrodinger, LLC). For the computational investiga-
7.88e7.84 (m, 1H), 7.73e7.70 (m, 1H), 7.18 (t, J ¼ 7.8, 1H), 3.86 (s,
tion of the receptor-inhibitor structures, the energy minimized
3H), 3.81 (s, 3H); 13C NMR (150 MHz, DMSO-d6)
d: 165.1, 155.7,

S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
263
137.0, 130.3, 126.8, 125.6, 118.1, 61.8, 51.4; HRMS (APCI/ASAP, m/z):
4.07 mmol). The crude product was purified by crystallization from
Et2O (50 mL) to yield 617 mg (2.12 mmol, 52%) of the desired
compound as white crystals, mp. 143e144 ꢃC; HPLC purity: 99%,
79
244.9813 (calcd. C9H BrO3, 244.9813, [MþH]þ). 1H NMR data are in
10
accordance with that reported previously [50].
tR ¼ 14.0 min; 1H NMR (600 MHz, DMSO-d6)
d: 8.18e8.16 (m, 1H),
4.5.3. Methyl 4-bromo-3-methoxybenzoate [51]
8.06e8.03 (m, 1H), 7.11e7.08 (m, 1H), 3.83 (s, 3H), 3.82 (s, 3H), 1.28
(s, 12H); 13C NMR (150 MHz, DMSO-d6)
d: 167.5, 165.8, 137.9, 134.4,
121.1, 117.2, 110.9, 83.4 (2C), 55.7, 51.8, 24.6 (4C); HRMS (APCI/ASAP,
11
m/z): 293.1563 (calcd. C
H
15 22
BO5, 293.1560, [MþH]þ). 1H NMR, 13
C
NMR and MS data were in accordance with reported literature [53].
4.6.2. Methyl 2-methoxy-3-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)benzoate
Methyl 4-bromo-3-methoxybenzoate was prepared as
described in Section 4.5, starting with 4-bromo-3-methoxybenzoic
acid (845 mg, 3.67 mmol). The procedure yielded 808 mg
(3.30 mmol, 90%) of methyl 4-bromo-3-methoxybenzoate a white
crystalline product, mp 52e53 ꢃC; HPLC-purity: 99%, tR ¼ 22.6 min;
1H NMR (600 MHz, DMSO-d6)
d: 7.75e7.72 (m, 1H), 7.54e7.53 (m,
1H), 7.48e7.45 (m,1H), 3.92 (s, 3H), 3.86 (s, 3H); 13C NMR (150 MHz,
DMSO-d6) d: 165.6, 155.5, 133.4, 130.4, 122.6, 116.6, 112.3, 56.4, 52.4;
The target compound was prepared as described in Section 4.6,
starting with methyl 3-bromo-2-methoxybenzoate (200 mg,
0.82 mmol). The crude product was purified by silica-gel column
chromatography (n-pentane/EtOAc, 3/1), Rf ¼ 0.46. This gave 129 mg
(0.44 mmol, 54%) of the desired compound as a clear oil, HPLC pu-
HRMS (APCI/ASAP, m/z): 243.9735 (calcd. C9H799BrO3, 243.9735,
[M]þ). 1H NMR and 13C NMR data are in accordance with that re-
ported previously [51].
rity: 96%, tR ¼ 14.2 min; 1H NMR (600 MHz, DMSO-d6)
d: 7.78e7.73
4.5.4. Methyl 4-bromo-3,5-dimethoxybenzoate [52]
(m, 2H), 7.21 (t, J ¼ 7.5, 1H), 3.82 (s, 3H), 3.76 (s, 3H), 1.31 (s, 12H); 13
C
NMR (150 MHz, DMSO-d6) d: 166.3, 163.9, 139.8, 133.3, 125.2, 123.4,
113.9, 83.6 (2C), 63.1, 52.1, 24.5 (4C); IR (neat, cmꢀ1): 3400, 2977,
2934, 1731, 1590, 1456, 1360, 1251, 1139, 981, 851, 673; HRMS (APCI/
11
ASAP, m/z): 293.1556 (calcd. C
H
BO5, 293.1560, [MþH]þ).
15 22
4.6.3. Methyl 3-methoxy-4-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)benzoate
Methyl 4-bromo-3,5-dimethoxybenzoate was prepared as
described in Section 4.5, starting with 4-bromo-3,5-
dimethoxybenzoic acid (2.99 mg, 11.49 mmol). The procedure
yielded 3.04
g (11.05 mmol, 96%) of methyl 4-bromo-3,5-
dimethoxybenzoate a white crystalline product, mp 122e123 ꢃC;
HPLC-purity: 99%, tR ¼ 21.7 min; 1H NMR (600 MHz, DMSO-d6)
d:
7.22 (s, 2H), 3.91 (s, 6H), 3.88 (s, 3H); 13C NMR (150 MHz, DMSO-d6)
d
: 165.6, 156.6 (2C), 130.0, 105.6, 105.2 (2C), 56.6 (2C), 52.5; HRMS
79
(APCI/ASAP, m/z): 273.9841 (calcd. C
H
10 12
BrO4, 273.9841, [MþH]þ).
The target compound was prepared as described in Section 4.6,
starting with methyl 4-bromo-3-methoxybenzoate (501 mg,
2.05 mmol). The crude oil (600 mg) was used without further pu-
rification. 1H NMR spectroscopy confirmed the presence of 54% of
the desired product by comparison with the literature [54]. 1H NMR
1H NMR and 13C NMR data are in accordance with that reported
previously [52].
4.6. General procedure for palladium catalysed boronylation
(600 MHz, DMSO-d6)
7.44e7.43 (m, 1H), 3.86 (s, 3H), 3.80 (s, 3H), 1.28 (s, 12H).
d: 7.65e7.63 (m, 1H), 7.54e7.52 (m, 1H),
The methyl benzoate (200e1500 mg), PdCl2(CH3CN)2 (0.02 eq)
and SPhos (0.08 eq) were added to a pressure tube. Then, 1,4-
dioxane (0.5e3.5 mL) and Et3N (0.5e2 mL) were added under N2
atmosphere, followed by dropwise addition of HBpin (1.5 eq) while
stirring under N2 atmosphere. The reaction mixture was heated to
80 ꢃC and stirred for 5e20 h. After cooling to rt, the reaction
mixture was filtered through a Celite pad. If further purification was
performed, it is specified in each specific example.
4.6.4. Methyl 3,5-dimethoxy-4-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)benzoate
4.6.1. Methyl 4-methoxy-3-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)benzoate [53]
The target compound was prepared as described in Section 4.6,
starting with methyl 4-bromo-3,5-dimethoxybenzoate (1.50 g,
5.45 mmol). The crude oil was used without further purification. 1H
NMR spectroscopy confirmed the presence of 33% of the desired
compound in the oil. A small amount of white crystals formed in
the oil, and were filtered off, mp. 162e163 ꢃC, HPLC purity: 97%,
The target compound was prepared as described in Section 4.6,
starting with methyl 3-bromo-4-methoxybenzoate (999 mg,
tR ¼ 14.9 min; 1H NMR (600 MHz, DMSO-d6)
d: 7.10 (s, 2H), 3.86 (s,
3H), 3.76 (s, 6H), 1.27 (s, 12H); 13C NMR (150 MHz, DMSO-d6)
d:

264
S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
166.0, 162.5 (2C), 132.7, 113.6, 104.0 (2C), 83.7 (2C), 55.7 (2C), 52.3,
24.4 (4C); IR (neat, cmꢀ1): 3001, 2977, 2944, 1710, 1573, 1347, 1316,
139.22, 131.4, 128.6 (2C), 127.3 (2C), 125.6 (2C), 123.3, 121.2 (2C),
118.4,114.9, 62.5; IR (neat, cmꢀ1): 3307, 3117, 2848,1618,1510,1448,
1379, 1207, 1029, 775, 749, 689; HRMS (APCI/ASAP, m/z): 334.1014
(calcd. C19H16N3OS, 334.1014, [MþH]þ).
1248, 1220, 1114, 1061, 996, 858, 699; HRMS (APCI/ASAP, m/z):
10
321.1629 (calcd. C
H
BO6, 321.1624, [M]þ).
16 23
4.7. General procedure for Suzuki coupling of on 6-bromothieno
[2,3-d]pyrimidines
4.7.3. 4-(4-(Benzylamino)thieno[2,3-d]pyrimidin-6-yl)-3-
methoxybenzaldehyde (4a)
Compounds II (200e1000 mg) were mixed with the selected
boronic acid (1.2 eq), fine powdered K2CO3 (3 eq), Pd(PPh3)4
(0.01 eq) and 1,4-dioxane/water (1/1 by vol.%, 4e8 mL). The reac-
tion was then stirred at 80 ꢃC for 2e5 h under nitrogen atmosphere.
The solvent was removed and the product was diluted with water
(25e50 mL) and extracted with Et2O or EtOAc (25e100 mL), the
water phase was extracted with more diethyl ether or EtOAc
(2 ꢂ 25 mL). The combined organic phases were washed with
saturated aq NaCl solution (25 mL), dried over anhydrous Na2SO4,
filtered and concentrated in vacuo. Purification was performed as
specified for each individual compound.
Compound 4a was prepared as described in Section 4.7, starting
with IIa (300 mg, 0.937 mmol) and (4-formyl-2-methoxyphenyl)
boronic acid (205 mg, 1.124 mmol). The crude product was purified
using silica-gel column chromatography (EtOAc/n-pentane, 1/1),
Rf ¼ 0.26. This gave 264 mg (0.703 mmol, 75%) of 4a as a yellow
solid, mp. 194e197 ꢃC; HPLC purity: 97%, tR ¼ 25.9 min; 1H NMR
4.7.1. (4-(4-(Benzylamino)thieno[2,3-d]pyrimidin-6-yl)phenyl)
methanol (2a)
(400 MHz, DMSO-d6)
d
: 10.03 (s, 1H), 8.61 (t, J ¼ 5.9, 1H), 8.38 (s,
1H), 8.37 (s, 1H), 7.94e7.92 (m, 1H), 7.67e7.65 (m, 2H), 7.40e7.32
(m, 4H), 7.28e7.24 (m, 1H), 4.79 (d, J ¼ 5.9, 2H), 4.06 (s, 3H); 13C
NMR (100 MHz, DMSO-d6) d: 192.3, 166.1, 156.6, 155.8, 154.4, 139.3,
136.6, 132.6, 128.4 (2C), 128.3, 127.5, 127.4 (2C), 126.9, 122.9, 119.6,
116.3, 112.1, 56.1, 43.3; IR (neat, cmꢀ1): 3262, 3029, 2946, 2836,
1688, 1574, 1527, 1257, 784, 691; HRMS (APCI/ASAP, m/z): 376.1114
(calcd. C21H18N3O2S, 376.1120, [MþH]þ).
Compound 2a was prepared as described in Section 4.7, starting
with IIa (300 mg, 0.937 mmol) and (4-(hydroxymethyl)phenyl)
boronic acid (171 mg, 1.124 mmol). The crude product was purified
using silica-gel column chromatography (EtOAc/n-pentane, 3/2),
Rf ¼ 0.21, and recrystallized from MeOH. This gave 188 mg
(0.541 mmol, 58%) of 2a as a light yellow solid, mp. 231e234 ꢃC;
4.7.4. 3-Methoxy-4-(4-(phenylamino)thieno[2,3-d]pyrimidin-6-yl)
benzaldehyde (4d)
HPLC purity: 99%, tR ¼ 22.0 min; 1H NMR (400 MHz, DMSO-d6)
d:
8.48 (t, J ¼ 5.8, 1H), 8.34 (s, 1H), 8.05 (s, 1H), 7.64e7.62 (m, 2H),
7.45e7.42 (m, 2H), 7.39e7.32 (m, 4H), 7.27e7.24 (m, 1H), 5.27 (t,
J ¼ 5.6, 1H), 4.77 (d, J ¼ 5.8, 2H), 4.54 (d, J ¼ 5.7, 2H); 13C NMR
(100 MHz, DMSO-d6) d: 164.9, 156.5, 153.9, 143.2, 139.3, 138.3, 131.6,
Compound 4d was prepared as described in Section 4.7, starting
with IId (500 mg, 1.633 mmol) and (4-formyl-2-methoxyphenyl)
boronic acid (353 mg, 1.960 mmol). The crude product was purified
using silica-gel column chromatography (EtOAc), Rf ¼ 0.30. This
gave 478 mg (1.323 mmol, 81%), of 4d as a yellow solid, mp.
218e225 ꢃC; HPLC purity: 98%, tR ¼ 26.8 min; 1H NMR (400 MHz,
128.4 (2C), 127.4 (2C), 127.3 (2C), 126.9, 125.4 (2C), 117.5, 114.9, 62.4,
43.4; IR (neat, cmꢀ1): 3236, 3127, 2966, 2847, 1585, 1259, 1086,
1081, 1012, 798, 752, 694; HRMS (APCI/ASAP, m/z): 348.1168 (calcd.
C
20H18N3OS, 348.1171, [MþH]þ).
4.7.2. (4-(4-(Phenylamino)thieno[2,3-d]pyrimidin-6-yl)phenyl)
methanol (2d)
DMSO-d6) d: 10.05 (s, 1H), 9.76 (s, 1H), 8.57 (s, 1H), 8.51 (s, 1H),
8.05e8.01 (m, 1H), 7.88e7.84 (m, 2H), 7.74e7.68 (m, 2H), 7.44e7.38
(m, 2H), 7.16e7.10 (m, 1H), 4.09 (s, 3H); 13C NMR (100 MHz, DMSO-
d6) d: 192.2, 167.0, 155.8, 154.5, 153.7, 139.1, 136.8, 133.4, 128.6 (2C),
127.3, 125.6, 123.4, 122.8, 121.4 (2C), 119.3, 117.0, 112.2, 56.2; IR
(neat, cmꢀ1): 3401, 3051, 2932, 2843, 1677, 1615, 1542, 1275, 1159,
781, 744, 685; HRMS (APCI/ASAP, m/z): 362.0961 (calcd.
C
20H16N3O2S, 362.0963, [MþH]þ).
Compound 2d was prepared as described in Section 4.7, starting
with IId (300 mg, 0.980 mmol) and (4-(hydroxymethyl)phenyl)
boronic acid (179 mg, 1.176 mmol). The crude product was purified
using silica-gel column chromatography (Et2O/EtOAc, 19/1),
Rf ¼ 0.26, and recrystallized from THF (10 mL) using n-pentane
(10 mL) as anti-solvent. This gave 24 mg (0.072 mmol, 7%) of 2d as a
beige solid, mp. 244e248 ꢃC; HPLC purity: 96%, tR ¼ 22.8 min; 1H
4.7.5. 4-(4-((3-Ethynylphenyl)amino)thieno[2,3-d]pyrimidin-6-yl)-
3-methoxybenzaldehyde (4e)
NMR (400 MHz, DMSO-d6) d: 9.65 (s, 1H), 8.49 (s, 1H), 8.27 (s, 1H),
7.89e7.84 (m, 2H), 7.73e7.68 (m, 2H), 7.50e7.45 (m, 2H), 7.44e7.37
(m, 2H), 7.14e7.08 (m, 1H), 5.31 (t, J ¼ 5.6, 1H), 4.56 (d, J ¼ 5.6, 2H);
13C NMR (100 MHz, DMSO-d6)
d: 165.8, 154.4, 153.3, 143.5, 139.26,

S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
265
Compound 4e was prepared as described in Section 4.7 using
compound IIe$HCl (500 mg, 1.364 mmol) and (4-formyl-2-
methoxyphenyl)boronic acid (294 mg, 1.636 mmol). Hot EtOAc
was used in the workup. The crude product was purified using
silica-gel column chromatography (CH2Cl2/EtOAc, 3/1), Rf ¼ 0.44.
This gave 168 mg (0.436 mmol, 32%) of 4e as a yellow solid, mp.
221e225 ꢃC; HPLC purity: 91%, tR ¼ 29.1 min; 1H NMR (400 MHz,
(2C), 126.9, 126.5, 124.0, 123.2, 117.6, 115.5, 64.7, 62.6, 56.4; IR (neat,
cmꢀ1): 3272, 2871, 1580, 1515, 1352, 1311, 1028, 772, 697; HRMS
(APCI/ASAP, m/z): 378.1280 (calcd.
C21H20N3O2S, 378.1276,
[MþH]þ).
4.7.8. (R)-4-(4-((1-Phenylethyl)amino)thieno[2,3-d]pyrimidin-6-
yl)benzoic acid (7b)
DMSO-d6) d: 10.05 (s,1H), 9.82 (s,1H), 8.60e8.54 (m, 2H), 8.12e8.10
(m, 1H), 8.05e8.01 (m, 1H), 7.92e7.88 (m, 1H), 7.74e7.70 (m, 2H),
7.46e7.40 (m, 1H), 7.24e7.21 (m, 1H), 4.22 (s, 1H), 4.09 (s, 3H); 13C
NMR (100 MHz, DMSO-d6) d: 192.2, 167.1, 155.8, 154.3, 153.6, 139.5,
136.9, 133.7, 129.1, 128.4, 127.2, 126.4, 123.9, 122.8, 121.9, 121.7, 119.1,
117.1, 112.2, 83.5, 80.7, 56.2; IR (neat, cmꢀ1): 3370, 3226, 1676, 1567,
1541, 1359, 1272, 1017, 780, 682; HRMS (APCI/ASAP, m/z): 386.0959
(calcd. C22H16N3O2S, 386.0963, [MþH]þ).
Compound 7b was prepared as described in Section 4.7, starting
IIb$HCl (308 mg, 0.831 mmol) and 4-boronobenzoic acid (165 mg,
0.997 mmol). Saturated aq NH4Cl solution (25 mL) was used in the
workup. The crude product was purified using silica-gel column
chromatography (Et2O/THF, 7/3), Rf ¼ 0.16, followed by recrystal-
lized from 1,4-dioxane/n-pentane (1/1, 4 mL). This gave 14 mg
4.7.6. 4-(4-((3-Chloro-4-fluorophenyl)amino)thieno[2,3-d]
pyrimidin-6-yl)-3-methoxybenzaldehyde (4f)
(0.037 mmol, 5%) of 7b as a white solid, mp 253e255 ꢃC (dec.);
20
HPLC purity: 99%, tR ¼ 21.3 min; [
a]
¼ ꢀ491.0 (c 0.51, DMSO); 1H
D
NMR (400 MHz, DMSO-d6)
d: 13.08 (s br, 1H), 8.36 (s, 1H), 8.33 (d,
J ¼ 8.0, 1H), 8.31 (s, 1H), 8.08e8.04 (m, 2H), 7.81e7.79 (m, 2H),
7.46e7.41 (m, 2H), 7.36e7.30 (m, 2H), 7.25e7.20 (m, 1H), 5.56e5.48
(m, 1H), 1.57 (d, J ¼ 7.0, 3H); 13C NMR (100 MHz, DMSO-d6)
d: 166.8,
Compound 4f was prepared as described in Section 4.7, starting
with IIf$HCl (500 mg, 1.266 mmol) and (4-formyl-2-
methoxyphenyl)boronic acid (273 mg, 1.519 mmol). The crude
product was purified by crystallization from THF (29 mL), using
toluene (15 mL) as anti-solvent, yielding 175 mg (0.423 mmol, 33%)
of 4f as a dark orange solid, mp 252e254 ꢃC; HPLC purity: 98%,
165.7, 156.0, 154.4, 144.5, 137.2, 136.7, 130.4 (2C), 130.3, 128.3 (2C),
126.8, 126.1 (2C), 125.6 (2C), 117.5, 117.4, 49.1, 22.5; IR (neat, cmꢀ1):
3292, 2982, 1676, 1574, 1293, 858, 767, 695; HRMS (APCI/ASAP, m/
z): 376.1114 (calcd. C21H18N3O2S, 376.1120, [MþH]þ).
4.7.9. (R)-4-(4-((1-Phenylethyl)amino)thieno[2,3-d]pyrimidin-6-
yl)benzamide (8b)
tR ¼ 31.3 min; 1H NMR (400 MHz, DMSO-d6)
d: 10.04 (s, 1H), 9.87 (s,
1H), 8.56 (s, 1H), 8.50 (s, 1H), 8.22 (dd, J ¼ 6.8, 2.8, 1H), 8.02e7.98
(m, 1H), 7.83e7.78 (m, 1H), 7.73e7.67 (m, 2H), 7.50e7.43 (m, 1H),
4.08 (s, 3H); 13C NMR (100 MHz, DMSO-d6)
d: 192.2, 167.1, 155.8,
154.1, 153.5,153.2 (d, J ¼ 242.6), 136.9, 136.5 (d, J ¼ 3.0), 133.9, 128.4,
127.2, 122.8, 122.6, 121.4 (d, J ¼ 7.1), 118.97 (d, J ¼ 18.6), 118.96, 117.0,
116.8 (d, J ¼ 21.9), 112.2, 56.2; 19F NMR (564 MHz, DMSO-d6, C6F6)
d
: ꢀ125.40 (s, dec.); IR (neat, cmꢀ1): 3357, 1674, 1534, 1437, 1267,
1199, 860, 775; HRMS (APCI/ASAP, m/z): 414.0471 (calcd.
C
Compound 8b was prepared as described in Section 4.7, starting
with IIb$HCl (750 mg, 2.023 mmol) and (4-carbamoylphenyl)
boronic acid (400 mg, 2.428 mmol). The crude product was purified
by hot filtration from MeOH (100 mL). This gave 425 mg
20H14ClFN3O2S, 414.0479, [MþH]þ).
4.7.7. (S)-2-((6-(3-(Hydroxymethyl)phenyl)thieno[2,3-d]
pyrimidin-4-yl)amino)-2-phenylethan-1-ol (6c)
(1.135 mmol, 56%) of 8b as an off-white solid, mp 304e306 ꢃC
20
(dec.); HPLC purity: 98%, tR ¼ 21.4 min; [
a]
¼ ꢀ441.0 (c 0.51,
D
DMSO); 1H NMR (400 MHz, DMSO-d6)
d
: 8.34e8.26 (m, 3H), 8.06 (s
br, 1H), 8.03e7.98 (m, 2H), 7.79e7.74 (m, 2H), 7.47e7.41 (m, 3H),
7.36e7.30 (m, 2H), 7.26e7.20 (m, 1H), 5.57e5.47 (m, 1H), 1.57 (d,
J ¼ 7.0, 3H); 13C NMR (100 MHz, DMSO-d6)
d: 167.1, 165.5, 155.9,
154.2, 144.5, 137.0, 135.7, 133.9, 128.6 (2C), 128.3 (2C), 126.7, 126.1
(2C), 125.3 (2C), 117.4, 116.8, 49.1, 22.5; IR (neat, cmꢀ1): 3432, 3286,
2964, 1658, 1580, 1497, 1311, 1106, 845, 752, 700; HRMS (APCI/ASAP,
m/z): 375.1275 (calcd. C21H19N4OS, 375.1280, [MþH]þ).
Compound 6c was prepared as described in Section 4.7 using
compound IIc (302 mg, 0.862 mmol) and (3-(hydroxymethyl)
phenyl)boronic acid (157 mg, 1.035 mmol). The crude product was
purified using silica-gel column chromatography (EtOAc/Et2O, 1/1),
Rf ¼ 0.18. This gave 283 mg (0.750 mmol, 87%) of 6c as a white
4.7.10. (R)-6-(4-(Methylsulfonyl)phenyl)-N-(1-phenylethyl)thieno
[2,3-d]pyrimidin-4-amine (9b)
crystalline solid, mp. 120e123 ꢃC; HPLC purity: 99%, tR ¼ 19.1 min;
20
[
a]
¼ ꢀ322.0 (c 1.02, DMSO); 1H NMR (400 MHz, DMSO-d6)
d:
D
8.30e8.26 (m, 2H), 8.23 (d, J ¼ 8.0, 1H), 7.73 (s br, 1H), 7.58e7.54 (m,
1H), 7.48e7.42 (m, 3H), 7.35e7.29 (m, 3H), 7.26e7.20 (m, 1H),
5.50e5.42 (m, 1H), 5.35 (t, J ¼ 5.6, 1H), 5.03 (t, J ¼ 5.6, 1H), 4.59 (d,
J ¼ 5.6, 2H), 3.82e3.72 (m, 2H); 13C NMR (100 MHz, DMSO-d6)
d:
165.0, 156.4, 153.8, 143.8, 141.1, 138.2, 133.0, 129.1, 128.2 (2C), 127.0

266
S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
20
Compound 9b was prepared as described in Section 4.7, starting
HPLC purity: 99%, tR ¼ 22.0 min; [
a]
¼ ꢀ446.9 (c 0.45, DMSO); 1H
D
with IIb$HCl (252 mg, 0.680 mmol) and (4-(methylsulfonyl)phenyl)
boronic acid (163 mg, 0.816 mmol). The crude product was purified
by silica-gel column chromatography (Et2O/THF, 19/1), Rf ¼ 0.19,
followed by recrystallized from 1,4-dioxane (2 mL). This gave 45 mg
NMR (400 MHz, DMSO-d6) d: 13.30 (s br,1H), 8.36 (s,1H), 8.35e8.31
(m, 2H), 8.30 (s, 1H), 7.98e7.92 (m, 1H), 7.92e7.87 (m, 1H),
7.67e7.60 (m, 1H), 7.46e7.41 (m, 2H), 7.37e7.29 (m, 2H), 7.26e7.19
(m, 1H), 5.57e5.46 (m, 1H), 1.58 (d, J ¼ 7.0, 3H); 13C NMR (100 MHz,
(0.109 mmol, 16%) of 9b as white needle shaped crystals, mp
DMSO-d6) d: 166.9, 165.3, 155.9, 154.2, 144.6, 136.9, 133.5, 131.8,
20
241e243 ꢃC; HPLC purity: 97%, tR ¼ 23.3 min; [
a
]
¼ ꢀ426.6 (c 1.02,
129.9, 129.8, 129.1, 128.3 (2C), 126.7, 126.0 (2C), 125.8, 117.4, 116.4,
49.1, 22.5; IR (neat, cmꢀ1): 3367, 1681, 1581, 1522, 1237, 1206, 1117,
753, 701, 692; HRMS (APCI/ASAP, m/z): 376.1114 (calcd.
D
DMSO); 1H NMR (400 MHz, DMSO-d6)
d
: 8.41 (s,1H), 8.37 (d, J ¼ 7.9,
1H), 8.32 (s, 1H), 8.07e8.03 (m, 2H), 7.95e7.91 (m, 2H), 7.46e7.42
(m, 2H), 7.36e7.30 (m, 2H), 7.26e7.20 (m, 1H), 5.58e5.48 (m, 1H),
C
21H18N3O2S, 376.1120, [MþH]þ).
3.27 (s, 3H), 1.58 (d, J ¼ 7.0, 3H); 13C NMR (100 MHz, DMSO-d6)
d:
165.9, 156.0, 154.5, 144.4, 140.0, 137.9, 135.7, 128.3 (2C), 128.2 (2C),
126.7, 126.1 (2C), 126.0 (2C), 118.3, 117.3, 49.1, 43.5, 22.4; IR (neat,
cmꢀ1): 3351, 2976, 1574, 1510, 1288, 1138, 1120, 1086, 827, 771, 702;
HRMS (APCI/ASAP, m/z): 410.0993 (calcd. C21H20N3O2S2, 410.0997,
[MþH]þ).
4.7.13. (R)-3-(4-((1-Phenylethyl)amino)thieno[2,3-d]pyrimidin-6-
yl)benzamide (13b)
4.7.11. (R)-3-Fluoro-4-(4-((1-phenylethyl)amino)thieno[2,3-d]
pyrimidin-6-yl)benzaldehyde (11b)
Compound 13b was prepared as described in Section 4.7,
starting with IIb$HCl (753 mg, 2.031 mmol) and (3-
carbamoylphenyl)boronic acid (0.402 mg, 2.438 mmol). The crude
product was purified by silica-gel column chromatography (Et2O/
EtOAc, 1/1), Rf ¼ 0.18. This gave 711 mg (1.899 mmol, 93%) of 13b as
an off-white solid, mp 135e138 ꢃC; HPLC purity: 99%, tR ¼ 22.0 min;
Compound 11b was prepared as described in Section 4.7, start-
ing with IIb$HCl (201 mg, 0.542 mmol) and (2-fluoro-4-
formylphenyl)boronic acid (182 mg, 1.084 mmol, 2 eq). The
boronic acid was added slowly over 1 h. The reaction time was 25 h.
The crude product was purified by silica-gel column chromatog-
20
[a
]
¼ ꢀ392.5 (c 1.02, DMSO); 1H NMR (400 MHz, DMSO-d6)
d:
D
8.36e8.30 (m, 1H), 8.32 (s, 1H), 8.30 (s, 1H), 8.28e8.25 (m, 1H), 8.16
(s br, 1H), 7.92e7.86 (m, 1H), 7.84e7.78 (m, 1H), 7.63e7.54 (m, 1H),
7.53 (s br, 1H), 7.47e7.40 (m, 2H), 7.37e7.28 (m, 2H), 7.26e7.18 (m,
1H), 5.58e5.46 (m, 1H), 1.57 (d, J ¼ 7.0, 3H); 13C NMR (100 MHz,
raphy (Et2O/n-pentane, 6/4), Rf
¼
0.17. This gave 128 mg
DMSO-d6) d: 167.3, 165.2, 155.8, 154.0, 144.6, 137.5, 135.3, 133.3,
(0.339 mmol, 63%) of 11b as a bright yellow solid, mp 143e146 ꢃC;
129.3, 128.3 (2C), 128.2, 127.1, 126.7, 126.0 (2C), 124.7, 117.4, 116.2,
49.0, 22.5; IR (neat, cmꢀ1): 3315, 2974, 1660, 1594, 1580, 1516, 1119,
760, 702; HRMS (APCI/ASAP, m/z): 375.1281 (calcd. C21H19N4OS,
375.1280, [MþH]þ).
20
HPLC purity: 99%, tR ¼ 27.9 min; [
a
]
¼ ꢀ508.7 (c 1.01, DMSO); 1H
D
NMR (400 MHz, DMSO-d6)
d
: 10.02 (d, J ¼ 1.5, 1H), 8.52 (d, J ¼ 8.0,
1H), 8.50 (s, 1H), 8.33 (s, 1H), 8.03e7.95 (m, 1H), 7.93e7.86 (m, 2H),
7.46e7.41 (m, 2H), 7.36e7.29 (m, 2H), 7.26e7.19 (m, 1H), 5.59e5.48
(m, 1H), 1.58 (d, J ¼ 7.0, 3H); 13C NMR (100 MHz, DMSO-d6)
d: 191.4
4.7.14. (R)-6-(3-(Methylsulfonyl)phenyl)-N-(1-phenylethyl)thieno
[2,3-d]pyrimidin-4-amine (14b)
(d, J ¼ 1.3), 166.2 (d, J ¼ 4.6), 158.6 (d, J ¼ 251.9), 156.1, 154.8, 144.4,
137.0 (d, J ¼ 6.7),129.7 (d, J ¼ 4.2),129.4 (d, J ¼ 2.9),128.3 (2C),126.7,
126.5 (d, J ¼ 12.5), 126.2 (d, J ¼ 2.8), 126.1 (2C), 121.3 (d, J ¼ 7.6),
116.9 (d, J ¼ 22.9), 116.8, 49.1, 22.4; 19F NMR (564 MHz, DMSO-d6,
C6F6)
d
: ꢀ114.74 (s, dec.); IR (neat, cmꢀ1): 3347, 2926, 1685, 1577,
1512, 1495, 1261, 1103, 776, 699; HRMS (APCI/ASAP, m/z): 378.1076
(calcd. C21H17FN3OS, 378.1076, [MþH]þ).
4.7.12. (R)-3-(4-((1-Phenylethyl)amino)thieno[2,3-d]pyrimidin-6-
yl)benzoic acid (12b)
Compound 14b was prepared as described in Section 4.7,
starting with IIb$HCl (250 mg, 0.674 mmol) and (3-(methyl-
sulfonyl)phenyl)boronic acid (162 mg, 0.809 mmol). The crude
product was purified by silica-gel column chromatography (EtOAc/
n-pentane, 1/1), Rf ¼ 0.22. This gave 195 mg (0.476 mmol, 71%) of
14b as
a
white solid, mp 182e184 ꢃC; HPLC purity: 99%,
20
tR ¼ 23.8 min; [
a
]
D
¼ ꢀ378.2 (c 0.98, DMSO); 1H NMR (400 MHz,
DMSO-d6)
d: 8.43e8.37 (m, 2H), 8.31 (s, 1H), 8.25e8.23 (m, 1H),
Compound 12b was prepared as described in Section 4.7,
starting with IIb$HCl (300 mg, 0.809 mmol) and 3-boronobenzoic
acid (161 mg, 0.971 mmol). Saturated aq NH4Cl solution (25 mL)
was used in the workup. The crude product was purified using
silica-gel column chromatography (Et2O/THF, 6/4), Rf ¼ 0.26, fol-
lowed by recrystallization from 1,4-dioxane (3 mL). This gave 61 mg
(0.162 mmol, 20%) of 12b as a white solid, mp 244e246 ꢃC (dec.);
8.00e7.93 (m, 2H), 7.81e7.75 (m, 1H), 7.46e7.41 (m, 2H), 7.36e7.30
(m, 2H), 7.26e7.20 (m, 1H), 5.57e5.48 (m, 1H), 3.32 (s, 3H), 1.58 (d,
J ¼ 7.0, 3H); 13C NMR (100 MHz, DMSO-d6)
d: 165.6, 156.0, 154.4,
144.5, 142.0, 135.9, 134.3, 130.8, 130.5, 128.3 (2C), 126.7, 126.7, 126.0
(2C), 123.3, 117.4, 117.3, 49.1, 43.5, 22.5; IR (neat, cmꢀ1): 3423, 3403,
2982, 1570, 1513, 1281, 1143, 770, 700; HRMS (APCI/ASAP, m/z):
410.0998 (calcd. C21H20N3O2S2, 410.0997, [MþH]þ).

S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
267
Compound 18b was prepared as described in Section 4.7,
starting with IIb$HCl (301 mg, 0.812 mmol) and (2,3-
dimethoxyphenyl)boronic acid (177 mg, 0.974 mmol). The
crude product was purified by silica-gel column chromatography
(Et2O/n-pentane, 3/1), Rf ¼ 0.21. This gave 261 mg (0.666 mmol,
4.7.15. (R)-2-Fluoro-5-(4-((1-phenylethyl)amino)thieno[2,3-d]
pyrimidin-6-yl)benzaldehyde (16b)
82%) of 18b as a white solid, mp 157e159 ꢃC; HPLC purity: 98%,
20
tR
¼
28.6 min;
[
a]
¼
ꢀ360.1 (c 1.04, DMSO); 1H NMR
D
(400 MHz, DMSO-d6)
d
: 8.28 (s br, 2H), 8.23 (d, J ¼ 7.9, 1H),
7.45e7.41 (m, 2H), 7.41e7.37 (m, 1H), 7.35e7.30 (m, 2H),
7.25e7.19 (m, 2H), 7.11e7.07 (m, 1H), 5.57e5.48 (m, 1H), 3.87 (s,
3H), 3.80 (s, 3H), 1.57 (d, J ¼ 7.0, 3H); 13C NMR (100 MHz,
Compound 16b was prepared as described in Section 4.7,
starting with IIb$HCl (400 mg, 1.079 mmol) and (4-fluoro-3-
formylphenyl)boronic acid (217 mg, 1.295 mmol). The crude prod-
uct was purified by silica-gel column chromatography (Et2O/n-
pentane, 6/4), Rf ¼ 0.18. This gave 293 mg (0.777 mmol, 72%) of 16b
DMSO-d6) d: 166.2, 155.6, 153.8, 153.2, 145.1, 144.7, 133.1, 128.3
(2C), 126.69, 126.66, 126.0 (2C), 124.6, 118.8, 117.3, 115.9, 112.7,
60.1, 55.9, 48.9, 22.5; IR (neat, cmꢀ1): 3031, 2971, 1578, 1506,
1472, 1308, 1266, 1085, 773, 698; HRMS (APCI/ASAP, m/z):
392.1429 (calcd. C22H22N3O2S, 392.1433, [MþH]þ).
as an orange solid, mp 143e145 ꢃC; HPLC purity: 97%, tR ¼ 26.4 min;
20
[
a]
¼ ꢀ360.9 (c 1.00, DMSO); 1H NMR (400 MHz, DMSO-d6)
d:
D
10.29 (s, 1H), 8.32e8.27 (m, 3H), 8.14e8.10 (m, 1H), 8.02e7.97 (m,
1H), 7.59e7.53 (m, 1H), 7.46e7.41 (m, 2H), 7.36e7.30 (m, 2H),
7.25e7.20 (m, 1H), 5.56e5.47 (m, 1H), 1.57 (d, J ¼ 7.0, 3H); 13C NMR
4.7.18. (R)-6-(4-Fluoro-2-methoxyphenyl)-N-(1-phenylethyl)
thieno[2,3-d]pyrimidin-4-amine (19b)
(100 MHz, DMSO-d6)
d: 187.5 (d, J ¼ 5.0), 165.4, 163.1 (d, J ¼ 259.3),
155.9, 154.2, 144.5, 135.5, 133.4 (d, J ¼ 9.4), 130.3 (d, J ¼ 3.6), 128.3
(2C), 126.7, 126.0 (2C), 125.3 (d, J ¼ 2.1), 124.3 (d, J ¼ 9.5), 118.1 (d,
J ¼ 21.3), 117.4, 116.8, 49.1, 22.5; 19F NMR (564 MHz, DMSO-d6, C6F6)
d
: ꢀ123.54 (s, dec.); IR (neat, cmꢀ1): 3440, 3382, 2925, 1701, 1580,
1517, 1486, 1353, 1303, 775, 762, 699; HRMS (APCI/ASAP, m/z):
378.1071 (calcd. C21H17FN3OS, 378.1076, [MþH]þ).
Compound 19b was prepared as described in Section 4.7,
starting with compound IIb$HCl (302 mg, 0.815 mmol) and (4-
fluor-2-metoxyphenyl)boronic acid (166 mg, 0.978 mmol). The
mixture was stirred at 60 ꢃC for 4 h. The crude product was purified
by silica-gel column chromatography (n-pentane/EtOAc, 13/7),
Rf ¼ 0.25. This gave 239 mg (0.630 mmol, 77%) of 19b as an off-
4.7.16. (R)-2-(4-((1-Phenylethyl)amino)thieno[2,3-d]pyrimidin-6-
yl)benzamide (17b)
white solid, mp. 142e148 ꢃC; HPLC purity: 95%, tR ¼ 29.6 min;
20
[a
]
¼ ꢀ292.3 (c 0.65, DMSO); 1H NMR (400 MHz, DMSO-d6)
d:
D
8.27 (s, 1H), 8.20 (d, J ¼ 8.0, 1H), 8.17 (s, 1H), 7.77e7.72 (m, 1H),
7.44e7.42 (m, 2H), 7.34e7.30 (m, 2H), 7.24e7.20 (m, 1H), 7.15e7.12
(m, 1H), 6.99e6.94 (m, 1H), 5.57e5.50 (m, 1H), 3.96 (s, 3H), 1.57 (d,
J ¼ 7.0, 3H); 13C NMR (100 MHz, DMSO-d6)
d: 165.4, 162.9 (d,
Compound 17b was prepared as described in Section 4.7, start-
ing with IIb$HCl (184 mg, 0.937 mmol) and (2-carbamoylphenyl)
boronic acid (131 mg, 1.124 mmol, 1.6 eq), and using 5 mol%
Pd(PPh3)4. The reaction time was 22 h. The crude product was
purified using silica-gel column chromatography (EtOAc/n-
pentane, 7/3), Rf ¼ 0.18. This gave 138 mg (0.367 mmol, 74%) of 17b
J ¼ 246.1), 156.9 (d, J ¼ 10.2), 155.6, 153.2, 144.7, 133.2, 129.4 (d,
J ¼ 10.4), 128.3 (2C), 126.7, 126.1 (2C), 118.5 (d, J ¼ 3.5), 117.0, 116.1,
107.8 (d, J ¼ 21.9), 100.7 (d, J ¼ 26.1), 56.4, 48.9, 22.5; 19F NMR
(564 MHz, DMSO-d6, C6F6): ꢀ112.56 (s, dec.); IR (neat, cmꢀ1): 3236,
2966, 1577, 1491, 1282, 1108, 1029, 962, 832, 776, 698, 554; HRMS
(APCI/ASAP, m/z): found 380.1229, (calcd. for
C21H19FN3OS,
as a white solid, mp. 152e156 ꢃC; HPLC purity: 99%, tR ¼ 19.4 min;
380.1233, [MþH]þ).
20
[
a]
¼ ꢀ265.6 (c 1.04, DMSO); 1H NMR (400 MHz, DMSO-d6)
d:
D
8.32 (d, J ¼ 7.9, 1H), 8.28 (s, 1H), 7.84 (s, 1H), 7.83 (s br, 1H),
7.66e7.46 (m, 5H), 7.44e7.40 (m, 2H), 7.35e7.29 (m, 2H), 7.25e7.19
(m, 1H), 5.59e5.49 (m, 1H), 1.56 (d, J ¼ 7.0, 3H); 13C NMR (100 MHz,
4.7.19. (R)-6-(2,4-Dimethoxyphenyl)-N-(1-phenylethyl)thieno[2,3-
d]pyrimidin-4-amine (20b)
DMSO-d6) d: 170.4, 166.0, 155.8, 153.6, 144.7, 137.6, 136.8, 131.5,
131.1, 130.2, 129.4, 128.3 (2C), 127.8, 126.6, 126.1 (2C), 118.5, 116.7,
48.9, 22.4; IR (neat, cmꢀ1): 3329, 3153, 2926, 1662, 1578, 1516, 1349,
1307, 1119, 756, 697; HRMS (APCI/ASAP, m/z): 375.1276 (calcd.
C
21H19N4OS, 375.1280, [MþH]þ).
4.7.17. (R)-6-(2,3-Dimethoxyphenyl)-N-(1-phenylethyl)thieno[2,3-
d]pyrimidin-4-amine (18b)
Compound 20b was prepared as described in Section 4.7,
starting with compound IIb$HCl (200 mg, 0.540 mmol) and (2,4-
dimetoksyphenyl)boronic acid (118 mg, 0.647 mmol). The
mixture was stirred at 60 ꢃC for 9 h. The crude product was purified
by silica-gel column chromatography (CH2Cl2/EtOAc, 4/1), Rf ¼ 0.24,
followed by recrystallization from CH2Cl2 using n-pentane as an
antisolvent. This gave 147 mg (0.375 mmol, 70%) of 20b as a white
solid, mp. 158e159 ꢃC; HPLC purity: 98%, tR
¼
28.9 min;

268
20
S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
[
a]
¼ ꢀ334.8 (c 0.55, DMSO); 1H NMR (400 MHz, DMSO-d6)
d:
4.7.22. (R)-N-(2-(Dimethylamino)ethyl)-4-(4-((1-phenylethyl)
amino)thieno[2,3-d]pyrimidin-6-yl)benzamide (22b)
D
8.24 (s, 1H), 8.14 (d, J ¼ 8.0, 1H), 8.09 (s, 1H), 7.66e7.64 (m, 1H),
7.44e7.42 (m, 2H), 7.34e7.30 (m, 2H), 7.23e7.20 (m, 1H), 6.74e6.73
(m, 1H), 6.71e6.69 (m, 1H), 5.46e5.49 (m, 1H), 3.94 (s, 3H), 3.84 (s,
3H), 1.57 (d, J ¼ 7.0, 3H); 13C NMR (100 MHz, DMSO-d6)
d: 165.0,
160.8, 156.7, 155.4, 153.2, 144.8, 134.4, 128.9, 128.3 (2C), 126.6, 126.1
(2C), 116.2, 115.3, 114.7, 106.3, 99.1, 55.7, 55.3, 48.9, 22.5; IR (neat,
cmꢀ1): 3224, 2934, 1588, 1493, 1308, 1204, 1026, 822, 777, 697, 554;
HRMS (APCI/ASAP, m/z): found: 392.1429, (calcd. for C22H22N3O2S,
392.1433, [MþH]þ).
Compound 22b was prepared as described in Section 4.7,
starting with IIb (200 mg, 0.598 mmol) and (4-((2-(dimethyla-
mino)ethyl)carbamoyl)phenyl)boronic acid hydrochloride (196 mg,
0.718 mmol). The crude product was purified using crystallization
from CH3CN (6 mL). This gave 136 mg (0.305 mmol, 51%) of 22b as a
4.7.20. (R)-6-(2,6-Dimethoxyphenyl)-N-(1-phenylethyl)thieno[2,3-
d]pyrimidin-4-amine (21b)
light yellow solid, mp. 178e181 ꢃC; HPLC purity: 98%, tR ¼ 17.9 min;
20
[
a
]
¼ ꢀ353.7 (c 0.49, DMSO); 1H NMR (400 MHz, DMSO-d6)
d:
D
8.48 (t, J ¼ 5.6, 1H), 8.33e8.26 (m, 3H), 8.00e7.94 (m, 2H),
7.80e7.49 (m, 2H), 7.46e7.41 (m, 2H), 7.36e7.30 (m, 2H), 7.26e7.20
(m, 1H), 5.57e5.47 (m, 1H), 3.41e3.34 (m, 2H), 2.41 (t, J ¼ 6.9, 2H),
2.19 (s, 6H), 1.57 (d, J ¼ 7.0, 3H); 13C NMR (100 MHz, DMSO-d6)
d:
165.5, 165.3, 155.9, 154.2, 144.5, 137.0, 135.6, 134.1, 128.3 (2C), 128.2
(2C), 126.7, 126.0 (2C), 125.3 (2C), 117.4, 116.8, 58.2, 49.1, 45.3, 37.4,
22.4; IR (neat, cmꢀ1): 3313, 2928, 2768,1638,1579,1540,1520,1497,
1307, 1119, 844, 760, 696; HRMS (APCI/ASAP, m/z): 446.2012 (calcd.
Compound 21b was prepared as described in Section 4.7,
starting with IIb$HCl (297 mg, 0.801 mmol) and (2,6-
dimethoxyphenyl)boronic acid (175 mg, 0.961 mmol). The crude
product was purified by silica-gel column chromatography (Et2O/n-
pentane, 7/3), Rf ¼ 0.22. This gave 235 mg (0.601 mmol, 75%) of 21b
C
25H28N5OS, 446.2015, [MþH]þ).
4.7.23. (S)-N-(2-(Dimethylamino)ethyl)-4-(4-((2-hydroxy-1-
as a white solid, mp 169e171 ꢃC; HPLC purity: 98%, tR ¼ 26.8 min;
phenylethyl)amino)thieno[2,3-d]pyrimidin-6-yl)benzamide (22c)
20
[
a]
¼ ꢀ274.8 (c 1.02, DMSO); 1H NMR (400 MHz, DMSO-d6)
d: 8.27
D
(s, 1H), 8.17 (d, J ¼ 8.0, 1H), 7.86 (s, 1H), 7.45e7.41 (m, 2H), 7.40e7.35
(m, 1H), 7.34e7.29 (m, 2H), 7.24e7.19 (m, 1H), 6.82e6.78 (m, 2H),
5.60e5.50 (m, 1H), 3.80 (s, 6H), 1.56 (d, J ¼ 7.1, 3H); 13C NMR
(100 MHz, DMSO-d6) d: 165.7, 157.6 (2C), 155.5, 153.3, 144.8, 130.4,
129.8, 128.3 (2C), 126.6, 126.1 (2C), 120.4, 116.0, 110.5, 104.5 (2C),
56.0 (2C), 48.9, 22.4; IR (neat, cmꢀ1): 3217, 2931, 1576, 1472, 1252,
1098, 723, 699; HRMS (APCI/ASAP, m/z): 392.1432 (calcd.
Compound 22c was prepared as described in Section 4.7, start-
ing with IIc (257 mg, 0.732 mmol) and (4-((2-(dimethylamino)
ethyl)carbamoyl)phenyl)boronic acid hydrochloride (299 mg,
1.095 mmol). The crude product was purified using crystallization
from CH3CN (20 mL). This gave 41 mg (0.088 mmol, 12%) of 22c as a
C
22H22N3O2S, 392.1433, [MþH]þ).
4.7.21. (S)-2-((6-(2,6-Dimethoxyphenyl)thieno[2,3-d]pyrimidin-4-
yl)amino)-2-phenylethanol (21c)
light brown solid, mp. 190 ꢃC (dec.); HPLC purity: 95%,
20
tR ¼ 15.6 min; [
a
]
D
¼ ꢀ278.3 (c 0.53, DMSO); 1H NMR (400 MHz,
DMSO-d6)
d
: 8.48 (t, J ¼ 5.7, 1H), 8.38 (s, 1H), 8.29 (s, 1H), 8.27 (d,
J ¼ 8.1, 1H), 8.00e7.95 (m, 2H), 7.81e7.77 (m, 2H), 7.47e7.42 (m,
2H), 7.36e7.30 (m, 2H), 7.27e7.21 (m, 1H), 5.50e5.42 (m, 1H), 5.05
(t, J ¼ 5.7, 1H), 3.82e3.72 (m, 2H), 3.40e3.36 (m, 2H), 2.42 (t, J ¼ 6.8,
2H), 2.19 (s, 6H); 13C NMR (100 MHz, DMSO-d6)
d: 165.5, 165.3,
156.6, 154.2, 141.0, 136.9, 135.6, 134.1, 128.24 (2C), 128.19 (2C), 127.0
(2C),126.9,125.3 (2C), 117.6, 116.9, 64.7, 58.2, 56.4, 45.3 (2C), 37.5; IR
(neat, cmꢀ1): 3344, 2944, 2873, 1645, 1584, 1516, 1320, 1309, 1063,
838, 762, 701; HRMS (APCI/ASAP, m/z): 462.1961 (calcd.
Compound 21c was prepared as described in Section 4.7, start-
ing with IIc (301 mg, 0.859 mmol) and (2,6-dimethoxyphenyl)
boronic acid (188 mg, 1.031 mmol). The crude product was purified
by silica-gel column chromatography (Et2O/EtOAc, 7/3), Rf ¼ 0.22.
C
25H28N5O2S, 462.1964, [MþH]þ).
This gave 292 mg (0.717 mmol, 83%) of 21c as a white solid, mp
4.7.24. (R)-N-(2-(Dimethylamino)ethyl)-3-(4-((1-phenylethyl)
amino)thieno[2,3-d]pyrimidin-6-yl)benzamide (23b)
20
155e157 ꢃC; HPLC purity: 98%, tR ¼ 22.7 min; [
a]
¼ ꢀ227.7 (c 0.56,
D
DMSO); 1H NMR (400 MHz, DMSO-d6)
d
: 8.25 (s, 1H), 8.11 (d, J ¼ 8.1,
1H), 7.87 (s,1H), 7.45e7.41 (m, 2H), 7.41e7.36 (m,1H), 7.34e7.28 (m,
2H), 7.25e7.20 (m, 1H), 6.83e6.79 (m, 2H), 5.51e5.43 (m, 1H), 5.00
(t, J ¼ 5.7, 1H), 3.80 (s, 6H), 3.79e3.69 (m, 2H); 13C NMR (100 MHz,
DMSO-d6) d: 165.7, 157.7 (2C), 156.1, 153.2, 141.3, 130.4, 129.7, 128.1
(2C), 127.0 (2C), 126.8, 120.5, 116.1, 110.5, 104.5 (2C), 64.6, 56.4, 56.0
(2C); IR (neat, cmꢀ1): 3344, 1582, 1474, 1349, 1257, 1101, 1026, 781,
718, 699; HRMS (APCI/ASAP, m/z): 408.1382 (calcd. C22H22N3O3S,
408.1382, [MþH]þ).
Compound 23b was prepared as described in Section 4.7,

S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
269
starting with IIb (214 mg, 0.640 mmol) and N-(2-(dimethylamino)
ethyl)-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzamide
(299 mg, 0.940 mmol). The pH was adjusted to 13 using 5 M aq
NaOH before the mixture was extracted with EtOAc (6 ꢂ 50 mL).
The crude product was purified by silica-gel column chromatog-
1H), 8.27 (d, J ¼ 8.1, 1H), 8.15 (s br, 1H), 7.93e7.88 (m, 1H), 7.73e7.71
(m, 1H), 7.71e7.67 (m, 1H), 7.55e7.48 (m, 3H), 7.41e7.35 (m, 2H),
7.31e7.26 (m, 1H), 5.57e5.49 (m, 1H), 5.09 (t, J ¼ 5.7, 1H), 4.08 (s,
3H), 3.88e3.77 (m, 2H); 13C NMR (100 MHz, DMSO-d6)
d: 167.0,
165.9, 156.4, 155.1, 153.8, 141.1, 135.0, 132.9, 128.2 (2C), 127.4, 127.0
(2C), 126.9, 124.4, 120.3, 118.1, 116.2, 111.5, 64.6, 56.3, 56.1; IR (neat,
cmꢀ1): 3296, 3184, 1661, 1581, 1510, 1411, 1355, 1257, 1030, 761, 700;
HRMS (APCI/ASAP, m/z): 421.1335 (calcd. C22H21N4O3S, 421.1334,
[MþH]þ).
raphy (CH2Cl2/MeOH, 5/1), Rf
¼
0.16. This gave 265 mg
(0.595 mmol, 93%) of 23b as a as a white crystalline powder, mp.
20
122e124 ꢃC; HPLC purity: 99%, tR ¼ 16.1 min; [
a
]
¼ ꢀ318.7 (c 0.98,
D
DMSO); 1H NMR (400 MHz, DMSO-d6)
d
: 8.62 (t, J ¼ 5.6,1H), 8.33 (d,
J ¼ 7.3, 1H), 8.32 (s, 1H), 8.30 (s, 1H), 8.21e8.19 (m, 1H), 7.86e7.83
(m, 1H), 7.83e7 0.80 (m, 1H), 7.59 (t, J ¼ 7.7, 1H), 7.45e7.43 (m, 2H),
7.35e7.31 (m, 2H), 7.24e7.21 (m, 1H), 5.55e5.49 (m, 1H), 3.42 (q,
4.7.27. (S)-2-Phenyl-2-((6-phenylthieno[2,3-d]pyrimidin-4-yl)
amino)ethan-1-ol (25c)
J ¼ 6.4, 2H), 2.51e2.49 (m, 2H), 2.25 (s, 6H), 1.57 (d, J ¼ 7.0, 3H); 13
C
NMR (100 MHz, DMSO-d6) d: 165.6, 165.3, 155.9, 154.1, 144.6, 137.3,
135.5, 133.3, 129.4, 128.3 (2C), 128.1, 126.9, 126.7, 126.1 (2C), 124.4,
117.4, 116.3, 58.0, 49.1, 45.0 (2C), 37.3, 22.5; IR (neat, cmꢀ1): 3291,
2974, 2940, 2820, 2773, 1579, 1514, 1304, 750, 698; HRMS (APCI/
ASAP, m/z): 446.2010 (calcd. C25H28N5OS, 446.2015, [MþH]þ).
4.7.25. (R)-3-Methoxy-4-(4-((1-phenylethyl)amino)thieno[2,3-d]
pyrimidin-6-yl)benzamide (24b)
Compound 25c was prepared as described in Section 4.7 using
compound IIc (305 mg, 0.87 mmol) and phenylboronic acid
(159 mg, 1.31 mmol). The crude product was purified using silica-
gel column chromatography (EtOAc/n-pentane, 3/2), Rf ¼ 0.21.
This gave 254 mg (0.730 mmol, 84%) of 25c as a white crystalline
solid, mp. 220e223 ꢃC; HPLC purity: 99%, tR
¼
21.5 min;
: 8.28
20
[a
]
¼ ꢀ338.8 (c 1.0, DMSO); 1H NMR (400 MHz, DMSO-d6)
d
D
(s,1H), 8.26 (s,1H), 8.20 (d, J ¼ 8.1,1H), 7.73e7.70 (m, 2H), 7.54e7.49
(m, 2H), 7.46e7.43 (m, 2H), 7.43e7.39 (m, 1H), 7.35e7.31 (m, 2H),
7.26e7.22 (m, 1H), 5.49e5.43 (m, 1H), 5.04 (t, J ¼ 5.7, 1H), 3.82e3.73
Compound 24b was prepared as described in Section 4.7 using
compound IIb$HCl (301 mg, 0.812 mmol) and (4-carbamoyl-2-
methoxyphenyl)boronic acid (190 mg, 0.974 mmol). The crude
product was purified using silica-gel column chromatography
(THF/n-pentane, 7/3), Rf ¼ 0.24, followed by recrystallization from
acetonitrile (6 mL). This gave 207 mg (0.512 mmol, 63%) of 24b as a
(m, 2H); 13C NMR (100 MHz, DMSO-d6)
d: 165.1, 156.5, 153.9, 141.0,
138.0, 133.2, 129.4 (2C), 128.5, 128.2 (2C), 127.0 (2C), 126.9, 125.6
(2C), 117.6, 115.7, 64.7, 56.4; IR (neat, cmꢀ1): 3424, 3211, 1578, 1492,
1348, 1076, 1043, 752, 702; HRMS (APCI/ASAP, m/z): 348.1169
(calcd. C20H18N3OS, 348.1171, [MþH]þ).
yellow solid, mp. 159e162 ꢃC; HPLC purity: 99%, tR ¼ 20.5 min;
20
[
a]
¼ ꢀ425.8 (c 1.07, DMSO); 1H NMR (400 MHz, DMSO-d6)
d:
4.7.28. (R)-2-Methoxy-3-(4-((1-phenylethyl)amino)thieno[2,3-d]
pyrimidin-6-yl)benzaldehyde (26b)
D
8.36 (s, 1H), 8.29 (s, 1H), 8.26 (d, J ¼ 8.0, 1H), 8.10 (s br, 1H),
7.85e7.81 (m, 1H), 7.68e7.65 (m, 1H), 7.65e7.61 (m, 1H), 7.48 (s br,
1H), 7.46e7.41 (m, 2H), 7.36e7.30 (m, 2H), 7.25e7.20 (m, 1H),
5.59e5.49 (m, 1H), 4.02 (s, 3H), 1.58 (d, J ¼ 7.0, 3H); 13C NMR
(100 MHz, DMSO-d6) d: 167.0, 165.9, 155.7, 155.1, 153.9, 144.6, 135.0,
132.9, 128.3 (2C), 127.4, 126.7, 126.1 (2C), 124.4, 120.3, 118.0, 116.1,
111.5, 56.1, 49.0, 22.5; IR (neat, cmꢀ1): 3313, 3165, 1669, 1575, 1558,
1379, 1257, 1103, 1033, 776, 691; HRMS (APCI/ASAP, m/z): 405.1379
(calcd. C22H21N4O2S, 405.1385, [MþH]þ).
4.7.26. (S)-4-(4-((2-Hydroxy-1-phenylethyl)amino)thieno[2,3-d]
pyrimidin-6-yl)-3-methoxybenzamide (24c)
Compound 26b was prepared as described in Section 4.7,
starting with compound IIb$HCl (400 mg, 1.079 mmol) and (3-
formyl-2-metoksyphenyl)boronic acid (233 mg, 1.295 mmol). The
crude product was purified by silica-gel column chromatography
(CH2Cl2/EtOAc, 4/1), Rf ¼ 0.27, followed by crystallisation from
Et2O/EtOAc (1/1) using n-pentane as anti-solvent. This gave 328 mg
(0.842 mmol, 78%) of 26b as a slight yellowish solid; mp. 71e90 ꢃC;
20
HPLC purity: 98%, tR ¼ 28.0 min; [
a
]
¼ ꢀ309.6 (c 0.60, DMSO); 1H
D
NMR (400 MHz, DMSO-d6)
d: 10.37 (s, 1H), 8.35 (s, 1H), 8.32 (d,
J ¼ 7.9, 1H), 8.32 (s, 1H), 8.09e8.07 (m, 1H), 7.83e7.81 (m, 1H),
7.50e7.46 (m, 1H), 7.45e7.43 (m, 2H), 7.35e7.31 (m, 2H), 7.24e7.21
(m, 1H), 5.58e5.51 (m, 1H), 3.85 (s, 3H), 1.58 (d, J ¼ 7.0, 3H); 13C
Compound 24c was prepared as described in Section 4.7, using
compound IIc (200 mg, 0.571 mmol) and (4-carbamoyl-2-
methoxyphenyl)boronic acid (134 mg, 0.685 mmol). The crude
product was purified using crystallization from CH3CN (4 mL). This
NMR (100 MHz, DMSO-d6) d: 189.8, 166.4, 159.0, 155.8, 154.2, 144.6,
131.7, 134.5, 129.9, 128.9, 128.3, 128.1 (2C), 126.7, 126.1 (2C), 125.3,
118.8, 116.2, 64.1, 49.0, 22.4; IR (neat, cmꢀ1): 2967, 1688, 1577, 1508,
1351, 1305, 1245, 1105, 989, 777, 698; HRMS (APCI/ASAP, m/z):
found 390.1272, (calcd. for C22H20N3O2S, 390.1276, [MþH]þ).
gave 168 mg (0.400 mmol, 70%) of 24c as a white solid, mp.
20
197e198 ꢃC (dec.); HPLC purity: 96%, tR ¼ 16.8 min; [
a
]
¼ ꢀ336.4
D
(c 1.00, DMSO); 1H NMR (400 MHz, DMSO-d6)
d: 8.46 (s,1H), 8.33 (s,

270
S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
Compound 32b was prepared as described in Section 4.7,
starting with compound IIb$HCl (1.027 g, 2.770 mmol) and (5-
formyl-2-metoksyphenyl)boronic acid (598 mg, 3.325 mmol).
The mixture was stirred at 60 ꢃC for 29 h. The crude product
was crystallised from MeOH (24 mL), giving 771 mg
4.7.29. Methyl (R)-2-methoxy-3-(4-((1-phenylethyl)amino)thieno
[2,3-d]pyrimidin-6-yl)benzoate (27b)
(1.980 mmol, 72%) of 32b as
a slight yellowish solid, mp.
20
173e174 ꢃC; HPLC purity: 97%, tR ¼ 27.5 min; [
a]
¼ ꢀ376.4 (c
D
0.96, DMSO); 1H NMR (400 MHz, DMSO-d6)
d: 9.99 (s, 1H), 8.39
(s, 1H), 8.35 (d, J ¼ 8.0, 1H), 8.31 (d, J ¼ 2.0, 1H), 8.29 (s, 1H),
7.98 (dd, J ¼ 8.5, 2.0, 1H), 7.45e7.43 (m, 2H), 7.45e7.42 (m, 1H),
7.35e7.31 (m, 2H), 7.24e7.20 (m, 1H), 5.58e5.50 (m, 1H), 4.08 (s,
Compound 27b was prepared as described in Section 4.7 using
compound IIb (655 mg, 1.960 mmol) and methyl 2-methoxy-3-
(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzoate (955 mg,
3.27 mmol). The crude product was purified using silica-gel column
chromatography (EtOAc/n-pentane, 3/2), Rf ¼ 0.24, followed by
trituration with n-pentane (50 mL). This gave 745 mg (1.776 mmol,
3H), 1.58 (d, J ¼ 7.0, 3H); 13C NMR (100 MHz, DMSO-d6)
d: 191.4,
166.0, 159.9, 155.7, 154.0, 144.7, 132.4, 132.4, 129.8, 128.3 (2C),
128.0, 126.7, 126.1 (2C), 122.6, 117.9, 115.9, 112.8, 56.5, 49.0, 22.5;
IR (neat, cmꢀ1): 3125, 1688, 1592, 1266, 1111, 1016, 820, 767, 703,
640, 553; HRMS (APCI/ASAP, m/z) found 390.1272, (calcd. for
C
22H20N3O2S, 390.1276, [MþH]þ).
91%) of 27b as an off-white crystalline solid, mp. 121e124 ꢃC; HPLC
20
purity: 98%, tR ¼ 25.2 min; [
a]
¼ ꢀ333.0 (c 1.03, DMSO); 1H NMR
D
4.7.32. Methyl (R)-4-methoxy-3-(4-((1-phenylethyl)amino)thieno
[2,3-d]pyrimidin-6-yl)benzoate (34b)
(600 MHz, DMSO-d6)
d
: 8.33 (s, 1H), 8.31 (s, 1H), 8.29 (d, J ¼ 7.9, 1H),
7.99e7.96 (m, 1H), 7.75e7.73 (m, 1H), 7.45e7.43 (m, 2H), 7.39 (t,
J ¼ 7.8,1H), 7.35e7.31 (m, 2H), 7.24e7.21 (m,1H), 5.57e5.50 (m,1H),
3.89 (s, 3H), 3.74 (s, 3H), 1.57 (d, J ¼ 7.0, 3H); 13C NMR (150 MHz,
DMSO-d6) d: 166.3, 165.9, 155.8, 155.4, 154.1, 144.6, 132.3, 132.0,
130.8, 128.3 (2C), 127.9, 126.7, 126.1 (2C), 125.9, 124.6, 118.5, 116.1,
62.1, 52.5, 49.0, 22.5; IR (neat, cmꢀ1): 3374, 3251, 2977, 2940, 1728,
1577, 1506, 1291, 1140, 994, 761, 699; HRMS (APCI/ASAP, m/z):
420.1375 (calcd. C23H22N3O3S, 420.1382, [MþH]þ).
Compound 34b was prepared as described in Section 4.7,
starting with compound IIb (123 mg, 0.368 mmol) and 4-
methoxy-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzoate
(179 mg, 0.613 mmol). The crude product was purified using
silica-gel column chromatography (EtOAc/n-pentane, 3/2),
Rf ¼ 0.21. This gave 124 mg (0.296 mmol, 80%) of 34b as a
4.7.30. Methyl (R)-3-methoxy-4-(4-((1-phenylethyl)amino)thieno
[2,3-d]pyrimidin-6-yl)benzoate (30b)
crystalline white solid, mp. 152e153 ꢃC; HPLC purity: 99%,
20
tR
¼
25.8 min;
[
a]
¼
ꢀ379.6 (c 1.01, DMSO); 1H NMR
D
(600 MHz, DMSO-d6)
d: 8.40e8.38 (m, 1H), 8.38e8.34 (m, 1H),
8.36 (s, 1H), 8.29 (s, 1H), 8.01e7.98 (m, 1H), 7.46e7.42 (m, 2H),
7.36e7.30 (m, 3H), 7.25e7.20 (m, 1H), 5.59e5.50 (m, 1H), 4.05 (s,
3H), 3.89 (s, 3H), 1.59 (d, J ¼ 7.0, 3H); 13C NMR (150 MHz,
Compound 30b was prepared as described in Section 4.7 using
compound IIb (231 mg, 0.691 mmol) and methyl 3-methoxy-4-
(4,4,5,5- tetramethyl-1,3,2-dioxaborolan-2-yl)benzoate (300 mg,
1.027 mmol). The crude product was purified using silica-gel col-
umn chromatography (EtOAc/n-pentane, 3/2), Rf ¼ 0.17. This gave
DMSO-d6) d: 165.9, 165.7, 158.8, 155.7, 153.9, 144.7, 132.5, 130.9,
128.6, 128.3 (2C), 126.7, 126.1 (2C), 122.4, 122.0, 117.5, 115.9,
112.6, 56.4, 52.1, 49.0, 22.5; IR (neat, cmꢀ1): 3358, 3271, 2970,
2944, 1716, 1578, 1511, 1267, 764, 699; HRMS (APCI/ASAP, m/z)
found 420.1378, (calcd. for C23H22N3O3S, 420.1382, [MþH]þ).
258 mg (0.615 mmol, 89%) of 30b as a yellow crystalline solid, mp.
20
126e127 ꢃC; HPLC purity: 98%, tR ¼ 29.0 min; [
a
]
¼ ꢀ434.2 (c
D
1.00, DMSO); 1H NMR (600 MHz, DMSO-d6)
d: 8.42 (s, 1H), 8.32 (d,
4.7.33. Methyl (R)-3,5-dimethoxy-4-(4-((1-phenylethyl)amino)
thieno[2,3-d]pyrimidin-6-yl)benzoate (36b)
J ¼ 7.9, 1H), 8.30 (s, 1H), 7.92e7.89 (m, 1H), 7.72e7.69 (m, 1H),
7.67e7.66 (m, 1H), 7.45e7.42 (m, 2H), 7.35e7.31 (m, 2H), 7.24e7.21
(m, 1H), 5.57e5.51 (m, 1H), 4.03 (s, 3H), 3.89 (s, 3H), 1.58 (d, J ¼ 7.0,
3H); 13C NMR (150 MHz, DMSO-d6)
d: 166.2, 165.7, 155.8, 155.2,
154.1, 144.6, 132.4, 130.1, 128.3 (2C), 127.8, 126.7, 126.5, 126.1 (2C),
122.0,118.9,116.0,112.6, 56.1, 52.3, 49.0, 22.5; IR (neat, cmꢀ1): 3248,
2970, 2930, 1714, 1576, 1294, 1229, 758, 697; HRMS (APCI/ASAP, m/
z): 420.1375 (calcd. C23H22N3O3S, 420.1382, [MþH]þ).
4.7.31. (R)-4-Methoxy-3-(4-((1-phenylethyl)amino)thieno[2,3-d]
pyrimidin-6-yl)benzaldehyde (32b)
Compound 36b was prepared as described in Section 4.7,
starting with compound IIb (380 mg, 1.137 mmol) and methyl 3,5-
dimethoxy-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzo-
ate (550 mg, 1.707 mmol). The crude product was purified using
silica-gel column chromatography (EtOAc/n-pentane, 3/2),
Rf ¼ 0.17. This gave 508 mg (1.130 mmol, 99%) of 36b as a yellow
solid, mp. 205e206 ꢃC; HPLC purity: 95%, tR
¼
25.3 min;
20
[a
]
¼ ꢀ306.6 (c 1.00, DMSO); 1H NMR (600 MHz, DMSO-d6)
d:
D
8.28 (s, 1H), 8.26 (d, J ¼ 8.0, 1H), 8.04 (s, 1H), 7.44e7.41 (m, 2H), 7.34

S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
271
(s, 2H), 7.34e7.30 (m, 2H), 7.23e7.20 (m, 1H), 5.59e5.52 (m, 1H),
3.91 (s, 3H), 3.90 (s, 6H), 1.56 (d, J ¼ 7.0, 3H); 13C NMR (150 MHz,
139.3, 135.2, 128.6 (2C), 127.7, 123.2, 121.3 (2C), 119.3, 118.9, 117.0,
116.3, 110.2, 62.5, 55.8; IR (neat, cmꢀ1): 3341, 3221, 2930, 1618, 1511,
1443, 1268, 1066, 1037, 767, 752, 691; HRMS (APCI/ASAP, m/z):
364.1115 (calcd. C20H18N3O2S, 364.1120, [MþH]þ).
DMSO-d6) d: 166.0, 165.7, 157.4 (2C), 155.6, 153.7, 144.7, 130.8, 128.3
(2C), 128.2, 126.6, 126.1 (2C), 121.6, 115.7, 115.3, 105.0 (2C), 56.3 (2C),
52.5, 48.9, 22.4; IR (neat, cmꢀ1): 3228, 3111, 2974, 2940, 1713, 1575,
1314, 1239, 1124, 998, 852, 762, 741, 697; HRMS (APCI/ASAP, m/z)
found 450.1482, (calcd. for C24H24N3O4S, 450.1482, [MþH]þ).
4.8.3. (4-(4-((3-Ethynylphenyl)amino)thieno[2,3-d]pyrimidin-6-
yl)-3-methoxyphenyl)methanol (3e)
4.8. General procedure for reduction of 4-N-substituted-thieno[2,3-
d]pyrimidinebenzaldehydes
The substrate (100e400 mg) was dissolved in THF/MeOH (3/1,
5e20 mL), and NaBH4 (1 eq.) was added. The reaction mixture was
stirred for 30 min before additional NaBH4 (1 eq.) was added. The
mixture was stirred for another 30 min, and concentrated to about
5 mL. The residue was diluted with EtOAc (50e100 mL) and washed
with water (3 x 25e50 mL). The organic layer was washed with
brine (50 mL), dried over anhydrous Na2SO4, filtered and concen-
trated in vacuo. Purification was performed as specified for each
individual compound.
Compound 3e was prepared as described in Section 4.8, starting
with 4e (100 mg, 0.259 mmol). The crude product was crystallized
from MeOH (18 mL), then from DMF (2 mL) and water (1 mL) and
again from MeOH (8 mL) to give 18 mg (0.046 mmol, 18%) of 3e as a
pale yellow solid, mp. 217e225 ꢃC; HPLC purity: 96%, tR ¼ 23.1 min;
1H NMR (400 MHz, DMSO-d6)
d: 9.68 (s, 1H), 8.53 (s, 1H), 8.33 (s,
4.8.1. (4-(4-(Benzylamino)thieno[2,3-d]pyrimidin-6-yl)-3-
methoxyphenyl)methanol (3a)
1H), 8.12e8.08 (m, 1H), 7.93e7.89 (m, 1H), 7.76e7.72 (m, 1H),
7.44e7.38 (m, 1H), 7.23e7.17 (m, 2H), 7.10e07 (m, 1H), 5.33 (t,
J ¼ 5.6, 1H), 4.57 (d, J ¼ 5.6, 2H), 4.21 (s, 1H), 3.98 (s, 3H); 13C NMR
(100 MHz, DMSO-d6) d: 166.4,155.6,153.9,152.9,145.2,139.7, 135.5,
129.1, 127.7, 126.2, 123.7, 121.9, 121.5, 119.8, 119.0, 117.2, 116.2, 110.2,
83.5, 80.6, 62.6, 55.8; IR (neat, cmꢀ1): 3236, 1539, 1443, 1412, 1355,
1005, 849, 774, 657; HRMS (APCI/ASAP, m/z): 388.1113 (calcd.
C
22H18N3O2S, 388.1120, [MþH]þ).
4.8.4. (4-(4-((3-Chloro-4-fluorophenyl)amino)thieno[2,3-d]
pyrimidin-6-yl)-3-methoxyphenyl)methanol (3f)
Compound 3a was prepared as described in Section 4.8, starting
with 4a (103 mg, 0.274 mmol). The crude product was purified using
silica-gel column chromatography (Et2O/THF, 92/8), Rf ¼ 0.21. This
gave 95 mg (0.252 mmol, 92%) of 3a as a pale yellow solid, mp.
195e197 ꢃC; HPLC purity 98%, tR ¼ 22.6 min; 1H NMR (400 MHz,
DMSO-d6)
d
: 8.47 (t, J ¼ 5.6,1H), 8.32 (s,1H), 8.12 (s,1H), 7.65e7.63 (m,
1H), 7.39e7.31 (m, 4H), 7.27e7.23 (m, 1H), 7.15 (s, 1H), 7.05e7.03 (m,
1H), 5.29(t, J ¼ 5.7,1H), 4.77 (d, J ¼ 5.6, 2H), 4.55(d, J ¼ 5.7, 2H), 3.95(s,
3H); 13C NMR (100 MHz, DMSO-d6)
d: 165.2,156.3,155.5,153.6,144.8,
Compound 3f was prepared as described in Section 4.8, starting
with 4f (120 mg, 0.290 mmol). The crude product was crystallized
from DMF (2 mL) to give 89 mg (0.214 mmol, 74%) of 3f as a pale
yellow solid, mp. 257e259 ꢃC; HPLC purity 99%, tR ¼ 27.2 min; 1H
139.5, 134.3, 128.3 (2C), 127.7, 127.3 (2C), 126.8, 120.0, 118.9, 116.5,
116.4, 110.2, 62.5, 55.7, 43.3; IR (neat, cmꢀ1): 3221, 3029, 2935, 2151,
1584, 1543, 1491, 1346, 1299, 1049, 857, 800, 701, 603; HRMS (APCI/
ASAP, m/z): 378.1269 (calcd. C21H20N3O2S, 378.1276, [MþH]þ).
NMR (400 MHz, DMSO-d6) d: 9.77 (s, 1H), 8.53 (s, 1H), 8.29 (s, 1H),
8.23 (dd, J ¼ 6.8, 2.8, 1H), 7.84e7.78 (m, 1H), 7.75e7.72 (m, 1H), 7.46
(t, J ¼ 8.0, 1H), 7.20e7.17 (m, 1H), 7.11e7.07 (m, 1H), 5.33 (t, J ¼ 5.6,
1H), 4.57 (d, J ¼ 5.6, 2H), 3.98 (s, 3H); 13C NMR (100 MHz, DMSO-d6)
4.8.2. (3-Methoxy-4-(4-(phenylamino)thieno[2,3-d]pyrimidin-6-
yl)phenyl)methanol (3d)
d
: 166.3, 155.5, 153.7, 153.0 (d, J ¼ 242.8), 152.8, 145.3, 136.7 (d,
J ¼ 3.0), 135.7, 127.7, 123.3, 121.2 (d, J ¼ 7.0), 119.8, 119.0, 118.9 (d,
J ¼ 18.3), 117.1, 116.7 (d, J ¼ 21.5), 116.0, 110.2, 62.5, 55.8; 19F NMR
(564 MHz, DMSO-d6, C6F6)
d
: ꢀ125.75 (s, dec.); IR (neat, cmꢀ1): 3313,
2862, 1623, 1512, 1443, 1359, 1199, 1008, 810, 776, 690; HRMS (APCI/
ASAP, m/z): 416.0630 (calcd. C20H16ClFN3O2S, 416.0636, [MþH]þ).
Compound 3d was prepared as described in Section 4.8, starting
with 4d (250 mg, 0.692 mmol). The crude product was purified using
silica-gel column chromatography (Et2O/EtOAc, 8/2), Rf ¼ 0.20, and
recrystallized from THF (15 mL) using n-pentane (15 mL) as anti-
solvent. This gave 138 mg (0.380 mmol, 55%) of 3d as a pale yel-
low solid, mp. 229e231 ꢃC; HPLC purity 96%, tR ¼ 22.8 min; 1H NMR
4.8.5. (R)-(3-Fluoro-4-(4-((1-phenylethyl)amino)thieno[2,3-d]
pyrimidin-6-yl)phenyl)methanol (10b)
(400 MHz, DMSO-d6)
d: 9.63 (s, 1H), 8.47 (s, 1H), 8.33 (s, 1H),
7.89e7.84 (m, 2H), 7.76e7.72 (m, 1H), 7.43e7.37 (m, 2H), 7.18 (s, 1H),
7.14e7.06 (m, 2H), 5.33 (t, J ¼ 5.6, 1H), 4.57 (d, J ¼ 5.6, 2H), 3.97 (s,
3H); 13C NMR (100 MHz, DMSO-d6)
d: 166.3, 155.5, 154.1, 153.0, 145.1,

272
S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
Compound 10b was prepared as described in Section 4.8,
(400 MHz, DMSO-d6)
d
: 8.29 (d, J ¼ 8.0, 1H), 8.26 (s, 2H), 7.76 (d,
starting with 11b (156 mg, 0.413 mmol). The crude product was
purified by silica-gel column chromatography (Et2O/n-pentane, 9/
1), Rf ¼ 0.20. This gave 125 mg (0.329 mmol, 80%) of 10b as a white
J ¼ 1.9,1H), 7.44e7.43 (m, 2H), 7.34e7.31 (m, 2H), 7.30e7.29 (m,1H),
7.24e7.20 (m, 1H), 7.17e7.14 (m, 1H), 5.57e5.50 (m, 1H), 5.25 (t,
J ¼ 5.5, 1H), 4.52 (d, J ¼ 5,4, 2H), 3.94 (s, 3H), 1.57 (d, J ¼ 7.0, 3H); 13
C
solid, mp 171e173 ꢃC; HPLC purity: 99%, tR
¼
24.1 min;
NMR (100 MHz, DMSO-d6) d: 165.6, 155.6, 154.3, 153.6, 144.8, 135.2,
20
[
a]
¼ ꢀ386.5 (c 1.04, DMSO); 1H NMR (400 MHz, DMSO-d6)
d:
134.1, 128.3 (2C), 127.9, 126.6, 126.1 (2C), 126.0, 121.4, 116.6, 116.0,
112.3, 62.4, 56.0, 48.9, 22.5; HRMS (APCI/ASAP, m/z): found
392.1430, (calcd for C22H22N3O2S, 392.1433, [MþH]þ).
D
8.38 (d, J ¼ 7.7, 1H), 8.30 (s, 1H), 8.27 (s, 1H), 7.76e7.67 (m, 1H),
7.47e7.39 (m, 2H), 7.37e7.26 (m, 4H), 7.25e7.18 (m, 1H), 5.58e5.49
(m, 1H), 5.43 (t, J ¼ 5.6, 1H), 4.57 (d, J ¼ 5.5, 2H), 1.57 (d, J ¼ 6.9, 3H);
13C NMR (100 MHz, DMSO-d6)
d
: 165.3 (d, J ¼ 4.3), 158.6 (d,
4.9. General procedure for the aminolysis of methyl esters to
secondary amides
J ¼ 249.2), 155.8, 154.1, 145.9 (d, J ¼ 7.7), 144.6, 131.3 (d, J ¼ 3.7),
128.5 (d, J ¼ 3.4), 128.3 (2C), 126.7,126.1 (2C),122.9 (d, J ¼ 2.8), 119.0
(d, J ¼ 12.6), 118.5 (d, J ¼ 6.6), 116.8, 114.0 (d, J ¼ 22.6), 61.8 (d,
The methyl ester (130e200 mg) was dissolved in a minimal
amount of methanol (1e5 mL), and N,N-dimethylethane-1,2-
diamine (8.90 mmol, 1.0 mL) was added under N2 atmosphere.
The reaction mixture was heated to reflux and stirred for 3e12
days. The solvents were removed in vacuo, and purification was
performed as specified for each individual compound.
J ¼ 1.0), 49.0, 22.4; 19F NMR (564 MHz, DMSO-d6, C6F6)
: ꢀ116.64
d
(s, dec.); IR (neat, cmꢀ1): 3249, 3118, 2856, 1577, 1495, 1314, 1103,
1023, 746, 694; HRMS (APCI/ASAP, m/z): 380.1231 (calcd.
C
21H19FN3OS, 380.1233, [MþH]þ).
4.8.6. (R)-(2-Methoxy-3-(4-((1-phenylethyl)amino)thieno[2,3-d]
pyrimidin-6-yl)phenyl)methanol (27b)
4.9.1. (R)-N-(2-(Dimethylamino)ethyl)-2-methoxy-3-(4-((1-
phenylethyl)amino)thieno[2,3-d]pyrimidin-6-yl)benzamide (29b)
Compound 27b was prepared as described in Section 4.8,
starting with compound 26b (151 mg, 0.388 mmol). The crude
product was purified by silica-gel column chromatography (EtOAc/
n-pentane, 6/4), Rf ¼ 0.24. The resulting oil was dissolved in Et2O/
EtOAc (1/1) and n-pentane was added dropwise until crystallisation
started. Isolation and drying gave 110 mg (0.281 mmol, 73%) of 27b
Compound 29b was prepared as described in Section 4.9,
starting with compound 28b (198 mg, 0.472 mmol). The crude
product was purified by silica-gel column chromatography (CH2Cl2/
MeOH, 4/1), Rf ¼ 0.15. This gave 44 mg (0.093 mmol, 20%) of a 29b
as a yellow crystalline solid, mp. 128e132 ꢃC; HPLC purity: 99%,
as
a
white powder, mp. 113e124 ꢃC; HPLC purity: 99%,
20
tR ¼ 17.7 min; [
a
]
D
¼ ꢀ318.6 (c 1.00, DMSO); 1H NMR (600 MHz,
20
tR ¼ 24.2 min; [
a
]
D
¼ ꢀ274.9 (c 0.59, DMSO); 1H NMR (400 MHz,
DMSO-d6)
d
: 10.03 (s br,1H), 8.29 (s, 1H), 8.25 (s,1H), 8.19 (d, J ¼ 8.0,
DMSO-d6)
d
: 8.29 (s, 1H), 8.26 (d, J ¼ 7.9, 1H), 8.26 (s, 1H), 7.68e7.64
1H), 7.90e7.86 (m, 1H), 7.86e7.83 (m, 1H), 7.46e7.42 (m, 2H),
7.34e7.30 (m, 2H), 7.23e7.19 (m, 1H), 6.85 (t, J ¼ 7.7, 1H), 5.56e5.50
(m, 1H), 4.10e3.20 (m br, 3H), 3.54e3.50 (m, 2H), 2.77e2.72 (m,
2H), 2.42 (s, 6H), 1.57 (d, J ¼ 7.0, 3H); 13C NMR (150 MHz, DMSO-d6)
(m, 1H), 7.50e7.48 (m, 1H), 7.45e7.43 (m, 2H), 7.35e7.31 (m, 2H),
7.31e7.29 (m, 1H), 7.24e7.20 (m, 1H), 5.56e5.49 (m, 1H), 5.25 (t,
J ¼ 5.6, 1H), 4.64 (d, J ¼ 5.4, 2H), 3.66 (s, 3H), 1.57 (s, 3H); 13C NMR
(100 MHz, DMSO-d6) d: 166.1, 155.7, 153.8, 153.7, 144.7, 136.7, 133.5,
d: 170.0, 165.4, 160.8, 155.4, 153.2, 144.9, 134.6, 130.9, 128.3 (2C),
128.8, 128.3 (2C), 127.0, 126.7, 126.3 (2C), 126.1, 124.6, 117.6, 116.2,
61.2, 57.7, 49.0, 22.5; IR (neat, cmꢀ1): 3849, 3740, 3242, 2353, 2156,
2026, 1574, 1501, 997, 774, 691, 613, 577; HRMS (APCI/ASAP, m/z):
found 392.1425, (calcd for C22H22N3O2S, 392.1433, [MþH]þ).
127.8, 126.6, 126.1 (2C), 122.3, 116.1, 115.9, 115.8, 115.4, 61.9, 57.5,
48.9, 44.3 (2C), 36.2, 22.5; IR (neat, cmꢀ1): 3251, 2967, 2937, 2864,
2817, 2773, 1577, 747, 698; HRMS (APCI/ASAP, m/z): found 476.2116,
(calcd for C26H30N5O2S, 476.2120, [MþH]þ).
4.8.7. (R)-(4-Methoxy-3-(4-((1-phenylethyl)amino)thieno[2,3-d]
pyrimidin-6-yl)phenyl)methanol (33b)
4.9.2. (R)-N-(2-(Dimethylamino)ethyl)-3-methoxy-4-(4-((1-
phenylethyl)amino)thieno[2,3-d]pyrimidin-6-yl)benzamide (31b)
Compound 33b was prepared as described in Section 4.8,
starting with compound 32b (400 mg, 1.027 mmol), but reacted at
50 ꢃC. The crude product was purified by silica-gel column chro-
matography (EtOAc/n-pentane, 7/3), Rf ¼ 0.24. The resulting oil was
dissolved in Et2O/EtOAc (1/1) and n-pentane was added dropwise
until crystallisation started. Isolation and drying gave 76 mg
Compound 31b was prepared as described in Section 4.9,
starting with compound 30b (130 mg, 0.310 mmol). The crude
product was purified by silica-gel column chromatography (CH2Cl2/
MeOH, 4/1), Rf ¼ 0.11. This gave 73 mg (0.153 mmol, 50%) of a 31b as
a white crystalline solid, mp. 162e163 ꢃC; HPLC purity: 99%,
20
tR ¼ 30.4 min; [
a
]
D
¼ ꢀ359.3 (c 0.99, DMSO); 1H NMR (600 MHz,
(0.194 mmol, 19%) of a 33b as white powder, mp. 196e197 ꢃC; HPLC
DMSO-d6)
d
: 8.48 (s, 1H), 7.65 (d, J ¼ 8.0, 1H), 7.62 (s, 1H), 7.59e7.57
20
purity: 98%, tR ¼ 23.2 min; [
a
]
¼ ꢀ350.3 (c 0.54, DMSO); 1H NMR
(m, 1H), 7.47e7.44 (m, 2H), 7.39e7.34 (m, 3H), 7.31e7.27 (m, 1H),
D

S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
273
7.12e7.08 (m, 1H), 5.66e5.60 (m, 1H), 5.60e5.57 (m, 1H), 4.0 (s, 3H),
3.59e3.55 (m, 2H), 2.62 (t, J ¼ 5.8, 2H), 2.35 (s, 6H), 1.69 (d, J ¼ 6.6,
3H); 13C NMR (150 MHz, DMSO-d6)
: 166.9, 166.8, 156.5, 156.1,
Contributions and acknowledgement
d
The work was planned by S. Bugge and B. H. Hoff. Synthetic
chemistry was by S. Bugge, A. F Buene, I. U. Moen and E. M.
Skjønsfjell. Computational studies were by E. Sundby. N. Jurisch-
Yaksi performed all experiments on zebrafish embryos and ana-
lysed the data on toxicity and teratogenicity. The paper was written
by S. Bugge and B. Hoff with input from all authors.
Susana Villa Gonzalez is acknowledged for the HRMS experi-
ments. Roger Aarvik, Svein Jacob Kaspersen and Kent-Ove Kragseth
Sylte are acknowledged for their contributions. Anders Jahres
foundation and Technology Transfer Office (TTO-NTNU Trondheim)
are thanked for financial support.
154.4,143.6,135.8,135.4,129.01,128.96 (2C),127.7,126.5 (2C),125.5,
119.1,117.0,116.2,111.4, 58.1, 56.2, 50.1, 45.3 (2C), 37.3, 22.2; IR (neat,
cmꢀ1): 3258, 2970, 2934, 2810, 2760, 1588, 1539, 1513, 1315, 1106,
697; HRMS (APCI/ASAP, m/z): found 476.2112, (calcd for
C
26H30N5O2S, 476.2120, [MþH]þ).
4.9.3. (R)-N-(2-(Dimethylamino)ethyl)-4-methoxy-3-(4-((1-
phenylethyl)amino)thieno[2,3-d]pyrimidin-6-yl)benzamide (35b)
NOTUR is acknowledged for CPU-time.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2015.11.012.
References
Compound 35b was prepared as described in Section 4.9,
starting with compound 34b (148 mg, 0.353 mmol). The crude
product was purified by silica-gel column chromatography (CH2Cl2/
MeOH, 5/1), Rf ¼ 0.11. This gave 109 mg (0.229 mmol, 65%) of a 35b
[1] S. Laufer, J. Bajorath, New frontiers in kinases: second generation inhibitors,
J. Med. Chem. 57 (2014) 2167e2168. http://dx.doi.org/10.1021/jm500195x.
[2] P. Cohen, D.R. Alessi, Kinase drug discovery e what's next in the field? ACS
Chem. Biol. 8 (2013) 96e104. http://dx.doi.org/10.1021/cb300610s.
[3] R. Roskoski Jr., ErbB/HER protein-tyrosine kinases: structures and small
molecule inhibitors, Pharmacol. Res. 87 (2014) 42e59. http://dx.doi.org/10.
1016/j.phrs.2014.06.001.
[4] K. Beaumont, S.M. Cole, K. Gibson, J.R. Gosset, ADMET for the medicinal
chemist, RSC Drug Discov. Ser. 1 (2010) 61e98. http://dx.doi.org/10.1039/
9781849731102-00061.
[5] Z. O'Brien, M.F. Moghaddam, Small molecule kinase inhibitors approved by
the FDA from 2000 to 2011: a systematic review of preclinical ADME data,
Expert Opin. Drug Metab. Toxicol. 9 (2013) 1597e1612. http://dx.doi.org/10.
1517/17425255.2013.834046.
[6] G.L. Amidon, H. Lennernaes, V.P. Shah, J.R. Crison, A theoretical basis for a
biopharmaceutic drug classification: the correlation of in vitro drug product
dissolution and in vivo bioavailability, Pharm. Res. 12 (1995) 413e420. http://
dx.doi.org/10.1023/A:1016212804288.
as a white crystalline solid, mp. 131e134 ꢃC; HPLC purity: 99%,
20
tR ¼ 16.2 min; [
a
]
D
¼ ꢀ313.1 (c 1.00, DMSO); 1H NMR (600 MHz,
DMSO-d6)
d
: 8.57e8.52 (m, 1H), 8.39 (s, 1H), 8.36 (d, J ¼ 7.0, 1H),
8.35 (s, 1H), 8.28 (s, 1H), 7.92e7.89 (m, 1H), 7.47e7.42 (m, 2H),
7.35e7.30 (m, 2H), 7.29e7.25 (m, 1H), 7.24e7.20 (m, 1H), 5.58e5.51
(m, 1H), 4.01 (s, 3H), 3.44 (q, J ¼ 6.2, 2H), 2.66e2.57 (m, 2H), 2.33 (s,
6H), 1.58 (d, J ¼ 7.1, 3H); 13C NMR (150 MHz, DMSO-d6)
d: 165.8,
165.4, 157.4, 155.7, 153.8, 144.7, 133.2, 128.4, 128.3 (2C), 127.4, 127.2,
126.6, 126.1 (2C), 121.6, 117.6, 116.0, 111.9, 57.9, 56.2, 49.0, 44.7 (2C),
36.8, 22.5; IR (neat, cmꢀ1): 3278, 2970, 2937, 2817, 2767, 1577, 1509,
1263, 699; HRMS (APCI/ASAP, m/z): found 476.2115, (calcd for
C
26H30N5O2S, 476.2120, [MþH]þ).
[7] L. Di, P.V. Fish, T. Mano, Bridging solubility between drug discovery and
development, Drug Discov. Today 17 (2012) 486e495. http://dx.doi.org/10.
1016/j.drudis.2011.11.007.
[8] M.E. Swarbrick, The learning and evolution of medicinal chemistry against
kinase targets, RSC Drug Discov. Ser. 19 (2012) 79e95. http://dx.doi.org/10.
1039/9781849733557-00079.
4.9.4. (R)-N-(2-(Dimethylamino)ethyl)-3,5-dimethoxy-4-(4-((1-
phenylethyl)amino)thieno[2,3-d]pyrimidin-6-yl)benzamide (37b)
[9] M.J. Waring, Lipophilicity in drug discovery, Expert Opin. Drug Discov. 5
(2010) 235e248. http://dx.doi.org/10.1517/17460441003605098.
[10] A.L. Gill, M. Verdonk, R.G. Boyle, R. Taylor, A comparison of physicochemical
property profiles of marketed oral drugs and orally bioavailable anti-cancer
protein kinase inhibitors in clinical development, Curr. Top. Med. Chem. 7
(2007) 1408e1422. http://dx.doi.org/10.2174/156802607781696819.
[11] F. Giordanetto, J. Bostrom, C. Tyrchan, Follow-on drugs: how far should
chemists look? Drug Discov. Today 16 (2011) 722e732. http://dx.doi.org/10.
1016/j.drudis.2011.05.011.
[12] J.J. Gagne, N.K. Choudhry, How many “me-too” drugs is too many? J. Am. Med.
Assoc. 305 (2011) 711e712. http://dx.doi.org/10.1001/jama.2011.152.
[13] X. Ji, T. Peng, X. Zhang, J. Li, W. Yang, L. Tong, R. Qu, H. Jiang, J. Ding, H. Xie,
H. Liu, Design, synthesis and biological evaluation of novel 6-alkenylamides
substituted of 4-anilinothieno[2,3-d]pyrimidines as irreversible epidermal
growth factor receptor inhibitors, Bioorg. Med. Chem. 22 (2014) 2366e2378.
http://dx.doi.org/10.1016/j.bmc.2014.01.035.
[14] F. Solca, G. Dahl, A. Zoephel, G. Bader, M. Sanderson, C. Klein, O. Kraemer,
F. Himmelsbach, E. Haaksma, G.R. Adolf, Target binding properties and cellular
activity of afatinib (BIBW 2992), an irreversible ErbB family blocker,
J. Pharmacol. Exp. Ther. 343 (2012) 342e350. http://dx.doi.org/10.1124/jpet.
112.197756.
[15] T. Beckers, S. Mahboobi, A. Sellmer, M. Winkler, E. Eichhorn, H. Pongratz,
T. Maier, T. Ciossek, T. Baer, G. Kelter, H.-H. Fiebig, M. Schmidt, Chimerically
designed HDAC- and tyrosine kinase inhibitors. A series of erlotinib hybrids as
dual-selective inhibitors of EGFR, HER2 and histone deacetylases, Med-
ChemComm 3 (2012) 829e835. http://dx.doi.org/10.1039/c2md00317a.
[16] T. Beckers, A. Sellmer, E. Eichhorn, H. Pongratz, C. Schaechtele, F. Totzke,
G. Kelter, R. Krumbach, H.-H. Fiebig, F.-D. Boehmer, S. Mahboobi, Novel in-
hibitors of epidermal growth factor receptor: (4-(arylamino)-7H-pyrrolo[2,3-
d]pyrimidin-6-yl)(1H-indol-2-yl)methanones and (1H-indol-2-yl)(4-(phenyl-
amino)thieno[2,3-d]pyrimidin-6-yl)methanones, Bioorg. Med. Chem. 20
(2012) 125e136. http://dx.doi.org/10.1016/j.bmc.2011.11.023.
Compound 37b was prepared as described in Section 4.9,
starting with compound 36b (203 mg, 0.452 mmol). The crude
product was purified by silica-gel column chromatography (CH2Cl2/
MeOH, 4/1), Rf ¼ 0.15. This gave 137 mg (0.271 mmol, 60%) of a 37b
as a white crystalline solid, mp. 194e195 ꢃC; HPLC purity: 99%,
20
tR ¼ 30.3 min; [
a]
¼ ꢀ260.8 (c 1.01, DMSO); 1H NMR (600 MHz,
D
DMSO-d6)
d
: 8.66e8.62 (m, 1H), 8.27 (s, 1H), 8.24 (d, J ¼ 8.0, 1H),
7.98 (s, 1H), 7.44e7.41 (m, 2H), 7.34e7.30 (m, 2H), 7.28 (s, 2H),
7.23e7.20 (m, 1H), 5.58e5.52 (m, 1H), 3.88 (s, 6H), 3.44 (q, J ¼ 6.5,
2H), 2.60e2.53 (m, 2H), 2.30 (s, 6H), 1.56 (d, J ¼ 7.0, 3H); 13C NMR
(150 MHz, DMSO-d6) d: 165.9, 165.4, 157.3 (2C), 155.6, 153.5, 144.8,
135.8, 128.8, 128.3 (2C), 126.6, 126.1 (2C), 121.1, 115.8, 113.2, 103.5
(2C), 58.0, 56.3 (2C), 48.9, 44.9 (2C), 37.2, 22.4; IR (neat, cmꢀ1):
3245, 3121, 2977, 2934, 2813, 1582, 1556, 1536, 1500, 1312, 1122,
785, 700; HRMS (APCI/ASAP, m/z): found 506.2218, (calcd for
C
27H32N5O3S, 506.2226, [MþH]þ).

274
S. Bugge et al. / European Journal of Medicinal Chemistry 107 (2016) 255e274
[17] E.R. Wood, L.M. Shewchuk, B. Ellis, P. Brignola, R.L. Brashear, T.R. Caferro,
S.H. Dickerson, H.D. Dickson, K.H. Donaldson, M. Gaule, R.J. Griffin,
A.M. Hassell, B. Keith, R. Mullin, K.G. Petrov, M.J. Reno, D.W. Rusnak,
S.M. Tadepalli, J.C. Ulrich, C.D. Wagner, D.E. Vanderwall, A.G. Waterson,
J.D. Williams, W.L. White, D.E. Uehling, 6-Ethynylthieno[3,2-d]- and 6-
ethynylthieno[2,3-d]pyrimidin-4-anilines as tunable covalent modifiers of
ErbB kinases, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 2773e2778. http://dx.
doi.org/10.1073/pnas.0708281105.
[18] T.R. Rheault, T.R. Caferro, S.H. Dickerson, K.H. Donaldson, M.D. Gaul,
A.S. Goetz, R.J. Mullin, O.B. McDonald, K.G. Petrov, D.W. Rusnak,
L.M. Shewchuk, G.M. Spehar, A.T. Truesdale, D.E. Vanderwall, E.R. Wood,
D.E. Uehling, Thienopyrimidine-based dual EGFR/ErbB-2 inhibitors, Bioorg.
Med. Chem. Lett. 19 (2009) 817e820. http://dx.doi.org/10.1016/j.bmcl.2008.
12.011.
[19] S. Mahboobi, A. Sellmer, M. Winkler, E. Eichhorn, H. Pongratz, T. Ciossek,
T. Baer, T. Maier, T. Beckers, Novel chimeric histone deacetylase inhibitors: a
series of lapatinib hybrids as potent inhibitors of epidermal growth factor
receptor (EGFR), human epidermal growth factor receptor 2 (HER2), and
histone deacetylase activity, J. Med. Chem. 53 (2010) 8546e8555. http://dx.
doi.org/10.1021/jm100665z.
[20] S. Bugge, S.J. Kaspersen, S. Larsen, U. Nonstad, G. Bjørkøy, E. Sundby, B.H. Hoff,
Structureeactivity study leading to identification of a highly active thieno-
pyrimidine based EGFR inhibitor, Eur. J. Med. Chem. 75 (2014) 354e374.
http://dx.doi.org/10.1016/j.ejmech.2014.01.042.
Disease, John Wiley & Sons, Hoboken, USA, 2012, pp. 101e114. http://dx.doi.
org/10.1002/9781118435762.ch6.
[35] C. Bissantz, B. Kuhn, M. Stahl, A medicinal chemist's guide to molecular in-
teractions, J. Med. Chem. 53 (2010) 5061e5084. http://dx.doi.org/10.1021/
jm100112j.
[36] C. Laurence, K.A. Brameld, J. Graton, Q.J.-Y. Le, E. Renault, The pKBHX data-
base: toward a better understanding of hydrogen-bond basicity for medicinal
chemists, J. Med. Chem. 52 (2009) 4073e4086. http://dx.doi.org/10.1021/
jm801331y.
[37] N. Hayashi, H. Higuchi, Twist angles and torsional potentials of 2,2'-biben-
zothiophene, 2,2'-biindole, and 2,2'-bibenzofuran, Heterocycles 74 (2007)
763e769. http://dx.doi.org/10.3987/COM-07-S(W)60.
[38] N. Hayashi, Y. Saito, H. Higuchi, Effect of fused benzene ring on rotational
barriers of 2,2'-bifuran, 2-phenylfuran, biphenyl, and their benzo analogues,
Heterocycles 77 (2009) 1261e1268. http://dx.doi.org/10.3987/COM-08-S(F)
112.
[39] Y. Takahashi, M. Hashizume, K. Shin, T. Terauchi, K. Takeda, S. Hibi, K. Murata-
Tai, M. Fujisawa, K. Shikata, R. Taguchi, M. Ino, H. Shibata, M. Yonaga, Design,
synthesis, and structure-activity relationships of novel pyrazolo[5,1-b]thia-
zole derivatives as potent and orally active corticotropin-releasing factor 1
receptor antagonists, J. Med. Chem. 55 (2012) 8450e8463. http://dx.doi.org/
10.1021/jm300864p.
[40] G. Keri, L. Orfi, G. Nemeth, Rational drug design of kinase inhibitors for signal
transduction therapy, Methods Princ. Med. Chem. 49 (2011) 87e114. http://
dx.doi.org/10.1002/9783527633470.ch4.
[41] T.T. Talele, M.L. McLaughlin, Molecular docking/dynamics studies of Aurora A
kinase inhibitors, J. Mol. Graph. Model. 26 (2008) 1213e1222. http://dx.doi.
org/10.1016/j.jmgm.2007.11.003.
[21] S. Bugge, I.U. Moen, K.-O.K. Sylte, E. Sundby, B.H. Hoff, Truncated structures
used in search for new lead compounds and in a retrospective analysis of
thienopyrimidine-based EGFR inhibitors, Eur. J. Med. Chem. 94 (2015)
175e194. http://dx.doi.org/10.1016/j.ejmech.2015.03.004.
[22] K.-F. Chen, K.-C. Pao, J.-C. Su, Y.-C. Chou, C.-Y. Liu, H.-J. Chen, J.-W. Huang,
I. Kim, C.-W. Shiau, Development of erlotinib derivatives as CIP2A-ablating
agents independent of EGFR activity, Bioorg. Med. Chem. 20 (2012)
6144e6153. http://dx.doi.org/10.1016/j.bmc.2012.08.039.
[42] A.M. Wassermann, M. Wawer, J. Bajorath, Activity landscape representations
for structureeactivity relationship analysis, J. Med. Chem. 53 (2010)
8209e8223. http://dx.doi.org/10.1021/jm100933w.
[43] A.F. Stepan, V. Mascitti, K. Beaumont, A.S. Kalgutkar, Metabolism-guided drug
[23] G.W. Rewcastle, W.A. Denny, A.J. Bridges, H. Zhou, D.R. Cody, A. McMichael,
D.W. Fry, Tyrosine kinase inhibitors. 5. Synthesis and structure-activity re-
lationships for 4-[(phenylmethyl)amino]- and 4-(phenylamino)quinazolines
as potent adenosine 5'-triphosphate binding site inhibitors of the tyrosine
kinase domain of the epidermal growth factor receptor, J. Med. Chem. 38
(1995) 3482e7348. http://dx.doi.org/10.1021/jm00018a008.
[24] S.J. Kaspersen, J. Han, K.G. Nørsett, L. Rydsa, E. Kjøbli, S. Bugge, G. Bjørkøy,
E. Sundby, B.H. Hoff, Identification of new 4-N-substituted 6-aryl-7H-pyrrolo
[2,3-d]pyrimidine-4-amines as highly potent EGFR-TK inhibitors with Src-
family activity, Eur. J. Pharm. Sci. 59 (2014) 69e82. http://dx.doi.org/10.
1016/j.ejps.2014.04.011.
design, MedChemComm
c2md20317k.
4 (2013) 631e652. http://dx.doi.org/10.1039/
[44] S. Boehrer, L. Ades, T. Braun, L. Galluzzi, J. Grosjean, C. Fabre, G. Le Roux,
C. Gardin, A. Martin, S. de Botton, P. Fenaux, G. Kroemer, Erlotinib exhibits
antineoplastic off-target effects in AML and MDS: a preclinical study, Blood
111 (2008) 2170e2180. http://dx.doi.org/10.1182/blood-2007-07-100362.
[45] J. Han, E. Sundby, B.H. Hoff, Solvent selection in synthesis of 4-(1-
arylfluoroethoxy)quinazolines and thienopyrimidines, J. Fluor. Chem. 153
(2013) 82e88. http://dx.doi.org/10.1016/j.jfluchem.2013.05.011.
[46] Induced Fit Docking Protocol 2013-3, Glide Version 6.1, Prime Version 3.4,
€
Schrodinger, LLC, New York, NY, 2013.
[25] N.A. Meanwell, Synopsis of some recent tactical application of bioisosteres in
drug design, J. Med. Chem. 54 (2011) 2529e2591. http://dx.doi.org/10.1021/
jm1013693.
[47] W. Sherman, H.S. Beard, R. Farid, Use of an induced fit receptor structure in
virtual screening, Chem. Biol. Drug Des. 67 (2006) 83e84. http://dx.doi.org/10.
1111/j.1747-0285.2005.00327.x.
[26] S.R. Langdon, P. Ertl, N. Brown, Bioisosteric replacement and scaffold hopping
in lead generation and optimization, Mol. Inf. 29 (2010) 366e385. http://dx.
doi.org/10.1002/minf.201000019.
[48] W. Sherman, T. Day, M.P. Jacobson, R.A. Friesner, R. Farid, Novel procedure for
modeling ligand/receptor induced fit effects, J. Med. Chem. 49 (2006)
534e553. http://dx.doi.org/10.1021/jm050540c.
[27] S. Bugge, S.J. Kaspersen, E. Sundby, B.H. Hoff, Route selection in the synthesis
of C-4 and C-6 substituted thienopyrimidines, Tetrahedron 68 (2012)
9226e9233. http://dx.doi.org/10.1016/j.tet.2012.08.090.
[49] T. Oberhauser, A new bromination method for phenols and anisoles: NBS/
HBF4$Et2O in CH3CN, J. Org. Chem. 62 (1997) 4504e4506. http://dx.doi.org/10.
1021/JO9622993.
[28] S. Bugge, E.M. Skjønsfjell, F.B. Willumsen, E. Sundby, B.H. Hoff, Improved and
scalable preparation of 6-bromo-4-chlorothieno[2,3-d]pyrimidine, Chem.
Heterocycl. Compd. 50 (2014) 1177e1187. http://dx.doi.org/10.1007/s10593-
014-1579-z.
[29] J.-J. Shie, J.-M. Fang, Direct conversion of aldehydes to amides, tetrazoles, and
triazines in aqueous media by one-pot tandem reactions, J. Org. Chem. 68
(2003) 1158e1160. http://dx.doi.org/10.1021/jo026407z.
[30] S.C. Ghosh, J.S.Y. Ngiam, A.M. Seayad, D.T. Tuan, C.L.L. Chai, A. Chen, Copper-
catalyzed oxidative amidation of aldehydes with amine salts: synthesis of
primary, secondary, and tertiary amides, J. Org. Chem. 77 (2012) 8007e8015.
http://dx.doi.org/10.1021/jo301252c.
[31] M.A. Ali, T. Punniyamurthy, Palladium-catalyzed one-pot conversion of alde-
hydes to amides, Adv. Synth. Catal. 352 (2010) 288e292. http://dx.doi.org/10.
1002/adsc.200900799.
[50] S. Bertini, T. Parkkari, J.R. Savinainen, C. Arena, G. Saccomanni, S. Saguto,
A. Ligresti, M. Allara, A. Bruno, L. Marinelli, V. Di Marzo, E. Novellino,
C. Manera, M. Macchia, Synthesis, biological activity and molecular modeling
of new biphenylic carboxamides as potent and selective CB2 receptor ligands,
Eur. J. Med. Chem. 90 (2015) 526e536. http://dx.doi.org/10.1016/j.ejmech.
2014.11.066.
[51] A.D. Burrows, C. Frost, M.F. Mahon, C. Richardson, Post-synthetic modification
of tagged metal-organic frameworks, Angew. Chem. Int. Ed. 47 (2008)
8482e8486. http://dx.doi.org/10.1002/anie.200802908.
[52] N. Varga, I. Sutkeviciute, C. Guzzi, J. McGeagh, I. Petit-Haertlein, S. Gugliotta,
J. Weiser, J. Angulo, F. Fieschi, A. Bernardi, Selective targeting of dendritic cell-
specific intercellular adhesion molecule-3-grabbing nonintegrin (DC-SIGN)
with mannose-based glycomimetics: synthesis and interaction studies of
bis(benzylamide) derivatives of
a pseudomannobioside, Chem. Eur. J. 19
[32] M. Ishikawa, Y. Hashimoto, Improvement in aqueous solubility in small
molecule drug discovery programs by disruption of molecular planarity and
symmetry, J. Med. Chem. 54 (2011) 1539e1554. http://dx.doi.org/10.1021/
jm101356p.
(2013) 4786e4797. http://dx.doi.org/10.1002/chem.201202764.
[53] H. Kikuchi, Y. Matsuo, Y. Katou, Y. Kubohara, Y. Oshima, Isolation, synthesis,
and biological activity of biphenyl and m-terphenyl-type compounds from
Dictyostelium cellular slime molds, Tetrahedron 68 (2012) 8884e8889. http://
dx.doi.org/10.1016/j.tet.2012.08.041.
[33] A.P. Li, Screening for human ADME/Tox drug properties in drug discovery,
Drug Discov. Today
6446(01)01712-3.
[34] L. Di, E.H. Kerns, ADME properties of drugs, in: N. Civjan (Ed.), Chemical
Biology: Approaches to Drug Discovery and Development to Targeting
6
(2001) 357e366. http://dx.doi.org/10.1016/S1359-
[54] G.B. Arhancet, A. Casimiro-Garcia, X. Chen, D. Hepworth, M.J. Meyers,
D.W. Piotrowski, R.K. Raheja, Preparation of 4,5-dihydro-1H-pyrazole Com-
pounds as Mineralocorticoid Receptor Antagonists, 2010. WO2010116282A1.