Overcoming imatinib resistance in chronic myelogenous leukemia cells using non-cytotoxic cell death modulators

Article abstract of DOI:10.1016/j.ejmech.2019.111748

Recent studies examined the possibility to overcome imatinib resistance in chronic myeloid leukemia (CML) patients by combination therapy with peroxisome proliferator-activated receptor gamma (PPARγ) ligands. Pioglitazone, a full PPARγ agonist, improved the survival of patients by the gradual elimination of the residual CML stem cell pool. To evaluate the importance of the pharmacological profile of PPARγ agonists on the ability to circumvent resistance, the partial PPARγ agonist 4‘-((2-propyl-1H-benzo[d]imidazol-1-yl)methyl)-[1,1’-biphenyl]-2-carboxylic acid, derived from telmisartan, and other related derivatives were investigated. The 4-substituted benzimidazole derivatives bearing a [1,1′-biphenyl]-2-carboxamide moiety sensitized K562-resistant cells to imatinib treatment. Especially the derivatives 18a-f, which did not activate PPARγ to more than 40% at 10 μM, retrieved the cytotoxicity of imatinib in these cells. The cell death modulating properties were higher than that of pioglitazone. It is of interest to note that all novel compounds were not cytotoxic neither on non-resistant nor on resistant cells. They exerted antitumor potency only in combination with imatinib.

Full text of DOI:10.1016/j.ejmech.2019.111748

Journal Pre-proof
Overcoming imatinib resistance in chronic myelogenous leukemia cells using non-
cytotoxic cell death modulators
Anna M. Schoepf, Stefan Salcher, Petra Obexer, Ronald Gust
PII:
S0223-5234(19)30900-6
DOI:
https://doi.org/10.1016/j.ejmech.2019.111748
EJMECH 111748
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 23 July 2019
Revised Date: 27 September 2019
Accepted Date: 27 September 2019
Please cite this article as: A.M. Schoepf, S. Salcher, P. Obexer, R. Gust, Overcoming imatinib resistance
in chronic myelogenous leukemia cells using non-cytotoxic cell death modulators, European Journal of
Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.111748.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier Masson SAS.

Overcoming imatinib resistance in chronic
myelogenous leukemia cells using non-cytotoxic cell
death modulators
Anna M. Schoepf a, Stefan Salcher b, Petra Obexer b,c, Ronald Gust a*
a Department of Pharmaceutical Chemistry, Institute of Pharmacy, CMBI - Center for Molecular
Biosciences Innsbruck, University of Innsbruck, CCB - Centrum for Chemistry and
Biomedicine, Innrain 80-82, 6020 Innsbruck, Austria
b Tyrolean Cancer Research Institute, Innrain 66, 6020 Innsbruck, Austria
c Department of Pediatrics II, Medical University of Innsbruck, Innrain 66, 6020 Innsbruck,
Austria
* Corresponding Author, E-mail: ronald.gust@uibk.ac.at
Highlights
• Benzimidazole derivatives sensitized resistant CML cells to imatinib treatment
• All compounds were non-cytotoxic against HS-5, K562, K562-resistant, or COS-7 cells
• STAT5 downregulation as mode of action
• Possibility to circumvent tyrosine kinase inhibitor resistance of cancers
Abstract
Recent studies examined the possibility to overcome imatinib resistance in chronic myeloid
leukemia (CML) patients by combination therapy with peroxisome proliferator-activated
receptor gamma (PPARγ) ligands. Pioglitazone, a full PPARγ agonist, improved the survival of
patients by the gradual elimination of the residual CML stem cell pool. To evaluate the
importance of the pharmacological profile of PPARγ agonists on the ability to circumvent
resistance, the partial PPARγ agonist 4‘-((2-propyl-1H-benzo[d]imidazol-1-yl)methyl)-[1,1’-
biphenyl]-2-carboxylic acid, derived from telmisartan, and other related derivatives were
investigated. The 4-substituted benzimidazole derivatives bearing a [1,1’-biphenyl]-2-
carboxamide moiety sensitized K562-resistant cells to imatinib treatment. Especially the
derivatives 18a-f, which did not activate PPARγ to more than 40% at 10 µM, retrieved the
cytotoxicity of imatinib in these cells. The cell death modulating properties were higher than that
of pioglitazone. It is of interest to note that all novel compounds were not cytotoxic neither on
non-resistant nor on resistant cells. They exerted antitumor potency only in combination with
imatinib.
1

Graphical Abstract
Keywords
peroxisome proliferator-activated receptor γ, structure-activity relationship, chronic myeloid
leukemia, imatinib resistance, cell death modulators.
Abbreviations
TKI, tyrosine kinase inhibitor; PPARγ, peroxisome proliferator-activated receptor gamma; SCs,
stem cells; CMR, complete molecular response; TZD, thiazolidinedione; SAR, structure-activity
relationship; ARB, angiotensin II type 1 receptor blocker; AcOH, glacial acetic acid; Pd/C,
palladium 10% on carbon; MeOH, methanol; EG, ethylene glycol; MCB, monochlorobenzene;
PyBOP,
benzotriazol-1-yl-oxytripyrrolidino-phosphonium
hexafluorophosphate;
HOBt,
hydroxybenzotriazole; DIPEA, N,N-diisopropylethylamine; EA, ethyl acetate; EtOH, ethanol;
PE, petroleum ether; DIPE, diisopropyl ether; eq, equivalents; HRMS, high-resolution mass
spectrometry; Amax, intrinsic activation; SD, standard deviation; SEM, standard error of the
mean; CTR, control; PIO, pioglitazone; TELMI, telmisartan; DMEM, dulbecco's modified eagle
medium; FCS, fetal calf serum; RPMI, Roswell Park Memorial Institute; PBS, phosphate-
buffered saline.
2

1 Introduction
Chronic myeloid leukemia (CML) is a myeloproliferative disease caused by the reciprocal
translocation and the subsequent proliferation of a monoclonal multipotent hematopoietic
progenitor cell. As a consequence of the resulting Philadelphia chromosome, the oncogene BCR-
ABL fusion protein arises with an increased tyrosine kinase activity [1].
The introduction of the first tyrosine kinase inhibitor (TKI) imatinib (Fig. 1) in 2001 was a
breakthrough in the therapy of CML, since the survival of patients was considerably improved
[1]. However, it is impossible to eradicate the disease with imatinib, because its use is associated
with severe limitations, e.g. due to poor efficacy on non-cycling CML cells and the inability of
TKIs to eliminate the quiescent stem cell (SC) pool [2-4]. Thus, achievement of sustained
complete molecular response (CMR) is tough to reach in a high percentage of patients and
resistances occur [3-6]. Stable patients with continuous CMR were examined to discontinue
imatinib therapy, though molecular relapse was observed in 40 to 60% of the patients [7-8].
The combination of TKIs with drugs targeting CML SCs was stated as new approach to
overcome resistance [9]. Recent studies confirm therapeutic benefits such as the induction of
durable CMR upon combining imatinib with other therapeutics, including peroxisome
proliferator-activated receptor gamma (PPARγ) ligands [6, 8-12].
PPARs belong to the family of nuclear receptors and cover the three subtypes α, β/δ and γ [13].
Ligands of PPARγ have been investigated as anticancer agents in a variety of cancers, e.g. colon,
pancreatic, breast, and prostate cancer as well as hematologic malignancies like leukemia [14-
15].
Thiazolidinediones (TZDs) comprise a class of drugs with agonistic activity towards PPARγ.
Although the most compatible representative pioglitazone (Fig. 1) was initially just used for the
therapy of type II diabetes mellitus, it has entered the clinical trial to treat hematopoietic
malignancies [11, 16]. Based on the gradual elimination of CML SCs, Rousselot et al. [8]
concluded from their studies in 2017 that combining pioglitazone with imatinib increases the
percentage of CML patients achieving CMR.
Fig. 1. Structures of the first-generation tyrosine kinase inhibitor imatinib and the full PPARγ agonist pioglitazone.
Telmisartan (Fig. 2), an angiotensin II type 1 receptor blocker (ARB), was identified as partial
PPARγ agonist, showing a safer pharmacological profile without TZD-like side effects [17].
Preclinical studies revealed promising outcomes for its use in different types of cancer, particular
non-small cell lung cancer with TKI resistance [14, 16].
3

In a structure-activity study we identified the 4‘-((2-propyl-1H-benzo[d]imidazol-1-
yl)methyl)-[1,1’-biphenyl]-2-carboxylic acid as essential core in telmisartan and used it as new
lead structure (17a, Fig. 2) to optimize the PPARγ activity [18-19].
Fig. 2. Structures of the partial PPARγ agonist telmisartan and the derived lead 17a.
As previous data already suggested that position 4 at the benzimidazole scaffold had major
influence on partial agonism [19], various substituents were introduced at this position within the
present study. The substituents were selected in terms of polarity or distribution of electrons (H,
F, Cl, Br, CH3, CF3, OCH3, OH, NO2, NH2). The resulting compounds were designed with the
intention to determine their influence on PPARγ activation and their effects upon co-application
with imatinib against resistant CML cells. Thus, all compounds were evaluated in a dual-
luciferase reporter assay using COS-7 cells, transiently transfected with the plasmids pGal5-TK-
pGL3, pGal4-hPPARγDEF, and pRenilla-CMV [20]. In a modified MTT assay, the influence of
the compounds on the metabolic activity of COS-7 cells was measured. Thus, unselective
cytotoxic properties that might have influence on the results of the luciferase assay were
excluded. Moreover, cytotoxicity was studied in the human non-malignant HS-5 cell line as well
as the CML cell line K562 by propidium iodide (PI) FACS analyses. The imatinib-sensitive
K562 cell line and the imatinib-resistant subclone K562-resistant were used to examine the
possibility to circumvent imatinib resistance [21].
Furthermore, it was tried to specifically reduce the intrinsic activation of PPARγ but to prevent
the effects on resistant CML cells. According to Obermoser et al. [22], the carboxylic acids 17a-j
were transformed to the respective carboxamides 18a-j and the biological effects were
comparatively evaluated.
2 Results and discussion
2.1 Synthesis
The synthesis pathway of the compounds is depicted in schemes 1 and 2 and was adapted
based on the approach described by Obermoser et al. [22-23]. The 1,2-phenylenediamines, if not
commercially available, were synthesized from the respectively substituted aniline derivatives
(Scheme 1).
2-Fluoroaniline 1 was transformed to the acetanilide 5 by reaction with acetic anhydride,
followed by nitration with HNO3 [24]. The subsequent reduction of 6 with SnCl2 yielded the 3-
fluoro-1,2-phenylenediamine 9b [25]. The 3-chloro-1,2-phenylenediamine 9c was obtained from
1,2-dichloro-3-nitrobenzene 2 after reaction with NH3 and reduction with SnCl2 [25-26].
4

Compound 8 resulted from O-methylation of 2-nitro-6-hydroxyaniline 3 and was reduced in the
same way to give 9g [25-27]. 2,6-Dinitroaniline 4 was reduced with Na2S to 9i [28].
Reagents and conditions: (a) acetic anhydride, 120 °C; (b) HNO3, AcOH, acetic anhydride, -5 °C; (c) SnCl2 x 2 H2O, HCl, EtOH,
rt; (d) NH3, EtOH, 80 °C; (e) CH3I, K2CO3, DMF, rt; (f) Na2S, NaHCO3, H2O, 65 °C.
Scheme 1. Syntheses of the 3-substituted 1,2-phenylenediamines 9b,c,g,i.
In the next step (Scheme 2), the appropriate benzimidazoles were synthesized. Reaction of 1,2-
phenylenediamine (9a), 3-methyl-1,2-phenylenediamine (9e), and 9b,c,g,i with butyric
anhydride formed a mixture of mono- and bis-butyric anilides, which were heated in diluted HCl
without separation to yield 10a-c,e,g,i [23]. 10d was synthesized in a Sandmeyer reaction
(NaNO2, HBr, Cu) from 10j [29], which was prior formed by reduction of 10i with H2, Pd/C
[30].
N-Alkylation of the benzimidazoles 10b-d,g,i with 4'-(bromomethyl)-[1,1'-biphenyl]-2-carbo-
nitrile in the presence of NaH (d: method A) gave the 2-carbonitriles as an isomeric mixture,
from which 15b-d,g,i were separated by column chromatography. Compound 15i was then used
to obtain 15j by reduction of the 4-NO2 group with Fe/HCl [22, 31].
In a comparable reaction, the benzimidazoles 10a-c,e were alkylated with 4'-(bromomethyl)-
[1,1'-biphenyl]-2-carboxylic acid methyl ester (d: method B) to get 16a-c,e.
5

Reagents and conditions: (a) butyric anhydride, HCl, 120 °C; (b) H2, Pd/C, MeOH; (c) NaNO2, HBr, Cu, H2O, 5 °C to 60 °C; (d)
method A: 4’-(bromomethyl)-[1,1’-biphenyl]-2-carbonitrile, NaH, DMF, rt; method B: 4’-(bromomethyl)-[1,1’-biphenyl]-2-
carboxylic acid methyl ester, NaH, DMF, rt; (e) method C: KOH, EG, H2O, 160 °C; method D: H2SO4, H2O, 120 °C; (f) NH4Cl,
PyBOP, DIPEA, DMF, 0 °C to rt; (g) HNO3, -10 °C; (h) SnCl2 x 2 H2O, HCl, EtOH, 80 °C; (i) Fe, HCl, MeOH, 70 °C; (j) BBr3,
DCM, MCB, -80 °C to rt.
Scheme 2. Syntheses of 4-substituted 4'-((2-propyl-1H-benzo[d]imidazol-1-yl)methyl)-[1,1'-biphenyl]-2-carboxylic acids 17a-j
and the respective carboxamides 18a-j.
6

For the synthesis of 15f, a slightly modified reaction pathway, first described by Obermoser et
al. [23], was used. This stereoselective reaction started from 3-trifluoromethylaniline 11, which
was transformed with butyric anhydride to the butyric anilide 12, followed by nitration with
HNO3 (13) and N-alkylation with 4'-(bromomethyl)-[1,1'-biphenyl]-2-carbonitrile. In the last
step, reduction of the nitro group in 14 with SnCl2 and ring closure was achieved in a one-pot
reaction. This approach, however, was not suitable for the syntheses of the other compounds,
since the overall yield decreased, due to poor solubility of the 2-nitro derivatives.
15b-d,f-g,i,j as well as 16a-c,e were then tried to be hydrolyzed to the carboxylic acids 17a-
g,i,j applying either basic (KOH in EG and H2O at 160 °C, e: method C) or acidic conditions
(aqueous H2SO4 at 120 °C, e: method D) [23, 32]. The compounds 15d,g,j and 16a-c,e gave the
appropriate carboxylic acids 17a-e,g,j. The hydrolysis of 15b and 15c was incomplete and
stopped at the carboxamide intermediates (18b and 18c). In case of 15f and 15i, both the
carboxylic acids (17f, 17i) as well as the carboxamide derivatives (18f, 18i) have been formed,
which were separated by column chromatography. Ether cleavage of the 4-OCH3 derivative 17g
with BBr3 yielded 17h (Scheme 2) [33].
In a final step, the carboxamides 18a,d,e,g,h were synthesized according to Shan et al. [34]
from the respective carboxylic acids (17a,d,e,g,h) applying the coupling reagent PyBOP with
DIPEA and NH4Cl. 18j was generated by reduction of 18i (Scheme 2) with H2, Pd/C [30].
1
13
17a-j and 18a-j were characterized applying H- and C-NMR spectroscopy as well as high-
resolution mass spectrometry (HRMS). Purity was assessed by HPLC. The synthesis protocols as
well as the analytic data of intermediates and target compounds are submitted as supplementary
material.
2.2 Biological activity
2.2.1 Transactivation assay
To assess the potency (EC50 values) and the intrinsic activation (Amax) of the novel compounds
towards PPARγ, they were investigated in a dual-luciferase reporter assay using COS-7 cells
(Table 1). Herein the cells were transiently transfected with the plasmids pGal5-TK-pGL3,
pGal4-hPPARγDEF, and pRenilla-CMV. For reasons of comparability, Amax of each compound
at the concentration of 10 µM is listed in Table 1 (and discussed below), which corresponds to
that applied by PI-FACS analyses with K562-resistant cells. The concentration-activation curves
(Fig. 3), however, comprise all tested concentrations, including the highest of 20 µM.
The full agonist pioglitazone was used as the reference and its maximum activation was set to
100%. The half-maximal effective concentration (EC50 = 1.40 µM) was calculated from the
concentration-activation curve depicted in Fig. 3. In accordance with the literature, the EC50 of
telmisartan was 4.28 µM and its Amax was 59.1%, confirming its partial agonistic
pharmacological profile [18, 23]. Moreover, the data of the lead 4‘-((2-propyl-1H-
benzo[d]imidazol-1-yl)methyl)-[1,1’-biphenyl]-2-carboxylic
acid
(17a:
Amax = 58.5%,
EC50 = 7.22 µM) relates to previous findings [18, 23].
7

Table 1. Results of the transactivation assay in COS-7 cells using the hPPARγ-LBD.
R
Compd.
Amax [%]a,c
EC50 [µM]b,c
R
Compd.
Amax [%]a,c
EC50 [µM]b,c
H
17a
17b
17c
17d
17e
17f
17g
17h
17i
58.5 ± 3.9
44.2 ± 5.6
44.0 ± 5.0
44.9 ± 5.7
54.3 ± 3.0
29.5 ± 3.6
17.9 ± 2.7
14.4 ± 2.7
0.42 ± 0.09
15.7 ± 2.6
100
7.22 ± 1.37
8.66 ± 1.49
8.67 ± 0.87
8.25 ± 0.55
11.1 ± 0.3
12.1 ± 1.4
n.d.
H
18a
18b
18c
18d
18e
18f
18g
18h
18i
39.5 ± 5.3
33.3 ± 3.9
21.1 ± 1.6
20.8 ± 0.8
19.6 ± 0.6
19.5 ± 3.6
13.3 ± 0.4
20.5 ± 0.9
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
F
F
Cl
Cl
Br
Br
CH3
CF3
OCH3
OH
NO2
NH2
CH3
CF3
OCH3
OH
NO2
NH2
n.d.
n.d.
7.15 ± 1.60 n.d.
25.1 ± 4.8 n.d.
n.d.
17j
18j
Pioglitazone
Telmisartan
1.40 ± 0.18
4.28 ± 0.42
59.1 ± 2.2
a The intrinsic activation (Amax) is the response of the compounds related to the maximum activation of pioglitazone (100%,
10 µM).
b Potency (EC50 values) was calculated based on the concentration-activation curves (Fig. 3).
c Data represent the mean ± SD of ≥ 3 independent experiments (n.d.: not determinable).
The introduction of substituents in position 4 of the benzimidazole core modulated the activity,
which is in accordance with the assessment of Goebel et al. [18]. Hence, this position can be
utilized for the synthesis of partial PPARγ agonists with different potency. The carboxylic acids
17a-j can be classified into three groups regarding their intrinsic activity. Derivatives bearing 4-F
(17b, 44.2%), 4-Cl (17c, 44.0%), 4-Br (17d, 44.9%), or 4-CH3 (17e, 54.3%) substituents at the
benzimidazole ring, activated PPARγ in a moderately diminished intensity compared to
telmisartan (59.1%) or the lead (17a, 58.5%). A stronger decrease of Amax was observed in case
of the 4-CF3 derivative (17f) with an Amax of 29.5%. The potency of 17a-d with
EC50 = 7.22-8.67
µM was less than that of telmisartan and it was even more reduced in case of
17e-f (EC50 = 11.1-12.1
µM). Finally, the 4-OCH3 (17g), the 4-NH2 (17j), and the 4-OH (17h)
8

moieties led to minimal effects on the receptor with Amax values of 17.9%, 15.7%, and 14.4%,
respectively. The 4-NO2 compound (17i) did not show any activation at all. These data clearly
demonstrate that it is possible to modulate the PPARγ activity by simple modification of the
substituent at position 4.
The transformation of the carboxylic group to a carboxamide group notably changed the
pharmacological profile. The curves of the transactivation assay (Fig. 3) were shifted towards
lower PPARγ activity in case of the unsubstituted derivative (18a) and all the compounds
bearing lipophilic substituents (18b-g). They activated the receptor to only 20-40% at 10 µM.
Compounds with more hydrophilic groups such as 4-OH (18h), 4-NO2 (18i), or 4-NH2 (18j)
retained their low receptor activation upon derivatization. The maximum effects of the
carboxamide derivatives were not achieved even at 20
failed, because of insufficient water solubility.
µM. The use of higher concentrations
Nevertheless, the concentration-activity curves suggest that the pharmacological profile of
most of the compounds corresponds more to the one of an impeded instead of a partial agonist.
The profile of partial agonists is similar to telmisartan and the derivatives 17a-d, which reached
saturation at 20
µM (Fig. 3). Therefore, it cannot be excluded that the compounds exert beside
agonistic also possibly antagonistic activity [35], which however was not evaluated here.
9

Fig. 3. Concentration-activation curves of the dual-luciferase reporter assay with respect to PPARγ. COS-7 cells were transiently
transfected with the plasmids pGal5-TK-pGL3 as well as pGal4-hPPARγDEF and subsequently stimulated with pioglitazone ( ),
telmisartan ( ), or the novel compounds. After 39 h of incubation, firefly-luciferase was measured and normalized with the
activity of the co-transfected pRenilla-CMV. The data represent the mean ± SD of ≥ 3 independent experiments with three
replicates each.
10

2.2.2 Induction of cell death
Since pioglitazone was successfully investigated on therapy-resistant leukemia cells [9], the
suitability of the new compounds to overcome the imatinib resistance in the K562-resistant cell
line was evaluated. This cell line was initially described as doxorubicin-resistant subclone of
K562 cells [21]. Based on our own data, it is also highly resistant to imatinib treatment, as
demonstrated by PI-FACS analyses (Fig. 4). Thus, K562 and K562-resistant cells were treated
with increasing concentrations of imatinib for 72 h and analyzed comparatively. At a
concentration of 1 µM, which was applied in further investigations, imatinib induced a cell death
rate of 46% in K562 cells and 15% in K562-resistant cells, respectively.
Fig. 4. K562 ( ) and K562-resistant ( ) cells treated with increasing concentrations of imatinib. The cells were incubated for
72 h and subjected to PI-FACS analyses to detect dead cells. The data represent the mean ± SEM of ≥ 3 independent
experiments.
To analyze the efficacy of the compounds, K562-resistant cells were treated with 10 µM of
pioglitazone, 17a-j, and 18a-j for 72 h, respectively. Pioglitazone as well as the compounds 17a-
j and 18g-j showed the same cell death rate as observed in control cells (CTR, Fig. 5A), while
18a-f induced slightly but less than 20% cell death as determined by PI-FACS analyses.
Application of pioglitazone and the telmisartan derivatives 17a-j as well as 18g-j at inactive
concentrations (10 µM, compare Fig. 5A) only marginally sensitized the cells to imatinib
treatment (1 µM, Fig. 5B). In contrast, co-application of imatinib with 18a-f induced 62-78%
cell death (Fig. 5B).
These findings clearly demonstrate that the novel compounds can act as effective cell death
sensitizers to imatinib.
11

Fig. 5. K562 cells resistant to imatinib were treated with pioglitazone (PIO), the compounds 17a-j, and 18a-j alone (10 µM each,
5A) or in combination with 1 µM of imatinib (5B). CTR represents the effects of the solvent DMSO (10 µM). After 72 h, the
cells were subjected to PI-FACS analyses to detect dead cells. The data represent the mean + SEM of ≥ 3 independent
experiments.
To get more information about the efficacy of 18a-f in sensitizing K562-resistant cells to
imatinib treatment, the cells were incubated with 1 µM of imatinib in combination with
increasing concentrations (0.1-10 µM) of the compounds for 72 h (Fig. 6). As a first important
result, pioglitazone showed no sensitizing effect at all tested concentrations. The same was
observed for 18a-f at low concentrations up to 2.5 µM. At 5 µM, 18a, 18c, and 18d induced
12

16.0 ± 4.6%, 15.3 ± 4.1%, and 21.4 ± 5.0% gain in cell death rate compared to imatinib alone.
The compounds 18b, 18e, and 18f were slightly less effective at 5 µM (8.1 ± 2.7%, 11.2 ± 4.2%,
and 7.1 ± 1.2%, respectively; Fig. 6). At 10 µM, the results depicted in Fig. 5 were confirmed.
Fig. 6. K562-resistant cells were treated with 1 µM of imatinib in combination with increasing concentrations (0.1-10 µM) of
pioglitazone (PIO) and the compounds 18a-f, respectively, for 72 h. PI-FACS analyses were performed to detect dead cells.
Shown are mean values + SEM of ≥ 4 independent experiments as gain in cell death [%] over K562-resistant cells treated with
imatinib alone.
2.2.3 Cytotoxicity assay
All compounds were also examined in the human non-malignant cells HS-5 (Fig. 7A) as well
as the chronic myelogenous leukemia cells K562 (Fig. 7B) by PI-FACS analyses to exclude an
induction of cell death by the compounds themselves. As a result, neither 17a-j nor 18a-j caused
cell death at the applied concentrations (10 µM, Fig. 7). The compounds can therefore be stated
as non-cytotoxic.
13

Fig. 7. Induction of cell death in HS-5 (Fig. 7A) or K562 (Fig. 7B) cells, treated with DMSO (CTR), 17a-j, or 18a-j, 10 µM
each. The mean values + SEM of ≥ 3 independent experiments are shown.
2.2.4 Metabolic activity assay
The non-malignant COS-7 cell line, which was applied as host for the transactivation assay,
was treated with 17a-j as well as 18a-j to exclude unselective cytotoxic properties that might
influence the results of the luciferase assay. The metabolic activity was assessed in a modified
MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) colorimetric assay. The
transformation of tetrazolium salts to their purple-colored formazan derivatives is an indication
of cell viability. As this process is based on NAD(P)H-dependent cellular oxidoreductases, the
number of cells with functional mitochondria can be determined [36]. Only actively growing
cells yield high metabolic activity, while cells exposed to cytotoxic molecules have diminished
activity.
This experiment reveals that none of the telmisartan derivatives influenced the proliferation
and viability of the COS-7 cells (Fig. 8). Even at the highest concentration used (20 µM) no
14

effects on the metabolic activity were observed. The achieved low Amax of some compounds in
the luciferase assay, especially within the carboxamide series, is therefore not related to an
impact on cell growth.
Fig. 8. Metabolic activity of COS-7 cells treated with DMSO (CTR), pioglitazone (PIO), telmisartan (TELMI), the compounds
17a-j, or 18a-j, respectively, at 20 µM. The mean values + SD of ≥ 3 independent experiments are shown.
2.2.5 Immunoblot analyses
Prost et al. [9] postulated that PPARγ modulators eradicate resistant cells via the STAT5
pathway. Glitazones such as pioglitazone decrease the expression of STAT5, which is a key
guardian of the quiescence and formation of CML SCs. Thus, the impact of selected compounds
on the expression of STAT5 was investigated by immunoblotting.
Remarkably, the compounds 18a as well as 18d efficiently repressed the STAT5 protein
expression in imatinib-treated K562-resistant cells (Fig. 9), which implicates that these
compounds might eradicate SCs. Clearly, the detailed molecular mechanism has to be defined in
ongoing in-depth studies.
Fig. 9. Immunoblot analyses of STAT5 expression in K562-resistant cells treated with 1 µM of imatinib alone (CTR) or in
combination with 10 µM of 18a or 18d for 24 h. GAPDH served as loading control.
15

2.3 Discussion
In continuation of our structure-activity relationship (SAR) study about telmisartan derivatives,
we focused our attention on position 4 of the benzimidazole core of the lead 4‘-((2-propyl-1H-
benzo[d]imidazol-1-yl)methyl)-[1,1’-biphenyl]-2-carboxylic acid. Various substituents were
introduced (F, Cl, Br, CH3, CF3, OCH3, OH, NO2, NH2) and the PPARγ activation was
determined in a luciferase assay using COS-7 cells transiently transfected with the plasmids
pGal5-TK-pGL3, pGal4-hPPARγDEF, and pRenilla-CMV.
In this assay, telmisartan showed the profile of a partial agonist, as it did not fully elicit the
response of pioglitazone (100%). The maximum activation of 78% was reached at the highest
concentration used (20 µM). The lead 17a caused a comparable concentration-activity curve.
Halide-substituted compounds (17b-d) were weaker partial agonists with a slightly lower
potency. The other compounds did not reach maximum activation at the applied concentrations.
Within the carboxylic acid series the potency decreased: 17a (4-H) = 17e (4-CH3) > 17b (4-
F) = 17c (4-Cl) = 17d (4-Br) > 17f (4-CF3) = 17g (4-OCH3) = 17h (4-OH) = 17j (4-NH2) >> 17i
(4-NO2). These data clearly show that substituents at position 4 modulate the activation of
PPARγ. Hydrophobic substituents had less influence, while the introduction of polar groups was
associated with a decrease of transactivation. The 4-NO2 derivative (17i) was even inactive.
In a previous study, it was determined that both, the position of the COOH group at the
biphenyl moiety of the lead as well as its transformation to a CONH2 group, change the
pharmacological profile. In the latter case, the potency (EC50) was nearly retained, whereas the
efficiency (Amax) markedly decreased [22]. Once the lead was substituted with a hydrophobic
group in position 4 of the benzimidazole (compare 17a-g and 18a-g) both potency and efficiency
were strongly reduced upon carboxamide formation. On the contrary, Amax was quite comparable
for the compounds 17h-j and 18h-j bearing more hydrophilic substituents (OH, NH2, NO2). The
EC50 values of the more hydrophilic compounds in each series could not be calculated, as the
concentration-activity curves did not reach the maximum. Escalation of the concentration was
impossible due to insolubility of the compounds in the medium at higher concentrations.
Another very interesting finding of the former study was the increased growth inhibitory
effects of the unsubstituted benzimidazole derivatives with a [1,1’-biphenyl]-2-carboxamide
moiety against COS-7 cells, dependent on the alkyl chain at position 2. These data indicate the
essential role of hydrophobicity concerning the cytotoxicity of PPARγ agonists [22]. Therefore,
all 4-substituted compounds were evaluated by an MTT assay with COS-7 cells and by PI-FACS
analyses with HS-5 as well as K562 cells in the current study. Notably, neither the carboxamides
nor the carboxylic acids affected cell growth or possessed cytotoxic properties.
The antitumor activity of pioglitazone as a representative of PPARγ agonists was extensively
analyzed. Rousselot et al. [8] found increased CMR to imatinib in CML patients upon
pioglitazone co-application. This finding inspired us to determine the effects of 17a-j and 18a-j
on K562-resistant cells. These cells showed high resistance to imatinib up to a concentration of
2 µM. All compounds were administered together with imatinib (1 µM) at inactive
concentrations (10 µM). Surprisingly, the carboxamides 18a-f strongly sensitized the K562-
16

resistant cells to imatinib treatment in a concentration-dependent manner, though none of the
compounds reduced cell growth or induced cell death per se. Pioglitazone was inactive in the
same concentration range.
The observed effect was shown to be depended on the substituents at position 4 of the
benzimidazole core and the COOH derivatization. While compounds 18a-f (H, F, Cl, Br, CH3)
sensitized resistant cells, the OCH3 (18g), OH (18h), NH2 (18i), and NO2 (18j) derivatives of the
carboxamide series as well as all the carboxylic acids 17a-j were ineffective at the applied
concentration of 10 µM. A correlation with the PPARγ activation cannot be deduced, but its
participation in the mode of action cannot be excluded. Anyway, the reduction of PPARγ
activation can also be of great advantage due to the reduction of TZD-like side effects.
Regarding the mode of action, several points have to be considered. One aspect, which needs
to be discussed, is the modulation of the imatinib accumulation in CML cells via the human
organic cation transporter 1 (hOCT1), which is described to be upregulated when applying PPAR
modulators [12, 37]. Another point is the interference with the STAT5 pathway. The observed
downregulation of the STAT5 expression, exemplarily demonstrated for 18a and 18d, suggests
that this pathway is involved in the modulation of cell death caused by a potential eradication of
the SC pool. Ongoing in-depth studies will define the detailed molecular mechanism of how the
compounds sensitize K562-resistant cells to imatinib.
3 Conclusions
In conclusion, modifications of the 4‘-((2-propyl-1H-benzo[d]imidazol-1-yl)methyl)-[1,1’-
biphenyl]-2-carboxylic acid in position 4 of the benzimidazole modulated the PPARγ agonistic
effects. Derivatization of the 2-COOH to a 2-CONH2 group reduced the agonism. Yet,
carboxamides 18a-f with H, F, Cl, Br, CH3, or CF3 substituents are sensitizers to imatinib
treatment in K562-resistant cells, without having cytotoxic properties themselves. The mode of
action is presumably correlated to STAT5, since downregulation of this pro-survival
transcription factor upon compound treatment was demonstrated by immunoblot analyses. Our
study clearly shows that PPARγ modulators derived from telmisartan are promising cell death
modulators to overcome imatinib resistance.
4 Experimental section
4.1 Chemistry
4.1.1 General materials and methods
The compounds 1-4, 9a, 9e, and 11, all reagents as well as other chemicals were purchased
from Sigma-Aldrich, TCI Chemicals, or Alfa Aesar and used without further purification.
Dichloromethane (DCM), ethyl acetate (EA), ethanol (EtOH), and methanol (MeOH) were
distilled before utilization, whereas petroleum ether (PE) was directly used. Ethylene glycol
(EG), monochlorobenzene (MCB), N,N-diisopropylethylamine (DIPEA), N,N-dimethyl
17

formamide (DMF), diisopropyl ether (DIPE), isopropanol, and tetrahydrofuran (THF) were
purchased in appropriate quality. Column chromatography was either performed applying classic
standard procedures or by medium pressure liquid chromatography. Thereto, an Isolera One 3.0
Flash purification system (Biotage) was employed. In all cases, silica gel 60 (particle size 40-
63 µm, 230-240 mesh) served as stationary phase. Thin-layer chromatography was carried out
using Polygram SIL G/UV254 polyester foils covered with a 0.2 mm layer of silica gel as well as
a fluorescence indicator (Macherey-Nagel) and were visualized with UV light (254 or 366 nm).
Nuclear magnetic resonance spectra (NMR) were recorded using a 200 MHz Gemini (now
Agilent), 400 MHz Avance 4 Neo (Bruker), 500 MHz Direct Drive 2 (Agilent), 600 MHz
Avance II (Bruker), or 700 MHz Avance 4 Neo (Bruker) spectrometer. Deuterated chloroform
(CDCl3), dimethyl sulfoxide (DMSO-d6), acetone (acetone-d6), methanol (CD3OD), and water
(D2O) (Eurisotop and Alfa Aesar) were used as solvents (all from Eurisotop or Alfa Aesar).
Chemical shifts (δ) were referenced to the solvent peak or tetramethylsilane (TMS) as internal
standard and are given in parts per million (ppm). Coupling constants (J) are reported in Hertz
(Hz). Signals are described as follows: s = singlet, br s = broad singlet, d = doublet, dd = doublet
of doublets, t = triplet, m = multiplet. HRMS was conducted with an Orbitrap Elite system
(Thermo Fisher Scientific). HPLC was used for the determination of purity. A Shimadzu Nexera-
i LC-2040C-3D and a Shimadzu Prominence HPLC with the autosampler SIL-20A HT, the
column oven CTO-10AS VP, the degasser DGU-20A, the detector SPD-M20A, and the pumps
LC-20AD were employed. An RP-18 column (dimension 125 × 4 mm, 5 µm particle size,
Knauer) was used and the chromatograms were analyzed with the program LabSolutions 5.86
(Shimadzu). The purity of ≥ 95% was assured for all compounds (except where noted).
4.1.2 General procedure for the hydrolysis of the carbonitriles and methyl esters
Method C
The respective carbonitrile or methyl ester (1 eq) was suspended in a solution of finely ground
potassium hydroxide (2 eq) in EG (0.5-1.5 ml/mmol) and a few drops of water were added. The
reaction mixture was heated to 160 °C for 12-72 h. Subsequently, the solution was cooled and
diluted with water to double the volume. The solution was acidified to a pH of 3-5 with 6 N
hydrochloric acid and stirred for several minutes. The formed precipitate was either sucked off,
washed with cold water and dried, or the solution was extracted with DCM or EA (3×), the
organic phases were combined, washed with brine, dried over anhydrous sodium sulfate and
filtered. The solvent was removed under reduced pressure. The resulting product was purified by
column chromatography (DCM/MeOH, 95:5) and recrystallized from MeOH.
Method D
The appropriate carbonitrile (1 eq) was suspended in water (~2 ml/mmol) and concentrated
sulfuric acid (~1 ml/mmol) before it was heated to 100 °C for 24 h. Then, the solution was
cooled and diluted with water to double the volume. After that, it was basified to a pH of 5-6
with a saturated aqueous solution of sodium bicarbonate and extracted with DCM (3×). The
organic layers were combined, washed with brine, dried over anhydrous sodium sulfate and
18

filtered. The solvent was removed under reduced pressure. The crude solid was purified by
column chromatography (DCM/MeOH, 95:5 and EA/MeOH, 8:2) and recrystallized from DIPE
or MeOH.
4’-((2-Propyl-1H-benzo[d]imidazole-1-yl)methyl)-[1,1’-biphenyl]-2-carboxylic acid (17a)
Method C: from 16a (0.50 g, 1.3 mmol) with potassium hydroxide (0.15 g, 2.6 mmol) in EG
(2 ml). Colorless solid, yield: 75%. 1H-NMR (400 MHz, DMSO-d6): δ 12.80 (br s, 1H, COOH),
3
4
7.67 (dd, 1H, J = 7.6 Hz, J = 1.5 Hz, H3), 7.62-7.56 (m, 1H, H5), 7.54-7.47 (m, 2H, H4’’,
H7’’), 7.44-7.38 (m, 1H, H4), 7.34-7.27 (m, 3H, H6, H6’, H2’), 7.20-7.14 (m, 2H, H6’’, H5’’),
3
7.11 (d, 2H, 3J = 8.2 Hz, H3’, H5’), 5.53 (s, 2H, NCH2), 2.84 (t, 2H, J = 7.5 Hz, CH2CH2CH3),
3
1.87-1.71 (m, 2H, CH2CH2CH3), 0.96 (t, 3H, J = 7.4 Hz, CH2CH2CH3). 13C-NMR (400 MHz,
DMSO-d6): δ 169.9, 155.4, 140.9, 140.8, 135.8, 134.9, 132.7, 131.4, 130.9, 129.6, 129.2, 127.9,
126.9, 123.3, 123.2, 117.8, 111.5, 46.6, 28.5, 20.7, 14.2. HRMS: m/z calculated for C24H22N2O2
[M+H]+: 371.1754, found: 371.1773.
4’-((4-Fluoro-2-propyl-1H-benzo[d]imidazole-1-yl)methyl)-[1,1’biphenyl]-2-carboxylic acid
(17b)
Method C: from 16b (0.01 g, 0.03 mmol) with potassium hydroxide (0.003 g, 0.05 mmol) in
1
EG (0.1 ml). Colorless solid, yield: 16%. H-NMR (600 MHz, CD3OD): δ 7.78 (dd, 1H,
4
3J = 7.8 Hz, J = 1.5 Hz, H3), 7.54-7.51 (m, 1H, H5), 7.44-7.39 (m, 1H, H4), 7.35-7.30 (m, 3H,
3
4
H6, H2’, H6’), 7.25 (dd, 1H, J = 8.1 Hz, J = 0.9 Hz, H7’’), 7.22-7.17 (m, 1H, H6’’), 7.12 (d,
3
2H, J = 8.1 Hz, H3’, H5’), 7.00-6.95 (m, 1H, H5’’), 5.55 (s, 2H, NCH2), 2.92 (t, 2H,
3J = 7.8 Hz, CH2CH2CH3), 1.84-1.77 (m, 2H, CH2CH2CH3), 1.01 (t, 3H, 3J = 7.4 Hz,
CH2CH2CH3). 13C-NMR (600 MHz, CD3OD): δ 157.8, 155.3, 153.6, 142.6, 136.5, 133.5, 132.1,
131.8, 130.6, 130.2, 129.6, 128.7, 128.4, 127.2, 126.9, 124.3, 115.5, 108.8, 107.8, 75.4, 47.9,
30.1, 22.3, 14.1. HRMS: m/z calculated for C24H21FN2O2 [M+H]+: 389.1660, found: 389.1694.
4’-((4-Chloro-2-propyl-1
(17c)
H-benzo[d]imidazole-1-yl)methyl)-[1,1’biphenyl]-2-carboxylic acid
Method C: from 16c (0.10 g, 0.2 mmol) with potassium hydroxide (0.03 g, 0.5 mmol) in EG
(0.5 ml). Colorless solid, yield: 59%. 1H-NMR (600 MHz, DMSO-d6): δ 7.67 (d, 1H,
3
3J = 7.7 Hz, H3), 7.56-7.51 (m, 1H, H5), 7.48-7.40 (m, 2H, H4), 7.31 (d, 1H, J = 7.7 Hz, H6),
7.29-7.23 (m, 3H, H2’, H6’, H7’’), 7.20-7.16 (m, 1H, H6’’), 7.11 (d, 2H, 3J = 7.9 Hz, H3’, H5’),
3
5.53 (s, 2H, NCH2), 2.86 (t, 2H, J = 7.6 Hz, CH2CH2CH3), 1.78-1.69 (m, 2H, CH2CH2CH3),
3
13
0.93 (t, 3H, J = 7.4 Hz, CH2CH2CH3). C-NMR (600 MHz, DMSO-d6): δ 170.1, 156.9, 140.7,
140.6, 139.5, 136.8, 135.9, 132.7, 131.4, 130.9, 129.5, 129.2, 127.9, 126.8, 125.8, 123.3, 122.8,
121.9, 109.9, 55.1, 46.7, 28.9, 20.9, 14.2. HRMS: m/z calculated for C24H21ClN2O2 [M+H]+:
405.1364, found: 405.1375.
19

4’-((4-Bromo-2-propyl-1H-benzo[d]imidazol-1-yl)methyl)-[1,1’-biphenyl]-2-carboxylic acid
(17d)
Method C: from 15d (0.18 g, 0.42 mmol) with potassium hydroxide (0.05 g, 0.84 mmol) in EG
1
(0.6 ml). Colorless solid, yield: 94%. H-NMR (500 MHz, DMSO-d6): δ 12.73 (br s, 1H,
COOH), 7.70 (d, 2H, 3J = 7.6 Hz, H3), 7.57-7.50 (m, 2H, H5, H7’’), 7.46-7.36 (m, 2H, H4, H6),
7.35-7.25 (m, 3H, H2’, H6’, H5’’), 7.16-7.08 (m, 3H, H3’, H5’, H6’’), 5.56 (s, 2H, NCH2), 2.87
3
3
(t, 2H, J = 7.6 Hz, CH2CH2CH3), 1.86-1.71 (m, 2H, CH2CH2CH3), 0.96 (t, 3H, J = 7.4 Hz,
CH2CH2CH3). 13C-NMR (500 MHz, DMSO-d6): δ 169.5, 156.2, 140.7, 140.4, 140.2, 136.0,
135.6, 132.4, 130.8, 130.4, 129.1, 128.7, 127.3, 126.2, 124.3, 123.0, 122.6, 121.3, 111.6, 110.0,
59.8, 46.2, 28.6, 20.5, 13.8. HRMS: m/z calculated for C24H21BrN2O2 [M+H]+: 449.0859, found:
449.0879.
4’-((4-Methyl-2-propyl-1
(17e)
H-benzo[d]imidazol-1-yl)methyl)-[1,1’-biphenyl]-2-carboxylic acid
Method C: from 16e (1.0 g, 2.7 mmol) with potassium hydroxide (0.92 g, 16.4 mmol) in EG
1
(3.2 ml). Colorless solid, yield: 63%. H-NMR (600 MHz, DMSO-d6): δ 7.65 (d, 1H,
3J = 7.5 Hz, H3), 7.52-7.46 (m, 1H, H5), 7.42-7.37 (m, 1H, H4), 7.32-7.26 (m, 4H, H6, H2’,
3
H6’, H7’’), 7.09 (d, 2H, J = 8.1 Hz, H2’, H6’), 7.06-7.02 (m, 1H, H6’’), 6.96 (d, 1H,
3
3J = 7.2 Hz, H5’’), 5.50 (s, 2H, NCH2), 2.83 (t, 2H, J = 7.6 Hz, CH2CH2CH3), 2.52 (s, 3H,
3
13
CH3), 1.83-1.72 (m, 2H, CH2CH2CH3), 0.96 (t, 3H, J = 7.3 Hz, CH2CH2CH3). C-NMR (600
MHz, DMSO-d6): δ 154.2, 141.7, 140.2, 136.0, 134.9, 130.3, 128.9, 128.7, 127.9, 127.2, 126.1,
121.7, 121.6, 107.7, 59.8, 45.8, 40.1, 28.7, 20.7, 16.4, 13.9. HRMS: m/z calculated for
C25H24N2O2 [M+H]+: 385.1911, found: 385.1933.
4’-((2-Propyl-4-(trifluoromethyl)-1H-benzo[d]imidazol-1-yl)methyl)-[1,1’-biphenyl]-2-
carboxylic acid (17f) and 4’-((2-propyl-4-(trifluoromethyl)-1
yl)methyl)-[1,1’-biphenyl]-2-carboxamide (18f)
H-benzo[d]imidazol-1-
Method D: from 15f (0.19 g, 0.5 mmol), water (1 ml) and concentrated sulfuric acid (0.4 ml).
Recrystallization from DIPE.
1
3
17f: Colorless solid, yield: 11%. H-NMR (500 MHz, CD3OD): δ 7.77 (d, 1H, J = 7.6 Hz,
3
H3), 7.68 (d, 1H, J = 8.2 Hz, H5’’), 7.58-7.49 (m, 2H, H5, H7’’), 7.44-7.39 (m, 1H, H4), 7.37-
7.30 (m, 4H, H6, H2’, H6’, H6’’), 7.12 (d, 2H, 3J = 7.9 Hz H3’, H5’), 5.62 (s, 2H, NCH2), 2.99
3
3
(t, 2H, J = 7.8 Hz, CH2CH2CH3), 1.84-1.71 (m, 2H, CH2CH2CH3), 1.01 (t, 3H, J = 7.4 Hz,
CH2CH2CH3). 13C-NMR (600 MHz, CD3OD): δ 173.2, 159.4, 142.7, 142.2, 139.8, 137.7, 136.4,
134.7, 131.7, 131.6, 130.9, 130.3, 128.3, 127.4, 127.2, 126.5, 124.7, 122.9, 120.8, 120.5, 115.8,
47.9, 30.1, 22.9, 14.2. HRMS: m/z calculated for C25H21F3N2O2 [M+H]+: 439.1628, found:
439.1633.
1
3
18f: Colorless solid, yield: 20%. H-NMR (500 MHz, CD3OD): δ 7.65 (d, 1H, J = 8.2 Hz,
H3), 7.56-7.45 (m, 3H, H5, H5’’, H7’’), 7.44-7.30 (m, 5H, H4, H6, H2’, H6’, H6’’), 7.13 (d, 2H,
3J = 8.0 Hz, H3’, H5’), 5.60 (s, 2H, NCH2), 2.97 (t, 2H, 3J = 7.8 Hz, CH2CH2CH3), 1.84-1.71 (m,
20

3
2H, CH2CH2CH3), 1.01 (t, 3H, J = 7.4 Hz, CH2CH2CH3). 13C-NMR (500 MHz, CD3OD): δ
175.5, 159.3, 141.7, 140.5, 139.8, 137.7, 137.3, 136.7, 131.3, 131.1, 130.4, 128.8, 128.5, 127.4,
127.1, 126.7, 124.5, 122.9, 120.9, 120.5, 120.5, 115.7, 47.9, 30.8, 30.1, 22.9, 14.2. HRMS: m/z
calculated for C25H22F3N3O [M+H]+: 438.1788, found: 438.1780.
4’-((4-Methoxy-2-propyl-1
acid (17g)
H-benzo[d]imidazol-1-yl)methyl)-[1,1’-biphenyl]-2-carboxylic
Method C: from 15g (0.49 g, 1.3 mmol) with potassium hydroxide (0.14 g, 2.5 mmol) in EG
1
(1.0 ml). Colorless solid, yield: 27%. H-NMR (400 MHz, DMSO-d6): δ 7.66 (d, 1H,
3J = 7.6 Hz, H3), 7.54-7.46 (m, 1H, H5), 7.44-7.36 (m, 1H, H4), 7.36-7.24 (m, 3H, H6, H2’,
H6’), 7.13-7.03 (m, 4H, H3’, H5’, H6’’, H7’’), 6.73-6.65 (m, 1H, H5’’), 5.49 (s, 2H, NCH2),
3
3.93 (s, 3H, OCH3), 2.80 (t, 2H, J = 7.5 Hz, CH2CH2CH3), 1.83-1.70 (m, 2H, CH2CH2CH3),
3
0.95 (t, 3H, J = 7.4 Hz, CH2CH2CH3). 13C-NMR (400 MHz, DMSO-d6): δ 153.4, 150.641,
140.2, 136.9, 135.9, 132.3, 130.3, 128.9, 128.7, 127.2, 126.1, 122.5, 103.2, 103.1, 55.5, 45.9,
28.5, 20.4, 13.8. HRMS: m/z calculated for C25H24N2O3 [M+H]+: 401.1860, found: 401.1888.
4’-((4-Nitro-2-propyl-1
(17i) and 4’-((4-nitro-2-propyl-1
carboxamide (18i)
H
-benzo[
d
]imidazole-1-yl)methyl)-[1,1’-biphenyl]-2-carboxylic acid
H
-benzo[ ]imidazole-1-yl)methyl)-[1,1’-biphenyl]-2-
d
Method D: from 15i (0.18 g, 0.45 mmol), water (1 ml) and concentrated sulfuric acid (0.4 ml).
Recrystallization from MeOH.
1
17i: Colorless solid, yield: 8%. H-NMR (400 MHz, DMSO-d6): δ 12.77 (br s, 1H, COOH),
3
4
3
8.01 (dd, 2H, J = 8.0 Hz, J = 2.6 Hz, H5’’, H7’’), 7.65 (d, 1H, J = 7.6 Hz, H3), 7.53-7.25 (m,
3
6H, H4, H5, H6, H2’, H6’, H6’’), 7.13 (d, 2H, J = 7.8 Hz, H3’, H5’), 5.67 (s, 2H, NCH2), 2.95
3
3
(t, 2H, J = 7.6 Hz, CH2CH2CH3), 1.86-1.72 (m, 2H, CH2CH2CH3), 0.98 (t, 3H, J = 7.3 Hz,
CH2CH2CH3). 13C-NMR (400 MHz, DMSO-d6): δ 159.5, 140.5, 139.8, 138.2, 137.9, 135.8,
135.1, 130.2, 128.9, 128.8, 127.2, 126.2, 121.3, 118.2, 117.2, 46.3, 28.7, 20.6, 13.8. HRMS: m/z
calculated for C24H21N3O4 [M+H]+: 416.1605, found: 416.1637.
18i: Colorless solid, yield: 16%. 1H-NMR (400 MHz, acetone-d6): δ 7.99 (dd, 1H, 3J = 8.0 Hz,
4J = 1.0 Hz, H5’’), 7.89 (dd, 1H, J = 8.0 Hz, J = 1.0 Hz, H7’’), 7.52 (dd, 1H, J = 7.5 Hz,
4J = 1.5 Hz, H3), 7.48-7.31 (m, 6H, H4, H5, H6, H2’, H6’, H6’’), 7.18 (d, 2H, 3J = 8.4 Hz, H3’,
H5’), 6.80 (br s, 1H, CONH), 6.52 (br s, 1H, CONH), 5.68 (s, 2H, NCH2), 2.98 (t, 2H,
3J = 7.6 Hz, CH2CH2CH3), 1.95-1.85 (m, 2H, CH2CH2CH3), 1.02 (t, 3H, 3J = 7.4 Hz,
CH2CH2CH3). 13C-NMR (400 MHz, acetone-d6): δ 170.8, 159.5, 140.6, 138.9, 138.6, 137.2,
137.1, 136.5, 135.4, 130.0, 129.4, 129.2, 127.9, 127.2, 126.4, 121.0, 117.9, 116.3, 46.7, 20.7,
13.3. HRMS: m/z calculated for C24H22N4O3 [M+H]+: 415.1765, found: 415.1791.
3
4
3
21

4’-((4-Amino-2-propyl-1H-benzo[d]imidazole-1-yl)methyl)-[1,1’-biphenyl]-2-carboxylic
acid (17j)
Method C: from 15j (0.25 g, 0.7 mmol) with potassium hydroxide (0.08 g, 1.4 mmol) in EG
1
(0.7 ml). Recrystallization from MeOH. Colorless solid, yield: 20%. H-NMR (600 MHz,
DMSO- d6): δ 7.69 (dd, 1H, 3J = 7.6 Hz, 4J = 1.4 Hz, H3), 7.56-7.51 (m, 1H, H5), 7.46-7.40 (m,
1H, H4), 7.33 (dd, 1H, 3J = 7.7 Hz, 4J = 1.2 Hz, H6), 7.28 (d, 2H, 3J = 8.2 Hz, H2’, H6’), 7.11 (d,
3
3
4
2H, J = 8.0 Hz, H3’, H5’), 6.87-6.83 (m, 1H, H6’’), 6.64 (dd, 1H, J = 8.0 Hz, J = 0.9 Hz,
3
4
H7’’), 6.35 (dd, 1H, J = 7.7 Hz, J = 0.9 Hz, H5’’), 5.42 (s, 2H, NCH2), 5.17 (br s, 2H, NH2),
2.80 (t, 2H, 3J = 7.6 Hz, CH2CH2CH3), 1.80-1.70 (m, 2H, CH2CH2CH3), 0.96 (t, 3H, 3J = 7.4 Hz,
CH2CH2CH3). 13C-NMR (600 MHz, DMSO-d6): δ 169.6, 151.9, 140.3, 139.9, 139.5, 136.3,
135.9, 130.9, 130.7, 130.4, 129.0, 128.6, 127.3, 126.2, 122.8, 114.6, 104.6, 98.1, 45.8, 28.6, 20.8,
13.9. HRMS: m/z calculated for C24H23N3O2 [M+H]+: 386.1863, found: 386.1876.
4’-((4-Fluoro-2-propyl-1H-benzo[d]imidazole-1-yl)methyl)-[1,1’-biphenyl]-2-carboxamide
(18b)
Method C: from 15b (0.08 g, 0.2 mmol) with potassium hydroxide (0.03 g, 0.4 mmol) in EG
(0.3 ml). Colorless solid, yield: 40%. 1H-NMR (500 MHz, CD3OD): δ 7.58-7.31 (m, 6H, H3, H4,
H5, H6, H2’, H6’), 7.27-7.05 (m, 4H, H3‘, H5‘, H6‘‘, H7‘‘), 7.01-6.90 (m, 1H, H5‘‘), 5.54 (s,
2H, NCH2), 2.91 (t, 2H, 3J = 7.7 Hz, CH2CH2CH3), 1.88-1.73 (m, 2H, CH2CH2CH3), 1.00 (t, 3H,
13
3J = 7.4 Hz, CH2CH2CH3). C-NMR (500 MHz, CD3OD): δ 175.5, 157.7, 155.5, 153.5, 141.6,
140.5, 139.5, 137.3, 136.8, 131.3, 131.1, 130.3, 128.8, 128.5, 127.4, 124.3, 124.2, 108.8, 108.7,
107.7, 47.9, 30.1, 22.3, 14.1. HRMS: m/z calculated for C24H22FN3O [M+H]+: 388.1820, found:
388.1813.
4’-((4-Chloro-2-propyl-1
(18c)
H-benzo[d]imidazol-1-yl)methyl)-[1,1’-biphenyl]-2-carboxamide
Method C: from 15c (0.56 g, 1.5 mmol) with potassium hydroxide (0.16 g, 2.9 mmol) in EG
1
(1.6 ml). Colorless solid, yield: 40%. H-NMR (600 MHz, CD3OD): δ 7.54-7.47 (m, 2H, H3,
3
4
H5), 7.44-7.34 (m, 5H, H4, H2’, H6’, H5’’, H7’’), 7.27 (dd, 1H, J = 7.8 Hz, J = 0.9 Hz, H6),
3
7.21-7.17 (m, 1H, H6’’), 7.13 (d, 2H, J = 8.3 Hz, H3’, H5’), 5.56 (s, 2H, NCH2), 2.94 (t, 2H,
3J = 7.8 Hz, CH2CH2CH3), 1.85-1.75 (m, 2H, CH2CH2CH3), 1.01 (t, 3H, 3J = 7.4 Hz,
CH2CH2CH3). 13C-NMR (600 MHz, CD3OD): δ 175.6, 158.3, 141.6, 140.5, 140.3, 137.7, 137.3,
136.8, 131.4, 131.1, 130.3, 128.8, 128.5, 127.4, 124.3, 123.4, 110.5, 47.9, 30.1, 22.7, 14.2.
HRMS: m/z calculated for C24H22ClN3O [M+H]+: 404.1524, found: 404.1537.
4.1.3 General procedure for the synthesis of the carboxamides
A solution of the respective carboxylic acid (1 mmol) in anhydrous DMF (3-3.5 ml/mmol) was
cooled to 0 °C. PyBOP (2 mmol) dissolved in anhydrous DMF (2-3 ml/mmol) was added and the
reaction mixture was stirred for 5 min. Then, DIPEA (4 mmol) was added, followed by a
suspension of ammonium chloride (2 mmol) in anhydrous DMF (2-4 ml/mmol). The mixture
22

was stirred for another 30 min at 0 °C, before warming to room temperature. After 16 h of
stirring, the solution was diluted with water to double the volume and 1 N sodium hydroxide
solution was used to obtain a pH of 9. The mixture was extracted with EA (3×) to remove HOBt
and unreacted carboxylic acid. The organic phases were combined, washed with brine, dried over
anhydrous sodium sulfate and filtered. EA was removed under reduced pressure. The resulting
product was purified by column chromatography (DCM/MeOH, 95:5) and flash column
chromatography with gradient elution (PE/EA, 7:3, 1:1, 3:7, 0:1).
4’-((2-Propyl-1H-benzo[d]imidazole-1-yl)methyl)-[1,1’-biphenyl]-2-carboxamide (18a)
From 17a (0.10 g, 0.3 mmol) in DMF (1 ml), PyBOP (0.28 g, 0.5 mmol) dissolved in DMF
(1 ml), DIPEA (0.2 ml) and ammonium chloride (0.03 g, 0.5 mmol) suspended in DMF (3 ml).
1
Colorless solid, yield: 59%. H-NMR (400 MHz, CD3OD): δ 7.66-7.60 (m, 1H, H3), 7.54-7.45
(m, 2H, H5, H7‘‘), 7.44-7.34 (m, 5H, H4, H6, H2‘, H6‘, H4‘‘), 7.28-7.20 (m, 2H, H5‘‘, H6‘‘),
3
7.12 (d, 2H, 3J = 8.3 Hz, H3‘, H5‘), 5.53 (s, 2H, NCH2), 2.92 (t, 2H, J = 7.8 Hz, CH2CH2CH3),
3
1.85-1.74 (m, 2H, CH2CH2CH3), 0.89 (t, 3H, J = 7.1 Hz, CH2CH2CH3). 13C-NMR (400 MHz,
CD3OD): δ 175.6, 157.2, 143.0, 141.5, 140.6, 137.3, 137.2, 136.5, 131.4, 131.1, 130.3, 128.8,
128.4, 127.4, 123.8, 123.4, 119.2, 111.3, 47.5, 32.6, 28.5, 28.2, 23.4, 14.3. HRMS: m/z
calculated for C24H23N3O [M+H]+: 370.1914, found: 370.1954.
4’-((4-Bromo-2-propyl-1H-benzo[d]imidazole-1-yl)methyl)-[1,1’-biphenyl]-2-carboxamide
(18d)
From 17d (0.10 g, 0.2 mmol) in DMF (1 ml), PyBOP (0.23 g, 0.4 mmol) dissolved in DMF
(1 ml), DIPEA (0.2 ml) and ammonium chloride (0.02 g, 0.4 mmol) suspended in DMF (2 ml).
Colorless solid, yield: 21%. 1H-NMR (400 MHz, CD3OD): δ 7.56-7.32 (m, 8H, H3, H4, H5, H6,
H2’, H6’, H5’’, H7’’), 7.16-7.07 (m, 3H, H3’, H5’, H6’’), 5.53 (s, 2H, NCH2), 2.93 (t, 2H,
3J = 7.8 Hz, CH2CH2CH3), 1.87-1.69 (m, 2H, CH2CH2CH3), 1.00 (t, 3H, 3J = 7.4 Hz,
CH2CH2CH3). 13C-NMR (400 MHz, CD3OD): δ 175.5, 158.3, 141.8, 141.6, 140.5, 137.3, 137.2,
136.8, 131.4, 131.1, 130.3, 128.8, 128.5, 127.4, 126.5, 124.7, 112.7, 110.9, 48.0, 30.1, 27.4, 27.3,
22.7, 14.2. HRMS: m/z calculated for C24H22BrN3O [M+H]+: 448.1019, found: 448.1037.
4’-((4-Methyl-2-propyl-1
(18e)
H-benzo[d]imidazole-1-yl)methyl)-[1,1’-biphenyl]-2-carboxamide
From 17e (0.20 g, 0.5 mmol) in DMF (1.3 ml), PyBOP (0.54 g, 1.0 mmol) dissolved in DMF
(3 ml), DIPEA (0.4 ml) and ammonium chloride (0.06 g, 1.0 mmol) suspended in DMF (3 ml).
Colorless solid, yield: 23%. 1H-NMR (400 MHz, CD3OD): δ 7.55-7.33 (m, 6H, H3, H4, H5, H6,
H2’, H6’), 7.20 (d, 1H, 3J = 8.0 Hz, H7’’), 7.16-7.01 (m, 4H, H3’, H5’, H5’’, H6’’), 5.52 (s, 2H,
NCH2), 2.92 (t, 2H, 3J = 7.8 Hz, CH2CH2CH3), 2.62 (s, 2H, CH3), 1.86-1.70 (m, 2H,
3
CH2CH2CH3), 1.00 (t, 3H, J = 7.4 Hz, CH2CH2CH3). 13C-NMR (400 MHz, CD3OD): δ 175.6,
156.4, 142.1, 141.4, 140.6, 137.3, 136.2, 131.4, 131.1, 130.3, 129.4, 128.8, 128.4, 127.4, 124.0,
23

123.7, 108.9, 47.6, 30.2, 22.9, 16.9, 14.2. HRMS: m/z calculated for C25H25N3O [M+H]+:
384.2070, found: 384.2095.
4’-((4-Methoxy-2-propyl-1
carboxamide (18g)
H-benzo[d]imidazole-1-yl)methyl)-[1,1’-biphenyl]-2-
From 17g (0.15 g, 0.4 mmol) in DMF (1 ml), PyBOP (0.39 g, 0.8 mmol) dissolved in DMF
(2 ml), DIPEA (0.3 ml) and ammonium chloride (0.04 g, 0.8 mmol) suspended in DMF (2 ml).
1
Colorless solid, yield: 7%. H-NMR (400 MHz, CD3OD): δ 7.54-7.45 (m, 2H, H3, H5), 7.44-
3
7.35 (m, 4H, H4, H6, H2’, H6’), 7.26-7.20 (m, 1H, H6’’), 7.14 (d, 2H, J = 8.3 Hz, H3’, H5’),
3
4
3
7.05 (dd, 1H, J = 8.3 Hz, J = 0.8 Hz, H7’’), 6.84 (d, 1H, J = 7.9 Hz, H5’’), 5.55 (s, 2H,
3
NCH2), 4.00 (s, 3H, OCH3), 2.94 (t, 2H, J = 7.6 Hz, CH2CH2CH3), 1.85-1.73 (m, 2H,
CH2CH2CH3), 0.99 (t, 3H, J = 7.4 Hz, CH2CH2CH3). 13C-NMR (400 MHz, CD3OD): δ 175.5,
3
155.5, 151.5, 141.6, 140.5, 137.4, 137.3, 136.7, 131.4, 131.1, 130.7, 130.3, 130.2, 128.8, 128.5,
127.5, 127.2, 125.5, 105.0, 104.6, 56.2, 48.0, 29.7, 22.4, 14.1. HRMS: m/z calculated for
C25H25N3O2 [M+H]+: 400.2020, found: 400.2041.
4’-((4-Hydroxy-2-propyl-1
carboxamide (18h)
H-benzo[d]imidazole-1-yl)methyl)-[1,1’-biphenyl]-2-
From 17h (0.02 g, 0.04 mmol) in DMF (0.5 ml), PyBOP (0.04 g, 0.08 mmol) dissolved in
DMF (1 ml), DIPEA (0.03 ml) and ammonium chloride (0.01 g, 0.08 mmol) suspended in DMF
1
(1 ml). Colorless solid, yield: 10%. Purity: 90%. H-NMR (700 MHz, DMSO-d6): δ 9.65 (br s,
1H, OH), 7.65-7-61 (m, 1H, H3), 7.48-7.36 (m, 5H, H4, H5, H2’, H6’, H6’’), 7.33 (d, 1H,
3J = 7.7 Hz, H7’’), 7.28-7.24 (m, 1H, H6), 7.11 (d, 2H, 3J = 7.9 Hz, H3’, H5’), 7.00-6.87 (m, 2H,
3
3
CONH2), 6.55 (d, 1H, J = 7.7 Hz, H5’’), 5.47 (s, 2H, NCH2), 2.83 (t, 2H, J = 7.0 Hz,
CH2CH2CH3), 1.81-1.74 (m, 2H, CH2CH2CH3), 0.97 (t, 3H, J = 7.4 Hz, CH2CH2CH3). 13C-
3
NMR (700 MHz, DMSO-d6): δ 171.5, 153.5, 138.8, 137.8, 136.5, 132.3, 130.3, 129.7, 129.1,
127.9, 127.5, 126.7, 70.2, 46.9, 29.5, 22.5, 21.2, 14.3, 12.0. HRMS: m/z calculated for
C24H23N3O2 [M+H]+: 386.1863, found: 386.1815.
4.1.4 Syntheses of compounds 17h and 18j
4’-((4-Hydroxy-2-propyl-1
acid (17h)
H-benzo[d]imidazol-1-yl)methyl)-[1,1’-biphenyl]-2-carboxylic
Compound 17h was synthesized from 17g by ether cleavage. 17g (0.10 g, 0.25 mmol) was
dissolved in anhydrous DCM and chlorobenzene (2 ml each), prior to cooling to -80 °C in a
liquid nitrogen-EtOH bath. Then, tribromoborane (0.25 g, 1.0 mmol) was added before the
reaction mixture was allowed to attain room temperature. It was stirred for 14 h overnight,
whereupon the mixture was treated with MeOH (5 ml) and evaporated. The residue was
dissolved in DCM and extracted with 2 N sodium hydroxide solution (3×). The combined
aqueous phases were acidified to a pH of 6 with 2 N hydrochloric acid and extracted with DCM
(3×). The combined organic phases were washed with brine, dried over anhydrous sodium
24

sulfate and filtered. DCM was removed under reduced pressure. The resulting product was
purified by column chromatography (DCM/MeOH, 95:5) and recrystallized from MeOH.
Colorless solid, yield: 27%. 1H-NMR (400 MHz, DMSO-d6): δ 12.73 (br s, 1H, COOH), 9.61 (s,
1H, OH), 7.69 (d, 1H, 3J = 7.6 Hz, H3), 7.57-7.50 (m, 1H, H5), 7.47-7.40 (m, 1H, H4), 7.33 (dd,
3
4
3
3
1H, J = 7.7 Hz, J = 1.3 Hz, H6), 7.28 (d, 2H, J = 8.2 Hz, H2’, H6’), 7.11 (d, 2H, J = 8.2 Hz,
H3’, H5’), 6.97-6.87 (m, 2H, H6’’, H7’’), 6.53 (dd, 1H, J = 7.5 Hz, 4J = 1.2 Hz, H5’’), 5.46 (s,
3
2H, NCH2), 2.81 (t, 2H, 3J = 7.5 Hz, CH2CH2CH3), 1.82-1.71 (m, 2H, CH2CH2CH3), 0.95 (t, 3H,
3J = 7.4 Hz, CH2CH2CH3). 13C-NMR (400 MHz, DMSO-d6): δ 169.5, 152.9, 148.6, 140.4, 139.9,
137.2, 136.2, 132.3, 131.8, 130.8, 130.4, 129.1, 128.6, 127.3, 126.2, 122.5, 106.7, 101.1, 45.9,
28.6, 20.7, 13.8. HRMS: m/z calculated for C24H22N2O3 [M+H]+: 387.1703, found: 387.1736.
4’-((4-Amino-2-propyl-1H-benzo[d]imidazole-1-yl)methyl)-[1,1’-biphenyl]-2-carboxamide
(18j)
To a solution of 18i (0.07 g, 0.2 mmol) in MeOH (6 ml) 10% palladium on activated carbon
(0.02 g, 0.02 mmol) was carefully added and the suspension was stirred under hydrogen
atmosphere for 3.5 h. It was filtered and the transparent filtrate was evaporated. Colorless solid,
1
yield: 49%. H-NMR (400 MHz, CD3OD): δ 7.54-7.45 (m, 2H, H3, H5), 7.43-7.34 (m, 4H, H4,
3
H6, H2’, H6’), 7.11 (d, 2H, J = 8.4 Hz, H3’, H5’), 7.01-6.95 (m, 1H, H6’’), 6.71 (dd, 1H,
3J = 8.1 Hz, 4J = 0.9 Hz, H7’’), 6.55 (dd, 1H, 3J = 7.7 Hz, 4J = 0.9 Hz, H5’’), 5.44 (s, 2H, NCH2),
2.86 (t, 2H, 3J = 7.6 Hz, CH2CH2CH3), 1.87-1.73 (m, 2H, CH2CH2CH3), 1.00 (t, 3H, 3J = 7.4 Hz,
13
CH2CH2CH3). C-NMR (400 MHz, CD3OD): δ 175.6, 154.6, 141.4, 140.6, 139.6, 137.4, 137.4,
137.3, 131.9, 131.4, 131.1, 130.2, 128.8, 128.4, 127.4, 124.9, 108.1, 100.8, 71.6, 70.3, 64.4, 47.6,
32.9, 30.7, 30.1, 26.6, 23.7, 22.4, 14.3, 14.2. HRMS: m/z calculated for C24H24N4O [M+H]+:
385.2023, found: 385.2035.
4.2 Biology
4.2.1 General cell culture methods
The monkey kidney-derived cell line COS-7 (ATCC) was cultured as a monolayer culture in
dulbecco's modified eagle medium (DMEM) with 4.5 g/l glucose and 584 mg/l L-glutamine (GE
Healthcare), supplemented with fetal calf serum (FCS 10%, Sigma-Aldrich), without phenol red
or sodium pyruvate. The human bone marrow stromal cell line HS-5 (ATCC) was cultivated in
DMEM supplemented with 10% FCS, 100 U/ml penicillin (Lonza), 100 µg/ml streptomycin
(Lonza) and 2 mM L-glutamine. The chronic myelogenous leukemia cell lines K562 (ATCC)
and K562-resistant were cultivated in Roswell Park Memorial Institute (RPMI) 1640 medium
(Lonza) supplemented with 10% FCS, 100 U/ml penicillin, 100 µg/ml streptomycin and 2 mM
L-glutamine. The K562-resistant cell line was obtained from Ernesto Yague and was originally
described as doxorubicin-resistant subclone of K562 cells (termed KD225 by Hui et al. [21]).
25

All cell lines were incubated in a humidified atmosphere (5% CO2/95% air) at 37 °C and
passaged twice a week. The final concentration of DMSO in cell-based assays never exceeded
0.1% and vehicle-treated controls were always included.
4.2.2 PPARγ transactivation assay
Transient transfection (TransIT-LT1, MoBiTec) and the dual-luciferase reporter assay
(Promega) were employed according to the manufacturer’s protocol. First, COS-7 cells were
seeded in 96-well plates (104 cells per well) in triplicates, followed by an incubation at 37 °C
under a humidified atmosphere (5% CO2/95% air) for 24 h. The cells were transiently transfected
with TransIT-LT1 as well as the three plasmids pGal5-TK-pGL3 (90 ng), pGal4-hPPARγDEF
(9 ng), and pRenilla-CMV (3 ng) in phosphate-buffered saline (PBS) prior to further incubation.
After 7 h, the selected compounds, telmisartan, pioglitazone, and the vehicle (DMSO),
respectively, were added at indicated concentrations. The samples were incubated for 39 h and
lysis was induced by washing with PBS and freezing (-80 °C) of the cells. Luciferase activity
was measured with the Multimode Plate Reader EnSpire (PerkinElmer) after adding the
appropriate buffers. Renilla luciferase activity served as internal control and for normalization
[20]. The results of pioglitazone, telmisartan, and the compounds 17a-j as well as 18a-j are
represented by the mean ± SD of ≥ 3 independent experiments with three replicates each.
4.2.3 Determination of cell death by flow cytometry
Cell death was measured as described previously [38]. Briefly, the cells were seeded in 24-well
plates (2×105 K562 or K562-resistant cells per well; 0.5×105 HS-5 cells per well) and the
compounds were added in the respective concentrations. After incubation for 72 h at 37 °C under
a 5% CO2/95% air atmosphere, the cells were harvested, stained with PI/Triton-X100 for 2 h at
4 °C, and subjected to forward/sideward scatter analyses using a CytomicsFC-500 Beckman
Coulter. Stained nuclei in the sub-G1 marker window were considered to represent dead cells.
The results are represented by the mean ± SEM of ≥ 3 independent experiments. All compounds
were proven to be soluble in the applied concentrations by microscopy.
4.2.4 Determination of metabolic activity
COS-7 cells were seeded in 96-well plates (2×103 cells per well) in triplicates. After 24 h, the
compounds were added at the respective concentrations, followed by incubation for 72 h at 37 °C
under a 5% CO2/95% air atmosphere. To analyze the metabolic activity a modified MTT (3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay (EZ4U kit, Biomedica) was
employed, according to the manufacturer’s instructions. Unspecific staining was excluded by
subtracting the optical density of the respective FCS-containing medium and the substrate.
Metabolic activity in the absence of the compounds (CTR, 20 µM DMSO) was set to 100%. The
results of pioglitazone, telmisartan, and the compounds 17a-j as well as 18a-j are represented by
the mean + SD of ≥ 3 independent experiments with three replicates each [39].
26

4.2.5 Western blot analyses
Preparation of protein extracts was performed as described previously [38]. Immunoblots were
incubated with antibodies specific to STAT5 (#94205) and GAPDH (#2118, Cell Signaling).
After incubation with anti-rabbit horseradish-peroxidase-conjugated secondary antibody (GE
Healthcare), the blots were developed by enhanced chemiluminescence (GE Healthcare), and
analyzed with the AutoChemi detection system (BioRad Laboratories).
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript.
Declarations of interest
None.
Acknowledgment
This work was supported by the Austrian Research Promotion Agency FFG [West Austrian
BioNMR 858017], the “Kinderkrebshilfe Südtirol-Regenbogen”, and the “Südtiroler
Krebshilfe”. We thank Christoph Kreutz (Institute of Organic Chemistry, University of
Innsbruck) for his assistance in recording 13C-NMR spectra and Peter Enoh (Department of
Pharmaceutical Chemistry, University of Innsbruck) for technical support.
Appendix. Supplementary data
The Supporting Information can be found online at https://www.journals.elsevier.com/european-
journal-of-medicinal-chemistry. Syntheses of intermediates. Analytical data including NMR
spectroscopy, HRMS and HPLC of all compounds.
References
[1] C.L. Sawyers, Chronic myeloid leukemia, N. Engl. J. Med., 340 (1999), pp. 1330-1340.
[2] S.M. Graham, H.G. Jorgensen, E. Allan, C. Pearson, M.J. Alcorn, L. Richmond, T.L. Holyoake,
Primitive, quiescent, Philadelphia-positive stem cells from patients with chronic myeloid leukemia
are insensitive to STI571 in vitro, Blood, 99 (2002), pp. 319-325.
27

[3] M.S. Holtz, M.L. Slovak, F. Zhang, C.L. Sawyers, S.J. Forman, R. Bhatia, Imatinib mesylate
(STI571) inhibits growth of primitive malignant progenitors in chronic myelogenous leukemia
through reversal of abnormally increased proliferation, Blood, 99 (2002), pp. 3792-3800.
[4] X. Jiang, Y. Zhao, C. Smith, M. Gasparetto, A. Turhan, A. Eaves, C. Eaves, Chronic myeloid
leukemia stem cells possess multiple unique features of resistance to BCR-ABL targeted therapies,
Leukemia, 21 (2007), pp. 926-935.
[5] S. Chu, T. McDonald, A. Lin, S. Chakraborty, Q. Huang, D.S. Snyder, R. Bhatia, Persistence of
leukemia stem cells in chronic myelogenous leukemia patients in prolonged remission with imatinib
treatment, Blood, 118 (2011), pp. 5565-5572.
[6] S. Saussele, J. Richter, A. Hochhaus, F.X. Mahon, The concept of treatment-free remission in
chronic myeloid leukemia, Leukemia, 30 (2016), pp.1638-1647.
[7] D.M. Ross, S. Branford, J.F. Seymour, A.P. Schwarer, C. Arthur, P.A. Bartley, C. Slader, C.
Field, P. Dang, R.J. Filshie, A.K. Mills, A.P. Grigg, J.V. Melo, T.P. Hughes, Patients with chronic
myeloid leukemia who maintain a complete molecular response after stopping imatinib treatment
have evidence of persistent leukemia by DNA PCR, Leukemia, 24 (2010), pp. 1719-1724.
[8] P. Rousselot, S. Prost, J. Guilhot, L. Roy, G. Etienne, L. Legros, A. Charbonnier, V. Coiteux, P.
Cony-Makhoul, F. Huguet, E. Cayssials, J.-M. Cayuela, F. Relouzat, M. Delord, H. Bruzzoni-Giovanelli,
L. Morisset, F.-X. Mahon, F. Guilhot, P. Leboulch, Pioglitazone together with imatinib in chronic
myeloid leukemia: a proof of concept study, Cancer, 123 (2017), pp. 1791-1799.
[9] S. Prost, F. Relouzat, M. Spentchian, Y. Ouzegdouh, J. Saliba, G. Massonnet, J.-P. Beressi, E.
Verhoeyen, V. Raggueneau, B. Maneglier, S. Castaigne, C. Chomienne, S. Chretien, P. Rousselot, P.
Leboulch, Erosion of the chronic myeloid leukaemia stem cell pool by PPARγ agonists, Nature, 525
(2015), pp. 380-383.
[10] B. Apsel Winger, N.P. Shah, PPARgamma: welcoming the new kid on the CML stem cell
block, Cancer Cell, 28 (2015), pp. 409-411.
[11] D. Heudobler, M. Rechenmacher, F. Luke, M. Vogelhuber, T. Pukrop, W. Herr, L. Ghibelli, C.
Gerner, A. Reichle, Peroxisome proliferator-activated receptors (PPAR)gamma agonists as master
modulators of tumor tissue, Int. J. Mol. Sci., 19 (2018), 3540.
[12] B. Yousefi, N. Samadi, B. Baradaran, V. Shafiei-Irannejad, N. Zarghami, Peroxisome
proliferator-activated receptor ligands and their role in chronic myeloid leukemia: therapeutic
strategies, Chem. Biol. Drug Des., 88 (2016), pp. 17-25.
[13] A.J. Kroker, J.B. Bruning, Review of the structural and dynamic mechanisms of PPAR
γ
partial agonism, PPAR Res., 2015 (2015), 816856.
[14] K.K.W. To, W.K.K. Wu, H.H.F. Loong, PPARgamma agonists sensitize PTEN-deficient
resistant lung cancer cells to EGFR tyrosine kinase inhibitors by inducing autophagy, Eur. J.
Pharmacol., 823 (2018), pp. 19-26.
[15] S. Ryu, D. Seong Kim, M. Lee, J. Won Lee, K. Woong Sung, H. Hoe Koo, K. Hee Yoo, Anti-
leukemic effects of PPARγ ligands, Cancer Lett., 418 (2018), pp. 10-19.
[16] T. Kozako, S. Soeda, M. Yoshimitsu, N. Arima, A. Kuroki, S. Hirata, H. Tanaka, O. Imakyure,
N. Tone, S.-i. Honda, S. Soeda, Angiotensin II type 1 receptor blocker telmisartan induces apoptosis
and autophagy in adult T‐cell leukemia cells, FEBS Open Bio, 6 (2016), pp. 442-460.
[17] F. Zhang, B.E. Lavan, F.M. Gregoire, Selective modulators of PPAR-gamma activity:
molecular aspects related to obesity and side-effects, PPAR Res., 2007 (2007), 32696.
28

[18] M. Goebel, M. Clemenz, B. Staels, T. Unger, U. Kintscher, R. Gust, Characterization of new
PPARγ agonists: analysis of telmisartan's structural components, ChemMedChem, 4 (2009), pp. 445-
456.
[19] M. Goebel, B. Staels, T. Unger, U. Kintscher, R. Gust, Characterization of new PPARgamma
agonists: benzimidazole derivatives - the importance of position 2, ChemMedChem, 4 (2009), pp.
1136-1142.
[20] E. Raspe, L. Madsen, A.M. Lefebvre, I. Leitersdorf, L. Gelman, J. Peinado-Onsurbe, J.
Dallongeville, J.C. Fruchart, R. Berge, B. Staels, Modulation of rat liver apolipoprotein gene
expression and serum lipid levels by tetradecylthioacetic acid (TTA) via PPARalpha activation, J.
Lipid Res., 40 (1999), pp. 2099-2110.
[21] R.C. Hui, R.E. Francis, S.K. Guest, J.R. Costa, A.R. Gomes, S.S. Myatt, J.J. Brosens, E.W. Lam,
Doxorubicin activates FOXO3a to induce the expression of multidrug resistance gene ABCB1
(MDR1) in K562 leukemic cells, Mol. Cancer Ther., 7 (2008), pp. 670-678.
[22] V. Obermoser, M. Urban, M. Murgueitio, G. Wolber, U. Kintscher, R. Gust, New telmisartan-
derived PPARγ agonists: impact of the 3D-binding mode on the pharmacological profile, Eur. J.
Med. Chem., 124 (2016), pp. 138-152.
[23] V. Obermoser, R. Mauersberger, D. Schuster, M. Czifersky, M. Lipova, M. Siegl, U. Kintscher,
R. Gust, Importance of 5/6-aryl substitution on the pharmacological profile of 4
benzo[d]imidazol-1-yl)methyl)-[1,1 -biphenyl]-2-carboxylic acid derived PPAR agonists, Eur. J.
Med. Chem., 126 (2017), pp. 590-603.
ʹ
-((2-propyl-1H-
ʹ
γ
[24] D. Xie, Y. Wang, J. Xie, J. Lu, J. Cui, M. Zhang, L. Fu, Y. Wang, Quinoxaline-2,3-diones:
potential d-amino acid oxidase (DAAO) inhibitors, Med. Chem. Res., 23 (2014), pp. 4977-4989.
[25] H. Chen, C. Nilsen, A. Choudhury, K.L. Sorgi, A safe and convenient synthesis of 4-
benzyloxy-3-chloroaniline, Arkivoc, 2008 (2008), pp. 1-6.
[26] F.D. Gunstone, S.H. Tucker, The preparation of 1-chloro-2,6-dinitrobenzene, J. Appl. Chem.,
2 (1952), pp. 204-208.
[27] D. Spinks, H.B. Ong, C.P. Mpamhanga, E.J. Shanks, D.A. Robinson, I.T. Collie, K.D. Read, J.A.
Frearson, P.G. Wyatt, R. Brenk, A.H. Fairlamb, I.H. Gilbert, Design, synthesis and biological
evaluation of novel inhibitors of trypanosoma brucei pteridine reductase 1, ChemMedChem, 6
(2011), pp. 302-308.
[28] V. Milata, J. Saloň, Simple, high yield preparation of 3-nitro-1,2-phenylenediamine, Org.
Prep. Proced. Int., 31 (1999), pp. 347-348.
[29] E.D. Bergmann, M. Bentov, Sandmeyer reactions of monoacyl arylenediamines, J. Org.
Chem., 20 (1955), pp. 1654-1656.
[30] A. Tanakit, M. Rouffet, D.P. Martin, S.M. Cohen, Investigating chelating sulfonamides and
their use in metalloproteinase inhibitors, Dalton Trans., 41 (2012), pp. 6507-6515.
[31] P.K. Pradhan, S. Dey, P. Jaisankar, V.S. Giri, Fe‐HCl: an efficient reagent for deprotection of
oximes as well as selective oxidative hydrolysis of nitroalkenes and nitroalkanes to ketones, Synth.
Commun., 35 (2005), pp. 913-922.
[32] A.R. Katritzky, C.J. Moody, O. Meth-Cohn, C.W. Rees, Comprehensive organic functional
group transformations, Pergamon Press (Elsevier ScienceLtd.), Tarrytown, NY., 1995.
[33] T.M. Kosak, H.A. Conrad, A.L. Korich, R.L. Lord, Ether cleavage re-investigated: elucidating
the mechanism of BBr3-facilitated demethylation of aryl methyl ethers, Eur. J. Org. Chem., 2015
(2015), pp. 7460-7467.
29