Synthesis, Characterization and in vitro Studies of a Cathepsin B-Cleavable Prodrug of the VEGFR Inhibitor Sunitinib

Article abstract of DOI:10.1002/cbdv.201800520

Since several decades, the prodrug concept has raised considerable interest in cancer research due to its potential to overcome common problems associated with chemotherapy. However, for small-molecule tyrosine kinase inhibitors, which also cause severe side effects, hardly any strategies to generate prodrugs for therapeutic improvement have been reported so far. Here, we present the synthesis and biological investigation of a cathepsin B-cleavable prodrug of the VEGFR inhibitor sunitinib. Cell viability assays and Western blot analyses revealed, that, in contrast to the non-cathepsin B-cleavable reference compound, the prodrug shows activity comparable to the original drug sunitinib in the highly cathepsin B-expressing cell lines Caki-1 and RU-MH. Moreover, a cathepsin B cleavage assay confirmed the desired enzymatic activation of the prodrug. Together, the obtained data show that the concept of cathepsin B-cleavable prodrugs can be transferred to the class of targeted therapeutics, allowing the development of optimized tyrosine kinase inhibitors for the treatment of cancer.

Full text of DOI:10.1002/cbdv.201800520

DOI: 10.1002/cbdv.201800520
FULL PAPER
Synthesis, Characterization and in vitro Studies of a
Cathepsin B-Cleavable Prodrug of the VEGFR Inhibitor Sunitinib
a
b
a
b
Claudia Karnthaler-Benbakka, Bettina Koblmüller, Marlene Mathuber, Katharina Holste,
b, c
b, c
a, c
a, c
Walter Berger, Petra Heffeter, Christian R. Kowol,* and Bernhard K. Keppler
a
Institute of Inorganic Chemistry, Faculty of Chemistry, University of Vienna, Waehringer Straße 42, AT-1090
Wien, e-mail: christian.kowol@univie.ac.at
b
c
Institute of Cancer Research and Comprehensive Cancer Center, Medical University of Vienna, Borschkegasse
A, AT-1090 Wien
Research Cluster ‘Translational Cancer Therapy Research’, University of Vienna and Medical University of Vienna,
AT-1090 Wien
8
©
2018 The Authors. Published by Wiley-VHCA AG. This is an open access article under the terms of the Creative Commons
Attribution Non-Commercial NoDerivs License, which permits use and distribution in any medium, provided the original work is
properly cited, the use is non-commercial and no modifications or adaptations are made.
Since several decades, the prodrug concept has raised considerable interest in cancer research due to its
potential to overcome common problems associated with chemotherapy. However, for small-molecule tyrosine
kinase inhibitors, which also cause severe side effects, hardly any strategies to generate prodrugs for therapeutic
improvement have been reported so far. Here, we present the synthesis and biological investigation of a
cathepsin B-cleavable prodrug of the VEGFR inhibitor sunitinib. Cell viability assays and Western blot analyses
revealed, that, in contrast to the non-cathepsin B-cleavable reference compound, the prodrug shows activity
comparable to the original drug sunitinib in the highly cathepsin B-expressing cell lines Caki-1 and RU-MH.
Moreover, a cathepsin B cleavage assay confirmed the desired enzymatic activation of the prodrug. Together,
the obtained data show that the concept of cathepsin B-cleavable prodrugs can be transferred to the class of
targeted therapeutics, allowing the development of optimized tyrosine kinase inhibitors for the treatment of
cancer.
Keywords: tyrosine kinase inhibitor, prodrugs, inhibitors, cathepsin B, sunitinib, vascular endothelial growth
factor receptor (VEGFR).
Introduction
tumor-specific. A broad spectrum of different targeting
With the aim of overcoming common drawbacks of strategies has been developed in the last decades.[3–5]
chemotherapeutic agents, such as a low therapeutic One of them is the activation via cleavage by
index due to high toxicities and side effects and a lack cathepsin B, a (usually lysosomal) cysteine protease,
of tumor selectivity, the prodrug concept has attracted which is overexpressed and secreted to the cell surface
considerable attention in the field of cancer research and pericellular space by a wide variety of cancer
since the 1950s.[ In general, this concept describes a types. High levels of cathepsin B are known to
pharmacologically inert derivative of a drug that can strongly contribute to the invasiveness of malignant
be converted to its active analogue in vivo by cells, and expression as well as activity of cathepsin B
enzymatic or non-enzymatic mechanisms,[ which, in have been correlated with worse prognosis in the
1]
[6]
2]
[
7,8]
the case of anticancer drugs, should preferably be clinic.
This tumor-specific overexpression of cathe-
psin B can be exploited for the design of prodrugs
carrying a cathepsin B-cleavable entity, resulting in the
Supporting information for this article is available on the release of the active species only in the surrounding of
WWW under https://doi.org/10.1002/cbdv.201800520
tumor cells that express cathepsin B on their cell
Chem. Biodiversity 2018, 15, e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG

Chem. Biodiversity 2018, 15, e1800520
surface. One example for the successful use of this Upon cathepsin B cleavage of the dipeptide, the
enzymatic activation is brentuximab vedotin (Adcet- PABC-linker is supposed to undergo 1,6-elimination
ris®), an antibody-drug conjugate that has already with subsequent release of the active drug (the
been approved for the treatment of relapsed or schematic overview of this concept is outlined in
refractory Hodgkin’s lymphoma and relapsed or refrac- Figure 1). The Phe-Lys dipeptide has demonstrated its
[
9]
tory systemic anaplastic large cell lymphoma. In efficacy with superior properties compared to other
addition, a range of other antibody conjugates as well peptides (often tetrapeptides) commonly used as drug
as doxorubicin prodrugs using cathepsin B cleavage as carrier linkers.[22] In addition, reduced toxicity and
activation mechanism are under (pre)clinical promising activity against different cancer types was
[
10–14]
evaluation.
Interestingly, most of the so far used recently reported for Ac-Phe-Lys-PABC-doxorubicin
[
24–26]
drugs for the design of cathepsin B-cleavable ther- in vitro and in vivo.
apeutics are either conventional small-molecule che-
motherapeutics or natural bacterial toxins.[ Thus, to
5]
the best of our knowledge, for the large class of small- Results and Discussion
molecule tyrosine kinase inhibitors (TKIs), no such
approach has been reported so far.
Synthesis
Within this study, we synthesized and investigated The synthesis of Z-Phe-Lys(alloc)-PABC-PNP was per-
cathepsin B-cleavable prodrugs of the vascular endo- formed by a straightforward four-step procedure from
[
23]
thelial growth factor receptor (VEGFR) inhibitor suniti- Dubowchik et al.
Sunitinib was then introduced
nib (Sutent®), which is approved for the treatment of using 4-DMAP in dry DMF with reaction times of up to
gastrointestinal stromal tumors (GIST) after failure of seven days followed by TLC. Alloc deprotection was
imatinib treatment, advanced renal cell carcinoma effected by treatment with Pd(PPh ) and, as a less
3
4
(
RCC), and unresectable or metastatic pancreatic toxic hydride source than Bu SnH, Me NH·BH3
3 2
[15]
neuroendocrine tumors (pNET).
Despite the belief (40 equiv.). Unfortunately, the final compound Z-Phe-
that targeted therapeutics (including sunitinib and Lys-PABC-Sun was not soluble at all in aqueous
other representatives of the class of TKIs) should solution at acceptable concentrations, not even by the
possess a much higher tolerability than conventional addition of 10% DMSO. Thus, we decided to introduce
chemotherapeutics, their clinical use is limited due to morpholine as water-solubilizing group at the peptide
severe side effects, such as hypertension, hepatotox- N-terminus. To this aim, Boc-Phe-OH was used as
icity, hypothyroidism, hand-foot syndrome, papulo- starting material, which was converted to Boc-Phe-Lys
[
16,17]
[23]
pustular rash, gastrointestinal perforation, etc.
(alloc)-PABOH (3) similarly to Dubowchik’s method.
Thus, also for TKIs, a strategy to circumvent these After cleavage of the Boc-protecting group, morpho-
problems would be a major clinical advantage. In this lin-4-yl-acetic acid was introduced to give Morph-Phe-
work, we designed cathepsin B-activatable prodrugs of Lys(alloc)-PABOH (5). Attempts to generate the PNP-
the VEGFR inhibitor sunitinib, since studies suggested active ester using PNP chloroformate as used in the
[
6,18–20]
a role of cathepsin B in inducing angiogenesis.
case of the Z-protected derivative failed, likely due to
This would offer the possibility of targeting tumor sites an attack of the morpholine nitrogen on the acid
which comprise high levels of both: cathepsin B and chloride. Alternatively, 5 was treated with bis(4-nitro-
VEGFR.
phenyl) carbonate and DIPEA in dry DMF to give the
In order to generate a prodrug of a TKI, the desired activated cathepsin B-cleavable unit 6. After
compound has to be inactivated by derivatizing a introduction of sunitinib, also the deprotection step
position essential for its tyrosine kinase inhibiting had to be adjusted, resulting in the use of Pd(PPh3)4
potential. Only after activation through cathepsin B and 1.1 equivalent of NDMBA (1,3-dimethylbarbituric
cleavage, the unmodified drug should be released. In acid) as hydride source. Final purification through RP-
the case of sunitinib, the oxindole nitrogen is a HPLC yielded the desired product Morph-Phe-Lys-
suitable site for derivatization, as co-crystal structures PABC-Sun·3 HCOOH (8) (Scheme 1,a). As a reference
of VEGFR2 and sunitinib revealed a crucial hydrogen compound, Morph-Gly-Gly-PABC-Sun (14; Scheme 1,b),
bond between this nitrogen and Glu917 in the which contains the dipeptide sequence Gly-Gly that
adenine pocket.[21] The cathepsin B-cleavable unit was should not be cleaved by cathepsin B, was prepared in
adopted from Dubowchik et al.[
22,23]
comprising the Z- a similar way (Scheme S1).
protected dipeptide Phe-Lys, followed by a self-
immolative p-aminobenzylcarbonyl (PABC) linker.
www.cb.wiley.com (2 of 12) e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG,
Zurich, Switzerland

Chem. Biodiversity 2018, 15, e1800520
Selection of Cell Lines
To allow the biological testing of our novel drugs, an
appropriate cell line panel of sunitinib-sensitive cell
lines with different cathepsin B expression levels
needed to be selected. To this end, a collection of 15
cancer cell models originating from different tissues
(including diverse renal cell carcinomas) were tested
for their sunitinib sensitivity, sunitinib target expres-
sion (VEGFR-1 and PDGFR β) as well as cathepsin B
protein expression (by Western blotting) and activity
(by zymography). A summary of the collected data is
shown in Table 1 (see also Figures S11–S13, Supple-
Table 1. Overview of sunitinib sensitivity, sunitinib target
receptor tyrosine kinase (RTK) expression, and cathepsin B
status of the used cell lines.
Cell line
Sunitinib RTK
IC50 [μM] SD Expression
Cathepsin B status
Western Zymo-
blot
graphy
Calu3
9.6
0.5
>10
0.1 VEGFR-1
+ + +
–
–
+
+
H1703
H1975
HCC827
0.0 VEGFR-1/PDFGR β –
a]
–
n.t.[
0.5 VEGFR-1
n.t.
0.2 VEGFR-1
n.t.
+
n.t.
+ + +
+
+ + +
+ + +
+ +
–
5.2
U87MG >10
–
n.t.
+ + +
–
+ + +
+ +
+ + +
–
–
–
+ +
Caki-1
Caki-2
ORMW/8
RU-MH
A431
A375
PC-3
HCT116
A549
6.8
>10
7.9
–
–
0.7 VEGFR-1
1.3 VEGFR-1
0.4 VEGFR-1
1.0 n.t.
0.1 n.t.
0.3 VEGFR-1
6.6
7.3
4.4
8.7
4.6
7.2
4.2
–
+/–
+/–
+
0.0 n.t.
0.4 VEGFR-1
H520
n.t.
[a]
n.t. – not tested
mentary Information). Based on this data, the cell lines
Caki-1 (strong cathepsin B expression and activity), RU-
MH (strong cathepsin B expression and activity) and
HCT116 (weak cathepsin B expression and activity)
were selected for the subsequent biological testing of
the drugs.
Figure 1. Schematic overview of the cathepsin B-cleavable Cytotoxicity Tests of Compounds 8 and 14
prodrug concept for sunitinib. Hydrogen bonds are indicated
by dashed lines.
After 72 h exposure, viability tests revealed that our
new sunitinib prodrug 8 had similar but about 1.1 to
1
.5-fold reduced anticancer activity compared to free
sunitinib (Table 2, Figure S14). In contrast, the reference
drug 14, which should not be cleaved by cathepsin B,
revealed distinctly less activity indicating that the
www.cb.wiley.com (3 of 12) e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG,
Zurich, Switzerland

Chem. Biodiversity 2018, 15, e1800520
Scheme 1. A) Synthetic route of prodrug 8. Reagents and conditions: a) NHS, DCC, THF abs., 0°C to room temperature; b) H-
Lys(alloc)-OH, NaHCO , DME, H O; c) PABOH, EEDQ, THF; d) HCl conc., EtOH; e) morpholin-4-yl-acetic acid, EEDQ, Et N, THF/EtOH; f)
3
2
3
bis(4-nitrophenyl) carbonate, DIPEA, DMF abs.; g) sunitinib, 4-DMAP, DMF abs.; h) Pd(PPh ) , NDMBA, DMF abs. B) Chemical structure
3
4
of the Gly-Gly reference compound 14.
Table 2. IC50 values of the tested cell lines after treatment with
the indicated drugs for 72 h.
attachment of the Morph-Gly-Gly-PABC-unit indeed
reduced sunitinib activity. Moreover, subsequent West-
ern blot analysis (Figure 2) indicated that after 4 h of
drug treatment 8 was able to inhibit phosphorylation
of ERK (a VEGFR-1-downstream protein, which is
frequently used as read-out for activity of this signal-
Cell line Sunitinib
8
14
IC50 [μM] SD IC50 [μM] SD IC50 [μM] SD
HCT116 5.2
1.8
1.8 12.2
0.5 7.9
7.8
1.1
2.7 >20
0.3 18.1
14.9
5.4
–
7.9
Caki-1
7.9
6.9
RU-MH
[
27,28]
ing pathway
). Comparable to the viability assays,
www.cb.wiley.com (4 of 12) e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG,
Zurich, Switzerland

Chem. Biodiversity 2018, 15, e1800520
However, the activation studies also revealed an
unexpected hydrolytic lability for both compounds.
Thus, additional stability measurements in 10 mM
sodium phosphate buffer, pH 7.4, were performed
with 8 and 14 at 37°C, monitored by reversed-phase
HPLC chromatography. As observed before, both
showed fast hydrolysis with half-lives of ca. 2 h and ca.
3
0 min, respectively, and only few percent of remain-
ing substrate after 24 h (Figures S5–S6, Supplementary
Information). Since Dubowchik stated that the corre-
sponding doxorubicin derivative was even stable in
Figure 2. Inhibitory effects of prodrug 8 in comparison to
sunitinib on RTK signaling. Caki-1 cells were grown in medium
with FCS and treated with the indicated drug for 4 h. Then, they
were harvested and lysed, and the impact on ERK phosphor-
[
23]
human plasma at 37°C, the reason for the lability of
8
and 14 has to be the carbamate linkage to the
ylation as readout for RTK signaling was analyzed by Western lactam nitrogen. In fact, Morph-Phe-Lys-PABOH was
blotting.
detected by HPLC-MS measurements as main frag-
ment of 8, indicating hydrolysis at this position.
Notably, in the literature, substances with a similar
8
was less active (full inhibition at 10 μM) than free substitution at the oxindole nitrogen were investi-
sunitinib (full inhibition at 2.5 μM), which can be gated without mentioning any stability problems.[
explained by the required drug cleavage before the Thus, it has to be questioned, whether only our
onset of drug activity, while the free drug is already particular design is responsible for the low stability or
capable of full anticancer activity. However, in contrast whether this problem just has not been recognized so
to the expectations, HCT116 cells, which were charac- far. Moreover, also fragments other than free drug
terized by only weak cathepsin B expression and were formed in both cases (Figures S7–S8, Supplemen-
activity, were sensitive to our new prodrug compara- tary Information). Especially for 14, an intense signal
ble to the high cathepsin B-expressing Caki-1 and RU- was observed, that could be assigned to the mass of
MH cells. Possibly already low levels of cathepsin B are the directly linked p-aminobenzyl unit to the sunitinib
sufficient to cleave the prodrug unit or other enzymes oxindole nitrogen. In fact, the presence of such a
29,30]
can also activate the compound.
rearrangement product that was also observed for 8
(however, with a much lower intensity) could be
confirmed by high resolution mass spectrometry. The
postulated mechanism of this rearrangement that
Cathepsin B Cleavage Assays and Activation Studies
Consequently, to better understand the activation of implies a keto-enol equilibrium is initiated by an attack
our new prodrug, cathepsin B cleavage assays were of the nitrogen double bond on the PABC moiety,
performed with 8 and 14 to demonstrate proof of followed by the elimination of CO (Scheme 2). The
2
principle for this concept. To this end, the substrates amount of this directly linked derivative, which was
were incubated with human liver cathepsin B at pH 5.0 stable over a period of 24 h according to HPLC-MS
and 37°C, and the reaction was followed by HPLC. In studies, also explains the rather low percentage of
order to distinguish between the effects of enzymatic hydrolytic sunitinib release (only 30–40%, Figures S5–
cleavage and hydrolysis, incubation of the compounds S6, Supplementary Information). Notably, further stabil-
without cathepsin B under the same conditions ity experiments revealed that the rearrangement
(denoted as ‘8 control’ and ‘14 control’ in Figure 3) product of both 8 and 14 was formed by an even
served as a reference experiment. As Figure 3,A, 3,C, higher extent in medium (Mc Coy’s) with 10% fetal calf
and 3,D shows, compound 8 was readily activated by serum than in phosphate buffer (Figures S9–S10,
cathepsin B with a half-life of <15 min resulting in Supplementary Information). As we assume that the
release of sunitinib via self-immolation of the PABC cytotoxicity of this directly linked derivative is quite
linker. By contrast, as expected, for the reference low (no release of sunitinib), this might be the reason
compound 14 no difference could be observed for the reduced cytotoxic activity of the non-cathepsin
between the measurements with/without cathepsin B, B-cleavable compound 14 in viability assays, despite
identifying it as the desired non-cathepsin B-cleavable its partial hydrolysis. For compound 8 (with much
negative control (Figure 3,B).
lower tendency to form the rearrangement product)
the activity in Caki-1 and RU-MH cells could be
www.cb.wiley.com (5 of 12) e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG,
Zurich, Switzerland

Chem. Biodiversity 2018, 15, e1800520
explained by cathepsin B cleavage and subsequent is caused by prodrug hydrolysis, if low levels of
self-immolation with release of free sunitinib. However, cathepsin B are sufficient to cleave the prodrug unit,
in case of HCT116 cells with only weak cathepsin B or whether another enzyme activates the compound.
expression, it is still uncertain whether the activity of 8
Figure 3. Cathepsin B assay. A), B): Time curves of (A) compound 8 and (B) compound 14 after incubation with and without
(
‘control’) human liver cathepsin B at pH 5 and 37°C. C), D): Chromatograms of the incubation study of compound 8 (C) with
human liver cathepsin B, and (D) without cathepsin B.
Scheme 2. Postulated mechanism of the rearrangement of the enol form of 8 and 14 from carbamate to direct linkage in aqueous
solution.
www.cb.wiley.com (6 of 12) e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG,
Zurich, Switzerland

Chem. Biodiversity 2018, 15, e1800520
Conclusions
were determined by RP-HPLC on a Dionex UltiMate
000 UHPLC system controlled by Chromeleon 6.8
3
In this study, a cathepsin B-cleavable prodrug of the chromatography software. The experimental condi-
VEGFR inhibitor sunitinib together with a non-cleav- tions were as follows: stationary phase: ethylene
able control, both containing a water-soluble unit, bridged hybrid C18; column: Acquity UPLC BEH C18,
have been successfully synthesized through a multi- 130 Å, 1.7 μm, 3.0 mm×50 mm (Waters Corp., Massa-
step procedure. Their antiproliferative potency was chusetts, USA); column temperature: 25°C; mobile
tested against suitable sunitinib-sensitive cancer cell phase: H O/MeCN (0.1% HCOOH), HPLC grade; flow
2
lines which were previously evaluated for their VEGFR- rate: 0.6 ml/min, injection volume: 10 μL; gradient: 5–
1
, PDGFR β, as well as cathepsin B expression and 95% MeCN in 6 min. Decomposition products were
cathepsin B activity by Western blot analysis and identified through HPLC-MS on a 1260 Infinity Bio-Inert
zymography, respectively, showing similar activity of 8 LC System from Agilent Technologies, controlled by an
compared to the original drug sunitinib, while the Agilent OpenLAB CDS ChemStation software (Edition
non-cleavable reference drug 14 was distinctly less Rev. C.01.06), coupled to an amaZon SL Ion Trap mass
active. Moreover, a cathepsin B cleavage assay con- spectrometer with HyStar 3.2 and Data Analysis 4.0
firmed the desired enzymatic activation of 8 and software package (Bruker Daltonics).
inactivity of 14, showing proof of principle for the first
time for this prodrug concept for the class of targeted
therapeutics. However, also hydrolytic lability of the
Chemicals and Materials
compounds with the formation of stable rearrange- All solvents and reagents were obtained from com-
ment products was observed, complicating the inter- mercial suppliers and used without further purification.
pretation of the obtained biological data, especially Anhydrous solvents were bought from Fisher Scientific
regarding the response of the only weak cathepsin B- GmbH (Austria) over molecular sieves. Sunitinib was
ɛ
expressing HTC116 cells. Thus, future investigations purchased from LC Laboratories®, N -allyloxycarbonyl-L-
with altered linker moieties will be performed to lysine from Chem-Impex International, Inc., and mor-
optimize the applicability of cathepsin B-based activa- pholin-4-yl-acetic acid from Activate Scientific GmbH.
tion approaches for targeted therapeutics. Moreover,
the in vivo evaluation of this strategy for tumor-
specific drug activation is necessary.
Preparation of Compounds
Boc-Phe-OSu (1).[31] A solution of Boc-Phe-OH (2.00 g,
7
.54 mmol) and N-hydroxysuccinimide (954 mg,
Experimental Section
1.1 equiv.) in dry THF (15 mL) at 0°C was treated with
DCC (1.63 g, 1.05 equiv.). The resulting mixture was
allowed to warm to room temperature. After 23 h the
General
Mass spectrometry was performed on a Bruker HCT solid byproduct was filtered off, washed with THF, and
plus ESI-QIT spectrometer, high resolution spectra were the filtrate was evaporated. The residue was dissolved
obtained from a Bruker maXis ESI-Qq-oaRTOF spec- in CH Cl (20 mL), allowed to stand for 1 h, and filtered
2
2
trometer. Expected and experimental isotope distribu- off again. Evaporation of the filtrate yielded the
1
13
1
tions were compared. H- and C-NMR spectra were product as a white powder (2.58 g, 94%). H-NMR
recorded in (D )DMSO, CDCl , and/or MeOD, referring (MeOD): 1.39 (s, 9 H); 2.87 (s, 4 H); 3.05 (dd, J=14, 10,
6
3
to the respective solubility of the synthesized com- 1 H); 3.32–3.36 (m, 1 H); 4.74–4.79 (m, 1 H); 7.24–7.35
pounds, with a Bruker Avance III 500 MHz spectrometer (m, 5 H).
1
13
at 500.32 ( H) and 125.81 ( C) MHz at 298 K, with
chemical shifts referenced to the solvent residual peak
Boc-Phe-Lys(alloc)-OH (2).[32] A mixture of H-Lys-
as an internal standard. NMR Data are reported (alloc)-OH (1.70 g, 7.48 mmol) and NaHCO (628 mg,
3
indicating the chemical shift (δ, [ppm]), the multiplicity 1 equiv.) in H O (25 ml) was added to a stirred solution
2
(s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet, of 1 (2.58 g, 0.95 equiv.) in DME (25 ml) which was
etc.), the coupling constant (J, [Hz]), and the integra- cooled in a cold water bath. The resulting mixture was
tion. All final compounds displayed �95% purity as allowed to warm to room temperature. After 23 h a
determined by elemental analysis performed by the small amount of solid was filtered off, and the filtrate
Microanalytical Laboratory of the University of Vienna. was diluted with H O (42 mL). The solution was
2
Stability and activation upon cathepsin B cleavage acidified to pH 3 with 15% aqueous citric acid and
www.cb.wiley.com (7 of 12) e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG,
Zurich, Switzerland

Chem. Biodiversity 2018, 15, e1800520
extracted with AcOEt (3×75 mL). The combined EtOH (5 mL), morpholin-4-yl-acetic acid (140 mg,
organic layers were washed with H O and brine 1 equiv.) and EEDQ (250 mg, 1.05 equiv.) were added.
2
(
100 mL each), dried over MgSO , filtered, and concen- After stirring for 24 h at room temperature, the solvent
4
1
trated in vacuo to give a white solid (2.95 g, 87%). H- was evaporated, the residue was treated with Et O
2
NMR (MeOD): 1.39 (s, 9 H); 1.39–1.46 (m, 2 H); 1.51– (5 mL), sonicated, and then left to stand for 2 h at
1
2
9
.58 (m, 2 H); 1.71–1.79 (m, 1 H); 1.87–1.94 (m, 1 H); room temperature. The solid was collected by filtra-
.84 (dd, J=14, 9, 1 H); 3.12–3.18 (m, 3 H); 4.36 (dd, J= tion, washed with Et O, and dried in vacuo. The crude
2
, 5, 1 H); 4.42 (dd, J=9, 5, 1 H); 4.53 (d, J=5, 2 H); 5.19 product was purified by flash chromatography over
(
7
d, J=11, 1 H); 5.30 (d, J=17, 2 H); 5.90–5.98 (m, 1 H); silica gel, eluting with AcOEt/MeOH 10:1, to give the
1
.22–7.31 (m, 5 H).
product as a white powder (283 mg, 48%). H-NMR
(
MeOD): 1.40–1.52 (m, 2 H); 1.53–1.60 (m, 2 H); 1.76–
Boc-Phe-Lys(alloc)-PABOH (3).[32] To a solution of 1.83 (m, 1 H); 1.89–1.96 (m, 1 H); 2.26–2.30 (m, 2 H);
(2.95 g, 6.18 mmol) and PABOH (799 mg, 1.05 equiv.) 2.37–2.41 (m, 2 H); 2.88 (d, J=16, 1 H); 3.02 (dd, J=14,
2
in THF (40 mL) was added EEDQ (N-ethoxycarbonyl-2- 9, 1 H); 3.06 (d, J=16, 1 H); 3.14 (t, J=7, 2 H); 3.26 (dd,
ethoxy-1,2-dihydroquinoline; 1.60 g, 1.05 equiv.) and J=14, 6, 1 H); 3.60 (t, J=5, 4 H); 4.49 (dd, J=9, 6, 1 H);
the resulting mixture was stirred for 23 h at room 4.52 (d, J=5, 2 H); 4.60 (s, 2 H); 4.77 (dd, J=9, 5, 1 H);
temperature. The solvent was removed, and the 5.18 (dd, J=11, 1, 1 H); 5.30 (dd, J=17, 2, 1 H); 5.89 –
residue was dissolved in AcOEt (200 mL). The organic 5.97 (m, 1 H); 7.18–7.19 (m, 1 H); 7.23–7.28 (m, 4 H);
1
3
layer was washed with 1M NaHCO , 10% aqueous citric 7.35 (d, J=9, 2 H); 7.59 (d, J=9, 2 H). C-NMR (MeOD):
3
acid, and brine (100 mL each), dried over Na SO , 24.1; 30.6; 32.9; 38.8; 41.5; 54.6 (2 C); 55.1; 55.4; 62.5;
2
4
filtered, and evaporated. The resulting solid was again 64.9; 66.3; 67.9 (2 C); 117.4; 121.3 (2 C); 128.0; 128.6
dissolved in AcOEt (200 mL) and treated as described (2 C); 129.6 (2 C); 130.4 (2 C); 134.6; 138.0; 138.6; 138.9;
1
+
above to yield an off-white solid (3.26 g, 91%). H- 158.9; 172.3; 172.6; 173.5. ESI-MS: 610.26 ([M+H] ).
NMR (MeOD): 1.34–1.43 (m, 12 H); 1.53–1.59 (m, 2 H);
1
9
.72–1.81 (m, 1 H); 1.86–1.93 (m, 1 H); 2.88 (dd, J=14,
, 1 H); 3.12–3.17 (m, 3 H); 4.36 (dd, J=9, 5, 1 H); 4.47 5 (283 mg, 0.464 mmol) and DIPEA (237 μL, 3 equiv.) in
Morph-Phe-Lys(alloc)-PABC-PNP (6). A solution of
(dd, J=9, 5, 1 H); 4.52 (d, J=5, 2 H); 4.58 (d, J=7, 2 H); dry DMF (9.5 mL) was treated with bis(4-nitrophenyl)-
5
5
7
.18 (dd, J=11, 1, 2 H); 5.29 (dd, J=17, 2, 1 H); 5.89– carbonate (212 mg, 1.5 equiv.), and the mixture was
.96 (m, 1 H); 7.23–7.31 (m, 5 H); 7.34 (d, J=9, 2 H); stirred for 25 h at room temperature. Then, AcOEt
.57 (d, J=9, 2 H).
(100 mL) was added and the organic layer was washed
with 0.01 M KOH (10×30 mL) and brine (1×10 mL),
H-Phe-Lys(alloc)-PABOH·HCl (4). A stirred solution dried over MgSO , filtered, and concentrated in vacuo.
4
of 3 (3.26 g, 5.59 mmol) in EtOH (30 mL) was treated The crude product was purified by flash chromatog-
with HCl conc. (2.34 mL, 5 equiv.). After 20 h the raphy over silica gel, eluting with AcOEt/MeOH 20:1,
solvent was evaporated, the residue was dissolved in a to give the product as a white foam (256 mg, 71%).
1
minimum amount of EtOH and precipitated by drop-
H-NMR (MeOD): 1.42–1.52 (m, 2 H); 1.54–1.61 (m,
wise addition to AcOEt (120 mL). The resulting white 2 H); 1.77–1.84 (m, 1 H); 1.90–1.97 (m, 1 H); 2.26–2.30
solid was filtered off, washed with AcOEt and Et O, (m, 2 H); 2.36–2.41 (m, 2 H); 2.88 (d, J=16, 1 H); 3.02
2
1
and dried in vacuo (1.84 g, 63%). H-NMR (MeOD): (dd, J=14, 9, 1 H); 3.06 (d, J=16, 1 H); 3.14 (t, J=7,
1
1
.42–1.53 (m, 2 H); 1.55–1.61 (m, 2 H); 1.76–1.84 (m, 2 H); 3.27 (dd, J=14, 5, 1 H); 3.60 (t, J=5, 4 H); 4.49–
H); 1.88–1.94 (m, 1 H); 3.07 (dd, J=14, 8, 1 H); 3.14 (t, 4.52 (m, 3 H); 4.77 (dd, J=9, 5, 1 H); 5.17 (dd, J=11, 1,
J=7, 2 H); 3.30–3.34 (m, 1 H); 4.20 (dd, J=8, 6, 1 H); 2 H); 5.29 (dd, J=17, 2, 1 H); 5.30 (s, 2 H); 5.89–5.96 (m,
4
5
.51–4.54 (m, 3 H); 4.60 (s, 2 H); 5.18 (dd, J=11, 1, 2 H); 1 H); 7.18–7.21 (m, 1 H); 7.25–7.28 (m, 4 H); 7.47 (d, J=
.30 (dd, J=17, 2, 1 H); 5.89–5.97 (m, 1 H); 7.25–7.31 9, 2 H); 7.48 (d, J=9, 2 H); 7.68 (d, J=9, 2 H); 8.33 (d,
1
3
+
(
m, 5 H); 7.36 (d, J=9, 2 H); 7.58 (d, J=9, 2 H). C-NMR J=9, 2 H). ESI-MS: 775.47 ([M+H] ), 797.47 ([M+
+
(MeOD): 24.0; 30.6; 33.0; 38.6; 41.4; 55.4; 55.5; 64.8; Na] ).
6
1
1
6.3; 117.4; 121.2 (2 C); 128.6 (2 C); 128.8; 130.1 (2 C);
30.5 (2 C); 134.5; 135.4; 138.6; 138.9; 158.9; 169.5;
Morph-Phe-Lys(alloc)-PABC-Sun (7). To a stirred
solution of 6 (330 mg, 0.426 mmol) in dry DMF
+
71.7. ESI-MS: 483.26 ([M+H] ).
(4.5 mL) were added 4-DMAP (57 mg, 1.1 equiv.) and
Morph-Phe-Lys(alloc)-PABOH (5). To a solution of sunitinib (187 mg, 1.1 equiv.), and the mixture was
4
(500 mg, 0.963 mmol) and Et N (134 μL, 1 equiv.) in stirred at room temperature in the dark. After 69 h
3
www.cb.wiley.com (8 of 12) e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG,
Zurich, Switzerland

Chem. Biodiversity 2018, 15, e1800520
AcOEt (100 mL) was added and the resulting mixture mol): C 59.60, H 6.48 N 11.58; found: C 59.73, H 6.40, N
was stirred for another 5 min. The orange solid was 11.82. (See Figure S1 and S2).
filtered off, washed with small amounts of AcOEt, and
1
dried in vacuo (295 mg, 67%). H-NMR (MeOD/CDCl ):
Boc-Gly-Gly-PABOH (9). To a mixture of Boc-Gly-
3
1
2
.08 (t, J=7, 6 H); 1.36–1.44 (m, 2 H); 1.51–1.56 (m, Gly-OH (200 mg, 0.861 mmol) and PABOH (111 mg,
H); 1.70–1.78 (m, 1 H); 1.86–1.92 (m, 1 H); 2.24–2.28 1.05 equiv.) in THF (3 mL) was added EEDQ (224 mg,
(
m, 2 H); 2.37–2.42 (m, 2 H); 2.49 (s, 3 H); 2.55 (s, 3 H); 1.05 equiv.), and the resulting mixture was stirred for
2
1
1
.62 (q, J=7, 4 H); 2.69 (t, J=6, 2 H); 2.86 (d, J=16, 24 h at room temperature. The solvent was removed,
H); 2.96–3.05 (m, 2 H); 3.12 (t, J=7, 2 H); 3.22 (dd, J= and the residue was dissolved in AcOEt (60 mL). The
4, 5, 1 H); 3.48 (t, J=7, 2 H); 3.55–3.60 (m, 4 H); 4.46 organic layer was washed with 1 M NaHCO , 10%
3
(dd, J=8, 6, 1 H); 4.50 (d, J=5, 2 H); 4.69 (dd, J=8, 5, aqueous citric acid, and brine (30 mL each), dried over
1
2
7
7
1
2
H); 5.16 (d, J=10, 1 H); 5.26 (d, J=17, 1 H); 5.47 (s, Na SO , filtered, and evaporated to yield a slightly
2 4
1
H); 5.84–5.92 (m, 1 H); 6.89 (ddd, J=11, 9, 2, 1 H); orange solid (187 mg, 64%). H-NMR (MeOD): 1.45 (s,
.16–7.23 (m, 5 H); 7.29 (dd, J=8, 2, 1 H); 7.47 (s, 1 H); 9 H); 3.77 (s, 2 H); 4.02 (s, 2 H); 4.56 (s, 2 H); 7.30 (d, J=
.51 (d, J=9, 2 H); 7.64 (d, J=9, 2 H); 7.75 (dd, J=9, 4, 8, 2 H); 7.58 (d, J=8, 2 H).
1
3
H). C-NMR (MeOD/CDCl ): 11.1; 11.5 (2 C); 13.9;
3
3.3; 29.8; 32.2; 37.5; 38.4; 40.9; 47.4 (2 C); 52.3; 54.1
H-Gly-Gly-PABOH (11). A stirred solution of 9
2
(2 C); 54.3; 54.6; 62.0; 65.9; 67.5 (2 C); 69.1; 104.9 ( J = (187 mg, 0.554 mmol) in EtOH (50 mL) was treated
CF
4
2
3
2
1
1
1
1
5); 112.7 ( J =2); 113.6 ( J =24); 116.9 ( J =9); with HCl conc. (231 μL, 5 equiv.). After 18 h the solvent
CF CF CF
17.5; 120.9 (2 C); 121.2; 125.7; 126.9; 127.6; 126.7; was evaporated yielding H-Gly-Gly-PABOH·HCl (10) as
28.0 ( J =10); 129.2 (2 C); 129.7 (2 C); 130.2 (2 C); a slightly orange powder which was used for the next
31.5; 132.2; 133.3; 133.6; 136.9 ; 139.2; 139.6; 151.5; step without further purification (151 mg, 100%). H-
60.9 ( J =241); 167.1; 168.2; 171.4; 171.7; 172.5. ESI- NMR (MeOD): 3.81 (s, 2 H); 4.12 (s, 2 H); 4.59 (s, 2 H);
3
CF
1
1
CF
2
+
MS: 517.76 ([M+2 H] ).
7.34 (d, J=9, 2 H); 7.56 (d, J=9, 2 H). A solution of 10
(
70 mg, 0.26 mmol) in H O (1 mL) was basified to pH 8
2
Morph-Phe-Lys-PABC-Sun·3HCOOH (8). A suspen- with sat. NaHCO . After removal of the solvent, the
3
sion of 7 (125 mg, 0.121 mmol) and 1,3-dimethylbarbi- crude product was purified by flash chromatography
turic acid (21 mg, 1.1 equiv.) in dry DMF (4 mL) was over silica gel, eluting with CH Cl /MeOH 6:1 (1%
2
2
treated with Pd(PPh ) (7 mg, 0.05 equiv.) and stirred Et N), to give the product as an off-white solid (26 mg,
3
4
3
1
for 30 min. Then, the solvent was evaporated. The 43%). H-NMR (MeOD): 3.41 (s, 2 H); 4.08 (s, 2 H); 4.58
residue was purified through RP-HPLC [H O/MeCN (s, 2 H); 7.33 (d, J=9, 2 H); 7.56 (d, J=9, 2 H).
2
(
0.1% HCOOH); 20–61% MeCN] to give the product as
1
an orange solid (43 mg, 33%). H-NMR (MeOD): 1.19 (t,
Morph-Gly-Gly-PABOH (12). To a solution of 11
J=7, 6 H); 1.46–1.59 (m, 2 H); 1.67–1.75 (m, 2 H); (440 mg, 1.85 mmol) and DIPEA (947 μL, 3 equiv.) in
1
2
.78–1.86 (m, 1 H); 1.92–1.99 (m, 1 H); 2.26–2.30 (m, dry DMF (30 mL), morpholin-4-yl-acetic acid (296 mg,
H); 2.35–2.40 (m, 2 H); 2.48 (s, 3 H); 2.51 (s, 3 H); 2.83 1.1 equiv.) and TBTU (715 mg, 1.2 equiv.) were added.
(q, J=7, 4 H); 2.87–2.95 (m, 5 H); 2.99–3.05 (m, 2 H); After stirring for 24 h at room temperature, the solvent
3
9
.23 (dd, J=14, 5, 1 H); 3.55–3.59 (m, 6 H); 4.54 (dd, J= was evaporated, and the crude product was purified
, 5, 1 H); 4.68 (dd, J=9, 5, 1 H); 5.46 (s, 2 H); 6.92 (ddd, by flash chromatography over silica gel, eluting with
J=9, 9, 3, 1 H); 7.16–7.20 (m, 1 H); 7.22–7.27 (m, 4 H); CH Cl /MeOH 16:1 (1% Et N). Further purification on
2
2
3
7
7
.50 (dd, J=9, 3, 1 H); 7.57 (d, J=9, 2 H); 7.61 (s, 1 H); aluminum oxide, eluting with CH Cl /MeOH 1:1, gave
2 2
.68 (d, J=9, 2 H); 7.78 (dd, J=9, 5, 1 H). 13C-NMR the product as a white powder (504 mg, 75%). H-
1
(
MeOD): 9.5; 9.7 (2 C); 12.1; 22.4; 26.9; 31.2; 36.2; 37.2; NMR ((D )DMSO): 2.47 (t, J=5, 4 H); 2.98 (s, 2 H); 3.62
6
3
9.2; 47.0 (2 C); 51.5; 53.2 (2 C); 53.6; 54.0; 61.0; 66.5 (t, J=5, 4 H); 3.81 (d, J=6, 2 H); 3.90 (d, J=6, 2 H); 4.44
2
4
(
(
2 C); 68.0; 104.3 ( J =26); 112.0 ( J =3); 112.5 (d, J=6, 2 H); 5.11 (t, J=6, 1 H); 7.25 (d, J=9, 2 H); 7.56
CF CF
2
3
J =25); 115.8 ( J =9); 119.9 (2 C); 120.3; 125.0; (d, J=9, 2 H); 8.06 (t, J=6, 1 H); 8.27 (t, J=6, 1 H); 9.80
CF
CF
3
1
1
1
1
26.2; 126.7; 127.4 ( J =9); 128.3 (2 C); 129.0 (2 C); (s, 1 H).
29.2 (2 C); 131.3; 131.7; 132.7; 136.5; 138.5 (2 C);
50.8; 160.3 ( J =241); 166.9; 167.3; 170.7; 171.3;
72.3. ESI-MS: 475.75 ([M+2 H] ), 950.49 ([M+H] ). (100 mg, 0.274 mmol) and DIPEA (152 μL, 3 equiv.) in
CF
1
Morph-Gly-Gly-PABC-PNP (13). A solution of 12
CF
2
+
+
Anal. calc. for C H FN O ·3 HCOOH (M =1088.18 g/ dry DMF (3 mL) was treated with bis(4-nitrophenyl)-
51
64
9
8
r
carbonate (125 mg, 1.5 equiv.), and the mixture was
www.cb.wiley.com (9 of 12) e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG,
Zurich, Switzerland

Chem. Biodiversity 2018, 15, e1800520
stirred for 24 h at room temperature. Then, AcOEt Biology
(100 mL) was added and the organic layer was washed
with 0.01 M KOH (10×30 mL), dried over MgSO4, Cathepsin B Cleavage Assay. Cathepsin B (human liver,
filtered, and concentrated in vacuo. The crude product 0.47 mg/mL, 324 U/mg) was obtained from Merck
was purified by flash chromatography over silica gel, Millipore, Merck KGaA. The following protocol was
eluting with CH Cl /MeOH 15:1, to give the product as applied to the test compounds: 5 μL of cathepsin B
2
2
1
a white foam (112 mg, 77%). H-NMR ((D )DMSO): 2.47 stock were added to 10 μL of activation buffer (30 mM
6
(
t, J=5, 4 H); 2.99 (s, 2 H); 3.62 (t, J=5, 4 H); 3.82 (d, J= DTT/15 mM EDTA-Na2 in H O) and incubated for
2
6
2
1
, 2 H); 3.92 (d, J=6, 2 H); 5.26 (s, 2 H); 7.44 (d, J=9, 15 min at room temperature. The solution was diluted
H); 7.58 (d, J=9, 2 H); 7.67 (d, J=9, 2 H); 8.07 (t, J=6, with 1.5 mL of 25 mM sodium acetate buffer (1 mM
H); 8.30 (t, J=6, 1 H); 8.33 (d, J=9, 2 H); 9.96 (s, 1H). EDTA, pH 5, preheated to 37°C), and the reaction was
initiated by the addition of 7 μL of compound stock
Morph-Gly-Gly-PABC-Sun (14). To a stirred solu- solution (10 mM in MeOH; final concentration: ca.
tion of 13 (62 mg, 0.12 mmol) in dry DMF (2.5 mL) 46 μM). Aliquots were withdrawn and the reactions
were added 4-DMAP (16 mg, 1.1 equiv.) and sunitinib were followed by RP-HPLC as a function of time. The
(
50 mg, 1.1 equiv.), and the mixture was stirred at HPLC chromatogram peak areas at 420 nm were used
room temperature in the dark. After 72 h, AcOEt to determine both the disappearance of the starting
40 mL) was added and the resulting mixture was material as well as the appearance of sunitinib.
(
stirred for another 10 min. The orange solid was Between the time points, all solutions were incubated
filtered off, washed with small amounts of AcOEt, and at 37°C. Reference solutions containing 7 μL of
1
dried in vacuo (38 mg, 41%). H-NMR ((D )DMSO): 0.98 compound stock solution, 1.5 mL of buffer solution,
6
(t, J=7, 6 H); 2.45–2.55 (m, 16 H); 2.97 (s, 2 H); 3.29 (dt, and 15 μL of activation buffer were prepared and
J=7, J=6, 2 H); 3.60 (t, J=5, 4 H); 3.80 (d, J=6, 2 H); analyzed under the same conditions and at the same
3
1
2
3
1
.91 (d, J=6, 2 H); 5.41 (s, 2 H); 7.05 (ddd, J=9, 9, 3, time points in order to distinguish between cathepsin
H); 7.50 (d, J=9, 2 H); 7.60 (t, J=6, 1 H); 7.66 (d, J=9, B-based activation and enzyme-free hydrolysis. A
H); 7.75 (dd, J=9, 5, 1 H); 7.78 (s, 1 H); 7.93 (dd, J=9, solution of sunitinib (prepared as described for the
, 1 H); 8.08 (t, J=6, 1 H); 8.32 (t, J=6, 1 H); 9.98 (s, reference measurements) served as standards for the
1
3
H); 12.75 (s, 1 H). C-NMR ((D )DMSO): 11.1; 12.3 quantification of the reaction product.
6
(
2 C); 13.8; 37.5; 42.3; 43.1; 47.0 (2 C); 52.1; 53.8 (2 C);
2
4
6
2
1
1
2
1.9; 66.6 (2 C); 68.4; 105.9 ( J =26.3); 111.3 ( J =
Chemicals for Cell Culture Tests. Sunitinib malate
CF
CF
2
3
.7); 113.0 ( J =24.0); 116.1 ( J =8.9); 119.6 (2 C); (from LC Laboratories®) and all other investigated
CF
CF
3
22.4; 126.3; 126.6; 128.0 ( J =10.0); 129.6 (2 C); compounds were dissolved in DMSO. These stock
CF
1
30.6; 131.8; 133.6; 139.3; 139.4; 150.8; 160.0 ( J = solutions were further diluted into culture media at
CF
38.9); 164.6; 166.8; 168.3; 169.8; 170.3. ESI-MS: 789.54 the indicated concentrations. The final DMSO concen-
+
(
[M+H] ). Anal. calc. for C H FN O ·H O (M = trations were always less than 1%.
40
49
8
8
2
r
8
06.88 g/mol): C 59.54, H 6.37, N 13.89; found: C 59.55,
H 6.33, N 13.76. (See Figure S3 and S4).
Cell Culture. The cell models used in this study are
summarized in Table S1 together with their culture
media as well as the respective sources. All cells were
Stability Measurements in Aqueous Buffer
cultivated in humidified incubators (37°C, 21% O , 5%
2
Compounds 8, 14, and sunitinib were dissolved in CO ) in the indicated culture medium containing 10%
2
1
0 mM sodium phosphate buffer (pH 7.4, 37°C) or Mc fetal calf serum (PAA, Linz Austria). Cell cultures were
Coy’s medium (+10% fetal calf serum) containing 1% periodically checked for Mycoplasma contamination.
DMSO at a concentration of 10 μM and incubated at
3
3
7°C. Aliquots were withdrawn at different time points
Cytotoxicity Assay. Cells were plated (2×10 cells/
(0–24 h, every 44 min) and subjected to RP-HPLC well) in 96-well plates and allowed to recover for 24 h.
analysis. The HPLC chromatogram peak areas at Subsequently, the dissolved drugs were added. After
4
20 nm were used to calculate the concentration of 72 h of drug exposure, the proportion of viable cells
the remaining prodrug. The sunitinib chromatograms was determined by MTT assay following the manufac-
were used as reference for the quantification of turer’s recommendations (EZ4U, Biomedica, Vienna,
formed free drug. Unknown peaks were identified Austria). Cytotoxicity was expressed as IC50 values
using mass spectrometry. (See Figures S5–S10).
www.cb.wiley.com (10 of 12) e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG,
Zurich, Switzerland

Chem. Biodiversity 2018, 15, e1800520
calculated from full dose-response curves using Graph References
Pad Prism software.
[1] N. J. Harper, ‘Drug Latentiation’, J. Med. Pharm. Chem.
1
959, 1, 467–500.
Western Blot Analysis. To assess the impact of the
tested compounds on receptor tyrosine kinase signal-
ing, cells were seeded in 6-well plates and (without
further manipulation) treated with the compounds as
indicated. For protein isolation, cells were harvested
and lysed, and the protein extracts resolved by SDS/
PAGE and transferred onto a polyvinylidene difluoride
membrane for Western blotting as previously de-
[2] Y. Singh, M. Palombo, P. J. Sinko, ‘Recent Trends in
Targeted Anticancer Prodrug and Conjugate Design’, Curr.
Med. Chem. 2008, 15, 1802–1826.
[
3] F. Kratz, I. A. Müller, C. Ryppa, A. Warnecke, ‘Prodrug
Strategies in Anticancer Chemotherapy’, ChemMedChem
2008, 3, 20–53.
[
4] W. R. Wilson, M. P. Hay, ‘Targeting hypoxia in cancer
therapy’, Nat. Rev. Cancer 2011, 11, 393–410.
[5] J. Vandooren, G. Opdenakker, P. M. Loadman, D. R. Ed-
wards, ‘Proteases in cancer drug delivery’, Adv. Drug
Delivery Rev. 2016, 97, 144–155.
[33]
scribed.
The details on the antibodies used are
given in Table S2. Additionally, horseradish peroxidase-
labeled secondary antibodies from Santa Cruz Biotech-
nology were used at working dilutions of 1:10000.
[
6] M. M. Mohamed, B. F. Sloane, ‘Cysteine cathepsins: multi-
functional enzymes in cancer’, Nat. Rev. Cancer 2006, 6,
7
64–775.
[
7] J. Kos, A. Mitrović, B. Mirković, ‘The current stage of
cathepsin B inhibitors as potential anticancer agents’,
Future Med. Chem. 2014, 6, 1355–1371.
Gelatin Zymography. Cathepsin B activity in total
cell extracts was determined by gelatin zymography.
To this end, 20 mg of the cell lysates were resolved by
SDS/Page using 10% SDS-polyacrylamide gels contain-
ing 10% gelatin (SigmaÀ Aldrich) under ice-cooled
conditions. The gels were washed four times in
[
[
8] C. S. Gondi, J. S. Rao, ‘Cathepsin B as a cancer target’, Expert
Opin. Ther. Targets 2013, 17, 281–291.
9] S. Borchmann, B. von Tresckow, A. Engert, ‘Current
developments in the treatment of early-stage classical
Hodgkin lymphoma’, Curr. Opin. Oncol. 2016, 28, 377–383.
washing buffer [containing 2.5% Triton X-100, 50 mM [10] U. H. Weidle, G. Tiefenthaler, G. Georges, ‘Proteases as
Activators for Cytotoxic Prodrugs in Antitumor Therapy’,
Cancer Genomics Proteomics 2014, 11, 67–79.
sodium acetate, 100 mM sodium chloride (pH 5.5) and
1
0 mM L-cystine] to remove the SDS. The gels were
[
11] K. Miller, R. Erez, E. Segal, D. Shabat, R. Satchi-Fainaro,
then incubated overnight in an incubation buffer
containing 50 mM sodium acetate, 100 mM sodium
chloride (pH 5.5), 10 mM L-cystine and 5 mM EDTA.
Next, the gels were stained with Coomassie Blue R250 [12] B. L. Barthel, D. L. Rudnicki, T. P. Kirby, S. M. Colvin, D. J.
for 2 to 3 h. Gelatinolytic activities were identified as
‘
Targeting Bone Metastases with a Bispecific Anticancer
and Antiangiogenic Polymer–Alendronate–Taxane Conju-
gate’, Angew. Chem. Int. Ed. 2009, 48, 2949–2954.
Burkhart, T. H. Koch, ‘Synthesis and Biological Character-
ization of Protease-Activated Prodrugs of Doxazolidine’, J.
Med. Chem. 2012, 55, 6595–6607.
clear zones of lysis against a dark background. (See
Figures S11–S13).
[
13] C. Potrich, R. Tomazzolli, M. Dalla Serra, G. Amderluh, P.
Malovrh, P. Maček, G. Menestrina, M. Tejuca, ‘Cytotoxic
Activity of a Tumor Protease-Activated Pore-Forming
Toxin’, Bioconjugate Chem. 2005, 16, 369–376.
14] Y.-J. Zhong, L.-H. Shao, Y. Li, ‘Cathepsin B-cleavable
doxorubicin prodrugs for targeted cancer therapy’, Int. J.
Oncol. 2013, 42, 373–383.
Acknowledgements
[
This work was financially supported by the Initiative
Krebsforschung of the Medical University Vienna (to
P. H.) and the Austrian Science Fund (FWF) (grant
[
15] Vol. 2016, U. S. Department of Health and Human Services,
http://www.fda.gov/Drugs/default.htm (03.05.2016).
number P28853, to C. R. K.). We thank Dr. Wolfgang [16] D. R. Shah, R. R. Shah, J. Morganroth, ‘Tyrosine Kinase
Inhibitors: Their On-Target Toxicities as Potential Indicators
of Efficacy’, Drug Saf. 2013, 36, 413–426.
17] M. E. Lacouture, ‘Mechanisms of cutaneous toxicities to
EGFR inhibitors’, Nat. Rev. Cancer 2006, 6, 803–812.
18] A. Bühler, S. Berger, F. Bengsch, G. Martin, H. Han, S.
Vierkotten, A. Pielen, D. Boehringer, G. Schlunck, S. Fauser,
H. T. Agostini, T. Reinheckel, A. Stahl, ‘Cathepsin proteases
promote angiogenic sprouting and laser-induced choroidal
neovascularisation in mice’, Exp. Eye Res. 2013, 115, 73–78.
19] R. R. Malla, S. Gopinath, C. S. Gondi, K. Alapati, D. H. Dinh,
M. Gujrati, J. S. Rao, ‘Cathepsin B and uPAR knockdown
inhibits tumor-induced angiogenesis by modulating VEGF
expression in glioma’, Cancer Gene Ther. 2011, 18, 419–
Kandioller, University of Vienna, for helpful discussions
and Ing. Ingo Polivka for synthetic support.
[
[
Author Contribution Statement
C. K. and M. M. performed the synthetic experiments
and analyzed the data. B. K. and K. H. performed the
biological experiments and analyzed the data. C. R. K.
and P. H. designed the experiments. C. K., C. R. K., P. H.,
W. B., and B. K. K. wrote/corrected the article.
[
4
34.
www.cb.wiley.com (11 of 12) e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG,
Zurich, Switzerland

Chem. Biodiversity 2018, 15, e1800520
[
[
[
20] E. Im, A. Venkatakrishnan, A. Kazlauskas, ‘Cathepsin B
synthesis and secretion in pheochromocytoma tumor cells
by blocking VEGF receptor 2 via PLC-gamma-related path-
ways’, Am. J. Physiol. Endocrinol. Metab. 2012, 303, E1006–
E1014.
Regulates the Intrinsic Angiogenic Threshold of Endothelial
Cells’, Mol. Biol. Cell 2005, 16, 3488–3500.
21] P. Wu, T. E. Nielsen, M. H. Clausen, ‘FDA-approved small-
molecule kinase inhibitors’, Trends Pharmacol. Sci. 2015,
[28] A. Piscazzi, E. Costantino, F. Maddalena, M. I. Natalicchio,
A. M. T. Gerardi, R. Antonetti, M. Cignarelli, M. Landriscina,
‘Activation of the RAS/RAF/ERK Signaling Pathway Contrib-
utes to Resistance to Sunitinib in Thyroid Carcinoma Cell
Lines’, J. Clin. Endocrinol. Metab. 2012, 97, E898–E906.
[29] I. Chiyanzu, E. Hansell, J. Gut, P. J. Rosenthal, J. H.
McKerrow, K. Chibale, ‘Synthesis and evaluation of isatins
and thiosemicarbazone derivatives against cruzain, falci-
pain-2 and rhodesain’, Bioorg. Med. Chem. Lett. 2003, 13,
3527–3530.
3
6, 422–439.
22] G. M. Dubowchik, R. A. Firestone, ‘Cathepsin B-sensitive
dipeptide prodrugs. 1. A model study of structural require-
ments for efficient release of doxorubicin’, Bioorg. Med.
Chem. Lett. 1998, 8, 3341–3346.
[
23] G. M. Dubowchik, R. A. Firestone, L. Padilla, D. Willner, S. J.
Hofstead, K. Mosure, J. O. Knipe, S. J. Lasch, P. A. Trail,
‘
Cathepsin B-Labile Dipeptide Linkers for Lysosomal Re-
lease of Doxorubicin from Internalizing Immunoconju-
gates: Model Studies of Enzymatic Drug Release and
Antigen-Specific In Vitro Anticancer Activity’, Bioconjugate
Chem. 2002, 13, 855–869.
[30] R. A. Iyer, P. E. Hanna, ‘N-(Carbobenzyloxy)Isatin - a Slow
Binding Alpha-Keto Lactam Inhibitor of Alpha-Chymotryp-
sin’, Bioorg. Med. Chem. Lett. 1995, 5, 89–92.
[
24] Q. Wang, Y.-J. Zhong, J.-P. Yuan, L.-H. Shao, J. Zhang, L.
Tang, S.-P. Liu, Y.-P. Hong, R. A. Firestone, Y. Li, ‘Targeting
therapy of hepatocellular carcinoma with doxorubicin
prodrug PDOX increases anti-metastatic effect and reduces
toxicity: a preclinical study’, J. Transl. Med. 2013, 11, 192.
25] L. Tang, R. Duan, Y.-J. Zhong, R. A. Firestone, Y.-P. Hong, J.-
G. Li, Y.-C. Xin, H.-L. Wu, Y. Li, ‘Synthesis, identification and
in vivo studies of tumor-targeting agent peptide doxorubi-
cin (PDOX) to treat peritoneal carcinomatosis of gastric
cancer with similar efficacy but reduced toxicity’, Mol.
Cancer 2014, 13, 44.
[31] N. I. Howard, T. D. H. Bugg, ‘Synthesis and activity of 5’-
Uridinyl dipeptide analogues mimicking the amino termi-
nal peptide chain of nucleoside antibiotic mureidomycin
A’, Bioorg. Med. Chem. 2003, 11, 3083–3099.
[32] R. Erez, E. Segal, K. Miller, R. Satchi-Fainaro, D. Shabat,
‘Enhanced cytotoxicity of a polymer–drug conjugate with
triple payload of paclitaxel’, Bioorg. Med. Chem. 2009, 17,
4327–4335.
[33] C. Karnthaler-Benbakka, D. Groza, K. Kryeziu, V. Pichler, A.
Roller, W. Berger, P. Heffeter, C. R. Kowol, ‘Tumor-Targeting
of EGFR Inhibitors by Hypoxia-Mediated Activation’, Angew.
Chem. Int. Ed. 2014, 53, 12930–12935.
[
[
26] J. Zhang, L. He, X. F. Geng, R. A. Firestone, Y. P. Hong, Y. Li,
‘Anti-cancer effects of novel doxorubicin prodrug PDOX in
MCF-7 breast cancer cells’, J. Huazhong Univ. Sci. Technol.
Med. Sci. 2014, 34, 521–528.
27] Y. Aita, K. Ishii, Y. Saito, T. Ikeda, Y. Kawakami, H. Shimano,
H. Hara, K. Takekoshi, ‘Sunitinib inhibits catecholamine
[
Received October 2, 2018
Accepted November 15, 2018
www.cb.wiley.com (12 of 12) e1800520 © 2018 The Authors. Chemistry & Biodiversity Published by Wiley-VHCA AG,
Zurich, Switzerland