A β-glucuronidase-responsive albumin-binding prodrug for potential selective kinase inhibitor-based cancer chemotherapy

Article abstract of DOI:10.1016/j.ejmech.2018.08.100

We report on the synthesis and in vitro biological evaluations of a nanomolar protein kinase inhibitor (PKI) and its β-glucuronidase-responsive albumin-binding prodrug. The highly potent PKI is 400–3400 times more cytotoxic than the well-known PKI Roscovitine. The prodrug is able to bind covalently to human serum albumin through Michael addition and release the cytotoxic PKI in the presence of β-glucuronidase, an enzyme over-expressed in the microenvironment of solid tumours.

Full text of DOI:10.1016/j.ejmech.2018.08.100

Accepted Manuscript
A β-glucuronidase-responsive albumin-binding prodrug for potential selective kinase
inhibitor-based cancer chemotherapy
Guillaume Compain, Nassima Oumata, Jonathan Clarhaut, Elodie Péraudeau, Brigitte
Renoux, Hervé Galons, Sébastien Papot
PII:
S0223-5234(18)30772-4
10.1016/j.ejmech.2018.08.100
EJMECH 10711
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 30 May 2018
Revised Date: 28 August 2018
Accepted Date: 31 August 2018
Please cite this article as: G. Compain, N. Oumata, J. Clarhaut, E. Péraudeau, B. Renoux, Hervé.
Galons, Sé. Papot, A β-glucuronidase-responsive albumin-binding prodrug for potential selective kinase
inhibitor-based cancer chemotherapy, European Journal of Medicinal Chemistry (2018), doi: 10.1016/
j.ejmech.2018.08.100.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
A β-glucuronidase-responsive albumin-binding prodrug for selective kinase inhibitor-
based cancer chemotherapy
Guillaume Compain, Nassima Oumata, Jonathan Clarhaut, Elodie Péraudeau, Brigitte Renoux,
Hervé Galons and Sébastien Papot

ACCEPTED MANUSCRIPT
A β-glucuronidase-responsive albumin-binding prodrug for potential
selective kinase inhibitor-based cancer chemotherapy
Guillaume Compain,a Nassima Oumata,b Jonathan Clarhaut,a,c Elodie Péraudeau,c Brigitte
Renoux,a Hervé Galonsb and Sébastien Papot*a
a Institut de Chimie des Milieux et des Matériaux de Poitiers (IC2MP), Université de
Poitiers, CNRS, groupe “Systèmes Moléculaires Programmés”, 4 rue Michel Brunet,
TSA 51106, F-86073 Poitiers, France. E-mail: sebastien.papot@univ-poitiers.fr.
b ManRos Therapeutics, Hôtel de Recherche, Centre de Perharidy, 29680 Roscoff,
France.
c CHU de Poitiers, 2 rue de la Miléterie, CS 90577, F-86021 Poitiers, France.
*Corresponding Author: Phone: +33 549 453 862. E-mail: sebastien.papot@univ-poitiers.fr.
Keywords: cancer, chemotherapy, drug delivery, self-immolative linker, β-glucuronidase-
responsive albumin-binding prodrug, protein kinase inhibitor, tumour microenvironment
Abstract. We report on the synthesis and in vitro biological evaluations of a nanomolar
protein kinase inhibitor (PKI) and its
β-glucuronidase-responsive albumin-binding
prodrug. The highly potent PKI is 400 TO 3400 times more cytotoxic than the well-known
PKI Roscovitine. The prodrug is able to bind covalently to human serum albumin
through Michael addition and release the cytotoxic PKI in the presence of β-
glucuronidase, an enzyme over-expressed in the microenvironment of solid tumours.

ACCEPTED MANUSCRIPT
Introduction
During the last two decades, protein kinase inhibitors have emerged as one of the most
important class of anticancer agents and more than 28 PKIs already received Food and Drug
Administration approval for applications in oncology [1–3]. However, almost all PKI
developed so far induce severe adverse effects that hamper their therapeutic efficacy in clinic
[4–7]. Therefore, the design of more selective PKIs is of great interest while remaining highly
difficult since 518 genes encodes for kinase proteins [8]. An attractive alternative to skirt this
challenging task relies on the development of molecular systems programmed for the
selective delivery of PKI at the tumour site. Such a therapeutic approach is already validated
in human with the clinical use of antibody-drug conjugates designed for the targeting of
potent cytotoxics [9]. Within this framework, our research team developed recently the
concept of β-glucuronidase-responsive albumin-binding prodrugs [10, 11]. These latter
combine a glucuronide trigger [12, 13], a self-immolative linker [14, 15], a potent anticancer
drug and a maleimide-bearing side chain in a single entity (Figure 1). Following intravenous
administration, such prodrugs bind covalently to plasmatic albumin (step 1), thereby enabling
their passive accumulation in tumours [16] where extracellular β-glucuronidase [17-20]
launches the selective release of the targeted drug (step 2). This targeting strategy has led to
outstanding therapeutic efficacy in mice, particularly for the treatment of triple-negative
mammary and pancreatic cancers [11]. Thus, with the aim to enhance the selectivity of kinase
inhibitor-based cancer chemotherapy, we now report on the study the β-glucuronidase-
responsive albumin-binding prodrug 1 of the novel nanomolar PKI 2 (Figure 2).
Results and Discussion
In the course of our intensive efforts devoted to the discovery of new potent PKI, we focused
on the synthesis of Roscovitine analogues (Figure 2) [21]. Roscovitine is indeed a well-known

ACCEPTED MANUSCRIPT
cyclin dependant kinase inhibitor that showed promising results in various animal models,
hence prompting its transfer to the clinic [22]. However, Roscovitine did not demonstrate
significant anticancer therapeutic efficacy in human while inducing severe side effects due to
its lack of selectivity. In the present study, compound 2 was selected from our screening since
it possesses an amine functional group that is required for the attachment of the drug on the
linker of our drug delivery system (Figures 1 and 2). PKI 2 was prepared in four steps from
the 8-isopropyl-2-(methylthio)pyrazolo[1,5-a][1,3,5]triazin-4-ol 3 previously described in the
literature (Scheme 1, see the supporting information). Thus, halogenation of the hydroxylated
pyrazolotriazine 3 carried out in the presence of POCl3 afforded the intermediate 4 (97%).
The 4-(2-pyridyl)benzylamine moiety was incorporated by nucleophilic substitution to
produce the derivative 5 in 98% yield. Oxidation of the thioether using 3-chloroperbenzoic
acid provided the sulfoxide 6 (90%) which reacted with trans-1,4-diaminocyclohexane to give
the expected PKI 2 (88%).
The inhibition profile of compound
compared to that of Roscovitine (Table 1, see the supporting information). In each
case, inhibited kinase activity at lower concentrations than Roscovitine. With the
exception of CLK3, PKI was both efficient and versatile for kinase activity inhibition
with IC50 values ranging from 0.049 to 0.12 µM. Since multiple oncogenic pathways
2 was then evaluated on 14 different kinases and
2
2
are often involved in tumour growth, the high affinity of
can be advantageous for cancer therapy.
2 for a wide range of kinases
We next pursued our investigations by measuring the antiproliferative activity of compound 2
on several cancer cell lines (Table 2). In these experiments, 2 was from 400- to 3400-fold
more cytotoxic than Roscovitine used as a reference. Furthermore, with IC50 values in the low
nanomolar (from 5 to 71 nM), the PKI 2 appeared as an excellent candidate for targeting
strategies employing glucuronide prodrugs such as 1 [13, 23].

ACCEPTED MANUSCRIPT
The synthesis of prodrug 1 was undertaken starting from the activated carbonate 7 [11] we
recently reported in the literature (Scheme 2). Coupling between 7 and the primary amine of
the drug 2 carried out in the presence of pyridine gave the alkyne 8 in 72% yield. The latter
reacted in CH2Cl2 with commercially available O-(2-aminoethyl)-O’-(2-azidoethyl)
nonaethylene glycol and Cu(CH3CN)4PF6 in order to form the triazole 9 through the well-
known copper(I)-catalyzed azide-alkyne 1,3-cycloaddition (CuAAC). After 1 hour under
these conditions, the reaction mixture was washed with an aqueous solution of EDTA, the
solvent was removed, and the crude product was engaged in the next step without any further
purification. The full deprotection of the glucuronide moiety using LiOH produced the
intermediate 10 which was directly coupled with the N-hydroxysuccinimide ester 11 to afford
the prodrug 1 in 40% yield over three steps after purification by reverse phase column
chromatography (purity > 95%).
Since β-glucuronidase-catalysed hydrolysis of the carbohydrate trigger is the key step
in the process of drug release, our first aim was to confirmed that the glucuronide
moiety was still substrate for the enzyme once bound to albumin. For this purpose,
prodrug
of the mixture was monitored by HPLC-HRMS (see the supporting information).
Under these conditions, prodrug disappeared along the time as the result of its
binding to HSA. In this experiment, no release of the PKI was observed after 20
1 was incubated with human serum albumin (HSA) at 37°C and the evolution
1
2
hours of incubation showing that our targeting device was highly stable in the absence
of the activating enzyme. However, addition of β-glucuronidase in the medium
triggered the release the drug indicating that the glucuronide was readily accessible for
the enzyme even with the proximity of the bulky HSA (Figure 3).

ACCEPTED MANUSCRIPT
Glucuronide
1
was then evaluated for its antiproliferative activity against A549 cancer
cell line (Table 3). When incubated alone, prodrug
the highest tested dose of 200 nM whereas PKI was highly toxic with an IC50 of 10.3
nM. As expected, derivatisation of drug
responsive albumin-binding prodrug drastically reduced its toxicity
1 did not affect viability of cells at
2
2
in the form of the
β
-glucuronidase-
1
. This decreased
cytotoxicity can be explained by the hydrophilicity imparted by both the glucuronide
and the polyethylene glycol-containing side chain that prevent passive penetration of
the prodrug through cell membrane and subsequent intracellular activation by
lysosomal
β-glucuronidase. In contrast, addition of β-glucuronidase in the culture
medium triggered the full release of the free drug thereby restoring its initial
cytotoxicity (IC50 9.4 nM). Overall, these results demonstrated that the biological
activity of the PKI
responsive targeting assembly such as
well as the resulting cytotoxicity of
2
can be turned off through its incorporation into an enzyme-
. Moreover, kinases inhibition properties as
can be fully recovered in the presence of β-
1
2
glucuronidase, an enzyme over-expressed in the microenvironment of almost all solid
tumours.
Conclusion
In summary, we synthesised a novel PKI that exhibits a nanomolar activity for the
inhibition of a wide range of protein kinase
albumin-binding prodrug is far less toxic than the free drug and is readily activated in the
presence of -glucuronidase. Since the selective delivery of potent anticancer drugs by β-
. The corresponding enzyme-responsive
β
glucuronidase-responsive albumin-binding prodrugs was already demonstrated in vivo, our
study can be of great interest to enhance the selectivity of kinase inhibitor-based cancer
chemotherapy.

ACCEPTED MANUSCRIPT
Experimental Section
General chemical procedure. All reactions were performed under a nitrogen atmosphere.
Unless otherwise stated, solvents used were of HPLC quality. Chemicals were of analytical
grade from commercial sources and were used without further purification. The reaction
progress was monitored on precoated silica gel TLC plates MACHEREY-NAGEL
ALUGRAM® SIL G/UV254 (0.2 mm silica gel 60). Spots were visualized under 254 nm UV
light. Flash columns chromatography of compounds were performed using a Combi-flash Rf
200i apparatus using a silica phase column or a C18 reversed phase column.
13
1H and C NMR spectra were respectively recorded at 400 MHz and 100 MHz on a Bruker
400 Avance III instrument, equipped with an ultra-shielded magnet and a BBFO 5 mm
13
broadband probe. The C NMR of the compound 2 was recorded by the CRMPO (Centre
Régional de Mesures Physiques de l’Ouest), at the University of Rennes 1, on a Bruker
Avance III HD 500 MHz. Chemical shifts (δ) are reported in parts per million (ppm) from low
to high field and referenced to residual solvent. Standard abbreviations indicating multiplicity
are used as follows: br = broad, s = singlet, d = doublet, t = triplet, quin = quintet, m =
multiplet. High resolution ESI mass spectrometry were carried out on a LC-QTOF MaXis
Impact spectrometer (Bruker).
Analytical RP-HPLC was carried out on a Dionex Ultimate 3000 system equipped with a
UV/Visible variable wavelength detector and with a reverse-phase column chromatography
Acclaim(R) (C18, 250x4.6 mm, 5 µm, 120 Å, column A) or Nucleoshell (C18, 150x4.6 cm, 5
µm, column B) at 30°C and 1 mL.min-1 with a linear gradient composed of A (0.2% TFA in
water) and B (CH3CN) beginning with A/B = 80/20 v/v and reaching A/B = 0/100 v/v within
30 min (method 1). All chromatograms were recorded at 254 nm.

ACCEPTED MANUSCRIPT
HPLC/HRMS experiments were performed on an Accela UHPLC system coupled to a
hybrid high resolution mass spectrometer (Q-ExactiveTM, Thermo Scientific). An
Acclaim® C18 column (250x4.6 mm, 5 µm, 120 Å) at 30°C was used for
chromatographic separation at a flow rate of 0.5 mL.min-1. The column effluent was
introduced into the electrospray ionisation source (ESI) of the mass spectrometer.
Analyses were performed in positive ion mode. The electrospray voltage was set at 4.0
kV. The capillary and heater temperatures were 275°C and 300°C respectively. The
sheath, sweep and auxiliary gas (nitrogen) flow rates were set at 35, 10 and 20
(arbitrary units). Analysis of data was performed with XcaliburTM software.
Enzymatic hydrolysis was monitored by HPLC/HRMS using a linear gradient
composed of A (0.1% formic acid in water) and B (0.1% formic acid in CH3CN)
(method 2). After injection of 20 µL of sample, compounds were separated through a
linear gradient starting from 20% of B and reaching 100% of B within 30 min.
Synthesis of compound 8: The carbonate 7 (0.0329 mmol, 22.6 mg) and the amine 2 (0.0329
mmol, 15 mg) were dissolved in 0.63 mL of dry DMF. 0.16 mL of pyridine was added
followed by HOBt (0.0329 mmol, 4.4 mg). The mixture was stirred at room temperature for
19h. Then, the solvent was evaporated under high vacuum. The crude was purified by flash
chromatography using combiflash® apparatus with a 4 gr gold column from redisep®. A
progressive gradient of DCM/MeOH was used from 100/0 to 95/5 which afforded 24 mg of
1
the desired compound 8 (72%). H NMR (400 MHz MHz, CDCl3): δ 8.69 (d, 1H, J = 4.5
Hz), 7.96 (d, 2H, J = 8.3 Hz), 7.86 (dd, 1H, J = 2.2 Hz, J = 3.6 Hz), 7.74 (ddd, 1H, J = 15.5
Hz, J = 8.1 Hz, J = 2.0 Hz), 7.71 (td, 1H, J = 8.0 Hz, J = 1.2 Hz), 7.63 (s, 1H), 7.55 (dt, 1H, J
= 8.6 Hz, J = 1.8 Hz), 7.44 (d, 1H, J = 8.2 Hz), 7.35 (dd, 1H, J = 8.6 Hz, J = 1.9 Hz), 7.24
(ddd, 2H, J = 7.1 Hz, J = 4.9 Hz, J = 1.3 Hz), 5.76 (t, 1H, J = 6.2 Hz), 5.39-5.28 (m, 4H), 5.20
(dd, 1H, J = 6.9 Hz, J = 3.8 Hz), 4.83 (br s, 1H), 4.76 (d, 3H, J = 5.8 Hz), 4.21 (d, 1H, J = 8.9

ACCEPTED MANUSCRIPT
Hz), 3.77 (br s, 1H), 3.73 (d, 3H, J = 0.9 Hz), 3.45 (br s, 1H), 3.03 (td, 1H, J = 13.7 Hz, J =
6.9 Hz), 2.74 (m, 2H), 2.12 (d, 3H, J = 1.4 Hz), 2.06 (s, 3H), 2.05 (d, 3H, J = 0.4 Hz), 2.01 (t,
13
1H, J = 2.5 Hz), 1.93-2.24 (m, 6H), 1.28 (d, 6H, J = 6.9 Hz), 1.18-1.37 (m, 2H). C NMR
(100 MHz, CDCl3): δ 170.0, 169.3, 169.2, 166.7, 166.7, 156.8, 156.1, 154.1, 149.7, 149.0,
148.8, 143.6, 141.1, 141.0, 139.0, 138.1, 136.8, 135.9, 135.8, 132.2, 132.1, 127.9, 127.3,
123.3, 123.0, 122.2, 120.5, 119.8, 119.7, 99.7, 99.7, 78.6, 78.5, 72.6, 72.6, 72.1, 71.7, 71.1,
71.1, 70.2, 68.7, 68.7, 53.0, 49.7, 44.1, 31.8, 31.5, 26.5, 26.5, 23.4, 23.2, 20.6, 20.5, 20.5.
13
Since the compound 8 was obtained as a mixture of two diastereoisomers, some C NMR
signals are split. HMRS (ESI+) for C50H55N9O14, calcd 1006.39467, found 1006.3948.
Synthesis of the compound 1: Under argon, the alkyne 8 (0.012 mmol, 12mg) was dissolved
in DCM (0.61 mL). Then, a solution of the O-(2-aminoethyl)-O’-(2-azidoethyl) nonaethylene
glycol in DCM (0.5 mL) was added, followed by Cu(CH3CN)4PF6. The mixture was stirred at
room temperature for 15h. A solution of EDTA (0.202 mmol, 75 mg) in water (1.1 mL) was
added. After stirring at room temperature for 1h, the mixture was diluted with water and
extracted four times with DCM. The organic layer was dried over MgSO4 and the solvent was
evaporated. The crude mixture was dissolved in 3 mL of MeOH and 0.2 mL of DCM and the
solution was cooled down at 0°C. Then, a solution at 0°C of LiOH.H2O (0.106 mmol, 4.4 mg)
in water (0.91 mL) was added. Two additional portions of LiOH.H2O (0.0.025 mmol, 1.1 mg)
in water (0.23 mL) were added after stirring for 1h30 and 2h respectively. After 0.5h,
Amberlyst acid resin was added and the mixture was stirred for 0.5h. The solution was filtered
on celite which was washed with MeOH and drops of water. The solvent was evaporated
under high vacuum. The crude mixture was dissolved in 0.29 mL of dry DMSO and NHS
ester 11 (0.014 mmol, 4.4 mg) was added. The reaction mixture was stirred at room
temperature overnight and DMSO was removed under high vaccum. The crude mixture was
purified on reverse phase column using combiflash apparatus with a H2O/MeCN gradient

ACCEPTED MANUSCRIPT
(from 80/20 to 0/100 in 35 minutes). This procedure afforded the desired prodrug 1 (7.5 mg,
1
40% over 3 steps). H NMR (400 MHz, CDCl3): δ 10.08 (bs, 1H), 8.66 (d, 1H, J = 4.5 Hz),
8.05 (d, 3H, J = 7.32 Hz), 7.94 (d, 2H, J = 8.2 Hz), 7.90 (m, 1H), 7.78 (m, 4H), 7.56 (d, 1H, J
= 8.8 Hz), 7.52 (m, 2H), 7.37 (m, 2H), 7.30 (m, 1H), 7.00 (s, 2H), 5.85 (t, 1H, J = 6.5 Hz),
5.25 (d, 1H, J = 7.0 Hz), 4.69 (d, 2H, J = 5.1 Hz), 4.44 (t, 2H, J = 4.8 Hz), 3.97 (d, 1H, J = 9.7
Hz), 3.12-4.20 (m, 50H), 2.95 (m, 2H), 2.03 (t, 2H, J = 7.3 Hz), 1.82 (m, 3H), 1.73 (br s, 1H),
1.46 (quin, 4H, J = 7.3 Hz), 1.11-1.37 (m, 12H). 13C NMR (125 MHz, CDCl3): δ 173.9,
172.1, 171.1, 169.9, 163.1, 157.9, 155.7, 154.5, 149.4, 148.6, 142.1, 139.8, 137.5, 134.9,
134.5, 128.2, 127.6, 126.6, 123.5, 122.7, 120.3, 116.6, 99.9, 75.9, 75.5, 72.8, 71.2, 69.8, 69.7,
69.6, 69.2, 68.9, 49.4, 38.5, 37.0, 35.1, 32.2, 30.8, 27.8, 26.7, 25.8, 24.8, 23.1. Since the
13
compound 1 was obtained as a mixture of two diastereoisomers, some C NMR signals are
split. HMRS (ESI-): for C75H103N14O24, calcd 1583.72697, found 1583.7234.
Cell Culture. Human lung carcinoma A549 cells were grown in RPMI 1640-GlutaMax
(Invitrogen) supplemented by 10% fetal bovine serum (Lonza), and 100u/ml
Penicillin/Streptomycin (Life Technologies) in a humidified incubator at 37°C and 5% CO2.
A549 cells were purchased from European Collection of Cell Cultures (ECACC). Cells were
used at early passages (<10) and cultured as recommended by the manufacturer.
Cell Viability Assay. The Cell Proliferation Kit II (XTT, Roche) was used to assess cell
viability. This assay is based on the cleavage of XTT by metabolic active cells resulting in the
production of an orange formazan dye quantified by spectrophotometry. Assays were carried
out essentially as described by the manufacturer. Briefly, 2.103 tumor cells/well were plated
in a 96-well plate. Cells were cultured for 24h before adding the compound at the indicated
concentration ± β-glucuronidase in the culture media. After 2 days of treatment, 25µl of the
XTT labeling mixture were added per well. Cells were further incubated for additional 4h at

ACCEPTED MANUSCRIPT
37°C before determination of the absorbance at 490nm on a 96-well microplate reader. IC50
values were determined using Graphpad software.

ACCEPTED MANUSCRIPT
Figure 1 The principle of tumour targeting with the β-glucuronidase-responsive albumin-
binding prodrugs: step 1: in the blood, the prodrug binds covalently to plasmatic albumin
through Michael addition; step 2: in the tumour microenvironment, hydrolysis of the
glycosidic bond triggered by extracellular β-glucuronidase initiates the self-decomposition of
the linker leading to the release of the drug.

ACCEPTED MANUSCRIPT
Figure 2 Structures of the β-glucuronidase-responsive albumin-binding prodrug 1, the protein
kinase inhibitor 2 and (R)-Roscovitine.

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis of PKI 2. (a) POCl3, NN-dimethylaniline 100°C, 4 h, 97%; (b) 2-(4-
(ammoniomethyl)phenyl)pyridin-1-ium bis-trifluoroacetate salt, Et3N, nBuOH, 80°C, 3 h,
98%; (c) 3-chloroperbenzoic acid, CH2Cl2, 0°C, 90%; (d) trans-1,4-diaminocyclohexane,
140°C, 4 h, 88%.

ACCEPTED MANUSCRIPT
Table 1. Inhibition of kinase activity by
2
and (R)-Roscovitine (IC50 values are
expressed in µM).
Kinase
Roscovitine
0.31
0.072
0.2
2
CDK1/ cyclin B
CDK2 / cyclin A
CDK5 / p25
CDK9 / cyclin T
CK1
0.07
0.068
0.05
0.049
0.071
0.1
0.22
0.73
1.2
CLK1
CLK2
CLK3
0.71
>10
0.13
2.9
CLK4
0.62
1.2
1.2
1.7
>10
0.089
0.12
0.11
0.12
0.38
0.32
DYRK1A
DYRK1B
DYRK2
DYRK3
GSK3
>10

ACCEPTED MANUSCRIPT
Table 2. IC50 (µM) of (R)-Roscovitine and
2
on different cancer cell lines (SHSY-5Y:
human neuroblastoma; IMR32: human neuroblastoma; Sc1: human B-lymphocyte,
(Epstein-Barr virus transformed); Sc2: human embryonic stem cells; SCOV-3 and
OVCAR-3: human ovarian adenocarcinoma)
Cell line
SHSY-5Y
IMR32
Sc1
Sc2
SCOV-3
OVCAR-3
Roscovitine
2
21
17
32
21
29
37
0.035
0.005
0.015
0.013
0.071
0.024

ACCEPTED MANUSCRIPT
O
O
O
O2N
H
N
NO2
MeOOC
N
O
O
a
N
N
O
AcO
AcO
O
O
7
N
H
N
N
OAc
H
N
MeOOC
AcO
8
NO2
O
b
AcO
H
N
OAc
N
N
O
O
N
O
N
H
N
N
H
N
O
N
N
N
NH2
10
MeOOC
NO2
9
O
AcO
AcO
OAc
c
O
O
O
O
H
N
N
N
N
O
N
N
11
O
O
O
N
N
N
H
N
N
O
H
N
N
NH2
10
HOOC
HO
HO
NO2
O
O
10
d
OH
1
Scheme 2. Synthesis of the prodrug
1
. (a) dimethylformamide, HOBt, pyridine, H
2, rt,
19h, 72%; (b) -(2-aminoethyl)-
O
O’-(2-azidoethyl) nonaethylene glycol,
Cu(MeCN)4PF6, CH2Cl2, rt, 1h; (c) LiOH, H2O/MeOH, 0°C, 4h; (d) 11, DMSO, rt,
overnight, 40% over 3 steps.

ACCEPTED MANUSCRIPT
Figure 3 Kinetics of PKI 2 release from prodrug 1 in the presence of β-glucuronidase (133
U.mL-1) at 37°C (0.02 M phosphate buffer, pH 7). The quantity of 2 released along the time
was determined by HPLC-HRMS.

ACCEPTED MANUSCRIPT
Table 3. Antiproliferative activities of PKI 2 and prodrug 1 (with or without β-glucuronidase)
against human lung carcinoma A549 cell line after 3 days of treatment.
Compound
1
IC50 (nM)
>200
10.3±1.0
9.4±1.2
2
1 + β-glucuronidase

ACCEPTED MANUSCRIPT
Appendix A. Supplementary data
Supplementary data related to this article can be found at
Acknowledgment
The authors thank CNRS, La Ligue Nationale contre le Cancer (for the postdoctoral
fellowship of GC), les Comités Vienne, Charente-Maritime and Deux-Sèvres de La Ligue
contre le Cancer, Sport et Collection, for financial support of this study.
References
[1] H.-Y. Zhao, H. Wei and X. Wang, The Reciprocal Interaction of Small Molecule Protein
Kinase Inhibitors and ATP-Binding Cassette Transporters in Targeted Cancer Therapy, J.
Cancer Res. Updat., 2 (2013) 68-86.
[2] O. Chahrour, D. Cairns and Z. Omran, Small Molecule Kinase Inhibitors as Anti-Cancer
Therapeutics, Mini Rev. Med. Chem. 12 (2012) 399-411.
[3] P. Wu, T. E. Nielsen and M. H. Clausen, Small-molecule kinase inhibitors: an analysis of
FDA-approved drugs, Drug Discov. Today 21 (2016) 5-10.
[4] S. K. Grant, Therapeutic Protein Kinase Inhibitors, Cell. Mol. Life Sci. 66 (2009) 1163-
1177.
[5] D. R. Shah, R. R. Shah and J. Morganroth, Tyrosine Kinase Inhibitors: Their On-Target
Toxicities as Potential Indicators of Efficacy, Drug Saf. 36 (2013) 413-426.
[6] S. Gross, R. Rahal, N. Stransky, C. Lengauer and K. P. Hoeflich, Targeting cancer with
kinase inhibitors, J. Clin. Invest. 125 (2015) 1780-1789.

ACCEPTED MANUSCRIPT
[7] R. R. Shah, Tyrosine Kinase Inhibitor-Induced Interstitial Lung Disease: Clinical
Features, Diagnostic Challenges, and Therapeutic Dilemmas, Drug Saf. 39 (2016) 1073-1091.
[8] G. Manning, The protein kinase complement of the human genome, Science 298 (2002)
1912.
[9] A. Beck, L. Goetsch, C. Dumontet, N. Corvaïa, Strategies and challenges for the next
generation of antibody drug conjugates, Nat. Rev. Drug Discov. 16 (2017) 315-337.
[10] T. Legigan, J. Clarhaut, B. Renoux, I. Tranoy-Opalinski, J.-M. Berjeaud, F. Guilhot and
S. Papot, Synthesis and Antitumor Efficacy of a β-Glucuronidase-Responsive Albumin-
Binding Prodrug of Doxorubicin, J. Med. Chem. 55 (2012) 4516-4520.
[11] B. Renoux, F. Raes, T. Legigan, E. Péraudeau, B. Eddhif, P. Poinot, I. Tranoy-Opalinski,
J. Alsarraf, O. Koniev, S. Kolodych, S. Lerondel, A. Le Pape, J. Clarhaut and S. Papot,
Targeting the Tumour Microenvironment with Enzyme-Responsive Drug Delivery System for
Efficient Therapy of Breast and Pancreatic Cancers, Chem. Sci. 8 (2017) 3427-3433.
[12] M. de Graaf, E. Boven, H. W. Scheeren, H. J. Haisma and H. M. Pinedo, Beta-
glucuronidase-mediated drug release, Curr. Pharm. Des. 8 (2002) 1391-1403.
[13] I. Tranoy-Opalinski, T. Legigan, R. Barat, J. Clarhaut, M. Thomas, B. Renoux and S.
Papot, β-Glucuronidase-Responsive Prodrugs for Selective Cancer Chemotherapy: an Update,
Eur. J. Med. Chem. 74 (2014) 302-313.
[14] I. Tranoy-Opalinski, A. Fernandes, M. Thomas, J.-P. Gesson and S. Papot, Design of
Self-Immolative Linkers for Tumour-Activated Prodrug Therapy, Anti-Cancer Agents Med.
Chem. 8 (2008) 618-637.

ACCEPTED MANUSCRIPT
[15] A. Alouane, R. Labruère, T. Le Saux, F. Schmidt and L. Jullien, Self-Immolative
Spacers: Kinetic Aspects, Structure-Property Relationships, and Applications, Angew. Chem.
Int. Ed. 54 (2015) 7492-7509.
[16] B. Elsadek and F. Kratz, Impact of albumin on drug delivery - New applications on the
horizon, J. Control. Release 157 (2012) 4-28.
[17] W. H. Fishman and A. J. Anlyan, Comparison of the beta-glucuronidase activity of
normal, tumor, and lymph node tissues of surgical patients, Science, 106 (1947) 66-67.
[18] N. Albin, L. Massaad, C. Toussaint, M. C. Mathieu, J. Morizet, O. Parise, A. Gouyette
and G. G. Chabot, Main drug-metabolizing enzyme-systems in human breast-tumors and
peritumoral tissues, Cancer Res. 53 (1993) 3541-3546.
[19] K. Bosslet, J. Czech and D. Hoffmann, A novel one-step tumor-selective prodrug
activation system Tumor Targeting, 1 (1995) 45-50.
[20] K. Bosslet, R. Straub, M. Blumrich, J. Czech, M. Gerken, B. Sperker, H. K.
Kroemer, J.-P. Gesson, M. Koch and C. Monneret, Elucidation of the mechanism
enabling tumor selective prodrug monotherapy Cancer Res., 58 (1998) 1195-1201.
[21] L. Meijer, H. Galons, B. Joseph, F. Popowycz, N. Oumata WO2010103486A1.
[22] J. Cicenas, K. Kalyan, A. Sorokinas, E. Stankunas, J. Levy, I. Meskinyte, V.
Stankevicius, A. Kaupinis and M. Valius, Roscovitine in cancer and other diseases, Ann.
Transl. Med. 3 (2015) 135.
[23] T. Legigan, J. Clarhaut, B. Renoux, I. Tranoy-Opalinski, A. Monvoisin, C. Jayle,
J. Alsarraf, M. Thomas and S. Papot, Synthesis and biological evaluations of a

ACCEPTED MANUSCRIPT
monomethylauristatin E glucuronide prodrug for selective cancer chemotherapy, Eur.
J. Med. Chem. 67 (2013) 75-80.

ACCEPTED MANUSCRIPT
Synthesis of a novel nanomolar protein kinase inhibitor.
Synthesis of the first β-glucuronidase-responsive albumin-binding prodrug for selective
kinase inhibitor-based cancer chemotherapy.
Covalent binding to human serum albumin through Michael addition.
Efficient release of the drug upon enzymatic activation.