Development of bioactive gemcitabine-D-Lys6-GnRH prodrugs with linker-controllable drug release rate and enhanced biopharmaceutical profile

Article abstract of DOI:10.1016/j.ejmech.2019.01.041

Peptide-drug conjugates have emerged as a potent approach to enhance the targeting and pharmacokinetic profiles of drugs. However, the impact of the linker unit has not been explored/exploited in depth. Gemcitabine (dFdC) is an anticancer agent used against a variety of solid tumours. Despite its potency, gemcitabine suffers mostly due to its unspecific toxicity, lack of targeting and rapid metabolic inactivation. To minimize these limitations and enable its targeting to tumours overexpressing the GnRH receptor, we examined the peptide-drug conjugation approach. Our design hypothesis was driven by the impact that the linker unit could have on the peptide-drug conjugate efficacy. Along these lines, in order to exploit the potential to manipulate the potency of gemcitabine through altering the linker unit we constructed three different novel peptide-drug conjugates assembled of gemcitabine, the tumour-homing peptide D-Lys⁶-GnRH and modified linker building blocks. Specifically, the linker was sculpted to either allow slow drug release (utilizing carbamate bond) or rapid disassociation (using amide and ester bonds). Notably, the new analogues possessed up to 95.5-fold enhanced binding affinity for the GnRH receptor (GnRH-R) compared to the natural peptide ligand D-Lys⁶-GnRH. Additionally, their in vitro cytotoxicity was evaluated in four different cancer cell lines. Their cellular uptake, release of gemcitabine and inactivation of gemcitabine to its inactive metabolite (dFdU) was explored in a representative cell line. In vitro stability and the consequent drug release were evaluated in cell culture medium and human plasma. In vivo pharmacokinetic studies were performed in mice, summarizing the relative stability of the three conjugates and the released levels of gemcitabine in comparison with dFdU. These studies suggest that the fine tuning of the linkage within a peptide-drug conjugate affects the drug release rate and its overall pharmaceutical profile. This could eventually emerge as an intriguing medicinal chemistry approach to optimize bio-profiles of prodrugs.

Full text of DOI:10.1016/j.ejmech.2019.01.041

Accepted Manuscript
6
Development of bioactive gemcitabine-D-Lys -GnRH prodrugs with linker-controllable
drug release rate and biopharmaceutical profile
Nisar Sayyad, Eirinaios I. Vrettos, Theodoros Karampelas, Christos M. Chatzigiannis,
Katerina Spyridaki, George Liapakis, Constantin Tamvakopoulos, Andreas G. Tzakos
PII:
S0223-5234(19)30051-0
DOI:
https://doi.org/10.1016/j.ejmech.2019.01.041
EJMECH 11048
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 20 November 2018
Revised Date: 16 January 2019
Accepted Date: 17 January 2019
Please cite this article as: N. Sayyad, E.I. Vrettos, T. Karampelas, C.M. Chatzigiannis, K. Spyridaki,
6
G. Liapakis, C. Tamvakopoulos, A.G. Tzakos, Development of bioactive gemcitabine-D-Lys -GnRH
prodrugs with linker-controllable drug release rate and biopharmaceutical profile, European Journal of
Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.01.041.
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
Development of bioactive gemcitabine-D-Lys6-GnRH prodrugs with linker-
controllable drug release rate and biopharmaceutical profile
Nisar Sayyad1, Eirinaios I. Vrettos1, Theodoros Karampelas2, Christos M.
Chatzigiannis1 Katerina Spyridaki3, George Liapakis3, Constantin Tamvakopoulos2,
,
Andreas G. Tzakos1*
1Department of Chemistry, Laboratory of Chemical Biology, University of Ioannina,
Ioannina, GR-45110, Greece;
2Center of Clinical Research, Experimental Surgery and Translational Research,
Biomedical Research Foundation Academy of Athens (BRFAA), Soranou Ephessiou
Street 4, Athens GR-11527, Greece;
3Department of Pharmacology, Faculty of Medicine, University of Crete, Heraklion,
Crete, Greece
Highlights
° 3 different peptide-drug conjugates (PDCs) targeting the GnRH-R were
synthesized
° The impact of the linker was explored in terms of PDCs stability and bioactivity
° Binding affinity of PDCs to GnRH-R was enhanced with respect to GnRH-II
° PDCs showed amplified plasma stability and pharmaceutical profile
Graphical Abstract

ACCEPTED MANUSCRIPT
Abbreviations
GnRH, Gonadotropin Releasing Hormone; GnRH-R, Gonadotropin Releasing
Hormone-Receptor; dFdC, 2'-2'difluorodeoxycytidine; dFdU, 2, 2’-
difluorodeoxyuridine; n-BuOH (n-butanol); MeCN, acetonitrile; HATU, 1-
[Bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid
hexafluorophosphate; DIPEA, N,N-Diisopropylethylamine; PDC, Peptide-Drug
Conjugate; EGFR, Epidermal Growth Factor Receptor;
Abstract
Peptide-drug conjugates have emerged as a potent approach to enhance the
targeting and pharmacokinetic profiles of drugs. However, the impact of the linker
unit has not been explored/exploited in depth. Gemcitabine (dFdC) is an anticancer
agent used against a variety of solid tumours. Despite its potency, gemcitabine suffers
mostly due to its unspecific toxicity, lack of targeting and rapid metabolic inactivation.
To minimize these limitations and enable its targeting to tumours overexpressing the
GnRH receptor, we examined the peptide-drug conjugation approach. Our design
hypothesis was driven by the impact that the linker unit could have on the peptide-
drug conjugate efficacy. Along these lines, in order to exploit the potential to
manipulate the potency of gemcitabine through altering the linker unit we constructed

ACCEPTED MANUSCRIPT
three different novel peptide-drug conjugates assembled of gemcitabine, the tumour-
homing peptide D-Lys6-GnRH and modified linker building blocks. Specifically, the
linker was sculpted to either allow slow drug release (utilizing carbamate bond) or
rapid disassociation (using amide and ester bonds). Notably, the new analogues
possessed up to 95.5-fold enhanced binding affinity for the GnRH receptor (GnRH-R)
compared to the natural peptide ligand D-Lys6-GnRH. Additionally, their in vitro
cytotoxicity was evaluated in four different cancer cell lines. Their cellular uptake,
release of gemcitabine and inactivation of gemcitabine to its inactive metabolite
(dFdU) was explored in a representative cell line. In vitro stability and the consequent
drug release were evaluated in cell culture medium and human plasma. In vivo
pharmacokinetic studies were performed in mice, summarizing the relative stability of
the three conjugates and the released levels of gemcitabine in comparison with dFdU.
These studies suggest that the fine tuning of the linkage within a peptide-drug
conjugate affects the drug release rate and its overall pharmaceutical profile. This
could eventually emerge as an intriguing medicinal chemistry approach to optimize
bio-profiles of prodrugs.
Keywords: Gemcitabine; Peptide-drug conjugate; Pharmacokinetics; Stability;
Binding affinity;
1. Introduction
The menace of cancer is affecting humanity with rapidly increasing rate and is
tending to become the number one cause of mortality worldwide. Therefore, cancer
has turned into a top priority for the scientific community which is continuously
attempting to unscramble its mechanism of action and discover new effective
therapies or improve the existing ones. In this frame, an immense variety of drug-
designing approaches has been developed, with the aim to improve conventional
chemotherapy through alleviation of severe side-effects caused by collateral toxicity
in healthy tissues [1]. This intriguing goal can be achieved through selective delivery
of cytotoxic agents to tumour cells by conjugating chemotherapeutic agents with
peptides [2-5] which selectively bind to cell surface proteins/receptors uniquely
expressed or over-expressed in cancer cells (e.g. GnRH-R, EGFR, integrin receptors)

ACCEPTED MANUSCRIPT
[6, 7]. Two peptide-drug conjugates (PDCs) have reached up to Phase-III clinical
trials typify this class of chemotherapeutics [8, 9].
In peptide-drug conjugates, bioconjugation is achieved through linkers which
may consist of a wide range of functional groups and therefore result in various bond
formations like amide, ester, oxime, carbamate etc. These linker units could have a
tremendous impact in the bioactivity profile of PDCs since they could instruct on the
time that a drug will be released [10-12]. Furthermore, they have to be carefully
designed for tethering the transporting vehicle to the cytotoxic warhead so as not to
perturb the binding affinity of the tumour-homing peptide to its receptor [4].
Interestingly, the carbamate unit has been included, as a structural moiety, in
many approved drugs/prodrugs including XP13512 [13] (a novel prodrug of
gabapentin), capecitabine and 5-fluorouracil (5-FU) [14]. Carbamate moiety is often
rationally selected to form drug-target interactions with certain receptors [15, 16] as
also are often manipulated for use in the design of drugs and prodrugs to offer
adequate systemic hydrolytic stability and enhanced overall bioavailability [17-20]. In
specific, aliphatic amine-derived carbamates are known to be sufficiently stable and
undergo hydrolytic degradation at a suitable rate for releasing the active drug [16, 21].
Intriguingly, this unit is frequently referred as a traceless linker, due to the fact that
after its hydrolysis, the only by-product produced is gas CO2 along with the starting
amine and alcohol [22, 23].
Nucleoside analogues constitute a major class of chemotherapeutic agents
administered to patients with progressed cancer. They are a part of a larger family of
anticancer drugs known as antimetabolites, including pyrimidine and purine
nucleoside derivatives with cytotoxic activity against various solid tumours [24].
Gemcitabine (2’ – deoxy – 2’, 2’ - difluorocytidine), a potent antimetabolite drug, is
known to be active against several solid tumour types [25] and has gained FDA-
approval for treating non-small cell lung, ovarian, bladder, pancreatic and breast
cancer. Gemcitabine enters the cells through nucleoside transporters and immediately
undergoes intracellular sequential conversion into the monophosphate (dFdCMP),
diphosphate (dFdCDP) and finally triphosphate (dFdCTP) form [26]. The main
mechanism of action of gemcitabine relies on the formation of dFdCTP, which
replaces deoxycytidine during DNA replication and leads to cell death [27] and to a
lesser extent on the formation of dFdCDP which interferes with thymidylate synthase
that normally catalyzes the conversion of uridine monophosphate into thymidine

ACCEPTED MANUSCRIPT
monophosphate [28]. However, gemcitabine is prone to certain limitations: i) it
suffers from rapid metabolic inactivation to its metabolite dFdU through deamination
either by cytidine deaminase (CDA) or by deoxycytidylate deaminase (DCTD),
resulting in the monophosphorylated form of dFdU (dFdUMP) [28]; ii) cancer cells
acquire drug resistance after extensive usage of gemcitabine, mostly by diminishing
the expressed nucleoside transporters that facilitate the internalization of the drug [29,
30]; iii) it is indiscriminately toxic, affecting both cancer and healthy tissues and
consequently result in severe side effects on patient’s health [31].
The limitations of gemcitabine motivated us to design novel analogues with
enhanced pharmacological properties and stability. Veltkamp et. al. has proved that
prolonged administration of gemcitabine in patients with pancreatic cancer resulted in
elevated intracellular levels of dFdCTP and higher overall toxicity in comparison with
the standard infusion protocol [32]. Intrigued by these results, along with similar
carbamate-based prodrugs of gemcitabine [33-35], we rationally designed and
synthesized three novel peptide drug conjugates of gemcitabine (Scheme 1),
consisting of either one traceless carbamate bond (small linker; compounds
designated 2G1 and 2G2) or two amide bonds and one ester bond (long linker;
compound designated GSHG) tethering the drug and the tumour homing D-Lys6-
Gonadotropin Releasing Hormone (D-Lys6-GnRH). The short linker (carbamate
bond) has been included in the design since it can offer stability to a bioconjugate by
decreasing the drug release rate (slow release), whereas the long linker consisted of a
mixture of amide and ester bonds, could offer a rapid disassociation and a consequent
faster drug release as these bonds are prone to intracellular esterases and amidases.

ACCEPTED MANUSCRIPT
Scheme 1. Architecture of the three peptide-drug conjugates (D-Lys6-GnRH-
gemcitabine) where the different building blocks are highlighted (gemcitabine in blue,
linkers in red and D-Lys6-GnRH in black).
The 2G1, 2G2 and GSHG peptide-drug conjugates have been designed with
different linker building blocks in order to afford different hydrolytic stabilities and
possibly might illustrate different potency. Specifically, 2G1 and 2G2 are expected to
be more stable, enabling slow release of the cytotoxic agent, while compound GSHG
is expected to afford rapid release of the cytotoxic agent.
The new analogues were evaluated regarding their binding affinity to the
GnRH-R compared to D-Lys6-GnRH. The stability and the intracellular
concentrations of the conjugates, gemcitabine and dFdU were also explored in vitro in
cell culture and plasma, as also in vivo in mice plasma.
2. Results and discussion
2.1 Design of D-Lys6-GnRH-gemcitabine conjugates
In our former studies [3], we reported the conjugation of gemcitabine and D-
Lys6-GnRH peptide via hemiglutarate and hemisuccinate linkers. These conjugates,
although are potent in vitro and in clinically relevant prostate cancer models, resulted
in low hydrolytic stability and consequent high levels of both dFdC and dFdU. The
results highlighted the dominant role of the linker towards the bioactive profile of
each compound. Based on these findings, we aimed to explore and exploit linker’s
impact towards the biopharmaceutical profile of the final PDCs. We thus synthesized
three peptide-drug conjugates bearing either a long linker (GSHG) or a very short
traceless carbamate linker (2G1 and 2G2). Among a wide variety of classic bonds, we
adopted carbamate due to its facile implementation and its hydrolytic stability offering
a sporadic drug release.
Gemcitabine contains two hydroxyl functionalities, a primary 5’-OH and a
secondary 3’-OH, which can be easily and rapidly conjugated with a tumour-targeting
peptide (Scheme 2). Specifically, compound 4 was synthesized according to our
previously reported synthetic scheme [3], while compound 3 was synthesized by
reacting the ε-amino group of D-Lys6-GnRH and the free carbonyl group of the
previously Boc- protected ε-aminocaproic acid [36] with the usage of HATU as a
coupling reagent. We would like to highlight here that, as we discovered recently, 1

ACCEPTED MANUSCRIPT
eq. of HATU has to be used in order to avoid formation of unwanted side-products
[37]. The ε-amine of compound 3 reacted with the free carboxylic acid of compound 4,
utilizing again 1 eq. of HATU as a coupling reagent, resulting to the relevant Boc-
protected compound. Another Boc- deprotection step under acidic conditions was
necessary to afford the final compound GSHG.
Compounds 2G1 and 2G2, bearing carbamate bonds, were synthesized by
reacting the ε-amino group of D-Lys6-GnRH with the activated intermediates 2a and
2b respectively, followed by Boc- deprotection under acidic conditions (Scheme 3).
Specifically, gemcitabine was selectively Boc- protected to afford compounds 1a and
1b [38], which were sequentially activated with the usage of bis(4-
nitrophenyl)carbonate to afford the intermediates 2a and 2b, respectively.
Intermediates 2a and 2b were treated with D-Lys6-GnRH, with the usage of HATU as
a coupling reagent, resulting in the final compounds 2G1 and 2G2.
Scheme 2. Synthesis of GSHG. Reagents and conditions: a) Boc2O, NaOH,
Dioxane:Water, rt, 6h; b) D-Lys6-GnRH, HATU, DIPEA, DMF, rt, 12h; c)
TFA/H2O/TIS (9.5/0.25/0.25, v/v), rt, 4h; d) 3, HATU, DIPEA, DMF, rt, 12h; e)
TFA/ H2O/ TIS (9.5/0.25/0.25, v/v), rt, 4h.

ACCEPTED MANUSCRIPT
Scheme 3. Syntheses of the carbamate-based 2G1 and 2G2. Reagents and conditions:
a) Bis(4-nitrophenyl)carbonate, DIPEA, MeCN, rt, 6h; b) D-Lys6-GnRH, DIPEA,
DMF, rt, 12h; c) TFA/ H2O/ TIS (9.5/0.25/0.25, v/v), rt, 4h.
2.2 Binding affinity towards human GnRH-R
The three synthesized conjugates (2G1, 2G2 and GSHG) were evaluated with respect
to their binding affinity to the GnRH-R as shown in Table 1.
Table 1. Binding assay values IC50 (nM) of 2G1, 2G2 and GSHG compared to native
peptide ligand D-Lys6-GnRH to GnRH-R.
Compounds
2G1
Radioligand Binding assay IC50 (nM)
0.69 ± 0.02 nM
0.11 ± 0.00 nM
2G2
2.40 ± 0.04 nM
GSHG
D-Lys6-GnRH
10.5 ± 0.2 nM
The presented data show that 2G2, 2G1 and GSHG bind to GnRH-R with
95.5-, 15.2-, and 4.4-fold higher affinity, respectively, than that of the native peptide
D-Lys6-GnRH (10.5 ± 0.2 nM, according to our former study [3]). When compared to
our previously published compound GSG (IC50 = 7.1 ± 0.2 nM, according to our
former study [3]), these three new conjugates presented enhanced binding affinity for
the GnRH-R (65-, 10.3-and 3-fold higher binding affinity for 2G2, 2G1 and GSHG,
respectively). The latter can be attributed to the fact that the carbamate group is a

ACCEPTED MANUSCRIPT
moiety of medium polarity, capable of forming hydrogen bonds as donor and acceptor
through the peptide amide backbone [39] and thus adjusting inter- and intramolecular
interactions with the targeted receptor(s). Also, the carbamate moiety induces a degree
of conformational restriction due to the delocalization of nonbonded electrons on
nitrogen into the carboxyl moiety [16] that could further favor the conformational
entropy upon binding.
2.3 GnRH-gemcitabine conjugates exhibit antiproliferative potential in vitro. The
antiproliferative effect of the three GnRH-gemcitabine analogues was evaluated in
two androgen prostate cancer cell lines (DU145 and PC3) and in two breast cancer
cell lines (MDA-MB-21 and MCF-7) using gemcitabine and D-Lys6-GnRH as their
corresponding controls as shown in Table 2. GSHG possesses the highest cytotoxic
effect among the three conjugates, which is comparable with that of gemcitabine in
the examined cell lines and especially regarding MCF-7 cells. Results are summarized
in Table 2 below:
Table 2. Antiproliferative effect (IC50 values in nM) of GnRH-gemcitabine
conjugates compared with gemcitabine and D-Lys6-GnRH in various cancer cell lines
(DU145, PC3, MDA-MB-231, MCF-7).
Cell lines
IC50 (nM)
2G2
GSHG
2G1
Gemcitabine
D-Lys6-
GnRH
>40,000
(n=3)
684 ± 203
(n=3)
>40,000
(n=3)
9,761 ±
213 ± 40
(n=3)
DU145
6278 (n=3)
937 ± 228
(n=3)
>40,000
(n=3)
>40,000
(n=3)
416 ± 13
(n=3)
>40,000
(n=3)
PC3
2,387 ±
>40,000
(n=3)
>40,000
(n=3)
1,567 ± 757
(n=5)
>40,000
(n=3)
MDA-MB-231
MCF-7
1,220 (n=3)
55.5 ± 7.2
(n=3)
621.3 ±
449.1 ±
38.0 ± 14.1
(n=3)
>40,000
(n=3)
113 (n=3)
87.0 (n=3)

ACCEPTED MANUSCRIPT
From the results presented above, it can be concluded that conjugates 2G1 and
2G2 bear less cytotoxicity in general when compared with the other compounds. This
could be correlated with the high proteolytic stability of the conjugates, which results
in an insufficient release of gemcitabine, associated with less toxicity. In order to
verify this hypothesis, the three conjugates were subjected to stability studies in cell
culture medium, as well as human plasma.
2.4 Stability studies and release of free gemcitabine in cell culture medium.
In Figure 1A, the degradation of the three conjugates over time is illustrated.
Analogues bearing the carbamate bond (2G1 and 2G2) appeared to possess similar
stability in the cell culture (stable even after 72 h), while analogue GSHG degraded
fully in less than 24 h. In Figure 1B, the amounts of gemcitabine released from the
three conjugates over time are presented.
Figure 1. Stability of the three conjugates (2G1, 2G2 and GSHG) over time in cell
culture medium (A) and release of gemcitabine (B).

ACCEPTED MANUSCRIPT
The data presented in Figure 1B, fully correlate with the stabilities of the
corresponding conjugates in cell culture medium, presenting low stability of GSHG
which results in higher drug release. In contrast, the high stability of 2G1 and 2G2
results in low amounts of free gemcitabine.
These findings provide a rationale on the reduced cytotoxicity of 2G2 and 2G1 and the
enhanced cytotoxicity of GSHG.
2.5 Stability studies of the GnRH-gemcitabine conjugates in human plasma.
To further understand the cell culture stability results, we performed stability
experiments in human plasma, a more clinically relevant system. The corresponding
analogues possess similar stabilities with those observed in cell culture medium.
Figure 2A clearly demonstrates the enhanced stability of conjugates 2G1 and 2G2,
while in contrast conjugate GSHG degrades within 4 h. Likewise before, in Figure
2B are presented the levels of gemcitabine released by the three conjugates, which
correlate with the stability pattern of Figure 2A.

ACCEPTED MANUSCRIPT
Figure 2. Stability of the three conjugates (2G1, 2G2 and GSHG) over time in human
plasma (A) and release of gemcitabine (B).
These results are in accordance with the cell culture stability experiments
presented in Figure 1. Similarly, GSHG that has the lowest stability in human plasma
releases the highest amount of gemcitabine, while the other two conjugates (2G1,
2G2) are more stable, thus resulting in reduced amounts of free gemcitabine in plasma
and lower cytotoxicity.
2.6 Pharmacokinetics of 2G1, 2G2 and GSHG in mice.
In order to further probe the pharmacokinetic profile of the three synthesized PDCs in
vivo we performed relevant studies in mice. Administration of the three conjugates in
mice led to significantly high levels of the conjugate in blood regarding the two short-
linked, via carbamate bond, analogues 2G1 and 2G2 (blood concentrations of at least 4

ACCEPTED MANUSCRIPT
µg/mL in 15, 30 min) as shown in Figure 3A. In contrast, long-linked, via
aminocaproic-succinic linkers, analogue GSHG led to low levels of the corresponding
conjugate (Figure 3A). The high stability of 2G1 and 2G2 resulted in minimum levels
of free gemcitabine in blood, in contrast with GSHG (Figure 3B).
In marked contrast to 2G1 and 2G2 blood levels in mice (4 µg/mL and 5
µg/mL, respectively in 30 min), the levels of gemcitabine in blood (approximately 3
ng/ml) were significantly low after administration of these conjugates. Administration
of the same dose (10 mg/kg) of the conjugate GSHG has the opposite results. Its
concentration in mice blood is considerably low compared to the other two conjugates,
while the levels of released gemcitabine from GSHG were significantly higher,
averaging approximately 300 ng/ml in 30 min after the administration.
Regarding the formation of the inactive metabolite of gemcitabine (dFdU), no
detectable levels were observed in 2G1 or 2G2 samples, while for GSHG it averaged
approximately 500 ng/mL in 1 h after the administration (Figure 3C).

ACCEPTED MANUSCRIPT
Figure 3. Pharmacokinetic evaluation of of 2G1, 2G2 and GSHG in mice. Blood
concentrations of the conjugates (A), gemcitabine (B) and dFdU (C) after 10 mg/kg IP
administration of conjugate in mice.

ACCEPTED MANUSCRIPT
2.7 Determination of intracellular concentrations of 2G1, 2G2 and GSHG in
cancer cell line DU145.
The intracellular levels measured over time in cancer cells DU145 (Figure 4) for the
three conjugates, indicated that 2G2 achieved the highest intracellular concentration
with a peak at approx. 15 ng/106 cells in 4 h. On the other hand, GSHG illustrated the
smallest intracellular quantity, with a peak at 8 ng/106 cells. The last could be related
to the fact that GSHG adopts the lowest binding affinity of these analogues for the
GnRH-R, while 2G2 possesses the highest one (as shown in Table 3). This trend
could potentially suggest that the highest affinity of the bioconjugate to the peptide
receptor could be associated with enhanced intracellular entrance, as has been also
observed in former studies on GnRH-R [40, 41].
Figure 4. Intracellular levels of 2G1, 2G2 and GSHG (A) and gemcitabine (B) over
time after incubation of the three conjugates with DU145 cells.

ACCEPTED MANUSCRIPT
Table 3. AUCs of intracellular levels of conjugates 2G1, 2G2 and GSHG after cell
uptake with DU145 cells
Compounds
2G1
AUC(1-8 hr) (ng/106 cells × hr)
62.8
88.7
38.6
2G2
GSHG
Based on the AUC values presented in Table 3, GSHG possesses inferior cell
uptake compared to conjugates 2G1 and 2G2.
Incubation of 2G1 and 2G2 in cancer cell lines DU145 did not result in
intracellular detection of gemcitabine, suggesting again that this linkage is very stable
and gemcitabine was not released even after 8 hours. On the other hand, GSHG
resulted in increasing levels of gemcitabine over time (Figure 4B), resulting in a
relatively rapid release of the drug.
Importantly, no detectable levels of the metabolite dFdU were observed in any
of the three incubations (data not shown). Although this was expected for the two
short-linked conjugates due to minimum release of gemcitabine, it was surprising for
the GSHG conjugate.
Apparently, GSHG has a more advantageous balance profile regarding cell
uptake and stability, whereas 2G1 and 2G2 although enter cells more readily than
GSHG cannot release rapidly gemcitabine, the free form of which exerts cytotoxic
actions. Thus, it can be envisioned that 2G1 and 2G2 could be potentially utilized for
long term administration of gemcitabine, while GSHG for short term.
3. Conclusions
Gemcitabine is a widely used anticancer nucleoside agent that possesses high toxicity
against both cancer and healthy cells. Peptide-drug conjugates have emerged as a
useful tool that enables the controlled release of the cytotoxic warhead at a specific
site and in time- and/or quantity- controlled manner. These characteristics are based
on the type of the labile linker utilized in each conjugate tethering the drug with the
peptide. Herein, we report the construction, synthesis and biological evaluation of
three novel gemcitabine-GnRH conjugates utilizing different linkers with the aim to
provide targeted delivery of gemcitabine to the cancer tissues in due time or

ACCEPTED MANUSCRIPT
sporadically. We synthesized two bioconjugates through a small carbamate linker, in
the case of 2G1 and 2G2, and one through a long linker which is composed of an ester
and two amide bonds, in the case of GSHG. Utilizing the specific types of linkers, we
provided PDCs with enhanced binding affinity to the GnRH-R and also, we were able
to manipulate the release of native gemcitabine as monitored through our in vivo
pharmacokinetics in mice, uptake in different cancer cell cultures, as also cell culture
and human plasma stability evaluations. In addition, we achieved an enhancement of
the bioavailability of the drug, especially in the cases of 2G1 and 2G2 which proved to
be hydrolytically stable. GSHG on the other hand, showed rapid release of
gemcitabine and consequently quite higher degradation rate, associated with reduced
levels of the inactive metabolite dFdU. In conclusion, the linker tethering the
cytotoxic drug and the carrier is of great significance regarding the overall bioactivity
of the final conjugate, offering unique release rates. Our proposed conjugates proved
to be very promising for selective drug delivery, with enhanced properties and a time-
controllable release of the parent drug.
4. Experimental Section
4.1 Chemicals
Gemcitabine was purchased from Carbosynth (Compton, Berkshire, UK). HATU ((O-
(7-azabenzotriazol-1-yl)- N, N, N’, N’-tetramethyluronium hexafluorophosphate) and
HOBt (1-hydroxybenzotriazole) were purchased from GLB (GLBiochem, China).
Bis(4-nitrophenyl)carbonate,
ε-aminocaproic
acid
and
DIC
(N,N’-
diisopropylcarbodiimide) were purchased from Sigma Aldrich. Boc2O (di-tert-
butyldicarbonate) from Alfa Aesar (A Johnson Matthey Company, Germany).
Triisopropylsilane (TIS), N, N-diisopropylethylamine (DIPEA), trifluoroacetic acid
(TFA) were purchased from Merck (Darmstadt, Germany). N, N Dimethylformamide
(DMF) was distilled over CaH2 and stored under pre-activated molecular sieves 4A.
Deuterated solvents were purchased from Deutero (Deutero GmBH, Kastellaun,
Germany). Ammonium acetate, formic acid and 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyl-tetrazolium bromide (MTT) were purchased from Sigma-Aldrich (Sigma-
Aldrich Chemie GmbH, Munich, Germany). Acetonitrile (LC-MS grade) was
purchased from Fisher Scientific (Fisher Scientific, Loughborough, UK). Water (LC-
MS grade) was purchased from Carlo Erba (Carlo Erba, Milan, Italy). Thin-layer

ACCEPTED MANUSCRIPT
chromatography (TLC) was performed on TLC plates, silica gel coated with
fluorescent indicator F254. Column chromatography was performed using SiO2 and
the appropriate elution solvent mixture as idnicated in each experimental procedure.
The 1H NMR and 13C NMR spectra were recorded on a Bruker AV400, using DMSO-
d6 and 298 K. Chemical shifts for 1H NMR spectra are reported as δ units of parts per
million (ppm) and relative to the signal of DMSO-d6 (δ 2.50 multiplet). Multiplicities
are given as follows: s (singlet); d (doublet); dd (doublet of doublets); t (triplet); bs
(broad singlet); m (multiplet). Coupling constants are reported as J values in Hz.
HPLC was accomplished on a Shimadzu preparative liquid chromatographer, with a
photodiode array detector (SPD-M20A) and an LC-8A pump. Purification by
preparative RP-HPLC was conducted on an Ascentis® C18 column (25 cm × 21.2
mm) at a flow rate of 20 mL/min and 10 µm particle size as stationary phase. For the
MS characterization of the synthesized compounds, the samples were injected directly
to an EVOQ™ Elite ER LC-TQ system (Bruker, Germany).
4.2 Synthesis of D-Lys6-GnRH-gemcitabine conjugates
D-Lys6-GnRH was synthesized manually via the classic Fmoc- solid phase peptide
synthesis on a Rink Amide resin (0.58 mmol/g). RP-HPLC, NMR and MS data for the
syntheses of 2G1, 2G2 and GSHG are presented in the Supporting Information (Fig.
S1 – Fig. S19).
4.2.1. 6-((tert-butoxycarbonyl)amino)hexanoic acid (2)
0
1 M aq. NaOH (0.762 mL, 0.76 mmol) was added dropwise at 0 C to a stirring
solution of 1 (100 mg, 0.76 mmol) in 10 mL Dioxane/H2O (2/1, v/v). After 5 mins,
Boc2O (166.4 mg, 1.52 mmol) was added and the reaction was stirred at rt. Reaction
was monitored via TLC (5 % methanol in CH2Cl2, Rf=0.32). After 3 h, the solvent
was evaporated under reduced pressure. Water (5 mL) was added, pH was adjusted to
pH=1 with 1 M aq. HCl and then extracted thrice with ethyl acetate (10 mL × 3).
Combined organic phases were washed once with brine and dried over sodium
sulphate. After filtration, the solvent was concentrated under reduced pressure to
o
1
afford compound 2 (170 mg, 96.47 %) as a colorless liquid (m.p.=35-40 C). H-
NMR of (2) (400 MHz, DMSO-d6, 25οC): δ = 12.01 (bs, 1H), 6.78 (t, J = 5.06 Hz,
1H), 2.91 (q, J = 6.66, 6.26 Hz, 2H), 2.21 (t, J = 7.33 Hz, 2H), 1.54 - 1.47 (m, 2H),
13
1.43 - 1.35 (m, 2H), 1.40 (s, 9H), 1.30 - 1.22 (m, 2H) ppm; C-NMR of (2) (400

ACCEPTED MANUSCRIPT
MHz, DMSO-d6, 25οC): δ = 174.44 (C), 155.58 (C), 77.31 (C), 40.17 (CH2), 33.62
(CH2), 29.20 (CH2), 28.27 (CH3), 25.84 (CH2), 24.21 (CH2) ppm; Mass: ESI-MS
(m/z) Calcd. for C11H21NO4: 231.15, found: 254.55 [M+Na]+
4.2.2. 6-amino-N-methylhexanamide - GnRH (3)
DIPEA (9.7 ul, 0.055 mmol) was added dropwise under inert atmosphere to a solution
of 2 (2.58 mg, 0.011 mmol) and HATU (4.18 mg, 0.011 mmol) in 3 mL anhydrous
DMF. After 5 mins, a solution of D-Lys6-GnRH (14 mg, 0.011 mmol) in 2 mL DMF
was added dropwise and the resulting mixture was stirred at rt for 12h. The progress
was monitored via TLC in a solvent system of n-/acetic acid/water (6/2/2, v/v), which
showed the complete consumption of both starting materials (2 and D-Lys6-GnRH)
and the formation of a new spot (Rf=0.37). DMF was evaporated under high vacuum
and the residue was washed thrice with MeCN (1 mL × 3) to afford 16 mg of a white
solid which was used for the next reaction (Boc- deprotection) without further
purification. Boc-deprotection was performed with 2 mL of cleavage cocktail
containing TFA/TIS/H2O (9.5/0.25/0.25, v/v) at rt for 1 h. The solvent was evaporated
under reduced pressure and the residue was washed thrice with cold diethyl ether (2
mL × 3) to afford compound 3 (14 mg, 91.92 %) as a white solid. Mass: ESI-MS
(m/z): [M+H]+ calcd. for C65H95N19O14: 1365.73; found: 1368.22 [M+H]+, 684.71
[M+2H]2+.
4.2.3.
4-((5-(4-((tert-butoxycarbonyl)amino)-2-oxopyrimidin-1(2H)-yl)-3-((tert-
butoxycarbonyl)oxy)-4,4-difluorotetrahydrofuran-2-yl)methoxy)-4-oxobutanoic acid
(4)
DIPEA (0.75 mL, 4.30 mmol) was added dropwise to a stirring solution of diboc-
protected gemcitabine (200 mg, 0.43 mmol) and succinic anhydride (108 mg, 1.08
mmol) in 20 mL of CH2Cl2. The resulting mixture was stirred at rt for 4 h (reaction
monitored via TLC in solvent system CH2Cl2/methanol 9/1, Rf= 0.04) and then 10
mL of water were added and the mixture was lyophilized. The residue was then
purified via RP-HPLC (60/40 % H2O*TFA/MeCN*TFA to 100 % MeCN*TFA for
30 min, at 254 nm, tR = 11.75 min) to afford compound 4 (190 mg, 78 %) as a white
solid.
1H-NMR of 4 (400 MHz, DMSO-d6, 25οC): δ = 12.31 (s, 1 H), 10.61 (s, 1 H), 8.01
(d, J = 7.6 Hz, 1 H), 7.15 (d, J = 7.6 Hz, 1 H), 6.33 (t, J = 8.7 Hz, 1 H), 5.33 (m, 1 H),

ACCEPTED MANUSCRIPT
4.51 (m, 1 H), 4.48 (m, 1 H), 4.41 (m, 1 H), 2.6 (t, 2 H), 2.54 (t, 2 H), 1.50 (s, 9 H),
13
1.49 (s, 9 H) ppm. C-NMR (400 MHz, DMSO-d6, 25οC): δ = 174.27 (C), 172.84
(C), 164.67 (C), 154.72 (C), 152.84 (C), 152.02 (C), 146.18 (C), 122.13 (CF2), 96.32
(CH), 85.60 (CH), 84.81 (C), 82.32 (C), 76.91 (CH), 73.88 (CH), 63.24 (CH2), 29.51
(CH2), 28.66 (CH3), 28.05 (CH3) ppm. Mass: ESI-MS (m/z) for C23H30F2N3O11:
563.5; found: 562.1 [M-H-]-.
4.2.4. GSHG
DIPEA (10.2 ul, 0.058 mmol) was added dropwise under inert atmosphere to a
stirring solution of 4 (6.6 mg, 0.011 mmol) and HATU (4.18 mg, 0.011 mmol) in 3
mL anhydrous DMF. After 5 mins, a solution of 3 (16 mg, 0.011 mmol) in 2 mL
anhydrous DMF was added dropwise and the resulting mixture was stirred at rt for
12h. TLC in solvent system of n-BuOH/acetic acid/water (6/2/2, v/v) showed the
complete consumption of both starting compounds 3 and 4 and the formation of a new
spot (Rf=0.31). DMF was removed under high vacuum and the residue was washed
thrice with MeCN (1 mL × 3) to afford 11 mg of white solid which was used for the
next reaction (Boc-deprotection) without further purification. Boc-deprotection was
performed with 2 mL of cleavage cocktail containing TFA/TIS/H2O (9.5/0.25/0.25,
v/v) at rt for 1 h. The solvent was evaporated under reduced pressure and the residue
was washed thrice with cold diethyl ether (2 mL × 3) to afford crude GSHG. After
RP-HPLC purification (90/10 % H2O*TFA/MECN*TFA to 70 % H2O*TFA for 30
t
min, at 214 nm, R=23.35 min), 5 mg (24.91 %) of pure GSHG were obtained as a
white solid. Mass: ESI-MS (m/z) calcd. for C78H108F2N22O20: 1710.81 [M+H]+;
found: 856.3 [M+2H]2+, 571.3 [M+3H]3+.
4.2.5. tert-butyl (1-(4-((tert-butoxycarbonyl)oxy)-3,3-difluoro-5-((((4-nitrophenoxy)
carbonyl)oxy)methyl)tetrahydrofuran-2-yl)-2-oxo-1,2-dihydropyrimidin-4-
yl)carbamate (2a) and tert-butyl (1-(5-(((tert-butoxycarbonyl)oxy)methyl)-3,3-
difluoro-4-(((4-nitrophenoxy)
carbonyl)oxy)tetrahydrofuran-2-yl)-2-oxo-1,2-
dihydropyrimidin-4-yl)carbamate (2b)
DIPEA (752.4 ul, 4.319 mmol) was added to a solution of 1a/2b (50 mg, 0.107 mmol)
0
in anhydrous MeCN (10 mL), under inert atmosphere at -10 C (ice/acetone bath).
After 5 min, a solution of Bis(4-nitrophenyl)carbonate (260.4 mg, 0.856 mmol) in 5
mL anhydrous MeCN was added dropwise and the resulting mixture was stirred at rt

ACCEPTED MANUSCRIPT
The progress of the reaction was monitored via TLC in acetone/CH2Cl2 (1/10, v/v,
Rf=0.79). After 4 h, solvent was removed under reduced pressure and the residue was
purified via column chromatography (eluant: CH2Cl2/acetone 9/1) to afford compound
2a/2b (60 mg, 88.58 %)/(26mg, 38.39 %) as white solids. 1H-NMR of 2a (400 MHz,
DMSO-d6, 25οC):δ = 10.58 (s, 1 H), 8.33 (d, J = 9.14 Hz, 2H), 8.02 (d, J = 7.67 Hz,
1H), 7.59 (d, J = 9.14 Hz, 2H), 7.08 (d, J = 7.55 Hz, 1H), 6.31 (t, J = 8.69 Hz, 1H),
5.46 - 5.36 (m, 1H), 4.69 - 4.67 (m, 1H), 4.64 - 4.56 (m, 2H), 1.46 (s, 9H), 1.45 (s,
9H) ppm; 13C-NMR of 2a (400 MHz, DMSO-d6, 25οC):δ = 163.84 (C), 155.17 (C),
153.80 (C), 151.86 (C), 151.73 (C), 151.11 (C), 145.60 (CH), 145.28 (C), 126.20 (C),
125.45 (CH), 122.62 (CH), 115.83 (C), 95.35 (CH), 83.95 (C), 81.45 (C), 75.69 (CH),
72.86 (CH), 67.13 (CH2), 27.73 (CH3), 27.13 (CH3) ppm; Mass: ESI-MS (m/z) calcd.
for C26H30F2N4O12: 628.18 [M+H]+, found: 629.93 [M+H]+.
1H-NMR of 2b (400 MHz, DMSO-d6, 25οC):δ = 10.60 (s, 1 H), 8.35 (d, J = 9.15 Hz,
2H), 8.04 (d, J = 7.7 Hz, 1H), 7.64 (d, J = 9.15Hz, 2H), 7.09 (d, J = 7.65 Hz, 1H), 6.35
(t, J = 8.56 Hz, 1H), 5.52 (m, 1H), 4.62 - 4.54 (m, 2H), 4.45 - 4.35 (m, 1H), 1.46 (s,
13
9H), 1.42 (s, 9H) ppm; C-NMR of 2b (400 MHz, DMSO-d6, 25οC): δ = 163.87
(C), 154.82 (C), 153.82 (C), 152.53 (C), 151.87 (C), 150.74 (C), 145.51 (CH), 126.19
(C), 125.56 (CH), 122.45 (CH), 115.80 (C), 95.30 (CH), 82.23 (CH), 81.45 (C), 75.77
(C), 74.60 (CH), 72.49 (CH), 65.24 (CH2), 28.74 (CH3), 27.29 (CH3) ppm; Mass:
ESI- MS (m/z) calcd. for C26H30F2N4O12: 628.18 [M+H]+, found: 629.87 [M+H]+.
4.2.6. 2G1 and 2G2
DIPEA (8.34 ul, 0.047 mmol) was added dropwise to a solution of GnRH (20 mg,
0.01596 mmol) in 2 mL anhydrous DMF under inert atmosphere at 0 oC. After 5 min,
a solution of 2a/2b (10 mg, 0.015 mmol) in anhydrous 1 mL DMF, was added
dropwise and the resulting mixture was stirred at rt. The reaction was monitored via
TLC using acetone/CH2Cl2 1/9 and then n-BuOH/AcOH/H2O (6/2/2, v/v, Rf=0.33).
After 12 h, solvent was removed under high vacuum and the residue was washed
twice with acetonitrile (2 mL × 2). The residues were dissolved in 3 mL of cleavage
o
cocktail (TFA/TIS/ H2O, 9.5/0.25/0.25, v/v) at 0 C and stirred at rt for 12 h. Solvent
was evaporated under reduced pressure and the residue was washed thrice with cold
diethyl ether (2 mL × 3). Crude compounds were purified with semi preparative RP-
HPLC (90/10 % H2O*TFA/MeCN*TFA to 70 % H2O*TFA for 30 min, at 214 nm,

ACCEPTED MANUSCRIPT
t
tR2G1=29.50 min and R2G2=28.50 min) and the fractions were lyophilized to afford
compound 2G1/2G2 (12.5 mg, 50.81 %) / (12 mg, 48.78 %).
Mass of 2G1: ESI-MS (m/z): calcd. for C69 H93F2N4O18: 1541.8 [M+H]+, found: 771.9
[M+2H]2+; 514.9 [M+3H]3+.
Mass of 2G2: ESI-MS (m/z): calcd. for C69 H93F2N4O18: 1541.8 [M+H]+, found:
771.8 [M+2H]2+; 514.9 [M+3H]3+.
4.3 Characterization and quantitative analysis of GnRH-gemcitabine conjugates,
gemcitabine and dFdU by Liquid Chromatography-Mass Spectrometry (LC-
MS/MS)
For the identification and quantification of gemcitabine, dFdU and GnRH-
gemcitabine conjugates, LC-MS/MS methodologies were developed and validated as
described previously [42]. HPLC was performed using a Dionex Ultimate 3000
system (Dionex Corporation, Germering, Germany) equipped with three pumps (two
for nano and one for micro LC), a temperature-controlled column compartment and an
autosampler. A C18 column (Agilent, ZORBAX Eclipse, 4.6 x 150 mm, 5 µM) was
used at a flow rate of 1 mL/min for the separation of analytes of interest. The mobile
phase consisted of A: 10% MeCN, 90% water, 2 mM ammonium acetate, and 0.1%
FA and B: 90% MeCN, 10% water, 2 mM ammonium acetate, and 0.1% FA. Mass
spectrometry was performed on an API 4000 QTRAP LC-MS/MS system fitted with
a TurboIonSpray source and a hybrid triple quadrupole/linear ion trap mass
spectrometer (Applied Biosystems, Concord, Ontario, Canada).
4.4 Binding to the human GnRH-R
Radioiodination of D-Tyr6-His5-GnRH, preparation of membrane homogenates from
HEK 293 cells stably expressing the GnRH-R and binding of the GnRH-gemcitabine
conjugates to the human GnRH-R was performed as described before [43, 44]. In
brief, aliquots of diluted membrane suspension (50 µL) were added into tubes
containing buffer B (25 mM HEPES containing 1 mM CaCl2, 10 mM MgCl2, 0.5%
o
BSA, pH 7.4 at 4 C) and 100,000-120,000 cpm [125I]-D-Tyr6-His5-GnRH with or
without increasing concentrations of GnRH-gemcitabine conjugates in a final volume
of 0.2 mL. The mixtures were incubated at 4 °C for 16-19 h and then filtered using a
Brandel cell harvester through Whatman GF/C glass fiber filters, presoaked for 1-2 h
in 0.5% polyethylenimine at 4 °C. The filters were washed four times with 1.5 mL of

ACCEPTED MANUSCRIPT
o
ice-cold 50 mM Tris-HCl, pH 7.4 at 4 C. Filters were assessed for radioactivity in a
gamma counter (LKB Wallac 1275 minigamma, 80% efficiency). The amount of
membrane used was adjusted to ensure that the specific binding was always equal to
or less than 10% of the total concentration of the added radioligand. Specific [125I]-D-
Tyr6-His5-GnRH binding was defined as total binding less nonspecific binding in the
presence of 1000 nM triptorelin (Bachem, Germany). Data for competition binding
were analyzed by nonlinear regression analysis, using GraphPad Prism 4.0 (GraphPad
Software, San Diego, CA). IC50 values were obtained by fitting the data from
competition studies to a one-site competition model.
4.5 Cell Cultures
The androgen-independent CaP cell lines DU145, PC3, the triple negative breast
cancer MDA-MB-231 and the luminal A breast cancer MCF-7 were used in this study.
Cells were maintained in RPMI (DU145, PC3, MDA-MB-231) or DMEM (MCF7)
medium, containing glutamine and supplemented with 10% Fetal Bovine Serum
(FBS), penicillin (100 U/mL) and streptomycin (100 mg/mL) at 37 °C and 5% CO2.
4.6 Cell Growth Assay
Cells were plated at a density of 5 x 103 cells per well on 96-well plates. After 24 h
o
incubation (37 C, 5% CO2), the cell medium was removed, and compounds were
added at selected concentrations (10-40,000 nM), followed by incubation for 72 h.
The medium was then removed and the MTT solution (0.3 mg/mL in PBS) was added
to cells for 3 h, after which the MTT solution was removed and the formazan crystals
were dissolved in 100 µL DMSO. The optical density was measured at 570 nm and a
reference wavelength of 690 nm using an absorbance microplate reader (SpectraMax
190, Molecular Devices, Sunnyvale, CA, USA). The 50% cytostatic concentration
(IC50) was calculated based on a four-parameter logistic equation using SigmaPlot 12
software (Systat Software, San Jose, CA, USA). Each point was the result of three
experiments performed in triplicate.
4.7 Determination of intracellular concentrations of conjugates, gemcitabine and
dFdU
Cells (DU145) were plated in 6-well plates at a density of 2 x 106 cells/well. Cells
were then incubated with 2G1, 2G2, GSHG or gemcitabine (10 µΜ) for selected time

ACCEPTED MANUSCRIPT
points (1 h, 4 h, 8 h). Incubations were terminated by removing the medium and
washing the cells twice with ice cold PBS to remove unbound gemcitabine or the
conjugates. The cells were then lysed by adding an ice-cold solution of MeCN-Water
(3:2) and scraping the cell monolayer. Samples were subsequently vortexed, sonicated
and centrifuged for three minutes at 16,060 g (Heraeus Biofuge Pico microcentrifuge,
Thermo Scientific, Bonn, Germany). The supernatants were collected, evaporated and
stored at -20 oC until the day of analysis. Intracellular accumulation of the conjugates,
gemcitabine and dFdU was determined by LC-MS/MS analysis using a stable internal
standard as well as gemcitabine and dFdU standards for the construction of analytical
standard curves.
4.8 In vitro stability in cell culture
2G1, 2G2 and GSHG 1 µΜ were incubated in cell culture medium (RPMI, 10%FBS,
o
1 % P/S) at 37 C. Samples (triplicates of 40 µL) were collected at selected time
o
points (t= 0, 1, 2, 4, 8, 24, 48, 72 hours) and were stored at -80 C after mixing with
160 µL of initial mobile phase (90 % H2O, 10 % MeCN, 2 mM ammonium acetate,
0.1 % formic acid). Samples were analyzed using LC-MS/MS.
4.9 In vitro stability in human plasma
1 µg/mL of either 2G1, 2G2 or GSHG were incubated in human plasma. Samples
(triplicates of 50 µL) were collected at selected time points (t= 0, 0.08, 0.5, 1, 4 hours)
o
were stored at -80 C after mixing with 150 µL acetonitrile. Samples were then
extracted using protein precipitation and analyzed using LC-MS/MS.
4.10 In vivo pharmacokinetics
All animal procedures were approved by the Project Evaluation Committee of the
Institution and the competent Veterinary Service of the Prefecture of Athens, in
accordance to the National legal framework on the protection of animals used for
scientific purposes (Presidential Decree 56/2013 in harmonization to the European
Directive 2010/63). For pharmacokinetic studies, animals at the age of 8-12 weeks
were weighed and fasted overnight before dosing (n = 5 per group, male C57BL/6N
inbred strain obtained from Charles River, Calco, Italy). Dosing solutions of GnRH
conjugates (10 mg/kg or 6.3 µmol/kg) or an equimolar dose of gemcitabine (1.65
mg/kg) in saline were administered intraperitoneally (IP). A serial tail bleeding

ACCEPTED MANUSCRIPT
protocol was used for the collection of blood samples. Blood samples (10 µL) were
collected at selected time points (0.25 h, 0.5 h, 1 h, 2 h, 4 h, 6 h, 8 h) in tubes
containing 40 µL citric acid (0.1 M, pH 4.5) and stored at -80 °C until sample
extraction. Samples were prepared for quantification by protein precipitation and
evaporation. 2G1, 2G2, GSHG, gemcitabine and dFdU were quantified by LC-
MS/MS analysis.
4.11 Statistical Analysis
The results presented herein are expressed as mean ± SD. Statistical analyses were
performed by the SigmaPlot 12 software. Statistical significance was determined by
using Student’s t test.
Supporting information
NMR, MS and HPLC spectra for all the synthesized compounds are presented in the
Supporting Information (Figure S1-S19).
Acknowledgments
This work was co-financed by the European Union (European Social Fund ESF) and
reek national funds through the Operational Program “Education and Lifelong
Learning” of the National Strategic Reference Framework (NSRF) – Research
Funding Program: ARISTEIA II [grant number: 5199]. Also, this research was co-
financed by Greece and the European Union (European Social Fund- ESF) through
the Operational Programme “Human Resources Development, Education and
Lifelong Learning 2014-2020” in the context of the project “Strengthening Human
Resources Research Potential via Doctorate Research – 2nd Cycle” (MIS 5000432).
References
[1] A.K. Mitra, V. Agrahari, A. Mandal, K. Cholkar, C. Natarajan, S. Shah, M. Joseph,
H.M. Trinh, R. Vaishya, X. Yang, Y. Hao, V. Khurana, D. Pal, Novel delivery
approaches for cancer therapeutics, Journal of Controlled Release, 219 (2015) 248-
268.
[2] O. Argyros, T. Karampelas, X. Asvos, A. Varela, N. Sayyad, A. Papakyriakou,
C.H. Davos, A.G. Tzakos, D. Fokas, C. Tamvakopoulos, Peptide-Drug Conjugate
GnRH-Sunitinib Targets Angiogenesis Selectively at the Site of Action to Inhibit
Tumor Growth, Cancer research, 76 (2016) 1181-1192.

ACCEPTED MANUSCRIPT
[3] T. Karampelas, O. Argyros, N. Sayyad, K. Spyridaki, C. Pappas, K. Morgan, G.
Kolios, R.P. Millar, G. Liapakis, A.G. Tzakos, D. Fokas, C. Tamvakopoulos, GnRH-
Gemcitabine conjugates for the treatment of androgen-independent prostate cancer:
pharmacokinetic enhancements combined with targeted drug delivery, Bioconjugate
chemistry, 25 (2014) 813-823.
[4] E.I. Vrettos, G. Mezo, A.G. Tzakos, On the design principles of peptide-drug
conjugates for targeted drug delivery to the malignant tumor site, Beilstein journal of
organic chemistry, 14 (2018) 930-954.
[5] M.A. El Mubarak, I. Leontari, G. Efstathia, E.I. Vrettos, A.K. Shaikh, S.E.
Konstantinos, C. Danika, H.P. Kalofonos, A.G. Tzakos, G.B. Sivolapenko,
Development of a novel conjugatable sunitinib analogue validated through in vitro
and in vivo preclinical settings, Journal of chromatography. B, Analytical
technologies in the biomedical and life sciences, 1092 (2018) 515-523.
[6] M.J. Akhtar, M. Ahamed, H.A. Alhadlaq, S.A. Alrokayan, S. Kumar, Targeted
anticancer therapy: overexpressed receptors and nanotechnology, Clinica chimica
acta; international journal of clinical chemistry, 436 (2014) 78-92.
[7] G.A. Gonzalez-Conchas, L. Rodriguez-Romo, D. Hernandez-Barajas, J.F.
Gonzalez-Guerrero, I.A. Rodriguez-Fernandez, A. Verdines-Perez, A.J. Templeton, A.
Ocana, B. Seruga, I.F. Tannock, E. Amir, F.E. Vera-Badillo, Epidermal growth factor
receptor overexpression and outcomes in early breast cancer: A systematic review and
a meta-analysis, Cancer Treatment Reviews, 62 (2018) 1-8.
[8] J.B. Engel, H.R. Tinneberg, F.G. Rick, E. Berkes, A.V. Schally, Targeting of
Peptide Cytotoxins to LHRH Receptors For Treatment of Cancer, Current drug targets,
17 (2016) 488-494.
[9] A. Nagy, A.V. Schally, P. Armatis, K. Szepeshazi, G. Halmos, M. Kovacs, M.
Zarandi, K. Groot, M. Miyazaki, A. Jungwirth, J. Horvath, Cytotoxic analogs of
luteinizing hormone-releasing hormone containing doxorubicin or 2-
pyrrolinodoxorubicin, a derivative 500-1000 times more potent, Proceedings of the
National Academy of Sciences, 93 (1996) 7269-7273.
[10] Y. Wang, A.G. Cheetham, G. Angacian, H. Su, L. Xie, H. Cui, Peptide-drug
conjugates as effective prodrug strategies for targeted delivery, Adv Drug Deliv Rev,
110-111 (2017) 112-126.
[11] L. Ma, C. Wang, Z. He, B. Cheng, L. Zheng, K. Huang, Peptide-Drug Conjugate:
A Novel Drug Design Approach, Current medicinal chemistry, 24 (2017) 3373-3396.
[12] V. Kushwah, S.S. Katiyar, A.K. Agrawal, I. Saraf, I.P. Singh, D.A. Lamprou,
R.C. Gupta, S. Jain, Implication of linker length on cell cytotoxicity, pharmacokinetic
and toxicity profile of gemcitabine-docetaxel combinatorial dual drug conjugate,
International journal of pharmaceutics, 548 (2018) 357-374.
[13] K.C. Cundy, S. Sastry, W. Luo, J. Zou, T.L. Moors, D.M. Canafax, Clinical
pharmacokinetics of XP13512, a novel transported prodrug of gabapentin, Journal of
clinical pharmacology, 48 (2008) 1378-1388.
[14] N. Masuda, S.J. Lee, S. Ohtani, Y.H. Im, E.S. Lee, I. Yokota, K. Kuroi, S.A. Im,
B.W. Park, S.B. Kim, Y. Yanagita, S. Ohno, S. Takao, K. Aogi, H. Iwata, J. Jeong, A.
Kim, K.H. Park, H. Sasano, Y. Ohashi, M. Toi, Adjuvant Capecitabine for Breast
Cancer after Preoperative Chemotherapy, The New England journal of medicine, 376
(2017) 2147-2159.
[15] M.-L. Ţînţaş, V. Gembus, F. Alix, A. Barré, G. Coadou, L. Truong, M. Sebban,
C. Papamicaël, H. Oulyadi, V. Levacher, Rational design of carbamate-based dual
binding site and central AChE inhibitors by a “biooxidisable” prodrug approach:

ACCEPTED MANUSCRIPT
Synthesis, in vitro evaluation and docking studies, European Journal of Medicinal
Chemistry, 155 (2018) 171-182.
[16] A.K. Ghosh, M. Brindisi, Organic carbamates in drug design and medicinal
chemistry, Journal of medicinal chemistry, 58 (2015) 2895-2940.
[17] Y. Gilad, E. Noy, H. Senderowitz, A. Albeck, M.A. Firer, G. Gellerman,
Synthesis, biological studies and molecular dynamics of new anticancer RGD-based
peptide conjugates for targeted drug delivery, Bioorganic & Medicinal Chemistry, 24
(2016) 294-303.
[18] Y. Li, Y. Wang, R. Zhang, C. Liu, Y. Wei, J. Sun, Z. He, Y. Xu, T. Zhang,
Improving the oral bioavailability of tapentadol via a carbamate prodrug approach:
synthesis, bioactivation, and pharmacokinetics, Drug delivery and translational
research, 8 (2018) 1335-1344.
[19] M.V. Chatziathanasiadou, E.G. Geromichalou, N. Sayyad, E.I. Vrettos, A.
Katsikoudi, E. Stylos, S. Bellou, G.D. Geromichalos, A.G. Tzakos, Amplifying and
broadening the cytotoxic profile of quercetin in cancer cell lines through
bioconjugation, Amino acids, 50 (2018) 279-291.
[20] T.F. Kellici, M.V. Chatziathanasiadou, M.S. Lee, N. Sayyad, E.G. Geromichalou,
E.I. Vrettos, A.D. Tsiailanis, S.W. Chi, G.D. Geromichalos, T. Mavromoustakos, A.G.
Tzakos, Rational design and structure-activity relationship studies of quercetin-amino
acid hybrids targeting the anti-apoptotic protein Bcl-xL, Organic & biomolecular
chemistry, 15 (2017) 7956-7976.
[21] N.A. Thiele, J. Kärkkäinen, K.B. Sloan, J. Rautio, K.M. Huttunen, Secondary
carbamate linker can facilitate the sustained release of dopamine from brain-targeted
prodrug, Bioorg Med Chem Lett, 28 (2018) 2856-2860.
[22] A. Mattarei, M. Azzolini, M. La Spina, M. Zoratti, C. Paradisi, L. Biasutto,
Amino Acid Carbamates As Prodrugs Of Resveratrol, 5 (2015) 15216.
[23] A. Primikyri, N. Sayyad, G. Quilici, E.I. Vrettos, K. Lim, S.W. Chi, G. Musco,
I.P. Gerothanassis, A.G. Tzakos, Probing the interaction of a quercetin bioconjugate
with Bcl-2 in living human cancer cells with in-cell NMR spectroscopy, FEBS letters,
592 (2018) 3367-3379.
[24] L.P. Jordheim, D. Durantel, F. Zoulim, C. Dumontet, Advances in the
development of nucleoside and nucleotide analogues for cancer and viral diseases,
Nat Rev Drug Discov, 12 (2013) 447-464.
[25] J. Carmichael, The role of gemcitabine in the treatment of other tumours, British
Journal of Cancer, 78 (1998) 21-25.
[26] M. Gusella, F. Pasini, C. Bolzonella, S. Meneghetti, C. Barile, A. Bononi, S.
Toso, D. Menon, G. Crepaldi, Y. Modena, L. Stievano, R. Padrini, Equilibrative
nucleoside transporter 1 genotype, cytidine deaminase activity and age predict
gemcitabine plasma clearance in patients with solid tumours, British journal of
clinical pharmacology, 71 (2011) 437-444.
[27] E. Mini, S. Nobili, B. Caciagli, I. Landini, T. Mazzei, Cellular pharmacology of
gemcitabine, Annals of Oncology, 17 (2006) v7-v12.
[28] L. de Sousa Cavalcante, G. Monteiro, Gemcitabine: metabolism and molecular
mechanisms of action, sensitivity and chemoresistance in pancreatic cancer, European
journal of pharmacology, 741 (2014) 8-16.
[29] D.R. Rauchwerger, P.S. Firby, D.W. Hedley, M.J. Moore, Equilibrative-sensitive
nucleoside transporter and its role in gemcitabine sensitivity, Cancer research, 60
(2000) 6075-6079.
[30] J. Spratlin, R. Sangha, D. Glubrecht, L. Dabbagh, J.D. Young, C. Dumontet, C.
Cass, R. Lai, J.R. Mackey, The absence of human equilibrative nucleoside transporter

ACCEPTED MANUSCRIPT
1 is associated with reduced survival in patients with gemcitabine-treated pancreas
adenocarcinoma, Clinical cancer research : an official journal of the American
Association for Cancer Research, 10 (2004) 6956-6961.
[31] B. Hryciuk, B. Szymanowski, A. Romanowska, E. Salt, B. Wasąg, B. Grala, J.
Jassem, R. Duchnowska, Severe acute toxicity following gemcitabine administration:
A report of four cases with cytidine deaminase polymorphisms evaluation, Oncology
letters, 15 (2018) 1912-1916.
[32] S.A. Veltkamp, J.H. Beijnen, J.H. Schellens, Prolonged versus standard
gemcitabine infusion: translation of molecular pharmacology to new treatment
strategy, The oncologist, 13 (2008) 261-276.
[33] Y. Zhou, Q. Chang, W. Wang, X. Zhang, F. Zhou, J. Sun, G. Wang, Y. Peng,
Sensitive analysis and pharmacokinetic study of a novel gemcitabine carbamate
prodrug and its active metabolite gemcitabine in rats using LC-ESI-MS/MS, Journal
of chromatography. B, Analytical technologies in the biomedical and life sciences,
1083 (2018) 249-257.
[34] H. Zhang, K. Wang, K. Na, D. Li, Z. Li, D. Zhao, L. Zhong, M. Wang, L. Kou, C.
Luo, H. Zhang, Q. Kan, H. Ding, Z. He, J. Sun, Striking a Balance between
Carbonate/Carbamate Linkage Bond- and Reduction-Sensitive Disulfide Bond-
Bearing Linker for Tailored Controlled Release: In Situ Covalent-Albumin-Binding
Gemcitabine Prodrugs Promote Bioavailability and Tumor Accumulation, Journal of
medicinal chemistry, 61 (2018) 4904-4917.
[35] J.T. Weiss, J.C. Dawson, C. Fraser, W. Rybski, C. Torres-Sanchez, M. Bradley,
E.E. Patton, N.O. Carragher, A. Unciti-Broceta, Development and bioorthogonal
activation of palladium-labile prodrugs of gemcitabine, Journal of medicinal
chemistry, 57 (2014) 5395-5404.
[36] W.E.K. Oskar Keller, Gert van Look, Gernot Wersin, tert-
BUTOXYCARBONYLATION OF AMINO ACIDS AND THEIR DERIVATIVES:
N-tert-BUTOXYCARBONYL-l-PHENYLALANINE, Organic Syntheses, (1985).
[37] E.I. Vrettos, N. Sayyad, E.M. Mavrogiannaki, E. Stylos, A.D. Kostagianni, S.
Papas, T. Mavromoustakos, V. Theodorou, A.G. Tzakos, Unveiling and tackling
guanidinium peptide coupling reagent side reactions towards the development of
peptide-drug conjugates, RSC Advances, 7 (2017) 50519-50526.
[38] Z. Guo Zw, J.M. Gallo, Selective Protection of 2',2'-Difluorodeoxycytidine
(Gemcitabine), The Journal of organic chemistry, 64 (1999) 8319-8322.
[39] X. Peng, B.I. Knapp, J.M. Bidlack, J.L. Neumeyer, High-affinity carbamate
analogues of morphinan at opioid receptors, Bioorganic & medicinal chemistry letters,
17 (2007) 1508-1511.
[40] H. Guo, J. Lu, H. Hathaway, M.E. Royce, E.R. Prossnitz, Y. Miao, Synthesis and
evaluation of novel gonadotropin-releasing hormone receptor-targeting peptides,
Bioconjugate chemistry, 22 (2011) 1682-1689.
[41] M. Schottelius, S. Berger, T. Poethko, M. Schwaiger, H.J. Wester, Development
of novel 68Ga- and 18F-labeled GnRH-I analogues with high GnRHR-targeting
efficiency, Bioconjugate chemistry, 19 (2008) 1256-1268.
[42] Z.D. Sofianos, T. Katsila, N. Kostomitsopoulos, V. Balafas, J. Matsoukas, T.
Tselios, C. Tamvakopoulos, In vivo evaluation and in vitro metabolism of leuprolide
in mice--mass spectrometry-based biomarker measurement for efficacy and toxicity,
Journal of mass spectrometry : JMS, 43 (2008) 1381-1392.
[43] D. Laimou, T. Katsila, J. Matsoukas, A. Schally, K. Gkountelias, G. Liapakis, C.
Tamvakopoulos, T. Tselios, Rationally designed cyclic analogues of luteinizing

ACCEPTED MANUSCRIPT
hormone-releasing hormone: enhanced enzymatic stability and biological properties,
European journal of medicinal chemistry, 58 (2012) 237-247.
[44] T. Katsila, E. Balafas, G. Liapakis, P. Limonta, M. Montagnani Marelli, K.
Gkountelias, T. Tselios, N. Kostomitsopoulos, J. Matsoukas, C. Tamvakopoulos,
Evaluation of a stable gonadotropin-releasing hormone analog in mice for the
treatment of endocrine disorders and prostate cancer, The Journal of pharmacology
and experimental therapeutics, 336 (2011) 613-623.