Design and Synthesis of an RGD Peptidomimetic-Paclitaxel Conjugate Targeting α v β 3 Integrin for Tumour-Directed Drug Delivery

Article abstract of DOI:10.1055/s-0036-1590898

A 1,2,3-triazole-based RGD peptidomimetic having nanomolar affinity for α _v β ₃ integrin was conjugated to the potent antimitotic paclitaxel via an oxime heterobifunctional linker. The resulting construct maintained nanomolar binding

Full text of DOI:10.1055/s-0036-1590898

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
©
Georg Thieme Verlag Stuttgart · New York
2017, 28, A–H
letter
A
en
Synlett
M. Piras et al.
Letter
Design and Synthesis of an RGD Peptidomimetic-Paclitaxel Conju-
gate Targeting α β Integrin for Tumour-Directed Drug Delivery
v 3
Monica Piras*a
Alexandra Andriua
Andrea Testaa
Paul Wieneckea
Ian N. Fleminga
Matteo Zanda*a,b
a
Institute of Medical Sciences and Kosterlitz Centre for
Therapeutics, School of Medicine, Medical Sciences and
Nutrition, University of Aberdeen, Foresterhill, Aberdeen,
AB25 2ZD, Scotland, UK
m.zanda@abdn.ac.uk
C.N.R. – I.C.R.M., via Mancinelli 7, 20131 Milan, Italy
b
This article is dedicated to Professor Victor Snieckus on the
th
occasion of his 80 birthday
Received: 16.06.2017
Accepted after revision: 07.08.2017
Published online: 01.09.2017
have been discovered and extensively investigated as valu-
able tools for tumour-targeted delivery of chemotherapeu-
tics.1 RGD-containing peptides targeting α
,2
β
integrin be-
DOI: 10.1055/s-0036-1590898; Art ID: st-2017-r0479-l
v
3
long to this class of tumour-homing vectors. The overex-
Abstract A 1,2,3-triazole-based RGD peptidomimetic having nanomo-
lar affinity for αvβ3 integrin was conjugated to the potent antimitotic
paclitaxel via an oxime heterobifunctional linker. The resulting con-
struct maintained nanomolar binding concentration to αvβ3 integrin
and showed 11-fold selectivity in terms of cytotoxicity towards highly
αvβ expressing U87MG cancer cells relative to non αvβ expressing
pression of α β on solid tumours, along with its capacity to
v
3
be internalised via endocytosis after binding of ligands such
as RGD peptides, makes it an attractive target for tumour
3
delivery of anticancer agents. A number of cytotoxic drugs
3
3
have been conjugated to tumour-homing peptides contain-
ing the RGD motif and the resulting constructs assessed in
animal models. Several studies in this field provided experi-
mental evidence that RGD-based strategies contribute to
reduce the toxicity and augment the therapeutic window of
MCF7 cells, indicating promising cancer cell targeting capacity.
1
2
3
4
5
6
7
Design
Retrosynthesis
Synthesis
Solid-Phase Receptor Binding Assay
Cell Cytotoxicity Assays
Metabolic Stability Assays
Conclusions
1,4,5
existing chemotherapeutics.
HN
NH2
Key words RGD, peptidomimetic, integrin, paclitaxel, click, cytotoxici-
HN
ty, cancer
COOH
O
O
H
N
The efficacy of chemotherapeutic agents clinically used
for the treatment of malignant tumours is generally limited
by their nonspecific toxicity against off-target cells, espe-
cially those characterised by a high proliferation rate, re-
sulting in a reduced therapeutic index and serious side ef-
fects (e.g. alopecia, nausea, vomiting, and immunosuppres-
sion). Several strategies have been developed over the years
in order to overcome these problems and achieve a tumour-
N
H
N
H
O
R
G
D
O
COOH
SO2
N
H
N
N
HN
N
NH
N
1
targeted delivery of cytotoxic agents. An attractive ap-
1
proach is represented by hybrid compounds, containing a
cytotoxic drug bound to a moiety that specifically identifies
protein markers overexpressed on cancer cells. Thanks to
the development of phage display technologies, several
peptide sequences capable of recognising tumour markers
CF3
F
Figure 1 Structure of the RGD parent peptide and a triazole-based
RGD peptidomimetic (1). The chemical groups mimicking the arginine
(R), glycine (G), and aspartate (D) amino acids are coloured in blue,
green, and red, respectively.
6
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–H

B
Synlett
M. Piras et al.
Letter
In previous studies we have described the rational de-
sign of a new class of integrin ligands based on a triazole
scaffold that led to the discovery of RGD peptidomimetics
highly active for α β and selective against the platelet re-
formed by (i) PTX-2′-succinate, (ii) the RGD mimetic 1
equipped with a chemical handle on the triazole ring, and
(iii) a heterobifunctional linker 4 connecting these two por-
tions (Scheme 1).
v
3
6
ceptor α β (compound 1, Figure 1). Herein, we report the
IIb
3
design and synthesis of a 1,2,3-triazole-based RGD peptido-
mimetic conjugated with the potent antimitotic agent pac-
litaxel (PTX), and its selectivity (11-fold) towards α β ex-
v
3
pressing cells relative to PTX alone.
1
Design
An important feature of RGD-cytotoxic conjugates is
that, once internalised into the target cells, they should be
promptly disassembled resulting in the release of the cyto-
toxic agent. Derivatives characterised by an excessive sta-
bility could irreversibly impair the therapeutic activity of
the parent drug. On the other hand, functionalities that
prematurely release the chemotherapeutic agent by meta-
bolic or chemical breakdown, will compromise the targeted
delivery strategy resulting in off-target toxicity. Usually,
amides, hydrazides, carbamates, and sterically hindered es-
ters are used to connect the cytotoxic drug to the tumour-
Figure 2 Predicted binding mode of triazole-based RGD peptidomi-
6
4,7
metics on integrin α β (top). The functional groups involved in the li-
v
3
homing moiety through a linker. Conjugation of PTX to
RGD ligands could be performed via cleavable succinyl ester
linker attached to the 2′-OH group of PTX (Scheme 1). As
the 2′-OH function is essential for PTX to exert its cytotoxic
and microtubule-stabilising activities, it should be easily
released from the ester function once delivered to the target
gand–receptor recognition process are circled in red (bottom). The
position 5 of the triazole ring (blue circle) has been selected as a suit-
able position for the attachment of the cytotoxic payload.
2
Retrosynthesis
8
organ. It has been shown that the succinyl ester bond at
the PTX-2′ position is sufficiently stable to guarantee tu-
mour-selective delivery of RGD-PTX conjugates and is enzy-
matically cleaved in vivo by esterases to provide PTX into its
active form. While the cytotoxic agent is converted into an
inactive prodrug by coupling to an RGD carrier, the RGD li-
The synthesis of conjugate 3 is based on the convergent
assembly of three building blocks (Scheme 1): (i) the PTX-
2′-succinate activated as N-hydroxysuccinimide (NHS) ester
2, (ii) the RGD ligand functionalised with a chemical handle
bearing an aldehyde function 5, and (iii) the heterobifunc-
tional linker 4-(aminooxy)butan-1-amine 4. Based on our
retrosynthetic plan, the aldehyde 5 can be generated by oxi-
dation of the key intermediate alcohol 6 obtained connecting
the iodinated precursor 7 and 4-pentynol 8 via Sonogashira
cross-coupling (Scheme 1).
4
gand should maintain its capacity to bind the α β receptor
v
3
when attached to the cytotoxic payload. In this regard, SAR
6
studies conducted on triazole-based RGD mimetics provid-
ed essential information for the rational design of active de-
rivatives functionalised with a handle necessary for the li-
gation of the cytotoxic cargo. We have demonstrated that
the functional moieties of this class of ligands mainly re-
sponsible for integrin recognition are (i) the carboxylic
function, (ii) the sulfonamide group, (iii) the electron-with-
drawing substituents on the phenyl ring, and (iv) the basic
3
Synthesis
For the synthesis of the building block PTX-2′-succinate-
2
-aminopyridine (Figure 2). Conversely, the triazole ring,
NHS ester 2 (Scheme 1), the 2′-hydroxyl function of PTX
was derivatised with succinic anhydride following a proce-
dure reported in literature.9 The resulting PTX succinate
was activated as NHS ester 2 using N,N′-dicyclohexylcarbo-
lying at the interface between the α and β subunits, acts as
a spacer between the major functional groups involved in
ligand–receptor interactions and, therefore, it appears to be
the most suitable site for functionalisation. Noteworthy is
the fact that incorporation of a bulky substituent – like io-
dine – in position 5 of the triazole ring did not affect the
9
diimide (DCC) as coupling reagent, as already described.
The key RGD intermediate 6 (Scheme 2) was synthe-
sised via Sonogashira cross-coupling between the iodinated
precursor 7 and 4-pentyn-1-ol (8). A number of conditions
were explored (Table 1). Application of typical Sonogashira
6
binding properties of this class of ligands. Taking into ac-
count all of these aspects, we designed the construct 3
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–H

C
Synlett
M. Piras et al.
Letter
PTX-2'-succinate
OAc
Ph
Ph
O
NH
O
OAc
O
O
NH
O
OH
O
Ph
O
OH
Ph
O
O
O
O
HO
H
O
O
HO
H
O
O
OAc
O
O
O
OAc
Ph
O
O
Ph
O
NH
O
N
heterobifunctional linker
O
PTX-2'-succinate-NHS 2
O
N
O
H2N
NH2
O
COOH
bifunctional linker 4
N
H
HN SO2
N
3
N
N
NH
O
N
F3C
F
O
COOH
functionalised
RGD mimetic
3
N
H
HN SO2
N
N
N
3
NH
5
N
F3C
F
oxidation
O
HO
HO
8
O
COOH
COOH
I
N
H
HN SO2
N
H
HN SO2
N
N
N
N
N
N
3
NH
3
NH
7
N
F3C
N
F
F3C
F
6
Scheme 1 Retrosynthetic approach to the target conjugate 3
protocol on this substrate (phosphine-ligated palladium
catalyst, CuI as co-catalyst, and an amine as a base)10 afford-
ed no reaction at room temperature and decomposition at
higher temperatures (Table 1, entries 1–4). Similarly, fol-
lowing a ligand-, copper-, and amine- free strategy report-
ed by Urgaonkar and Verkade, based on the use of palladi-
um acetate (Pd(OAc) ) and tetrabutylammonium acetate
2
11
(Bu NOAc) as a base, no reaction was observed at room
4
temperature (Table 1, entry 5). By increasing the tempera-
ture no decomposition occurred, although only starting
material was recovered (Table 1, entry 6). Addition of Et N
3
to Pd(OAc) and Bu NOAc gave rise to the formation of trace
2
4
HO
O
COOH
2
I
4
-pentyn-1-ol
N
O
COOH
8
H
HN SO2
N
N
DMF
HN SO2
N
H
3
N
N
NH
N
N
3
N
NH
F3C
7
F
6
N
F3C
F
Scheme 2 Synthesis of 6 via Sonogashira cross-coupling
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–H

D
Synlett
M. Piras et al.
Letter
Table 1 Reaction Conditions Tested for Sonogashira Cross-Coupling
Entry
Catalyst
8 (equiv)
Temp (°C)
Base
Time (h)
Yield (%)a–c
1
2
3
4
5
6
7
Pd(PPh
Pd(PPh
Pd(PPh
Pd(PPh
3
3
3
3
)
)
)
)
4
4
2
2
+ CuI
+ CuI
1
25
60
25
60
25
60
60
60
Et
Et
Et
Et
3
3
3
3
N
N
N
N
12
6
NR
1
D
Cl
Cl
2
2
+ CuI
+ CuI
1
12
6
NR
1
D
Pd(OAc)
Pd(OAc)
Pd(OAc)
Pd(OAc)
2
1.25
1.25
1.25
10
Bu
Bu
Bu
Bu
4
4
4
4
NOAc
12
12
6
NR
2
2
2
NOAc
NR
NOAc + Et
3
3
N
N
low yield
70
8
NOAc + Et
6
a
The reaction course was monitored via RP-HPLC coupled to tandem ESI-MS.
NR = no reaction, starting material recovered.
D = decomposition.
b
c
amounts of 6 at 60 °C after six hours (Table 1, entry 7). Us-
4 Solid-Phase Receptor Binding Assay
ing Et N (3 equiv) and a large excess of alkyne (10 equiv)
3
the desired alkynylated product 6 was eventually obtained in
good yields (70%) in less than six hours (Table 1, entry 8).12
The heterobifunctional linker 4 (Scheme 3) was synthe-
sised starting from 1-bromo-4-cloro butane (9) that, react-
ing with sodium azide at room temperature in DMF, afford-
ed 1-cloro-4-azido butane (10). Treatment of 10 with N-
Boc-protected hydroxylamine in the presence of a base led
to the formation of 11, which was converted into the amino
derivative 12 by Staudinger reaction. N-Boc cleavage of 12
with TFA afforded the final product 4 in good overall yields.
The assembly of the conjugate 3 was straightforward
The RGD mimetic 6 functionalised with a chemical han-
dle and the conjugate construct 3 were submitted to biolog-
ical assays to measure their affinity for the α β receptor.
v
3
Competitive affinities were calculated analysing the inter-
actions between immobilised αvβ3 and the commercially
available c[RGDfK(Biotin-PEG-PEG)] peptide in the pres-
ence of increasing amounts of ligands, as already de-
scribed.15 The IC50 values of 6 and 3 were compared to that
of the nonfunctionalised peptidomimetic 1 and the α β in-
v
3
hibitor c(RGDfK) (Table 2). Notably, functionalisation of the
highly active RGD mimetic 1 with an alkynyl chemical han-
dle to give the derivative 6 was performed without perturb-
(Scheme 4). Conversion of alcohol 6 into aldehyde 5 was
performed under mild conditions using Dess–Martin perio-
dinane (DMP) reagent. The reaction was monitored by RP-
HPLC coupled to tandem ESI-MS. After total consumption of
the starting material 6, the heterobifunctional linker 4 was
added to the same reaction flask generating the intermedi-
ate 13 via chemoselective oxime bond formation. Purifi-
cation of 13 was performed by semipreparative RP-HPLC.
The free amino function of 13 was then reacted with PTX-
Table 2 Inhibition of Cyclo[RGDfK(Biotin-PEG-PEG)] Binding to Immo-
bilised α
β
v 3
Entry
Compound
IC50 (nM)a
13
1
2
3
4
c(RGDfK)
138 ± 51
25 ± 3
1
6
3
20 ± 1
2
′-succinate-NHS ester 2 in the presence of DIPEA to pro-
14
vide the final conjugate 3 in 54% yield after chromato-
200 ± 46
graphic purification.
a
Data are the average of two independent experiments performed in tripli-
cate; standard deviations are shown.
H
N
H
i
ii
N3 iii
N
NH2
Br
N3
Boc
O
Boc
O
Cl
Cl
3
3
3
3
9
10
11
12
iv
H2N
NH2
O
3
4
Scheme 3 Synthesis of 4. Reagents and conditions: (i) NaN , DMF, r.t., overnight, 99%; (ii) BocNHOH, DBU, DMF, r.t., 2 h, 40%; (iii) Ph P, THF, r.t., 1 h then
3
3
H O, 60 °C, 4h, 99%; (iv) TFA–CH Cl (1:1), r.t., 2 h, 99%.
2
2
2
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–H

E
Synlett
M. Piras et al.
Letter
HO
2
O
O
COOH
2
N
H
O
COOH
HN
HN SO2
i
N
N
N
H
SO2
N
3
NH
N
N
N
3
N
6
NH
F3C
F
N
5
F3C
F
4
ii
OAc
O
NH
O
O
OH
O
O
H2N
4
O
HO
iii
O N
O
O
O
OAc
2
Ph
O
COOH
HN SO2
O
2
N
HN
4
O
H
N
N
N
2
3
N
COOH
HN
O
NH
N
N
13
F3C
H
SO2
F
N
N
N
3
NH
N
3
F3C
F
Scheme 4 Synthesis of 3. Reagents and conditions: (i) DMP, MeCN, r.t., 6 h, then (ii) 4, r.t., 1 h, 60%; (iii) DIPEA, DMF, r.t., 5 h, 54%.
ing the binding affinity of the parent ligand (Table 2, entries
and 3). Conversely, the large increase of steric hindrance
In order to reliably assess the selective cytotoxicity of con-
2
jugate 3 against α β positive cell lines, a ‘cell washout’ cy-
v
3
in the conjugate 3 caused a ca. tenfold drop in binding affin-
ity (Table 2, entries 2 and 4). However, the IC50 value of 3
was still in the nanomolar range and not significantly dif-
ferent from that of the reference c(RGDfK) peptide used as
positive control (Table 2, entries 1 and 4).
totoxicity assay was performed.16 Following this procedure
– aimed at mimicking the in vivo conditions where the ad-
ministered drug is rapidly cleared from the extracellular tu-
mour environment – cells were incubated with the cytotox-
ic agents for a short period of time (6 h) and then the medi-
um was washed away to remove the noninternalised
conjugate 3. Using this method, the possible antiprolifera-
tive activity arising from extracellular cleavage of the linker
from 3 – and consequent undesired drug release – can be
minimised.17 As expected, cytotoxicity assays clearly indi-
cated that the tumour-homing peptidomimetic 1 was not
cytotoxic (Table 3) when used at concentrations up to
10,000 nM. In contrast, PTX was potently cytotoxic to both
5
Cell Cytotoxicity Assays
Conjugate 3 was tested in vitro to assess its cytotoxicity
towards MCF7 and U87MG tumour cells, which express no
and high levels of α β integrin, respectively. The unconju-
v
3
gated ligand 1 and free PTX were also included in the study.
Table 3 Cell Sensitivity of U87MG and MCF7 Tumour Cells to PTX, 3, and 1
Cell line
Compound 1 IC50 (nM)a
Compound 3 IC50 (nM)a
PTX IC50 (nM)a
3
3
3
MCF7 (α
v
β
3
–)
++)
>10,000
>10,000
–
21.6·10 ± 10.6·10
41.3 ± 14
44.4 ± 26
1.0
3
U87MG (α
v
β
3
1.99·10 ± 1.59·10
Selectivityb,c
11
a
IC50 values were calculated as the concentration of compound required for 50% inhibition of cell viability in culture. The number of cells in each well was mea-
18
sured using an 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) cell proliferation assay, as described.
b
Cytotoxicity ratio for U87MG vs MCF7 is statistically significant; (p value <0.01 in one way ANOVA). Compound 1 exhibited no significant cytotoxicity when
tested at concentrations up 10,000 nM.
c
Selectivity: IC50(MCF7)/IC50(U87MG).
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–H

F
Synlett
M. Piras et al.
Letter
cell lines, as a result of its capacity to readily enter the cells
and inhibit microtubule disassembly. Conjugate 3 was also
cytotoxic for both cell lines, although IC50 values were con-
siderably higher than for PTX, indicating that the construct
was not hydrolysed and did not release free PTX during in-
cubation.
struct 3 caused a ca. tenfold drop in binding affinity, how-
ever, the IC50 value of 3 remained in the nanomolar range
and was not significantly different relative to that of the ref-
erence c(RGDfK) peptide. PTX and conjugate 3 were then
tested in vitro to assess their cytotoxic activity towards
MCF7 and U87MG tumour cells, which express no and high
levels of α β integrin, respectively, using a ‘cell washout’
As expected, PTX was not α β selective, showing essen-
v
3
v 3
tially the same IC50 on both highly α β integrin expressing
assay. Under these conditions, PTX was found to be cytotox-
v
3
cells (U87MG) and nonexpressing cells (MCF7). In contrast,
the conjugate 3 showed good selectivity index = 11 (deter-
mined by the IC (MCF7)/IC (U87MG) ratio), indicating a
ic on both cell lines irrespectively of their α β integrin-ex-
v
3
pression level, while the cytotoxic activity of conjugate 3
was dependent on α β integrin expression. These results
50
50
v 3
higher cytotoxicity for U87MG vs. MCF7 cells. This is con-
sistent with the hypothesis that compound 3 can selectively
target cancer cells via α β integrin.
demonstrate the capacity of 3 to selectively deliver the cy-
totoxic paclitaxel cargo through α β integrin interaction. In
v
3
vitro human and rat liver microsomal stability tests showed
a remarkable stability of both conjugate 3 and RGD carrier 1
against CYP-mediated hepatic metabolism. Further studies
will be performed in the future to assess the in vivo capaci-
ty of the conjugate 3 to increase the therapeutic window
and limit the peripheral toxicity of PTX.
v
3
6
Metabolic Stability Assays19
The conjugate 3 and the RGD mimetic 1 were subjected
to human and rat liver microsomal stability tests to deter-
mine in vitro their stability towards cytochrome P (CYP)-
mediated metabolism. Test compounds were incubated
with rat and human microsomes at 37 °C in the presence of
the co-factor NADPH. The reaction was terminated by the
addition of methanol. Following centrifugation, the super-
natant was analysed on the LC–MS/MS. From this study
emerged that both the tumour-homing moiety 1 and the
cytotoxic construct 3 were considerably stable both in hu-
man and rat microsome suspensions (half-life 770 and 470
min, respectively). The fate of the conjugate 3 upon incuba-
tion with U87MG cells was also investigated. HPLC/MS
analysis and F NMR spectroscopy of both cell lysate and
culture medium confirmed that after 24 hours the conju-
gate 3 is still present unchanged, whereas a minor HPLC
peak corresponding to free PTX could also be detected. This
confirm the substantial stability of 3 and the slow release of
PTX in the presence of cells.
Funding Information
We thank the Development Trust, University of Aberdeen, for funding
a fellowship to M.P. and a studentship to A.A.
Unvieytrsi
of
A
b
e
dr
e
e
n
)(
Supporting Information
Supporting information for this article is available online at
https://doi.org/10.1055/s-0036-1590898.
S
u
p
p
ortioIgnfmr oaitn
S
u
p
p
o
nrtogI
i
f
rm oaitn
References and Notes
19
(
1) Svensen, N.; Walton, J. G.; Bradley, M. Trends Pharmacol. Sci.
2012, 33, 186.
(2) Arap, W.; Pasqualini, R.; Ruoslahti, E. Science 1998, 279, 377.
(3) Chen, K.; Chen, X. Theranostics 2011, 1, 189.
(
4) Colombo, R.; Mingozzi, M.; Belvisi, L.; Arosio, D.; Piarulli, U.;
Carenini, N.; Perego, P.; Zaffaroni, N.; De Cesare, M.; Castiglioni,
V.; Scanziani, E. J. Med. Chem. 2012, 55, 10460.
(
5) Liu, Y.; Bajjuri, K. M.; Liu, C.; Sinha, S. C. Mol. Pharm. 2012, 9,
7
Conclusions
168.
(
6) Piras, M.; Testa, A.; Fleming, I. N.; Dall’Angelo, S.; Andriu, A.;
Menta, S.; Mori, M.; Brown, G. D.; Forster, D.; Williams, K. J.;
Zanda, M. ChemMedChem. 2017, 12, 1142.
In this work, we described the rational design of a novel
tumour-targeted cytotoxic conjugate 3 incorporating the
potent antimitotic paclitaxel (PTX). The triazole-based RGD
mimetic 1 was designed as the tumour-homing motif of the
chemotherapeutic construct. The targeting vector 1 was
then functionalised with a chemical handle on the triazole
ring, which was identified as the most suitable site for deri-
vatisation of the RGD mimetic scaffold. Measurement of the
binding affinity of the functionalised mimetic 6 demon-
strated that the alkynyl chain on the triazole ring did not af-
fect the high affinity of 1 for α β integrin. The synthesis of
(
7) Dal Pozzo, A.; Ni, M. H.; Esposito, E.; Dallavalle, S.; Musso, L.;
Bargiotti, A.; Pisano, C.; Vesci, L.; Bucci, F.; Castorina, M.; Foderà,
R. Bioorg. Med. Chem. 2010, 18, 64.
(8) Fu, Y.; Li, S.; Zu, Y.; Yang, G.; Yang, Z.; Luo, M.; Jiang, S.; Wink,
M.; Efferth, T. Curr. Med. Chem. 2009, 16, 3966.
(9) Gao, Y.; Kuang, Y.; Guo, Z. F.; Guo, Z.; Krauss, I. J.; Xu, B. J. Am.
Chem. Soc. 2009, 131, 13576.
10) Chinchilla, R.; Nájera, C. Chem. Soc. Rev. 2011, 40, 5084.
11) Urgaonkar, S.; Verkade, J. G. J. Org. Chem. 2004, 69, 5752.
12) Synthesis of 6
(
(
(
v
3
To a solution of 7 (as TFA salt, 22 μmol, 20 mg, 1 equiv) in dry
DMF (500 μL) Et N (66 μmol mmol, 10 μL, 3 equiv), n-Bu NOAc
the conjugate 3 was performed by linking the cytotoxic
agent 2 and the RGD carrier 5 through a heterobifunctional
linker 4. The strong increase of steric hindrance in the con-
3
4
(66 μmol, 20 mg, 3 equiv), a catalytic amount of Pd(OAc) , and
2
alkyne 8 (0.22 mmol, 18 mg, 10 equiv) were added. The reaction
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–H

G
Synlett
M. Piras et al.
Letter
was stirred at 60 °C for 6 h. The mixture was cooled to r.t.,
diluted with water (4 mL), and then filtered using a 0.20 μm
syringe filter. The filtrate was purified by semipreparative RP-
J = 6.3 Hz, 1 H), 3.89 (t, J = 6.3 Hz, 1 H), 3.76–3.66 (m, 2 H), 3.60–
3.45 (m, 1 H), 3.45–3.35 (m, 1 H), 3.34–3.25 (m, 2 H), 3.09–2.97
(m, 2 H), 2.72–2.53 (m, 5 H), 2.48–2.34 (m, 4 H), 2.33–2.19 (m, 8
H), 2.10–1.97 (m, 4 H), 1.80 (s, 3 H), 1.75–1.60 (m, 2 H), 1.60–
HPLC. Analytical RP-HPLC conditions (solvent A = H O + 0.1%
2
19
TFA, solvent B = MeCN, gradient: from 30% B to 50% B in 15 min;
1.37 (m, 8 H), 1.03 (s, 3 H), 1.02 (s, 3 H). F NMR (376 MHz,
flow: 1 mL/min; t = 7 min). The product was lyophilized afford-
ing 12 mg of a yellow solid (70% yield, isolated as TFA salt): H
CD OD): δ = –63.06* (d, J = 12.7 Hz), –63.07 (d, J = 12.7 Hz), –77.19
R
3
1
13
(s), –110.47 to –110.61 (m). C NMR (101 MHz, CD OD): δ = 203.8,
3
NMR (400 MHz, CD OD): δ = 8.21–8.05 (m, 2 H), 7.73 (d, J = 6.6
Hz, 1 H), 7.51–7.37 (m, 1 H), 6.85 (s, 1 H), 6.80 (dd, J = 6.7, 1.4
Hz, 1 H), 4.62 (t, J = 6.6 Hz, 2 H), 4.29 (dd, J = 8.9, 4.9 Hz, 1 H),
172.4, 172.2, 170.8, 170.1, 169.9, 169.1, 169.0, 166.2, 161.4 (d,
J = 260.6 Hz), 160.3, 152.6, 149.3*, 148.6, 141.7, 141.6, 141.0,
138.1 (d, J = 3.7 Hz), 137.0, 134.2, 134.1, 133.6 (d, J = 9.9 Hz),
133.4, 133.2, 131.5, 129.9, 129.8, 128.7, 128.3, 128.1, 127.2,
127.2, 122.5 (d, J = 2.8 Hz), 121.9 (q, J = 272.6 Hz), 117.7 (d,
J = 22.1 Hz), 114.6, 105.1, 84.4, 80.8, 78.0, 77.6, 76.0, 75.3, 74.6,
73.2, 72.7, 71.4, 70.9, 70.0, 65.3, 65.2, 57.8, 55.3, 54.0, 46.4, 45.9,
43.1, 40.7, 38.8, 38.6, 36.0, 34.9, 29.8, 29.3, 28.6, 27.7, 27.5, 26.1,
26.0, 25.5, 25.5, 23.9, 21.8, 20.9, 20.6, 19.4, 16.7, 16.2, 13.6, 9.0.
3
3.82 (dd, J = 13.8, 4.9 Hz, 1 H), 3.74 (t, J = 6.1 Hz, 2 H), 3.52 (dd,
J = 13.8, 8.9 Hz, 1 H), 3.42 (t, J = 6.7 Hz, 2 H), 2.68 (t, J = 7.0 Hz, 2
H), 2.43 (s, 3 H), 2.41–2.29 (m, 2 H), 1.93–1.83 (m, 2 H).
1
9
F NMR (376 MHz, CD OD): δ = –63.15 (d, J = 12.7 Hz), –76.96 (s),
3
13
–
1
3
2
5
110.64 (q, J = 12.7 Hz). C NMR (101 MHz, CD OD): δ = 170.9,
3
61.4 (d, J = 261.5 Hz), 160.4, 156.7, 152.7, 141.4, 138.2 (d, J =
.7 Hz) , 134.2, 133.6 (d, J = 10.3 Hz), 126.3, 122.8, 121.9 (q, J =
71.4 Hz), 117.7 (d, J = 22.2 Hz), 114.5, 111.6, 106.2, 64.6, 60.0,
+
ESI-MS: m/z calcd for C82H91F4N10O22S [M + H] : 1675.59; found:
1675.7.
5.4, 45.9, 40.7, 38.6, 30.3, 27.5, 20.5, 15.7. ESI-MS: m/z calcd for
(15) Piras, M.; Fleming, I. N.; Harrison, W. T.; Zanda, M. Synlett 2012,
23, 2899.
+
C27H30F N O S [M + H] : 655.18; found. 655.2.
4
7
6
(
13) Synthesis of 13
(16) Dal Corso, A.; Caruso, M.; Belvisi, L.; Arosio, D.; Piarulli, U.;
Albanese, C.; Gasparri, F.; Marsiglio, A.; Sola, F.; Troiani, S.;
Valsasina, B. Chem. Eur. J. 2015, 21, 6921.
To a suspension of 6 (65 μmol, 50 mg, 1 equiv) and DMP (97.5
μmol, 41 mg, 1.5 equiv) in MeCN (1 mL) DMSO was added drop-
wise until the solution became clear. The mixture was stirred at
r.t. for 6 h and then filtered to remove a white precipitate
formed during the reaction. To the filtered solution 4 (195 μmol,
(17) Cell Cytotoxic Assay
Cells were seeded in 96-well plates at either 2,500 cells/well
(MCF7 cells) or 4,000 cells/well (U87MG cells) and left to grow
for 48 h in RPMI containing 10% foetal calf serum (FCS). Stock
solutions of compound 1, 3, and paclitaxel (PTX) were prepared
in DMSO and diluted to test concentrations in RPMI containing
10% FCS. Cells were exposed to a range of concentrations of test
compound for 6 h in triplicate wells. Medium containing the
compounds was then removed by aspiration and replaced with
fresh RPMI containing 10% FCS. Cells were left to grow for 72 h.
The number of cells in each well was then measured using an 3-
25 mg, 3 equiv) was added, and the mixture was stirred for 1 h
at r.t. The product was purified by semipreparative RP-HPLC.
Analytical RP-HPLC conditions (solvent A = H O + 0.1% TFA,
solvent B = MeCN, gradient: from 30% B to 50% B in 15 min;
flow: 1 mL/min; tR = 4.8 min). The product was lyophilized
affording 38 mg of a pale yellow solid (60% yield, isolated as bis-
TFA salt) containing a mixture of oxime E/Z isomers (3:2).
2
1
Minor isomer resonances are denoted by an asterisk*: H NMR
(
400 MHz, CD OD): δ = 8.08–8.00 (m, 2 H), 7.63 (d, J = 6.2 Hz, 1
(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide
3
H), 7.49 (t, J = 5.7 Hz, 0.6 H), 7.38–7.28 (m, 1 H), 6.80* (t, J = 5.1
Hz, 0.4 H), 6.69 (s, 1 H), 6.65 (d, J = 6.1 Hz, 1 H), 4.50 (t, J = 6.3
Hz, 2 H), 4.10–3.92 (m, 3 H), 3.78–3.62 (m, 1 H), 3.48–3.34 (m, 1
H), 3.30 (t, J = 6.1 Hz, 2 H), 2.91–2.79 (m, 2 H), 2.67 (t, J = 6.8 Hz,
(MTT) cell proliferation assay. Cells convert the MTT reagent
into formazan which can be detected at 562 nm in a microplate
reader. The concentration of each compound that inhibits cell
proliferation by 50% (IC50) was calculated using Graph PAD
PRISM. In addition, for each experiment a selectivity ratio was
calculated for compound 3 by dividing the IC50 for compound 3
with the paclitaxel IC50. Each IC50value and ratio is the average ±
s.e. of 4 independent experiments. A one way ANOVA was per-
formed using IBM SPSS Statistics version 24, to compare the
selectivity ratios obtained in the different cell lines.
2
H), 2.62–2.53 (m, 1 H), 2.43 (m, 1 H), 2.29 (s, 3 H), 2.27–2.17
1
9
(
m, 2 H), 1.72–1.57 (m, 4 H). F NMR (376 MHz, CD OD): δ =
3
–
63.10 (d, J = 12.7 Hz), –76.95, –110.39 to –110.82 (m). ESI-MS:
+
m/z calcd for C31H38F N O S [M + H] : 740.25; found: 740.2.
4
9
6
(
14) Synthesis of 3
To a solution of PTX-2′-succinate-NHS ester 2 (24 μmol, 25 mg,
equiv) in DMF (0.5 mL) at r.t., DIPEA (72 μmol, 13 μL, 3 equiv)
1
(18) Lamidi, O. F.; Sani, M.; Lazzari, P.; Zanda, M.; Fleming, I. N.
J. Cancer Res. Clin. Oncol. 2015, 141, 1575.
(19) In Vitro Stability Assays
and 13 (24 μmol, 23 mg, 1 equiv) were added. The reaction
mixture was stirred for 5 h at r.t., then filtered and purified by
semipreparative RP-HPLC. Analytical RP-HPLC conditions (sol-
All assays conducted by Cyprotex Ltd (Macclesfield, UK).
Human and Rat Liver Microsomal Stability Assays
vent A = H O + 0.1% TFA, solvent B = MeCN, gradient: from 50% B
2
to 1000% B in 15 min, flow: 1 mL/min; t = 6.9 min). The product
was lyophilized affording 23 mg of a white solid (54% yield, iso-
Briefly: Test compound (3 μM) was incubated with pooled liver
microsomes. Test compound was incubated at 5 time points
over the course of a 45 min experiment, and the test compound
was analysed by LC–MS/MS.
R
lated as TFA salt) containing a mixture of oxime E/Z isomers
1
(
3:2). Minor isomer resonances are denoted by an asterisk*: H
NMR (400 MHz, CD OD): δ = 9.03 (d, J = 8.7 Hz, 1 H), 8.29–8.20
Experimental Procedure
3
(
m, 1 H), 8.05–7.98 (m, 4 H), 7.76–7.70 (m, 2 H), 7.62–7.55 (m, 2
Pooled human or mouse liver microsomes were purchased from
a reputable commercial supplier. Microsomes were stored at
–80 °C prior to use. Microsomes (final protein concentration 0.5
mg/mL), 0.1 M phosphate buffer pH 7.4, and test compound
(final substrate concentration 3 μM; final DMSO concentration
0.25%) were pre-incubated at 37 °C prior to the addition of
H), 7.55–7.26 (m, 11 H), 7.15 (t, J = 7.1 Hz, 1 H), 6.78* (t, J = 5.1
Hz, 0.4 H), 6.73 (s, 1 H), 6.67 (dd, J = 6.6, 1.3 Hz, 1 H), 6.33 (s, 1
H), 5.94 (t, J = 9.0 Hz, 1 H), 5.68 (d, J = 6.6 Hz, 1 H), 5.52 (d, J = 7.2
Hz, 1 H), 5.33 (dd, J = 6.7, 1.2 Hz, 1 H), 4.88 (d, J = 8.1 Hz, 1 H),
4
.48 (t, J = 6.0 Hz, 2 H), 4.31–4.13 (m, 2 H), 4.07 (s, 2 H), 3.96 (t,
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–H

H
Synlett
M. Piras et al.
Letter
NADPH (final concentration 1 mM) to initiate the reaction. The
final incubation volume was 50 μL. A minus cofactor control
incubation was included for each compound tested where 0.1 M
phosphate buffer pH 7.4 was added instead of NADPH (minus
NADPH). Two control compounds were included with each spe-
cies. All incubations were performed singularly for each test
compound. Each compound was incubated for 0, 5, 15, 30, and
bate to 60 μL MeOH at the appropriate time points. The termi-
nation plates were centrifuged at 2,500 rpm for 20 min at 4 °C
to precipitate the protein.
Quantitative Analysis
Following protein precipitation, the sample supernatants were
combined in cassettes of up to 4 compounds, internal standard
was added and samples analysed using Cyprotex generic LC–
MS/MS conditions.
45 min. The control (minus NADPH) was incubated for 45 min
only. The reactions were stopped by transferring 20 μL of incu-
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–H