Co-liposomes comprising a lipidated multivalent RGD-peptide and a cationic gemini cholesterol induce selective gene transfection in αvβ3 and αvβ5 integrin receptor-rich cancer cells

Article abstract of DOI:10.1039/c4tb00701h

The αvβ3 and αvβ5 integrins, transmembrane glycoprotein receptors, are over-expressed in numerous tumors and in endothelial cells that constitute tumor blood vessels. As this protein selectively binds to the Arg-Gly-Asp (RGD) sequence containing peptides, it is an attractive way to target tumors. Herein we have developed novel formulations for integrin mediated selective gene delivery. These formulations are composed of a novel palmitoylated tetrameric RGD containing scaffold (named RAFT-RGD), cationic gemini cholesterol (GL5) and a natural helper lipid 1,2-dioleoyl-l-α- glycero-3-phosphatidylethanolamine (DOPE). We have optimized a co-liposomal formulation to introduce the multivalent RGD-containing macromolecule in GL5: DOPE (GL5D) mixture to produce GL5D-RGD. We have unambiguously shown the selectivity of these formulations towards cancer cells that over express αvβ3 and αvβ5 integrins. Two reporter plasmids, pEGFP-C3 and PGL-3, were employed for the transfection experiments and it was shown that GL5D-RGD liposomes increased exclusively the transfection in αvβ3 and αvβ5-overexpressing HeLa cells. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4tb00701h

Journal of
Materials Chemistry B
View Article Online
View Journal | View Issue
PAPER
Co-liposomes comprising a lipidated multivalent
RGD-peptide and a cationic gemini cholesterol
induce selective gene transfection in avb3 and
avb5 integrin receptor-rich cancer cells†
Cite this: J. Mater. Chem. B, 2014, 2,
758
5
Santosh K. Misra, Paturu Kondaiah, Santanu Bhattacharya,* Didier Boturyn*de
a
b
ac
ef
and Pascal Dumy*
The avb3 and avb5 integrins, transmembrane glycoprotein receptors, are over-expressed in numerous
tumors and in endothelial cells that constitute tumor blood vessels. As this protein selectively binds to
the Arg–Gly–Asp (RGD) sequence containing peptides, it is an attractive way to target tumors. Herein we
have developed novel formulations for integrin mediated selective gene delivery. These formulations are
composed of a novel palmitoylated tetrameric RGD containing scaffold (named RAFT-RGD), cationic
gemini cholesterol (GL5) and a natural helper lipid 1,2-dioleoyl-L-a-glycero-3-phosphatidylethanolamine
(DOPE). We have optimized a co-liposomal formulation to introduce the multivalent RGD-containing
macromolecule in GL5: DOPE (GL5D) mixture to produce GL5D–RGD. We have unambiguously shown
the selectivity of these formulations towards cancer cells that over express avb3 and avb5 integrins. Two
reporter plasmids, pEGFP-C3 and PGL-3, were employed for the transfection experiments and it was
shown that GL5D–RGD liposomes increased exclusively the transfection in avb3 and avb5-
overexpressing HeLa cells.
Received 3rd May 2014
Accepted 13th June 2014
DOI: 10.1039/c4tb00701h
www.rsc.org/MaterialsB
2
such selective transfection. Integrins belong to a major class of
ab heterodimeric transmembrane glycoprotein receptors. It is
Introduction
3
–5
Selective cell targeting has been considered as a major objective known that avb3 and avb5 integrins are expressed on various
8
1
6
7
in gene therapy to combat different diseases. The gene carrier cell types such as endothelial cells, osteoclasts, macrophages,
9
10
species should have specic ligands which can help it to platelets, and melanomas. They play a signicant role in
recognize the victim cell, so that it can deliver the cargo to the angiogenesis, vascular intimal thickening and proliferation of
11
desired cells only. The concept of cell targeting introduces an malignant tumors. Some important cancer cell lines highly
effort to recognize natural targets on tumor cell surfaces. expressing avb3 integrin are A375 (melanoma), M21 (mela-
Integrin receptors have been recognized as natural targets for noma), SK-MEL-28 (melanoma), and M19-MEL (melanoma)
while HeLa cells (carcinoma) over-expresses avb3 as well as
12
avb5. However, some cell lines do not express avb3 and avb5
Department of Organic Chemistry, Indian Institute of Science, Bangalore 560 012, integrins. These include HEK293, 293T and NIH3T3 cells,
a
13
14
15
India. E-mail: sb@orgchem.iisc.ernet.in; Fax: +91-80-2360-0529; Tel: +91-80-2293-
which express another integrin receptor as a5b1 in abundance.
Cells expressing avb3 or avb5 integrins bind with a broad
range of ligands such as bronectin and vitronectin mainly
through the recognition of the ubiquitous triad sequence RGD
2
664
b
Department of Molecular Reproduction, Development and Genetics, Indian Institute
of Science, Bangalore 560 012, India
c
JNCASR, Bangalore 560 064, India
16
d
e
(Arg–Gly–Asp). The latter has served as a basis for the devel-
opment and discovery of high and selective integrin–peptide
Univ Grenoble Alpes, D ´e partement de Chimie Mol ´e culaire, 38400 Grenoble, France
CNRS, UMR 5250, FR 2607, France
17
f
IBMM-UMR-5247 Ecole Nationale Sup ´e rieure de Chimie de Montpellier, 34296 ligands such as c[–RGDfV–] that was exploited for a wide range
18–23
Montpellier cedex 5, France
of applications.
Additionally, the principle of multivalency
†
Electronic supplementary information (ESI) available: Materials and
equipment, general procedures for peptide syntheses, synthesis of lipopeptide
, the RP-HPLC prole of 3, 1H-NMR of 3, lipodecapeptide 5, the RP-HPLC
prole of 6, ESI-MS analysis of compound 6, palmitoyl-RAFT-RGD (1), the
has then been recognized as an important approach for the
24
design of synthetic ligands. In this context, enhancements of
1
biological activity are obtained from multivalent RGD peptide
4
ligands used to target cell surface receptors such as the avb3
RP-HPLC prole of 1, ESI-MS analysis of compound 1, synthesis of gemini
lipids (GLs), characterization of gemini lipids (GLs), transfection efficiencies of integrin.
gemini lipids (GLs), stability of lipoplexes in FBS, untransfected lipoplex and
surface binding of lipoplexes. See DOI: 10.1039/c4tb00701h
25–27
For instance, we have shown that tetrameric RGD-
5758 | J. Mater. Chem. B, 2014, 2, 5758–5767
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
28
containing scaffolds exhibit advantageous biological proper- Synthesis of palmitoyl-RAFT-RGD4
29–31
32,33
ties for imaging
and for targeted drug delivery.
The synthesis of lipopeptide 1 was produced via solid-phase
peptide syntheses (SPPS) according to the methods already
developed by our group (Scheme 1). Stable oxime bonds were
chosen to append aldehyde-bearing RGD motifs to a cyclo-
decapeptide scaffold. Building blocks 2 and 3 including
respectively the protected amino-oxy function and the palmitoyl
group were directly introduced during the SPPS. This strategy
decreases the number of steps and the combination of pro-
tecting groups required so far for the production of lipopeptide
Two strategies may be developed for achieving cell selective
targeting using carrier liposomes. Firstly the covalent conjuga-
34
tion of RGD based molecules to the lipid carrier and secondly
the non-covalent loading of RGD based molecules to lipid
35
carriers by physical mixing in optimized molar ratios. Very
recently, we prepared the RGD-containing lipopeptide to
36
construct uid supported lipid bilayers (SLB). We then studied
RGD-mediated cell adhesion to an SLB surface providing the
average ligand spacing to trigger cell adhesion and spreading
on a uid substrate. Herein we present the results of non-
4. The following head-to-tail cyclization on resin and the
subsequent deprotection under acidic conditions furnished the
lipopeptide 6. The graing of glyoxylyl aldehyde containing-
RGD targeting elements 7 provided the expected palmitoyl-
37
covalent RAFT-RGD inclusion in cationic liposomes for cancer
cell targeting in vitro (Fig. 1).
4
RAFT-RGD 1 in satisfying 15% overall yield.
Results
Optimization of the GL : DOPE ratio
The concept of cell targeting introduces an effort to recognize
natural targets viz. integrin receptors for selective transfection.
The multivalency has been established as an important
approach for the design of ligands to increase the targeting
efficiency. We applied the non-covalent palmitoylated RAFT-
RGD inclusion strategy to develop a co-liposomal formulation of
GL, DOPE and a palmitoylated RAFT-RGD macromolecule 1 to
transfect pEGFP-C3 and PGL-3 in avb3 and avb5 positive HeLa
and negative HEK293 and NIH3T3 cells. This methodology of
using lipopeptide 1 was found highly benecial for introducing
high selectivity in gene therapy.
Gemini lipid GL5 was screened for the optimization of the lip-
id : DOPE ratio from 1 : 0 to 1 : 5. Here the number of trans-
fected cells (% GFP cells) was higher for the GL : DOPE at ratios
1
: 3 and 1 : 5 compared to that in 1 : 4. However, the amount of
DNA transfected to each cell (MFI) was very high when a molar
ratio of 1 : 4 was used. A gemini lipid to DNA ratio of 1 : 4 was
used as the optimal ratio for further studies (Fig. S1†).
Optimization of palmitoyl-RAFT-RGD
4
inclusion in liposomes
1 into cationic
Incorporation of the palmitoyl-RAFT-RGD
4
liposomes was achieved either by the preparation of co-lipo-
somes or via a post-injection method. In the rst method, an
optimized GL5D was mixed with the lipidated RGD in molar
ratios of 2000 : 1 and 1000 : 1, respectively. Liposomes were
prepared from these mixtures as mentioned earlier and were
termed as co-liposomal formulations, i.e., Co2000 and Co1000,
respectively. A second method was used by injecting an etha-
nolic solution of the lipidated RGD in pre-prepared GLD
suspensions. These formulations have been termed as
Scheme 1 Synthesis of 1: (i) 1 equiv. Pd(PPh
Fig. 1 Molecular structures of the gemini lipid (A) GL5 and (B) a 2 equiv. PyAOP, DIPEA (pH 8.0); TFA, TIS, H O (90 : 5 : 5); (iii) 4.4 equiv.
palmitoyl-RAFT-RGD (1) ligand used in the present investigation. 7, acetic acid, H O, CH CN (50 : 25 : 25).
3 4 3
) , 100 equiv. PhSiH ; (ii)
2
4
2
3
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 5758–5767 | 5759

View Article Online
Journal of Materials Chemistry B
Paper
post-injection formulations, i.e., Po2000 and Po1000, respec-
tively. FACS analysis of the pEGFP-C3 reporter gene transfection
showed a distinct advantage of the co-liposomal preparation
over the one prepared by the post-injection protocol. Liposome
GL5D–RGD prepared by a co-liposomal method showed at least
50% increase in the transfection efficiency compared to the
corresponding gemini lipid–DOPE–RGD formulations prepared
by the post injection method (Fig. S2A†).
Optimization of the GLD : RAFT-RGD molar ratio
Molar ratios of GL5D : RGD were varied from 1 : 0 to 50 : 1 for
the pEGFP-C3 transfection in HeLa cells. A molar ratio of 100 : 1
was found to be the best ratio for achieving high and cell-
specic transfection. Beyond this ratio, the transfection effi-
ciency decreased considerably (Fig. S2B†). This is probably
because of the alteration in the liposomal organization and/or
aggregation leading to the formation of signicantly larger
aggregates of the GL5D–RGD formulations at a GL5 : RGD
molar ratio of 50 : 1 (Fig. S2C†).
Palmitoyl-RAFT-RGD mediated cell targeting
4
Optimized GLD and GLD–RGD formulations were used for
transfecting three different cell lines, HeLa, HEK293 and
NIH3T3 using 0.8 mg pEGFP-C3 per well using the GL/DNA
charge ratio (N/P) 0.5 (Fig. 2). In HeLa cells, GL5D induced
ꢀ
50% GFP cell transfection with a MFI of ꢀ25 only while GL5D–
RGD furnished ꢀ75% GFP cells with a MFI of ꢀ75 at the N/P
charge ratio of 0.5. In HEK293 cells, GL5D–RGD gave ꢀ40% GFP
cells with a MFI of ꢀ20 at the N/P charge ratio of 0.5. Lip-
ofectamine2000® in the presence of 10% FBS (Lipo2000*) was
used as a positive control in all the cell lines. GL5D–RGD was
found to be better than Lipo2000* at least in HeLa and
NIH3T3 cells.
Fig. 2 Transfection efficiency of the GL5D gemini lipid with and
We further measured the luciferase activity with respect to
4
without the incorporation of palmitoyl-RAFT-RGD (100 : 1) and
N/P charge ratios and the relevant data are shown as RLU per Lipo2000* in the presence of serum (ꢂFBS + FBS) at a N/P charge ratio
mg of protein. In HeLa cells, GL5D–RGD gave around 5-fold 0.5. The concentration of the plasmid DNA (pEGFP-C3) ¼ 0.8 mg per
well was used for the experiment, which was performed on HeLa
higher transfection efficiency compared to that of GL5D at the
(Fig. 3A), NIH3T3 (Fig. 3B) and HEK293 (Fig. 3C) cells. The statistical
N/P of 0.5. On the other hand in NIH3T3 cells, the overall
transfection was found to be strikingly low and GL5D–RGD
showed no improvements in poorer transfection efficiency
compared to GL5D. Thus GL5D–RGD was found to be a signif-
icantly better transfecting agent in HeLa cells compared to
GL5D with a high biostatistical signicance with p < 0.001
difference in MFI from GL5D–RGD and GL5D is labelled as *** indi-
cating P < 0.001.
ꢀ
1, GL5D–RGD/DNA ꢀ60% release of DNA was recorded while
GL5D/DNA could release only 40% even at an SDS/GL of 5.
DNA protection ability of the cationic liposomes was also
(Fig. 3).
ꢁ
examined by incubating DNase I (1 unit) for 30 min at 37 C.
Whereas 1 unit DNase I could digest/degrade whole DNA under
this condition, all lipoplexes at N/P ratio 3 could protect DNA
Gel electrophoresis
DNA complexation efficiency of the GLD and GLD–RGD completely from any degradation (Fig. 6).
formulations were examined at different N/P charge ratios Stability of the lipoplexes in the presence of FBS was
0
1
.125, 0.25, 0.75 and 1. The GL5D and GL5D–RGD gave around measured both using a xed N/P ratio while varying the % of
00% complexation at an N/P charge ratio of 0.5. Inclusion of FBS (v/v) (Fig. S3A†) and by keeping the % of FBS xed while
palmitoyl-RAFT-RGD4 (1) did not however affect the DNA altering the N/P charge ratio (Fig. S3B†). It was found that even
binding efficiency of the GLDs (Fig. 4). Miceller SDS mediated 50% of FBS could not dissociate the lipoplexes at N/P charge
de-complexation of DNA from the lipoplexes was analyzed using ratios for the liposomes with or without RGD whereas in the
lipoplexes of an N/P ratio 3 followed by incubation with SDS at presence of 70% FBS, an N/P charge ratio of 2 was enough to
an SDS/GL ratio varying from 1 to 4 (Fig. 5). At a SDS/GL ratio of protect the DNA from being released.
5760 | J. Mater. Chem. B, 2014, 2, 5758–5767
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
Fig. 5 SDS mediated release of the DNA from GLD and GLD–RGD
based lipoplexes. The experiment was performed using 0.2 mg plasmid
DNA per well. Lipoplexes were prepared at an N/P ratio of 3. Numbers
above the lanes indicate SDS/GL charge ratios while DNA and D/N
indicates only DNA and GL/DNA charge ratios respectively.
Fig. 3 Transfection efficiency of the GL5D gemini lipid with and
without RGD-RAFT (100 : 1) in the presence of serum (ꢂFBS + FBS) at
an N/P (lipid/DNA) charge ratio of 0.5. Plasmid DNA (PGL-3) [0.8 mg per
well] was used for the experiment. Data are expressed as relative light
unit per mg of protein as obtained from the single luciferase assay. The
experiment was performed on HeLa (A) and NIH3T3 (B) cells. The
statistical difference in transfection efficiency of GL5D–RGD from the
control GL5D is labelled as *** indicating P < 0.001.
Fig. 6 DNA protection ability of various liposomes against DNase
I. The experiment was performed using 0.2 mg plasmid DNA per well.
Lipoplexes with and without RAFT-RGD (N/P ratio 2) were treated with
ꢁ
1
unit of DNase I possessing 1ꢃ DNase buffer for 30 min at 37 C. The +
and ꢂ sign indicate the presence and absence of DNase I in formu-
lations respectively. Numbers below the lanes indicates the N/P charge
ratio in the formulation while DNA indicates only DNA in the lane.
HeLa cells treated with GL5D–RGD : pyrene, selectively
expressed ꢀ95% pyrene positive cells compared to ꢀ70% in the
case of GL5D : pyrene. In NIH3T3, GL5D–RGD : pyrene as well
as GL5D : pyrene both expressed ꢀ65% pyrene positive cells
Fig. 4 DNA binding efficiency of GLD and GLD–RGD liposomes. The
experiment was performed using 0.2 mg pEGFP-C3 plasmid DNA per
well. Numbers above the lanes indicate an N/P charge ratio used in the without any selectivity. Presumably the presence of the RAFT-
formulation.
RGD in GL5D–RGD : pyrene assisted selective and highly effi-
cient surface binding to cells containing the integrin receptors,
avb3 and avb5 (HeLa cells). In contrast, in NIH3T3 cells (which
are devoid of avb3 and avb5) the RAFT-RGD could not inuence
The effect of RGD on the lipoplex delivery efficiency
Overall the transfection efficiency of liposomes was found to be
related to the DNA delivery efficiency. The UV absorbance of the
undelivered DNA was found to be higher in the case of HeLa
cells when they were transfected with GL5D compared to GL5D–
RGD. In NIH3T3 and HEK293, however, it was found to be the
opposite (Fig. S4†).
the surface binding percentage to any signicant extent.
Confocal uorescence microscopy
Cells were observed under a confocal microscope to visualize
the expression of green uorescent protein (GFP) in the cell
cytoplasm. Green uorescence was clearly visible in the cell
cytoplasm, which was obtained via cell transfection using the
Cell surface binding
A ow cytometry experiment was performed on both the HeLa cationic gemini cholesterol formulations with or without the
and NIH3T3 cells to nd out the selective cell surface binding RAFT-RGD in both HeLa and NIH3T3 cell lines (Fig. 8) similar to
efficiency of the lipoplexes with the palmitoyl-RAFT-RGD
Fig. 7).
4
the cells transfected with the commercially available agents
Lipofectamine2000® and Effectene.
(
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 5758–5767 | 5761

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 8 Confocal images showing GFP expression of pEGFP-C3 gene
transfected to HeLa (A–F) and NIH3T3 (G–L) cells. Liposomal formu-
lations were used at an N/P charge ratio of 0.5 using 0.8 mg DNA per
well. (A) HeLa; (B) +Effectene; (C) +GL5D; (D) +GL5D–RGD; (E) +only
pEGFP-C3; (F) +Lipo2000*; (G) NIH3T3; (H) +Effectene; (I) +GL5D; (J)
+GL5D–RGD; (K) +only pEGFP-C3; (L) +Lipo2000*.
Fig. 7 Surface binding efficiency of GLD and GLD–RGD formulations
in HeLa and NIH3T3 cells. Bar diagrams shows high surface binding of
formulations with RAFT-RGD in HeLa cells compared to NIH3T3 cells.
Cytotoxicity
The MTT assays of optimized GLDs and GLD–RGDs, with and
without DNA were obtained in HeLa, NIH3T3 and HEK293 cell
lines (Fig. 9). All formulations without DNA were screened for
concentrations 0.75 mM to 6 mM while formulations with DNA
were screened for N/P charge ratios varying from 0.125 to 1. In
the case of the nal optimized formulations, the % cell viability
was found to be ꢀ80% which dropped to ꢀ60% at higher
concentrations and at higher N/P charge ratios, specically in
the case of other non-optimized formulations across each cell
line. Importantly an inclusion of addition of RAFT-RGD in the
liposomal formulation did not reduce it at all, in any cells, e.g.,
HeLa, NIH3T3 and HEK293 examined.
Discussion
There are many barriers that impede effective transfection
using non-viral gene vectors. The stability of vector/DNA
complexes in the bloodstream and specic cellular uptake by
the target tissues/organs are some of the main obstacles for
in vivo gene delivery. Cationic lipids are well known candidates
38,39
for gene delivery.
Liposomal formulations of the cationic
Fig. 9 Cytotoxicity assay (MTT) performed on HeLa (A and B), NIH3T3
C and D) and HEK293 (E and F) cells. The experiment was performed
by using 0.2 mg DNA per well in 96 well plates. The experiment was
cholesterol derivatives are known to achieve DNA transfection
in cells with greater efficiency than a number of commercial
transfection reagents. These cationic cholesterol based formu-
lations are more efficient than the cationic lipid suspensions
(
performed in triplicate.
5762 | J. Mater. Chem. B, 2014, 2, 5758–5767
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
based on the pseudoglyceryl skeleton in terms of gene trans- formulations were signicantly better transfectants (Fig. 2A)
40,41
fection.
Moreover gemini lipids have been shown to be more compared to the GL5D (P < 0.001) whereas in NIH3T3 and
efficient than their monomeric counterparts in terms of their HEK293 cells, inclusion of RGD did not lead to an enhancement
37
ability to transfect eukaryotic cells. Earlier we have reported in the transfection efficiency (Fig. 2B and C). Similar results
serum compatible cholesterol based systems as efficient trans- were obtained in transfection of PGL-3 plasmids in HeLa and
42,43
fecting agents.
But to establish a precisely targeted gene NIH3T3 cells as well. In this instance RGD inclusion in GLDs
delivery, there was a need to design and develop newer formu- increased the transfection efficiency by ꢀ5 fold in HeLa cells
lations with an efficient cell targeting moiety also capable of with P < 0.001 in bio-statistical analysis (Fig. 3A). In contrast in
efficient gene transfection. Accordingly we have utilized the NIH3T3 it did not change to any considerable extent (Fig. 3B).
cationic cholesterol derivatives as cytofectins for our studies.
An introduction of RAFT-RGD in the GL5D formulation at an
The spacer lengths between the two monomeric units in such N/P charge ratio of 0.5 causes signicant improvement in HeLa
gemini lipids play an important role in determining the trans- cell selectivity as revealed from the FACS analysis. Similarly the
38,41
fection efficacies of the cationic lipids.
luciferase assay shown in Fig. 3 indicates that an introduction
Recognition of specic integrin receptors on cell surfaces by of RAFT-RGD in GL5D improves the transfection efficiency at N/
the particular tripeptidic sequence (RGD), allows the binding of P charge ratios of 0.5 in HeLa cells only. It may be further noted
RGD loaded liposomal formulations with cell surfaces having here that the luciferase assay was performed using the PGL-3
abundance of compatible integrin receptors. One can employ plasmid while plasmid pEGFP-C3 was used for the FACS
different ways of using RGD templates in gene delivery formu- analysis.
lations for achieving integrin receptor specic targeting; rstly
It is possible that either an improved DNA binding, a better
via a covalent modication of the cationic lipid molecules with DNA release, an improved serum tolerance or better DNase I
34
RGD based tripeptides or secondly via a non-covalent incor- sensitivity of the formulations in the presence of RGD might
poration of an optimized amount of RGD based lipid soluble have resulted in a high transfection efficiency with specic
molecules. To avoid the complexity of molecular geometry and targeting. But gel electrophoresis experiments performed on
in order to achieve convenient variation of ratios of RGD to GLs, formulations with and without RAFT-RGD indicated that the
we decided to proceed with the non-covalent strategy.
presence of RGD did not enhance the binding of DNA (Fig. 4).
For the present work, cationic gemini cholesterol (GL5) was This also did not enhance the SDS mediated release of DNA
used with DOPE at dened proportions to form liposome GL5D, (Fig. 5) or even afford better DNA protection against DNase I
which also contained specied amount of palmitoylated (Fig. 6). Even the stability of lipoplexes in the presence of serum
multivalent RGD to prepare GL5D–RGD via co-liposomal and was not affected by the presence of RGD (Fig. S3†). These results
post-injection methods. During the co-liposome formation, clearly suggest that there might be another factor, which
presumably the long fatty acid chain in the palmitoylated governs the enhancement of transfection efficiency by the co-
multivalent RGD ligand has full freedom to be anchored into liposomes loaded with the RGD. The HeLa cells possess high
the hydrophobic region of the vesicular aggregates. Also in the abundance of avb3 and avb5 transmembrane integrin receptors
case of the palmitoyl-RAFT-RGD , the presence of the sterically while the HEK293 and NIH3T3 cells are devoid of them. As
4
demanding cyclic peptide residues might ensure preferential expected, palmitoyl-RAFT-RGD4 present in the co-liposomal
partitioning of these molecules on the outer periphery of the co- formulations improves the transfection efficiency in HeLa cells.
liposomes. It is certainly the reason for the increase of the
In support of the RGD-mediated targeting of the avb3 and
transfection efficiency by using the co-liposomal methodology. avb5 integrins, the undelivered lipoplex measurement assay
Co-liposomal formulation was used for the optimization of was performed. This indicates that in HeLa cells, GLD–RGD
the GLD : RGD molar ratio for highest transfection with best could deliver more lipoplexes in the cytoplasm compared to that
selectivity. It was found that the 100 : 1 molar ratio was the best of the GLD alone (Fig. S4A†). In contrast, in NIH3T3 and
transfecting formulation. Incorporation of RGD templates in HEK293 cell lines, the observed pattern was just the opposite.
molar ratios more or less than 100 : 1 led to decreases in the Thus with GLD–RGD, the higher amount of lipoplexes was le
transfection efficiency (Fig. 2B). Probably, at lower ratios undelivered. Similarly, GLD–RGD : pyrene was able to tag
(2000 : 1 or 1000 : 1), the number of RGD tags onto the co- higher % of cells compared to the GLD : pyrene while NIH3T3
liposomes was insufficient to achieve maximum while at a showed no enhancement in surface binding (Fig. 7), indicating
higher ratio (50 : 1) too many RGD tags might have altered the a selective targeting induced by the RGD peptide (Fig. S5†).
surface properties of the co-liposomes (GLD) in such a way that
Confocal uorescence microscopy was performed to nd out
it reduced the transfection efficiency by probably due to the the GFP expression of the pEGFP-C3 plasmid in HeLa and
instability and/or aggregation of liposomes with a signicantly NIH3T3 cells (Fig. 8). It was found that each formulation, with
larger hydrodynamic diameter for the GL5D–RGD formulation or without RGD along with commercially available Effectene
at the GL5 : RGD molar ratio 50 : 1 (Fig. S2C†).
and Lipofectamine2000® (Lipo2000*) (two positive controls),
The overall optimized GL5D–RGD liposomes were then used were able to transfect the cells. Confocal uorescence micros-
for transfection of HeLa, HEK293 and NIH3T3 cells using copy data shown here indicate a qualitative expression of GFP.
plasmid pEGFP-C3 and PGL-3 vectors. In all the experiments, These data cannot be used here for any quantitative comparison
GL5D, devoid of RGD, was used as negative controls. Trans- of different formulations including positive controls, Effectene
fection of pEGFP-C3 revealed that in HeLa cells GL5D–RGD and Lipofectamine2000®. The results obtained from FACS and
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 5758–5767 | 5763

View Article Online
Journal of Materials Chemistry B
Paper
luciferase assays are however, quantitative for this purpose.
A trans-sonic T 460H Elma bath sonicator (40 W) was used
Finally, the cytotoxicity of each formulation was measured for sonic dispersal of lipid suspension. Blood serum used in cell
using the MTT assay protocol. It conrmed the considerably low culture experiments was obtained from GIBCO. Cell culture
toxicity of the formulations with the palmitoylated-RAFT-RGD4 asks and multi-well plates were obtained from NUNC. Dul-
(Fig. 9). Thus, the co-liposomal methodology of RGD tagging on becco's modied phosphate buffer saline (DPBS) and Dulbec-
cationic liposomes has been demonstrated to be an efficient co's modied eagles medium (DMEM) was obtained from
and convenient way of cell targeting in the gene delivery SIGMA. Public domain soware, WinMDI, was used for the
experiments.
FACS data analysis.
Preparation of lipopeptides
Conclusions
Palmitoylated multivalent RGD-peptide 1 was prepared as
described in the literature (see the ESI†).
A new gene delivery formulation has been developed which is
composed of a cationic gemini cholesterol (GL5), helper lipid
(DOPE) and a palmitoyl-RAFT-RGD4 template to imbibe the
Preparation of liposomes
selectivity in liposomes towards cells therefore endowed with
abundance of avb3 and/or avb5 integrins. Stable formulations
were developed by a non-covalent co-liposomal mixing of GL,
DOPE and palmitoylated-RGD-RAFT in different optimized
molar ratios. The expression of pEGFP-C3 and PGL-3 in avb3
and avb5 positive cells (HeLa) was found to be considerably
high for the GL5D–RGD formulations compared to other avb3
and avb5 negative cells (HEK293 and NIH3T3). Similar gel
electrophoretic patterns of liposomes with (GL5D–RGD) or
without (GL5D) RGD conrmed that the RGD played no role in
enhancing the DNA binding, anionic micelle induced DNA
release, serum stability or even DNA protection against DNase
I to improve the transfection. Measurement of the extent of
undelivered lipoplexes and % GLD–RGD : pyrene labeled posi-
tive cells support the role of RGD in avb3 and avb5 mediated
cell selectivity by the present gemini lipid based transfectants.
The formulations were good transfecting agents in comparison
to commercially available Effectene or Lipofectamine2000®
3
7
Liposomes were prepared using a literature protocol. In brief,
each of the gemini lipid was mixed with DOPE in an appropriate
molar ratio as solution in chloroform. The organic solvent was
evaporated under a rotary evaporator and nally dried under
vacuum. The vacuum dried lms of lipid mixtures were sub-
jected to hydration with addition of an appropriate amount of
buffer, repeated freeze-thaw cycles followed by sonication at
0 C for 15 min. Each liposomal suspension was found to be
transparent and very stable while stored at 4 C under sterile
conditions. Formulations were re-sonicated at room tempera-
ture for 5 min before use for each experiment.
ꢁ
7
ꢁ
Plasmid DNA
Plasmids pEGFP-C3 and PGL-3, which encode for an enhanced
green uorescence protein (GFP) under a CMV promoter and
rey luciferase protein under a SV40 promoter respectively,
were amplied in Escherichia coli (DH5a) and puried using a
Qiagen Midi Prep plasmid purication protocol (Qiagen, Ger-
many). The purity of the plasmid was checked by electropho-
resis on 1.0% agarose gel. The concentration of DNA was
estimated spectroscopically by measuring the absorption at

(Lipo2000*). As inclusion of RGD did not increase the cytotox-
icity of the liposomes and corresponding lipoplexes, it may be
highly benecial to use this formulation for in vivo gene delivery
processes. We are currently developing this strategy and in vivo
evaluation will be reported in due course.
2
60 nm and conrmed by gel electrophoresis. Plasmid pre-
parations showing a value of OD260/OD280 > 1.8 were used.
Experimental
Materials and methods
Cell culture
Cells (HeLa, HEK293 and NIH3T3) were cultured in Dulbecco's
modied Eagle's medium (DMEM, Sigma) supplemented with
Gemini lipid (GL), GL5 has been synthesized (ESI†) as described
37
elsewhere. Palmitoyl-RAFT-RGD
4
was prepared through a
1
0% fetal bovine serum (FBS) in T25 culture asks (Nunc,
combination of solid and solution-phase syntheses according to
ꢁ
36
Denmark) at 37 C in a humidied atmosphere containing 5%
the methods previously described (ESI†). All reagents,
solvents, and chemicals used in this study were of the highest
purity available. The solvents were dried prior to use. Column
chromatography was performed using 60–120 mesh silica gel.
NMR spectra were recorded using a Jeol JNM-300 (300 MHz
CO
trypsin (EDTA 0.02%, dextrose 0.05%, and trypsin 0.1%) in PBS
pH 7.4).
2
. Cells were regularly passaged by trypsinization with 0.1%
(
1
13
Transfection procedure
for H and 75 Hz for C) spectrometer. The chemical shis (d)
are reported in ppm downeld from the internal standard, TMS, Transfection experiments were performed using a literature
1
13
36
for H-NMR and C-NMR. Mass spectra were recorded on a protocol. In brief, 24 h before transfection, 24-well plates were
Kratos PCKompact SEQ V1.2.2 MALDI-TOF spectrometer, a seeded with 60 000 cells per well in antibiotic-free media.
MicroMass
HP1100LCMSD) spectrometers. Infra-red (IR) spectra were serum (ꢂFBS + FBS) where lipoplex preparation was carried out
recorded on a Jasco FT-IR 410 spectrometer using KBr pellets or in the absence of serum but the lipoplexes were incubated with
ESI-TOF
spectrometer
or
on
ESI-MS Transfection experiments were performed in the presence of
(
as neat.
the cells in the presence of serum. At the time of transfection,
5764 | J. Mater. Chem. B, 2014, 2, 5758–5767
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
cells were at least 70% conuent. Liposomal formulations and concentration was estimated in each case using Bradford's
45
plasmid DNA were diluted in DMEM. Lipoplexes were prepared method with bovine serum albumin as a standard. Compar-
in DMEM by mixing liposomes and DNA in a required N/P ison of the transfection efficiencies of the individual lipids was
charge ratio and the total volume was made up to 200 ml. The N/ presented using data for luciferase assay expressed in relative
P values represent the ratio of charges on the cationic gemini light units (RLU) per mg of protein or luciferase activity per mg
cholesterol (in mol) to nucleotide base molarity and were of protein.
calculated by considering the average nucleotide mass of 330.
Lipid–DNA complexes were incubated for 30 min at room Statistical analysis
temperature followed by dilution with 200 ml of 20% FBS con-
taining DMEM (nal DNA concentration 12.12 mM). Formula-
Statistical signicance of differences between control and
samples were evaluated using two-way ANOVA using GraphPad
tions were added to the wells and incubated for 6 h under
Prizm 5.0 with Bonferroni analysis wherever applicable. Results
optimal conditions. Aer 6 h of incubation, old medium was
were considered statistically signicant when the p value was
less than 0.05 (* represents p < 0.05; ** represent p < 0.005 and
replaced with new medium having 10% FBS. Cells were further
incubated for a period of 42 h before checking the reporter gene
*
** represent p < 0.001).
expression. The extent of reporter gene expression was exam-
ined by a uorescence microscopy and/or luciferase assay
technique and expression was quantied either by ow cytom-
etry analysis or by luminometry. All the experiments were per-
formed in duplicate, and the results presented are the average
of at least duplicates of two such independent experiments
performed on different days. Lipofectamine2000 was used as
positive control using the manufacturer's protocol in 10%
serum, abbreviated as Lipo2000* throughout the manuscript.
Optimization of the lipid : DOPE ratio
DOPE, a natural lipid, is known to enhance gene transfection
46
efficacy of different cationic liposomal formulations, has been
used to prepare high output transfecting agents. The Gemini
lipid : DOPE ratio was varied from 1 : 0 to 1 : 5 of molar ratio to
optimize the efficiency of transfection. Efficiency of each lipo-
somal formulation was checked by performing pEGFP-C3,
reporter gene transfection on HeLa cells at an N/P charge ratio
of 0.5 under 10% FBS condition.
Flow cytometry
Transfection efficiency of different liposomal formulations for
the reporter gene pEGFP-C3 was examined by performing ow
cytometry assay (FACS). Transfected cells were regularly
Optimization of palmitoyl-RAFT-RGD inclusion methodology
4
44
Two different methodologies were used for non-covalent
incorporation of the palmitoyl-RAFT-RGD in GLD liposomes.
4
observed under a uorescent microscope for the intensity of
In the rst method, the optimized amount of GLD was mixed
with the RGD-containing scaffold in molar ratios of 2000 : 1 and

uorescence due to green uorescence protein (GFP) expressed
by the cells. Finally, 42 h post transfection, cells were trypsi-
nized by adding 100 ml of 0.1% trypsin and cells were collected
in 5% FBS containing DPBS. Duplicate cultures were pooled and
analyzed by ow cytometry immediately using a Becton and
Dickinson ow cytometer equipped with a xed laser source at
1000 : 1. Liposomes were then prepared from each of these
mixtures as mentioned earlier. These formulations were termed
as co-liposomal formulations Co-2000 and Co-1000 respectively.
A second method was used via injection of an ethanolic solution
of RGD-containing scaffold in pre-formed GLD suspensions in
buffered media. Formulations were termed as post-injection
formulation Po-2000 and Po-1000 respectively. Efficiency of
formulations was checked by performing pEGFP-C3, reporter
gene transfection on HeLa cells at an N/P charge ratio of 0.5
under 10% FBS condition.
488 nm. FACS data were analyzed using public domain
WinMDI2.8 soware.
Luciferase assay
Transfection efficiency of various formulations for the reporter
gene PGL-3 expression was quantied further by the luciferase
assay. Transfected cells were assayed using the single-luciferase
assay kit, provided by Promega (U.S.A.). In a typical assay, aer
Optimization of the GLD : palmitoyl-RAFT-RGD
Palmitoyl-RAFT-RGD was mixed with the GL5D formulation in
ratios of GLD : RGD-peptide 1 : 0, 2000 : 1, 1000 : 1, 100 : 1 and
0 : 1 respectively. GL5D was optimized as the 100 : 1 molar
4
molar ratio
4
48 h of transfection, the old medium was removed from the
wells and the cells were washed twice with 200 ml of PBS and
5
1
00 ml of 1ꢃ cell lysis buffer was added to each well. Cells were
ratio as best with respect to the transfection efficiency. This
optimized ratio was used for further optimization of the N/P
charge ratio.
lysed for 30 min in a horizontal rocker at RT. The cell lysate was
transferred to centrifuge tubes and centrifuged (4000 rpm, RT)
ꢁ
for 2 min. The supernatant was stored at ꢂ70 C until used for
the assay. For the assay, 10 ml of supernatant was mixed with 10
ml of the luciferase assay substrate (Promega). The lysate and the
Gel electrophoresis
substrate were both thawed to RT before performing the assay. Complexation and de-complexation ability of the GLD and
47
The luciferase activity was measured in a luminometer (Turner GLD–RGD at different GL/DNA ratios, DNase I stability of
48
designs, 20/20, Promega, U.S.A.) in standard single-lumines- lipoplexes derived from GLD/DNA and GLD–RGD/DNA and
49
cence mode. The measurement was performed for 10 s. A delay the stability of lipoplexes in the presence of FBS were all
of 2 s was given before each measurement. The protein assayed by gel electrophoresis. Lipoplexes were prepared at
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 5758–5767 | 5765

View Article Online
Journal of Materials Chemistry B
Paper
different GL/DNA charge ratios. Aer 30 min of incubation, conuency. Further, transfection was performed as mentioned
lipoplexes were run on 1% agarose gel to perform the gel above. Plates were then incubated for a period of 42 h before
retardation assay. Lipoplexes were incubated for 15 min in the checking under confocal microscopy. Control transfections
presence of micellar SDS at different SDS : GL molar ratios and were performed in each case by using commercially available
run on the 1% agarose gel to perform the de-complexation assay transfection agents ‘Lipofectamine2000’ and ‘Effectene’ using
whereas DNase I (1 unit) was incubated along with lipoplexes manufacturer's protocol in the presence (ꢂFBS + FBS) of serum.
ꢁ
for 30 min at 37 C and run on the 1% agarose gel to nd out the Aer 42 h of incubation, all the media were removed from the
DNA protection efficiency of lipoplexes against DNase I. wells and the cells were washed with DPBS buffer carefully to
Stability of lipoplexes in high serum percentage was performed remove all the cell debris without disturbing the monolayer of
by incubating the lipoplexes with high FBS percentage and run cells. Cells were xed for 10 min with 1 ml of 4% para-
on the 1% agarose gel. We also performed gel electrophoresis formaldehyde in each well. Cells were washed with DPBS for 3 ꢃ
for the lipoplexes that remained un-internalized to the cell aer 10 min and kept with 1 ml 0.1% Triton-x-100 for 5 min to
6
h of incubation. Observed band intensity proportional to the increase the membrane permeabilization. Cells were washed
amount of lipoplexes could not enter inside the cells even aer 6 with DPBS for 3 ꢃ 10 min. Glass slips having xed and per-
h. The high intensity band in the electrophoretic gel indicates mealizable monolayer of cells were taken out from the wells and
ꢂ1
low uptake of lipoplexes by cells.
kept on the glass slides and incubated for 5 min with 1 mg ml
of PI (propidium iodide) to specically stain the nucleus of the
cells. Finally cells were washed with DPBS for 3 ꢃ 10 min to
remove extra PI to reduce over-staining. The vector shield was
used to mount the glass slips upside down on the glass slides
which were observed under a confocal microscope (Zeiss) at an
objective of Plan Apochromat 63ꢃ/1.4 Oil DIC and analyzed
using a Zeiss LSM image browser.
Effect of palmitoyl-RAFT-RGD
4
on lipoplex delivery efficiency
inclusion on the
To measure the effect of palmitoyl-RAFT-RGD
4
DNA delivery efficiency of liposomes in each of HeLa, HEK293
and NIH3T3 cells, we performed UV absorption spectral
5
0
measurements for the lipoplexes which could not enter inside
the cells even aer 6 h of incubation. The experiment performed
was similar to the general transfection experiment using a
pEGFP-C3 plasmid complexed with GLD and GLD–RGD. Aer 6
h of incubation, culture medium possessing undelivered lip-
oplexes was removed. Medium was centrifuged at 13 000 rpm
Cytotoxicity
Cytotoxicity of GL5D and GL5D–RGD formulations was deter-
mined using a 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyl tetra-
zolium bromide (MTT) reduction assay in HeLa, HEK293 and
ꢁ
for 20 min at 4 C. Centrifuged lipoplexes were further sus-
pended in water and equally divided into two parts. One part
was run on 1% agarose gel and the second part was separated
using phenol : chloroform. The absorbance due to DNA was
measured using a BioRad UV-spectrometer. Absorbance values
(A260) were plotted against individual formulations which indi-
cate the amount of lipoplex le outside the cells. Results were
conrmed by gel electrophoresis.
52
NIH3T3 cells. In brief, 15 000 cells per well were plated in 96
well plates. Aer 24 h, GL5D, GL5D–RGD, GL5D/DNA and
GL5D–RGD/DNA formulations were added to the cells in the
absence of serum. Liposomal formulations GL5D and GL5D–
RGD were complexed with 0.2 mg of the DNA at various N/P
ratios for 30 min to give GL5D/DNA and GL5D–RGD/DNA
respectively. Aer 6 h of incubation, lipoplexes were removed
and 200 ml of media with 10% FBS was added. Aer 42 h, 20 ml of
MTT solution was added and the cells were incubated further
for 4 h. Blue formazan crystals were seen in the wells under a
microscope. The entire media were removed and 200 ml of
DMSO was added per well. The absorbance was measured using
a microtiter plate reader. The % viability was then calculated as
Cell surface binding analysis
The experiment was performed to nd out the specicity of
palmitoyl-RAFT-RGD
line that over-expresses avb3 and avb5 integrin receptors. A
uorescent probe, pyrene was included in GLD (GLD : pyrene
4
loaded liposomes toward the HeLa cell

molar ratio 100 : 1) with and without the palmitoyl-RAFT-RGD4.
Aer 24 h, cells were incubated with formulations possessing
either GLD : pyrene or GLD–RGD : pyrene for 2 h. Cells were
then washed with DPBS and collected for FACS analysis. The
FACS data were analyzed using WinMDI2.8 soware.
¼
[{A590 (treated cells) ꢂ background]/[A590 (untreated cells) ꢂ
background}] ꢃ 100.
Acknowledgements
We thank the DST grant (J C Fellowship to SB) and the ARCUS-
INDE-project for the support of this work.
Confocal uorescence microscopy
Transfection of pEGFP-C3 reporter plasmid was visualized
under a confocal uorescent microscope and the experiment
51
was performed as described. Cells were cultured in T25 culture
Notes and references

asks, trypsinized and plated in 12-well plates, having auto-
claved glass slips in the wells. Then, cells were plated on glass
slips in antibiotic free 10% FBS containing DMEM as medium.
Cells were grown for 24 h at 99% humidity, 37 C temperature
1 M. A. Barry, W. J. Dower and S. A. Johnston, Nat. Med., 1996,
2, 299.
2 D. Hallahan, L. Geng, S. Qu, C. Scarfone, T. Giorgio,
E. Donnelly, X. Gao and J. Clanton, Cancer Cell, 2003, 3, 63.
ꢁ
and 5% CO
2
condition till cell-monolayer gained ꢀ70%
5766 | J. Mater. Chem. B, 2014, 2, 5758–5767
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
3
4
A. Maria, E. W. Croyle, J. Sonia, J. R. Blake and L. A. Gordon, 29 J. Razkin, V. Josserand, D. Boturyn, Z. Jin, P. Dumy,
Hum. Gene Ther., 1998, 9, 561.
M. Favrot, J. L. Coll and I. Texier, ChemMedChem, 2006, 1,
D. Pramanik, B. K. Majeti, G. Mondal, P. P. Karmali, R. Sistla,
1069.
O. G. Ramprasad, G. Srinivas, G. Pande and A. Chaudhuri, J. 30 J. Dimastromatteo, L. M. Riou, M. Ahmadi, G. Pons,
Med. Chem., 2008, 51, 7298.
A. Mistry, R. Harbottle, S. Hart and K. M. Hodivala-Dilke,
Curr. Opin. Mol. Ther., 2003, 5, 603.
E. Pellegrini, A. Broisat, L. Sancey, T. Gavrilina, D. Boturyn,
P. Dumy, D. Fagret and C. Ghezzi, J. Nucl. Cardiol., 2010,
17, 435.
5
6
7
8
9
P. K. Dubey, D. Singodia, R. K. Verma and S. P. Vyas, J. Pharm. 31 Z.-H. Jin, T. Furukawa, M. Galibert, D. Boturyn, J. L. Coll,
Pharmacol., 2011, 63, 33–40.
M. A. Chellaiah and K. A. Hruska, Calcif. Tissue Int., 2003, 72,
T. Fukumura, T. Saga, P. Dumy and Y. Fujibayashi, Nucl.
Med. Biol., 2011, 38, 529.
1
97.
32 S. Foillard, Z. Jin, E. Garanger, D. Boturyn, M. Favrot,
J.-L. Coll and P. Dumy, Chem-Bio Inf. J., 2008, 9, 2326.
33 S. Foillard, L. Sancey, J.-L. Coll, D. Boturyn and P. Dumy, Org.
Biomol. Chem., 2009, 7, 221.
34 A. J. Schraa, R. J. Kok, A. D. Berendsen, H. E. Moorlag,
E. J. Bos, D. K. F. Meijer, L. F. M. H. de Leij and
G. Molema, J. Controlled Release, 2002, 83, 241.
D. C. Paik, C. Fu, J. Bhattacharya and M. D. Tilson, Exp. Mol.
Med., 2004, 36, 524.
K. Chitkara, K. Hogrefe, M. Vasa-Nicotera, N. Swanson and
A. H. Gershlick, J. Invasive Cardiol., 2006, 18, 417.
10 O. H. P. Ramos, A. Kauskot, M. R. Cominetti, I. Bechyne,
C. L. S. Pontes, F. Chareyre, J. Manent, R. Vassy,
M. Giovannini, C. Legrand, H. S. Selistre-de-Araujo, 35 T. Nishiya and S. Sloan, Biochem. Biophys. Res. Commun.,
M. Crepin and A. Bonnefoy, Clin. Exp. Metastasis, 2008, 25, 53. 1996, 224, 242.
1 P. C. Brooks, R. A. F. Clark and D. A. Cheresh, Science, 1994, 36 L. Sandrin, L. Coche-Guerente, H. Basit, A. Bernstein,
1
1
1
1
264, 569.
P. Labb ´e , P. Dumy and D. Boturyn, Org. Biomol. Chem.,
2 T. J. Wickham, P. Mathias, D. A. Cheresh and
G. R. Nemerow, Cell, 1993, 73, 309.
2010, 8, 1531.
37 A. Bajaj, P. Kondiah and S. Bhattacharya, J. Med. Chem.,
3 L. Erguang, L. B. Swati, G. S. Dwayne, S. P. Xose, A. C. David
and R. N. Glen, J. Virol., 2001, 75, 5405.
2007, 50, 2432.
38 S. Bhattacharya and A. Bajaj, Chem. Commun., 2009, 4632–
4 X. Zhengming, C. Zhen, Z. Xianzhong, P. Manish,
4656.
C. W. Joseph, S. G. Sanjiv and C. Xiaoyuan, J. Nucl. Med., 39 M. A. Ilies, W. A. Seitz, B. H. Johnson, E. L. Ezell, A. L. Miller,
2006, 47, 130.
E. B. Thompson and A. T. Balaban, J. Med. Chem., 2006, 49,
15 S. Klein, A. Bikfalvi, T. M. Birkenmeieri, F. G. Giancotti and
3872.
D. B. Riin, J. Biol. Chem., 1996, 271, 22583.
6 F. M. Fusi and R. A. Bronson, J. Androl., 1992, 13, 28.
40 Y. K. Ghosh, S. S. Visweswariah and S. Bhattacharya, FEBS
Lett., 2000, 473, 341.
1
1
7 R. Haubner, R. Gratias, B. Diefenbach, S. L. Goodman, 41 A. Bajaj, P. Kondaiah and S. Bhattacharya, J. Med. Chem.,
A. Jonczyk and H. Kessler, J. Am. Chem. Soc., 1996, 118, 7461. 2008, 51, 2533.
8 J.-X. Chen, H.-Y. Wang, C.-Y. Quan, X.-D. Xu, X.-Z. Zhang and 42 A. Bajaj, P. Kondaiah and S. Bhattacharya, Bioconjugate
R.-X. Zhuo, Org. Biomol. Chem., 2010, 8, 3142. Chem., 2008, 19, 1640.
9 E. Garanger, D. Boturyn, J. L. Coll, M. C. Favrot and P. Dumy, 43 A. Bajaj, P. Kondaiah and S. Bhattacharya, Bioconjugate
Org. Biomol. Chem., 2006, 4, 1958. Chem., 2007, 18, 1537.
0 Y.-b. Lim, O.-J. Kwon, E. Lee, P.-H. Kim, C.-O. Yun and 44 A. Sandrine, M. Grietje, L. Lou de and H. Dick, J. Gene Med.,
M. Lee, Org. Biomol. Chem., 2008, 6, 1944. 2000, 2, 465.
1 I. Martin, C. Dohmen, C. Mas-Moruno, C. Troiber, P. Kos, 45 A. Noguchi, N. Hirashima and M. Nakanishi, Pharm. Res.,
1
1
2
2
D. Schaffert, U. L ¨a chelt, M. Teixid ´o , M. G u¨ nther, H. Kessler
and E. Giralt, Org. Biomol. Chem., 2012, 10, 3258.
2 R. Xu, M. Fisher and R. L. Juliano, Bioconjugate Chem., 2011,
2002, 19, 933.
46 S. Bhattacharya and P. V. Dileep, Bioconjugate Chem., 2004,
15, 508.
2
2
2
2
2
2
2, 870.
47 S. Bhattacharya and S. S. Mandal, Biochemistry, 1998, 37,
7764.
48 M. A. Arangoa, N. Duezguenes and C. Tros de Ilarduya, Gene
Ther., 2003, 10, 5.
49 S. Fumoto, S. Kawakami, K. Shigeta, Y. Higuchi,
F. Yamashita and M. Hashida, J. Pharmacol. Exp. Ther.,
2005, 315, 484.
3 M. N. Bongiovanni, D. B. Scanlon and S. L. Gras,
Biomaterials, 2011, 32, 6099–6110.
4 M. Mammen, S. K. Choi and G. M. Whitesides, Angew.
Chem., Int. Ed., 1998, 37, 2755.
5 H. D. Maynard, S. Y. Okada and R. H. Grubbs, J. Am. Chem.
Soc., 2001, 123, 1275.
6 R. J. Kok, A. J. Schraa, E. J. Bos, H. E. Moorlag, 50 S. F. Ausar, S. B. Joshi and C. R. Middaugh, Methods Mol.
S. A. Asgeirsdottir, M. Everts, D. K. F. Meijer and
G. Molema, Bioconjugate Chem., 2002, 13, 128.
7 G. Thumshirn, U. Hersel, S. L. Goodman and H. Kessler,
Chem.–Eur. J., 2003, 9, 2717.
8 D. Boturyn, J. L. Coll, E. Garanger, M. C. Favrot and P. Dumy, 52 D. Fischer, T. Bieber, Y. Li, H. P. Elsasser and T. Kissel,
J. Am. Chem. Soc., 2004, 126, 5730. Pharm. Res., 1999, 16, 1273.
Biol., 2008, 434, 55.
51 R. Mukthavaram, S. Joyeeta, S. Marepally, M. Koushik,
S. Bojja and A. Chaudhuri, J. Am. Chem. Soc., 2007, 129,
11408.
2
2
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 5758–5767 | 5767