Spacer structure and hydrophobicity influences transfection activity of novel polycationic gemini amphiphiles

Article abstract of DOI:10.1016/j.bmcl.2017.06.026

Three novel polycationic gemini amphiphiles with different spacers were developed and evaluated in terms of their physiochemical properties and transfection efficiencies. Cationic liposomes formed by these amphiphiles and the helper lipid DOPE were able to successfully condense DNA, as shown by gel mobility shift and ethidium bromide intercalation assays. Transfection activity of the liposomes was superior to Lipofectamine 2000 and was dependent on spacer structure, hydrophobicity, and nucleic acid type (pDNA or siRNA). We demonstrated that the cationic liposomes 2X6/DOPE and 2X7/DOPE are potential non-toxic vehicles for gene delivery.

Full text of DOI:10.1016/j.bmcl.2017.06.026

Accepted Manuscript
Spacer structure and hydrophobicity influences transfection activity of novel
polycationic gemini amphiphiles
Pavel A. Puchkov, Irina A. Kartashova, Elena V. Shmendel, Anastasya S.
Luneva, Nina G. Morozova, Marina A. Zenkova, Mikhail A. Maslov
PII:
S0960-894X(17)30627-3
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.06.026
BMCL 25060
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
12 April 2017
7 June 2017
8 June 2017
Please cite this article as: Puchkov, P.A., Kartashova, I.A., Shmendel, E.V., Luneva, A.S., Morozova, N.G., Zenkova,
M.A., Maslov, M.A., Spacer structure and hydrophobicity influences transfection activity of novel polycationic
gemini amphiphiles, Bioorganic & Medicinal Chemistry Letters (2017), doi: http://dx.doi.org/10.1016/j.bmcl.
2017.06.026
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.

Bioorganic & Medicinal Chemistry Letters
jo u r n a l h o m e p a g e : w w w . e ls e v i e r . c o m
Spacer structure and hydrophobicity influences transfection activity of novel
polycationic gemini amphiphiles
Pavel A. Puchkova, Irina A. Kartashovaa, Elena V. Shmendela, Anastasya S. Lunevaa, Nina G. Morozovaa,
Marina A. Zenkovab, Mikhail A. Maslova,
aInstitute of Fine Chemical Technologies, Moscow Technological University, Vernadsky Ave. 86, Moscow 119571, Russia
bInstitute of Chemical Biology and Fundamental Medicine SB RAS, Lavrentiev Ave. 8, Novosibirsk 630090, Russia
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
Three novel polycationic gemini amphiphiles with different spacers were developed and
evaluated in terms of their physiochemical properties and transfection efficiencies. Cationic
liposomes formed by these amphiphiles and the helper lipid DOPE were able to successfully
condense DNA, as shown by gel mobility shift and ethidium bromide intercalation assays.
Transfection activity of the liposomes was superior to Lipofectamine® 2000 and was dependent
on spacer structure, hydrophobicity, and nucleic acid type (pDNA or siRNA). We demonstrated
that the cationic liposomes 2X6/DOPE and 2X7/DOPE are potential non-toxic vehicles for gene
delivery.
Keywords:
Cationic liposomes
Spacer
Gemini amphiphiles
Transfection
2017 Elsevier Ltd. All rights reserved.
Gene therapy
Among the different gene delivery systems, non-viral systems
such as cationic liposomes (CLs) and cationic polymers remain
the most safe and attractive carriers. CLs are formed by cationic
amphiphiles (CAs) that are cheap, easy to manufacture, and
scalable, but still not particularly effective.1 The transfection
efficiency of CLs is highly dependent on the structure of the CA
forming CLs. Typical CAs consist of hydrophilic and
hydrophobic domains, a linker connecting them, and a spacer that
maintains the steric arrangement of the domains.1,2 Each CL
component plays a key role in liposome structure maintenance
and influences the resulting efficiency of gene delivery.
As mentioned above, each CA component plays a role in the
biological activity of CA, with the spacer possibly being one of
the key domains that should be optimised to improve gene
delivery activity. It has been shown that spacer length affects
DNA condensation and the physicochemical properties of CL-
DNA complexes.15 Gemini cholesterol-based CAs with different
spacer lengths (3-12 carbon atoms) were tested, and the highest
transfection efficiency (TE) observed for CAs with spacer
lengths of five or six carbon atoms.16 Similar results were
observed for amino acid-based CAs, for which the optimal spacer
length was shown to be seven carbon atoms.17 Our own
investigations also showed that gemini CAs with a spacer length
of six carbon atoms mediated improved gene delivery compared
to analogous CAs with a spacer length of four carbon atoms.14 In
addition to the hydrocarbon spacers, which are hydrophobic,
hydrophilic oxyethylene-type spacers have also been used.17–19
Bajaj et al. demonstrated that introduction of an oxyethylene unit
had a positive effect on gene delivery.18 By contrast, gene
delivery mediated by amino acid-based CAs with an
oxyethylene-type spacer was poor.17 This discrepancy in
transfection efficiency is likely linked with differences in the cell
lines used as well as the influence of other CA structural
components. Supporting evidence for this hypothesis was
provided by an investigation of the structure-activity
relationships of gemini CAs bearing hydrophobic or hydrophilic
spacers.20 CAs with hydrophobic spacers were shown to be
effective in HeLa cells, whereas CAs with hydrophilic spacers
To date, a large number of CAs utilised for gene delivery were
based on various polyamines.3 Foremost is the natural polyamine
spermine, which has the unique ability to efficiently condense
nucleic acids.4 CLs that included an amphiphile as a core
component were shown to be promising gene delivery vectors.5–9
Gemini amphiphiles (including those that are polyamine-based)
have also been found suitable for nucleic acid delivery. The
advantages of these amphiphiles are a much lower critical micelle
concentration (CMC) and an ability to reduce surface tension,
which favours improved cell membrane permeability.10,11 We
recently developed the polycationic gemini amphiphile 2X3,
which is based on cholesterol and spermine (Fig. 1).12 CLs
containing this amphiphile exhibited high transfection efficiency,
significantly exceeding the transfection efficiency of both
Lipofectamine® 2000 and monomeric counterparts.13,14
———
 Corresponding author. Tel.: +7-499-600-8202; fax: +7-495-434-9287; e-mail: mamaslov@mail.ru

organoborane compound with a high regioselectivity in the
hydroboration of terminal alkenes.24 Subsequent oxidative
cleavage of boron derivatives with H2O2 in an aqueous alkali
base gives terminal alcohols with high yields. After the removal
of the Вос protecting group by treatment with 3 М HCl in
dioxane, compound 4 was isolated by column chromatography
with a three-step yield of 81%. The hydrophobic and hydrophilic
spacers 7а,b, each with a chain length of eight atoms, were
synthesised in two steps. The azides 6а,b were obtained from the
chlorinated alcohols 5a,b by treatment with sodium azide (Fig.
2). Catalytic hydrogenation in the presence of 5% palladium on
carbon subsequently produced the amines 7а,b in 80% yields.
An activated hydrophobic motif needed for coupling with
spacers 4 and 7а,b was next obtained. We specifically used two
activated cholesterol derivatives, namely cholest-5-en-3β-yl
imidazole-1-carboxylate (8a)25 and commercially available
cholesterol chloroformate (8b) (Fig. 3). Imidazolide 8a did not
react with 7a since its corresponding alcohol was not detected.
We thus used cholesterol chloroformate instead. The alcohols
obtained were subsequently brominated by tetrabromomethane in
the presence of triphenylphosphine. The product 9b was
crystallised from diethyl ether with a two-step yield of 64%. The
bromo derivatives 9a,c were isolated by column chromatography
with 43–56% yields.
Figure 1. Structures of the polycationic gemini amphiphiles.
were more effective in HT1080 cells. Introduction of hydroxyl
groups into the spacer chain also did not lead to appreciable
changes in transfection efficiency.19 Further investigations are
therefore required to better understand the influence of spacer
structure on TE.
To synthesise amphiphiles 2X5, 2X6, and 2X7, 4,9-di(tert-
butoxycarbonyl)-1,12-bis(2-nitrobenzenesulfonylamino)-4,9-
diazadodecane12 was treated with the corresponding bromo
derivatives 9a-c, in the presence of cesium carbonate under
Fukuyama reaction conditions,26 and gave compounds 10a-c with
51–66% yields (Fig. 3). Removal of the 2-nitrobenzenesulfonyl
protecting group by treatment with thiophenol in the presence of
potassium carbonate, followed by N-Вос deprotection with 3 М
HCl in dioxane, produced CAs 2X5, 2X6, and 2X7 with two-step
yields of 47–65%. The synthesised compounds were
characterised by 1H-NMR, 13C-NMR, and mass spectrometry
(See Supplementary materials). The critical micelle concentration
values (25–30 μM) were also determined by dynamic light
scattering and shown to be independent on spacer structure.
In this study, we designed, synthesised, and performed a
detailed analysis of the transfection activity of three novel
polycationic gemini amphiphiles (2X5, 2X6, 2X7; Fig. 1). These
new CAs consisted of cholesterol and the natural polyamine
spermine as the lipophilic and cationic domains and varied in
spacer structures and hydrophobicity. CLs formed by the novel
CAs and the helper lipid 1,2-dioleoyl-sn-glycero-3-
phosphoethanolamine (DOPE) were tested in terms of
physicochemical properties, toxicity, and transfection efficiency.
Moreover, we compared the transfection activities of the novel
amphiphiles with the previously studied CA 2X314 which has a
similar
structure,
and
the
commercial
transfectant
Lipofectamine® 2000.
Based on the CAs 2X3,12 2X5, 2X6, 2X7, and the zwitter-
ionic helper lipid DOPE, the CLs 2X3/DOPE, 2X5/DOPE,
2X6/DOPE, and 2X7/DOPE were prepared at a lipid molar ratio
of 1:1 using the thin film hydration method. The hydrodynamic
diameter of each CL was determined by dynamic light scattering
and found to be between 40 and 60 nm; however, 2X5/DOPE had
an additional minor fraction of particles with an approximate
diameter of 200 nm (Table 1). All CLs were positively charged.
Several approaches have been used to connect polyamine and
hydrophobic domains upon synthesis of polycationic
amphiphiles. Most approaches are primarily based on coupling of
the spacer to the polyamine, followed by reaction with the
activated hydrophobic domain.21–23 We used an alternative
approach and first coupled the spacer with the hydrophobic
domain, followed by condensation of the intermediate with the
spermine derivative.
Spacers were synthesised from available commercial reagents.
To obtain the hydrophobic spacer 4, the amino group of
2-aminoethanol (1) was routinely protected with the Boc group
(Fig. 2). Introduction of the terminal allyl group was carried out
using 3-bromo-1-propene in the presence of tetrabutylammonium
The ability of CLs to bind and condense nucleic acids was
studied using the gel mobility shift assay.27 CLs and plasmid
DNA (pDNA) complexes were prepared at various N/P ratios
(number of polycationic amino groups of CAs per phosphate
groups of nucleic acids) and the resulting mixtures were resolved
on an ethidium bromide (EB) containing agarose gel, under
natural conditions, to detect the presence of non-bound pDNA
(Fig. 4A). Low N/P ratios (1/1 and 2/1) resulted in pDNA not
bromide
9-borabicyclo[3.3.1]nonane (9-BBN) was used for the conversion
of alkene into corresponding alcohol. 9-BBN is an
(TBAB)
in
aqueous
NaOH.
Then,
3
Figure 2. Synthesis of spacers 4 (A) and 7a,b (B). a) Boc2O, NaOH, THF, 24 °C, 5 h; b) 3-bromo-1-propene, TBAB, NaOH, 24 °C, 19 h; c) 9-BBN, THF,
9 °С, 4 h, then 30 % aq. H2O2, NaOH, 5 °C, 1 h; d) 3 M HCl/dioxane, DCM, 20 °C, 1.5 h; e) NaN3, DMF, 70 °C, 27–74 h; f) H2, 5 % Pd/C, MeOH, 24 °C,
2-10 h.

Figure 3. Synthesis of the polycationic amphiphiles 2X5, 2X6, and 2X7. a) Et3N, DCM, for 8a reflux, 58–70 h, for 8b 24 °C, 20–24 h; b) CBr4, Ph3P,
MeOH, 24 °C, 3–50 h; c) 4,9-di(tert-butoxycarbonyl)-1,12-bis(2-nitrobenzenesulfonylamino)-4,9-diazadodecane, Cs2CO3, DMF, 80 °C, 32–37 h; d) PhSH,
K2CO3, DMF, 24 °C, 5–35 h; e) 3 M HCl in dioxane, DCM, 24 °C, 3–5 h.
being fully bound to CLs since free pDNA remained visible in
the gel. On increasing the N/P ratio to 4/1 (or to 2/1, in the case
of 2X6/DOPE and 2X7/DOPE, both with spacer lengths of eight
atoms), all pDNA appeared to be complexed with CLs. Analysis
of the physicochemical characteristics of CL/pDNA complexes
formed at N/P 6/1 (Table 1) showed that compact particles are
formed under these conditions; the smallest was formed with
pDNA and 2X7/DOPE, while the largest was formed with
2X5/DOPE. The CLs 2X6/DOPE and 2X3/DOPE formed
CL/pDNA complexes approximately 100 nm in diameter, which
was not surprising since these two CLs only differ by two
methylene groups in their spacers. Complexes of 2X7/DOPE
with small interfering RNA (siRNA) were also the smallest in
diameter. By contrast, 2X5/DOPE and siRNA formed
abnormally large particles that above 1500 nm in diameter. Also,
increasing the spacer length is considered to lead to a lower
polydispersity index.
Fluorescent titration with EB also was performed to evaluate
the binding ability of the CLs with respect to DNA (calf thymus
DNA was used in this set of experiments). Increasing the
proportion of CL in complexes decreased the fluorescence
intensity of EB, indicating its displacement from the DNA double
helix (Fig. 4B). All CLs demonstrated high DNA binding ability,
with 2X7/DOPE being slightly superior to the others.
Interestingly, the major differences observed between the
capacities of each CL to bind DNA was at an N/P ratio of 1/1.
Under these conditions, 2X3/DOPE and 2X6/DOPE only
Figure 4. DNA binding studies at various N/P ratios. A) Gel retardation
displaced half the DNA-bound EB (48.6 ±0.7% and 45.4 ±0.6%
assay of CL-DNA complexes, where K = pDNA as control; B) Fluorescence
EB remained, respectively), while 2X5/DOPE and especially
spectroscopy of EB displacement by the respective CLs. Each measurement
2X7/DOPE displaced much higher levels of DNA-bound EB
was taken in triplicate.
Table 1. Physicochemical characterisation of CLs and complexes formed with plasmid DNA and siRNA at a N/P of 6/1
Liposomes only
pEGFP-C2 plasmid
Anti-EGFP siRNA
Mean
diameter
(nm)
CLs
Mean diameter
ξ-potential
ξ-potential
Mean
diameter (nm)
ξ-potential
(mV)
PIa
PI
PI
(nm)
(mV)
66.5
59.8
(mV)
46.9
48.4
2X3/DOPE
2X5/DOPE
51.0±9.5
0.307
0.290
99±0.6
0.238
0.352
170.3±0.1
0.359
0.369
n.d.
60.5±11.9
193.7±58.9
152.8±0.3
1665.3±126.2
26.0
2X6/DOPE
2X7/DOPE
58.1±4.4
0.217
0.232
60.9
57.8
102.4±0.3
66.6±0.7
0.147
0.282
47.1
41.9
226.8±1.9
158.7±1.3
0.079
0.261
43.4
32.0
39.7±2.8
aPI: polydispersity index. n.d.: no data

(33.0 ±1.0% and 19.2 ±0.3% EB remained, respectively).
To evaluate the TE of the novel CLs, complexes were
prepared at different N/P ratios with the pEGFP-C2 plasmid,
which encodes enhanced green fluorescent protein (EGFP).
Experiments were performed in HEK 293 cells in the presence of
10% FBS (foetal bovine serum) to mimic in vivo conditions. The
TE of all CLs increased with increasing N/P ratios, except for
2X7/DOPE with an ethoxyethoxyethylene spacer, which reached
a maximal TE at an N/P ratio of 6/1 followed by a decrease in TE
when increasing the N/P ratio to 8/1 (Fig. 5A and 5B). The CLs
2X3/DOPE, 2X5/DOPE, and 2X6/DOPE all mediated a 1.5–2.0
fold higher percentage of transfected cells at N/P 6/1 and 8/1 than
Lipofectamine® 2000 (Lf2000). The same results were
demonstrated by 2X7/DOPE at lower N/P ratios of 4/1 and 6/1.
The highest level of EGFP expression was mediated by
2X7/DOPE at N/P 6/1 (Fig. 5B). We suggest that the high
transfection efficiency of 2X7/DOPE may be explained by the
smaller diameter of these complexes with pDNA (67 nm, Table
1). The TE of CL 2X5/DOPE with an ethoxypropylene spacer
was comparable to 2X3/DOPE with an hexamethylene spacer,
whereas 2X6/DOPE with an octamethylene spacer showed a
slightly lower mean fluorescence intensity (Fig. 5B). These
findings therefore indicate 2X7/DOPE as being the most
promising CL for pDNA delivery at an optimal N/P ratio of 6/1.
To further study the transfection profiles of the novel CLs,
analysis of the delivery of siRNA targeting EGFP mRNA was
carried out. The siRNA duplex, which is homologous to EGFP
messenger RNA, was delivered into BHK IR-780 cells stably
expressing EGFP. A decrease in the cells fluorescence intensity
can therefore be used as an indication of the efficiency of siRNA
delivery into the cytoplasm through silencing of EGFP via a
RNAi mechanism.28 As was the case with pDNA, TE was
enhanced with N/P increasing ratios (Fig. 5C). At an N/P ratio of
10/1, CLs with oxyethylene units (2X5/DOPE and 2X7/DOPE)
exhibited EGFP inhibition similar to both 2X3/DOPE with an
hexamethylene spacer and Lf2000. The highest level of EGFP
inhibition was observed in the case of 2X6/DOPE-mediated
siRNA delivery at an N/P ratio of 10/1 (90% inhibition).
Figure 5. Delivery of pDNA in HEK 293 cells (A, B) and siRNA in
BHK IR-780 cells (C) mediated by CLs. Complexes were prepared at
various N/P ratios (for pDNA 2/1, 4/1, 6/1, 8/1; for siRNA 4/1, 6/1, 8/1, 10/1
– bars from left to right). The data are presented as a percentage of
fluorescent cells (A), a mean fluorescence intensity of the cell population
(B), and as a percentage of EGFP inhibition in comparison with its levels in
untreated cells (C), as determined by flow cytometry. 2X3/DOPE and
Lf2000 were used as positive controls. Experiments were performed in the
presence of 10% FBS, each in triplicate. Standard deviation did not exceed
7%–9%.
Elongation of the spacer from six (2X3, 2X5) to eight (2X6,
2X7) atoms is therefore considered to have a positive effect on
the TE of the corresponding liposomes, while introduction of
oxyethylene units (2X5, 2X7) leads to contradictory effects on
transfection activity. In the case of 2X5/DOPE, oxyethylene
spacer caused no significant changes in TE, while for
2X7/DOPE, a sensitivity to nucleic acid type was observed. The
size of the complexes also appeared to influence transfection
efficiency; the biggest complexes were formed for 2X5/DOPE
with both pDNA and siRNA in comparison to the other CLs.
These larger CLs were characterized by a much lower TE.
structure and hydrophobicity of the spacer used as well as the
type of nucleic acid. The CA 2X7, with an ethoxyethoxyethylene
spacer, was the best transfection agent since the CL 2X7/DOPE
was superior to both the 2X3/DOPE and Lf2000 in the delivery
of pDNA. On the other hand, the CA 2X6, with an octamethylene
spacer, provided the most efficient delivery of siRNA in BHK
IR-780 cells as a component of a liposomal formulation. The CLs
2X6/DOPE and 2X7/DOPE are thus potential non-toxic vehicles
for nucleic acid delivery.
The cytotoxic effects of the CLs 2X6/DOPE and 2X7/DOPE
were tested on HEK 293 cells by using the xCelligence
system®.29 Experiments were carried out at CLs concentrations of
3–24 μM, which correspond to the amount of CLs used for
transfection. Results showed that IC50 values (concentration of
the liposomes providing 50% cell viability) were not reached for
both 2X6/DOPE and 2Х7/DOPE at these concentrations.
In this study, three novel polycationic amphiphiles, 2X5, 2X6,
and 2X7 with different spacers were successfully synthesised and
used to prepare liposomal formulations with the helper lipid
DOPE. The resulting CLs were used for delivery of pDNA and
siRNA, and TE was evaluated by either measuring the level of
transgene expression or by the level of inhibition of target gene
expression. We demonstrated that TE was affected by the
Acknowledgments
This work was supported by the President's Program in
Support of Leading Scientific Schools SS-7946.2016.11 and by
the Russian Foundation for Basic Research (grant nos. 13-04-
40183 comfi, and 13-04-40181 comfi).

Supplementary Material
Supplementary data associated with this article can be found
in the online version,
References and notes
1.
2.
Zhao Y, Huang L. Adv Genet. 2014; 88:13–36.
Sanchez A, Pensado A. Expert Opin Drug Deliv. 2014;11:1721–
1731.
3.
Vijayanathan V, Agostinelli E, Thomas T, Thomas TJ. Amino Acids
2014;46:499–509.
4.
5.
6.
Ramani D, De Bandt JP, Cynober L. Clin Nutr. 2014;33:14–22.
Gruneich JA, Diamond SL. J. Gene Med. 2007;9:381–391.
Randazzo RAS, Bucki R, Janmey PA, Diamond SL. Bioorg Med
Chem. 2009;17:3257–3265.
7.
Blagbrough IS, Metwally AA, Ghonaim HM. Mol Pharm.
2012;9:1853–1861.
8.
9.
Ghonaim HM, Li S, Blagbrough IS. Pharm Res. 2010;27:17–29.
Dewa T, Asai T, Tsunoda Y, Kato K, Baba D, Uchida M, Sumino
A, Niwata K, Umemoto T, Iida K, Oku N, Nango M. Bioconjug
Chem. 2010;21:844–852.
10.
11.
Menger FM, Keiper JS. Angew Chemie Int Ed. 2000;39:1906–1920.
Kirby AJ, Camilleri P, Engberts JBFN, Feiters MC, Nolte RJM,
Söderman O, Bergsma M, Bell PC, Fielden ML, García Rodríguez
CL, Guédat P, Kremer A, McGregor C, Perrin C, Ronsin G, Van
Eijk MCP. Angew Chemie Int Ed. 2003;42:1448–1457.
Petukhov IA, Maslov MA. Russ Chem Bull, Int Ed. 2010, 59, 260–
268.
Markov OO, Mironova NL, Maslov MA, Petukhov IA, Morozova
NG, Vlassov VV, Zenkova MA. J Control Release. 2012;160:200–
210.
12.
13.
14.
15.
Maslov MA, Kabilova TO, Petukhov IA, Morozova NG,
Serebrennikova GA, Vlassov VV, Zenkova MA. J Control Release.
2012;160:182–193.
Luciani P, Bombelli C, Colone M, Giansanti L, Ryhänen SJ, Säily
VMJ, Mancini G, Kinnunen PKJ. Biomacromolecules.
2007;8:1999–2003.
16.
17.
18.
19.
20.
21.
Bajaj A, Kondiah P, Bhattacharya S. J Med Chem. 2007;50:2432–
2442.
Obata Y, Saito S, Takeda N, Takeoka S. Biochim. Biophys. Acta,
Biomembr. 2009;1788:1148–1158.
Bajaj A, Kondaiah P, Bhattacharya S. Bioconjug Chem.
2007;18:1537–1546.
Misra SK, Biswas J, Kondaiah P, Bhattacharya S. PLoS One.
2013;8:e68305.
Bajaj A, Kondaiah P, Bhattacharya S. J Med Chem. 2008;51:2533–
2540.
Ronsin G, Perrin C, Guédat P, Kremer A, Camilleri P, Kirby AJ,
Mook RA, Hendren RW, Wagner RW. Chem Commun.
2001;35:2234-2235.
22.
Miller KA, Kumar EVKS, Wood SJ, Cromer JR, Datta A, David
SA. J. Med. Chem. 2005;48:2589-2599.
23.
24.
25.
Geall AJ, Blagbrough IS. Tetrahedron. 2000;56:2449–2460.
Bracher F, Litz T. J Prakt Chem. 1996;338:386–389.
Medvedeva DA, Maslov MA, Serikov RN, Morozova NG,
Serebrenikova GA, Sheglov DV, Latyshev AV, Vlassov VV,
Zenkova MA. J. Med. Chem. 2009;52:6558–6568.
Fukuyama T, Jow CK, Cheung M. Tetrahedron Lett.
1995;36:6373–6374.
Aljaberi A, Spelios M, Kearns M, Selvi B, Savva M. Colloids Surf,
B. 2007;57:108–117.
Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC.
Nature. 1998;391:806–811.
26.
27.
28.
29.
Ke N, Wang X, Xu X, Abassi YA. Methods in Molecular Biology.
2011;740:33–43.