Optically triggered release of DNA from multivalent dendrons by degrading and charge-switching multivalency

Article abstract of DOI:10.1002/anie.200701200

(Figure Presented) Catch and release: Multivalent dendrons that have o-nitrobenzyl-linked spermine surface groups bind DNA efficiently by multivalent interactions. Cleavage of the o-nitrobenzyl groups from the dendron framework, with short UV irradiation,

Full text of DOI:10.1002/anie.200701200

Communications
DOI: 10.1002/anie.200701200
Controlled DNA Binding
Optically Triggered Release of DNA from Multivalent Dendrons by
Degrading and Charge-Switching Multivalency**
Mauri A. Kostiainen,* David K. Smith,* and Olli Ikkala
Multivalent binding between nanoscale objects has recently
emerged as one of the most powerful methods for the
assembly of functional supramolecular materials with appli-
cations in nanotechnology.[1,2] Controlling the self-assembly of
nanomaterials by using external stimuli, such as pH, temper-
ature, light, electric potential, or magnetic field, is an
important requirement for the preparation of functional and
responsive materials for a wide range of potential applica-
tions.[3,4] Stimuli-responsive materials have been extensively
pursued, with special focus on medicinal applications; for
example, controlled drug and DNA delivery systems,[5,6]
reactivation of caged enzymes,[7] and switchable membrane
proteins.[8] The use of light as an external stimulus offers a
number of advantages, because light is easy to apply,
relatively harmless to living organisms, and—importantly—
controllable both spatially and temporally.[5,7,9] DNA-binding
compounds that can be manipulated by light are especially
interesting in nonviral gene therapy, which relies on synthetic
compounds that protect, transport, and release DNA into
target cells.[10] Cationic dendritic systems have been of
particular interest in this regard.[11–13] Pioneering gene-ther-
apy research has been conducted with polyamidoamine
(PAMAM)[14] dendrimers, while dendritic poly(l-lysine)[15]
and poly(propylene imine)[16] have also been studied. Many
of these compounds bind DNA. However, unpacking of the
complexes and release of the DNA is difficult to achieve if the
binding is very strong, which results in a low transfection
efficiency.[17] DNA release is therefore of direct importance.
There have been some recent studies with a focus on
photocleavable dendrimers and dendrons. The studies include
self-immolative dendrimers[18] and dendrimers based on
photocleavable cores[19,20] or photoactive surfaces.[21] These
systems release covalently bound units from the dendritic
structure in a process triggered by UV irradiation. It is worth
noting that gold nanoparticles can also be employed in DNA
binding and delivery in an analogous manner to dendritic
structures.[9] Despite much progress, systems in which non-
covalent multivalent binding could be controlled by external
stimuli have not yet been fully developed.
Multivalency is defined as a type of binding in which
multiple ligands are attached to a single molecular scaffold
and used to interact with another entity that displays multiple
binding sites which are complementary to the ligands.[1,2] In
recent studies, we reported a series of Newkome-type[22]
dendritic ligands, with multiple protonated spermine groups
on their surfaces, which exhibit multivalent DNA bind-
ing.[17,23,24] Given our interest in multivalent DNA recogni-
tion, we decided to explore whether our receptor could be
developed in such a way as to achieve photoresponsivity. We
therefore modified our previously reported dendrons by
attaching the spermine surface groups through an o-nitro-
benzyl linker (Scheme 1a). The o-nitrobenzyl group under-
goes photolytic degradation (Scheme 1b) when submitted to
long-wavelength UV light (l = 350 nm), thus allowing a
controlled release of the covalently attached spermine surface
groups and the noncovalently bound DNA. Once the
spermine groups are cleaved from the surface of the dendron,
the cationic multivalency effect is destroyed, thereby leaving
just individual spermine groups, with only a weak affinity for
DNA. In this way, the DNA molecule will be effectively
decomplexed upon photolysis. Importantly, as the surface
groups are cleaved, they leave behind an anionic carboxylic
acid surface that will further repel DNA (Scheme 1c).
UV-responsive spermine derivatives (Scheme 1a) were
synthesized and characterized by using standard methods (see
the Supporting Information). Their photolytic degradation
was first studied by means of UV/Vis spectroscopy by
irradiating an aqueous solution of pll-G0, pll-G1 (see the
Supporting Information), or pll-G2 (Figure 1) and following
the time course of the reaction. Irradiation of these com-
pounds with UV light at 350 nm led to significant changes in
the UV/Vis spectra. A decrease of absorbance was observed
at 245 nm, along with a clear increase at 268 and 349 nm—
changes which typically indicate the photolytic reaction
proposed in Scheme 1b.[19,25] Figure 1b shows that the degra-
dation of the dendritic systems reaches a plateau after about
200 s. Longer irradiation times lead to further changes in the
absorption spectra, for example, to a decrease of the
absorption at 330–400 nm (see the Supporting Information).[8]
[*] M. A. Kostiainen, Prof. O. Ikkala
Department of Engineering Physics and Mathematics
Helsinki University of Technology
P.O. BOX 2200, 02015 HUT, Espoo (Finland)
Fax: (+358)9-451-3155
E-mail: kostiainen@gyroid.hut.fi
Prof. D. K. Smith
Department of Chemistry
University of York
Heslington, York, YO10 5DD (UK)
Fax: (+44)1904 432516
E-mail: dks3@york.ac.uk
[**] This work was supported by the Finnish National Graduate School
in Nanoscience (M.A.K.) and partly carried out in the Centre of
Excellence of Academy of Finland (Bio- and Nanopolymers Research
Group, 77317). The Research Foundation of Orion Corporation is
gratefully acknowledged for a study grant. We acknowledge Dr.
Markus Linder and Samuli Hirsjärvi for discussions and support.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
7600
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7600 –7604

Angewandte
Chemie
known to effectively
relaxweak DNA–cation
complexes.[27]
These
results are presented as
function of sulfonic
a
acid/protonatable den-
dron amine (S/N) ratio
(see the Experimental
Section).
We used two differ-
ent salt concentrations at
a physiological pH value
of 7.2, and employed
spermine as a reference
compound. Under low-
salt conditions, with
9.4mm NaCl, the non-
dendritic
compounds
spermine and pll-G0
bind to DNA, although
not particularly effec-
tively (CE50 = 6 and 32,
respectively, Figure 2a
and Table 1). The den-
dritic systems pll-G1 and
pll-G2, however, bind to
DNA very strongly and
with similar strength
(CE50 = 0.5 and 0.4,
respectively, Figure 2a
and Table 1). At a high
salt concentration of
150mm NaCl, spermine
and pll-G0 almost com-
pletely lose their DNA-
binding ability (CE50
>
Scheme 1. Spermine derivatives. a) Target photolabile dendrons pll-G0, pll-G1, and pll-G2. b) The photolysis of
pll-G1 and pll-G2 liberates the spermine surface and exposes the carboxylic acids. c) Self-assembly of multivalent
dendrons and DNA, and subsequent optically triggered degradation of the cationic surface and the release of
DNA. Blue spheres: photocleavage sites, red spheres: cationic spermine amines, yellow spheres: anionic
carboxylic acid groups exposed after photolysis.
200,
Figure 2b
and
Table 1).
Conversely,
compounds pll-G1 and
pll-G2 are barely
Table 1: Results obtained from an ethidium bromide displacement assay
for spermine, pll-G0, pll-G1, and pll-G2.[a]
The DNA-binding affinities of the spermine derivatives
and the optically triggered release of DNA from the
complexes were evaluated by using an ethidium bromide
(EthBr) fluorescence-quenching displacement assay.[26] This
assay measures the competition between the ligands and
EthBr for binding to DNA (as EthBr is displaced by the
ligands, its fluorescence, which is enhanced when bound to
DNA, decreases in intensity). Such a study leads to profiles
that define the CE50 values (Table 1), which represent the
“charge excess” (see the Experimental Section) required to
achieve a 50% reduction in the relative fluorescence inten-
sity. Notably, the fluorescence of EthBr can increase again if
re-intercalation in DNA becomes possible. DNA release can
therefore be monitored directly over the UV-irradiation time.
The strengths of the resulting DNA–dendron complexes were
also studied by DNA relaxation using chondroitin sulfate B
(csB), which is a sulfated polyanionic glycosaminoglycan
Compound
Nominal
Charge
CE50
(9.4mm NaCl)
CE50
(150mm NaCl)
Spermine
pll-G0
pll-G1
4+
3+
9+
27+
6
32
0.5
0.4
>400
>200
1.0
pll-G2
0.7
[a] Conditions: buffered water, pH 7.2 (2mm HEPES, 0.05mm EDTA).
The concentrations of DNA (1mm) and ethidium bromide (1.26mm) were
kept constant. The added polyamine solution did not exceed 5% of the
total volume; therefore no corrections were made for sample dilution.
The results are an average of three titrations.
affected by the increase in the concentration of competitive
Na+ ions, because of the multivalent nature of these dendritic
Angew. Chem. Int. Ed. 2007, 46, 7600 –7604
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7601

Communications
systems. The larger dendron pll-G2 binds slightly stronger
than pll-G1 (CE50 = 0.7 and 1.0, respectively, Figure 2b and
Table 1). These CE50 values are in good accordance with—
although slightly lower than—our previously reported values
for spermine derivatives without the pll-linker.[23]
Complexrelaxation with csB (at a salt concentration of
9.4mm NaCl) shows that pll-G0 and spermine pack DNA into
weak complexes, which are easily opened by a relatively small
amount of csB (Figure 2c). However, pll-G1 and pll-G2 form
extremely strong complexes with DNA that cannot be
opened—not even with very high S/N ratios (Figure 2c). If
the salt concentration is increased to 150mm, the pll-G1–
DNA complex can be fully relaxed with an approximately 10-
fold excess of csB, whereas the pll-G2 complexes are stronger
and can be relaxed with a 50-fold excess (Figure 2d).
The dendron–DNA complexdisassembly was then
directly monitored as a function of the irradiation time.
DNA was first fully complexed with the dendritic polycation
(CE = 2), and the resulting complexes were irradiated under
UV light. The EthBr fluorescence was then recorded after
different time periods (if dendron disassembly occurs, EthBr
should be able to compete effectively for DNA binding). At
9.4mm NaCl, the fluorescence of EthBr increases, which
indicates that both pll dendrons release DNA after 90 s
(Figure 2e), while at a salt concentration of 150mm, the
dendron pll-G1 releases DNA after 40 s and pll-G2 after 55 s
Figure 1. Photolysis of spermine derivatives. a) UV/Vis spectra of pll-
G2 (40 mgmLÀ1) after different UV-irradiation times (0–260 s). b) The
molar absorption coefficients of pll-G0, pll-G1, and pll-G2 (at 390 nm)
plotted against the UV-irradiation time indicate that pll-G2 releases
approximately three times more surface groups than pll-G1.
Figure 2. Titration curves for spermine, pll-G0, pll-G1, or pll-G2. EthBr fluorescence quenching in the presence of a) 9.4mm and b) 150mm NaCl.
DNA–polycation complex relaxation with csB in the presence of c) 9.4mm and d) 150mm NaCl. Release of DNA from the complexes by UV
irradiation in the presence of e) 9.4mm and f) 150mm NaCl. The added polyamine solution did not exceed 5% of the total volume; therefore, no
corrections were made for sample dilution. The results are the average of triplicates; error bars: standard deviation.
7602 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7600 –7604

Angewandte
Chemie
(Figure 2 f). This faster release might be expected as a result
of the slightly weaker complexation between the dendrons
and DNAunder the high-salt conditions. Dendron G1 (that is,
pll-G1 without the photolabile o-nitrobenzyl linker) was used
as a reference, and no release of DNA was observed from this
complexas a result of UV irradiation. Importantly, this same
trend—namely, that the dendrons pll-G1 and pll-G2 behave
similarly at 9.4mm NaCl, while the latter compound binds
more strongly to DNA than the former one at 150mm NaCl—
was observed with all three fluorescence titration methods.
G2 bind efficiently to DNA—through complementary elec-
trostatic interactions—but they can also release their target
very rapidly upon long-wavelength UV irradiation (l =
350 nm). Effectively, the high-affinity multivalent interactions
are “switched off” by UV irradiation. It is therefore possible
to control DNA binding and release, which makes these
dendrons very promising for applications in nanobiotechnol-
ogy.
Light scattering and z-potential measurements were used Experimental Section
Full details of the experimental methods can be found in the
Supporting Information.
to further investigate the DNA-binding and releasing proper-
ties of pll-G2. DNA was complexed with pll-G2 (CE = 2), and
the particle-count rate and the z-potential were measured
before and after one minute of irradiation. Before the UV
treatment, the observed count rate was 420.9 kcps (kilo
counts per second) and the z-potential was (14.7 Æ 3.4) mV,
which indicates the formation of a large number of positively
charged particles. After UV irradiation, however, the particle-
count rate dropped to only 8.8 kcps and the z-potential to
À(27.8Æ7) mV. This result confirms the complexbreakdown
and the formation of anionic species with a high negative
charge. The decrease in the count rate is attributed to the
change in the refractive indexof the dendron–DNA com-
plexes as they undergo a transition from condensed globules
to loose coils, which have a lower refractive index.
Charge Excess (CE): is defined as the nominal “number of
positive charges” of the polyamine divided by the “number of
negative charges” present on the DNA. A molecular weight of
330 gmolÀ1 and one negative charge per nucleotide were assumed.
S/N Ratio: is defined as the nominal “number of negative
charges” of the csB divided by the “number of positive charges” of the
polyamine. A molecular weight of 444 gmolÀ1 and one negative
charge per repeat unit were assumed.
Received: March 19, 2007
Revised: May 16, 2007
Published online: August 29, 2007
Keywords: dendrimers · DNA · multivalency · self-assembly ·
.
UV-degradation
Finally, DNA binding and release by pll dendrons was
confirmed by means of gel electrophoresis. The photolabile
dendritic constructs pll-G1 and pll-G2 retarded the electro-
phoretic mobility of pDNA as a consequence of charge
neutralization and complexformation, whilst pll-G0 was
ineffective (Figure 3). After UV irradiation, pll-G1 and pll-
[1] J. D. Badjic, A. Nelson, S. J. Cantrill, W. B. Turnbull, J. F.
Stoddart, Acc. Chem. Res. 2005, 38, 723 – 732.
[2] A. Mulder, J. Huskens, D. N. Reinhoudt, Org. Biomol. Chem.
2004, 2, 3409 – 3424.
[3] W. R. Browne, B. L. Feringa, Nat. Nanotechnol. 2006, 1, 25 – 35.
[4] G. ten Brinke, J. Ruokolainen, O. Ikkala, Adv. Polym. Sci. 2007,
207, 113 – 177.
[5] N. Nishiyama, A. Iriyama, W.-D. Jang, K. Miyata, K. Itaka, Y.
Inoue, H. Takahashi, Y. Y, Y. Tamaki, H. Koyama, K. Kataoka,
Nat. Mater. 2005, 4, 934 – 941.
[6] Y. Wang, S. Gao, W.-H. Ye, H. S. Yoon, Y.-Y. Yang, Nat. Mater.
2006, 5, 791 – 796.
[7] A. Deiters, D. Groff, Y. Ryu, J. Xie, P. G. Schultz, Angew. Chem.
2006, 118, 2794 – 2797; Angew. Chem. Int. Ed. 2006, 45, 2728 –
2731.
[8] A. Kocer, M. Walko, W. Meijberg, B. L. Feringa, Science 2005,
309, 755 – 758.
[9] G. Han, C.-C. You, B.-j. Kim, R. S. Turingan, N. S. Forbes, C. T.
Martin, V. M. Rotello, Angew. Chem. 2006, 118, 3237 – 3241;
Angew. Chem. Int. Ed. 2006, 45, 3165 – 3169.
Figure 3. Gel electrophoresis of pDNA (250 ng per lane). Lane 1:
pDNA. Lane 2: pDNA+pll-G0 (CE=30), no UV irradiation. Lane 3:
pDNA+pll-G1 (CE=2), no UV irradiation. Lane 4: pDNA+pll-G2
(CE=2), no UV irradiation. Lane 5: pDNA+pll-G1 (CE=2), 1 min UV
irradiation. Lane 6: pDNA+pll-G2 (CE=2), 1 min UV irradiation.
Lane 7: pDNA, 1 min UV irradiation.
[10] D. W. Pack, A. S. Hoffman, S. Pun, P. S. Stayton, Nat. Rev. Drug
Discovery 2005, 4, 581 – 593.
[11] U. Boas, J. B. Christensen, P. M. H. Heegard, Dendrimers in
medicine and biotechnology, RSC, Cambridge, 2006.
[12] S. Svenson, D. A. Tomalia, Adv. DrugDelivery Rev. 2005, 57,
2106 – 2129.
[13] D. Joester, M. Losson, R. Pugin, H. Heinzelmann, E. Walter,
H. P. Merkle, F. Diederich, Angew. Chem. 2003, 115, 1524 – 1528;
Angew. Chem. Int. Ed. 2003, 42, 1486 – 1490.
[14] J. Haensler, F. C. Szoka, Bioconjugate Chem. 1993, 4, 372 – 379.
[15] M. Ohsaki, T. Okuda, A. Wada, T. Hirayama, T. Niidome, H.
Aoyagi, Bioconjugate Chem. 2002, 13, 510 – 517.
[16] B. H. Zinselmeyer, S. P. Mackay, A. G. Schatzlein, I. F. Uchegbu,
Pharm. Res. 2002, 19, 960 – 967.
G2 clearly released DNA, thus allowing its free electro-
phoretic mobility. Interestingly, the conformation of the pll-
G1- or pll-G2-bound pDNA shifted from the supercoiled to
the open-circular form (lanes 5 and 6 in Figure 3) after UV
irradiation. Importantly, UV-irradiation-induced pDNA frag-
mentation was not observed.
In conclusion, we have demonstrated a new approach for
achieving reversible DNA binding by using photolabile
multivalent dendrons, which release DNA by degrading and
charge-reversing multivalency. In particular, pll-G1 and pll-
Angew. Chem. Int. Ed. 2007, 46, 7600 –7604
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7603

Communications
[17] J. G. Hardy, M. A. Kostiainen, D. K. Smith, N. P. Gabrielson,
[23] M. A. Kostiainen, J. G. Hardy, D. K. Smith, Angew. Chem. 2005,
117, 2612 – 2615; Angew. Chem. Int. Ed. 2005, 44, 2556 – 2559.
[24] M. A. Kostiainen, G. R. Szilvay, D. K. Smith, M. B. Linder, O.
Ikkala, Angew. Chem. 2006, 118, 3618 – 3622; Angew. Chem. Int.
Ed. 2006, 45, 3538 – 3542.
[25] M. S. Kim, S. L. Diamond, Bioorg. Med. Chem. Lett. 2006, 16,
4007 – 4010.
[26] B. F. Cain, B. C. Baguley, W. A. Denny, J. Med. Chem. 1978, 21,
658 – 668.
D. W. Pack, Bioconjugate Chem. 2006, 17, 172 – 178.
[18] R. J. Amir, N. Pessah, M. Shamis, D. Shabat, Angew. Chem. 2003,
115, 4632 – 4637; Angew. Chem. Int. Ed. 2003, 42, 4494 – 4499.
[19] M. Smet, L. X. Liao, W. Dehaen, D. V. McGrath, Org. Lett. 2000,
2, 511 – 513.
[20] M. L. Szalai, D. V. McGrath, Tetrahedron 2004, 60, 7261 – 7266.
[21] I. Grabchev, V. Bojinov, J.-M. Chovelon, Polymer 2003, 44,
4421 – 4428.
[27] M. Ruponen, S. Yla-Herttuala, A. Urtti, Biochim. Biophys. Acta,
Biomembr. 1999, 1415, 331 – 341.
[22] G. R. Newkome, X. Lin, Macromolecules 1991, 24, 1443 – 1444.
7604 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7600 –7604