Synthesis of stable peptide nucleic acid-modified gold nanoparticles and their assembly onto gold surfaces

Article abstract of DOI:10.1002/anie.201209684

PNA does it better: Peptide nucleic acid (PNA) oligomers were attached to gold nanoparticles (AuNPs) through a variety of mono- and trithiol linkers. These functionalized particles had sufficient stability for sequence-specific self-assembly onto gold surfaces (see figure) in the absence of ions or surfactants. The nanoparticle surface densities obtained were superior to comparable DNA-modified AuNPs. Copyright

Full text of DOI:10.1002/anie.201209684

Angewandte
Chemie
DOI: 10.1002/anie.201209684
Gold Nanoparticles
Synthesis of Stable Peptide Nucleic Acid-Modified Gold Nanoparticles
and their Assembly onto Gold Surfaces**
Philipp Anstaett, Yuanhui Zheng, Thibaut Thai, Alison M. Funston, Udo Bach,* and
Gilles Gasser*
DNA-based gold-nanoparticle (AuNP) systems are currently
employed in a growing range of applications,[1] which include
gene regulation,[2] nanofabrication,[3] sensing,[4] and plasmonic
rulers.[5] However, a disadvantage of the assembly of DNA-
modified NPs is the need for salt to keep their assemblies
stable. Halogens are known to damage silver nanoparticles
(AgNPs),[6] while the addition of salt in general destabilizes
colloidal particles. Additionally, the presence of substantial
amounts of ions is problematic for the study and use of
physical phenomena that rely on electrostatics.[7] Thus, the
functionalization of nanoparticles with the non-natural DNA
analogue peptide nucleic acid (PNA), instead of DNA, is
extremely attractive. PNA has many advantages over DNA
including a higher stability against biodegradation, greater
mismatch sensitivity, and higher binding efficiency to PNA,
DNA, and RNA.[8] Hence, shorter oligonucleotide strands
could potentially be used for the assembly of PNA-function-
alized AuNPs compared to those required for stable DNA-
based assembly. An increase in the resolution control of close-
packed gold nanoparticles could therefore be observed.
Additionally, the stability of PNA-DNA and PNA-PNA
hybrids is independent of the ionic strength of the
medium.[9] In comparison, DNA alone is not able to form
assemblies under ion-free conditions as the electrostatic
repulsion between the negatively charged strands is too
high. Therefore, AuNP assemblies relying on PNA hybrid-
ization could be formed without the addition of salt, unlike
DNA-AuNP hybrids. Despite the huge potential of PNA-
based AuNPs, there are only a few articles reporting the direct
attachment of PNA onto AuNPs.[10] Part of the reason for this
is the strength of the interaction between gold and the neutral
PNA, which is much stronger than with the negatively
charged DNA.[11] The stronger tendency of PNA to adsorb
flat onto the gold particle through direct interaction of the
bases with the gold surface, rather than the preferred covalent
linkage through a thiol–gold bond, results in a significantly
thinner PNA layer and a strongly reduced PNA loading,
leading to poor colloidal stabilization. This, combined with
the lack of surface charge, results in inherently poor colloidal
stability of PNA-modified metal nanoparticles.[12] This prob-
lem has not been sufficiently overcome with the PNA-
functionalized AuNPs reported so far (see below). Herein, we
use a conjugation technique that allows, for the first time, the
synthesis of highly stable PNA-functionalized metal nano-
particles. We first functionalized the AuNPs with a stabilizing
surfactant commonly used for DNA functionalization[13] and,
in a second step, covalently attached the PNA to the AuNP
surface. A variety of linkers were used, showing the wide
applicability of the described method. Furthermore, we
demonstrated the binding specificity of the PNA-functional-
ized nanoparticles to complementary oligomers, and the
accessibility of the PNA strands, by directed assembly of
PNA-AuNP conjugates to PNA-functionalized gold sub-
strates. Scanning electron microscopy (SEM) imaging was
used to show the sequence-specific binding to unstructured
and micropatterned surfaces.
Functionalization of gold nanoparticles or surfaces with
PNAs requires the modification of the oligomers with
a terminal thiol group. Usually, either a cysteine or a similar
monothiol-containing linker is inserted at the amino end of
a PNA strand (Table 1, entry 1). Trithiol-capped DNA
oligomers have been shown to bind more efficiently to
[*] P. Anstaett, Prof. Dr. G. Gasser
Dr. A. M. Funston
Institute of Inorganic Chemistry, University of Zurich
Winterthurerstrasse 190, 8057 Zurich (Switzerland)
E-mail: gilles.gasser@aci.uzh.ch
School of Chemistry, Monash University
Clayton, Victoria 3800 (Australia)
[**] This work was supported by the Swiss National Science Foundation
(Professorship No. PP00P2_133568 to G.G.), the University of
Zurich (G.G.), the Australian Research Council (Australian Research
Fellowship to U.B. and Future Fellowship to A.M.F. (FT110100545))
and the Commonwealth Scientific and Industrial Research Organ-
ization (OCE Science Leader position to U.B.). This work was
performed in part at the Melbourne Centre for Nanofabrication, an
initiative partly funded by the Commonwealth of Australia and the
Victorian Government.
Homepage: http://www.gassergroup.com
Dr. Y. Zheng, T. Thai, Prof. Dr. U. Bach
The Melbourne Centre for Nanofabrication
151 Wellington Road, Clayton 3168, Victoria (Australia)
E-mail: udo.bach@monash.edu
Homepage: http://www.udobach.com
Dr. Y. Zheng, Prof. Dr. U. Bach
Commonwealth Scientific and Industrial Research Organization,
Materials Science and Engineering
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201209684.
Clayton South, Victoria 3169 (Australia)
T. Thai, Prof. Dr. U. Bach
Department of Materials Engineering, Faculty of Engineering,
Monash University
Clayton 3800, Victoria (Australia)
Angew. Chem. Int. Ed. 2013, 52, 4217 –4220
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4217

.
Angewandte
Communications
Table 1: Overview of the PNA and DNA conjugates with thiol linkers.
also synthesized (Table 1, entry 4). Importantly, the final step
of SPPS, which necessitates trifluoroacetic acid (TFA) to
cleave the PNA sequence from the solid support, also
removes the trityl protecting groups of the thiols. An excess
of a reducing agent such as tris(2-carboxyethyl)phosphine
(TCEP) is required in this step to prevent formation of
a disulfide bridge network.
Entry Linker
Sequence[a] Type Corre-
sponding
NP[b]
a
a’
1
2
3
4
9
PNA AuNP-4
PNA AuNP-5
PNA AuNP-6
With the thiol-containing PNA sequences made, we then
investigated their ability to form stable PNA-functionalized
AuNPs. Duy et al. reported the synthesis of PNA-modified
particles in highly concentrated Tween 20 solutions.[10b] How-
ever, we found that the particles prepared using this method
were not sufficiently stable for self-assembly purposes and
therefore, did not provide the anticipated advantages of PNA
on AuNPs. In the presence of Tween 20, we observed only
non-selective aggregation. If the Tween 20 concentration was
lowered, the PNA-functionalized NPs aggregated, even in the
absence of ions. For DNA-modified particles, it was demon-
strated that the additional attachment of 1-mercapto-6-hexyl-
tri(ethylene glycol) improves salt stability through steric
shielding without decreasing the hybridization efficiency.[18]
Because PNA, in contrast to DNA, lacks negative charges, 32-
mercapto-3,6,9,12,15,18,21-heptaoxadotriacontan-1-oic acid 3
was chosen for pre-functionalization of the AuNPs, because 3
incorporates steric shielding as well as additional electrostatic
stabilization (Scheme 2).[13a] The final particles were, there-
fore, envisioned to remain stable in the presence of ions as
well as under a wide range of pH values.[19]
5
6
10
11
4
5
7
8
12
13
PNA AuNP-7
DNA AuNP-8
[a] The sequences, denoted N!C terminus (PNA), or 5’!3’ (DNA),
respectively, are as follows: a: TCTCAGTATT; a’: AATACTGAGA;
[b] 40 nm AuNPs, functionalized with 3 and the respective oligonucleo-
tide.
AuNP surfaces and are displaced more slowly by other thiols
compared to mono- and cyclic dithiol-capped derivatives.[14]
As this stability towards subsequent ligand-exchange manip-
ulation is of interest, for example, for sensing applications, we
wanted to link the PNA strand in an analogous manner.
However, to the best of our knowledge, trithiol ligands have
not previously been used with solid-phase peptide/PNA
synthesis (SPPS).[15] To fill this gap, we designed a novel and
easy-to-prepare trithiol linker (1, Scheme 1). More specifi-
cally, the tribromide 2[16] was converted into the trithioether
1 by reaction with triphenylmethanethiol in the presence of
NaH. 1 could then be attached directly to different PNA
sequences using standard SPPS procedures (Table 1,
entry 2).[17] To assess the impact of the distance between the
PNA and the AuNPs, PNA sequences containing ethylene
glycol spacers between the PNA and the tripodal ligand were
Scheme 2. a) Illustrations of 3 and 6 and chemical formula of 3;
b) Overview of the two functionalization steps in the synthesis of PNA-
modified AuNPs with stabilizing 3 and a representative PNA.
Additionally, 3 was used as a linker molecule for PNA
(Table 1, entry 3). After in situ trityl protection of the thiol
group of 3, the carboxylate was coupled to the PNAs on the
solid support (see Supporting Information for more details).
This reaction did not proceed in the commonly used solvent
for such coupling reactions, namely dimethylformamide
(DMF). By switching to tetrahydrofuran (THF), which has
a significantly lower dielectric constant, the desired product
could be obtained.[20] We believe that the critical micelle
concentration (CMC) of the PEG-derivative 3 and/or its
Scheme 1. a) Synthesis of SPPS-compatible trithiol linker; b) Synthesis
of a trithiol-linker-containing PNA oligomer. Conditions: 1) Ph3CSH,
NaH, THF, 508C, 62 h, 63%; 2) Fmoc-SPPS: i) 20% piperidine in
DMF; ii) Fmoc-protected PNA-monomer (5 equiv), HATU (4.5 equiv),
DIPEA (10 equiv), 2,6-lutidine (10 equiv), DMF; 3) 1 (3 equiv), HATU
(2.75 equiv), DIPEA (13 equiv), 2,6-lutidine (13 equiv), procedure
repeated again; 4) TFA:phenol:triisopropylsilane 38:2:1 (v/v/v), TCEP
(1000 equiv). Fmoc=fluorenylmethoxycarbonyl; Bhoc=benzhydryloxy-
carbonyl; HATU=O-(7-azobenzotriazol-1-yl)-1,1,3,3-tetramethyluro-
nium hexafluorophosphate; DIPEA=N,N-diisopropylethylamine.
4218
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 4217 –4220

Angewandte
Chemie
activated ester is too low in DMF, rendering the carboxylate
and/or active ester site of the respective molecules inacces-
sible. For comparison, DNA-modified AuNPs were prepared
using a linker with a 15-mer thymine spacer. Such linkers are
often used for DNA-modified AuNPs because thymine has
a very low affinity towards gold.[21]
The thermal stability of the PNA-PNA, PNA-DNA, and
DNA-DNA hybrids in the absence of AuNPs was measured
using UV/Vis spectroscopy (see Supporting Information).
Consistent with previous reports,[22] the melting temperatures
of the PNA-PNA duplexes were significantly higher than for
the corresponding DNA-PNA or DNA-DNA duplexes,
indicative of their greater stability. This highlights a major
advantage of using PNA (as compared to DNA), because the
PNA-modified particles have the potential to create assem-
blies with either shorter inter-particle distances or stronger
hybridization than DNA-based systems, as fewer bases are
required to achieve sufficient binding. Moreover, the PNA is
not responsible for the stabilization of the particles and
therefore the colloidal stability is largely independent of the
oligomer length, unlike in the case of the common DNA-
functionalized particles.
Figure 1. SEM images of surfaces modified with a) complementary
PNA 7 and b) non-complementary PNA 12 after incubation with
AuNP-6. Scale bars=500 nm.
Table 2: Number of bound nanoparticles per mm2.[a]
Nanoparticle
Complementary 7
Non-Complementary 12
[mmꢀ2
]
[mmꢀ2
]
AuNP-4
AuNP-5
AuNP-6
AuNP-7
AuNP-8
38
39
39
43
26
3.6
5.1
2.8
5.5
1.1
The preparation of PNA-modified AuNPs is described in
detail in the Supporting Information. In brief, Tween 20,[23]
phosphate buffer (pH 7.4), and 3 were successively added to
a solution of AuNPs (40 nm diameter). After incubation for
12 hours, a sufficiently dense layer of 3 formed, protecting the
AuNPs. The thiolated PNA was then added, and the
incubation continued for another eight hours. Using this
procedure, similar surface charges were predicted for all
AuNPs, including the DNA modified AuNP-8.
The stability of particles functionalized with the different
PNAs was investigated using UV/Vis spectroscopy. Partial
aggregation results in a decrease of the extinction coefficient
and/or red shifting/broadening of the localized surface
plasmon resonance and can therefore easily be detected. In
1 mm phosphate buffer solutions, no changes in the UV/Vis
spectra could be observed after 20 hours for any of the AuNPs
(see Supporting Information for spectra). Similarly, in 10 mm
buffer solutions, only minor decreases of lmax were observed.
At a phosphate buffer concentration of 0.1m, aggregation and
precipitation were visible within seconds for AuNP-6 while
the lmax decreased less than 20% over the course of 15 hours
for all other PNA-functionalized AuNPs.
[a] As determined by particle counting in the SEM images.
densities of about 40 mmꢀ2 (i.e. 40 particles per mm2). These
observations along with the even distribution of particles on
the surface suggest that the average distance between NPs
adsorbed to the surface is dictated by the electrostatic
repulsion forces between the negatively charged NPs. Also
the trithiol-substituted AuNPs AuNP-5 and AuNP-7 reached
this threshold and are expected to withstand oxidation or
replacement by analytes if used, for example, for sensing
applications in future studies.[14] Moreover, the similar surface
coverage density of AuNP-5 with AuNP-6 demonstrates
effective PNA hybridization, despite the use of a linker that is
significantly shorter than the stabilizing surfactant 3. Notably,
the coverage density of the DNA-modified AuNP-8 is lower
compared to the PNA-modified AuNPs. This is in line with
the results of the thermal denaturation experiments; different
charge densities of the particles could also be a reason for this
result, but this is unlikely because of the identical preparation
conditions. Some non-complementary binding occurs in all
cases, although slightly less in case of AuNP-8.
Self-assembly experiments were carried out in the
absence of additional ions or surfactants. Gold-coated sub-
strates were modified with complementary PNA 7 or with
non-complementary PNA 12, as a negative control. After
incubating 5 mL of NP solution on the PNA-modified
substrates for four hours in a humidification chamber, the
substrates were washed four times with water. After drying,
SEM measurements were performed to determine the cover-
age densities of AuNPs on the substrate surface (see Figure 1
and Table 2). To the best of our knowledge, these are the first
images proving the sequence-specific self-assembly of PNA
modified nanoparticles.
DNA-based systems have been shown to be usable for
specific deposition of AuNPs onto patterned substrates,
a feature especially of interest in the context of nanoelec-
tronics.[24] To test if our PNA-modified AuNPs were capable
of fulfilling this task as well, we incubated AuNP-6 on
micropatterned substrates with lines of Au and silica. The
substrates were first modified with hydrophilic PEGylated
silanes or hydrophobic fluoroalkyl silanes and then with
PNAs 7 or 12, as described above for gold-coated substrates.
AuNP-6 assembles exclusively onto the areas with comple-
mentary PNA (Figure 2). Hence, the nanoparticles stick to
neither hydrophilic nor hydrophobic surfaces but rather only
to complementary oligonucleotide sequences. This further
demonstrates the rigidity and practicality of our PNA-based
system.
All PNA- and DNA-functionalized nanoparticles selec-
tively bound to the surfaces modified with complementary 7.
The PNA-modified AuNPs showed almost identical particle
Angew. Chem. Int. Ed. 2013, 52, 4217 –4220
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4219

.
Angewandte
Communications
[3] a) J. D. Le, Y. Pinto, N. C. Seeman, K. Musier-Forsyth, T. A.
Taton, R. A. Kiehl, Nano Lett. 2004, 4, 2343 – 2347; b) S. J. Tan,
M. J. Campolongo, D. Luo, W. Cheng, Nat. Nanotechnol. 2011, 6,
268 – 276; c) S. J. Barrow, A. M. Funston, D. E. Gꢀmez, T. J.
Davis, P. Mulvaney, Nano Lett. 2011, 11, 4180 – 4187; d) Y.
Zheng, C. H. Lalander, T. Thai, S. Dhuey, S. Cabrini, U. Bach,
Angew. Chem. 2011, 123, 4490 – 4494; Angew. Chem. Int. Ed.
2011, 50, 4398 – 4402; e) J. Zheng, P. E. Constantinou, C.
Micheel, A. P. Alivisatos, R. A. Kiehl, N. C. Seeman, Nano
Lett. 2006, 6, 1502 – 1504; f) J. Zhang, Y. Liu, Y. Ke, H. Yan,
Nano Lett. 2006, 6, 248 – 251.
[4] a) C. A. Mirkin, R. L. Letsinger, R. C. Mucic, J. J. Storhoff,
Nature 1996, 382, 607 – 609; b) R. Elghanian, J. J. Storhoff, R. C.
Mucic, R. L. Letsinger, C. A. Mirkin, Science 1997, 277, 1078 –
1081; c) J. Liu, Y. Lu, J. Am. Chem. Soc. 2003, 125, 6642 – 6643.
[5] a) B. M. Reinhard, S. Sheikholeslami, A. Mastroianni, A. P.
Alivisatos, J. Liphardt, Proc. Natl. Acad. Sci. USA 2007, 104,
2667 – 2672; b) C. Sçnnichsen, B. M. Reinhard, J. Liphardt, A. P.
Alivisatos, Nat. Biotechnol. 2005, 23, 741 – 745.
[6] a) Y. Cao, R. Jin, C. A. Mirkin, J. Am. Chem. Soc. 2001, 123,
7961 – 7962; b) S. Liu, Z. Zhang, M. Han, Anal. Chem. 2005, 77,
2595 – 2600.
[7] a) M. Krishnan, N. Mojarad, P. Kukura, V. Sandoghdar, Nature
2010, 467, 692 – 695; b) N. Mojarad, M. Krishnan, Nat. Nano-
technol. 2012, 7, 448 – 452.
[8] P. E. Nielsen, M. Egholm, R. H. Berg, O. Buchardt, Science 1991,
254, 1497 – 1500.
[9] S. Tomac, M. Sarkar, T. Ratilainen, P. Wittung, P. E. Nielsen, B.
Nordꢁn, A. Grꢂslund, J. Am. Chem. Soc. 1996, 118, 5544 – 5552.
[10] a) R. Chakrabarti, A. M. Klibanov, J. Am. Chem. Soc. 2003, 125,
12531 – 12540; b) J. Duy, L. Connell, W. Eck, S. Collins, R. Smith,
J. Nanopart. Res. 2010, 12, 2363 – 2369.
Figure 2. SEM images of micropatterned surfaces with a) complemen-
tary PNA 7 and PEG-modified silica, b) non-complementary PNA 12
and PEG-modified silica, c) complementary PNA 7 and fluoroalkyl-
modified silica, d) non-complementary PNA 12 and fluoroalkyl-modi-
fied silica, after incubation with AuNP-6. Scale bars=1 mm. The
magnification of the insets is double that of the parent image.
In summary, we have demonstrated the preparation of
stable PNA-modified particles through a novel approach
involving the use of a thiolated alkyl PEG carboxylate
surfactant. Standard, as well as novel, mono- and trithiol
linkers were found to be compatible with this approach. The
potential of these PNA-nanoparticles as new building blocks
for self-assembling systems was confirmed by synthesizing
particles that were able to self-assemble under additive-free
conditions, an endeavor which has, to the best of our
knowledge, not been directly shown with any other DNA/
PNA-based systems. The superior properties of PNA com-
pared to DNA make these particles highly interesting for
future studies. Ion-free conditions are needed for investiga-
tion of electrostatic phenomena[7] and are not compatible with
DNA hybridization. Sensing applications[4] could also benefit
from the increased binding strength of PNA hybrids. This
strength also allows for the investigation of optical phenom-
ena at unprecedentedly small inter-particle distances.[5]
[11] A. Gourishankar, S. Shukla, K. N. Ganesh, M. Sastry, J. Am.
Chem. Soc. 2004, 126, 13186 – 13187.
[12] R. Kanjanawarut, X. Su, Anal. Chem. 2009, 81, 6122 – 6129.
[13] a) T. R. Tshikhudo, Z. Wang, M. Brust, Mater. Sci. Technol. 2004,
20, 980 – 984; b) Y. Zheng, C. H. Lalander, U. Bach, Chem.
Commun. 2010, 46, 7963 – 7965.
[14] Z. Li, R. Jin, C. A. Mirkin, R. L. Letsinger, Nucleic Acids Res.
2002, 30, 1558 – 1562.
[15] T. Sakata, S. Maruyama, A. Ueda, H. Otsuka, Y. Miyahara,
Langmuir 2007, 23, 2269 – 2272.
[16] T. Axenrod, K. K. Das, H. Yazdekhasti, P. R. Dave, A. G. Stern,
Org. Prep. Proced. Int. 1997, 29, 358 – 361.
[17] N. Hꢃsken, G. Gasser, S. D. Kçster, N. Metzler-Nolte, Biocon-
jugate Chem. 2009, 20, 1578 – 1586.
[18] T. Stakenborg, S. Peeters, G. Reekmans, W. Laureyn, H. Jans, G.
Borghs, H. Imberechts, J. Nanopart. Res. 2008, 10, 143 – 152.
[19] M. R. Jones, R. J. Macfarlane, A. E. Prigodich, P. C. Patel, C. A.
Mirkin, J. Am. Chem. Soc. 2011, 133, 18865 – 18869.
[20] B. T. Houseman, M. Mrksich, J. Org. Chem. 1998, 63, 7552 –
7555.
[21] S.-J. Park, A. A. Lazarides, J. J. Storhoff, L. Pesce, C. A. Mirkin,
J. Phys. Chem. B 2004, 108, 12375 – 12380.
[22] A. M. Sosniak, G. Gasser, N. Metzler-Nolte, Org. Biomol. Chem.
2009, 7, 4992 – 5000.
Received: December 4, 2012
Revised: January 30, 2013
Published online: March 4, 2013
[23] a) It should be noted that in this procedure, Tween 20 was only
present during the deposition of the thiols to the AuNPs. During
this time, it provided additional stabilization. Afterwards, it was
not needed anymore and was removed. This is in contrast to the
particles described in Ref. [10b]; b) K. Aslan, V. H. Pꢁrez-Luna,
Langmuir 2002, 18, 6059 – 6065.
Keywords: gold nanoparticles · nanoparticles ·
peptide nucleic acids · self-assembly · solid-phase synthesis
.
[1] J. I. Cutler, E. Auyeung, C. A. Mirkin, J. Am. Chem. Soc. 2012,
134, 1376 – 1391.
[2] N. L. Rosi, D. A. Giljohann, C. S. Thaxton, A. K. R. Lytton-Jean,
M. S. Han, C. A. Mirkin, Science 2006, 312, 1027 – 1030.
[24] C. H. Lalander, Y. Zheng, S. Dhuey, S. Cabrini, U. Bach, ACS
Nano 2010, 4, 6153 – 6161.
4220 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 4217 –4220