Preferential End-to-End Assembly of Gold Nanorods by Biotin-Streptavidin Connectors

Article abstract of DOI:10.1021/ja037969i

A major challenge in nanoscience and nanotechnology is the rational assembly of nanoscale objects. Here we report that gold nanorods, aspect ratio 18, can be functionalized with a biotin disulfide, and subsequent addition of streptavidin links the rods together in an end-to-end manner much more often than expected. Copyright

Full text of DOI:10.1021/ja037969i

Published on Web 10/23/2003
Preferential End-to-End Assembly of Gold Nanorods by Biotin-Streptavidin
Connectors
†
,†
K. K. Caswell, James N. Wilson, Uwe H. F. Bunz,* and Catherine J. Murphy*
Department of Chemistry and Biochemistry, UniVersity of South Carolina, 631 Sumter Street,
Columbia, South Carolina 29208
Received August 16, 2003; E-mail: murphy@mail.chem.sc.edu
The synthesis of inorganic materials with dimensions on the
1-7
nanometer scale has greatly improved in the last 10 years. Future
2
applications in nanoscale electronic devices, magnetic data storage
systems,3 optical devices,8 and biological sensing
9-11
require
monodisperse semiconductor or metallic materials of controlled size
and shape. The next major challenge in the field is to assemble
these nanoscale objects into functional devices. Self-assembly
procedures governed by surface chemistry are already proving to
3
be useful in this respect. Here we report that gold nanorods,
surface-functionalized with biotin, can be assembled preferentially
in an end-to-end fashion upon the addition of a streptavidin linker
(Figure 1). These results provide evidence that the chemical
reactivity of nanomaterials is shape-dependent and may be useful
in construction of future nanoscale assemblies.
Biotin, a small molecule, is well-known to bind tightly to the
protein streptavidin, with four biotins binding per protein.12 The
synthesis of the biotin disulfide (Figure 1) was analogous to that
reported by Connelly et al.13 (Supporting Information). We also
employed a commercially available biotin disulfide, EZ-Link Biotin
HPDP (Pierce; Figure 1). Disulfides are known to bind gold surfaces
to produce self-assembled monolayers very similar to those obtained
with thiols, with some improvement in compositional control for
Figure 1. Scheme showing the assembly of gold nanorods (golden ovals)
by surface functionalization with the biotin disulfide (red), and subsequent
addition of streptavidin (blue), to produce aggregates of nanorods. The
chemical structures of the two biotin disulfides are also shown, homemade
mixed thiol species.1
4-16
Other groups
13,17,18
have used biotin-
streptavidin (or the related protein avidin) linkers to assemble
inorganic nanospheres into three-dimensional (3D) structures, and
Mann’s group has shown that gold nanorods coated with DNA can
be similarly assembled into 3D aggregates by the addition of a
complementary linker DNA.19 Thus, the spacing of the nanoparticles
is governed by the size of the linker molecule.
We synthesized gold nanorods of aspect ratio 18 according to
our seed-mediated growth approach in water in the presence of a
shape-directing surfactant, cetyltrimethylammonium bromide
(
top) and commercial EZ-Link Biotin-HPDP (bottom).
-4
functionalized by adding 1 mL of a 1 × 10 M stock solution of
EZ-Link Biotin-HPDP. This mixture was stirred, shaken, sonicated,
and allowed to set overnight. Nanoparticles with shapes other than
rods and excess unbound biotin disulfide were purified away by a
second round of centrifugation. This batch was split, and one-half
was coated with a saturating amount of streptavidin. Excess,
unbound streptavidin was centrifuged away. Aliquots of 2 µL, at a
time, of the streptavidin-coated rods were titrated into the biotin-
coated rods. This was mixed and repeated several times. A final
centrifugation was then employed to separate the linked rods from
unlinked.
,20,21
(
CTAB).7
The nanorods thus prepared are covered with at least
22
a bilayer of CTAB and are stable and soluble in water. High-
resolution crystallography on individual nanorods has shown that
our nanorods are pentatetrahedral twins, with the {111} faces of
gold at the ends, and {100} faces along the length of the rods.23
We currently postulate that the CTAB preferentially binds to the
The addition of streptavidin to the biotin-coated gold nanorods
produced an unexpectedly high proportion of end-to-end-linked gold
nanorods (Figure 2), which were separated by 4-5 nm spacingss
{
{
100} faces, along the length of the rods, compared to the end
111} Au faces, due to the size of the CTAB headgroup,
22,23
3
the size of a streptavidin molecule (∼4.5 × 4.5 × 5.8 nm )sand
although the hydrocarbon tail of CTAB also plays a role in the
growth of gold nanorods via a postulated “zipping” mechanism.24
Binding of the homemade biotin disulfide, in ethanol, to the gold
at angles approximately what one would expect for the four biotin
25
binding sites on the protein projected into two dimensions. Control
experiments in which no streptavidin, or no biotin, were used gave
statistically distinct results (Figure 3). Both biotin derivatives
worked, but the commercial linker was more water-soluble and
easier to work with.
An average of multiple batches revealed that end-to-end linkages
could be realized ∼30% of the time while other binding events
comprised ∼34%. We defined binding as any rod within 5 nm of
another; anything but clear end-to-end binding was called
-
6
nanorods was accomplished by adding 2.5 mL of 3.1 × 10
M
biotin disulfide to 50 mL of the mixed solution of rods, stirring,
sonicating, and allowing to set overnight. For the EZ-Link biotin
HPDP, a mixed solution of rods (rehydrated to 14 mL), which had
been centrifuged once to remove the majority of spheres, was
†
Present address: School of Chemistry and Biochemistry, Georgia Institute of
Technology, Atlanta, GA 30332.
13914
9
J. AM. CHEM. SOC. 2003, 125, 13914-13915
10.1021/ja037969i CCC: $25.00 © 2003 American Chemical Society

C O MMU N I C A T I O N S
contradictory to the observation of the closest-packed face of gold
being at the ends of the nanorods23 but is consistent with
Fitzmaurice’s notion of full biotin coverage blocking streptavidin
sites.13 Other workers have shown that the disordered sites at
metallic interfaces yield more disordered molecular monolayers at
those sites;27 thus, it is possible that the biotin disulfide localizes
preferentially to such sites at the pentahedral twin boundaries on
the ends of the rods. Elemental analyses of biotinylated gold
nanorods, of which unfortunately only small quantities are available,
are consistent with a monolayer or less of the biotin disulfide on
the surface, based on %Au and %S in the samples, and thus cannot
distinguish between the hypotheses at present. Nonetheless, our
current results provide evidence that chemical linkages can organize
nanorods in a nonrandom fashion, which may be exploited for the
assembly of higher-order arrays of nanomaterials.
Supporting Information Available: Details of the synthesis and
characterization of the homemade biotin disulfide (PDF). This material
is available free of charge via the Internet at http://pubs.acs.org.
Figure 2. Transmission electron micrographs of gold nanorods, surface-
derivatized with either biotin disulfide, after addition of streptavidin. Scale
bars ) 100 nm (a), 20 nm (b), 100 nm (c), and 500 nm (d).
References
(
(
(
(
1) Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chem. Soc. 1993,
15, 8706-8715.
2) Hu, J. T.; Odom, T. W.; Lieber, C. M. Acc. Chem. Res. 1999, 32, 435-
45.
3) Sun, S.; Murray, C. B.; Weller, D.; Folks, L.; Moser, A. Science 2000,
87, 1989-1992.
4) Puntes, V. F.; Krishnan, K. M.; Alivisatos, A. P. Science 2001, 291, 2115-
117.
1
4
2
2
(
(
(
(
5) Wu, Y.; Yang, P. J. Am. Chem. Soc. 2001, 123, 3165-3166.
6) Peng, Z. A.; Peng. X. J. Am. Chem. Soc. 2002, 124, 3343-3353.
7) Murphy, C. J.; Jana, N. R. AdV. Mater. 2002, 14, 80-82.
8) Maier, S. A.; Kik, P. G.; Atwater, H. A.; Meltzer, S.; Harel, E.; Koel, B.
E.; Requicha, A. A. G. Nature Materials 2003, 2, 229-232.
9) Bruchez, M., Jr.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P. Science
1998, 281, 2013-2016.
Figure 3. Summary of numbers of end-to-end and end-to-side linkages
observed in transmission electron micrographs for experiments with (1) rods
only, (2) rods, and streptavidin only, (3) biotinylated rods (commercial
linker) and streptavidin. In each case 500 rods were counted and the
interparticle distances measured.
(
(10) Chan, W. C. W.; Nie, S. Science 1998, 281, 2016-2018.
(
11) Storhoff, J. J.; Elghanian, R.; Mucic, R. C.; Mirkin, C. A.; Letsinger, R.
L. J. Am. Chem. Soc. 1998, 120, 1959-1964.
(
12) Green, N. M. AdV. Protein Chem. 1975, 29, 85-133.
“
end-to-side.” Crudely, one would expect on the basis of relative
(13) Connolly, S.; Cobbe, S.; Fitzmaurice, D. J. Phys. Chem. B 2001, 105,
2222-2226.
surface areas that side-by-side binding would predominate over end-
to-end binding by 20-fold, should biotinylation and subsequent
streptavidin linking be random. There were rather large “islands”
on the TEM grids in control experiments (1 and 2 from Figure 3)
where distinctly liquid crystalline side-by-side behavior was
observed, as documented previously, but not in 3.26 Control
experiments with bovine serum albumin, which does not bind to
biotin, produced results strikingly similar to 2 of Figure 3.
(
14) Porter, L. A.; Ji, D.; Westcott, S. L.; Graupe, M.; Czernuszewicz, R. S.;
Halas, N. J.; Lee, T. R. Langmuir 1998, 14, 7378-7386.
15) Gronbeck, H.; Curioni, A.; Andreoni, W. J. Am. Chem. Soc. 2000, 122,
(
3839-3842.
(16) Shon, Y. S.; Mazzitelli, C.; Murray, R. W. Langmuir 2001, 17, 7735-
7741.
(
(
(
(
17) Sastry, M.; Lala, N.; Patil, V.; Chavan, S. P.; Chittiboyina, A. G. Langmuir
1998, 14, 4138-4142.
18) Mann, S.; Shenton, W.; Li, M.; Connolly, S.; Fitzmaurice, D. AdV. Mater.
2000, 12, 147-150.
19) Dujardin, E.; Hsin, L.-B.; Wang, C. R. C.; Mann, S. Chem. Commun.
001, 1264-1265.
20) Jana, N. R.; Gearheart, L.; Murphy, C. J. J. Phys. Chem. B 2001, 105,
065-4067.
(21) Busbee, B. D.; Obare, S. O.; Murphy, C. J. AdV. Mater. 2003, 15, 414-
16.
2
The mechanism behind the higher-than-expected proportion of
end-to-end linkages of gold nanorods is still under investigation.
There are two current hypotheses we are exploring: (i) the biotin
disulfide was unsuccessful at displacing the CTAB bound to the
length of the gold nanorods, and thus the biotin preferentially bound
to the {111} ends of the rods, resulting in preferential end-to-end
linkages upon the addition of streptavidin; the bilayer structure of
CTAB on the nanorod surface may make the surfactant less easy
4
4
(
22) Nikoobakht, B.; El-Sayed, M. A. Langmuir 2001, 17, 6368-6374.
(23) Johnson, C. J.; Dujardin, E.; Davis, S. A.; Murphy, C. J.; Mann, S. J.
Mater. Chem. 2002, 12, 1765-1770.
(
24) Gao, J.; Bender, C. M.; Murphy, C. J. Langmuir 2003, 19, 9065-9070.
(25) Weber, P. C.; Ohlendorf, D. H.; Wendoloski, J. J.; Salemme, F. R. Science
1989, 243, 85-88.
(
26) Jana, N. R.; Gearheart, L. A.; Obare, S. O.; Johnson, C. J.; Edler, K. J.;
Mann, S.; Murphy, C. J. J. Mater. Chem. 2002, 12, 2909-2912.
27) Aizenberg, J.; Black, A. J.; Whitesides, G. M. Nature 1998, 394, 868-871.
22,24
to displace than one might expect;
(ii) the biotin did indeed
(
bind all over the rods, but streptavidin is sterically constrained to
bind to the ends of the rods. This second hypothesis is somewhat
JA037969I
J. AM. CHEM. SOC.
9
VOL. 125, NO. 46, 2003 13915