actin-based metallic nanowires as bio-nanotransporters

LETTERS

692 nature materials | VOL 3 | OCTOBER 2004 | www.nature.com/naturematerials

The synthesis of conductive nanowires or patterned conductivenanoelements is a challenging goal for the future fabrication ofnanoscale circuitry1. Similarly, the realization of nanoscale

mechanics might introduce a new facet to the area ofnanobiotechnology. Here we report on the design of conductive andpatterned actin-based gold nanowires, and on the ATP-drivenmotility of the nano-objects.The polymerization of G-actin labelledwith Au nanoparticles, followed by the catalytic enlargement of thenanoparticles, yields gold wires (1–4 µm long and 80–200 nm high)exhibiting high electrical conductivity. The polymerization of theAu nanoparticle/G-actin monomer followed by the polymerizationof free G-actin, or alternatively the polymerization of the Au-nanoparticle-labelled G-actin on polymerized F-actin followedby the catalytic enlargement of the particles, yields patternedactin–Au wire–actin or Au wire–actin–Au wire nanostructures,respectively.We demonstrate the ATP-fuelled motility of the actin–Auwire–actin filaments on a myosin interface. These actin-basedmetallic wires and their nanotransporting funcionality introduce newconcepts for developing biological/inorganic hybrid devices.

Different metal wires consisting of silver2, gold3, platinum4,palladium5 or copper6 have been deposited on DNA templates by the reduction of metal ions or metal complexes associated with DNA or by the intercalation of functionalized Au nanoparticles into aDNA template, followed by the enhanced deposition of gold.Another approach to the synthesis of metal wires involved thedeposition of metals on protein fibrils7 or the deposition of metals insidemicrotubule fibrils8.The selective patterning of DNA templates by goldhas been examined by using RecA as an addressable protein for theselective shielding of DNA domains against metal deposition9. Here wereport on a novel method for the preparation of predesignedprotein–metal nanowires based on the use of G-actin as a molecularbuilding block. We demonstrate a stepwise polymerization ofAu-nanoparticle (NP)-functionalized G-actin monomer units andunlabelled G-actin units that gives metal-patterned protein filamentsafter catalytic metallization of the particles. Extensive research effortswere directed to understanding and imaging the ATP-driven motility ofthe myosin motor-protein on the F-actin filaments, or the sliding of F-actin filaments on myosin-modified surfaces10–16. The synthesis ofpatterned Au-nano-objects based on actin filaments is a new method of designing metallic nanotransporters, in which the actin filament

acts as the track counterpart of a complex with the myosin motorprotein. We demonstrate the ATP-driven motility of the actin-basednanotransporter on a myosin-modified surface.

Figure 1a depicts the preparation of the Au-metallic nanowires onthe F-actin template. First, an F-actin filament was prepared by thepolymerization of G-actin units in the presence of ATP, Mg2+ and K+ ions.The resulting filaments were then reacted with 1.4-nm Au-NPsmodified with a single N-hydroxysuccinimide active ester (fromNanoprobes). The resulting filaments were then dissociated to the G-actin monomer units by dialysing off the ATP, Mg2+ and K+ ions.The UV–visible spectrum of the resulting G-actin/NP hybrids shows anaverage loading of one Au-NP per G-actin monomer. Figure 1b showsthe atomic force microscopy (AFM) image of the individual unlabelledG-actin units, which were about 4 nm high, as expected11. Figure 1cshows the scanning transmission electron microscopy (STEM) image ofthe Au-NP-labelled G-actin monomer. The single Au-NP tethered to asingle actin monomer unit is clearly visible.

The resulting Au-NP-labelled G-actin was repolymerized, and the resulting filaments were subjected to catalytic enlargement of the Au-NP to generate a continuous metal wire (see Methods). Note that itis essential to follow this route.Attempts to couple the Au-NP directly tothe G-actin monomers (rather than to filaments) failed to form apolymerizable active unit, possibly because the covalent attachment ofthe Au-NP to the free monomer blocks the polymerization sites, whichare protected in the filament structure.Figure 1d shows the AFM imageof the Au nanowire obtained after the enhanced deposition of gold onthe Au-NP-functionalized protein filament. Wires with lengthscorresponding to 2–3 µm are obtained with an average height of80–100 nm, depending on the metal deposition time. Longer catalyticenhancement periods yield thicker nanowires.

Figure 2 shows the patterning of the actin-based metal wire by thepolymerization process. In Fig. 2a the wire structuring was initiated bythe primary polymerization of the Au-NP-functionalized G-actinmonomers to form the NP-functionalized F-actin filament as describedabove.The resulting polymer wire was then reacted with the unlabelledG-actin monomer units, a process that further elongated both ends ofthe Au-NP-derivatized actin filament. Catalytic deposition of gold onthe nanoparticles thus yields the patterned actin–metal-wire–actinfilament. Figure 2b shows the AFM image of the resulting patternedwire. The metallic wire (about 80–100 nm high) is linked at its two

Actin-based metallic nanowires as bio-nanotransportersFERNANDO PATOLSKY, YOSSI WEIZMANN AND ITAMAR WILLNER*Institute of Chemistry,The Hebrew University of Jerusalem,Jerusalem 91904, Israel*e-mail:[email protected]

Published online:12 September 2004; doi:10.1038/nmat1205

nmat1205-print 9/10/04 12:03 PM Page 692

© 2004 Nature Publishing Group


LETTERS

nature materials | VOL 3 | OCTOBER 2004 | www.nature.com/naturematerials 693

ends to bare protein actin filaments (about 4–6 nm high). Figure 2cshows the generation of a metal–actin–metal nanowire by thepredesigned controlled polymerization approach. The wire structurewas initiated by the primary polymerization of the unlabelled actinmonomers to form the actin filaments. The resulting filaments werethen reacted with the Au-NP-labelled actin monomers to yield thepatterned nanowire terminated at both ends with Au-NP-decoratedfilaments. The subsequent deposition of Au on the nanoparticlesyielded the Au wire–actin–Au wire system. Figure 2d shows an AFMimage of the resulting wire containing two gold sections (about120 nm high) separated by a bare protein filament section about 6 nmhigh. In most (more than 95%) of the observed nanowires, theattachment of the Au-NP-labelled actin monomers at the ends ofthe preformed unlabelled actin filament occurred at different rates.This led to gold wires of different lengths at the ends of the bare actinfilament (0.4–1.0 µm and 1.0–3.0 µm, respectively), an observationconsistent with the fact that polymerization of the G-actin unit on the(−)-end of the filament is slower than polymerization of the monomerunits at the (+)-end of the actin polymer17.Some non-labelled F-actinfilaments are also observed on the surface.

The conductivity of the Au wires generated on the actin templatewas then characterized by deposition of the metal wires across 1.0–1.2-µmgaps separating two gold electrodes. Figure 3a, b shows the AFM and

high-resolution scanning electron microscopy (HRSEM) images of theactin-based metallic nanowires deposited on the gap separating the twoAu surfaces (the bottom support is a glass surface).The enhancement ofthe NP-decorated filaments yields continuous metallic Au nanowireswith good metal conductivity. Figure 3c shows the I–V curve of thesystem, giving a resistance of about 300 Ω (for wires 70–150 nm inwidth). Reproducibility of the preparation of the conductive Aunanowires is excellent, and in ten experiments the resistance of thedeposited wires was 300 ± 40 Ω.Neither unlabelled actin filaments norunenhanced NP-labelled filaments yielded any conductivity(R > 1014 Ω). This is consistent with the fact that both the proteinfilaments and the Au-NP-decorated filaments are insulators.The use ofwider gaps between the gold electrodes (10–15 µm)—distances longerthan the metallic nanowire lengths—also does not yield conductivitybetween the gaps after the deposition of the metallic nanowires.These results verify that the observed conductivity is a result of metallicnanowires crossing the gap.

As stated before, one of the main aims of this work was the use ofactin as a building block for the creation of predesigned actin–Au-nanoblock patterned nanotransporters, where the unlabelled filamentsections act as engine parts (tracks) that move the Au nanostructures on motor protein-functionalized surfaces. Towards this end, weperformed a series of motility assays on myosin-coated surfaces, using

AuAu

AuAu

Au

Au

Au

Au

Au

Au

Au

AuAu

Au

Au

Au

Au

AuAu

AuAu

AuAu

Au

Au

Au

Au

Au

Au

Au

Au

Au

AuAu

Au

Au

AuAu

N

O

O

O O

ATP, NaCI, MgCl2polymerization

(–)-end

(+)-end

Dialysis

Polymerization Au3+, NH2OH

5 nm 00

1,000

1,000

2,000

2,000

0

20

40

60

80

00400

400 800

8001,200

1,200

4

2

a

b c d

Figure 1 The assembly of actin-based Au nanowires. a,Scheme for the synthesis of the nanowires.The shading on the final image indicates continuous wire formation.b,AFM image of the Au-NP-labelled G-actin monomer on a mica surface.c,STEM image of the Au-NP-labelled G-actin monomer.d,AFM image of the Au nanowire generated on the actintemplate.All dimensions in b and d are in nanometres.




LETTERS

694 nature materials | VOL 3 | OCTOBER 2004 | www.nature.com/naturematerials

the actin–metal–actin hybrids as nanotransporters. In the motilityassay,a flow cell consisting of two glass slides separated by an O-ring wasuseed. The bottom slide consisted of a nitrocellulose-modified glassslide on which heavy meromyosin was deposited (0.2 mg ml−1, for20 min at 25 °C). The upper and lower slides were blocked with bovineserum albumin to prevent non-specific adsorption (by passing asolution containing 1 mg ml−1 bovine serum albumin through the cell).Actin–Au–actin filaments were stabilized by phalloidin by thepolymerization of the Au-NP-labelled actin in the presence of phalloidin(1:1 actin/phalloidin molar ratio), followed by polymerization in thepresence of free actin and phalloidin (1:1). Subsequently, the Au-NPswere enlarged to Au-nanowire, and the phalloidin-stabilized actin–Au-nanoblock–actin filaments were allowed to interact with the myosininterface (for 10 min). Unbound filaments were washed away with abuffer solution (25 mM KCl, 1.0 mM EGTA, 4.0 mM MgCl2, 0.1 mMdithiothreitol, 10 mM imidazole, pH 7.3). Before the addition of ATP(which induces filament movement on surfaces), none of the metalnanowires was observed to slide on the surface (the nanowires weretightly bound to the surface by the F-actin–myosin interactions).The addition of the assay buffer containing 1.0 mM Mg-ATP triggeredthe sliding of the metal–actin rods on the surface. Figure 4 shows fourconsecutive images of the same frame taken at intervals of 5 s.We foundthat about 30–40% of about 100 of the Au-functionalized filamentsmoved on the surface. From the distances that the rods were

transported, the motility rate of the filaments was 250 ± 50 nm s−1. In afurther experiment, proteinase K, a hydrolytic enzyme that degradesproteins,was added to the cell.This resulted in the blocking of nanowiremotility, presumably due to the degradation of the actin filaments andthe myosin motor protein. Interestingly, the actin–Au nanowire–actinsystem slid on the surface at a substantially slower speed than thatreported for the sliding rate of actin on heavy meromyosin as a surface-confined motor:up to 4µms−1 (refs 17, 18).Thus,our nanotransportersmove on the surface at a speed that is about 16-fold less than that ofunlabelled actin.Although the origin of the slower motility of actin–Auwire–actin is not fully understood, several effects might have led to the observed phenomenon. Previous studies have emphasized that thecomposition and structure of the base myosin motor interfaceinfluences the actin motility. Inherent features of the motor interface,such as the coverage of the myosin layer, different conformations of themyosin–actin complex and the partial oxidation of the myosin layer,were found to affect the actin motility.In the present system,the need toslide the ‘heavy’ Au element by the actin filaments might have been acause of the lower motility. Previous studies that used actin filamentsmodified by the addition of a polystyrene bead at the barbed end (beaddiameter 1 µm) claimed similar motilities to that of unmodified actin,results that stand against our latter explanation19. Nonetheless, oneshould note that, in addition to the need to slide the ‘heavy’nanowire bythe actin filament, other interactions between the myosin amino acids

Au

AuAu

Au

Au

Au

Au

AuAu

AuAu

Au

Au

Au

Au

AuAu

AuAu

Au

Au

Au

Au

Au

Au

Au

Au

AuAu

Au

Au

Au

Au

Au

Au

Au

Au

Au

AuAu

Au

Au

Au

Au

Au

Au

a

c d

Polymerization

Au

Au

Polymerization

Au3+, NH2OH

Au3+, NH2OH

b

00 1,000

1,000

2,000

2,000

3,000

3,000

4,000

4,000

80

00

1,000

1,000

2,000

2,000

Gold F-actin

Pure F-actin

Figure 2 Synthesis and structure of patterned actin-based Au nanowires. a,Scheme for the synthesis of the patterned actin–Au-nanowire-actin filaments.b,AFM image of theresulting filament actin–Au nanowire–actin.c,Scheme for the synthesis of the Au nanowire–actin–Au nanowire filament.d,AFM image of the resulting patterned Au nanowire–actin–Aunanowire filament.All dimensions in b and d are in nanometres.




LETTERS

nature materials | VOL 3 | OCTOBER 2004 | www.nature.com/naturematerials 695

(such as lysine residues)20 and the Au surface might add ‘frictioninteractions’that retard the motility of the system.

METHODS

METHOD FOR PREPARATION OF CONDUCTIVE AU NANOWIRESThe polymerization of the Au-NP-labelled G-actin was induced for 1 h in a 2 mM HEPES buffer solution,

pH 8.0, containing 50 mM KCl, 2 mM MgCl2, 0.25 mM dithiothreitol and 2 mM ATP in the presence of

phalloidin at a 0.5 molar ratio to G-actin. The resulting actin filaments were separated with a Centricon

filtration device (cut-off 100,000 Da). The deposition of gold on the resulting filaments was

accomplished by treatment of the filaments for 2 min in a 2 mM HEPES buffer solution containing

0.15 mM HAuCl4 and 0.15 mM hydroxylamine, to yield nanowires 80–150 nm in height. The deposition

of the Au nanowires on the gap separating the two electrodes was accomplished by placing a 2-µl drop on

the surface and allowing it to dry. The region of the deposited drop was scanned by AFM and only gaps

that included the Au wires were subjected to the conductivity measurements.

Received 21 April 2004; accepted 23 April 2004; published 12 September 2004.

References1. Niemeyer, C. M. Nanoparticles, proteins, and nucleic acids: biotechnology meets materials science.

Angew. Chem. Int. Edn 40, 4128–4158 (2001).

2. Braun, E., Eichen, Y., Sivan, U. & Ben-Yoseph, G. DNA-templated assembly and electrode attachment

of a conducting silver wire. Nature 391, 775–778 (1998).

3. Patolsky, F., Weizmann, Y., Lioubashevski, O. & Willner, I. Au-nanoparticle nanowires based on DNA

and polylysine templates. Angew. Chem. Int. Edn 41, 2323–2327 (2002).

4. Richter, J. et al. Construction of highly conductive nanowires on a DNA template. Appl. Phys. Lett. 78,

536–538 (2001).

5. Mertig, M. et al. DNA as a selective metallization template. Nano Lett. 2, 841–844 (2002).

6. Manson, C. F. & Wooley, A. T. DNA-templated construction of copper nanowires. Nano Lett. 3,

359–363 (2003).

7. Scheibel, T. et al. Conducting nanowires built by controlled self-assembly of amyloid fibers and

selective metal deposition. Proc. Natl Acad. Sci. USA 100, 4527–4532 (2003).

8. Reches, M. & Gazit, E. Casting metal nanowires within discrete self-assembled peptide nanotubes.

Science 300, 625–627 (2003).

9. Keren, K. et al. Sequence-specific molecular lithography on single DNA molecules. Science 297, 72–75

(2002).

10.Yildiz, A. et al. Myosin V walks hand-over-hand: single fluorophore imaging with 1.5-nm localization.

Science 300, 2061–2065 (2003).

11.De La Cruz, E. M. The kinetic mechanism of myosin V. Proc. Natl Acad. Sci. USA 96, 13726–13731

(1999).

12.De La Cruz, E. M. Kinetic mechanism and regulation of myosin VI. J. Biol. Chem. 276, 32373–32381

(2001).

13.Mehta, A. Myosin learns to walk. J. Cell Sci. 114, 1981–1998 (2001).

14.Yanagida, T. & Iwane, A. H. A large step for myosin. Proc. Natl Acad. Sci. USA 97, 9357–9359 (2000).

15.Vale, R. D. & Milligan, R. A. The way things move: looking under the hood of molecular motor

proteins. Science 288, 88–95 (2000).

16.Rief, M. et al. Myosin-V stepping kinetics: a molecular model for processivity. Proc. Natl Acad. Sci.

USA 97, 9482–9486 (2000).

17.Sellers, J. R. et al. Myosin-specific adaptations of the motility assay. Methods Cell Biol. 39, 23–49

(1993).

18.Le Goff, L., Hallaschek, O., Frey, E. & Amblard, F. Tracer studies on F-actin fluctuations. Phys. Rev. Lett.

89, 258101 (2002).

19.Suzuki, N. et al. Preparation of bead-tailed actin filaments: estimation of the torque produced by the

sliding force in an in vitro motility assay. Biophys. J. 70, 401–408 (1996).

20.Doron, A., Katz, E. & Willner, I. Organization of Au colloids as monolayer films onto ITO glass

surfaces—application of the metal colloid films as base interfaces to construct redox-active

monolayers. Langmuir 11, 1313–1317 (1995).

AcknowledgementsWe thank J. F. Hainfeld, Biology Department, Brookhaven National Laboratory, USA, for providing us

with the STEM image of the Au-nanoparticle-modified G-actin. This research is supported by the

German–Israeli Foundation (GIF).

Correspondence and requests for materials should be addressed to I.W.

Competing financial interestsThe authors declare that they have no competing financial interests.

00 1,000

1,000

2,000

2,000

3,000

3,000

200

250 nm

I (mA)

E (mV)

0.03

0.02

0.01

00

–0.03

–0.02

–0.01

2 4 6 8 10–10 –8 –6 –4 –2

a

b

c

Figure 3 Electrical conductivity of an actin-based Au nanowire. a,b,AFM (a) andHRSEM (b) images of the actin-based Au wire that bridges the gap between electrodes.All dimensions in a are in nanometres.c, I–V curve corresponding to the actin-based Au nanowire.

1 µm 1 µm 1 µm 1 µm

a b c d

Figure 4 ATP-fuelled motility of the actin–Au nanoblock–actin filaments on amyosin interface. Four consecutive microscopy images, taken at intervals of 5 s,areshown,corresponding to the motility of the actin–Au nanoblock–actin nanotransporterassociated with the myosin interface after the addition of ATP.




actin-based metallic nanowires as bio-nanotransporters

Documents