NANO LETTERS
Self-Assembled Peptide Nanotubes Are Uniquely Rigid Bioinspired Supramolecular Structures
2005 Vol. 5, No. 7 1343-1346
Nitzan Kol,†,⊥ Lihi Adler-Abramovich,‡,⊥ David Barlam,§ Roni Z. Shneck,| Ehud Gazit,‡ and Itay Rousso*,† Department of Structural Biology, Weizmann Institute of Science, RehoVot 76100, Israel, Department of Molecular Biology and Biotechnology, Tel-AViV UniVersity, Tel-AViV 69978, Israel, Department of Mechanical Engineering and Department of Materials Engineering, Ben-Gurion UniVersity of the NegeV, Beer-SheVa 84105, Israel Received March 27, 2005; Revised Manuscript Received May 15, 2005
ABSTRACT We recently presented a novel class of self-assembled diphenylalanine-based peptide nanotubes. Here, for the first time, we present their mechanical properties, which we directly measured through indentation type experiments using atomic force microscopy. We find that the averaged point stiffness of the nanotubes is 160 N/m, and that they have a correspondingly high Young’s modulus of ∼19 GPa, as calculated by finite element analysis. This high value places these peptide nanotubes among the stiffest biological materials presently known, making them attractive building blocks for the design and assembly of biocompatible nanodevices.
The self-assembly of simple building blocks into wellordered structures on the nanoscale is a central theme in bottom-up nanotechnology. Bio-derived building blocks are of special interest due to their biocompatibility, advanced molecular recognition elements, and versatile chemical properties. Peptide-based nanostructures have gained much attention in recent years due to their relatively simple, yet biocompatible and modifiable, structure.1-12 We recently described a novel class of peptide nanotubes constructed from much simpler building blocks than those commonly used. These tubes are formed by the self-assembly of the diphenylalanine core-recognition motif of the Alzheimer’s diseaseassociated β-amyloid polypeptide.13 These biocompatible and water-soluble tubes form under mild conditions and are inexpensive and easy to manufacture.13,14 The peptide nanostructures assemble very efficiently and have remarkable thermal and chemical stability. The chemical nature of the peptide nanostructures is compatible with a variety of chemical and biochemical modifications, including specific functionalization and decoration with biological molecules. * Corresponding author. Tel: 972-8-9343479; Fax: 972-8-9344136; E-mail:
[email protected]. † Weizmann Institute of Science. ‡ Tel-Aviv University. § Department of Mechanical Engineering, Ben-Gurion University of the Negev. | Department of Materials Engineering, Ben-Gurion University of the Negev. ⊥ N.K. and L.A. contributed equally to this work. 10.1021/nl0505896 CCC: $30.25 Published on Web 06/08/2005
© 2005 American Chemical Society
The self-assembled peptide nanotubes appear to be rather stiff supramolecular entities, as evidenced by their remarkable persistent length. These high aspect ratio nanostructures revealed a persistent length of over 10 microns13 (Figure 1A), indicating that the peptide structures should be suitable for numerous future applications in self-assembled nanodevices. Yet, the mechanical parameters of the tubes were not determined explicitly. Here, we probe the mechanical properties of single peptide nanotubes by indentation type experiments conducted with an atomic force microscope (AFM) tip. The AFM is capable of achieving a combination of low penetration depths and high lateral resolution, which makes it a powerful and attractive tool to measure the mechanical properties of nanoscale biological structures.15-19 Indeed, recently the AFM has been used to measure the mechanical properties of microtubules.20 Peptide nanotubes were deposited on a mica surface and imaged using AFM (Figure 1A,B). The structure of the deposited nanotubes under these conditions was consistent with previous electron microscopy analysis,7 yet it allowed direct topographical measurement of the self-assembled tubular structures. The diameters of the nanotubes varied from 80 to 300 nm, as indicated by their cross-sectional profiles (a cross-sectional profile of a typical tube is shown in Figure 1C). Due to convolution between the AFM tip and the sample, measurement of the diameter of the nanotube in the lateral direction is less accurate than in the vertical direction, and the nanotube’s lateral diameter is consequently
Figure 2. Measuring the point stiffness of the peptide nanotube. (A) Typical force-distance curve of a peptide nanotube in reference to the cantilever deflection curve. Both curves were shifted along the z-axis to set the tip-sample contact point to a distance of zero. Each curve represents the average of 100 consecutive measurements. (B) The 100 individual measured point stiffness values obtained for a nanotube during a single experiment against experiment number (count). These were averaged to calculate the graph shown in (A). The horizontal solid line represents the averaged stiffness. (C) Histogram of the 100 measured nanotube spring constants derived from the consecutive force-distance curves, with Gaussian curve fitted. Analysis of the narrow distribution of the individual measured spring constant shows that the tube did not undergo irreversible deformation during the indentation measurements.
Figure 1. Peptide nanotube shape and dimensions. (A) Lowresolution AFM image of a nanotube deposit on freshly cleaved mica in air (scan area 10×10 µm, 512×512 pixels). (B) Highresolution AFM image of a nanotube (scan area 1×1 µm, 512×512 pixels). (C) The cross sectional profile of a nanotube revealing its 100 nm diameter.
larger. To measure the mechanical properties of the peptide nanotube, the tip was positioned at the center of a single nanotube surface, and force-distance curves were acquired at the same position. One such force-distance curve for a peptide nanotube is shown in Figure 2A, together with the corresponding cantilever deflection curve. Both curves are shifted along the z-distance axis to set the tip-sample contact point to zero. The cantilever deflection graph is obtained by measuring a force-distance curve for mica. Relative to the cantilever’s stiffness, mica is an infinitely stiff material that the AFM probe cannot penetrate or deform in any way. Cantilever deflection is then simply converted to loading force by multiplying the deflection reading by the cantilever spring constant. The difference between the z-displacement of the peptide nanotube and the cantilever deflection for a 1344
given loading force represents the indentation of the nanotube by the AFM probe. Before focusing on the quantitative interpretation of the indentation experiments, we verified that, during each force measurement comprising approximately 100 force-distance curves, the sample does not undergo plastic deformation. The mean calculated point stiffness derived from these 100 curves was plotted as a function of the measurement count (Figure 2B) and as a histogram to which a Gaussian curve was fitted (Figure 2C). During each experiment, the point stiffness varied slightly between individual force-distance curves, with a standard deviation value of ∼3%. The narrow distribution of point stiffness values refutes the possibility of plastic deformation during measurement. Figure 3 shows the distribution of the measured point stiffness values acquired at 24 different sites spread over several nanotubes. Our dataset includes tubes with diameters ranging from 150 to 300 nm. However, the diameter of the nanotubes had no apparent effect on their stiffness values, as indicated by Figure 3 inset. The measured point stiffness (kmeas) is comprised of the stiffness constants of the cantilever (kcan) and the nanotubes (knano). Assuming that the Nano Lett., Vol. 5, No. 7, 2005
Figure 3. Histogram distribution of the measured stiffness of the peptide nanotube. Each value was calculated as the average of approximately 100 force-distance curves obtained from different sites on several tube surfaces. The average stiffness derived from all of the measurements is 32 N/m (SD ) 3, n ) 24). The inset shows a scatter plot of the measured point stiffness as a function of the diameters of the tubes.
Figure 4. Finite element simulation for indentation of a tube. The deformed shape of a thick (t/Rext ) 0.8) and a thin (t/Rext ) 0.2) wall tubes are shown in (A) and (B), where t is the wall thickness and Rext is the external radius of the tube. The contours of the tubes prior to the indentation are shown by blue lines.
nanotube and the cantilever act as two springs oriented in a series, the point stiffness of the nanotubes can be calculated using the following relation: knano )
kcan‚kmeas kcan - kmeas
Using the above equation and an averaged measured value of 32 N/m (SD ) 3) for kmeas, we calculated that the average stiffness of the peptide nanotubes is 160 N/m. To estimate the material property of the tube from tube stiffness, we assume that the mechanical behavior of the tube can be described as linear elastic, which is a good approximation for solids under small strains. In this framework the deformation of the tube was calculated by the finite element method. Figure 4A shows the deformed and undeformed shapes of the model tube. Intense local deformation is observed under the indenter. The ratio of the model tube’s inner and outer radii to its length was 1:5:20, according to the estimated real dimensions of the tube. The indenter was Nano Lett., Vol. 5, No. 7, 2005
Figure 5. Stiffness of elastic tubes under indenters as a function of the wall thickness t. The finite element results of our models with E ) 1 MPa, Poisson’s ratio ν ) 0.3, tube length l ) 200 nm, and external radius Rext ) 50 nm (blue line). Our simulation results are compared with two asymptotic results for thin tubes: an analytic approximation for finite length thin shells claimed to be valid for t/R < 0.0124 (magenta line): F/D ) πEt/{l3/6R3 + 0.802‚[3‚(1 V2]3/4 [R/t]3/2} (R is the average radius of the tube; D is the indentation) and the numerically fitted results of de Pablo et al.20 for t/R
