Preparation of Pep11

The Pep11 powder was dissolved in 10 mM PBS (phosphate 10 mM, NaCl 20 mM, pH 7.0) to a certain concentration. Then, the solutions were centrifuged on a centrifuge (Hitachi, Japan) at a rate of 12,000 rpm for 20 min to eliminate the possible aggregates, and the supernatant was immediately used in the experiment.

Preparation of actin

G-Actin was dissolved to a concentration of 10 mg/ml in Tris-HCl buffer (pH 8.0, 5 mM) containing 0.2 mM CaCl₂, 0.2 mM ATP, and 5% (w/v) sucrose. Then, the G-actin solution was quickly frozen in liquid nitrogen and stored at -80℃. Before experiments, the prepared G-actin solution was diluted to 0.4 mg/ml with general actin buffer (GB), which consisted of 5 mM Tris-HCl (pH 8.0), 0.2 mM CaCl₂, 0.2 mM ATP, and 0.5 mM DTT. The diluted G-actin solution was incubated on ice for 60 min to depolymerize possibly existing actin oligomers, which might form during storage. After that, the solution was centrifuged in a 4℃ microfuge tube at 14,000 rpm for 30 min to remove any large actin crystals generated by the snap-freezing process. Then, the supernatant was transferred to a new microfuge tube and further diluted with GB to a desired working concentration for experiments. F-Actin was polymerized from G-actin by addition of polymerization buffer (PB), which was made up of 100 mM Tris-HCl (pH 7.5), 20 mM MgCl₂, 500 mM KCl, 50 mM guanidine carbonate, and 10 mM ATP.

Formation of supported lipid bilayers (SLBs)

The SLBs were formed by the vesicle rupture method [23-26]. An appropriate amount of a phospholipid chloroform solution was placed into a small vessel, and the chloroform was removed using a stream of dry nitrogen. The dry phospholipid was then resuspended in pure water to its final concentration of 0.5 mg/ml and sonicated at room temperature until the solution appeared to be clear. Next, 100 μL of the solution was deposited on freshly cleaved pure mica (KAl₂(Si₃AlO₁₀)(OH)₂, 1 cm × 1 cm, Sichuan Meifeng Co., China). If the lipid samples had phase transition temperature above 25℃, they were incubated in a sealed humid chamber within a heater at 60℃ for 30 min. Otherwise, the lipid samples were incubated at room temperature. After a controlled period of time, the samples were gently rinsed with pure water to remove any lipids weakly attached to the substrate and remaining in solution. Rinsing involved 10 repeated washes, each with approximately 100 μL of purified water from a pipette. The prepared SLBs were then carefully inserted into an AFM.

AFM study

AFM operations were conducted on a commercial Multimode AFM system (Multimode 8, Bruker) equipped with a liquid cell and J-scanner. Oxide-sharpened silicon nitride cantilevers (Bruker) with typical spring constants of 0.32 N/m were used. The AFM was operated in tapping mode or Peak Force Tapping Mode with loading ranging from 50 to 400 pN depending on different substrates and peptides [13, 27].

3.1 Mechanical-force–promoted peptide self-assembly on mica

Pep11 contains two hydrophilic amino acid residues (Arg and Glu), which can be oppositely charged and provide strong electrostatic interaction during peptide assembly [18]. Peptide Pep11 was first deposited onto a negatively charged mica surface in an aqueous solution [28, 29]. In situ AFM scanning was then used to apply a mechanical force of piconewton-to-nanonewton levels to the sample on the mica surface and to monitor the force-induced peptide assembly in real time. The mechanical force applied by the AFM tip was manipulated by adjusting the tip oscillation amplitude and setpoint [13]. Fig. 1 illustrates the self-assembly of Pep11 (0.9 mg/mL) during scanning by an AFM tip in tapping mode in PBS. First, several Pep11 nanofibers appeared on the substrate after first scanning over the area of 1 μm × 1 μm (Fig. 1(a)). Then, the nanofibers covered almost the whole scanning area after 4 subsequent scans (Fig. 1(b)). AFM analysis of a zoomed-out image (Fig. 1(c)) indicates that the formed peptide fibers have a height of ∼2.5 nm. It is clear that the AFM scans greatly promoted Pep11 self-assembly because of the strong contrast between the scanned area and peripheral regions. Therefore, AFM scanning is a crucial factor for inducing and accelerating Pep11 self-assembly on a mica surface and can be used to fabricate patterns with specific shapes at desired locations.

Figure 1:Full-screen View

(Color online) AFM topological images of Pep11 peptide nanofibers formed on a bare mica substrate. (a) After the first tapping-mode scan of the substrate by the AFM tip. (b) Pep11 nanofibers almost covered the whole area after continuous scanning of the same area by the AFM tip 4 times. (c) A zoomed-out image indicating the promoting effect of AFM scanning on peptide assembly (left) and section analysis of the white dashed line (right).

3.2 Mechanical-force–promoted polymerization of G-actin on SLBs

In addition to the mechanical-force–promoted peptide self-assembly on the bare mica, the mechanical force also has an effect on actin polymerization on SLBs. G-Actin polymerizes into F-actin faster under a scanning AFM tip on 4:1 DOPC/DOTAP bilayers. An AFM image (Fig. 2(a)) indicates that a fluid-phase bilayer, 4:1 DOPC/DOTAP, has a height of 5 nm on the mica substrate. After addition of G-actin (∼20 μg/ml) into the system, G-actin tended to polymerize into F-actin on the bilayers and to spread to the whole imaging area to form a densely packed pattern in the next 50 min (Fig. 2(b)) during continuous scanning by the AFM tip in tapping mode. Interestingly, although the SLBs were in the fluid phase and thus could move around on the substrate, the formed actin fibers were stable at least for hours during the scanning by the AFM tip. In addition, it was found that G-actin polymerized into fibers without the help of polymerization buffer, which was normally required. This phenomenon may be ascribed to the cationic lipid DOTAP (present in the SLBs), which may facilitate the deposition and polymerization of negatively charged G-actin on the SLBs. When the AFM imaging was zoomed out to a large area, it was found that most of the actin fibers were located at the center of the new image (Fig. 2(c)). Thus, it is clear that continuous scanning by the AFM tip in tapping mode accelerated the polymerization of G-actin into F-actin on the fluid-phase bilayer.

Figure 2:Full-screen View

(Color online) Tapping-mode AFM images indicating the polymerization of G-actin on a specific substrate of the 4:1 DOPC/DOTAP fluid-phase lipid bilayer. (a) An AFM image of a 4:1 DOPC/DOTAP cationic fluid-phase lipid bilayer. (b) Mechanical-force–promoted polymerization of G-actin into F-actin on the lipid bilayer. (c) A zoomed-out AFM image indicating that F-actin was mostly present in the tip-scanned area. (d) Actin continued to grow and filled the entire area after a further scan by the AFM tip in the next 18 min.

To validate the impact of the mechanical force on the actin polymerization on more specific surfaces, SLBs composed of gel phase lipids at room temperature were examined. Bilayers of 80% DPPC and 20% EDPPC were formed at 60℃ and cooled to room temperature (Fig. 3(a)). Fig. 3(b) shows the closely packed F-actin formed on the cationic gel phase lipid bilayer after scanning of the same area by the AFM tip for 106 min at a G-actin concentration of ∼20 μg/ml. In a zoom-out image, we can see that F-actin grew from the previously scanned area (Fig. 3(c)) and quickly covered almost the entire area (Fig. 3(d)). The fast expanding actin fibers during AFM scanning indicated a strong effect of the AFM nanomechanical stimulus on the self-assembly of G-actin.

Figure 3:Full-screen View

(Color online) Tapping-mode AFM images indicating that F-actin formed on a 4:1 DPPC/EDPPC cationic gel phase lipid bilayer. (a) An image of SLBs just after addition of G-actin into the system. (b) A large amount of F-actin formed after a continual scan by the AFM tip for 106 min. (c) A zoomed-out image indicating that F-actin was mostly formed in the area that was previously scanned by the AFM tip. (d) An AFM image indicating that F-actin continued to grow and filled the entire area. Images were obtained with a very small load to avoid disassembling the formed actin fibers on the substrate.

The above results suggested that AFM tip scanning of the substrate facilitated the actin assembly. Because the AFM tip was scanning the surface in oscillating mode in the Z direction and back and forth laterally, actin monomers should be mechanically pushed around on the substrate, and this action should increase the probability of interaction with other monomers to form a trimer that is necessary to start polymerization. The phase of the lipid bilayers also played a role in the actin polymerization. It was found that the fluid-phase bilayers took less time to form actin fibers than did gel phase bilayers even under the same AFM scanning conditions. This result may be due to the fact that fluid-phase bilayers have lower lipid density and faster lipid flip-flop rates than do gel phase bilayers, and these properties helped G-actin to interact with other monomers. Be that as it may, we presented a method for site-specific creation of a stable F-actin network on lipid membranes.

3.3 Mechanical-force–promoted deposition of F-actin on SLBs

An F-actin solution was placed onto the 4:1 DOPC/DOTAP bilayers to check the AFM tip effect on the deposition of prepolymerized F-actin on lipid membranes. AFM imaging indicated that the preformed actin fibers deposited and continued to elongate on the bilayers (Fig. 4). This result indicates that the mechanical stimulus promotes deposition of F-actin on SLBs.

Figure 4:Full-screen View

(Color online) Tapping-mode AFM images indicating the deposition of F-actin on the 4:1 DOPC/DOTAP lipid bilayer as promoted by the scanning tip. (a) The smooth lipid bilayer. (b) The same area after addition of 40 μg/ml F-actin. The AFM tip continuously scanned this area for ∼4 min. (c) A zoomed-out image of the deposited F-actin.