Advances in cellular nanoscale force detection and manipulation

2.1 Surface characterization

Atomic force microscope (AFM) (MFP-3D™, Asylum Research, Santa Barbara, CA) was used to characterize the surface topography and to acquire force measurements on a broad range of biological materials. Because of the soft characteristics of the cell, it is necessary to choose a cantilever with an appropriate elasticity coefficient or spring constant to prevent cell surface damage by the AFM probe (Clifford and Seah, 2005). Here, a silicon nitride cantilever (Nanoworld, Switzerland, PNP-TR) that had a spring constant of 0.05 N/m, as determined by the thermal noise calibration method (Hutter and Bechhoefer, 1993), was deemed appropriate for cellular surface characterization. Force measurements were acquired either at selected points or by collecting a force map array and then evaluated using the AFM data analysis tools.

2.2 Force–distance curves and analysis of adhesion force

For single cell mechanical measurements, a typical AFM experiment consisted of first scanning a region of the sample topographically and then using that image as a map to perform force measurements at specific sites (Dufrêne et al., 1999; Gotsmann et al., 1999; Hsieh et al., 2012). By collecting force curves on the cell surface, mechanical properties, including hardness, elasticity, and modulus, can be probed, and pulling measurements may yield data such as those related to adhesion and receptor site bonding energies. For force measurements, the AFM probe can be thought as a small ball (tip) attached to a weak spring (the cantilever). The spring is used to measure forces between the tip and the surface, with attractive forces stretching the spring and repulsive ones compressing it. Thus, in contrast to imaging in which the cantilever (or sample) is scanned in the XY direction, for force measurements, the cantilever moves only in the Z direction. Force measurements are made by collecting a force distance curve (force curve), which is a plot of cantilever deflection as a function of tip-sample separation.

Fig. 1 is a schematic representation of a force curve, and important events in the measurement are numbered 1–5. Point #1 is the cantilever position above the surface before initiating the force curve. Once the measurement has begun, the cantilever is moved vertically toward the surface (red line) until it reaches point #2. This is called the jump-to-contact point, where attractive forces acting on the tip cause it to snap to the surface. Between point #2 and point #3, the cantilever deflects upward as the system continues moving toward the surface. This region may be analyzed to yield the stiffness and the elasticity (Young’ modulus) of a sample. Point #3 is the trigger point, which is specified in software, and when this point is reached, the system stops and starts to retract the cantilever from the surface. This setpoint controls the maximum force applied to the sample during the measurement. Point #4 is the snap off, which corresponds to the adhesion force. Point #5 is when the cantilever retracts to its initial starting position. Moreover, adhesion is determined by the difference between the force measured at the lowest point (jump-off contact) in the retracting curve and the baseline force measured when the cantilever is away from the surface (Point #5).

Close

Figure 1 Approach (red) and withdrawal (blue) curves as shown in a typical AFM force-distance plot (left). The illustrations of the cantilever (right) depict its position at each of the five points (1–5) shown in the graph.

Export to PPT

2.3 Statistical analysis of AFM force curves

Force curves were acquired as the AFM probe scanned a sample surface, and these force data were then collected for subsequent analysis. By analyzing these force curves, the elastic properties of the sample can be extracted using the Hertzian contact model (Wang and Dennis, 2007). The hardness or Young’s modulus of the surface can be obtained by analyzing the AFM force curves using the following relationship: $$E=\frac{3F(1-{\nu }^2)}{4\sqrt{R}}{\delta }^{-3/2}$$ where E is the Young’s modulus, F is the load, ν is the Poisson ratio of the sample, R is the tip radius (10 nm) and δ is the indentation depth. A Poisson ratio of ν = 0.5 is commonly used for cells, and the tip geometry is assumed to be spherical. The cantilever spring constant in our example was determined using the thermal method (Hutter and Bechhoefer, 1993), and the elastic modulus (E) was calculated using the Hertzian-based contact theory. $$E_c={\left(\frac{1-{\nu }_1^2}{E_1}+\frac{1-{\nu }_2^2}{E_2}\right)}^{-1}\text{,}$$ where E_c is the fit parameter for the Hertz model. This parameter is directly related to the Young’s modulus (E) and Poisson ratio (ν) of the indenter and the indented material, where E₁ and ν₁ are the indented properties and E₂ and ν₂ are the indenter properties (for a silicon nitride tip, ν₂ is 0.25 and E₂ is 290.0 GPa). Thus, by fitting the force curve data using the above equation, the Young’s modulus of the sample (E₁) can be determined.

2.4 Cell culture and sample preparation

The cell lines of human hepatocellular carcinoma (SK-Hep-1 and Hep G2) were purchased from the Bioresource Collection and Research Center, HsinChu, Taiwan, and were cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS) (Gibco, Invitrogen, Carlsbad, California) at 37 °C in a humidified atmosphere of 5% CO₂. To prepare cell samples for subsequent AFM experiments, cells were seeded on glass cover slides in a 12-well culture plate and incubated overnight. Cells were washed twice with 1× PBS (phosphate buffer saline, pH 7.4, Sigma–Aldrich), and then, fixation was induced with 2.5% glutaraldehyde (Sigma–Aldrich). Cells were again washed twice with 1× PBS and then subjected to the AFM-based experiments in liquid (1× PBS, pH 7.4).

3.1 Ultra-sensitive detection of cellular mechanics

Several methods have been developed to study cell mechanics in the pico Newton (pN) to nano Newton (nN) force range. The three most common techniques include optical tweezers (Ashkin, 1997; Ashkin et al., 1986), magnetic tweezers (Alenghat et al., 2000; Lim et al., 2009; Matthews et al., 2006; Puig-de-Morales et al., 2004), and AFM (Binnig et al., 1986). Optical tweezers utilize a focused laser beam to “trap” micrometer-sized dielectric particles based on photon momentum and refractive index differences between the particles and a medium. Nanometer displacement of the particles can be achieved with angstrom level precision by generating forces in the pN range. For cell studies, a dielectric particle may be coated with specific proteins or antibodies that adhere to sites on a cell surface. The optical tweezers are then used to pull or push the particle, and thus exert a force by which molecular interactions between the particle and cell can be measured.

Magnetic tweezers are similar to optical tweezers, but use small superparamagnetic particles and magnetic field gradients to generate forces in the fN to pN force range. As with optical tweezers, magnetic tweezers use functionalized particles to interact with the cell surface. In addition to pulling and pushing, magnetic tweezers can also apply a torsional force, which provides additional opportunities for molecular characterization. Due to the simplicity of this technique, it is one of the most widespread methods for biophysical research.

Both optical and magnetic tweezers are limited to pN (or less) force detection, thus limiting the types of interactions that can be probed. Determining how the cytoskeleton and ECM assemble into 3D biopolymeric networks involves an understanding of the connection between molecular and cellular length scale mechanics. Therefore, methods capable of characterizing the mechanical properties of single macromolecules, sub-cellular macromolecular assemblies, and whole cells are needed, and forging connections between these different length scales (from microns to nanometers) in a quantitative fashion is required.

3.2 Molecular mechanical measurements by AFM

AFM has emerged as a powerful tool for both imaging and measuring forces on a wide range of samples and across many length scales (Binnig et al., 1986). This technique possesses several significant advantages for researching biological systems. For example, there is no postfixation as in electron microscopy; there is no staining for immunofluorescence observations; the ability to conduct experiments in physiological media exists; and a force range from piconewtons to hundreds of nanonewtons can be studied. In addition, because AFM is a probe technique, such measurements can be made at specific locations on a sample with nanometer scale precision. Thus, AFM is well suited to studying the structure and mechanical properties of a broad range of biological specimens under ambient and physiological conditions.

AFM is a scanning probe technique that uses a sharp tip on the free end of a flexible cantilever to probe the sample surface. Both topographical images and force measurements can be made using AFM (as well as many more types of measurements not relevant to this report). Interactions between the tip and sample cause the cantilever to deflect, which is measured using an optical lever system composed of a laser, cantilever, and photodiode. Deflections on the order of tens of picometers can be detected. Thus, by selecting an appropriate cantilever (spring constant), one can measure a wide range of forces on a variety of biological materials and specimens.

In addition, AFM-based force spectroscopy can be used to characterize the molecular mechanisms underlying substrate adhesion by careful analysis of tip–substrate interactions. Additional performance measurement techniques, such as using a calibrated colloidal probe, may provide much needed data at different scales because the contact area of the probe can influence mechanical measurements (Francius et al., 2006; Richert et al., 2004).

Fig. 2 shows a 3D topographical image of a SK-Hep-1 cell on a glass surface, with drawings of two cantilevers depicting the locations where force measurements were made, and the data acquired at each point. A qualitative comparison of the two data profiles shows that there is a significant difference in the forces measured (the cell surface is softer with much higher adhesion). A more detailed description of how we can extract force information from AFM measurements is provided in a later section.

Close

Figure 2 AFM height image (40 × 40 μm²) of a cell with drawings depicting the analysis of adhesion force on the cell surface and on the substrate.

Export to PPT

3.3 AFM-based cell adhesion interaction

Fig. 3 shows the results of force measurements on Hep G2 cells on glass, where both adhesion and elasticity were characterized. The AFM tip vertically approaches the cell and makes contact with the cell surface, and then, the free (tip) end of the cantilever deflects as the system moves the base of the cantilever closer to the substrate. This region of the force curve is used to calculate the stiffness of the substrate. As the tip is retracted, the force curve profile provides information on the adhesion between the tip and the cell surface. The force curves in Fig. 3b indicate that multiple adhesion events (all <2 nN) occurred as the cantilever tip was retracted from the cell surface. This can occur because of multiple binding sites between the tip and the cell or other features on the cell surface. In contrast, the force curve measurements on the glass substrate showed little or no adhesion force.

Close

Figure 3 (a) AFM topographical image (55 × 55 μm²) of cells was cultured on a glass surface. Force distance curves were collected at each of the points shown in the image and are displayed in the graph (b). The inset of (b) displays a magnified view showing multiple adhesions.

Export to PPT

Table 1 summarizes the AFM imaging of the cellular morphology and the cell cluster and movement of two different types of cells. These results are consistent with the results observed from optical microscopy and confirm that the AFM scanning process does not destroy the morphology of cells.

Additionally, we use the Hertzian model to calculate the Young’s moduli of Hep G2 and SK-Hep-1. As shown in Fig. 4, a total of 305 points on the Hep G2 cell and 303 points on the SK-Hep-1 cell from the array of AFM force curves were collected and analyzed. Our results display two peaks in the histogram of the Young’s modulus values for the SK-Hep-1 cell surface, one at 54.77 ± 62.39 kPa and the other at 99.43 ± 23.39 kPa, which resulted in an average value of less than 120 kPa. For Hep G2 cell, a narrow distribution of Young’s moduli was localized at approximately 10.11 ± 12.26 kPa. Moreover, a wide range of Young’s modulus values were distributed between 112.74 kPa and 497.77 kPa for the HepG2 cell surface and showed an average value of 299.53 kPa. Finally, a high stiffness and a low adhesion force were also clearly observed in the Hep G2 cell compared to the SK-Hep-1 cell. These results are in agreement with previous studies (Bao and Suresh, 2003; Swaminathan et al., 2011). Based on these data, we can distinguish between the surface properties of two human hepatocarcinoma cell lines with dissimilar differentiated stages. These detailed results are summarized in Table 2.

Close

Figure 4 Histogram of all Young’s modulus values shown in the map from SK-Hep-1 cells and HepG2 cells.

Export to PPT

Indeed, scanning the cells force alteration using AFM can be applied to confirm the stiffness nanotomography and the mechanical properties in different differentiated tumors. For example, (Fuhrmann et al., 2011) utilized AFM to validate stiffness tomograms in normal (EPC2), metaplastic (CP-A) and dysplastic (CP-D) human esophageal cells, which revealed that poorly differentiated cells were softening (the Young’s moduli were calculated as 2.9 kPa and 2.1 kPa in CP-A and CP-D, respectively) more than normal cells (9.9 kPa). Furthermore, the mechanical properties of the nuclei and the nucleoli revealed a significantly stiffening in normal cells and an obvious deformability in metaplastic cells, as well as a softening of subcellular structures in dysplastic cells.

3.3.1

3.3.1 Chemically functionalized probes for specific detection

Beyond mechanical and basic chemical or adhesion measurements, AFM also affords opportunities for more specific chemical probing of samples at nanometer length scales by chemical functionalization of the tip. CH₃-modified AFM tips have been used to investigate hydrophobic forces associated with mycolic acids on the surface of mycobacteria (Alsteens et al., 2007; Dorobantu et al., 2009). The functionalized AFM tips revealed that two bacterial species, venetianus RAG-1 and R. erythropolis 20S-E1-c, exhibited different cell surface hydrophobicities.

Chemical modification of silicon-based probes can be accomplished using silicon chemistry and a variety of precursors with specific chemical functional groups. A schematic diagram of this concept is shown in Fig. 5. In this case, alkylsilane molecules were covalently bonded to the AFM tip, similar to a self-assembled monolayer on a silicon substrate. As the tip scans over a surface or is used for force measurements, chemical interactions between the tip and the surface are monitored and recorded. In addition, one may attach Au nanoparticles to the AFM tip using aminopropylsilane (APS) as an adhesive layer (Hsieh et al., 2009) to probe the interactions between cell surface components and noble metal nanoparticles at specific locations.

Close

Figure 5 Schematic depicting the process of chemically attaching functional molecules to an AFM tip.

Export to PPT

3.3.2

3.3.2 Cellular level manipulation and treatment with AFM

AFM uses piezoelectric transducers with sub-nanometer resolution to position the tip relative to the sample. Such precision enables unique applications, for example, the use of a very sharp object as a custom probe that can be used to insert molecular entities into the cytosol of individual cells.

AFM-based nano-surgical techniques would allow manipulated cells to remain viable and avoid tumorigenic transformations. Such cells could be used for repeated or sequential experiments or for therapeutic purposes. Han et al. (2005) used an ultra-thin needle fabricated from a commercially available AFM tip to inject plasmid DNA into a living cell to modulate gene expression. The genetic material was injected into the cell in a controlled manner without causing irreversible cell damage. Obataya et al. (2005) demonstrated the use of a needle-shaped AFM probe to controllably apply a constant force to a cell membrane. By monitoring the cantilever deflection, they were able to measure a 1–2 μm needle penetration depth into the cell membrane following indentation. The ability to measure displacements and sense the force on the needle is similar to a surgeon’s finger during surgery. By modifying the surface of a needle, one can load various molecules such as nucleic acids, proteins or other chemicals through standard immobilization chemistries.

Cellular level manipulation involving membrane protein pulling, extraction and identification of mRNA and insertion of plasmid DNA into defined loci on cell membranes has been reported. In studies by Han et al. (2005) and Harley et al. (2008), cell viability was evaluated after holes were created using an AFM tip at specific locations on a cell membrane surface. Phospholipase A2 coated beads were attached to the AFM cantilever and then allowed to contact the cell membrane surface for durations of 5–10 min, creating holes 5–10 μm in diameter. The creation of a hole was confirmed by fluorescence imaging, before and after the bead contacted the surface, and by AFM (Afrin et al., 2009). In addition to verifying cell viability, intracellular filamentous structures were visualized, and targeted gene delivery was attempted and confirmed. Moreover, using AFM to probe interaction forces between a ligand and a receptor (e.g., an AFM probe replaced by a biological cell or by isolated biomolecules such as DNA, antibody, and protein) has become a multifunctional molecular toolbox in screening assays. For example, the AFM cantilever probes a prostate cancer cell (PC3), which is coupled with concanavalin A to contact bone marrow endothelial (BME) cells. The strong adhesive interactions between PC3 and BME cells can be significantly disrupted by anti-ICAM-1, anti-β1 and anti-P-selectin. Therefore, AFM can determine and quantify the nanoscale adhesion events between different cell types (Muller and Dufrene, 2008; Reeves et al., 2013).

Manipulation of single cells for transgenesis, in vitro fertilization, individual cell-based diagnosis, and pharmaceutical applications has recently become a topic of great interest. Each of these techniques requires precise injection and manipulation of cells; thus, issues related to penetration force can arise. In AFM studies, Kwon et al., 2009 showed that the penetration force for a variety of cells (L929, HeLa, 4T1, and TA3 HA II, wherein L929 is the mouse fibrosarcoma cell line, HeLa is the human cervical cancer cell line, 4T1 is the mouse mammary carcinoma cell line, and TA3 HA II is the mouse mammary adenocarcinoma cell line) ranged from 2 to 22 nN. The authors also found that the point of entry for the tip on the cell surface and the cantilever stiffness significantly affected the penetration force. An example of a force curve depicting tip penetration through a cell surface (a human SK-Hep-1 hepatoma cell) is shown in Fig. 6. Double penetration events were detected due to the multi-membrane layers of the cell. These results may aid in the development of precision micro-medical instruments for cell manipulation and treatment.

Close

Figure 6 AFM force distance curve showing the point of tip penetration through the cell membrane.

Export to PPT