Fluorescence-Guided Force Spectroscopy & Recognition Imaging on Cells via ILM/AFM

Topics Covered

Introduction
    Fluorescence Microscopy in Living Cell Analysis
    Combining AFM methods with Fluorescence Recognition Imaging
Overview of Experiments
Glycosylphosphatidylinositol (GPI) Anchored Green Fluorescent Protein (GFP)
CD4 Molecules Fused with Yellow Fluorescent Protein (YFP)
Summary
References
About Agilent Technologies

Introduction

Fluorescence Microscopy in Living Cell Analysis

Fluorescence microscopy has become an important tool for localizing receptor/ligand interactions in living cells. Labeling different proteins with different spectral fluorophores allows imaging of different cellular, subcellular, or molecular components and determination of the specific localization of the proteins in cells. The rejection of unwanted, short-wavelength background (Rayleigh scattering of excitation light) by spectral filtering improves the contrast of specifically labeled cellular structures in fluorescence microscopy. The lateral and axial resolutions are limited by the diffraction limit of light and result in ~200nm resolution. Recently, techniques with higher resolution have been developed, such as stimulated emission depletion (STED), photo-activated localization microscopy (PALM), and stochastic optical reconstruction microscopy (STORM), which achieve lateral resolution of 10-30nm.

Combining AFM methods with Fluorescence Recognition Imaging

Optical techniques, however, cannot provide any information of the sample topography. By contrast, atomic force microscopy (AFM) allows topographical images to be obtained at the nanometer scale in liquid environments and at room temperature. In addition, the recognition of receptor/ligand pairs can be investigated at the single-molecule level.

A recently developed simultaneous topography and recognition imaging technique, performed utilizing PicoTREC from Agilent Technologies for the work detailed herein, is capable of yielding a topographical image as well as a map of recognition sites of the same area with a single scan at 5nm lateral resolution. The new technique’s operating principle is the same as for dynamic mode AFM imaging. A cantilever is oscillated near its resonance frequency and scanned over the sample surface, but in this case the cantilever is made chemically sensitive by attaching a ligand via a short linker to its tip. The binding sites are evident from the reduction of oscillation amplitude. Enhanced signal processing, in combination with a modified feedback loop, provides a recognition image simultaneously acquired alongside a topography image.

Essentially, the separation of topographical and recognition events is achieved by splitting the cantilever’s oscillation amplitude into lower and upper parts (with respect tothe cantilever’s resting position). The maxima of these parts are then used to record the topography (lower parts) and recognition image (upper parts) at the same time.

This article discusses the ability of the Agilent 6000ILM atomic force microscope to easily position the AFM tip to regions of interest using a contrast-enhanced optical image and a fluorescence image, and the subsequent TREC imaging and force spectroscopy, which show high correlation among fluorescence intensity, binding probability, and recognition area, as well as the visualization of the recognized nanodomains on cells with different levels of protein expression.

Overview of Experiments

One of the most intriguing topics in life science and bionanotechnology is the investigation of cells and the organization and function of proteins in the cell membrane. Cell signaling, communication to neighboring cells, and transport to adjacent tissue is all organized through membrane proteins. Two examples of important protein systems for cell signaling and cell membrane organization are presented here.

AFM experiments were performed using an Agilent 6000ILM AFM mounted on a Zeiss Axio Observer A1 (for measurement of cells with GPI-GFP) or a TILL Photonicsfluorescence microscope (for measurement of cells with YFP-CD4). All images were acquired in PBS at room temperature using Agilent MAC Mode (i.e., magnetic AC). Agilent type VI and VII MAC Levers were utilized and AFM images were acquired using Agilent PicoView software. Cantilevers for force spectroscopy were washed in chloroform three times, dried in air, washed in piranha (30% H2O2 and 70% H2SO4) for 30 min, rinsed with water, and dried by heating at 100°C. Cantilevers for TREC imaging were washed in chloroform three times and dried in air. All cantilevers were then treated with APTES, conjugated with NHS-PEG- aldehyde or NHS-PEG-acetal, and subsequently linked with antibody [1].

Glycosylphosphatidylinositol (GPI) Anchored Green Fluorescent Protein (GFP)

In the plasma membrane of cells, there are cholesterol- and sphingolipid-enriched domains, termed ‘lipid rafts’. These lipid rafts serve as organizing centers for the assembly of signaling molecules, influence membrane fluidity and membrane protein trafficking, and also influence receptor trafficking. For example, upon specific triggering, membrane receptors such as T-cell receptors or B-cell receptors translocate into lipid rafts, which is the prerequisite for efficient receptor-mediated signal transduction.

The Agilent 6000ILM AFM allowed the investigation of both the morphology and the distribution of lipid rafts on cells. For this, T24 cells (human bladder carcinoma cells) - which were transfected to express GPI anchor derived from DAF fused to GFP [termed GPI-(DAF)-GFP], a highly effective lipid raft marker [2] - were grown on glass-bottom Petri dishes. Before imaging, the cells were fixed with 4% paraformaldehyde for 30 min.

The topographical image (Figure1A) of the cells was obtained using Agilent MAC Mode. The imaging force was so gentle that the filaments of the cytoskeleton underneath the plasma membrane can hardly be seen. Lamellipodia at the cell border, however, can be clearly detected. The level of GPI-(DAF)-GFP expression in the cells was examined by obtaining GFP fluorescence images (Figure1B).

Figure 1. (A) AFM topography of fixed T24 cells expressing GPI-GFP (image size: 50μm). The image was measured with MAC Mode in PBS by using an Agilent type VI MAC Lever oscillating at 16kHz and at about 7nm with an imaging speed of 15μm/s. The imaging force was so gentle that the filaments of the cytoskeleton underneath the plasma membrane can hardly be seen. Lamellipodia at the cell border, however, can be clearly detected. (B) Fluorescence image (FITC filter set, 100 x magnification) of the same cells showing the expression level of GPI-GFP. Some bright regions close to the nucleus indicate the endoplasmic reticulum (ER), where the GPI-anchored proteins are synthesized. Beyond the ER regions, the distribution of GPI-GFP in the plasma membrane looks homogeneous, except for several bright dots (e.g., those in the red square). (C) By using the cantilever functionalized with the anti-GFP antibody, the recognition image of the region marked with the red square in (B) revealed nanodomains with a size of about 200-300nm. In the region marked with the dashed blue circle, the nanodomains have been aggregated, which correlates well to the microdomain shown in the fluorescence image. During the recognition imaging, the cantilever was oscillating at 16 kHz and at about 7nm with an imaging speed of 6.5μm/s. Image size: 5μm.

Close to the nucleus of the cells there are some regions with very high fluorescence intensity, which is in accordance with the fact that the GPI-anchored proteins are synthesized in the endoplasmic reticulum (ER) surrounding the nucleus of the cells (see Figure 2). Beyond the ER regions, there is no correlation between the topographical height and the fluorescence intensity, which suggests that most of the GPI-(DAF)-GFP molecules are located in the plasma membrane rather than in the cytosol. Occasionally, some bright fluorescence dots can be found far away from the ER region, as marked with the red square (Figure1B). Such bright dots may be vesicle in the cytosol or microdomain in the plasma membrane. In general, the conventional fluorescence microscopy with the highest magnification (100x) shows a homogeneous distribution of GPI-(DAF)-GFP in the plasma membrane.

Figure 2. Schematic diagram of the combined optical and atomic force microscopy for recognition measurements on cell. GPI-anchored GFP is synthesized in ER, modified in Golgi, and transported via vesicles to plasma membrane. While the optical microscope can examine the overall expression level and micrometer-scale distribution of GPI-GFP through the fluorescence measurement, the recognition imaging by the antibody-functionalized cantilever can visualize the distribution of GPI-GFP at the nanometer scale. For recognition imaging, the cantilever oscillates at constant amplitude. When the antibody on the tip binds with the GFP on the cell surface, the upper part of the oscillation wave is reduced. This can be detected by the Agilent PicoTREC box, where the oscillation wave is split into upper and lower parts. The upper part of the oscillation wave is used to construct the recognition image. Here, recognition events are shown in red.

To investigate the distribution of GPI-(DAF)-GFP at the nanometer level, TREC imaging was utilized. The principle of TREC imaging using a cantilever tip functionalized with anti-GFP antibody is shown in Figure 2. During imaging, the cantilever oscillates (driven by a magnetic field) at a constant amplitude. When the antibody on the tip binds with the GFP on the cell surface, the upper part of the oscillation is reduced. This can be detected by the Agilent PicoTREC box, where the oscillation wave is split into upper and lower parts. From the upper part of the oscillation wave, the recognition image can be constructed; the area with a binding event is shown as a dark spot.

The recognition image of the region marked with the red square in Figure1B is shown in Figure 1C. From the recognition image, it can be seen that the GPI-(DAF)-GFP molecules form nanodomains with a size of about 200-300nm. In the region marked with the dashed blue circle, the nanodomains have been aggregated, which looks like a microdomain in the fluorescence image. Such a microdomain provides a unique bridge to show the correlation between the fluorescence and recognition images.

CD4 Molecules Fused with Yellow Fluorescent Protein (YFP)

CD4 (cluster of differentiation 4), a glycoprotein originally found on the surface of T helper cells, plays a very important role in immunology and the disease of HIV. The distribution of the CD4 protein was investigated on T24 cells that were transfected to express YFP-fused CD4. Such cells were fixed and imaged by using fluorescence microscopy and AFM. From the brightfield image and the fluorescence image (Figures 3A and 3B), it was found that some of the cells (e.g., cell 1 in Figure 3) have high expression of YFP-CD4, while some other cells (e.g., cell 2 in Figure 3) have very low expression.

Figure 3. (A) Brightfield image and (B) fluorescence image of fixed T24 cells transfected with YFP-CD4 reveal that some of the cells (e.g., cell 1) have high expression of YFP-CD4, while some other cells (e.g., cell 2) have very low expression. With the guidance of the optical images, force spectroscopy and recognition imaging were performed on cell 1 and cell 2. On cell 1, 244 force-distance curves were measured, from which 220 curves show binding events, resulting in a binding probability of 90.2%. (F-I) AFM topography and recognition images measured on cell 1 and cell 2 with an Agilent type VII MAC Lever functionalized with anti-CD4 antibody. Images on the two cells were measured with the same tip at the same frequency (8kHz) and amplitude (about 50nm) with an imaging speed of about 1.9μm/s. Cell 1 (panel H) showed many large recognition spots with a size ranging from 100 to 200nm, while cell 2 (panel I) showed much smaller spots with farther distance.

With the guidance of the optical images, force spectroscopy and recognition imaging were performed on cells 1 and 2. For these measurements, the cantilever tips were functionalized with anti-CD4 antibody. On cell 1, 244 force-distance curves were measured, from which 220 curves show binding events, resulting in a binding probability of 90.2% (Figure 3D). A typical force-distance curve with a binding event is shown in Figure 3C. On cell 2, 241 force- distance curves were measured, from which only 10 curves show binding events, resulting in a binding probability of 4.1%. A typical force-distance curve without a binding event is shown in Figure 3E. The force-distance curve measurements on the two cells were performed with the same tip. Binding probability for cells1 and 2 is in very good agreement with the fluorescence intensity.

Figures 3F to 3I show the AFM topography and recognition images measured on cells 1 and 2. Images on these two cells were measured with the same tip at the same frequency and amplitude. It was found that there are many large recognition spots on cell 1 (Figure 3H) with a size ranging from 100 to 200nm. Most of the recognition spots are located in hole regions in the topography, but recognition spots can also be found in protruding regions in the topography (Figure 3F).

Many of the nanodomains are already connected with each other. Such a high density of recognition spots coincides with the strong fluorescence signal on cell 1, and is identical with the high binding probability from force spectroscopy. On the other hand, the recognition spots on cell 2 (Figure 3I) are normally smaller than those on cell 1, and the distance between recognition spots is basically greater than that on cell 1. The overall area of recognition spots on cell 2 is much less than that on cell 1, which correlates very well to the expression level revealed by the fluorescence image.

To further examine the specificity of the recognition spots, a block experiment was performed by injecting free anti-CD4 antibody into the measurement solution. Figure 4 shows that the topography of the cell membrane looked similar after the block; however, the recognition spots were significantly reduced by the injected free antibody. For imaging before and after the block, the same cantilever was used at the same oscillation frequency and amplitude. The clear reduction of the recognition spots confirmed the specificity of the recognition-imaging-revealed CD4 nanodomains.

Figure 4. (A) Topography and (B) recognition images on fixed T24 cells expressing YFP-CD4 before block, imaged by Agilent type VII Mac Lever functionalized with anti-CD4 antibody. About 2.5 hours after injection of free anti-CD4 antibody molecules (with a final concentration of 0.05mg/ml), (C) topography and (D) recognition images were measured at the same position with the same cantilever tip, which showed significant reduction of recognition spots. All images were measured at a cantilever oscillation frequency of 8.37kHz, an amplitude of about 50nm, and with an imaging speed of about 2.7μm/s.

Summary

The fully integrated combination of atomic force microscope and inverted light microscope capabilities afforded by the Agilent 6000ILM AFM recently enabled an easy and quick investigation of the distribution of cell membrane proteins at the micrometer and nanometer scales. The simple and convenient operation of this new instrument will make it the preferred tool of biologists and biophysicists for numerous future applications in life science at the nanoscale, thereby bridging the worlds of optics and atomic resolution under physiological conditions.

References

1. Linda Wildling, Barbara Unterauer, Rong Zhu, Anne Rupprecht, Thomas Haselgrübler, Christian Rankl, Andreas Ebner, Doris Vater, Philipp Pollheimer, Elena E. Pohl, Peter Hinterdorfer, and Hermann J. Gruber, "Linking of sensor molecules with amino groups to amino-functionalized AFM tips," Bioconjugate Chem 22 (2011): 1239-1248.
2. Julian Weghuber, Stefan Sunzenauer, Birgit Plochberger, Mario Brameshuber, Thomas Haselgrübler, and Gerhard J. Schütz, "Temporal resolution of protein- protein interactions in the live-cell plasma membrane," Anal Bioanal Chem 397 (2010): 3339-3347.

About Agilent Technologies

Agilent Technologies nanotechnology instruments let you image, manipulate, and characterize a wide variety of nanoscale behaviors - electrical, chemical, biological, molecular, and atomic. Our growing collection of nanotechnology instruments, accessories, software, services and consumables can reveal clues you need to understand the nanoscale world.

Agilent Technologies offers a wide range of high-precision atomic force microscopes (AFM) to meet your unique research needs. Agilent's highly configurable instruments allow you to expand the system's capabilities as your needs occur. Agilent's industry-leading environmental/ temperature systems and fluid handling enables superior liquid and soft materials imaging. Applications include material science, electrochemistry, polymer and life-science applications.

Date Added: May 17, 2012 | Updated: Jul 15, 2013
Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback
Submit