ATOMIC FORCE MICROSCOPY STUDY OF EMODIN TREATED MCF-7 HUMAN BREAST CANCER CELLS DIVISION OF BIOENGINEERING NATIONAL UNIVERSITY OF SINGAPORE 2010... CHAPTER 3 MATERIALS AND METHODS 15
Trang 1ATOMIC FORCE MICROSCOPY STUDY OF EMODIN TREATED MCF-7 HUMAN BREAST CANCER CELLS
DIVISION OF BIOENGINEERING NATIONAL UNIVERSITY OF SINGAPORE
2010
Trang 2
3 Dr. Ong Choon Nam and Dr. Alan Prem Kumar for providing the human breast cancer cell lines
Trang 31.1.3 Anti-cancer properties of emodin 3
2.1 Micromechanical properties of cells 5
2.2 Cytoskeleton and chemotherapy 6
2.3 Chemical compounds affecting actins 7
2.5 Atomic force microscopy for cell biology 9
Trang 4
CHAPTER 3 MATERIALS AND METHODS 15
3.1 Human breast cancer cell line 15
3.7 Young’s modulus calculation of the cells 19
CHAPTER 4 AFM IMAGING STUDY OF MCF-7 CELLS 21
CHAPTER 6 AFM STUDY OF THE EFFECTS OF EMODIN ON THE
MICROMECHANICAL PROPERTIES OF MCF-7 CELLS 41
Trang 5CHAPTER 7 CONCLUSIONS AND FUTURE WORK 48
REFERENCES 51
APPENDIX OPTIMIZATION OF AFM INDENTATION 62
Trang 6SUMMARY
Cancer has long been one of the most fatal diseases worldwide. A substantial
understanding of cancer cells will lead to improved strategies in cancer diagnosis and treatment. Atomic force microscopy (AFM) has recently provided great progress in the study of cancer cells. This emerging technique allows the study of the
morphology and mechanical properties of cells in aqueous environment with high spatial resolution and force sensitivity.
In this study, AFM was used to probe the cell cortical filamentous network as well as the elasticity of living MCF‐7 cell, a human breast cancer cell line. The effects of emodin, an anti‐cancer drug on cell cortex and cell elasticity were also investigated.
Using an optimized scanning setting, it was found that a scanning force of 1 nN was effective in probing the cell cortical filamentous network. Quantitative cell elasticity measurement was done based on AFM indentation test. Young’s modulus values were extracted from AFM force curves using Hertz’s contact model, and were found
to be 437.0 ± 208.2 Pa.
Comparing with control samples, the pre‐treatment of 20µM emodin for 1 hour reduced the mean Young’s modulus values of MCF‐7 cells from 437.0 ± 208.2 Pa to 380.1 ± 138.2 Pa. According to corresponding AFM images, the main reason was that emodin treatment decreased the density of cortical filamentous network, thus reducing the mechanical strength of MCF‐7 cells.
Trang 7AFM technique as a surface analyzing tool has the limitation of not being able to probe the cell interior. However, it can still be a powerful method in detecting anti‐cancer drug effects on cancer cell mechanics.
Trang 8LIST OF FIGURES
Figure 2 Principle and implementation of AFM 11
Figure 3 The chemical formula of emodin 16
Figure 4 An illustration of the AFM indentation test on cells 20
Figure 5 The effects of deformation force on AFM images of cell cortex 23
Figure 6 The effects of scan direction on AFM images of living MCF-7 cells 25
Figure 7 Illustration of photodiode detection mechanism 26
Figure 8 The effects of tip sharpness on AFM images of living MCF-7 cells 27
Figure 9 Contact mode AFM images of a living MCF-7 cell 29
Figure 10 Correlated AFM and CFM of cells double stained for actin and vimentin
30 Figure 11 Micromechanical architecture of the mammalian cell cortex 33
Figure 12 Force curve and Z sensor signal collected in AFM indentation on living
Trang 9Figure 13 Force versus indentation curves for AFM indentation on living MCF-7
cells 37 Figure 14 The effects of emodin on the micromechanical properties of MCF-7 cells
43 Figure 15 AFM images reveal the effects of emodin on the cortical filamentous
Trang 11CHAPTER 1 INTRODUCTION
1.1 Background
1.1.1 Cancer and metastasis
Cancer is one of the leading causes of death worldwide. According to World Health Organization (http://www.who.int), cancer accounted for 7.9 million deaths (around 13% of all deaths) worldwide in 2007.
The term “cancer” refers to a group of diseases that is characterized by the
uncontrolled proliferation and malignant dissemination of cells (Hanahan and
Weinberg 2000; Chambers et al., 2002; Alberts 2008). From the viewpoint of
molecular biology, accumulation of genetic mutation could be the ultimate cause of cancer (Alberts 2008).
Cancer metastasis, which accounts for more than 90% of cancer death (Sporn 1996; Entschladen et al., 2004), is a horrible feature in the pathology of cancer. During metastasis (Fig. 1), cancer cells detach from a primary tumor, intravasate into the blood vessel, are carried by circulation to distant sites in the body, and finally
colonize in these distant secondary sites (Lee and Lim 2007). Cancer cell adhesion, migration and invasion are the crucial steps to enable cancer metastasis, and
attempts to interrupt these steps are the logical strategy for cancer prevention and treatment.
Trang 12
Figure 1
Cancer metastasis (Lee and Lim 2007) The schematic drawing shows the critical steps in cancer metastasis: tumor cells are detached from the primary site, penetrate through the basement membrane, and flow in the blood or lymphatic circulation system to distant sites where they get arrested and invade back into the tissues These tumor cells can initiate angiogenesis, and form secondary tumor in the organs
Trang 13
1.1.2 Atomic force microscopy
In the 1980s, a series of microscopy with similar principles were invented, and they formed the family of scanning probe microscopy (SPM). As a member of this family, scanning force microscopy (SFM), also known as atomic force microscopy (AFM), was widely applied in probing cells and biological materials, because it can be easily adapted to work in liquid environment (Binnig et al., 1986; Hoh and Hansma 1992; Radmacher et al., 1992; Parot et al., 2007).
AFM is useful in measuring the micromechanical properties of living cells as well as probing the micromechanical architectures (Mahaffy et al., 2000; Park et al., 2005; Pesen and Hoh 2005). With the help of this technique, it is possible to get more insights into the cell cortex and mechanical properties of cancer cells.
1.1.3 Anti-cancer properties of emodin
Rhubarb root was classified as a top botanical drug in traditional Chinese medicine,
according to the Classics of the Materia Medica, Shen Nong Ben Cao Jing (Li and Luo
2003). Traditionally it is used to treat various diseases like constipation, jaundice, gastro‐intestinal hemorrhage, and ulcers. Emdoin (3‐methyl‐1,6,8‐
trihydroxyanthraquinone) is one of the main active components of rhubarb, and its anti‐cancer metastasis potentials were revealed by some recent pharmacological studies (Huang et al., 2007).
Firstly, emodin can inhibit cancer cell invasion by reducing the expression of MMPs, which can promote degrading of extracellular matrix (Huang et al., 2004). Secondly,
Trang 14et al., 2005). Thirdly, emodin can inhibit cancer cell adhesion by suppressing the formation of focal adhesion complex (FAC) (Huang et al., 2006).
1.2 Objectives
A substantial understanding of the structural and physical properties of cancer cells arising from drug treatment can potentially contribute to the improvement of cancer diagnosis and treatment. Thus, the objectives of this project are to:
1) Probe the cortical filamentous network of MCF‐7 cells, a cultured human breast cancer cell line using AFM,
2) Quantify their elastic properties by performing AFM indentation tests, and 3) Investigate the effects of an anti‐cancer drug, emodin, on the cortex and corresponding elasticity of these cells.
Trang 15CHAPTER 2 LITERATURE REVIEW
In this chapter, the concept of structure‐property‐function relationship of cells will
be introduced first. Then, some literatures about chemotherapy agents targeting cytoskeleton will be presented. Lastly, the principle of AFM and its application on cancer cell research will be presented.
2.1 Micromechanical properties of cells
The ultimate aim of investigating the mechanical properties of living cells is to
further understand the relationship between cell structure and cell function (Elson 1988). Investigations have been carried out to probe both the global and local
mechanical properties of living cells, with techniques like micropipette aspiration (Hochmuth 2000), optical tweezers (Ashkin and Dziedzic 1989), cell poker (Zahalak et al., 1990), and magnetic beads (Crick and Hughes 1950). Several studies on Fibroblast and bovine pulmonary artery endothelial cells (BPAECs) have suggested that actin microfilament is the major contributor to the micromechanical properties of
cultured mammalian cells (Henderson et al., 1992; Mahaffy et al., 2000; Rotsch and Radmacher 2000; Pesen and Hoh 2005). One of the clues is that disrupting actin filaments results in a distinct decrease in the elastic property of the Fibroblast cells, while disaggregation of microtubule filaments yields no effect on cell elasticity (Rotsch and Radmacher 2000). Thus, in vitro reconstitution of actin network can be used as a simplified model system for cell mechanics study. The mechanical
Trang 16individual components but also the architecture (Wainwright 1982). Tensegrity model was proposed as a framework for cell mechanics study (Ingber 1993; Ingber 1998; Ingber 2000).
2.2 Cytoskeleton and chemotherapy
Cytoskeletal filaments form a complex network within the cell, governing many cellular processes like cell division, cell adhesion, cell movement, and maintaining cell shape (Rao and Li 2004; Alberts 2008). There are three types of cytoskeletal filaments: microfilaments (actin), intermediate filaments (vimentin), and
microtubules. In cultured mammalian cells, actin and vimentin provide the basic micromechanical architecture for sustaining the morphology of the cell by forming the cortical filamentous network beneath the plasma membrane, while microtubules are critical for segregating chromosomes in mitosis (Pesen and Hoh 2005; Alberts 2008).
Taxol and Taxol‐like compounds are clinically used in chemotherapy, and the
mechanism is targeting microtubules to disrupt cellular mitosis (Rao and Li 2004). It has been known for more than three decades that microfilament actin is
substantially modified in malignant transformed cells, and these changes are closely
Trang 17invasion (Jordan and Wilson 1998). But there has been no clinically proven drugs developed to target actin filaments (Rao and Li 2004). So developing drugs to target actin filaments can be promising to control cancer metastasis as actin filaments are known to be involved in active cell migration.
There are many well studied chemical compounds directly disrupting actin filaments, like Cytochalasins, Lantrunculins and Jasplakinolide. All these compounds are
isolated from natural sources, and possess different mechanisms in affecting actin filaments (Rotsch and Radmacher 2000). The limitations of these chemicals as clinical anti‐cancer agents are their toxicity and non‐selectivity between cancer and healthy cells. From the signal transduction point of view, all these compounds are
downstream regulators, and thus are very toxic. Upper regulators like extracts from traditional Chinese herbs are less toxic, therefore providing great clinical potential in chemotherapy. Interestingly, chitosan shows selectivity between malignant and non‐malignant bladder tumor cell lines (Lekka et al., 2001).
2.3 Chemical compounds affecting actins
Microfilament actin is considered as a major contributor to the micromechanical properties of cultured mammalian cells, and it provides the basic framework for maintaining the morphology and functions of living cells (Janmey and Chaponnier 1995; Rotsch and Radmacher 2000). To understand the role of actin in cultured
Trang 18Cytochalasin, Phalloidin, Latrunculin and Jasplakinolide.
Cytochalasins are derived from mold metabolites. Their principle function is to bind
to the barbed end of actin filaments to prevent actin elongation (Cooper 1987; Rotsch and Radmacher 2000; Rao and Li 2004). They also stabilize actin filaments by capping the barbed end. Although cytochalasins are wildly used actin affecting probes, their specificity to actin is not evident. The effects of Cytochalasins on the cell elasticity were first studied using cell poker, and decrease in elasticity was
detected as a sequence of Cytochalasin B treatment (Petersen et al., 1982).
Phalloidin is extracted from poisonous mushrooms. They shift the equilibrium of actin from monomers towards filaments (Cooper 1987). Fluorescent derivatives of phalloidin are commonly applied for visualizing actin filaments in vitro. However, they are not useful in living cell study since they cannot permeate cell membrane, thus have no effect on cell elasticity (Frimmer 1977; Coluccio and Tilney 1984; Rotsch and Radmacher 2000). Besides, they are quite toxic. Cells treated with phalloidin often die, while treated with Cytochalasin D are still alive (Cooper 1987).
Latrunculins are isolated and purified from the Red Sea sponge Latrunculia magnifica
specimens (Spector et al., 1983). They can disrupt actin filaments in various cells by shifting the actin equilibrium between filaments and monomers (Coue et al., 1987; Spector et al., 1989; Lyubimova et al., 1997; Oliveira et al., 1997). They also have anti‐cancer potential since they can inhibit proliferation in some tumor cells (Spector
et al., 1989). Besides, Latrunculins are very valuable for probing microfilament actin functions in cells (Ayscough et al., 1997).
Trang 19Interestingly, this compound can promote actin polymerization while disrupt
filament actin (Duncan et al., 1996). However, the anti‐proliferation effects on
several breast and prostate cancer cell lines make it a promising anti‐cancer agent (Stingl et al., 1992; Duncan et al., 1996).
2.4 Atomic force microscopy for cell biology
Microscopes have been the primary techniques for cell biology study. Light
microscopy (LM) is the most commonly used microscopy. For live cell imaging, LM requires contrast‐enhancing procedures. However, its resolution is limited by the wavelength of light used. The invention of Electron microscopy (EM) is a revolution for microscopy techniques, for it can acquire images with ultra high resolution.
However, the sample preparation procedure for EM is invasive, thus not capable for live cell study (Braet et al., 1998).
Started with scanning tunneling microscopy (STM), a series of scanning probe
microscopy (SPM) were invented in the 1980s (Butt et al., 2005). Among them, AFM has become a standard tool of biological surface science due to its ability to image biological samples under near physiological conditions with nanometer scale
resolution (Binnig et al., 1986; Henderson et al., 1992; Radmacher et al., 1992; Braet
et al., 1998). As a micro‐nano scale force sensor, AFM is also wildly applied to probe the mechanical properties of biological samples including living cells (Mahaffy et al., 2000; Radmacher 2002; Costa 2003).
Trang 20piezoelectric ceramics scanner, and the deflection of the cantilever is detected by a sensitive photodiode via a laser that has been reflected off the back at the free end
of the cantilever.
In the contact working mode of the AFM, the feedback control loop drives the
piezoelectric scanner up and down to maintain the cantilever deflection constant during the scanning process. Thus the sample will lower down when the profile gets higher, while it extends upwards when the profile becomes lower. By recording the piezoelectric scanner’s movement, the 3D topography of the sample surface can be constructed.
In the case of hard sample whose surface is much stiffer than the cantilever, sample surface topology and tip geometry determine the image results. But when scanning soft samples like living cells, the sample will be deformed by a constant force over surface (Heinz and Hoh 2005). In general, the micromechanical structure on the cell surface dominates the formation of AFM images, while the viscoelastic properties of the cell surface contribute significantly to the images obtained (Radmacher et al., 1992).
Trang 21
Principle and implementation of AFM (Heinz and Hoh 2005) A tiny tip is mounted
on the free end of a micro-fabricated cantilever The sample is held by a piezoelectric ceramic scanner The scanner can move the sample in 3D space with high accuracy The interaction force between the tip and sample can be detected by the optical
detector and processed by an electrical feedback controller By keeping the interaction force constant, the spatial topography of the sample can be obtained As a force sensor, the mechanical property of the sample can also be obtained Inset: The “small ball and spring” model of AFM This concept models the AFM as using a sample ball attached
to a weak spring to interact with sample surface
Trang 22
2.5 AFM in cancer cell research
The potential of AFM in living cell imaging under near physiological conditions was illustrated since the early 1990s (Henderson et al., 1992; Radmacher et al., 1992). Since then, high resolution AFM images of various types of living cells were acquired (Hoh and Schoenenberger 1994; Braet et al., 1998; Nagao and Dvorak 1998;
Leporatti et al., 2009). Recently, human bladder and breast cancer cells were
extensively studied with AFM (Lekka et al., 1999; Lekka et al., 2001; Chen and Cai 2003; Li et al., 2008; Leporatti et al., 2009). Other types of cancer cells like lung cancer cells, malignant transformed fibroblasts were also investigated with AFM (Weisenhorn et al., 1993; Goldmann and Ezzell 1996; Goldmann et al., 1998; Rotsch
et al., 2001; Park et al., 2005).
Lekka et al was the first to compare the elastic properties of normal human bladder
epithelial cell lines (Hu609 and HCV29) and cancerous ones (Hu456, T24, BC3726) (Lekka et al., 1999). The results showed that the Young’s modulus of cancerous cells
is about one order of magnitude lower than healthy ones. The effect of chitosan on the stiffness and gycolytic activity was tested on these human bladder cells. A strong correlation between the decrease of the energy production and the increase in Young’s modulus was revealed on chitosan‐treated human bladder cancer cells. It is quite interesting to notice the selectivity of chitosan between cancerous cells and healthy ones. Cancerous cells became stiffer due to addition of chitosan, whilst healthy ones were not affected.
Chen and Cai investigated the interaction between human breast carcinoma cells and extracellular matrix with AFM (Chen and Cai 2003). MCF‐7‐R cells were fixed in
Trang 23Li et al probed the elasticity of human breast cancer cells with AFM. Micro‐sized
spherical probe was adapted to measure the apparent Young’s modulus of benign (MCF‐10A) and cancerous (MCF‐7) human breast cells (Li et al., 2008). Cancerous MCF‐7 is found to be two times softer than benign MCF‐10A at room temperature. Both loading rate and temperature were found to influence the characterization of apparent Young’s modulus. AFM and confocal fluorescence microscopy images together revealed the different F‐actin organization under cell membrane which probably contributes to the cell elasticity directly. These mechanical changes in cancerous cells may facilitate their migration and invasion in cancer metastasis (Lee and Lim 2007).
Leporatti et al combined confocal fluorescence microscopy with atomic force
microscopy to study the topological and mechanical properties of two human breast carcinoma cell lines MCF‐7 and HeLa (Leporatti et al., 2009). AFM images of living MCF‐7 cells, as well as fixed cells either in air or in phosphate buffer solution (PBS) were obtained. Young’s moduli on both living and fixed cells were measured. The interaction between MCF‐7 and fibronection‐coated substrate was also studied. Short and fine protrusions in cell edges were noted microscopically and proved to be actin‐rich extensions by phalloidin staining.
Trang 24In conclusion, AFM can be applied in the investigation of anti‐cancer drug effects on the cortical filamentous network of cancer cells, and thus can potentially contribute insights on chemotherapy in the cytoskeleton level.
Trang 25CHAPTER 3 MATERIALS AND METHODS
In this chapter, the details of the sample preparation and AFM experiments will be described.
3.1 Human breast cancer cell line
MCF‐7, a human breast cancer cell line, was kindly provided by Dr. Alan Prem Kumar (Yong Loo Lin School of Medicine, National University of Singapore). The cells were seeded in 25 cm2 flasks at 37 ⁰C, 5% CO2 in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), and 1% Penicillin. The cells were passaged when confluent. Only passages 3‐10 were used for experiments
to avoid distinct genetic changes.
3.2 Anti-cancer drug, emodin
6‐methyl‐1,3,8‐trihydroxyanthranquinone (emodin) was purchased from Sigma. The chemical formula of emodin was shown in Fig. 3. It was dissolved in Dimethyl
sulfoxide (DMSO) to obtain a 20 mM stock solution. The working concentration of emodin is 20 ~ 40 µM (Huang et al., 2006).
Trang 26
Figure 3 The chemical formula of emodin
3.3 Sample preparation
MCF‐7 cells were sub cultured on round glass cover slips (13mm diameter) in 24 well tissue culture plates for two days. Before performing the AFM experiments, the cells were washed with phosphate buffered saline (PBS) and treated with 20µM emodin for 1 hour in the incubator. The cells were then washed with PBS again. Finally, the glass cover slip was glued onto a round steel holder to be mounted on the AFM scanner.
For control samples, the cells were washed with PBS and treated with culture
medium alone. To prepare DMSO samples, the volume of DMSO diluted in cell culture medium was the same as the volume of emodin stock solution diluted in cell culture medium.
Trang 27
3.4 Atomic force microscopy
A Multimode AFM with Nanoscope IV controller (Veeco Instruments, Santa Barbara,
CA, USA) was used for both living cell imaging and force curves collection. This AFM was equipped with a PicoForce module. The Pico‐Force scanner had a maximum XY scan range of 50 µm × 50 µm and Z range of 20 µm.
3.5 AFM imaging
Standard V‐ shape silicon nitride 0.01N/m cantilevers (model MLCT and MLST;
VEECO Probes, Camarillo, CA, USA) were used for imaging. A standard fluid cell was used without the o‐ring. Cell imaging was performed in a liquid drop of PBS buffer. Prior to the imaging process, spring constants of the cantilevers were calibrated with the attached thermal noise module on unloaded cantilevers, and the deflection sensitivity of the cantilever was calibrated on a bare part of the glass surface.
Imaging was done at room temperature .The tip was localized at the centre of the cell as observed under an optical microscope.
The effects of emodin on the filamentous structures of MCF‐7 cells were examined
by acquiring AFM images from three sets of samples: MCF‐7 cells, MCF‐7 cells pre‐treated with 20µM emodin, and MCF‐7 cells pre‐treated with DMSO. About 5 to 10 well spread cells were scanned for each set. The scanning parameters were
30µm×30µm for scan size, 0.2 and 0.4 for PI control, 60µm/s for scan rate, 1nN for scan force, and 90⁰ for scan direction. All the imaging was performed over the center
of the nucleus region to make the results comparable.
Trang 283.6 AFM force curves
Standard V‐ shape 0.03N/m colloidal probes with polystyrene beads (Novascan Technologies, Armes, IA, USA) were used for cell elasticity measurement. The
diameter of the polystyrene bead was 4.5 µm. A standard fluid cell was used without the o‐ring. All the calibrations and elasticity measurements were performed in a liquid drop of PBS buffer at room temperature. Prior to the measurements, spring constants of the cantilevers were calibrated with the attached thermal noise module
on unloaded cantilevers, and the deflection sensitivity of the cantilever was
calibrated on a bare part of the glass surface. Probes were located over the center of the cells under optical microscopy monitoring.
Force curves were collected by recording cantilever deflection while ramping the piezo scanner in the z direction. A series of plots of force versus sample position were obtained. All the force curves were collected under 1nN absolute trigger mode with 10µm ramp size and 1Hz scan rate.
The effects of emodin on the elasticity of MCF‐7 cells were examined by collecting force curves on three sets of samples: MCF‐7 cells, MCF‐7 cells pre‐treated with 20µM emodin, and MCF‐7 cells pre‐treated with DMSO. Twenty well spread cells were investigated for each set, and three force curves were collected successively on each single cell.
Trang 29
3.7 Young’s modulus calculation of the cells
The elastic modulus can be extracted from the force curves. Up to now, Hertz’s contact model and its variation were widely used in analyzing AFM indentation
results (Lekka et al., 1999; Mahaffy et al., 2000; Rotsch and Radmacher 2000; Li et al., 2008; Leporatti et al., 2009). Hertz’s contact model was originally applied to study the contact deformation between two perfectly homogeneous non‐conforming smooth bodies. When applying Hertz’s contact model on indentation of a living cell (Fig. 4), the cell was treated as an elastic half space and assumed as incompressible since the indentation was made small. As such, Hertz’s contact model, can still,
within reasonable limits, fit the experimental data and give a good approximation of the Young’s modulus of the cells (Kuznetsova et al., 2007).
When using Hertz’s contact model to extract Young’s modulus values from AFM
indentation study on cells, if the radius of the indenter is R, exerting a force F on this elastic half space will cause an indentation δ. And the following relationship is
Trang 30(3)
Figure 4
An illustration of the AFM indentation test on cells (Li et al., 2008) The cell is
assumed as an elastic half space and incompressible material To fulfill this
assumption, a 4.5μm diameter bead was used to indent a 5μm high cell in a rate of 1um/s to obtain a 1μm indentation depth
Trang 31CHAPTER 4 AFM IMAGING STUDY OF MCF-7
4.1 Introduction
In this chapter, AFM will be used to scan the cortical filamentous network of living MCF‐7 cells in a physiological buffer. Some early studies suggested that AFM images
of living cells can reveal some submembranous structures (Henderson et al., 1992; Braet et al., 1998). However it was not clear whether the AFM tip is deforming the cell membrane or penetrating the cell membrane (Henderson et al., 1992). Some following studies showed that the cell membrane was not damaged during scanning (Haydon et al., 1996; Schaus and Henderson 1997). Therefore, membrane
deformation arising from contact with AFM tip was considered as the major
mechanism in the acquisition of AFM images of living cells under near physiological conditions.
4.2 Optimizing experimental parameters
When using AFM to probe the cell surface morphology, many control parameters need to be optimized. These mainly include scan size, computer feedback control, scan rate, scan direction, tip sharpness, and scan force. Here, tests were done to optimize all the parameters for the scanning of living MCF‐7 cells.
Trang 32
Trang 33
Figure 5
The effects of deformation force on AFM images of cell cortical filaments: (A, B) Images of a living MCF-7 cell acquired at imaging forces of ~0.1nN and 1.2 nN, respectively When the loading force is small, the tip does not maintain good contact with the sample surface The image quality is poor and the cell surface features are difficult to distinguish When the loading force is above 1 nN, a clear image with good contrast can be acquired It should be noted that the cell became deformed when increasing loading force The cross-section profiles in the middle of the images show that the cells were deformed by 500nm when increasing the imaging force from 0.1nN
to 1.2nN
Trang 34
When conducting AFM study on living cells in liquid, 0⁰ and 90⁰ scanning will have different effects depending on both scan direction and cell alignment (Fig. 6). From the AFM instrumentation point of view, 90⁰ scan is more stable, because the lateral bending of tip does not affect the laser detection (Fig. 7B). Besides triangle shape cantilever does not have much torsional bending. 0⁰ scan actually apply slightly higher and unstable force on the sample during scanning, since lateral scratching of the tip can lead to a small scale shifting of laser spot in the detector, and the
feedback control do have some delay (Fig. 7A). From the cytoskeleton point of view,
it would be good that the tip can actually scan across the aligning of the stress fibers which are the major micromechanical structures. So in this project, we will scan cells with 90⁰ scan direction, and choose images with less lifting up effects as those
Trang 35
Figure 6
The effects of scan direction on AFM images of living MCF-7 cells: (A, B) Images of
a living MCF-7 cell acquired at scan direction of 0 and 90 degree, respectively (C, D) Illustration of the scanning processes for 0 and 90 degree, respectively The arrows
with dash line show the images acquisition direction, therefore Fig B was rotated 90 degree clockwise after the original AFM image was acquired, to make it comparable with Fig A Comparing the two AFM images, it is noted that there are ‘coarse
filaments’ aligned from top to bottom paralleled, and ‘fine filaments’ in between to
form a well organized network In Fig A, the scan direction is perpendicular to the
‘coarse filaments’, thus leading to a ‘lifting up’ effect, while Fig B is more smooth
90⁰ 0⁰
Trang 36
Figure 7
Illustration of photodiode detection mechanism (Heinz and Hoh 2005): (A) The photodiode was divided into segments ○A and ○B The deflection of the cantilever was magnified by the ‘light lever’ and measured by the current signal change from (○A – ○B ) / (○A + ○B ) (B) Lateral force on the tip can torque the cantilever and lead to laser spot rotation on the photodiode But such rotation does not influence too much on the deflection measurement
Trang 37
Normally, increasing the sharpness of AFM tips can improve the resolution of the images. However, it was not clear whether high resolution cell morphology images can be acquired by using sharper AFM tips for living cells. Here we tested two typical AFM tip models MLCT and MSCT, with tip radius of 20nm and 10nm, respectively. The results showed that ultra‐sharp tips like MSCT could not acquire images of living MCF‐7 cell (Fig. 8). This is mainly due to the sharp tip penetrating into cell