Capturing the elasticity and morphology of live fibroblast cell cultures during degradation with atomic force microscopy

Authors


K. E. Aifantis, Lab of Mechanics and Materials, Aristotle Univeristy, Thessaloniki 54124, Greece. Tel: 30-2310 995921; fax: 30-2310995921; e-mail: k.aifantis@mom.gen.auth.gr

Summary

Atomic force microscopy, in a liquid environment, was used to capture in vitro the morphological and mechanical changes that cultured fibroblasts undergo as time elapses from the completion of the cell culture. Topography images illustrated that initially, the nucleus had a height of 1.18 ± 0.2 μm, and after 48 h it had decreased to 550 ± 60 nm; similarly, the cell membrane exhibited significant shrinkage from 34 ± 4 to 23 ± 2 μm. After each image scan, atomic force microscopy indentation was performed on the centre of the nucleus, to measure the changes in the cell elasticity. Examination of the force-distance curves indicated that the membrane elastic modulus at the nucleus remained the same within the time frame of 48 h, even though the cell morphology had significantly changed.

Introduction

Recent studies have suggested that the mechanical properties of cells are of particular interest as they can assist in disease diagnosis. It has been shown, for example, that blood cells infected with malaria have a higher elastic modulus than healthy red blood cells (Suwanarusk et al., 2004). Furthermore, recent studies have documented that the elastic modulus of cancer cells is lower than that of the respective healthy cells (Sokolov, 2007; Cross et al., 2007; Zhou et al., 2012).

Although various methods exist for measuring the elastic modulus of cells, such as poking (Zaharak et al., 1990), micropipette aspiration (Shao et al., 1998), scanning acoustic microscopy (Bereiter-Hahn et al., 1995), magnetic tweezers (Bausch et al., 1999), and optical tweezers (Lenormand et al., 2001), atomic force microscopy (AFM) can be considered as the most efficient method since, in addition to measuring the elastic modulus, through nanoindentation, it can also capture simultaneously the morphology of the cells with a high spatial resolution (Radmacher et al., 1996). Furthermore, AFM may operate in a liquid environment making it possible to study live cell cultures in vitro. Particularly, AFM was used to document that the morphology of cells is the first cell property that is affected by apoptosis (Hessler et al., 2005), as cell shrinkage was noted to take place before other biochemical changes, such as the loss of the mitochondrial membrane potential. More recently, AFM has been used to examine changes in cancer cell morphology and reduction in stiffness as they undergo apoptosis (Kim et al., 2012), and a protocol was developed to measure the elastic modulus of live cancer cells cultures (Zhou et al., 2012).

In the present study, AFM was used to examine the elasticity of cell cultures as time elapses from completion of the culture. Fibroblast cells were left to mature after completion of the culture, and the morphological and elasticity changes they experienced until decomposition were documented.

It should be noted that the morphology, as well as the elasticity of as cultured fibroblast cells, have been studied using AFM (Haga et al., 2000). It was found that the elastic modulus ranged between 4 and 100 kPa throughout the fibroblast area, with the nucleus being the softest (low elastic modulus) and the cell membrane being the stiffest (high elastic modulus); these changes, however, were also due to the substrate effect.

No study has examined how time affects elasticity of cell cultures. In the sequel therefore, AFM will be employed to capture the changes that fibroblasts (cell line L929) experience from the completion of the cell culture until decomposition of the cells, from both a structural and a mechanical point of view. Therefore, the cells studied can be categorized as follows: Stage 1 cells, which were examined upon completion of the cell culture; Stage 2 cells, which were examined 48 hours after completion of the cell culture; and Stage 3 cells, which were examined 60 hours after completion of the culture.

Sample preparation

Reagents

Fetal bovine serum (FBS), alpha-minimum essential medium (α-MEM), L- glutamine (200 mM), Trypsin-EDTA 0.5% solution and phosphate-buffered saline (PBS) were purchased from GIBCO, Paisley, Scotland, whereas the Penicillin/Streptomycin 100× solution was purchased from Life Technologies. The tissue culture plastic ware (pipettes, flasks, 24-well plates, conical tubes, 10-mm diameter cover slips) were purchased from Corning Costar Corporation (Cambridge, MA, USA).

The culture medium in which the cells were cultured and stored was a mixture of α- MEM medium supplemented with 10% FBS (v/v), 100 Units/mL penicillin, 100 μg/mL streptomycin, and 2 mM/L L-glutamine. This culture medium resembled the natural environment, in which cells are grown. The antibiotics penicillin and streptomycin were used to prevent contamination.

Cells and cell cultures

Fibroblasts were derived from the mouse L929 cell line (alpha sub-line), which was purchased from the European Collection of Cell Cultures (Salisbury, UK). After thawing, the cells were suspended in the culture medium (mentioned above), and seeded in T-75 flasks at 1.25×106 cells in a 10-mL complete medium and left to grow in a humidified incubator at 37°C in a 5% CO2_95% air atmosphere, as previously described (Milikovic et al., 2003; Coimbra et al., 2007) with some modifications. Cells that did not adhere to the glass slip and cell debris were removed by changing the culture medium after 24 and 72 h of cultivation. Fibroblasts were grown until the cultures reached confluence. At that time the cells were detached by trypsinization, re-suspended in the culture medium and seeded into 24-well plates (with a 10-mm diameter cover slip in each well) at 1.0 × 104 cells/well and allowed to attach for at least 24 h. Thereafter, dead cells were removed by a washing step in the culture medium. Viability of the cells was checked by means of the inverted microscope, as well as with a hemacytometer using trypan blue dye (Phipps et al., 2007).

Experimental process

Upon completion of the cell culture (confluence was reached), the cells were transferred to the microscope lab to perform the AFM experiments. One set of as cultured cells (Stage 1 cells) was set to be studied under the AFM, and the remaining cells were stored in a sterile environment at 4°C, without CO2 supplementation and medium changing. This allowed for cell decomposition to occur at a slower pace than if left in an incubator.

Before placing the cover slip with the Stage 1 cultured fibroblasts on the AFM, the culture fluid was carefully rinsed off the cells with sterile PBS through a pipette, so that no additional proteins were attached on the fibroblasts. Cell imaging was performed using a Veeco Multimode AFM in tapping mode, with a SiC tip that had a tip radius of 12 nm. An optical microscope was attached to the AFM and hence it was possible to choose the appropriate cells to examine. The AFM software allowed for the “point and shoot” method, and therefore, after imaging, the centre of the cell nucleus was indented with the AFM tip to examine the cell elastic modulus. Once the indentations were completed the cover slip with the cells was disposed of.

After 48 (Stage 2) and 60 h (Stage 3), AFM imaging and indentation on the centre of the nucleus, was performed on a second and third set of cells (taken from storage), respectively.

It should be noted that while in storage, the cells were fully covered by the culture medium; the plate was also covered by a microfilm to avoid possible evaporation and, hence, the cells did not experience any dehydration.

Results

Microstructural changes

Figures 1 and 2 depict the fibroblasts upon completion of the cell culture (Stage 1). To clearly see the dimensions of the cell, the height and length profile of the cell between specific points was provided through AFM line profiles. In Figure 2 it can be seen that, initially, the total cell length was about 38 μm, whereas the nucleus height was 1 μm; the synapses connecting the cells are also clearly seen. Performing line profiles on all the cells of Figure 1 for Stage 1 it was found that their average nucleus height was 1178 ± 200 nm, whereas the average cell membrane length was 34 ± 4 μm.

Figure 1.

AFM topography image of fibroblasts at stage 1.

Figure 2.

(a) Higher magnification of fibroblast indicated in Fig. 1; (b) line profile corresponding to colored marks of cell in (a).

Figures 3 and 4 depict the cells studied 48 h (Stage 2) after the culture was completed. It can be seen that the cells experienced significant shrinkage; particularly the line profile (Fig. 4b) illustrates that the height of the nucleus had decreased to about 500 nm, whereas the cell length to 25 μm. Performing line profiles on all the cells of Figure 2 for Stage 2, it was found that their average nucleus height was 550 ± 60 nm, whereas the average cell length was 23 ± 2 μm.

Figure 3.

AFM topography image of cells at Stage 2.

Figure 4.

(a) AFM topography image of fibroblast at stage 2; (b) line profile corresponding to marks of cell in (a).

The images taken for the cells that had been left sixty hours in storage (Stage 3) indicate that severe decomposition had occurred, as Figure 5 illustrates. The nucleus and the cell membrane are not distinguishable, and only remains of the cell are seen; the total height of this decomposed fibroblast is 30 nm, whereas the length is about ∼1.4 μm. It can, therefore, be said that this was the final degradation stage of the fibroblast cell culture.

Figure 5.

(a) AFM topography image of fibroblast remains at Stage 3; (b) line profile corresponding to marks in (a).

Elasticity

To study how the elasticity of the cells was affected as time elapsed, AFM indentations were performed on the centre of the nucleus for the cells at each stage, depicted in Figures 2, 4, as well as on the centre of the degraded cell of Figure 5. These indentations provided force-distance curves that captured the distance that the tip penetrated into the nucleus as a function of the applied force (Fig. 6a). Furthermore, the piezoelectrics driving the indentation can provide both the displacement of the tip in the sample as a function of time (nm vs. sec) and also the voltage changes throughout the indentation time (V vs. sec); hence each indentation provides two respective curves as shown in Figure 6b. It is these two curves (Fig. 6b) that allow for the determination of the cell elasticity. According to (Ngan & Tang, 2009; Tang et al., 2008) the nm vs. sec curve at the onset of unloading gives the displacement rate, while differentiation of the V vs. sec curve provides the loading rate. Figure 7 illustrates how the loading rate was computed; the displacement rate was evaluated in a similar manner from Figure 6b.

Figure 6.

Representative force-distance and deflection-time curves obtained from AFM indentations for cells of Stage 1.

Figure 7.

Deflection-time curves obtained from AFM indentations for cells of all Stages. The slope of the curves between 0.5∼0.7 s gives the loading rate used in Fig. 8.

In Figure 8, the displacement rate obtained from the various indentations for each Stage is plotted against the respective loading rate, resulting in linear plots for each set of cells examined. Performing a fit to the data of Figure 8 gives a function of the form y=Kx where the slope K is related to the elastic modulus of the cell nucleus through the expression (Ngan & Tang, 2009)

image(1)

where A is the cantilever sensitivity of the AFM tip, k is the spring constant, a is the tip radius, vtip and vcell are the Poisson's ratios of the tip and cell, whereas Etip and Ecell are the elastic moduli of the tip and the cell. It is, therefore, seen that K is inversely proportional to the elastic modulus of the cell (K∼1/Ecell).

Figure 8.

Comparison of stiffness of the nucleus as the fibroblasts decompose. The slope of nm/s vs. V/s data is inversely proportional to the elastic modulus.

In concluding, indentations on gold were also performed, to have a standard to which the elasticity of the cells could be compared with, as cells are much softer than gold.

Discussion

The AFM images, Figures 1–5 illustrate that as time elapses from the cell culture completion the morphology of fibroblasts changes significantly since after 48 h, the cell height and length were approximately 50% less than their initial values. Despite, however, the structural changes, the elasticity of the cells does not seem to change within 48 h. According to Figure 8: Kgold<Kstage3_cell<Kstage2_cell=Kstage1_cell, which implies that Egold>Estage3_cell>Estage2_cell=Estage1_cell. Indentations on gold were provided as a standard to verify that the results are correct as it is well known that Egold has to be greater than Ecell and indeed this is what Figure 8 suggests. The exact modulus cannot be deduced, because to do so two standards with known elastic moduli should have also been indented according to Ngan and Tang (2009), to determine the constants A and a in Eq. (1). However, the purpose here is not to determine the elastic modulus of fibroblasts [because it has already been found to depend on the substrate (Solon et al., 2007)] but to see how degradation affects it.

It can, therefore, be concluded from Figure 8 that the cells at Stage 1 and Stage 2 had a similar elastic modulus. This suggests that as long as the cells are alive they have similar elastic properties, regardless of how much time has elapsed since completion of the cell culture. Once, however, 60 h had passed, decomposition had significantly progressed and the decomposed remains of the fibroblast at Stage 3 had a higher elastic modulus than the live cells. Such an increase of the elastic modulus and hence stiffness of the decomposed cells was anticipated, since at decomposition dehydration occurs; after dehydration the biomaterials behave as ‘solids’ rather than viscoelastic, and hence become stiffer (Balooch et al., 1998; McDaniel et al., 2007). It should be noted, however, that the increase in stiffness of the Stage 3 cells could be attributed to the substrate effect, as the cell height for the Stage 3 cells was 30 nm (Fig. 5b), whereas the indentation depth was 20 nm.

Figure 8, therefore, indicates that as time elapsed from completion of the cell culture the elasticity of the fibroblast nucleus did not change as long as cell decomposition had not initiated. A more detailed follow-up study to this would be to continuously scan and indent the same cell over a period of 12 h to capture the initiation of cell shrinkage and possible degradation, because if left at room temperature under the AFM, this process will occur much faster.

Conclusions

In the present study, it was shown that AFM can be used to capture in vivo the changes in the morphology and elasticity of cells during a biological process.

Once the cell culture was completed, the cells were stored in a sterile environment and were completely covered with culture fluid so dehydration would not occur as time elapsed.

Within 48 h after completion of the culture, significant shrinkage of the cells occurred as their height decreased from 1.18 ± 0.2 μm to 550 ± 60 nm and the cell membrane length from 34 ± 4μm to 23 ± 2μm, while after 60 h significant decomposition of the cell had taken place, since the cell height was 30 nm and no distinction between the nucleus and the membrane could be done. Despite the significant morphological changes observed, the stiffness of the cells did not change within the initial 48 h, since AFM indentation on the nucleus indicated a similar elastic modulus at the onset of the culture completion and after 48 h.

Having successfully employed AFM to capture in vitro the morphological and mechanical changes of fibroblasts, this high resolution microscope can be used to study more involved biological processes, such as cell aging, cancer cell proliferation, and plasmodium penetration into healthy red blood cells (Povelones et al., 2009).

Acknowledgements

K.E.A. and S.S. are grateful for support from the European Research Council Starting Grant MINATRAN 211166. The authors thank Prof. N. Grigoriadis, for providing the mouse L929 cell line, as well as, Dr. R. Lagoudaki, MSc, and Prof. A. Kyritsis for helpful discussions.

Ancillary