Dynamic Fibroblast Cultures

Abstract

Mechanical forces play an important role in the organization, growth and function of tissues. Dynamic extracellular environment affects cellular behavior modifying their orientation and their cytoskeleton. In this work, human fibroblasts have been subjected for three hours to increasing substrate deformations (1–25%) applied as cyclic uniaxial stretching at different frequencies (from 0.25 Hz to 3 Hz). Our objective was to identify whether and in which ranges the different deformations magnitude and rate were the factors responsible of the cell alignment and if actin cytoskeleton modification was involved in these responses. After three hours of cyclically stretched substrate, results evidenced that fibroblasts aligned perpendicularly to the stretch direction at 1% substrate deformation and reached statistically higher orientation at 2% substrate deformation with unmodified values at 5–20%, while 25% substrate deformation induced cellular death. It was also shown that a percentage of cells oriented perpendicularly to the deformation were not influenced by increased frequency of cyclical three hours deformations (0.25–3 Hz). Cyclic substrate deformation was shown also to involve actin fibers which orient perpendicularly to the stress direction as well. Thus, we argue that a substrate deformation induces a dynamic change in cytoskeleton able to modify the entire morphology of the cells.

Introduction

Native tissues are composed of three-dimensional matrices and well aligned cellular layers. This composition enables the tissues to operate with proper functionality, for instance to move peristaltically and to be compliant in vascular and visceral tubular-structured tissues. The orientation of cells and the alignment of cytoskeleton elements are affected by the dynamic extracellular environment that results from the application of mechanical stress.¹ It is well established that connective tissues adapt to changes in mechanical environment, i.e., in bones trabeculae align in response to compressive and tensile stresses.²

More recently in vitro studies evidenced that cyclic substrate deformations cause changes in cell orientation^3–6 and in actin cytoskeleton.^6,8 Cyclic substrate deformation was shown to induce alignment of fibroblast perpendicular to the stretch direction; osteoblasts,⁹ smooth muscle cells,⁵ embryonic myoblasts¹⁰ and endothelial cells^6,11 have also shown to modify the alignment when cultured on stretched substrate.

Thus, literature reports large evidences which support cell alignment after applying a substrate deformation but a comparison between different deformations and frequencies applied to cell lines is not reported yet. In this work, human fetal lung fibroblasts have been subjected to increasing substrate deformations (from 1% to 25%) and different frequencies (from 0.25 Hz to 3 Hz) in order to identify whether and in which ranges the different deformations magnitude and rate were the factors responsible of the cell alignment and if actin cytoskeleton modification was involved in these responses.

Materials and Methods

Materials.

Silicone sheets 0.010″ non-reinforced vulcanized (SMI Specialty Manufacturing Inc. Saginaw, Michigan, USA) have been used as deformable substrates. Sheets (1 x 1 cm) have been sterilized by autoclaving for 20 minutes at 121°C and coated with sterile fibronectin 10 µg/ml (Sigma, Milan, Italy) for one hour at room temperature before cell seeding. Unless otherwise specified, all chemical reagents were purchased from Sigma.

Cell culture.

Human fibroblasts MRC5 (ATCC CRL 171) derived from normal lung tissue have been used at 15 x 10³ cells/cm². Cells have been cultured in DMEM enriched with 10% fetal bovine serum, glutamine (2 mM), penicillin (100 U/ml) and streptomycin (100 µg/ml) (Euroclone, Italy). In order to obtain the optimal attachment to the membrane, cells have been maintained on silicon at 37°C in humidified atmosphere with 5% CO₂ in static conditions for 24 hours before applying mechanical stress.

Application of mechanical stretching to the cells.

Instron 5564 testing Instrument (Instron Corporation, Canton, Massachusets, USA) has been used to tensing silicon substrates on which cells have been allowed to adhere. The device comprises an electronic control console and a loading frame with a load capacity of 2.5 N in tension or in compression and 2,500 mm/minute-0.05 mm/minute respectively maximum and minimum speed of the moving crosshead. Silicon samples with cells, connected with tweezers to the Instron's load cell, have been immerged for all the experimental time (three hours) in a culture vertical chamber (Ugo Basile, Milan, Italy) filled with culture medium and maintained at 37°C with 5% CO₂ in a closed bath. Different substrate deformations have been applied as cyclic uniaxial stretching and compared with a not stressed control. The range of magnitude tested was from 1 to 25% at 0.5 Hz, whereas different frequencies tested were from 0.25 to 3 Hz at 2% magnitude. The orientation and number of cells were determined from photographs of cells obtained in random fields from at least three separate experiments as described in next section.

MTT test.

MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-tetrazolium bromide) is a water-soluble chemical. Active mitochondrial dehydrogenases of living cells convert the yellowish MTT to an insoluble purple formazan. This conversion does not take place in dead cells. This water-insoluble formazan can be solubilized using dimethylsulfoxide (DMSO), and the dissolved material can be measured spectrophotometrically. After applying mechanical stress on cell, cells were incubated for additional three hours with a MTT solution (50 µg/ml final concentration). Samples were rinsed and formazan salt formed were dissolved. Absorbance at 570 nm was read. Data obtained were expressed as mean percentage with respect to control ± standard deviation.

Orientation counting.

Cell orientation has been considered by the longest aspect of the cells. To evaluate the percentage of oriented cells, samples have been fixed with 3.7% formaldehyde for 30 minutes and stained ten minutes with toluidine blue 0.01% in dH₂O pH8. Seven pictures for each sample have been acquired and tests for each stress condition have been performed in triplicate (n = 21). A grid with different oriented angle (0–30°; 30–60° and 60–90°) has been applied on each picture and the cellular orientation has been measured and reported in graphs as percentage of oriented cells in each of the three angle considered. The direction of uniaxial deformation has been considered as 0°. The picture below shows an example of the method used to define and measure cell orientation (Fig. 1).

Figure 1.

Example of cell orientation counting. This method has been used for measure cell and actin filaments orientation.

Actin filaments.

To determine the role of the actin cytoskeleton in cell alignment, separate experiments performed in the same conditions described above have been used. Cells cultured on unstretched control or on silicon stretched for three hours in different conditions have been fixed in formaldehyde 3.7% for 30 minutes and then labeled with phalloidin-TRITC conjugated (Sigma, Italy). Actin filaments have been observed by fluorescence microscopy (Leica, DM 2500) at 40x magnification. A grid with different oriented angle (0–30°; 30–60° and 60–90°) has been applied on pictures acquired and the actin filaments orientation has been measured and reported in graphs as percentage of oriented filaments in each of the three angles considered.

Statistical analysis.

Means of group were compared by analysis of variance, and the significance of differences was obtained by post hoc testing using Bonferroni's method. p value was obtained from the ANOVA table; the conventional 0.05 level was considered to reflect statistical significance.

Results

Cell viability.

Figure 2 shows results obtained from viability test. Mechanical stress did not interfere with cells viability until 20% of magnitude deformation was reached, as the mean of percentages close to 100% showed in all the cases. At 25% of deformation magnitude cells viability decreased reaching a mean of 60% ± 6 with statistical relevance. In this case, the decreasing value was due to both cell detachment and damage of the cellular structure as confirmed in Figure 3D.

Figure 2.

MTT test. Cell viability has been verified after applying different deformation magnitude for three hours. Results are expressed as mean of percentage with respect to control ± standard deviation.

Figure 3.

Picture shows a cellular random distribution in the control (A) and when 1% substrata deformation was applied (B). (C) represents the morphology and alignment of cells cultured on substrata deformed from 2 to 20%. (D) evidenced damaged cells at 25% substrate deformation. (At the top of the figure the grid used to measure the cell alignment).

Cell alignment on deformed substrate.

Pictures shown in Figure 3 (magnification 25x) are representative of the morphology and the distribution of cells when subjected to a substrate deformation (Fig. 3B–D) compared with cells cultured in a static control (Fig. 3A). Cells were randomly distributed on control not deformed substrata while cells subjected to a mechanical stress ≥2% of substrate deformation align perpendicularly to the stress direction. No modifications in cell morphology were seen from 2 to 20% of silicon deformation while cells subjected to a 25% deformation shown a severely altered morphology (Fig. 3D).

Figure 4 shows percentage of oriented cells in the three angles considered to study cell alignment modifications after three hours of different degrees of deformation. Cells cultured on control, not deformed substrata, showed a random distribution with no statistical differences between cell numbers in the three degrees of alignment. Starting from the 2% deformation, 60% of fibroblasts aligns perpendicularly to the stretch direction, as shown by a statistically higher number of cells in the range 60–90° together with lower percentage of cells in the range 0–30° and 30–60° orientation. No statistically significant differences were observed between cells stressed from 2 to 20% while when 25% of substrate deformation was applied, cell morphology, as shown in Figure 3D, was altered thus cell reorientation was technically difficult to quantify for the high percentage of cells damaged and detached (Fig. 4).

Figure 4.

Percentage of oriented cells in the three angle-range considered varying the degrees of substrate deformation. (* indicates significant results compared with the control with p ≤ 0.05).

Cells subjected for three hours to 2% of substrate deformation were then tested modifying the frequency of substrate deformation from 0.25 Hz to 3 Hz. As shown in Figure 5, when comparing the percentage of oriented cells in control not stressed culture with stressed substrata it was evident a statistically increase of cells oriented in the range 60–90°. No statistical differences were seen when comparing all the frequency applied.

Figure 5.

Percentage of oriented cells in the three angle-range considered varying the frequency of substrate deformation. (* indicates significant results compared with the control with p ≤ 0.05).

Actin filaments.

Figure 6 shows fluorescent micrographs (40x magnification) of cells stained with phalloidin-TRITC labeling actin filaments and morphometrical measures of actin filaments. Human fibroblasts cultured on stretched substrata showed that the actin cytoskeleton remodeling follows cell alignment. Control unstretched cells (Fig. 6A) and cells cultured on 1% deformed silicon (Fig. 6B) evidenced dense and randomly oriented actin filaments while after three hours of elongation in a deformation range between 2 and 20%, cells formed bundles of stress fibers oriented near perpendicular to the deformation direction (Fig. 6C). Seventy-seven ± 8% of actin filaments was oriented in the range of 60–90° angle with respect to the uniaxial stretch direction. A significant lower percentage of cells was in the range of 0–30° angle (2 ± 3%) and 30–60° (34 ± 3%) (Fig. 6D). At 25% substrate deformation, cell sufferance shown in contrast microscopy (Fig. 3D) was evidenced also in fluorescence microscopy (data not shown) where cells became smaller with few actin filaments randomly oriented.

Figure 6.

Fluorescence microscopy of phalloidin-TRITC stained cells for actin filaments on control (A) and when 1% substrata deformation was applied (B). (C) represents the morphology and alignment of actin stress fibers when cells were cultured on substrata deformed from 2 to 20%. (D) shows the graph obtained from the morphometric measures on actin filaments reorientation after applying mechanical stress (2% 0.5 Hz three hours). (* indicates significant results compared with the control with p ≤ 0.05).

Discussion

Cyclic stretching plays an important role in creating the architecture of tissues (e.g., bone, arteries, ligament) but the mechanisms involved in cell orientation as response to mechanical stress remains for certain aspects unclear even if actin cytoskeleton seems to be strongly involved. Literature reported several studies concerning cell response to mechanical stretching^12–15 but in none of them cell response to different stretching magnitude or to different stretching frequencies have been compared. In the present work, silicon deformable substrates have been elongated for three hours uniaxially and magnitudes have been applied from values that produce no response on cellular behavior (1%) until a cellular damage that was reached where many cells are detached, damaged or round-shaped (25%). Cell counting of oriented cells after applying a mechanical stress showed that a substrate deformation from 2 to 20% is well tolerated by the cells and enable more than 60% of cells to align perpendicularly to the stress direction with a percentage of cell orientation significantly higher respect to cells cultured on control. Our results, obtained after three hours of substrate deformation, suggest that the molecular mechanism which regulates cell orientation involves a really quick response affecting directly cytoskeleton proteins and other structural components (i.e., focal adhesions sites) which have an established role in mechanotransduction, being able to transmit and modulate tension within the cell.

Moreover, within few hours of mechanical stimuli applied to cells cultured on deformable substrates, cells flatten and spread and active filaments form actin stress fibers.¹⁶ The actin filaments resulted aligned perpendicularly to the stress direction, confirming data obtained by Wang and colleagues, which stated that stress fibers can only be formed in a direction with minimal substrate deformation, probably guiding cell spreading in the same direction. In fact, the time frame for formation of this new cell morphology and reorganization of actin microfilaments is consistent with the duration of the experiments used in our work. Our data show that, short stressing time (three hours) is sufficient to induce fibroblasts reorientation and 1% substrate deformation is not sufficient to evidence statistical significant reorientation obtained at ≥2% substrate deformation. After three hours of stretching we found that at constant percentage of substrate deformation, different frequencies applied do not modify cell alignment. These findings support the hypothesis that cell orientation is a defense which allows minimizing the stress on cells. In fact, the alignment direction corresponds to the less strength applied on the cells, thus the substrate deformation has a stronger role compared to the frequency. Thus, at short stretching time, the frequency of mechanical stress applied seems not to have a key role for cell orientation which resulted influenced by stretching magnitude. In fact, it has been evidenced that cell orientation in the angle range 60–90° occur from a 2% to a 20% substrate deformation, with no differences if the stress is applied at low (0.25 Hz) or high (3 Hz) frequencies. Cells do not sustain substrate deformations higher than 25% and do not modify their orientation at magnitude lower than 2% substrate deformation. Actin filaments work as a guide to modify cell alignment, orienting their direction as well perpendicularly to the stress direction. These findings are useful considering their application in tissue engineering in order to mimic as close as possible the physiological conditions as for instance in the cardiovascular system.