Cell Culture

The human cervical cancer cell line, HeLa, was obtained from the American Type Culture Collection (Rockville, MD). The cells were grown in 90% DMEM supplemented with 10% FBS, 100 units/mL penicillin G and 100 μg/mL streptomycin.

The incubation was carried out at 37°C under humidified atmosphere of 5% CO2 in air. The pH values of media and buffers used in this study were adjusted to 7.4.

Medium was filtered through a 0.22-μm-pore size filter before use. Medium was changed twice weekly. All cells were used between four and seven passages after receipt.

3.4 Cell Growth on Micron-patterned Substrates

HeLa cells were cultured in 100-mm tissue-culture dishes for maintenance. When the cells were 80% confluent, the culture medium was replaced with new medium and then subcultured for experiments. To explore the influence of geometrically patterned substrate on cell behaviors, HeLa cells were seeded to patterned PDMS substrate. The sterile PDMS stamps were put into traditional culture dish, and immerged in culture medium. Cell suspension at density of 5×10⁵ cells/ml were added into each dish and further incubated for 24 hours. PDMS stamps, with the characteristic of light transmittable, were used for microscopic observation. In addition, cells were also grown for flowcytometric analysis. After seeding, most cells were attached to the PDMS stamps. There were two types of patterns with the dimension of 500 nm, 1 μm, 5 μm, 10 μm and 20 μm, respectively. The smooth areas between the patterned fields acted as internal controls.

3.5 Cell Cycle Analysis by Flowcytometry

For flow cytometric analysis, a FACSCalibur flow cytometry (Becton Dickinson, NJ) equipped with a single Argon ion laser was used. Forward light scatter (FSC), which is correlated with the size of the cell, and the right-angle light scatter (SSC), which is correlated with the complexity of the cytoplasm, were used to establish size gates and exclude cellular debris from the analysis. At the end of incubation, the silicon wafers were removed to another clear dish. After rinsed twice with PBS, the adherent cells on silicon wafers were trypsinized for detach, washed with PBS, then fixed in PBS-methanol (1:2, volume/volume) solution and, finally, maintained at 4°C for at least 18 h. Following two more washes with PBS, the cell pellet was stained with the fluorescent probe solution containing PBS, 40 μg/ml propidium iodide and 40 μg/ml DNase-free RNaseA for 30 min at room temperature in the dark. DNA fluorescence of PI-stained cells was evaluated by excitation at 488 nm and monitoring through a 630/22-nm band pass filter. A minimum of 10,000 cells were analyzed per sample, and the DNA histograms were gated and analyzed further using Modfit software on a Mac workstation to estimate the percentage of cells in various phases of the cell cycle.

3.6 BrdU Incorporation

At the end of incubation after 48 hr, cells were labeled with 10 μM BrdU for 6 h.

After fixation with ice cold 70% ethanol for 1 h, store at -20°C overnight or for several days. Pellet cells were resuspended in 2 mM HCl/0.5% TritonX-100 and incubated for 30 min at room temperature. After centrifugation, cell pellets were resuspended in 0.5 ml of 0.1 M Na2B4O7, pH 8.5, and after a PBS wash, cells were stained with 1ml of antibody solution (1 ml PBS containing 0.5% Tween-20/1% BSA + 10μl FITC-conjugated mouse anti-BrdU monoclonal antibody) for 1 h at room

temperature in the dark, washed, and resuspended in 5 mg/ml of propidium iodide solution. After 30 min, the cells were analyzed by two-dimensional flow cytometry.

3.7 Western-Blot Analysis

3.7.1 Extraction of Whole­cell Protein

After 24 hr of culturing, cells were scrapped from culture dishes or plates with a policeman and collected in 1.5 ml of eppendorfs. Cell pellets were spined down at 3,000 g at 4℃ for 5 min and the supernatants were discarded. Cell pellets were resuspended with proper amount of cell lysis buffer, incubated on ice for 30 min, and disrupted by vortex every 5 min. After that, cell lysate was centrifuged at 14,000 g at 4℃ for 15 min and transferred to a new 1.5 ml of eppendorf. Proteins were stored at

­80℃.

3.7.2 Determination of Protein Concentration

The protein concentration was measured by the Bio­Rad protein assay kit. Protein samples were diluted in 1/25 with 1x PBS. For quantification, the standard BSA protein was diluted from 400 μg/ml to 25 μg/ml by a 1:2 serial dilution. 10 μl of each standard diluent or protein sample was pipetted into 96­well ELISA plates in triplicate. 200 μl of dye reagent was added into each sample­containing well and another three empty wells for blank. The absorbance of protein solution was read at 590 nm by ELISA reader.

3.7.3 SDS­PAGE and Immunoblotting

About 30­50 μg of cell lysate was mixed with 3x sample buffer in a 1.5 ml of eppendorfs, denatured by boiling water for 10 min then chilled on ice. Samples were loaded into 10% polyacrylamide gel, stacked at 80 V for 30 min, and then resolved at

100 V for 60­90 min. Separated protein was transferred onto PVDF membranes in protein transfer buffer at 330 mA for 60­90 min. Membrane blocking was done by rinsing the PVDF membranes with TBST containing 5% skim­milk and incubated at 37℃ for 1 hour with shacking. The PVDF membranes were then incubated overnight at 4℃ with primary antibodies in proper dilution. After removal of the primary antibodies, the PVDF membranes were washed with TBST three times with five min each. Hybridization of secondary antibody was done at 37℃ for 1 hour. The membranes were washed with TBST three times with five min each. Immuno­

reaction protein bands were visualized in ECL chemiluminescence system.

3.8 Gelatin Zymography

Gelatinase secretion of cell culture supernatants was determined by zymography with gelatin as substrate for MMP-2 and MMP-9. Briefly, serum-free conditioned medium was collected from confluent culture of the cells after 24 hr culturing. 20 µl were mixed with sodium dodecyl sulphate (SDS) sample buffer, and without prior denaturation, were run on an 10% SDS-polyacrylamide gel electrophoresis containing 1 mg/mL of gelatin. After electrophoresis, the gels were washed in 2.5% Triton X-100 for 1 h at room temperature to remove the SDS and incubated for 48 h at 37°C in a renaturing buffer containing 50 mmol/L Tris (pH 7.5), 10 mmol/L CaCl2, 150 mmol/L NaCl, and 0.05% NaN3 to allow digestion of the gelatin. The gels were subsequently stained in a solution of 0.25% Coomassie brilliant blue G-250 for 30 minutes and destained for 1 h with acetic acid and methanol. Proteolytic activity appeared as clear bands (zones of gelatin degradation) against the blue background of stained gelatin. For quantitation, the bands were scanned and densitometry was performed.

3.9 Statistical Analysis

All numerical values reported represent mean values ± SD. The statistical significance comparing differences between the experimental and control values was evaluated using Student's t test. Asterisks were used to graphically indicate statistical significance (P < 0.05) in the figures. All experiments were performed in at least triplicate. Figures were derived from representative experiments.

Chapter 4: Results and Discussions

4.1. Fabrication and characterization of systematic effects of patterned elastic PDMS substrate on cell adhesion.

In order to develop next-generation tissue engineering materials, the understanding of cell responses to novel material surfaces needs to be better understood.

Topography presents powerful cues for cells, and it is becoming clear that cells will react to micro-scale surface features, intensely.

Two types of patterned PDMS membranes were fabricated in this study as shown in Fig. 4.1. One is one dimensional (1D) periodic lines with equal width and space. The other one is two dimensional (2D) arrayed pillars and dimension of the pillars is half of their period. Fig. 1B-F represented the cells grew on 1D periodic line/space pattern.

Cell elongation and alignment on grooves and ridges were observed and in Fig. 1G-K showed the cells grew on 2D periodic pillar pattern. By contrast, when the cells grew on the substrate of 2D periodic pillar pattern, they showed poor adhesion and spreading properties accompanying with growth retardation. The dimension of patterns is 500 nm, 1 μm, 5 μm, 10 μm and 20 μm, respectively. The smooth areas between the patterned fields were acted as internal controls (A). All patterns shown here were on the same PDMS substratum. The cells were grown in the same condition except the difference on substrate patterns.

Figure 4.1 Effects of different patterned substrates on cell morphology and growth. (A) Cell cultured on a smooth silicon oxide substrate which serves as an internal control surface. (B-F) represented the cells grew on 1D periodic line/space pattern. (G-K) showed the cells grew on 2D periodic pillars pattern. The dimension of pattern is 500 nm, 1 μm, 5 μm, 10 μm and 20 μm, respectively.

To compare the gradient dimension of patterns, we found that HeLa cells did not spread on 2D periodic pillars when pillar diameter is larger than 5 μm and the adherent cells decreased by increasing pillar dimension (Fig.1I-K). So, here we chose the dimension of 1 μm (1D) periodic lines and (2D) arrayed pillars for our further study because cells adhered on these patterns with conspicuous elongate on 1D periodic lines and opposite circular form on 2D periodic pillars.

This result is consist with the report of Teixeira et al. who demonstrated that [61]

with 600 nm deep grooves, the percentage of aligned cells was constant on patterns with pitches ranging from 400 nm to 2000 nm and decreased on 4000 nm pitch patterns.

Figure 4.2 SEM images of flat PDMS surface (A), 1D periodic line/space (1μm width) PDMS surface (B) and 2D periodic pillar (1μm diameter) PDMS surface (C).

Optical microscope images of HeLa cells cultured for 24h on bacteriological grade polystyrene Petri dish (D), flat PDMS surface (E), 1D periodic line/space patterned PDMS surface (F), and 2D periodic pillar patterned PDMS surface (G). (H) The average number of adherent HeLa cells on different patterned PDMS surface.

Statistical significance assessed by one-way ANOVA tests is shown as *:p＜0.05 when compared with polystyrene Petri dish control and #: p＜0.05 when compared with flat PDMS control.

After the patterned and non-patterned PDMS is carefully removed from silicon mold, the images of features on PDMS surface were collected by scanning electron microscope (shown in Fig. 4.2A-C). Feature pitches (sum of the groove and ridge widths) were uniform in each field. And 2D dense lines and dense dots are both on a 2μm pitch with 1μm line width and 1μm dot diameter, both patterns were about 350nm depth. HeLa cervical cancer cells were cultured on patterned and non-patterned PDMS surface for 24hr and direct observed of living cells under an

Petri-dish PDMS 1D periodic ridge/groove

1D periodic pillar array

optical microscope. As shown in Fig. 2D-G, when cells cultured on the 2D periodic pillars and smooth PDMS substrates, cells aggregated and were mostly round and worse adherent. Contrast with cells growth on 1D periodic lines which expressed a normal morphology similar to cells adhere on Petri dish. The growth rate was further quantified. Equal numbers of the cells were seeded at identical conditions and the number of cells at 24 h after seeding was counted at different three positions on each substrates. Relative to those on flat PDMS and 2D periodic pillars with 1μm wide diameter (p＜0.05, Fig. 4.2H), the average number of adherent HeLa cells (cells/mm²) after 24hr of culturing was significantly greater on 1D periodic lines with 1μm wide ridges/grooves PDMS surface which revealed no significant differences with Petri dish control.

Furthermore, according to the Ar/O2-based plasmas treated PDMS surface, untreated PDMS is not a good substrate for supporting cells adhesion and growth well as an extracellular scaffold because of the higher the hydrophobicity, the lesser the cell adhered and cell surface [62]. In agreement on the characteristic of PDMS substrate, we found it was more bio-compatible than untreated smooth PDMS when 1D periodic line patterned on, it is likely that the topography has this positive effect on cell survival (improved adhesion is usually associated with improved survival).

4.2. Cells adhere to flat PDMS and 2D periodic pillars patterned PDMS substrate show a retarded G1/S transition.

Petri-dish PDMS 1D periodic 1D periodic

Figure 4.3 Analysis of cell cycle phases. The cell cycle of HeLa cells cultured on smooth and patterned PDMS substrates for 24h. Cellular DNA was stained with propidium iodide and analyzed by flow cytometry. (A) The DNA content of cells of the cell cycle. (B-C) A histogram represents the percentage of total cells in G1 phase, S phase and G2/M phase, respectively. Statistical significance assessed by one-way ANOVA tests is shown as *:p＜0.05 when compared with polystyrene Petri dish control and #: p＜0.05 when compared with flat PDMS control.

The cell cycle was investigated by comparing the DNA contents of the HeLa cells cultured on Petri dish, smooth and patterned PDMS substrates for 24h and analyzed by flow cytometry (Fig. 4.3A). The DNA content of HeLa cells cultured on Petri dish

Petri-dish PDMS 1D periodic ridge/groove

1D periodic pillar array

Petri-dish PDMS 1D periodic ridge/groove

1D periodic pillar array

showed a cell cycle distribution typical for exponentially growing cells. HeLa cells adhered to flat PDMS and 2D periodic pillars PDMS surface showed a decreased percentage of cells in the S phase and, in contrast an increased percentage of cells in G1 phase (Fig. 4.3B-C). The proportion of tetraploid cells in G2/M phase was much lower when cells adhere on 1D periodic line than on the other surface (Fig. 4.3D).

The reduced S phase in the HeLa cells which adhered to flat PDMS and 2D periodic pillars PDMS surface were further verified by measuring the BrdU incorporation. The BrdU incorporation was determined by immunofluorescence and a similar trend was noted for percentage of cells in S phase, as shown by BrdU incorporation. (Fig. 4.4)

Figure 4.4 BrdU incorporation in HeLa cells cultured on Petri dish, smooth and patterned PDMS substrates. Exponentially growing HeLa cells were incubated with BrdU and the incorporation rate was analysed by immunofluorescence. Statistical analysis was done using ANOVA. The star marks the statistically significant difference in BrdU incorporation at P＜0.05 when compared with polystyrene Petri dish control and # marks the statistically significant difference at p＜0.05 when compared with flat PDMS control.

Cells adhered on flat PDMS and 2D periodic pillars PDMS substrate showed a significantly decreased incorporation of BrdU. The impaired BrdU incorporation

displayed essentially unchanged on flat PDMS and 2D periodic pillars PDMS surface and was maximal with only 50% BrdU incorporation as compared with those adhered on 1D periodic lines surface. These data confirm the finding of S phase obtained by the analysis of the cellular DNA content, and in agreement with the theory that if a cell cannot flatten fully, it will not enter the S-phase of the cell cycle so readily [63].

4.3. Protein expressed by integrin-mediated intracellular signal transduction.

(A)

Figure 4.5 (A) Immunoblotting of transmembrane integrin proteins (α5); focal adhesion-associated adaptor protein, paxillin; tumor suppressor protein, p53 (B) Activity of matrix metalloproteinase 9 (MMP-9) analyzed by gelatin zymography.

Different surface topography regulates integrin expression occurs selectively on specific integrin subunits [64]. α5β1 integrin is a cell surface receptor that mediates cell extracellular matrix adhesions by interacting with fibronectin. Studies have shown that FN-α5β1 interaction regulates a variety of cellular responses including gene induction [65], oncogenic transformation [66], differentiation [67], proliferation and cell survival [68] and adhesion and migration [66,69]. Numbers of cranial neural crest cells are undergoing apoptosis along their migration pathways in α5 subunit-deficient embryos could be due to failure of these neural crest cells to migrate to their correct destinations. Nevertheless, they found that α5 is not essential for the survival or normal proliferation of mesodermal cells[70]. In contrast with in vitro studies of Zhang et al. [68] α5 protects against cell death in cultured chinese hamster ovary (CHO) cells, perhaps the difference in the cell growth behavior might be caused by the type of cells involved or differences in cellular environment presented.

Up till now it is not clear how topography affects integrin expression.

In our study, western blotting revealed that α5 integrin displayed the variations with respect to micro-scale patterns during the 24hr culture period (Fig. 4.5 A). Level of integrin α5 expression was higher in HeLa cells cultured on 1D periodic lines close to

(B)

the cells on Petri dish. The variation of integrin α5 is generally consistent with the tendency of cell attachment (Fig. 4.2H) suggesting that micro-topographic PDMS surfaces effect on cells may be mediated by integrin α5 expression.

Paxillin is a cytoskeleton protein involved in actin-membrane attachment at sites of cell adhesion to the extracellular matrix (focal adhesion) [71]. The predicted structure of paxillin suggests that it is a unique cytoskeleton protein capable of interaction with a variety of intracellular signaling and structural molecules important in growth control and the regulation of cytoskeleton organization [72]. According to these reports, our results also found that when cells adhered to 1D periodic lines they spread well accompanying with integrin α5 and paxillin expresse. That is to say, reducing integrin clustering may have resulted in reduced transduction of cell signals to the nucleus, and therefore low rates of proliferation and tissue formation (as shown in Fig. 4.5A and bromodeoxyuridine incorporation above-mentioned).

It has been reported when fibroblast contacts a material surface, it must adhere first of all, otherwise it will undergo apoptosis via anoikis (which means homelessness in Greek) [73]. Furthermore, in 1998s, Almeida et al. first report that p53 monitors survival signals from ECM/FAK in anchorage-dependent cells, if FAK or the correct ECM is absent, cells enter apoptosis through a p53-dependent pathway. And this pathway is suppressible by dominant-negative p53. In other words, upon inactivation of p53, cells survive even if they lack matrix signals or FAK [74]. For this reason, it’s important to understand how signals from ECM suppress cell death, and what apoptotic pathway is triggered in cells when these signals are lost. In order to investigate the correlation between the signals from external topography and the transduction via transmembrane protein integrin α5 and the anoikis when cells are detached from the ECM, we analyzed the p53 expression on HeLa cells after 24hr culturing on different patterned PDMS surface. Interestingly, there was significantly

reduced p53 as cells cultured on 1D periodic line. In agreement with the theory of p53 controls both the G1 and the G2/M checkpoints and mediates growth arrest [75], cells represented a retarded G1/S transition significantly as HeLa cells adhering to 2D periodic pillars and flat PDMS surface compared with polystyrene Petri dish control (p＜0.05, Fig. 3). Oppositely, cells adhered on 1D periodic line with much lower p53 promoted cells to progress into S phase.

Previous reports have shown that MMP-9 was expressed and contributed to early-stage B-cell chronic lymphocytic leukemia (B-CLL) migration through artificial basement membranes or endothelial cells, moreover also contributes to B-CLL progression by facilitating malignant cell migration and tissue invasion [76]. Thus, MMP-9 plays a key role in cell invasion and transendothelial migration and the physiologically up-regulated by integrin also be demonstrated in B-CLL [77].

In this study, we investigated the relation between cell adhesion and expression of active MMP-9 in human epithelial carcinoma HeLa cell line, and also found that accompany with increased α5 integrin protein there were highly expressed MMP-9 when HeLa cells adhered to 2D periodic pillars surface (Fig. 4.5B).. Furthermore, p53 is a negative regulator of MMP-9 gene expression [78], it indicated that MMPs are implicated in tumor cell resistance to the synergistic proapoptotic effect of p53.

This is similar to our finding above-mentioned that when HeLa cells were cultured on 1D periodic lines and Petri dish surface, they grow well with suppressed p53 expression. On the contrary, as the HeLa cells adhered on 2D periodic pillars, they showed poor adhesion and spreading properties and accompanied with MMP-9 expression to protect against the proapoptotic effect of p53. For these reasons, we suggest when cells adhere to the proper position of ECM through receptor-integrin interaction, cell progressed and if cells adhere to a worse environment, invasion can be initiated.

4.4. Investigation of cell morphology using scanning electron microscopy.

Figure 4.6 Morphology of human epithelial carcinoma HeLa cells cultured on Petri dish (A), smooth bare silicon (native oxide) surface (B), 1D periodic line silicon (native oxide) surface, and 2D periodic pillar silicon (native oxide) surface for 24hr imaged by scanning electron microscopy. Areas of lower cell density were selected to facilitate observation of individual cell shapes. The images of the cells shown in the selected micrographs are typical of cells throughout the culture.

The cells exhibited different shapes on Petri dish, patterned and non-patterned

The cells exhibited different shapes on Petri dish, patterned and non-patterned