Variation in cell–substratum adhesion in relation to cell cycle phases

Experimental Cell Research 293 (2004) 58 – 67 www.elsevier.com/locate/yexcr

Variation in cell–substratum adhesion in relation to cell cycle phases D.O. Meredith, a G.Rh. Owen, a I. ap Gwynn, b and R.G. Richards a,* a

b

Interface Biology, AO Research Institute, Davos-Platz, Switzerland Institute of Biological Sciences, The University of Wales, Aberystwyth, UK Received 6 June 2003, revised version received 18 September 2003

Abstract The quantification of focal adhesion sites offers an assessable method of measuring cell – substrate adhesion. Such measurement can be hindered by intra-sample variation that may be cell cycle derived. A combination of autoradiography and immunolabelling techniques, for scanning electron microscopy (SEM), were utilised simultaneously to identify both S-phase cells and their focal adhesion sites. Electronenergy ‘sectioning’ of the sample, by varying the accelerating voltage of the electron beam, combined with backscattered electron (BSE) imaging, allowed for S-phase cell identification in one energy ‘plane’ image and quantitation of immunogold label in another. As a result, it was possible simultaneously to identify S-phase cells and their immunogold-labelled focal adhesions sites on the same cell. The focal adhesion densities were calculated both for identified S-phase cells and the remaining non-S-phase cells present. The results indicated that the cell cycle phase was a significant factor in determining the density of focal adhesions, with non-S-phase cells showing a larger adhesion density than S-phase cells. Focal adhesion morphology was also seen to correspond to cell cycle phase; with ‘dot’ adhesions being more prevalent on smaller non-S-phase and the mature ‘dash’ type on larger S-phase cells. This study demonstrated that when quantitation of focal adhesion sites is required, it is necessary to consider the influence of cell cycle phases on any data collected. D 2003 Elsevier Inc. All rights reserved. Keywords: Cell cycle; Focal adhesions; Immunocytochemistry; Scanning electron microscopy; Tritiated thymidine; Autoradiography

Introduction In vivo and in vitro, nontransformed cells require attachment as a stimulus to proliferate (anchorage dependency) [1]. In vitro, cells attach to a substratum, usually glass or plastic, with several different mechanisms. These include extracellular matrix (ECM) contacts, close contacts and focal adhesions. These adhesive areas can be differentiated by their proximity to the substratum [2,3]. The closest and strongest adhesive points, normally within 10 –20 nm to the substrate, are focal adhesions [4,5]. A focal adhesion is a cascade of proteins situated near the capped end of cytoskeletal actin filaments. The complex terminates with transmembrane proteins known as integrins [6,7] that adhere to adsorbed extracellular matrix components on the substrate surface. Not only do these focal adhesions provide anchorage, but they also provide a channel for intracellular signal transduction [1,8 – 16], reg-

* Corresponding author. Interface Biology, AO Research, Clavadelerstrasse, CH-7270 Davos-Platz, Switzerland. Fax: +41-81-4142-288. E-mail address: [email protected] (R.G. Richards). 0014-4827/$ - see front matter D 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2003.10.005

ulation of cytoskeletal dynamics [15] and cell cycle progression [8,13,17]. Quantification of the number and area of focal adhesions is sought as an assessable method of measuring cell adhesion and cytocompatibility. Evidence from numerous studies suggests that focal adhesion complexes have a finite strength [18]. Therefore, the adhesion strength of a cell could be directly attributed to the number and size of focal adhesions a cell expresses. A variety of substrata have been used to determine whether they would induce the formation of different amounts of focal adhesions in the cells cultured upon them. Hunter et al. [19] utilises an immunofluorescence measuring technique while Richards et al. [20] modifies it for electron microscopical analysis. Difficulties arise from the occurrence of intra-sample variation in focal adhesion measurements. Considerable variation is found in the adhesive interactions of normal adherent cell populations [21,22]. Davies et al. [23] comments that the numbers of focal adhesion sites can vary by up to 10% on normal growth substrates. Richards et al.’s [20] biomaterial comparison is hindered by similar variability within substrate groups. It is probable that this cell adhesion variation derives from the cell cycle [24,25].

D.O. Meredith et al. / Experimental Cell Research 293 (2004) 58–67

The cell cycle is the growth and proliferation cycle of the cell [26]. It has two distinctive stages: interphase and mitosis. Interphase is concerned with an increase in cell mass and replication of the cells’ DNA [27,28]. This is split into three stages, respectively, G1, S and G2. Cell growth takes considerably longer than DNA replication and continues throughout interphase. Replication of the cells’ DNA takes a shorter period of time and is confined to the ‘S-phase’. Upon entering mitosis, the cell sets about distributing the DNA into two complete sets, and cleaving the cell into two daughter cells (cytokinesis) [8,29]. Significant rearrangement of the cytoskeleton occurs during this process, resulting in visible changes in the cell’s morphology—indicative of the cycle stage [1,30]. This rearrangement also signifies movement and reorganisation of cytoskeletal terminal complexes such as focal adhesions [24,25,31]. The cell cycle-regulated rearrangement of focal adhesions may be the cause of the variation found when attempting to quantify cell –substrate adhesion. This paradigm was tested here by quantifying the adhesion of an identified cell phase and comparing it to that of residual cells. A measurement of focal adhesion density (the percentage of the entire cell – substratum contact area identified to be composed of focal adhesions) provided a more meaningful assay of cell –substrate adhesion. Hunter et al. [19] observed that flattened cells display the greatest number and area of focal adhesions. The strength of adhesion of these flatter cells, to the substratum, was also greater [18]. A flattened morphology is a prominent feature of S-phase cells [8,32]. Therefore, it would be reasonable to assume that the most adhesive cells (cell to substratum) are found during the S-phase. S-phase is a distinctive morphological and physiological cell cycle stage with unique characteristics that can be exploited in its identification. It is during the S-phase that nucleosides are incorporated into the cell’s DNA and a duplicate set of chromosomes are synthesised [27,28]. When radiolabelled (tritiated) thymidine is introduced into the cell growth medium, S-phase cells incorporate an indelible marker for identification directly into the nucleus, without causing any systemic disruption to the cell. Vinculin is an integral and well-characterised component of the focal adhesion protein complex and is especially well conserved throughout many species [33]. Its omission or overexpression in transformed cells strongly affects cell adhesion and motility [34,35]. Identifying vinculin by means of a specific antibody provides a definite detection system for focal adhesion sites [36]. The combination of thymidine labelling of the nucleus, detected by autoradiography, with the immunocytochemical labelling of vinculin realises the possibility of identifying Sphase cells and visualising their focal adhesion sites [37]. Quantification of the focal adhesion sites, by image analysis, can determine the focal adhesion density a cell produces during the S-phase of the cell cycle and the residual non-‘S’

59

cell phases. The comparison of this data provides the basis for determining the involvement of the cell cycle on cell – substratum adhesion.

Materials and methods Cell culture and seeding Cell culture techniques complied within UK Coordinating Committee on Cancer Research (UKCCCR) guidelines for cell culture (http://www.ncrn.org.uk/Csg/publications. htm#cell). Swiss Balb/c 3T3 (mouse murine mesenchymal embryo fibroblasts, ECACC #85022108) were plated on polyethylene terephthalate plastic (Thermanox, Nunc, USA) at a cell density of 40,000 cells/ml in a 24well plate with 0.5 ml of Dulbecco’s Modified Eagle Medium (DMEM) solution containing 10% foetal calf serum (FCS), without the addition of antibiotics [38]. Seeded cells were allowed 48 h for settling and spreading on the surface. Application of tritiated (6-3H) thymidine pulse Autoradiographic identification of S-phase cells was conducted following a method described by Owen et al. [37]. The equivalent of 4 ACi/ml of tritiated (6-3H) thymidine (Amersham Pharmacia Biotech Europe GmbH, Dubendorf, Switzerland) was placed in the culture medium and mixed thoroughly. Incubation time was 30 min. The radioactive medium was then ‘chased’ out with nonradioactive medium for 2 h and the cells were then processed for immunolabelling. All procedures were carried out at 20jC. Immunogold labelling The immunolabelling method is described previously [36,39]. Following the ‘pulse/chase’, the cells were rinsed in 0.1 M PIPES buffer, permeabilised in 0.1% Triton for 1 min, fixed in buffered 4% paraformaldehyde for 5 min, rinsed again in buffer and then nonspecific antigenic sites were blocked with buffered 1% bovine serum albumin (BSA) and 0.1% Tween 20 for 15 min. Cells were then incubated in a solution of mouse antihuman vinculin (clone hVin-1; SIGMA, USA) at a dilution of 1:300 for 1 h and nonspecific binding sites were blocked with 5% goat serum in buffered 1% BSA and 0.1% Tween 20 for 15 min. The cells were labelled with goat anti-mouse 5 nm gold conjugate (BBI, Cardiff, UK) at 1:200 dilution for 2 h. Cells were then fixed permanently with buffered 2.5% glutaraldehyde for 5 min. The enlargement of the gold probes was performed by gold enhancement for 5 min (Nanoprobes, New York, USA) [39]. Additional contrasting of the cells was accomplished by incubating with 1% osmium tetroxide, buffered to pH 6.8, for 1 h.

60

D.O. Meredith et al. / Experimental Cell Research 293 (2004) 58–67

The cells were dehydrated through an ethanol and Fluorisol series, critical point dried, mounted on stubs and coated with a thin layer of carbon (4 nm). This carbon coat serves as an insulator to reduce ‘background’ labelling of the sample. Only areas with high levels of tritium incorporation would yield a signal detectable by the nuclear emulsion.

within the polymerised resin were mounted on SEM viewing stubs and coated with 10 nm carbon. Cells were imaged using high current backscattered electron (BSE) detection mode [41] in a Hitachi S-4100 FESEM, using an Autrata yttrium aluminium garnet (YAG) BSE detector, at 4 and 8 kV accelerating voltages. Image and statistical analysis

Electron microscopy autoradiography (tritiated thymidine detection) Nuclear emulsion was prepared and used to thinly coat the samples. The h-radiation emitted from the tritium, incorporated into S-phase cells, denatures the silver bromide crystals suspended in the emulsion directly adjacent S-phase cell nuclei. When developed, all the emulsion was washed away leaving only the silver adjacent to the sites of tritium location. All the work involving the emulsion was carried out in a dark room at 20jC, except where stated differently, with the appropriate safelight (Ilford 904, Ilford Imaging UK Limited, Knutsford, UK). The nuclear emulsion was produced by dissolving 2.5 g of Ilford L4 emulsion in 5 ml ultra-pure water, at 45jC for 20 min. The emulsion was cooled in an ice bath for 5 min and then left at room temperature for 30 min in a light tight box. Samples were coated with a thin film of emulsion using the ‘loop method’ [40]. The method involved dipping a retractable loop of platinum wire into the liquid emulsion. A film of emulsion was produced and extension of the loop created a large film area, similar to the sample size. This film was then passed ‘through’ the stub mounted on a stand giving the sample a thin even coat of emulsion. The coated samples were kept in light tight boxes wrapped in aluminium foil for 1 week in a refrigerator at 4jC. The samples were placed into a polystyrene holder so that development could be conducted in batches. The samples were developed by exposure, at 20jC, to the following chemicals: (a) undiluted developer (Kodak D19, USA), for 1 min; (b) double-distilled water wash with two changes at 10 s duration prewash; (c) fixer (Ilford paper fix) 1:3 dilution for 2 min; (d) two subsequent long washes in ultra-pure water.

Modifying the method of Richards et al. [20], PC_ Image 2.2.03 image analysis software (Foster Findlay Associates, Newcastle, GB) was used to quantify the amount of gold label present on selected cells. Digital electron micrographs of the cells (4 kV) were imported into PC_Image. The image was calibrated so that measurements related to actual specimen dimensions. A ‘region of interest’ (ROI) was drawn freehand around each cell; this highlighted the cell area exclusively for consequent steps (Fig. 1a). The image of the cell area was modified using the ‘Square’ function; this increased the contrast in the higher/whiter grey-level regions, at the expense of the darker regions (Fig. 1b). The image ‘threshold’ values were set and a binary image formed to represent the result of applying the ‘thresholding’ process (Figs. 1c and d). Setting a threshold provides grey-level boundary values. A pixel is only recorded on the binary image if the greylevel value for the corresponding pixel falls within the set range. Fifty identified S-phase and fifty non-S-phase cells, all immunolabelled, were analysed using the PC_ Image analysis protocol. The measurement arrays were exported to Excel 2000 (Microsoft) for initial analysis and sorting. From the array of measurements generated, two PC_ Image outputs were utilised as statistical inputs; ‘contact area’ and ‘adhesion density’. ‘Contact area’ denoted the area of the cell in contact with the substrate, measured in Am2. The ‘adhesion density’ was calculated as the percentage of the contact area of the cell that was identified by the application of the threshold feature and then measured (i.e. the immunolabel). In addition to these, the ‘contact area’ and ‘adhesion density’ were used to calculate the total area of adhesion present on the cells in Am2. This was calculated using the formula shown below;

Scanning electron microscopy (SEM) preparation and imaging The samples were left to dry in a dessicator, containing a copper sulphate bed, for 24 h. The SEM stubs were removed and the dehydrated samples were placed in silicone rubber moulds and immersed in LR White resin, leaving the samples for 8 h, at 4jC, to allow resin penetration. The resin was substituted by fresh resin, sealed with paraffin oil (to keep from oxygen contact) and cured thermally at 55jC for 12 h. The Thermanox substrates were then removed, exposing the cell undersurface. The cells

 Contact area

Adhesion density 100

 ¼ Total adhesion area

The data for the three factors (inputs) were then processed using the SPSS Base 10.0 (SPSS Inc., USA) statistical analysis package. For the statistical analysis, the input differences were first compared with an ‘independent t test’. The variations were then displayed graphically using box plots produced from the three data sets. Normality of the

D.O. Meredith et al. / Experimental Cell Research 293 (2004) 58–67

61

Fig. 1. Image analysis protocol. (a) A ‘region of interest’ (ROI) was drawn freehand around each cell; this highlighted the cell area exclusively for consequent steps. (b) The image of the cell area was modified using the ‘Square’ function; this increased the contrast in the higher/whiter grey-level regions, at the expense of the darker regions. (c) The image ‘threshold’ values were set and a binary image formed to represent the result of applying the ‘thresholding’ process (c and d). Setting a threshold provides grey-level boundary values. A pixel is only recorded on the binary image if the grey-level value for the corresponding pixel falls within the set range.

distributions was tested using ‘Shapiro –Wilks’ test and the homogeneity of the variants between groups was examined using ‘Levene’s Test’. Statistical significance was assumed at a level of P V 0.05.

Results Cells embedded within the resin were imaged at an accelerating voltage of 8 kV to visualise both the whole

Fig. 2. Cells embedded within the resin were imaged at an accelerating voltage of 8 kV to visualise both the whole cell and labels (autoradiographic marker and immunolabelled focal adhesions), whereas a lower beam energy of 4 kV was used to visualise exclusively the labelled focal adhesions.

62

D.O. Meredith et al. / Experimental Cell Research 293 (2004) 58–67

cell and labels (autoradiographic marker and immunolabelled focal adhesions), whereas a lower beam energy of 4 kV was used to visualise exclusively the labelled focal adhesions (Figs. 2 and 3).

Table 1 Statistical observations Cell contact area measurements

S-phase cells were found to have significantly greater contact area than non-S-phase cells (14.7 F 1.6 vs. 9.5 F 0.9, mean F SE, P = 0.001) (Fig. 7a). adhesion density The adhesion density was significantly less for S-phase than non-S-phase cells (1.3 F 0.1 vs. 2.3 F 0.2, mean F SE, P = 0.002) (Fig. 7b). total adhesion There was no difference in adhesion area area between S- and non-S-phase cells (0.2 F 0.04 vs. 0.3 F 0.04, mean F SE, P = 0.8) (Fig. 7c).

SEM observations Viewing all the images of both identified S-phase cells and non-S-phase cells (Fig. 3), it appeared that the immunolabelled focal adhesions became more organised as the cell contact area increased (Figs. 5 and 6). This organisation manifested itself in a less ‘random’ distribution of the labelled focal adhesions, seen in rounded cells (Figs. 5a and b), progressing towards lines of adhesion originating at the furthermost points of the cell periphery and spreading inwards towards the nucleus (Figs. 5c and d). The identified S-phase cells were generally the larger cells in terms of surface contact area (Figs. 6a– d). Some S-phase cells showed what appeared to be significant labelling underlying the area of the nucleus. As the larger non-S-phase cells also has some apparent label patterning underlying their nuclei (Fig. 5d), it was reasonable to assume that this label originated from the immunocytochemical labelling of vinculin.

Variations

Cell contact area varied more for S-phase cells than non-S-phase cells (Fig. 8a). Cell adhesion density was less varied for S-phase cells than non-S-phase cells (Fig. 8b).

accurately from the generalised statistical tests. Therefore, the non-S-phase data could only give a general indication of a ‘typical’ cell’s focal adhesion density when not in S-phase.

Statistical observations Discussion All of the distributions were skewed to the left; therefore, the input variables were log10 transformed. The following observations could be gathered from the data (Table 1). Non-S-phase cell data encompassed all that were derived from the remaining cell phases. During these phases the cells were subject to significant changes. Their specific adhesion or contact area could not be determined

Fig. 3. Autoradiographically identified S-phase cells are clearly distinguishable from their unlabelled counterparts cells embedded in resin. The patterned arrow identifies a labelled nucleus while the white arrow singles out an unlabelled cell.

It was hypothesised that focal adhesions are rearranged during the cell cycle. This would result in a variation of focal adhesion density. The S-phase cells were expected to show the most focal adhesions, as they should have been the most adhesive cells. The SEM images indicated that the immunolabelled focal adhesions had become more organised as the cell contact area increased. The apparent randomness of focal adhesions on the smaller cells could be explained by the ‘dot’ and ‘dash’ morphology of focal adhesions (Figs. 4a and b) [42]. Bershadsky et al. showed that focal adhesions are initiated in the form of small ‘dot’ adhesion points, about 0.2– 0.5 Am in diameter, located at the active edge of a cell (Fig. 4a). Such ‘dot’ adhesions were dispersed over the cell undersides, rather than being concentrated at the cell periphery (Figs. 5a and b). Bershadsky et al. further stated that subsequently some of adhesion points ‘matured’ into larger ‘dash’ contacts, 2 –10 Am long and 0.5 Am wide (Fig. 4b). Such ‘dash’ adhesions were dispersed in the lamellae, endoplasm and below the nucleus. Only as part of the maturation process of the adhesion point, from ‘dot’ to ‘dash’, did the actin microfilaments attach to the adhesion point [42]. In this work, the mature focal adhesions detected were predominantly located at the cell periphery (Figs. 6a– d). The adhesion morphology transition from ‘dot’ to ‘dash’ coincided with the change from smaller to larger contact areas and could indicate that cell spreading required the

D.O. Meredith et al. / Experimental Cell Research 293 (2004) 58–67

Fig. 4. (a) Example of small ‘dot’ adhesions points. (b) Example of larger ‘mature’ dash adhesion areas called ‘dash’ contacts.

development of mature anchorage points. Additionally, the S-phase cells were significantly larger, in terms of surface contact area. Larger non-S-phase cells did display ‘dash’ adhesions, but the data set was derived predominantly from

63

smaller cells, displaying dot adhesions. The focal adhesions were probably rearranged when the cell entered S-phase. The ‘dot’ adhesion morphology found in non-S-phase cells agreed with current opinion on cytokinesis, whereby cells in mitosis might have two reasons not to assemble mature focal adhesions [21,25,43]. Firstly, the cytoskeletal role is modified so that connections to adhesion sites are disassembled and the actin is relocated and reorganised into the rounding cortex of the cells, therefore downplaying its role in adhesion [25]. Secondly, the nonpermanent adhesions (dot) allow the cells to adjust their adhesion as mitosis progresses and shift the function of cell adhesion from anchorage to migration. Migration is required to terminate the cleavage process by severing the intercellular connection left by the incomplete separation of the daughter cells [43]. Mature focal adhesions, shown here to be prevalent in Sphase cells, are related to anchorage dependency, signal transduction and cell-shape mediated growth. The anchorage dependency phenomenon provides evidence confirming that there is an intrinsic link between adhesion and proliferation control, in nontransformed cells. A cell in vitro is required to adhere and remain adhered to the substratum well into the G1 phase, before it can progress further in the cell cycle. The trigger is a G1 checkpoint, controlled by the transduction of signals from outside to the inside of the cell [28]. This checkpoint signals the end of required external stimulation by growth factors—in vitro these factors are supplied by the serum [44,45]. Most of the signalling is

Fig. 5. Examples of non-S-phase cells. The immunolabelled focal adhesions become more organised as the cell contact area increases. This organisation manifests itself in a less ‘random’ distribution of the labelled focal adhesions, seen in rounded cells (a and b), progressing towards lines of adhesion originating at the furthermost points of the cell periphery and moving inwards towards the nucleus (c and d).

64

D.O. Meredith et al. / Experimental Cell Research 293 (2004) 58–67

Fig. 6. (a – d) Examples of identified S-phase cells. These were generally the larger cells in terms of surface contact area and had a ‘dash’ focal adhesion morphology located predominantly at the cell periphery.

transmitted through the focal adhesions [1,8 – 13]. The phenomenon known as ‘cell-shape mediated growth’ requires that the cell has to spread to a sufficient degree before DNA synthesis, the S-phase, can occur [1,32]. All these phenomena require the presence of mature focal adhesions with attached cytoskeleton. Cell adhesion measurement, defined as ‘adhesion density’ demonstrated that there was a significant difference between the adhesion of S-phase and non-S-phase cells. This suggested strongly that the cell cycle was the source of significant variation when attempting to quantify cell adhesion. It was interesting that the ‘adhesion density’ values pointed to the non-S-phase cells having the greater adhesion (Fig. 7b). In mechanical terms, this result appeared logical. The non-S-phase cells displayed a more spherical morphology, with less contact area with the substrate surface (Fig. 7a). Such a shape increased the height of the cell and thus exposed it to greater shear forces. If the cells had a similar density of adhesion area to S-phase cells, on the smaller contact area, this could significantly decrease the adhesion strength potential of the cell. The increase in adhesion density facilitated more of an adhesive area within the smaller membrane contact area needed to accommodate the probable increase in shear forces experienced by the raised cells. In this in vitro culture system, there was no imposed shear force and the viscosity of the medium was increased by the presence of serum [46], thus decreasing the influence of shear further. Therefore, this could not be compared to a

dynamic system. Adhesion in this case was only a requirement of cell cycle progression and morphological modification (anchorage dependency) [44]. The development of the mature focal adhesions, seen in the SEM images, suggested that these were required for the observed morphological changes to occur [42,47]. That there was no significant difference in total adhesion area between S- and non-S-phase cells (Fig. 7c) also suggested that there were probably a finite number of focal adhesions a cell could produce, regardless of its size and contact area with the substrate and that only reorganisation was occurring. This agreed with the work of Davies et al. [23]. Some evidence to support this can be found indirectly from theories concerning the finite amounts of cellular materials available. Numerous authors maintain that the blebbing and ruffling of rounded up cells are expressions of storage areas for excess membrane [48,49]. In terms of the cytoskeleton, cell morphology can be changed in seconds (e.g. when trypsinised) without a change in the cellular levels of F-actin [50]. Linking the concept of finite amounts of vinculin and finite strength for individual focal adhesions, Yamamoto et al. [51] suggests that a cell has a mechanical limit to its adhesion strength, in terms of cell– substrate adhesion. Theorising about cell adhesion properties becomes difficult when the material properties attributed to a cell, such as cytoskeletal attachment to focal adhesions, are considered. Lotz et al. [18] produced an elegant analogy to explain this relationship, comparing the basal membrane of the cell to a piece of ‘sticky tape’ with the focal adhesions providing

D.O. Meredith et al. / Experimental Cell Research 293 (2004) 58–67

65

Fig. 7. Histograms of the means for each of the three outputs including the standard error. (a) Contact area: S-phase cells were found to have significantly greater contact area than non-S-phase cells (14.7 F 1.6 vs. 9.5 F 0.9, mean F SE, P = 0.001). (b) Adhesion density: The adhesion density was significantly less for S-phase than non-S-phase cells (1.3 F 0.1 vs. 2.3 F 0.2, mean F SE, P = 0.002). (c) Total adhesion area: There was no difference in adhesion area between S- and non-S-phase cells (0.2 F 0.04 vs. 0.3 F 0.04, mean F SE, P = 0.8) (**P < 0.01).

the adhesive sections of the tape. When peeling the sticky tape away, the focal adhesions are removed, one at a time, and the force required for removal is quite low. By placing a stiff backing on the sticky tape, thus modifying its mechanical properties, each adhesive area would now be responsible for adhering all of the backed tape and removal of the adhered tape would require a significant increase in force. The ‘sticky tape’ here is analogous to a cell whose focal adhesions are connected to the cytoskeleton. It could be termed a two-way mechanism, the cell’s rigidity is aided by adhesion and, reciprocally, the maintenance of adhesion is assisted by the rigidity of the cell [7]. This reinforced material property would, most logically, occur during S-phase, because the required cytoskeletal changes are present for its progression [52]. This brings into question the amount of adhesion that can be attributed to a single focal adhesion. Large variations in focal adhesion size and shape still produces constant stress in the cell [53]. Total cell contact area varied more for S-phase cells than non-S-phase cells while focal adhesion density was less

varied for S-phase cells than non-S-phase cells (Fig. 8). This indicated that the focal adhesion area and cell contact area were not closely interlinked. As with most biological systems, there was also a natural variation to be expected in the cell population.

Conclusion The SEM imaging suggested that there was a cell cyclerelated change in focal adhesion morphology, coinciding with a modification in cellular morphology. Focal adhesion density varied significantly suggesting that the cell cycle had an influence on cell adhesion. The focal adhesion density of S-phase cells was shown to be less than that for non-S-phase cells, contrary to our original expectations. In the context of biomaterial cytocompatibility testing by focal adhesion quantification, defining the cell cycle phase appeared to be beneficial. The method could be improved by providing sufficient sample replicates to normalise the data. However, if sample and cell number are limited, then

Fig. 8. (a) Box plot showing the variation in cell contact area (Am2) between S-phase and non-S-phase cells. The cell contact area varied more for S-phase cells than non-S-phase cells. (b) Box plot showing the variation in adhesion density (%) between S-phase and non-S-phase cells. The cell adhesion density was less varied for S-phase cells than non-S-phase cells.

66

D.O. Meredith et al. / Experimental Cell Research 293 (2004) 58–67

the exclusive quantification of S-phase cells still provide useful results. It has a further advantage, in that there is very low sample variation with respect to adhesion density measurements. This method of combining labelling procedures provided an easily applied high-resolution co-identification of two distinct criteria.

Acknowledgment The authors would like to thank Dr. Keita Ito for his assistance with the statistical analysis of the data.

References [1] J. Folkman, A. Moscona, Role of cell shape in growth control, Nature 273 (1978) 345 – 349. [2] R. Cornell, Cell – substrate adhesion during cell culture. An ultrastructural study, Exp. Cell Res. 58 (1969) 289 – 295. [3] A.S. Curtis, The mechanism of adhesion of cells to glass, J. Cell Biol. 20 (1964) 199 – 215. [4] C.S. Izzard, L.R. Lochner, Cell-to-substrate contacts in living fibroblasts: an interference reflexion study with an evaluation of the technique, J. Cell Sci. 21 (1976) 129 – 159. [5] W.T. Chen, S.J. Singer, Immunoelectron microscopic studies of the sites of cell – substratum and cell – cell contacts in cultured fibroblasts, J. Cell Biol. 95 (1982) 205 – 222. [6] R.O. Hynes, B.L. Bader, Target mutations in integrin and their ligands: their implications for vascular biology, Thromb. Haemostasis 78 (1997) 83 – 87. [7] S.M. Schoenwaelder, K. Burridge, Bidirectional signaling between the cytoskeleton and integrins, Curr. Opin. Cell Biol. 11 (1999) 274 – 286. [8] R. Ohnishi, Dynamics of cultured L cells as studied by cinemicroscopy and scanning electron microscopy, Biomed. Res. Suppl. 2 (1981) 1 – 12. [9] W.J. Pledger, C.D. Stiles, H.N. Antoniades, C.D. Scher, An ordered sequence of events is required before BALB/c-3T3 cells become committed to DNA synthesis, Proc. Natl. Acad. Sci. U. S. A. 75 (1978) 2839 – 2843. [10] M. Iwig, D. Ngoli, D. Glaesser, Autoradiographic investigations on cell shape-mediated growth regulation of lens epithelial cells in culture, Biomed. Biochim. Acta 48 (1989) 121 – 127. [11] M. Iwig, E. Czeslick, A. Muller, M. Gruner, M. Spindler, D. Glaesser, Growth regulation by cell shape alteration and organization of the cytoskeleton, Eur. J. Cell Biol. 67 (1995) 145 – 157. [12] B. Geiger, A. Bershadsky, Exploring the neighborhood: adhesioncoupled cell mechanosensors, Cell 110 (2002) 139 – 142. [13] C.S. Chen, M. Mrksich, S. Huang, G.M. Whitesides, D.E. Ingber, Geometric control of cell life and death, Science 276 (1997) 1425 – 1428. [14] K. Burridge, M. Chrzanowska-Wodnicka, Focal adhesions, contractility, and signaling, Annu. Rev. Cell Dev. Biol. (1996) 463 – 519. [15] S.K. Sastry, K. Burridge, Focal adhesions: a nexus for intracellular signaling and cytoskeletal dynamics, Exp. Cell Res. (2000) 25 – 36. [16] N.J. Boudreau, P.L. Jones, Extracellular matrix and integrin signalling: the shape of things to come, Biochem. J. 339 (Pt. 3) (1999) 481 – 488. [17] C.E. Turner, K. Burridge, Transmembrane molecular assemblies in cell – extracellular matrix interactions, Curr. Opin. Cell Biol. 3 (1991) 849 – 853. [18] M.M. Lotz, C.A. Burdsal, H.P. Erickson, D.R. McClay, Cell adhesion to fibronectin and tenascin: quantitative measurements of initial binding and subsequent strengthening response, J. Cell Biol. 109 (1989) 1795 – 1805.

[19] A. Hunter, C.W. Archer, P.S. Walker, G.W. Blunn, Attachment and proliferation of osteoblasts and fibroblasts on biomaterials for orthopaedic use, Biomaterials 16 (1995) 287 – 295. [20] R.G. Richards, G.R. Owen, B.A. Rahn, I. ap Gwynn, A quantitative method of measuring cell adhesion areas. Review, Cell and Materials 7 (1997) 15 – 30. [21] Y. Yamakita, G. Totsukawa, S. Yamashiro, D. Fry, X. Zhang, S.K. Hanks, F. Matsumara, Dissociation of FAK/p130(CAS)/c-Src complex during mitosis: role of mitosis-specific serine phosphorylation of FAK, J. Cell Biol. 144 (1999) 315 – 324. [22] E. Zamir, B.Z. Katz, S. Aota, K.M. Yamada, B. Geiger, Z. Kam, Molecular diversity of cell – matrix adhesions, J. Cell Sci. 112 (Pt. 11) (1999) 1655 – 1669. [23] P.F. Davies, A. Robotewskyj, M.L. Griem, Endothelial cell adhesion in real time. Measurements in vitro by tandem scanning confocal image analysis, J. Clin. Invest. 91 (1993) 2640 – 2652. [24] N. Wang, D.E. Ingber, Control of cytoskeleton mechanics by extracellular matrix, cell shape, and mechanical tension, Biophys. J. 66 (1994) 2181 – 2189. [25] A.S. Maddox, K. Burridge, RhoA is required for cortical retraction and rigidity during mitotic cell rounding, J. Cell Biol. 160 (2003) 255 – 265. [26] A. Howard, S.R. Pelc, Nuclear incorporation of P32 as a demonstrated by autoradiographs, Exp. Cell Res. (1951) 178 – 187. [27] L.P. Everhart Jr., R.W. Rubin, Cyclic changes in the cell surface: I. Change in thymidine transport and its inhibition by cytochalasin B in Chinese hamster ovary cells, J. Cell Biol. 60 (1974) 434 – 441. [28] A.B. Pardee, G1 events and regulation of cell proliferation, Science 246 (1989) 603 – 608. [29] A. Murray, T. Hunt, The Cell Cycle an Introduction, Oxford Univ. Press, USA, 1993. [30] K. Porter, D. Prescott, J. Frye, Changes in surface morphology of Chinese hamster ovary cells during the cell cycle, J. Cell Biol. 57 (1973) 815 – 836. [31] M.S. Zana, G. Albrecht-Buehler, What structures, beside adhesions, prevent spread cells from rounding up? Cell Motil. Cytoskeleton 13 (1989) 195 – 211. [32] D.E. Ingber, D. Prusty, Z. Sun, H. Betensky, N. Wang, Cell shape, cytoskeletal mechanics, and cell cycle control in angiogenesis, J. Biomech. 28 (1995) 1471 – 1484. [33] B. Geiger, A 130 K protein from chicken gizzard: its localization at the termini of microfilament bundles in cultured chicken cells, Cell 110 (1979) 139 – 142. [34] J.L. Rodriguez, R. Fernandez, B. Geiger, D. Salamon, A. Ben-Ze’ve, Overexpression of vinculin suppresses cell motility in BALB/c 3t3 cells, Cell Motil. Cytoskeleton 22 (1992) 127 – 134. [35] J.L. Rodriguez, R. Fernandez, B. Geiger, D. Salamon, A. Ben-Ze’ve, Suppression of vinculin expression by antisense transfection confers changes in cell morphology, motility and anchorage-dependent growth of 3t3 cells, J. Cell Biol. 122 (1993) 1285 – 1294. [36] R.G. Richards, M. Stiffanic, G.R. Owen, M. Riehle, I. ap Gwynn, A.S. Curtis, Immunogold labelling of fibroblast focal adhesion sites visualised in fixed material using scanning electron microscopy, and living, using internal reflection microscopy, Cell Biol. Int. 25 (2001) 1237 – 1249. [37] G.R. Owen, D.O. Meredith, I. ap Gwynn, R.G. Richards, Simultaneously identifying S-phase labelled cells and immunogold-labelling of vinculin in focal adhesions, J. Microsc. 207 (2002) 27 – 36. [38] P. Elvin, C.W. Evans, The adhesiveness of normal and SV40-transformed BALB/c 3T3 cells: effects of culture density and shear rate, Eur. J. Cancer Clin. Oncol. 18 (1982) 669 – 675. [39] G.R. Owen, D.O. Meredith, I. ap Gwynn, R.G. Richards, Enhancement of immunogold-labelled focal adhesion sites in fibroblasts cultured on metal substrates: problems and solutions, Cell Biol. Int. 25 (2001) 1251 – 1259. [40] A.W. Rogers, Techniques of Autoradiography, second ed., Elsevier Scientific Publishing Company, Amsterdam, 1973.

D.O. Meredith et al. / Experimental Cell Research 293 (2004) 58–67 [41] R.G. Richards, B.A. Rahn, I. ap Gwynn, Scanning electron microscopy of the undersurface of cell monolayers grown on metallic implants, J. Mar. Sci.: Mater. Med. 6 (1995) 120 – 124. [42] A.D. Bershadsky, I.S. Tint, A.A. Neyfakh Jr., J.M. Vasiliev, Focal contacts of normal and RSV-transformed quail cells. Hypothesis of the transformation-induced deficient maturation of focal contacts, Exp. Cell Res. 158 (1985) 433 – 444. [43] K. Burton, D.L. Taylor, Traction forces of cytokinesis measured with optically modified elastic substrata, Nature 385 (1997) 450 – 454. [44] S. Huang, D.E. Ingber, The structural and mechanical complexity of cell-growth control, Nat. Cell Biol. 1 (1999) E131 – E138. [45] E.K. Han, T.M. Guadagno, S.L. Dalton, R.K. Assoian, A cell cycle and mutational analysis of anchorage-independent growth: cell adhesion and TGF-beta 1 control G1/S transit specifically, J. Cell Biol. 122 (1993) 461 – 471. [46] R.I. Freshney, Culture of Animal Cells: A Manual of Basic Technique, fourth ed., Wiley Inc., New York, 2000. [47] E. Zamir, B. Geiger, Molecular complexity and dynamics of cell – matrix adhesions, J. Cell Sci. 114 (2001) 3583 – 3590. [48] R.W. Rubin, L.P. Everhart, The effect of cell-to-cell contact on the

[49]

[50]

[51]

[52]

[53]

67

surface morphology of Chinese hamster ovary cells, J. Cell Biol. 57 (1973) 837 – 844. C. O’Neill, P. Jordan, G. Ireland, Evidence for two distinct mechanisms of anchorage stimulation in freshly explanted and 3T3 Swiss mouse fibroblasts, Cell 44 (1986) 489 – 496. D.J. Mooney, R. Langer, D.E. Ingber, Cytoskeletal filament assembly and the control of cell spreading and function by extracellular matrix, J. Cell Sci. 108 (Pt. 6) (1995) 2311 – 2320. A. Yamamoto, S. Mishima, N. Maruyama, M. Sumita, Quantitative evaluation of cell attachment to glass, polystyrene, and fibronectinor collagen-coated polystyrene by measurement of cell adhesive shear force and cell detachment energy, J. Biomed. Mater. Res. 50 (2000) 114 – 124. M. Iwig, D. Glaesser, Cell – substratum interactions and the cytoskeleton in cell shape-mediated growth regulation of lens epithelial cells, Lens Eye Toxic. Res. 8 (1991) 281 – 309. N.Q. Balaban, U.S. Schwarz, D. Riveline, P. Goichberg, G. Tzur, I. Sabanay, D. Mahalu, S. Safran, A. Bershadsky, L. Addadi, B. Geiger, Force and focal adhesion assembly: a close relationship studied using elastic micropatterned substrates, Nat. Cell Biol. 3 (2001) 466 – 472.