Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effects of surface topography on organization of cells and extracellular matrix Glass-Brudzinski, Jeanette 2000

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2000-0407.pdf [ 6.4MB ]
Metadata
JSON: 831-1.0089536.json
JSON-LD: 831-1.0089536-ld.json
RDF/XML (Pretty): 831-1.0089536-rdf.xml
RDF/JSON: 831-1.0089536-rdf.json
Turtle: 831-1.0089536-turtle.txt
N-Triples: 831-1.0089536-rdf-ntriples.txt
Original Record: 831-1.0089536-source.json
Full Text
831-1.0089536-fulltext.txt
Citation
831-1.0089536.ris

Full Text

E F F E C T S O F S U R F A C E T O P O G R A P H Y O N O R G A N I Z A T I O N O F C E L L S A N D E X T R A C E L L U L A R M A T R I X by J E A N N E T T E G L A S S - B R U D Z I N S K I B . S c , The University of British Columbia, 1997 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L M E N T O F R E Q U I R E M E N T S F O R T H E D E G R E E OF M A S T E R OF S C I E N C E in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Department of Oral Biological and Medical Sciences, Faculty of Dentistry) We accept this thesis as conforming to the registered standard T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A M A Y 2000 © Jeannette Glass-Brudzinski, 2000 In p resent ing this thesis in partial fu l f i lment of the requ i rements for an advanced degree at the Univers i ty of Brit ish C o l u m b i a , I agree that the Library shall make it freely available for re ference and study. I further agree that pe rm i s s i on for extens ive c o p y i n g of this thes is f o r scholar ly p u r p o s e s may b e granted by the h e a d of my depa r tmen t or by his o r her representat ives . It is u n d e r s t o o d that c o p y i n g or pub l i c a t i on o f this thesis fo r f inancia l ga in shal l no t be a l l o w e d w i t h o u t my wr i t t en pe rm iss ion . The Univers i ty of Brit ish C o l u m b i a Vancouve r , C a n a d a Depa r tmen t DE-6 (2/88) A B S T R A C T The long-term success of an implant is governed to a large extent by its surface properties that w i l l influence cell behaviour reflected in their migration, attachment, morphology, and proliferation. Also , the integration of the surrounding tissue around the implant is dependent on the reorganization of the surrounding E C M in a morphogenetic manner. It is of great importance to understand the interactions of cells with the implant surface, with the E C M , and with other cells so that failure of implanted devices can be avoided. This thesis examined the in vitro process of connective tissue reorganization and how it was affected by the topography of substratum using the following techniques: substrata micromachining, collagen gels as substrata, Light Microscopy, Polarized Light Microscopy (PSR stained collagen), Confocal Laser Scanning Microscopy (PI stained nuclei), and Time Lapse Video Microscopy. This thesis measured the cell orientation on 6 types of surfaces: T C dish, smooth titanium, 30IP175S, 3G30P, 30G40P, and 30G175P grooved titanium. The effect of collagen matrix on cell orientation under three conditions (Control, Cel lGel , and Gel) was determined. Also , the effects of cell distribution within a gel on the gel contraction was examined. The distribution of cells between collagen and titanium surface under the Ce l lGel and Gel conditions was compared. Gel penetration into grooves was determined. The following observations and conclusions were made. It appeared that collagen gel did not penetrate into the grooves of a grooved surface. In the absence of collagen gel, cells formed patchworks of parallel arrays on T C dish, smooth, and pitted surfaces. Cells aligned with the grooves on grooved surfaces. The presence of collagen matrix atop a stable culture (CellGel) disrupted the orientation of cells leading to a lesser alignment. The presence of collagen matrix around the cells (Gel), resulted in lowest level of alignment. The collagen fibers seemed to ii conform with the orientation of cells and also become aligned with the grooves. Orientation of cells within the three dimensional matrix was not affected by either the surface topography or the collagen fibers. Cells were found to be essential for the gel contraction. When most cells were found at the bottom of the gel throughout the experiments (CellGel) a thin sheet of "orthogonally" arranged fibers was formed after a period of 3-4 weeks. When cells were distributed throughout a matrix (Gel) a "R ing" of fibers was created. Cells were not only able to leave titanium surface and make new attachments with the overlaying matrix (CellGel) but also to leave a collagenous matrix and settle on titanium surface (Gel). The Cel lGel condition that leads to a production of a continuous sheet of "tissue" can be used to coat a device prior to implantation to improve a long term success of such a device. iii T A B L E O F C O N T E N T S Abstract i i Table of Contents iv List of Figures vi i i Acknowledgments x C H A P T E R I: INTRODUCTION 1. Overview 1 2. Cell-Titanium Interactions 2 a) Titanium as Substratum 2 b) Substratum Topography and Contact Guidance 4 c) Ce l l Attachment to Substratum (Serum Proteins) 6 d) Titanium Versus Collagen as Substratum 8 3. Cell-Collagen Interactions 10 a) Collagen as Substratum 10 b) Collagen and Cel l Behaviour 12 c) Collagen and Contact Guidance 14 d) Ce l l Locomotion, Moti l i ty, and Traction Force 15 e) Collagen Gel Contraction 18 4. Cel l -Cel l Interactions 23 a) Monolayering 23 b) Orthogonality.... 24 5. Objective of the Thesis 25 C H A P T E R II: M A T E R I A L S AND M E T H O D S 1. Cel l Culture 28 2. Culture Conditions 28 a) Control: Cells + Surface 28 b) Control: Surface + Collagen Gel 29 iv c) "Ce l lGe l " Condition: (Surface + Cells) + Collagen Gel 29 d) " G e l " Condition: Surface + (Cells + Collagen Gel) 29 e) "Boxes" with opposing microfabricated surfaces + (Cells + Collagen Gel) 29 f) Distribution of Cells Between Collagen Gel and Titanium 30 g) Collagen Gel Penetration into Grooved Surface 30 3. Collagen Gels 30 4. Micromachined Substrata and "Boxes" with opposed microfabricated surfaces 31 a) Micromachining 31 b) "Boxes" with Opposed Microfabricated Surfaces 34 c) Preparation of Substrata for Use 34 5. Light Microscopy 35 6. P S R and Polarized Light Microscopy 36 7. PI and Eosin: Epifluorescence Microscopy 37 8. Time-Lapse Studies 39 9. Data Collection 39 10. Statistics 40 C H A P T E R I H : R E S U L T S I. Quantitative Data 45 1. Cel l Orientation Within the Gels 45 a) T C dish 45 i) Control Condition 46 ii) Cel lGel Condition 47 ii i) Ge l Condition 47 b) Smooth Titanium Surface 47 i) Control Condition : 48 ii) Cel lGel Condition 48 iii) Gel Condition 49 v c) 3G30P:Titanium with Shallow (3um) Grooves 49 i) Control Condition 50 ii) Cel lGel Condition 50 iii) Gel Condition 51 d) 30G40P:Titanium with Deep (30um) Grooves and a Narrow Ridge .. 51 i) Control Condition 52 ii) Cel lGel Condition 52 ii i) Ge l Condition 53 e) 30G175P: Titanium with Deep (30pm) Grooves and a Wide Ridge.... 54 i) Control Condition 54 ii) Cel lGel Condition 55 iii) Gel Condition 55 f) Summary of the Grooved Surfaces 56 g) 30IP175S: Titanium Surface with Inverted Pyramids 57 i) Control Condition 58 ii) Cel lGel Condition 58 iii) Gel Condition 58 2. Ce l l Distribution by Zones 59 II. Qualitative Data • 59 1. Collagen Gel Contraction 59 a) T C Dish and Smooth Titanium Surface 60 i) Control Condition 60 ii) Cel lGel Condition 60 iii) Gel Condition 61 b) 3G30P: Titanium with Shallow (3um) Grooves 61 i) Control Condition 61 ii) Cel lGel Condition 61 iii) Gel Condition 62 c) 30G175P:Titanium with Deep (30um) Grooves and a Wide Ridge 62 i) Control Condition 62 vi ii) Cel lGel Condition 62 ii i) Gel Condition 63 d) 30IP175S:Titanium Surface with Inverted Pyramids 63 i) Control Condition 63 ii) Cel lGel Condition 63 iii) Gel Condition 63 2. Observations on the Distribution of Cells Between Collagen and Titanium 64 3. Collagen Gel Penetration on Grooved Surfaces 64 4. Fiber Alignment in Boxes with Opposed Microfabricated Surfaces 65 Summary of Results 106 C H A P T E R I V : D I S C U S S I O N 1. Discussion 108 2. Future Work 119 Bibliography 122 vu L I S T O F F I G U R E S 1. Cross sections of surface topographies. 2. Methods of measuring cell orientation. 3. Cel l orientation on T C dish (1 week). 4. Cel l orientation on smooth titanium (1 week). 5. Cel l distribution and orientation within a three dimensional lattice on T C , smooth, and pitted surfaces. 6. Ce l l orientation on 3G30P grooved surface (1 week). 7. Ce l l orientation on 3G30P grooved surface (2 weeks). 8. Ce l l orientation on 30G40P grooved surface (1 week). 9. Cel l orientation on 30G40P grooved surface (2 weeks). 10. Cel l orientation on 30G175P grooved surface (1 week). 11. Cel l orientation on 30G175P grooved surface (2 weeks). 12. Cel l distribution and orientation within a three dimensional lattice on grooved surfaces. 13. Cel l orientation and cell density distribution throughout a gel (grooved surface). 14. Proportion of aligned cells in Zone 1 on grooved surfaces under different conditions. 15. Cel l orientation on 30IP175S pitted surface (1 week). 16. Cel l distribution by Zones on six surfaces under the Cel lGel and the Gel conditions. 17. Gel contraction on smooth surface under the Cel lGel condition. 18. Gel contraction on smooth surface under the Gel condition. 19. Gel contraction on T C dish under the Cel lGel condition. 20. Gel contraction on T C dish under the Gel condition. 21. Gel contraction on 3G30P grooved surface under the Cel lGel condition. 22. Gel contraction on 3G30P grooved surface under the Gel condition. V l l l 2 3 . Gel contraction on 30G175P grooved surface under the Ce l lGel condition. 2 4 . Gel contraction on 30G175P grooved surface under the Gel condition. 2 5 . Gel contraction on 30IP175S pitted surface under the Cel lGel condition. 2 6 . Gel contraction on 30IP175S pitted surface under the Gel condition. 2 7 . "R ing" structure. 28. "R ing" formation: time lapse movie frames. 2 9 . Orthogonal multilayers. 3 0 . Ce l l distribution between titanium and collagen gel under the Ce l lGe l and Gel conditions (model for gel contraction). ix A C K N O W L E D G M E N T S This is to acknowledge a number of people Who helped me a lot more than just a little First, and foremost, my supervisor, Dr. Brunette Whose guidance and advice many times made me sweat A n d even though, I may have overused my passport Throughout my studies, He maintained His support I am also grateful to my committee members Dr. Waterfield, Putnins, Uitto, and Dr. Clive Roberts Next, F d like to thank Mrs. Weston Lesley Whom I owe a lot more than a 5 dollar fee She taught me everything, from simple microscopy To tissue culture, staining, and saving data on a floppy M r . Adnre Wong helped me with the Confocal C L S M was Greek to me, but now it seems banal Dr. N ick Jeager from U B C Electrical Engineering Provided his expertise in substratum micromachining Medical Research of Canada supported this thesis So, thank you all for the help in my thesis synthesis C H A P T E R I INTRODUCTION 1. O V E R V I E W Insertion of any implant, whether it is partial hip, other joint implant or prostheses, cardiovascular shunts and catheters or simply dental implants, disrupts the organization of the surrounding tissue. Healing of a wound is related to an intricate cascade of biological events and to a number of interactions that are initiated upon placement of the artificial device replacing the missing tissue. Among those interactions are: cell-material surface interactions, cell-ExtraCellular Matrix ( E C M ) interactions, and cell-cell interactions. The long-term success of an implant is governed to large extent by the surface properties (chemistry and topography) of such a device that w i l l promote appropriate cell behavior reflected in their migration, attachment, morphology, and proliferation and the induction of the reorganization of surrounding E C M in a morphogenetic manner creating integration of the surrounding tissue around the implant. It is of great importance to understand the interactions of cells with E C M , with the implant surface, and with other cells so that failure of implanted devices can be avoided thereby decreasing risks to patient's health. This thesis examined how cells react to micromachined titanium surfaces in the presence of three dimensional collagen lattices. Cel l orientation (with respect to topographical cues) on the surface as well as within the three dimensional lattice was determined. The thesis examined the effects of collagen matrix on cell orientation when the cells are embedded within the matrix or just overlayed with it. It also determined the distribution of cells between the surface and the collagen matrix under these two conditions and used that information to explain the result of collagen contraction under the different conditions. The ability of cells to alter the orientation of l collagen and to create new sheets of tissues or new strands of tissues (depending on the collagen environment conditions: whether cells are suspended within the matrix, overlayed with the matrix, cultured on top of the matrix, or sandwiched in between) was determined. 2. C E L L - T I T A N I U M INTERACTIONS a) T I T A N I U M AS S U B S T R A T U M Biomaterials used for direct contact with hard and soft tissue have to fulfil some critical requirements such as: 1) biocompatibility (i.e. the material can induce an appropriate host response in a specific application (Williams, 1989) or as stated by Ratner (1993) "the material can exploit the proteins and cells of the body to meet a specific performance goal" with low intrinsic toxicity and low inflammatory activation), 2) surface texture promoting cellular adhesion and limiting relative interfacial motion (Brunette, 1988), 3) surface chemistry that could elevate extracellular matrix protein adhesion thereby increasing cellular adhesion, 4) biofunctionality (mechanical requirements which may differ depending on a particular application), 5) corrosion resistance (withstanding being oxidized by components of the surrounding environment) and associated with it low material dissipation out into tissue, 6) prevention of bacterial adhesion (Tengvall and Lundstrom, 1992). It has been found that titanium and its alloys fulfil these requirements to a high extent, especially in comparison to other metallic biomaterials. Titanium and titanium alloy biomaterials are widely used in a variety of dental and orthopedic applications ( Merritt, 1984). There seems to be no biologically significant differences in terms of initial cell attachment and morphology between a commercially pure titanium and titanium alloy surfaces (Keller et al, 1994). Titanium is not only an excellent bone-2 anchoring material (Bothe et al, 1940) but it is also well accepted by soft tissue (Branemark et. al, 1969 and Meachim and Williams, 1973), without adverse allergic or immune reactions (Rae, 1981). The corrosion, wear problems, bone resorption and inflammation in other metal systems, such as the stainless steel, hastened the progress to develop T i prostheses (Williams and Meachim , 1974). However, even small amounts of ion released from a metal surface can interfere with cell differentiation (Thompson and Puleo, 1995). Therefore, to avoid such effects, titanium metals used in implant therapy are usually passivated to induce a formation of a more substantial oxide layer that reduces ion release from the surface and therefore improves the corrosion resistance of the material (Raikar et al, 1995). This dense passivating oxide,by its low solubility, protects titanium surfaces (Baes and Mesmer, 1976), and high breakdown potentials (Zitter and Plenk, 1987). It can also increase the E C M protein adsorption onto the surface (El-Ghannam et al, 1998). These proteins are required for cell attachment under in vivo conditions since cells do not normally bind directly to the implant (Kothari et al, 1995). The passivated oxide layer on a titanium surface can be altered by a cleaning technique known as the Radio Frequency Glow Discharge (RFGD) which ashes away organic contaminants and renders the surface sterile (Doundoulakis, 1987). The oxide coating has a higher surface energy, as determined by contact angle measurements (Baier and Meyer, 1988). The contact angle created by a liquid droplet spreading on a surface indicates the wettability of a surface and allows classification of that surface as either hydrophilic or hydrophobic. The higher the surface energy (ranging between y s of 18-116 erg/cm 2), the smaller the contact angle (ranging between OPBS of 26-105 degrees) and therefore the more hydrophilic is the surface. Consequently, higher surface energy leads to higher wettability of a surface which in turn increases adhesion of matrix 3 and serum derived proteins, thereby enhancing cell attachment and spreading on a surface (Stanford etal, 1994). It can be concluded that titanium is a very useful biomaterial because of its biocompatibility, corrosion resistance, ease and effectiveness of cleaning, and high cell-binding affinity, as well as its mechanical properties. b) S U B S T R A T U M T O P O G R A P H Y A N D C O N T A C T G U I D A N C E A l l cell types live in an environment with some type of topography which can influence them in a variety of ways. Often the features of a surface can restrict the flow of molecules through diffusion channels in and out of cells, limit cells' access to nutrients and can also entrap them such that the escape of waste products is difficult (Curtis and Clark, 1990). Most of the time, though, the topographical reactions of cells are considered with respect to their altered movements, polarity and their overall shape (e.g. Oakley and Brunette, 1993). The change in cell shape can lead to a change in other vital processes such as cell proliferation (Folkman and Moscona, 1978), cytoskeletal gene expression (Ben Ze'ev, 1984), collagenase gene expression (Werb et al, 1989), extracellular matrix metabolism and cell differentiation (Watt, 1986). The particular topography experienced by a cell can be created by other cells, extracellular matrix components, or artificial materials. Interactions between cells and surrounding topography is thought to be involved in many natural processes such as embryogenesis, tissue morphogenesis, regeneration, and wound healing (Dunn 1982). The responses in which the direction of orientation or locomotion of cells is determined by the shape, alignment or curvature of the substratum have been assigned the name of "Contact Guidance" (Weiss, 1934), more recently referred to as "Topographic Guidance" 4 (Wilkinson et al, 1982). This phenomenon was first observed by Harrison in one of the earliest experiments on cell culture using cells cultured in plasma clots and on spider-web fibers (Harrison, 1914). Similar observations were made by Weiss who employed substrata as diverse as plasma clots, fish scales, aligned collagen gels, fibrin strands, and engraved glass (Weiss and Garber, 1952; Weiss, 1959; Weiss and Taylor, 1956; Weiss, 1945). It is generally agreed that cells align with the topography of their substratum and rearrange their cytoskeleton (Dunn and Heath, 1976); this process being one of the theories suggested to explain Contact Guidance. The topography of a surface can be responsible for inducing changes in a cell 's migration patterns and a cell 's shape leading to further alterations in the cell 's behavior. Therefore, it is not only the chemical composition of an implant material that w i l l determine an implant's long term success but also its micro topography that w i l l eventually enhance or lessen cellular interactions and attachment to the biomaterial. Highly ordered and precise topographies can be fabricated by micromachiriing, a technique which involves etching and photolithography. These techniques allow excellent control over production of surface topography because different aspects of surface such as depth, repeat spacing, and shape can be precisely controlled by different steps in the process (Brunette et al. 1983). Many different types of topographies and their effects on cells have been previously studied. Some of these include: single cliffs, grooves/ridges, dots, spikes, pits, tunnels and tubes, cylinders, and random roughness (Curtis and Wilkinson, 1997), with multiple cues eliciting a much stronger effect. Many studies have found that cells w i l l align on a micromachined grooved substrata (Brunette et al, 1983, Brunette 1986a, Oakley and Brunette, 1995; Clark et al, 1987). The grooves seem to guide the cells in their orientated migration resulting in cell polarization with the direction of the grooves (Oakley et al, 1997). Studies have shown that fibroblasts respond hierarchically to grooves of different dimensions, with large 5 grooves dominating the effects of smaller grooves (Brunette 1986b). The degree to which cells w i l l align with topographic cues (such as grooves) depends on several factors: groove depth, pitch size, and ridge width. It has been shown that shallow grooves were less effective in orienting cultured cells then deeper grooves; 0.3\xm deep grooves eliciting less alignment than a 1.9um deep grooves (Clark etal, 1990). Larger pitched (24um spacing) grooved substrata were also less effective in aligning cells than the smaller pitched (4um spacing) grooves (Ohara and Buck, 1979). However, alignment of fibroblasts on a grooved substrata was most dependent on ridge width, alignment being inversely proportional to ridge width (Dunn and Brown, 1986). One theory explaining contact guidance suggested that a response of a cell to a topographical feature may be a probabilistic process (Brunette, 1986), meaning that the topographical feature alters the probability of a cell making a successful protrusion and contact in a given direction and implies that topography does not necessarily induce an all-or-none response. c) C E L L A T T A C H M E N T T O S U B S T R A T U M ( S E R U M P R O T E I N S ) A cell does not attach to a metallic surface directly through its membrane. Cel l attachment and spreading on a non-biological surface is dependent not just on the nature of the material to which cells adhere but rather on the presence of other adhesive molecules that are vital to this process (Baier, 1986). Cel l adhesion differs in vitro and in vivo, but the process can resemble cell behavior in vivo i f serum is added to medium. This is because the serum contains a number of extracellular adhesion molecules, such as fibronectin and vitronectin (Hayman et al, 1985). I f such molecules are available, the adhesion of cells to an artificial substratum w i l l be a multi-step process involving: 1) adsorption of medium and adhesion molecules to the foreign substratum, 2) actual contact between the cell and surface adsorbed molecules (through 6 membrane receptors like integrins), 3) reorganization of cytoskeleton leading to cell spreading (Revel and Wolken, 1973; Weiss, 1975). Other studies have also shown that in vitro cell cultures actually require cell-surface adhesion which is mediated by that sub-layer of adsorbed proteins (Bennett and Gill igan, 1993). Cel l adhesion and spreading through a protein layer (fetal calf serum proteins) has been found to be related to the surface free energy of a solid substratum (Schakenraad et al, 1986). In general, high surface energy substrata promote cellular adhesion and spreading, whereas low surface free energy substrata do not (Baier, 1984) (note: the R F G D cleaning technique discussed earlier increases surface free energy and increases cell attachment and spreading). It has also been found that the maximal spreading area of human skin fibroblasts on protein pre-coated substrata (fetal calf serum, fibronectin, or bovine serum albumin) is significantly higher than on bare substrata. The increase in spreading is related to substratum surface free energy; with the lowest spreading being found on a surface with surface 2 2 energy Ys =12erg/cm and the highest spreading found on a surface with Ys= :70erg/cm (Schakenraad et al, 1989). The adsorbed proteins, with their conformation often being altered after binding due to structural rearrangements (Eckert et al, 1997), are capable of modifying the substratum surface characteristics and rendering it wettable. It has been suggested that the adhesive molecules bend out of shape exposing new attachment sites for cells (Baier, 1970) thereby also increasing the cell binding to a substratum through the ligand-receptor interactions. Glycoproteins are probably the most important extracellular adhesion molecules. Two proteins which are of particular relevance to adhesion of cells to non-living surfaces are fibronectin and laminin (Yamada and Olden, 1978). It has been argued that fibronectin is an essential requirement to achieve focal contacts between a cell and substratum (Grinell, 1986). However, others have found that attachment can take place in the absence of fibronectin (Curtis and Forrester, 1984). Nevertheless, fibronectin is the major adhesive protein in vivo as it has the 7 ability to bind to both the cell membrane (through membrane receptors: integrins) and to substratum (whether it is artificial or natural extracellular components like collagen and proteoglycans). Proteoglycans are also involved in cell membrane binding to other extracellular components. Structural proteins such as collagen and elastin can also bind to cell membranes through either a specific amino acid sequence or through the proteoglycans and glycoproteins. In addition, many other proteins have been demonstrated to be involved in cell-substratum interactions, including gamma-globulin, fibrinogen, prothrombin, vinculin, and albumin (Schakenraad et al, 1986). It is evident that in vivo cells are capable of attachment, spreading and migration on/in their substratum due to the presence of binding molecules. In vitro, however, the cells' environment has to be augmented with such proteins for the cell-substratum binding to occur. d) T I T A N I U M V E R S U S C O L L A G E N A S S U B S T R A T U M In vivo, fibroblasts are embedded in an extracellular matrix containing mainly strength-providing collagen, elasticity-providing elastin, and ground substance which is involved in filtration processes (Uitto and Larjava, 1991). Since collagen is involved in cell binding (as discussed previously), cell spreading and migration are also favored in a collagenous environment. Collagen fibers occurring in nature are highly ordered and aligned and serve as substrata for the E C M producing cells causing them to orient in the direction of the fibers (Dickinson et al, 1994). In other words, these collagen fibers provide contact guidance for the locomoting cells in vivo. Collagen gels have been used as a model for in vitro studies of cell interactions with the extracellular matrix. 8 A s mentioned previously, titanium surfaces with the proper topography and supplemented with extracellular adhesive proteins, can also serve as advantageous substrata. However, according to Lowenberg (et al, 1987) cells prefer the more natural extracellular material substrata to the metallic bio-material. One study involving a control and laminin-5 (a member of E C M glycoproteins) coated titanium substrata indicated that the protein coated metallic surfaces enhanced epithelial cell attachment and spreading as compared to uncoated titanium surfaces (El-Ghannam et.al., 1998). When titanium alloy surfaces were compared with demineralized root slices, attachment of cells and their orientation occurred with greater facility to the deminerilized substratum with the exposed collagen that has a high binding affinity to cells and fibronectin (Lowenberg et al, 1987). Surface-demineralization of roots that results in exposure of collagen of the cementum and dentin, appears to improve new attachment of periodontal tissue following surgery (Register and Burdic, 1975). The revealed collagen could act as a haptotactic agent attracting cells towards itself in preference to other E C M components. It has been proposed that such collagen may offer a more hospitable surface for fibroblast attachment and growth (Boyko et al, 1980). Lowenberg (et al, 1985) due to their finding of collagen-coating enhancing the cell migration, attachment, and orientation to untreated root slices suggested that such collagen coating could increase cell adhesion to surfaces used for implant applications (e.g. titanium) that cannot be otherwise demineralized (Lowenberg et al,1985). A study by Cooper et al (1993) showed that even though protein-coating (fibronectin, keratan sulfate, and the fibronectin derived peptide) of titanium increases the number of cells that bind to substratum, the relative binding to titanium was 5 to 10 times lower than binding to collagen I gels. Also , the collagen I matrices successfully compete with commercially pure titanium for binding of cells from solution when collagen was added to an agarose matrix surrounding a titanium disk. This means that in the presence of collagen I matrix, cell binding to titanium is greatly reduced (Cooper et al, 1993). 9 These studies confirm that specific cell receptor-matrix protein interactions can be a more effective means of attachment than the undefined process of cell-titanium interaction. The attachments to matrix components form faster, with higher affinity, and also seem to be more stable than the attachments to a metallic surface (Cooper et al, 1993). It should be noted, however, that cleaning procedure of a metallic surface prior to interaction with cells can also affect the cell binding; for example the R F G D treatment increases the affinity of titanium surface for cells (as discussed previously). Thus, comparisons of the preference of cells for titanium or collagen probably depend on the techniques used for the preparation of the titanium surface. 3. C E L L - C O L L A G E N INTERACTIONS a) C O L L A G E N AS S U B S T R A T U M Fibroblasts in vivo are surrounded by a macromolecular matrix of which type I collagen is a major constituent. The extracellular matrix plays an important role in regulating cell characteristics. In vitro, type I collagen in the form of three-dimensional gels is used as a substratum for cells. Such a three dimensional (3D) culture made of a reconstituted native type I collagen gel provides an environment that is thought to be more similar to in vivo conditions than the conventional culture dish. The fibroblasts cultured in gels have distinct characteristics compared to those plated on flat plastic or metallic surfaces. Such collagen gels populated with cells have been termed as "tissue equivalents" (Tranquillo, 1999) or "Fibroblast Populated Collagen Lattices" (FPCL) (Stephens et al, 1996). 10 Collagen is normally insoluble because of extensive covalent cross-linking. However it can be brought into solution without denaturation by cold salt or low ionic strength acidic solution from some vertebrate tissues (Elsdale and Bard, 1972). Solution can occur because of various factors: 1) the collagen is newly synthesized and not yet extensively cross-linked, as in skin of young rats and other mammals; 2) intermolecular aldimine cross-links that are readily opened at acidic p H predominate, this occurs in rat tail tendon; 3) a lathyritic agent was used which either inhibits aldehyde formation or combines with cross-link precursor and inhibits cross-linking, this occurs in animals which have been fed beta-aminopropionitrile (Chandrakasan etal , 1976). Upon raising the ionic strength (by addition of medium) and p H to physiological levels (by adding NaOH) the collagen wi l l be caused to precipitate and aggregate into native bundles forming a gel. The fibril formation is a self-assembly process and in vitro it requires at least three steps: The first step, initiation, involves a temperature dependent change which leads to an unidentified intermediate. The second step is linear growth of filaments (remaining very thin) by a process that is apparently temperature-independent. The third step is lateral association of filaments (formation of covalent cross-links) by a temperature-dependent process. The reaction times of both the second and the third steps are inversely proportional to collagen concentration suggesting that both linear and lateral growth occur by accretion (Gelman et al, 1979). It should be mentioned though, that the extent to which cross-linking can occur, and thereby the diameter of fiber, depends on whether a lathyritic agent was used in animals from which the collagen was isolated. Nevertheless collagen gels provide a useful model substratum and have been employed in many in vitro studies 11 b) C O L L A G E N A N D C E L L B E H A V I O U R Evidence has accumulated suggesting that physical cues, in addition to chemical signals, play an important role in modulating cell response to their environment. A use of 3D matrices often results in a better cell differentiation and enables the reconstitution of more in v/vo-like culture system (Berthiaume et al, 1996). For example, it is possible to generate HBEC-der ived (human breast epithelial cells) colonies with gross morphologies of alveoli and ducts (reminiscent of what is seen in a fetal mammary tissue) when a single cell is cultured in a collagen gel (Stingl et al, 1998). Epithelial cells cultured on floating collagen gels have also been found to rearrange themselves to form alveolus-like structures. When cultured on such floating collagen substrata, these cells develop secretory and myoepithelial specializations (Emerman and Pitelka, 1977). A mammary epithelial cell culture isolated from pre-lactating mice seems to acquire the ultrastruCtural and biochemical characteristics of differentiated mammary secretory cells in vivo when cultured on floating collagen membranes (Emerman et al, 1977). In vivo, fibroblasts are non-dividing or slowly dividing cells, while fibroblasts cultured on glass or plastics are induced to divide by fetal calf serum. The proliferation of gel-cultured fibroblasts has been found to be slower than that of monolayer cultured cells and to require higher concentrations of serum. This reduced proliferation of fibroblasts in 3D culture may be explained by their reduced responsiveness to growth stimulators (e.g. platelet-derived growth factor P D G F ) but equivalent response to growth inhibitors (e.g.prostaglandin E2 PGE2) (Mio et al, 1996). A s might be expected from the lower proliferation rate, the D N A synthesis by cells cultured in 3D lattices is decreased (Souren and Wijk, 1993). The decrease of D N A synthesis in cells cultured in a collagen lattices can be explained by the changes in cell cycle which cause an inhibition of cell advancement to the S phase and an accumulation of most cells in the either GO or Glphase (Kono et al, 1990). The proliferation can also be affected by cell shape. Cells 12 cultured on 2D and in 3D substrata have different shapes which may play a role in the different growth rates (Nakagawa et al, 1989). Seeded sparsely onto tissue culture plastic or glass, normal fibroblasts adopt the "ruffling membrane" form. These cells typically possess one or more broad, flattened pseudopodia, the edges of which may show the so-called ruffling activity. Also there seems to be no stable differentiation between the pseudopodial and non-pseudopodial surface, the pseudopodia can be produced from any point on the cell surface, giving the cell a constantly changing outline in time (Elsdale and Bard, 1972). Seeded sparsely onto a Hydrated Collagen Lattice, fibroblasts characteristically adopt the "bipolar spindle" form. These cells are greatly stretched with a clear differentiation between pseudopodial and non-pseudopodial surface, pseudopodia being confined to the two ends of a cell (Noble, 1987). Their pseudopodial processes are generally cylindrical and not as flattened as the lamella of cells moving on 2D substrata (Heath and Peachey, 1989 and Bard and Elsdale, 1973). Other studies also indicate that cells surrounded by type I collagen gel adopt a morphology and intracellular organization that closely mimic that found in vivo (Berthiaume et al, 1996). The bipolar spindle shape of cells in a collagen gel may reflect their orientation with respect to the axis of a collagen fiber. Studies show, that fibroblasts are responsive to mechanical load (provided by orientation of fibers), and mechanical stimulation can be a "cue" for the alignment of cells (necessary for many tissue engineering applications) (Eastwood et al, 1998). Normal fibroblasts adhere strongly to collagen lattices and it appears that the attachments they make to collagen-containing substrate are different from those made to plastic and other substrata. Not only are cells more easily removed from a non-collagenous substrate than from a 13 collagenous substrata, but also cells which are overlaid by a collagen gel release their attachments to the other substrata and move into collagen (Eladale and Bard, 1972). Within a 3D randomly polymerized gel, in the absence of known chemotactic factors, cells exhibit random locomotion in a zig-zag fashion. They retrace their path due to the fact that cells w i l l choose a path of least resistance. Cells alter their locomotory activity between the two polar ends of the cell, with a rest state between the change of direction (Noble, 1987) Metabolically, collagen production is decreased and proteinases and fibronectin production are increased when fibroblasts are cultured in gels (Nusgens et al, 1984). c) C O L L A G E N A N D C O N T A C T G U I D A N C E The morphology and behavior of tissue cells when surrounded by a net work of protein fibers, such as for a tissue-equivalent (T.Eq.) comprising cells entrapped in a type I collagen gel, is distinct from that when cells are cultured on a rigid surface, and physiologically relevant. The interaction of fibroblasts with extracellular matrix components like collagen appears to be a dynamic process integral to the formation of oriented fiber systems like the periodontal ligament (Fernyhough et al, 1987). It has been demonstrated that fibroblasts generate tension and change their orientation along the tensile direction. A t the same time collagen fibers become tense due to the forces exerted by the cells. The cells then, stretch themselves along such tense fibers and further increase tension in that direction creating a positive feedback loop (Takakuda and Miyar i , 1996). The consolidation of fibrillar network (in vitro know as gel compaction) results from the traction forces exerted by locomoting cells. This fiber alignment induces a cell orientation parallel to the fibers as a consequence of contact guidance (Tranquillo, 1999). The movement of a cell exhibiting contact guidance is characterized as "bi-directional", a cell having the maximum 14 probability of migrating in opposite directions along an axis of oriented extracellular matrix fibers (Guido and Tranquillo, 1993). Contact guidance is an important morphogenetic mechanism, where traction forces exerted by cells might even create the fiber orientation that then serves to guide their migration during development (Stopak and Harris, 1982). The ability of aligned collagen to induce contact guidance was shown by demineralizing dentin, thereby exposing collagen, and observing cell behavior in close contact. It appeared that cells migrate into and along dentinal tubules (Lowenguth et al, 1993). This phenomenon has been studied in vitro using collagen gels. Such fibrous substrata has also been found to guide cell migration in a particular direction (Dunn and Ebendal, 1978). A quantitative correlation for cell contact guidance in an oriented fibrillar network in terms of biased cell migration indicated that the measure of such biased cell migration in contact guidance increases with increasing collagen fiber orientation. Such directed cell locomotion can be explained by the fact that at first when the fibril orientation is at low levels cell migration along the axis of fibers is rapidly enhanced. When the fibril orientation reaches high levels, on the other hand, cell migration normal to the axis of fibers is suppressed (Dickinson et al, 1994). d) C E L L L O C O M A T I O N , M O T I L I T Y , AND T R A C T I O N F O R C E Nearly all cell types: leukocytes, fibroblasts, epithelial cells, pigment cells, muscle cells, even chondrocytes, exhibit active locomotion termed as a "crawling locomotion" (Abercrombie, 1980). This migration of tissue culture cells, however, is dependent on the presence of a solid substratum since such cells cannot swim (Harrison, 1914). In order to propel itself forward, a cell must exert rearward forces on something in its immediate environment (Harris, 1994). 15 Locomotion results from steady pulling forces, called traction, being exerted tangentially through the plasma membrane in the areas just behind the leading margins by a combination of "actin treadmilling" and "membrane raking". The direction of the traction forces is always centripetal, directly away from the part of cell margin undergoing protrusive activity, generally along the axis of stress fibers (Harris and Dmytric, 1988). Traction forces are thought to be exerted, in part, by contraction of a cytoskeletal-based actomyosin motor that transfers forces to the substrate through cell-matrix adhesive contacts. The crawling activities of tissue cells have several functions in addition to simply moving cells from one place to another. Fibroblasts exert forces very much larger than those needed for locomotion. The primary function of fibroblast traction has been proposed to be the morphogenetic rearrangement of fibers, whereby strong traction alters fiber orientation. Collagen can be displaced laterally by fibroblast traction as well as be induced to form a long-range alignment between two points separated by some distance (Harris et al, 1981). Distortions caused by cell traction play an important role in the organization of extracellular matrix during developmental morphogenesis (Stopak and Harris, 1982; Stopak et al, 1985) and wound healing (Ehrlich and Rajartnam, 1990). The migration of cells over substrata is a fundamental and critical function that requires the co-ordination of several cellular processes which operate in a repeated cycle. Five steps are involved in a cell migration cycle: 1) extension of the leading edge, 2) adhesion to matrix contacts, 3) contraction of the cytoplasm, 4) release from the contact site, and 5) recycling of membrane receptors from the rear to the front of the cell (Sheetz et al, 1999). These w i l l now be considered in sequence. Directed movement and polymerization occurs when one area of extension predominates. This extension is thought to be caused by directed actin assembly and polymerization. This process of actin assembly may involve significant addition of subunits to 16 one end of the filament. Such an elongation of the cytoskeletal element has to generate a protrusive force enabling the plasma membrane extension through the environment which is exerting forces in the opposite direction (Sheetz and Dai , 1996). Extension of the leading edge to new extracellular matrix (or other substrata) molecules w i l l enable receptors to bind and to initiate the adhesion process. The newly extended cellular domains of the membrane are stabilized by the complex created between the cell adhesion-receptor and the extracellular matrix component, allowing the cell to exert traction forces on the substratum. Involved in this process are integrins which are receptors for the E C M molecules and which are involved in cell migration. A n inverse relationship has been found between a cell 's adhesive and contractile ability and its speed of locomotion (Couchman and Rees, 1979). This means that the more adhesive cells must exert greater forces against cell adhesions to propel themselves forward and, hence, to exhibit more cell traction (and matrix distortion) with less cell locomotion. Also , in the absence of adhesive molecules when cells are cultured without serum, fibroblasts show an inability to form extensive microfilament bundles and focal adhesion complexes, suggesting reduced adhesion, locomotion, and contractility (Ridley and Hal l , 1992). The force required to pull the body of the cell forward is generated within the cytoskeleton and is expressed outside the cell as a traction force being exerted tangentially under the lamellipodium on the substratum (Sheetz et al, 1998). In leading lamellipodia, traction forces are oriented rearward as a result of the attachment of the ECM-bound integrins to cytoskeletal material that is moving centripetally rearward. This w i l l cause the contraction of the cytoplasm. For a forward migration to occur there must be an asymmetry in the adhesion process, such that traction forces in the front are greater than those in the rear. If the contractile forces are smaller in the rear of the cell, that also means that there must be less adhesion taking place in that part of the membrane. This can be achieved by a release from attachment A decrease in integrin attachment to the cytoskeleton at the rear of the cell makes the physical release from the substratum possible as the integrins are 17 pulled out of the cell. Translocating cells must replenish their matrix receptors at the leading edge. Although the insertion of newly synthesized protein accounts for some of the available receptor, several lines of evidence suggest that protein must also be recycled (Bretscher, 1996). The protein recycling can be accomplished by either one or both of the processes: a)endocytosis and reinsertion of integrins into the leading edge, b) forward transport of protein in the plasma membrane (Sheetz et al, 1999). e) C O L L A G E N G E L C O N T R A C T I O N Studies in vitro utilizing three-dimensional Fibroblast Populated Collagen Lattices (FPCL) have shown that there is a relationship between the tension developed by contracting fibroblasts and their orientation and the orientation of collagen fibers of a gel as well as the subsequent contraction of that gel. (Bellows et al, 1982). A development of tension is critical to this process since gels lacking cells or cultured in the presence of colcemid and cytochalasin D do not undergo any transformational rearrangements (Harris et al, 1981). The gel contraction has also been found to be controlled by cell-matrix and cell-cell interactions as well as by the cell migration which is related to the effectiveness of the contraction (Andujar et al, 1992). The sparse and randomly packed collagen fibrils originally found in the gels become densely packed and aligned by the cells during gel reorganization (Grinnell and Lamke, 1984). The rate and extent of gel contraction depends upon the concentration of cells and collagen, the type of cell incorporated into the gel (Bellows et al, 1981), the cell to collagen ratio, and the process of collagen purification (Frey et al, 1995). It also relies on the ability of the cells to attach to and spread along the collagen fibers. It has been found that the time of initial collagen gel reorganization correlates with initial cell spreading , and that cells that spread poorly compared 18 to normal cells have a decreased ability to cause collagen gel contraction (Steinberg et al, 1980). Others have also found that collagen contraction was decreased when fewer collagen fibrils were bound at the cell surface due to the absence of serum during incubation (Grinnell and Minter, 1978). The process of gel contraction can be divided into several steps: 1) the attachment of cells to collagen, 2) the spreading of cells within the collagen fiber matrix, 3) the organization and alignment of collagen fibers by cellular processes, 4) cell migration, 5) the establishment of intercellular contacts either by cell migration or by extension of cell processes along aligned collagen fibers (Bellows et al, 1981). After contraction is complete, the collagen fibrils are stabilized by two different mechanisms. Initially the collagen fibrils are reorganized mechanically by the cells, and the continued presence of cells is required to maintain this arrangement. Fibroblasts protrude their processes, bind the individual collagen fibrils and organize them into clusters and hold them in place (Yamato et al, 1995). With time, the mechanically rearranged collagen molecules became stabilized by non-covalent chemical interactions that are independent of cells (Guidry and Grinnell, 1986). Cells may either be grown on the surface of such gel cultures or actually embedded within the three-dimensional collagen matrix as a single-cell suspension when the gel is initially cast. Also , the gels may remain attached to a T C dish or they may become dislocated and be in a floating form. Depending on a particular combination of gel and cell plating , fibroblasts cause a gel contraction in either two or three dimensions. The contraction of unattached (not directly bound to underlying substratum), round, three-dimensional cell-containing collagen gels occurs in a radial fashion and results in a decrease in the diameter and thickness of the gel, meaning that the contraction happened in three dimensions (Bell et al 1979). In the condition in which cells were placed on top of gels attached to underlying solid support, the organization of collagen gels occurs in two dimensions rather than three. There is a decrease in gel thickness without a 19 decrease in diameter (Guidry and Grinnell, 1985). During contraction, collagen fibrils in attached collagen gels become aligned in the plane of cell spreading (Nakagawa et al, 1989). The final shape of a contracted gel depends on the number of fixed points at the time of gel formation. Fixed points have a different makeup from that of collagen gel and they can be in a form of T C dish walls, polystyrene cylinders, tissue explants, root slices, etc.. I f there are only two fixed points of attachment, as in two explants or cell aggregates with high cell densities, the collagen fibers became realigned parallel to the axis connecting the two points. A t points distributed from several millimeters to even several centimeters apart through out the matrix,,the actively motile fibroblasts coalesce and cluster gradually together. These aggregating cells pull on the collagen fibers and draw both the collagen and other fibroblasts together into developing aggregations. B y this process of fibroblasts traction pulling on collagen, the developing aggregations progressively enlarge themselves, and collagen becomes further compacted into and onto the accumulating masses of fibroblasts. The traction, here, has an additional effect of aligning collagen fibers into linear tracts running directly between adjacent concentrations of cells (Harris et al, 1984). The response of fibroblasts to the axial alignment of the matrix is to orient their movement axially, presumably through contact guidance all resulting in formation of a connecting bridge. This process is based on a positive feedback between cell orientation by contact guidance along matrix fibers and matrix orientation by the traction exerted by orienting cells (Stuart et al, 1972). This phenomenon was first observed by Weiss (1955) and was given a name of a "two center effect". When additional fixed points are distributed in a collagen gel they establish a pattern which consists of a set of axial traction fields interconnecting pairs of nearest neighbors. When fibroblasts are dispersed through the gel, very often the contracted gel w i l l end up in a form of a ring as many culture dishes are circular providing a circular outline of fixed points. Before contraction begins, a condensation of cells forms around the perimeter of 20 the gel first creating a "cellular ring" (Gullberg et al, 1990). The gel, then, pulls away from the sides of the dish, contraction proceeds quickly and results in a creation of a disk-like structure. In the interior of the gel, the matrix is compressed to form a very thin sheet which over time is slowly incorporated into the margins until they meet and form a single strand of tissue (Stopak and Harris, 1982). This interaction between cells and extracellular matrix is believed to be involved in such in vivo processes as embryogenesis, morphogenesis, wound healing and the processes of tissue engineering. During embryogenesis a number of intricate networks of tube, rods, cables, bearings and many other structures are created by the combination of molecular mechanisms and physical forces. Classical embryology describes how the movements of the primary germ layers lead to the formation of the primary body plan. Individual cells or sheets of cells sort out and create new organization patterns based on their embryonic germ layer origin. Their polarity is very important in this process since it leads to the non-random distribution of cellular structures which ultimately determine the morphology of cells and tissues. Therefore, environmental geometry plays a significant role establishing the morphology (Klebe et al, 1989). The orientation of collagen was shown to determine the orientation of cells with respect to their neighbors and reciprocally those cells have been found to organize collagen into a wide variety of different spatial patterns whose mechanical properties lend structural support and give form to tissues (Stopak et al, 1985). This intrinsic ability of cells to self assembly into organized structures is also one of the mechanisms involved in morphogenesis. Wound healing is directly related to an insertion of an implant since such a device creates a wound. The process of wound healing is complex and can be affected by many factors that 21 w i l l lead to dynamic cell-matrix interactions. During the healing process contraction takes place so that the final scar is considerably smaller than the original wound (Walter and Path, 1976). Wound contraction involves a volume change whereby normal dermal and adipose tissues are pulled into the defect by forces generated within fibroblasts (Berry et al, 1998). A n open wound is closed by the centripetal movement of the surrounding connective tissue whereby the cells organize their surrounding extracellular material by exerting traction (Enhrlich and Rajaratnam, 1990) and collagen fibers are laid down parallel to the preexisting matrix (Layfer et al, 1974). In tissue engineering applications, three dimensional insoluble scaffolds composed of synthetic or natural polymers are used as cell transplant devices. Collagen-based materials in the form of sponges frequently are used as solid three-dimensional networks for the preparation of artificial dermis or skin. Those matrices are chosen not only because of their biocompatibility, their modulable biodegradability but also because of their ability in guiding tissue regeneration (Chevallay et al, 2000). Another application of tissue engineering is the formation of skin substitutes as a treatment for chronic wounds and burn patients. This bio-engineered skin is made up of keratinocytes cultured on a mature (contracted and remodeled) tissue equivalent. Many studies in bio-engineering tissues also concentrate on the development of blood vessels, ligaments, and bronchi (Auger et al, 1998). 22 4. C E L L - C E L L INTERACTIONS a) M O N O L A Y E R I N G A s mentioned earlier, cells require a solid substratum to be able to move from one place to another. Cells in a liquid medium culture are presented not only with interface between the dish and the medium but also with the interface between the exposed surface of the neighboring cells and the medium. In the early stages of a culture's life, there seems to be a restriction on freedom of movement of fibroblasts over each other's surface. This restriction does not operate at a distance to hinder the mutual approach of cells but it rather operates only after contact has been established. Since contact is required for the directional prohibition of movement to occur, this phenomenon has been named "contact inhibition" (Abercrombie et al, 1957). Fibroblasts, are still free to make contact with each other, and in fact to adhere together. This leads to a tendency of cells to form a single layer of cells at the interface, a process termed "monolayering" (Abercrombie and Heaysman, 1954). A s a result of the mutual restriction of movement, neighboring fibroblasts appear to move in the same direction and at the same speed. The velocity of bodily displacement of a cell is affected by its contacts with other cells. It has been found that the velocity of movement of fibroblast is depressed by an established contact with another fibroblast and that it is momentarily increased after breaking the contact. There is an inverse relationship between the speed of movement of a cell and the number of cells with which it is in contact (Abercrombie and Heaysman, 1953). In a confluent culture, the same speed and direction of cell movement in a monolayer leads to formation of so called "parallel arrays" (Elsdale and Bard, 1972). A l l cells in such a group are found to move together en masse, thereby reaffirming the array (Erickson, 1978). A complete monolayer consists of a patchwork of many 23 independent parallel arrays separated by a frontier across which there is no exchange of cells. Cells in close contact can merge together in a horizontal plane to create a single larger array. The critical angle which determines whether the arrays merge or not has been found to be <20 degrees (Elsdale and Bard, 1972) or <55 degrees (Erickson, 1978) depending on the breadth of the lamellipodia. After a cell contacts the side margin of another, adhesion is created and localized inhibition of ruffling begins at that contact site and continues until alignment is achieved. In should be noted, that maintainance of parallel arrays requires the presence of collagen. The contact inhibition of cell locomotion described above, is thought to be responsible for immobilizing cells within an organism under normal conditions. However, in a case of a malignant tumour, such a lack of motility is not seen. Studies show that sarcoma cells do not undergo any detectable changes in behaviour after encountering a normal fibroblast. These invader cells are not inhibited by the cells they invade. It has been proposed that all malignant cells are characterized by an absence, or at least a reduction, of contact inhibition (Abercrombie etal , 1957). b) O R T H O G O N A L I T Y Dense confluent cultures in a monolayer arrangement produce extracellular matrix sheets which serve as new substrata for cellular colonization. Cells that colonize one matrix substratum may secrete another above themselves, which wi l l in turn be colonized. This multilayering has been observed only in the presence of collagen. Elsdale and Foley (1969) have found that the cells in the frontier line of a monolayer are stationary. On the basis of that finding they proposed that such non-motile cells are the most likely producers of new collagenous substrate, and 24 therefore, the frontier is the most likely place where the two-layered overlap is initiated. Furthermore, they agreed that it is possible for the motile cells to produce collagen. However, the locomotion of these cells creates a certain level of disturbance that prevents assembly of the fibers into a utilizable substratum (such is not the case around the stationary cells) while the fibers above the frontier cells remain unchanged. Primary overlaps are long and narrow, with the overlapping cells oriented perpendicular to the longer axis. Some of the primary overlaps become foci of aggregation, attracting cells in their vicinity and developing into wider ridges. The new layers of cells are oriented at approximately 90 degrees to their sublayers and that is why they have been termed "orthogonal multilayers" (Elsdale and Foley, 1969). 5. O B J E C T I V E O F T H E THESIS This thesis carefully examined the in vitro process of connective tissue reorganization and how it was be affected by the topography of an underlying surface. It observed cell behavior under three conditions. In the first, the fibroblasts were affected only by the topography of a titanium surface produced by micromachining. In the second, the cells responded to the addition of a collagen gel overlaying the established fibroblasts culture on a titanium surface. A n d in the third condition, cells were influenced by the presence of collagen gel within which they were entrapped and only after contact with titanium surface was achieved were they affected by its topography. The thesis also examined gel contraction and how it was affected by the presence or absence of cells or by the location of the cells with respect to the gel (within the gel or underneath it) leading to different fiber organizations and different end results of gel contraction. It also considered the distribution of cells between titanium and collagenous substrata. The 25 process of gel distortion was observed over time to gain insight into cell-titanium, cell-collagen, and cell-cell interactions. This study has found that micromachined topography affected cells in two ways depending on the pattern of a surface. If the surface was smooth or pitted, the cells formed parallel arrays. If the surface was grooved, the cells aligned with the direction of the groove. The alignment was decreased in the presence of collagen gel and cells within the gel were not affected by the underlying surface. When gel contraction was tested, it was found that fibroblasts were essential and their location would determine the final shape of contracted gel creating either a thin sheet or a ring. It has also been found that cells in close contact with both the titanium surface and collagen gel tend to prefer the collagenous matrix and their attachments to such a matrix are stronger and more stable. 26 C H A P T E R II M A T E R I A L S A N D M E T H O D S 1. Cel l Culture 2. Culture Conditions 3. Collagen Gels 4. Micromachined Substrata and "Boxes" 5. Light Microscopy 6. P S R and Polarized light Microscopy 7. PI, Eosin and Epifluorescence Microscopy 8. Time lapse studies 9. Data Collection 10. Statistics 27 1. C E L L C U L T U R E Fibroblasts were isolated from human gingival explants as described by Brunette et. al., (1976). H G F between the 6th and 11th subculture were cultured in Alpha Min imal Essential Medium (ocMEM) (Terry Fox Labs, Vancouver, B .C. ) supplemented with antibiotics (lOOug/ml penicil l inG (Sigma, St. Louis, Missouri), 50ug/ml gentamycin (Sigma), 3mg/ml amphotericin B (Fungizone, Gibco)) and 15% bovine serum (Fetal Clone III, HyClone, Utah) at 37° C in a humidified atmosphere with 5% CG*2. The medium was changed twice a week. In order to subculture, fibroblasts were removed from culture flasks using trypsin solution (0.25% trypsin, 0.1%) glucose, citrate saline buffer, p H 7.8), centrifuged, and resuspended in medium. Ce l l density (cell/ml) was determined using aCoulter Counter. Ce l l concentration used in experiments varied between 2.5-3.1x105 cell/ml (per ml of either medium or collagen mixture), the density at which cells were plated varied between 3.73-4.63x10 4 cel l /cm 2 , and the total number of cells on each surface varied between 7.5-9.3x10 4. 2. C U L T U R E CONDITIONS a) Control: Cells + Surface The behaviour of cells on different substrata in the absence of exogenous ExtraCellular Matrix (ECM)was determined. For these control conditions H G F at concentrations of 2.5x105 cells/ml of medium (density of 3.73xl0 4 cells/cm 2 and the total number of cells on the surface of 7 .5xl0 4 ) were plated on different surfaces and were cultured for 48 hours to allow the cells to be influenced by the topography of their substratum (TC dish, Smooth T i , 30G175P T i , 30G40P T i , 28 2G30P T i , 30IP175S Ti) . b) Control: Surface + Collagen Gel Gels without any cells embedded in them were plated on al l 6 substratum types, so that it could be determined whether a collagen gel without embedded cells has the capacity of self contraction and shrinkage, Time was allowed for solidification, after which the gels were treated the same way as cell cultures, i.e. supplemented with medium which was changed twice a week for a duration time of 4 weeks. The gel thickness was 3mm at the beginning and end of this period. c) "CellGel" Condition: (Surface + Cells) + Collagen Gel To determine the effects of a 3D collagen lattice on fibroblasts behaviour, as well as the competition between titanium substrata and collagen substrata, a gel was plated on top of a previously established cell culture. First, cells at concentrations o f 2.7-3.2xl0 5 cells/ml (with total cell number between 8.1-9.6x104 cells) were plated on top of the different substrata. Forty eight hours were allowed for the cells to attach and to be influenced by the topography. On day two a collagen gel was overlaid on top o f the cell culture. The gel was 3mm thick at this time which was to be the day 0 of these experiments which lasted up to 4 weeks. This condition was referred to as the " C e l l G e l " Condition. d) "Gel" Condition: Surface + (Collagen Gel + Cells) To determine the role of H G F in organizing collagen fibers, cells were embedded within collagen gels. Cells at concentrations of 2.7-3.2x10 5 cells/ml (with total cell number between 8.1-9.6xl0 4 cells) were re-suspended in collagen solution (see next section) and plated on substrata. The gel thickness was 3mm at the beginning of experiments that were terminated after 2 weeks. Again, these "tissue equivalents" were cultured in medium which was changed twice a week. This condition was referred to as the " G e l " Condition. e) "Boxes" with opposing microfabricated surfaces + (Collagen Gel + Cells) 29 To determine whether two micromachined surfaces separated by a distance of 1 cm could affect cell migration and collagen fiber alignment thereby inducing a so called two-centre effect, cells within a collagen gel (i.e. " G e l " condition) were placed inside a "box" of 1cm by 0.5cm in which two opposing sides separated by 1 cm had a microfabricated surface. f) Selective distribution of cells between Collagen Gel and Titanium The Cel lGel condition and the Gel condition were used to study the competition between titanium surfaces and a collagen gel. On day 7 of experiments the gels were gently stripped and the presence or absence of cells on the titanium surface was determined. g) Collagen Gel penetration into grooved surfaces To determine whether the collagen gel had the capacity to enter the grooves and settle there, gels were plated on grooved surfaces and their penetration into grooves was determined. 3. C O L L A G E N G E L S Many combinations of cell concentrations, collagen concentrations, p H , buffers, setting time and setting temperature were examined to determine the optimal conditions for gel formation. The cell concentrations tested varied between 1-10x105 cell/ml of collagen. The collagen concentrations used were between 0.5-2.9mg/ml (Bellows et. al., 1982, Lorimier S. et. al., 1998, Stephens et. al., 1996, MacNe i l R . L . et. al., 1996). Setting time was between 30 minutes to overnight. Setting temperature was either room temperature or 37°C. p H was adjusted by either N a O H or H E P E S buffer and was between 7-7.4. Those different variations resulted either in no gelation of collagen mix, cell death or no gel contraction. The selected system is based on that of R . L . MacNe i l et. al. (1996) with slight variations. In brief, the collagen used in all assays was Vitrogen 100 (Cohesion, Palo Al to , Ca), a pepsin solubilized type I bovine dermal collagen dissolved in 0.012N HC1. It is 99.9% pure collagen as judged by SDS Polyacrylamide Gel Electrophoresis in conjunction with bacterial collagenase 30 sensitivity. It is 95-98% type I collagen with the remainder being comprised of type III collagen. Vitrogen is a native collagen as judged by polarimetry and trypsin sensitivity, although it does contain a low percent of nicked or shortened helices. The Vitrogen 100 stock was at a concentration of 3mg/ml and p H of 2.0. It was refrigerated at 4°C when not in use. Collagen gels were prepared by the addition of 0.65ml Vitrogen (at 3.0mg/ml) to 0.15ml distilled sterile H2O, 0.1ml 0.1 M N a O H , and 0.1 ml lOx Min imal Essential Medium ( M E M , Sigma, St Louis, M o ) supplemented with lOOmM H E P E S buffer (Sigma, St. Loius, Mo. ) to obtain p H of 7.4. The collagen mix was allowed to set at 37°C for 45 minutes. The final collagen concentration in the gels was 1.95mg/ml. For Control Conditions: Surface + Collagen Gel this Collagen mix was plated by itself. For Gel conditions, a cell pellet was resuspended in the collagen mix to give the final cell concentration between 2.5-3.1x10 s cell/ml of collagen. For Cel lGel conditions, the collagen mix by itself was plated atop the previously established cell culture. 4. M I C R O M A C H I N E D SUBSTRATA AND "BOXES" W I T H OPPOSED M I C R O F A B R I C A T E D S U R F A C E a) Micromachining Micromachining is a useful tool in creating microscopic surfaces as it allows excellent control over the production of grooved or pitted surfaces' characteristics (depth, pitch, tapering, ridge and groove width) during different steps involved in the process. The techniques used to prepare the surfaces were originally developed by Camporese et. al. (1981) in the Department of Electrical Engineering at U B C for the fabrication of high quality photomasks for solar cells. Substrata were produced in the laboratories of the Center for Microelectronics, Department of Electrical Engineering at U B C , directed by Dr. N . Jeager. The wafers used in this study were 5 centimeters in diameter, 200-3 5 Oum in thickness, and had polished front surface and bright-31 etched rear surfaces. The overall process of creating the final micromachined wafers involved several stages (Chehroudi, 1991): a. Cleaning: The n-type (1-0-0) silicon wafers (Virginia Semiconductor Inc., Fredericksburg, V a , U S A ) were cleaned as follows: • 10 minutes in a solution of 300ml H 2 O , 60ml H 2 O 2 , and 60ml N H 4 O H at 75-85° C followed by a 10 minute rinse in distilled H 2 O • 30 seconds in 10% H F and rinsed in distilled water for another 10 minutes • 10 minutes in solution of 300ml H 2 O , 60ml 37% HC1, and 60ml H 2 O 2 followed by 10 minute rinse in distilled water • 4 minutes in isopropyl alcohol • blow drying in filtered nitrogen b. Oxidation A layer of Sil icon Dioxide (about 0.6um thick) on both sides of the silicon wafers was obtained by using wet oxygen at 1150°C for two hours in a two inch tube furnace [no. 7] (Fairchild Semiconductor Corporation, U S A ) . c. Photolithography A computerized optical pattern generator produced the desired pattern on a photomask. The polished front surface of the silicon wafer was coated with negative photoresist (Microposit S-1400 Series, Shipley, Newton, Massachusetts) and the photomask was placed atop the silicon wafer. This setup was exposed to 320 nm wavelength UV-l ight , developed, and baked at 160°C. The alignment of the photomask, with the desired pattern, on the silicon wafer is critical for obtaining the precise dimensions of depth and spacing. Misalignment of the two may lead to a less predictable etching. 32 d. Oxide Patterning The oxide layers that were not protected during the exposure to U V were removed by the buffered H F , and the remaining developed layer was removed in microstrip ( N M P ) solvent (Microstrip 2001, Ol in Hunt Specialty, Inc., West Paterson, N J , U S A ) . A s a result, the desired pattern on the silicon wafer was in the form of an oxide layer. e. Final Etching The wafers were etched in 19% potassium hydroxide solution at 80° C which removes the silicon but leaves the oxide layer intact. The speed of etching is typically 1.4 um/minute on the 1-0-0 wafers, but it varies with the change in crystalline arrangement of the silicon wafer and so it is about 300 times slower i f the crystal orientation is "111". A s mentioned earlier, the pitch (width of one groove plus the width of one ridge) is closely controlled by the pattern created on the photomask. The shape of the grooves is determined by the crystal orientation in the silicon wafer. The 1-0-0 wafers used in this study produce V-shaped grooves. The depth of the designed features (e.g. grooves), on the other hand, is controlled by the time of etching. f. Titanium Coating To ensure chemical homogeneity of all micromachined substrata used in the experiments, a titanium-coating protocol adopted by Brunette (1986b) was used. The process involved a sputter coating of wafers with 50nm of titanium g. Topographies The surfaces used in these experiments were as follows: • Smooth: used as control • 30G175P: Grooves with 30um depth, 140um wide ridge, and 35um wide groove. • 30G40P: Grooves with 30um depth, 5um wide ridge, and 35um wide groove • 3G30P: Grooves with 3um depth, 15um wide ridge, and 15um wide groove • 30IP175S: Pits with 30um depth, that are 175um apart and in a shape of Inverted 33 Pyramids (IP). These surfaces are illustrated in Fig . 1. b) "Boxes" with opposed microfabricated surfaces The design was developed by M s . K i m . and the process involved the following steps: a. A 1cm by 0.5 cm "Lego" piece was surrounded by Putty Soft material (President, Coltene, Mahwah, NJ) to create a "working bath". b. Small amounts of impression material (Provil light, Heraeus, South Band, IN) was placed on the T i micromachined wafers and was forced in with an Interplak toothbrush. Those wafers then were placed on the opposite sides of the "working bath" t (i.e. surfaces separated by 1 cm) and the bath was filled with dental impression material to create a negative "box". c. Negative "boxes" were surrounded, but not covered, by epoxy resin (Epo-Tek 302-3M, Epoxy Technology, Billeria, Ma.) to create the working "boxes" that would have the desired micromachined patters on the opposite sides. The epoxy replicas were allowed to air dry overnight following 3 day baking at 60°C to increase polymerizing of the epoxy. d. The "boxes" were coated with 50nm of titanium. However both epoxy and titanium-coated "boxes" were used in experiments. c) Preparation of substrata for use a. Cutting A l l Titanium coated silicon wafers were 5cm in diameter and contained 6 or 9 different patterns. Therefore, the desired pieces were cut out using a diamond-tipped glass cutter. b. Washing Both the wafers and "boxes" were cleaned by placing them into beakers with 50:50 water to 7 X detergent (7X, I C N Biomedicals Inc., Costa Mesa, California) solution and ultrasonicating for 15 minutes. The surfaces were then extensively rinsed first in tap water then by distilled water. A 15 minute ultrasonication in distilled water followed. The surfaces were air dried in laminar-34 flow hood. A t the end of each experiment the Titanium wafers as well as the forts were reused. The collagen gels were discarded and the surfaces were immersed in trypsin solution for at least 30 minutes at 37°C to detach the remaining cells. The previously described cleaning process followed. c. Radio-frequency glow-discharge treatment (Baier and Meyer, 1988) In this process, a controlled electrical gas discharge (or plasma) is used to generate an electric ion field that physically bombards the surface. Samples are exposed to an ionized low pressure (0.1-1 mm Hg) noble gas (argon) which removes the existing passivated oxide layer by a process of ion stripping. Upon a release of vacuum, a new oxide coating is spontaneously created upon reaction with ambient oxygen (Stanford et. al., 1994). The resultant oxide layer hasa higher surface energy, leading to an increased cell attachment and spreading (Swart et. al., 1992) This process w i l l ash away organic contaminants to render the surfaces sterile. 5. L I G H T M I C R O S C O P Y Live cultures were observed under Zeiss Jena Microscope (lOx objective, Zeiss Jena, Germany) and digital pictures were taken with a Hammamatsu Video camera (Hammamatsu, Japan) on day 0, 1, 3, 7, 10, 14, 20, 22, and 28. To determine the depth of focus the following formula (James J. and Tanke H.J . , 1991) was used: D 0 =n[X/2(N.A.)2 + 0.34/M(N.A.)] where: D 0 is the depth of field, n is the refractive index of surrounding medium (i.e. collagen 35 gel), X is the wavelength of light (nm), N . A . is the numerical aperture of the objective, and M is the magnification. The Refractive Index of the collagen gel was measured in the department of Chemistry, U B C and determined to be n= 1.335. Wavelength value used is 656nm. The Numerical aperture of the objective used is 0.2 and the overall magnification is lOOx. Therefore the final depth of field is about 11 um. To determine the vertical location of cells within the gel the microscope focusing knob was calibrated against surfaces of known depth to determine the rotation value equivalent to certain vertical dimension. 6. PSR AND P O L A R I Z E D L I G H T M I C R O S C O P Y One of the classic characteristics of a highly ordered structure such as collagen is birefringency, that is the ability to rotate the plane of plane-polarized light. A specific and stable dye frequently used to stain collagen is Sirius Red. Sirius Red not only stains red but it also enhances the normal birefringency of a collagen molecule. The configuration of both the collagen and dye molecules after a bond is created is well matched. The long axes of Sirius Red and the fiber are aligned parallel allowing the examination of collagen fiber orientation (Puchtler etal , 1973). The "tissue equivalents" and simple collagen gels were stained using picrosirious red method on day 7, 14, 20, 28. The staining of collagen gels was as follows: 1. Gels were fixed for 1 hour in 100% Buffered Formalin Phosphate (Fisher Scientific, NJ) at room temperature. 2. A quick rinse in distilled water followed. 3. A quick rinse in 95% alcohol. 4. A 10 minute treatment in alkaline alcohol (95% alcohol with ammonium hydroxide (Fisher Scientific, NJ) to raise the p H above 8.0) at 60°C was the next step 5. A quick rinse in distilled water 36 6. Gels were stained for 1 hour at room temperature in Picrosirius Red (Saturated aqueous picric acid 100ml with O.lg of Gurr's sirius red F3B ( B D H , England)). 7. Several changes of 1% acetic acid. 8. Mount in aqueous mount (50% glycerol in PBS) to prevent cell shrinkage and to preserve cell height. Place cover slip over the gels. 9. Examination of stained "tissue equivalents" was done using lOx objective on Zeiss Jena Microscope (Zeiss Jena, Germany) with crossed polarizers (ie. the transmission plane of the polarizer located above the light source is perpendicular to the transmission plane of the analyzer positioned above the objective lens). Pictures were taken with the Hammamatsu video camera (Hammamatsu. Japan). Polarization microscopy depends on a modified form of illumination to gain specific information about a specimen. It is a light microscopic technique in which a polarizer located close to the source of illumination and an analyzer located above the objective lens are oriented at right angles to each other so that only light waves whose plane of polarization has been altered are visible in the eyepiece. The P S R staining of collagen enhances the birefringence and therefore makes collagen fibers more apparent. 7. PROPIDIUM IODIDE AND EOSIN: E P I F L U O R E S C E N C E M I C R O S C O P Y To determine cell orientation on different substrata and within the 3D collagen gels, cells were stained for their nuclei and the orientation of nuclei was measured. Propidium Iodide (Sigma Cgemicals, St. Loius, Mo.) was used to stain the nuclei and Confocal Microscopy was used to scan different sections of a gel. Precise focal planes can be selected because out-of-focus information is eliminated from the final image. Confocal imaging of cells within the gels makes it possible to distinguish the orientation of cells and their distribution in the 3D lattice. 37 PI staining procedure as developed by M r . A . Wong, Faculty of Dentistry, U B C , was as follows: 1. A quick rinse in warm medium without serum and antibiotcs. 2. Fixation in 100% ethanol for 2 minutes. 3. 2 minutes in each: 95%, 90%, 75%, 50%, 30% ethanol, and distilled water. 4. Stain gels and cells with 50-200ul of Propidium Iodide ( l g / L in PBS) in the dark for 20 minutes. 5. Rinse quickly in distilled water 3 times. 6. Rinse quickly in 100% ethanol. 7. Mount in immersion oi l and place cover slip on top. 8. Store in the fridge wrapped in foil to keep dark. 9. Examine under fluorescent microscope with U V using a rhodamine filter (with wavelength of A,max=543nm; To reduce background information the following were used: the excitation filter which transmits radiation in the band 540-560nm, the emission filter which transmits radiation above 580nm, and the chromatic beam splitter which allows light whose wavelength is greater than 580nm to be transmitted). Confocal Laser Scanning Microscope (Zeiss) was used with a helium-neon laser. 20x objective was used and sections were taken every 15um. The depth of focus was about 1.5um. To ensure that no cells were missed since the scanning sections were so thick, 3 sections scanned 15 urn apart were compared with 31 sections scanned l u m apart. The same cells showed up on the composite of both methods. To determine the extent to which collagen gels infiltrated grooves, the gels were stained with Eosin in the following manner: 1. Gels were fixed for 1 hour in 100% Buffered Formalin Phosphate ((Fisher Scientific, NJ) at room temperature. 2. A rinse in distilled water followed. 3. A 2 minute treatment in each 75%), 90%, and 95% alcohols. 38 4. A 2 minute treatment in 1% Eosin in 95% alcohol. 5. Three rinses in 100% alcohol. 6. Mount in immersion oi l . 7. Cover slip placed on top. Examination was done under the C L S M , as described before. 8. T I M E - L A P S E STUDIES The collagen gel with cells was placed on top of a grooved titanium-coated wafer inside a small vertical rectangle. This set-up was attached to a glass slide and placed inside the Pentz chamber (Bachofer, Reutlingen, Germany). This chamber was fixed inside the stage incubator (heated and supplied with air with 5%C02) on a upright microscope (Reichert, Vienna, Austria) equipped with an Epiplan 16x objective (Zeiss) and with reflected Nomarski Differential Interference Contrast (DIC). Reflected light was used because the titanium coated substrata are opaque. The D I C was used to improve the contrast specimen. A cell has different optical densities in different areas which affects the path of light going through those different areas. DIC exaggerates the contrast between such areas producing a clear image. A 3 C C D color camera (Sony Corp., Japan) video camera was used with a combination of a image program Northern Eclipse (Empix, Missasagua, Ont) . The pictures were taken every 5 minutes for up to 5 weeks. Movies were transferred to an Apple computer and were edited using Scion Image 1.62. 9. D A T A C O L L E C T I O N Data was collected from the images obtained from Confocal Laser Scanning Microscopy. During the transfer from C L S M to a computer the images were slightly distorted. To compensate for the distortion images were converted back to their original proportions. The N I H 39 Image 1.61 program was used to measure the orientation of cells. The method used for measuring orientation angle of cells on different surfaces is illustrated in Fig . 2. The orientation angle was measured with respect to a randomly selected axis (in the case of smooth surfaces and T C dishes), with respect to the axis of the grooves (in the case of all grooved patterns), and with respect to the axis of either side of the pit (in the case of 30IP175S and smooth for control comparison). The two reference orientation axis on the pitted surface restricted the maximum cell angle to 45 degrees (45 degrees being half way between the two axis, passing a 45° angle brings the cell closer to a second axis and decreases the angel below 45°). The measurements were expressed as the percent of cells in each 10 degree increment over the range of 0-90 degrees (for Smooth, T C dish, 30G175P, 30G40P, and 3G30P) and a range of 0-45 degrees (for 30IP175S). The increments of angles were distributed in the following manner; 1 (0.00-9.99°), 2 (10.00-19.99°), 3 (20.00-29.99°), 4 (30.00-39.99°), 5 (40.00-49.99° or 40.00-45.00°), 6 (50.00-59.99°), 7 (60.00-69.99°), 8 (70.00-79.99°), 9 (80.00-90.00°). The cells were grouped depending on where in the gel they were found. The cells found at the bottom of the gel and on the surface (Oum) were said to be in Zone 1. Cells found between Oum and 30um from the bottom of the gel were said to be in Zone 2, while cells found more than 3 Oum away from the surface were said to be located in Zone 3. 10. S T A T I S T I C S A l l data was entered into S P S S - X (SPSS Inc., Chicago, Illinois). Data was analyzed by a non-parametric Kruskal-Wallis test. The orientation of cells in different zones (Zones 1, 2, and3) and under different conditions (Control, Cel lGel , and Gel) on each surface type at different time points (one and two weeks) was compared. The P value less than 0.05 indicated statistical significance. 40 Figure 1. The cross section of surface topographies used in this study. Figure 2. A diagrammatic representation of the methods used to measure the orientation angel on different surfaces: Grooves: angle measured with respect to the axis of the grooves; Smooth and T C dish: angle measured with respect to a randomly chosen line; Pitted: angle measured with respect to the closer axis created by the two sides of the pits . 41 Figure 1: CROSS SECTIONS OF SURFACE TOPOGRAPHIES 3G30P ^ 15fxm 15(im 30G40P 35fim 5u.m r 30um I V W W W W X 30G175P 140fim 35nm 30jimT V 30IP175S 130fuii 45Pm • • 30^ mT 42 Figure 2: MEASUREMENTS OF C E L L ORIENTATION GROOVES: Relative to the direction of grooves 1 4 e7 II SMOOTH: Relative to randomly chosen line^ • • • PITTED: Smallest angle to x or y • y • • y • • y ^ • • • • • 4 • • • • • 43 C H A P T E R III R E S U L T S A number of experiments were performed to determine the result of various conditions and various surfaces on cell orientation and collagen gel contraction. The contraction of the gel depended on the presence of cells as well as their distribution throughout the matrix and resulted in either no contraction (in the absence of cells), in formation of thin sheets of cells in fibers (under the Cel lGel Condition), or in the formation of a ring structure (under the Gel Condition). The different surfaces either caused the cells to form patchworks of parallel arrays of cells and fibers (TC dish, smooth titanium, and pitted titanium) or to align with the direction of the grooves (all grooved surfaces). Cells above surfaces did not seem to have any particular orientation. After the preliminary observations were made from a number of initial experiments, the data was collected from experiments that were repeated three times. In each experiment, each condition and surface was studied in duplicate. The quantitative and qualitative evaluations of these experiments w i l l be presented as follows. First, the numerical data on the orientation of cells within the gels w i l l be presented. A s different surfaces were used, their effect on cell orientation w i l l be discussed in order of increasing complexity of topography. First, results on T C dish w i l l be presented, then on a titanium coated smooth surface. Grooves present a more complex topography, and so the results on grooved titanium substrata w i l l be discussed next in the following order: the shallow grooves (3G30P), the deep grooves with a narrow ridge (30G40P) and last the deep grooves with a wide ridge (30G175P). The final surface to be discussed w i l l be the most complex substratum used in this study, the pitted (30IP175S) titanium surfaces that provide two possible axis for cell alignment. Each of those surfaces w i l l be discussed with respect to three experimental conditions, starting with the simplest one (no gel) and ending with the most complex one (cells 44 in a gel), that is in the following order: 1. Control: Cells + Surface, 2. Ce l lGel : (Surface + Cells) + Collagen Gel , 3. Gel : Surface + (Collagen Gel + Cells). Also data on cell distribution in different zones under the Cel lGel and the Gel conditions on all the surfaces w i l l be presented. Secondly, some qualitative observations w i l l be reported. These observations comprise descriptions of cell or gel behaviour. One aspect of this observed data is concerned with the end result of gel contraction on different surfaces (presented once again in the order of increasing complexity of topography under different experimental conditions (presented in the order of increasing condition complexity)). Another aspect of the qualitative data is the distribution of cells between collagen and titanium under the two collagen gel conditions (Cel lGel and Gel) that was observed after the stripping of the collagen gel off the titanium surface. Collagen gel distribution on a surface w i l l also be described. Finally, the results from experiments involving collagen gels with cells (Gel condition) in boxes with opposed micromachined surfaces wi l l be presented. I. Q U A N T I T A T I V E D A T A 1. C E L L O R I E N T A T I O N WITHIN T H E G E L S a) TISSUE C U L T U R E PLASTIC DISH Fig.3 (1 week) shows data on orientation of cell alignment on Tissue Culture plastic dish under three conditions: the control, the Cel lGel conditions, and the Gel conditions. On this surface, a total of 615 cells were measured for orientation. A non-parametric Kruskal-Wallis test, comparing cell orientation under different conditions and in different Zones was performed (i.e. 7 groups: 1. Gel/Opm (70cells), 2. Gel/<30um (49cells), 3. Gel/>30um (164cells), 4. CellGel/Oum (70cells), 5. CellGel/<30pm(45cells), 6. CellGel/>30uum(67cells), 7. Control/0uum(149cells)). In order to test the hypothesis that cell 45 orientation on T C dish did not differ between the three conditions, groups 1, 4, and 7 were compared together one by one to determine whether there was a significant difference between cell orientation in Zone 1 under the three conditions (1:4 P=0.62, 4:7 P=0.08, 7:1 P=0.3). Also , in order to test the hypothesis that cells in the different Zones did not differ in their orientation groups 1, 2, and 3 (for Gel condition) as well as 4, 5, and 6 (for Cel lGel condition) were compared to determine whether cell orientation differed in the three Zones under a particular condition. The statistical tests (P value> 0.05) indicated that the groups do not differ significantly, indicating that the cell orientation on the surface as well as within the gel under different conditions was similar. Indeed, in figure 3 there are no clear trends of cell orientation . i) Control Condition: Cells + Surface A l l cells are found on the surface since a collagen gel was not present in this condition. Therefore, only cell orientation in Zone 1 is shown. Clark et al. (1990) defined cells as aligned when their orientation angle is less than 10 degrees (increment 1) with respect to a given axis. The T C dish surface provided no cues for directing cell orientation and it would be expected that the orientation of cells would be random. Indeed, the data presented in figure 3 shows no particularly favored orientation. The average orientation angle in this group was 30.8 degrees ± 4 . 1 3 (95% confidence interval: CI) It should be noted that cells grown under this condition formed parallel arrays and it was those arrays that dictate the orientation of cells. A s all the cells in a field were counted, the existence of these parallel arrays resulted in peaks in the frequency distribution of angles of cell orientation. Nevertheless, it should be pointed out, that in these confluent cultures neighboring cells act on each other to bring each other into self alignment. It has been shown in previous studies that topographies which have only small degree of alignment at low cell densities, have a much greater effect at high cell densities where a population pressure drives the cells to become locally aligned at confluence (Trinkaus, 1984). 46 ii) CellGel Condition: (Surface + Cells) + Collagen Gel Cells were distributed throughout the three dimensional matrix. They were found in the first zone right on the surface as well as in the second zone close to the surface (<30um from the surface) and in the third zone further away from the surface (>30um). Cells in all three Zones: 1, 2, and 3 showed no preference in their orientation. The average orientation angles were 44.1° ± 2.09 (95%CI), 38.9° ± 6.85 (95% CI), and 41.1° ± 5.68 (95% CI) in Zones 1, 2, and 3 respectively. The cells in Zone 1, however, formed a patchwork of parallel arrays, just like in the control condition. The presence of gel above the arrays, did not induce noticeable new arrangements of the cells on the surface. iii) Gel Condition: Surface + (Collagen Gel + Cells) Cells in Zone 1, 2 and 3 showed no trend in their orientation. The average orientation angles were 40.5° ± 6.06 (95% CI), 36.7° ± 6.64 (95% CI), and 44.6° ± 3.87 (95% CI) in Zones 1, 2, and 3 respectively. After migrating to the bottom of the gel, cells in Zone 1 formed a patchwork of parallel arrays with no favored direction of orientation. It appeared that the cells adopted the same pattern of orientation regardless of whether they started out on the surface or whether they migrated from the gel to the surface, b) S M O O T H T I T A N I U M S U R F A C E Figure 4 (1 week) shows cell orientation by Zones within a three dimensional matrix on a smooth titanium surface under three conditions: the control, the Cel lGel condition, and the Gel condition. A total of 2035 cells were measured for orientation. A non-parametric Kruskal-Wallis test, comparing cell orientation under different conditions and in different Zones was performed (i.e.7 groups: 1. Gel/Oum (386cells), 2. Gel/<30um (64cells), 3. Gel/>30um (153cells), 4. CellGel/Oum (862cells), 5. CellGel/<30um (171cells), 6. CellGel/>30uum(l 17cells), 7. Control/Oum (282 cells)). Groups 1, 4, and 7 were compared together one by one to determine whether there was a significant difference between cell 47 orientation in Zone 1 under the three conditions (1:4 P=0.63; 4:7 P=0.09; 7:1 P=0.16). Also , groups 1, 2, and 3 (for Gel condition) as well as 4, 5, and 6 (for Ce l lGel condition) were compared to determine whether cell orientation differed in the three Zones. The statistical tests indicated that the groups did no differ significantly (P > 0.05), indicating that the cell orientation on the surface as well as within the gel under different conditions was similar. Indeed, in figure 4 no clear trends of cell orientation are evident. i) Control Condition: Cells + Surface A l l cells were found in Zone 1 as there was no superior layer. The smooth surface did not provide any cues for directed orientation of cells, and so the cells did not show any preference in their orientation, with the average orientation angle being 44.3° ± 2.47 (95% CI). Cells on this surface behaved similarly to the cells on T C dish. The cells also formed parallel arrays which were reflected by peaks in a few of the increments, because of sampling artifacts. The orientation of the parallel arrays with respect to a randomly chosen axes appeared to be random. ii) CellGel Condition: (Surface + Cells) + Collagen Gel Because a three dimensional matrix was present in this condition, cellular orientation in the three Zones is presented in figure 4. Cells in Zone 1 behaved just like the cells in control condition. Their orientation on the surface showed no particular favored orientation (average angle of 42.0° ± 3.66 (95% CI)). The cells did form a patchwork of parallel arrays. A s noted earlier, formation of parallel arrays leads to peaks of cell orientation in some of the increments. The collagen gel overlaying the cells did not seem to dramatically alter their orientation nor induced any new orientational pattern. Cells in Zones 2 (average orientation angle of 49.9° + 3.53 (95% CI)) and 3 (average orientation angle of 45.7° ± 4.66 (95%> CI)), as on the T C dish, did not exhibit any preference directional 48 orientation. iii) Gel Condition: Surface + (Collagen Gel + Cells) Cells in Zone 1 (with average orientation angle of 40.1° ± 1.90 (95% CI))formed parallel arrays oriented randomly. The peaks in increment 5 and 6 simply indicated either more parallel arrays or bigger arrays oriented at angles 40-59.99 degrees to randomly chosen axes in some samples. Again, no clear trends in cell orientation were seen in the collagen gel. Cells in Zones 2 (with average angle of 41.8° ± 6.45 (95% C I ) ) and 3 (with average angle of 46.5° ± 3.66 (95% CI)) showed no preference in their arrangement within the gel. A representative orientation of cells in gels on T C dishes and smooth titanium surfaces (cells on pitted titanium behave in a similar fashion as well) is diagrammatically illustrated in figure 5. To summarize, it seems that neither of the two surfaces: T C dish and smooth titanium induced a guided cell orientation on the surface in Zone 1 or above the surface in the other two zones: 2 and 3. The cells in Zone 1 formed a patchwork of parallel arrays, probably due to their higher cell density. The cells in Zones 2 and 3 were individually distributed throughout the gel and showed no preferred pattern in their orientation. c) 3G30P: T I T A N I U M W I T H S H A L L O W (3um) G R O O V E S Figure 6(1 week) and Figure 7 (2 weeks) show cell orientation by Zones within a three dimensional matrix on the 3 pm deep grooved substrata with a 15 pm wide groove and a 15 pm wide ridge under three conditions: the control, the Cel lGel condition, and the Gel condition. A total of 1411 cells after 1 week and 2099 cells after two weeks were measured for orientation. A non-parametric Kruskal-Wallis test, comparing cell orientation under different conditions and in different Zones was performed (i.e.7 groups: 1. Gel/Oum (205cells andl95 cells after 1 and 2 49 weeks respectively), 2. Gel/<30p,m (164cells and 69cells after 1 and 2 weeks respectively), 3. Gel/>30um (231cells and 301cells after 1 and 2 weeks respectively), 4. CellGel/Oum (461cells and 768cells after 1 and 2 weeks respectively), 5. CellGel/<30um (156cells and 359cells after 1 and 2 weeks respectively), 6. CellGel/>30um (40cells and 253 cells after 1 and 2 weeks respectively), 7. Control/Opm (154cells)). Groups 1, 4, and 7 were compared together one by one to determine whether there was a significant difference between cell orientation in Zone 1 under the three conditions (1:4 P=0.0000; 4:7 P=0.0000, 7:1 P=0.0000). Also , groups 1, 2, and 3 (for Gel condition) as well as 4, 5, and 6 (for Cel lGel condition) were compared to determine whether cell orientation differed in the three Zones under a particular condition. The statistical tests (P< 0.05) indicated that the difference in cell orientation in Zone 1 under the three conditions was significant and the difference in cell orientation in Zone 1 as compared to Zones 2 and 3 was also significant. i) Control Condition: Cells + Surface Again, all cells were located on the surface and so all cells were affected by the topography. A l l cells were found to be aligned with the axis of the grooves and the average orientation angle was 2.2° ± 0.27 (95%CI). ii) CellGel Condition: (Surface + Cells) + Collagen Gel Cells in Zone 1 were affected by the topography since they were in contact with the surface. 81.1% after 1 week (average angle of 8.7° ± 1.43 (95% CI)) and 89.5% of the cells after 2 weeks (average angle of 7.3° ± 1.16 (95% CI)) were found to be aligned with the grooves. The orientation of the remaining cells varied from 10 degrees to 90 degrees. Since the only difference between this condition and the control condition was the presence of collagen gel atop the cells, it can be assumed than that the decrease in cell alignment was caused by the overlaying gel. Cells in Zone 2 (with average orientation angle of 44.0° ± 3.66 (95% CI) after 1 week and 43.1° ± 2.58 (95% CI) after two weeks) and in Zone 3 (with average angle of 38.2° ± 6.26 (95% 50 CI) after one week and 44.3° ± 3.2 (95% CI) after two weeks) showed no particular trend in their orientation with respect to the underlying substrata. iii) Gel Condition: Surface + (Collagen Gel + Cells) 72.2% of cells in Zone 1 were aligned with the grooves after 1 week (average angle 12.0° ± 1.65 (95% CI)). A t two weeks, 84.6% (average angle 6.6° ± 2.22 (95% CI)) of cells were aligned with the grooves. The cells not classified as aligned were oriented over a wide range of angles with respect to the axis of the groove (increments 2-9). Relative to the Control and Cel lGel condition the cell orientation was decreased in the Gel condition (as described previously tested by Kruskal-Wallis test). Over the period of two weeks more cells became aligned with the grooves (discussed in detail in the Discussion section). Ce l l orientation in Zone 2 (with average angle of 46.6° ± 3.34 (95% CI) after one week and 40.3° ± 5.19 (95% CI) after two weeks) and in Zone 3 (with average angle of 40.1° ± 3.12 (95% CI) after one week and 32.3° ± 2.34 (95% CI) after two weeks) was distributed throughout all 9 increments with no preferred orientation. To summarize, the topography of this 3G30P surface produced a high level of alignment of fibroblasts under control conditions. The alignment was decreased by placing a gel atop an established culture and the effects of topography were further masked by the effect of collagen when cells were embedded within the gel at plating. d) 30G40P: T I T A N I U M W I T H D E E P (30um) G R O O V E S A N D N A R R O W RIDGE Figure 8(1 week culture) and Figure 9 (2 weeks culture) show data on cell orientation by Zones within a three dimensional matrix on the 30um deep grooved substrata with a 35pm wide groove and a 5 urn wide ridge under three conditions: the control, the Ce l lGe l condition, and the Gel condition. A total of 1756 cells after 1 week and 2582 cells after two weeks were measured for orientation. A non-parametric Kruskal-Wallis test, comparing cell orientation under 51 different conditions and in different Zones was performed (i.e.7 groups: 1. Gel/Oum (175cells and 52 cells after 1 and 2 weeks respectively), 2. Gel/<30pm (29cells and 160cells after 1 and 2 weeks respectively), 3. Gel/>30p,m (315cells and 594cells after 1 and 2 weeks respectively), 4. CellGel/Oum (589cells and 822cell after 1 and 2 weeks respectively), 5. CellGel/<30um (198cells and 473cells after 1 and 2 weeks respectively), 6. CellGel/>30uum(218cells and 249 cells after 1 and 2 weeks respectively), 7. Control/Oum (232cells)). Groups 1, 4, and 7 were compared together one by one to determine whether there was a significant difference between cell orientation in Zone 1 under the three conditions (1:4 P=0.0000; 4:7 P=0.0000, 7:1 P=0.0000). Also , groups 1, 2, and 3 (for Gel condition) as wel l as 4, 5, and 6 (for Cel lGel condition) were compared to determine whether cell orientation differed in the three Zones under a particular condition. The statistical tests (P< 0.05) indicated that the difference in cell orientation in Zone 1 under the three conditions was significant and the difference in cell orientation in Zone 1 as compared to Zones 2 and 3 was also true. i) Control Condition: Cells + Surface A l l cells were found on the surface since a collagen gel was not present in this condition. One hundred percent of the cells were aligned with the axis of the grooves, using the criteria of Clark et. al. (1990). The average orientation angle was 1.2° ± 0.16 (95% CI). This surface produced the maximum cell alignment observed in this study. ii) CellGel Condition: (Surface + Cells) + Collagen Gel The orientation of cells right on the surface was affected by the features of the surface. Those cells exhibited a very high level of alignment with the grooves. 98.5% in the first week and 98.3% of cells in the second week were found to be orientated within the first increment. The average orientation angles in Zone 1 were 1.5° ± 0.28 (95% CI) and 2.1° ± 0.33 (95% CI) after one and two weeks respectively. The alignment of cells was only slightly less than under the control conditions. 52 Cells were distributed throughout the three dimensional matrix since the collagen gel was present in this condition. The data indicates that cell orientation was affected by the topography of a substratum only when in the immediate vicinity, as the cells present in zones 2 and 3 showed no orientation with the grooves (fig.8 and fig.9). The percent of cells in each increment is roughly the same. The average orientation angles in Zone 2 were 51.8° ± 2.91 (95% CI) and 45.1° ± 2.22 (95% CI) after one and two week respectively, and in Zone 3 the angles were 45.5° ± 3.25 (95% CI) and 49.2° ± 2.92 (95% CI) after one and two weeks respectively, iii) Gel Condition: Surface + (Collagen Gel + Cells) On the surface, the percent of cells with orientation angle less than 10 degrees were 95.4 and 94.2 after 1 and 2 weeks respectively, slightly less than under the Control and Cel lGel conditions. The average orientation angles in Zone 1 were 5.4° ± 1.50 (95% CI) and 2.8° ± 1.27 (95% CI) degrees after one and two weeks respectively. A t the time of gelation cells were evenly distributed throughout the lattice. However, with time, the cells migrated downward toward the titanium surface where they could sense the topography and be influenced by it. Again, cells were distributed throughout the three dimensional matrix since the cells were suspended in the collagen gel at plating. Just as in the previous condition, cell orientation was influenced by the topography only in the cells located in the first zone. Since cells in zones 2 and 3 did not display any preferred orientation (average angles in Zone 2 were 49.6° ± 7.44 (95% CI) and 42.4° ± 3.73 (95% CI) and in Zone 3 46.0° ± 2.93 (95% CI) and 36.6° ± 1.82 (95% CI) degrees after one and two weeks respectively), regardless of the surface type, only orientation of cells in Zone 1 w i l l be discussed below. This substratum (the 30 um depth, 40um pitch and a very narrow ridge (5um)) seemed to maintain a high level of cell alignment in the presence and absence of collagen. To summarize, this 30G40P surface induced high levels of alignment in the surface Zone under all three conditions, with the alignment being greatest under control condition and 53 decreasing only slightly under the Cel lGel and the Gel condition. e) 30G175P: T I T A N I U M W I T H D E E P (30um) G R O O V E S A N D WIDE RIDGE Figure 10(1 week culture) and Figure 11 (2 week culture) show cell orientation by Zones within a three dimensional matrix on the 30 um deep grooves with a 3 pm wide groove and 140um wide ridge under three conditions: the control, the Cel lGel condition, and the Gel condition. A total of 1730 cells after one week and 2523 cells after two weeks were measured for orientation. A non-parametric Kruskal-Wallis test, comparing cell orientation under different conditions and in different Zones was performed (i.e.7 groups: 1. Gel/Oum (361cells and 74 cells after 1 and 2 weeks respectively), 2. Gel/<30pm (172cells and 178cells after 1 and 2 weeks respectively), 3. Gel/>30um (208cells and 425cells after 1 and 2 weeks respectively), 4. CellGel/Oum (96cells and 330cells after 1 and 2 weeks respectively), 5. CellGel/<30um (505cells and 853cells after 1 and 2 weeks respectively), 6. CellGel/>30pm (125cells and 400 cells after 1 and 2 weeks respectively), 7. Control/0um(263cells)). Groups 1, 4, and 7 were compared together one by one to determine whether there was a significant difference between cell orientation in Zone 1 under the three conditions (1:4 P=0.0028, 4:7 P=0.0023, 7:1 P=0.0006). Also , groups 1, 2, and 3 (for Gel condition) as well as 4, 5, and 6 (for Cel lGel condition) were compared to determine whether cell orientation differed in the three Zones under a particular condition. The statistical tests (P< 0.05) indicated that the difference in cell orientation in Zone 1 under the three conditions was significant and the difference in cell orientation in Zone 1 as compared to Zones 2 and 3 was also significant, i) Control Condition: Cells + Surface A l l cells were found on the surface since a collagen gel was not present in this condition. Cells exhibited a very high degree of alignment with the grooves. Ninety seven percent of the cells were found to be aligned within the first increment, that is their orientation angle was less than 10 degrees to the axis of the grooves. The average orientation angle of the cells found on 54 the surface was 4.3° ± 0.52 (95% CI). ii) CellGel Condition: (Surface + Cells) + Collagen Gel Cells that were found in Zone 1 on the surface were affected by the topography. 91.7% after 1 week and 84.8% of cells after 2 weeks were aligned with the axis of the grooves. The remaining cells' orientation angles were found in increments 2, 3, 4, and 5. The average orientation angles of the cells found on the surface were 4.3° ± 0.84 (95% CI) and 6.8° ± 1.34 (95%) CI) degrees after one and two weeks respectively. The overlaying of cells with a gel decreased cell alignment, which was especially evident after two weeks. The alignment on this surface under this condition was decreased by the gel to a greater extent than that observed on the deep grooved titanium surface with a narrow ridge (30G40P) and the shallow grooved titanium surface (3G30P). With time more cells lost their alignment with the grooves (1 week versus 2 weeks). Cells in Zones 2 and 3 exhibited no apparent preference in their orientation. The average orientation angles of cells in Zone 2 were 29.8° ± 2.07 (95% CI) and 37.5° ± 1.53 (95% CI) degrees after one and two weeks respectively, and in Zone 3 the angles were 46.9° ± 3.76 (95% CI) and 46.7° ± 2.24 (95% CI) degrees after one and two weeks respectively. iii) Gel Condition: Surface + (Collagen Gel + Cells) In Zone 1 many cells (59% after 1 week and 73% after 2 weeks) were aligned with the axis of the grooves. However, the number of cells with orientation angle in the first increment was greatly reduced from both the Control and Cel lGel conditions. Not only were less cells aligned but also the remaining cells were oriented at a wide range of angles, reaching as high as the ninth increment. The average orientation angles of cells found on the surface were 13.34° ± 1.68 (95% CI) and 10.1° ± 3.45 (95% CI) degrees after one and two weeks respectively. The model explaining how the cells were affected by the surrounding collagen matrix w i l l be discussed in the discussion section. 55 Cells in Zones 2 and 3 showed no particular favored orientation. The average orientation angles in Zone 2 were 44.3° ± 3.32 (95% CI) and 36.1° ± 3.28 (95% CI) degrees after one and two weeks respectively, and in Zone 3 the angles were 45.4° ± 3.31 (95 % CI) and 40.1° ± 2.14 (95% CI) degrees after one and two weeks respectively. To summarize, this 30G175P surface, as compared to the other grooved surfaces, produced the smallest level of cell alignment. The alignment was further decreased when a collagen gel was placed on top of an established culture. Least alignment in Zone 1 was seen when cells were first suspended within a collagen gel. f) S U M M A R Y O F T H E G R O O V E D SURFACES A schematic representation of cell orientation in gels on grooved substrata is diagrammatically illustrated in figure 12. In Zone 1 most cells were aligned with the direction of the grooves, while cells above the surface in Zones 2 and 3 had no particular orientation. Ce l l orientation and cell density distribution throughout the gel (i.e. in Zone 1, 2, and 3) under both the Cel lGel and the Gel condition (i.e. Zones 1, 2, and 3) on one surface (the 3G30P) is illustrated in F ig . 13 ( C L S M images of Propidium Iodide stained cells at various Zones within a gel). The highest cell density was found in Zone 1 and it decreased as the distance from the surface increases. The cells in Zone 1 were mostly aligned with the direction of the grooves, with a higher proportion of the cells being aligned under the Ce l lGel condition than the Gel condition. Cells in Zones 2 and 3 had no particular orientation. Figure 14 summarizes the effects of both the groove dimensions and the experimental conditions on alignment of cells close to the surface, that is cells in Zone 1. Kruskal-Wallis test was performed to compare (one by one) the effects of the topographies of the three grooved surfaces on cell orientation in Zone 1 under each condition separately, that is the three surfaces were compared under the Control condition, then under the Cel lGel condition, and finally under the Gel condition (Note: the results of statistical testing comparing cell orientation between the 56 surfaces and zones are given on pages 49-55). It is evident, that control condition produced the highest alignment of cells. In the substrata with the same depth the wide ridge was associated with slightly lesser alignment. The highest level of alignment was seen on the 30G40P surface and the lowest on the 30G175P. The alignment decreased as the cells are presented with an alternate substratum such as collagen gel. Relative to the Control condition, the overlaying of a cell culture with a collagen gel decreased cell orientation. Suspending cells within the collagen matrix at plating, as in the Gel condition, was associated with a further decrease in cell alignment. g) 30IP175S: T I T A N I U M S U R F A C E W I T H I N V E R T E D PYRAMIDS Figure 15 shows data on cell orientation by Zones within a three dimensional matrix on a 30um deep pitted surface under three conditions: the control, the Ce l lGel condition, and the Gel condition. The cell orientation was measured with respect to the two sides of the pit, both of which created possible cues involved in directing cell orientation. The two reference axis on the pitted surface restricted the maximum cell angle to 45 degrees and therefore the increment range to 5. A total of 1620 cells were measured for orientation. A non-parametric Kruskal-Wallis test, comparing cell orientation under different conditions and in different Zones was performed (i.e.7 groups: 1. Gel/Oum (163cells), 2. Gel/<30um(54cells), 3. Gel/>30um (273cells), 4. CellGel/Oum (643cells), 5. CellGel/<30um(l51 cells), 6. CellGel/>30um (87cells), 7. Control/Oum (249 cells)). Groups 1, 4, and 7 were compared together one by one to determine whether there was a significant difference between cell orientation in Zone 1 under the three conditions (1:4 P=0.07, 4:7 P=0.46, 7:1 P-0.17). Also , groups 1, 2, and 3 (for Ge l condition) as well as 4, 5, and 6 (for Cel lGel condition) were compared to determine whether cell orientation differed in the three Zones under a particular condition. The statistical tests (P> 0.05) indicated that there was no significant difference between the cell orientation in Zone 1 under the three conditions and that there also was no difference in cell orientation in Zone 1 as compared to Zones 2 and 3. Indeed, 57 no clear preference of cell orientation is observed in figure 15. i) Control Condition: Cells + Surface Due to the lack of the gel in this condition, only data on cells in Zone 1 is presented. Just as observed on T C dish and smooth titanium, cells on this pitted surface showed no preference in their orientation. The average orientation angle was 22.1° ± 12.33 (95% CI) (note: maximum angle was 45 degrees, expected angle i f no orienting effects of substratum were present would be 22 degrees). The cells also formed parallel arrays in different directions. The two cue axes did not seem to affect the orientation of individual cells or the arrays composed of many cells. ii) CellGel Condition: (Surface + Cells) + Collagen Gel Cells were distributed throughout the three dimensional matrix as the collagen gel was placed atop the previously established culture. The cells in Zone 1 formed a number of variably oriented parallel arrays (as in the control condition). The cells average orientation angle in Zone 1 was 22.7° ± 13.13 (95% CI). Cells in Zone 2 and 3 showed no trends in their orientation, their orientation angles were almost evenly distributed between the 5 increments. The average orientation angle in Zone 2 was 23.8° ± 12.69 (95% CI) and in Zone 3 the angle was 23.9° ± 11.78 (95% CI). iii) Gel Condition: Surface + (Collagen Gel + Cells) The data did not differ from the Cel lGel condition. Cells in Zone 1 (average angle of 20.7° ± 13.79 (95% CI)) formed a patchwork of parallel arrays while cells in Zone 2 (average angle of 21.5° ± 12.52 (95% CI)) and in Zone 3 (average angle of 23.6° ± 12.36 (95% CI)) seemed to be randomly distributed. A schematic representation of orientation of cells in gels on T C dishes, smooth titanium, and pitted titanium surfaces is diagrammatically illustrated in figure 5. A s discussed previously, neither of these surfaces induced a guided cell orientation right on the surface in Zone 1 or above the surface in the other two zones: 2 and 3. The cells in Zone 1 formed a patchwork of parallel 58 arrays, probably due to their higher cell density. The cells in Zones 2 and 3 were individually distributed throughout the gel and showed no pattern in their orientation. 2. C E L L DISTRIBUTION B Y ZONES This data w i l l be important in explaining the different end results of gel contraction under the Cel lGel and the Gel conditions (Discussion section). Figure 16 shows data on the cell distribution in the three Zones under the Cel lGel and the Gel conditions. Under the Cel lGel condition most of the cells remain in Zone 1 throughout the duration of the experiments. One hundred percent of the cells were found at the bottom of the gel at the onset of experiments and by the end of the first week up to 65-70% of them were still found in Zone 1. The exception here, is the 30G175P (wide ridged) surface on which only less than 20%) of the cells remained on the surface. Under the Gel condition on the other hand, cells were suspended throughout all three Zones at the beginning of the experiments, and at the end of the first week most of the cells were not located in Zone 1. The percent of the cells in various Zones varied under the Gel condition and no clear pattern of distribution was observed. II. Q U A L I T A T I V E D A T A 1. C O L L A G E N G E L C O N T R A C T I O N O N D I F F E R E N T S U R F A C E S Figs. 17, 18, 19, 20, 21, 22, 23, 24, 25, and 26, show collagen gel contraction on different surfaces after different time of culture. The odd numbered figures represent the Cel lGel condition and the even numbered figures represent the Gel condition. Ge l contraction occurred under both Gel and Cel lGel conditions, but "R ing" formation was observed under the Gel condition whereas "Orthogonal Multilayers" formation was observed under the Cel lGel 59 condition. Under both conditions, however, the gels contracted vertically to about 1/10 of its original size (3mm thick at the onset and about 300um thick at the end). When a " R i n g " was formed the collagen fibers conformed with the ring (Fig.27, page 101). It usually took anywhere from two to three weeks for the initiation of ring formation. The formation of a ring over time is shown in Figure 28 (page 102) which is composed of several frames taken from a Time Lapse Video Microscopy movie. When Orthogonal Multilayers were created under the Cel lGel condition, the collagenous matrix formed orthogonally positioned sheets composed of parallel arrays of collagen fibers (Figure 29, PSR, page 103). a) T C DISH A N D S M O O T H T I T A N I U M S U R A F C E These surfaces w i l l be discussed together as there were no observable differences in their effects on cell behaviour. i) Control Condition: Surface + Collagen Gel Collagen gels without cells embedded in them did not change their appearance over time. The gels did not contract in any dimension and they remained as at plating: fill ing the dish and 3mm thick. Also , P S R staining revealed an apparently random fiber alignment (data not shown). ii) CellGel Condition: (Surface + Cells) + Collagen Gel Fig 17 and 19 represent the process under Cel lGel conditions on smooth titanium substratum and T C dishes respectively. Again, the process involved formation of parallel arrays of cells in the absence of collagen gel (fig 17a, 19a), after plating the gel overtop, cells escaped into the collagen gel and lost their alignment with the array (indicated by arrows: fig 17b, c, d and fig 19b, c, d representing days 1, 3, 7 after gel plating), more cells coming into one array and overall condensation of cells and collagen fibers (fig.l7e and 19e, day 10), resulting in a formation of large parallel arrays composed of cells and collagen fibers on the surface ( f ig l7f and 19f) creating a very thin sheet. 60 Results suggested an orthogonal arrangement of layers of fibers in sheets (PSR). In fig. 17f only one layer of such fiber alignments is seen, while in fig. 19f a faint orthogonal overlap of the two sheets can be seen (arrows indicate the two directions), iii) Gel Condition: Surface + (Collagen Gel + Cells) Fig . 18 and 20 represent the process under Ge l condition on smooth titanium substrata and T C dishes respectively. A t plating (fig 18a and 20a) the cells appeared round as a result of previous trypsinization. Through days 1 and 3 (Figs 18b, c and 20b, c) cells were first randomly distributed throughout the gel as they had been randomly suspended in it at plating. With time, cells formed parallel arrays at the bottom of surface (Fig 18d) and remained randomly distributed in the 3D lattice (Arrows point to cells above the bottom array, their orientation is different from that of the array: figs 18c, 20c,d). A s always under these conditions a " R i n g " was formed (not seen in this picture) by the end of two weeks with a thin sheet in the middle composed of highly ordered cells with collagen fibers that create parallel arrays (arrow indicates the direction of array: F ig 18e, P S R stained collagen fibers). b) 3G30P: T I T A N I U M W I T H S H A L L O W (3um) G R O O V E S i) Control Condition: Surface + Collagen Gel Collagen gels without cells embedded in them did not change their appearance over time. N o gel contraction occurred (data not shown). ii) CellGel Condition: (Surface + Cells) + Collagen Gel Figure 21 represents the process under the Cel lGel condition. Before plating the gel (Fig. 21a) all the cells were aligned with the grooves (also supported by numerical data (Fig.6)). A day after plating the collagen lattice on top of cell culture (Fig21. b) many cells migrated into the gel losing their former orientation (indicated by arrows). F ig . 21c (day 3) again shows the apparently random distribution of cells within the gel, and F ig . 21d (Day7) shows collagen fibers being aligned with the grooved substratum when on the surface (PSR stain; arrow 61 indicates the direction of fiber orientation). iii) Gel Conditions: Surface + (Collagen Gel + Cells) Figure 22 illustrates collagen gel contraction under the Gel condition. A t plating (Fig. 22a) the cells suspended within the 3D collagen were rounded as a result of previous trypsinization. During days 1, 3, 7, the cells migrate through the lattice and some cells were located on the titanium surface (Fig. 22b, c, d) and were aligned with the grooves. The cells above the surface (arrow, Fig . 22c,d) appeared to have a random orientation (also indicated by numerical data in fig.6). On day ten (Fig 22e) gels were very condensed and by day 14 (Fig.22f) a ring has formed (lower left corner) with a thin sheet of collagen and cells in the middle. Note, the orientation of collagen fibers (confirmed by P S R staining) close to the ring follows the direction of the ring, regardless of the orientation of the underlying grooved substrata (arrows), c) 30G175P: T I T A N I U M W I T H D E E P (30um) G R O O V E S A N D WIDE RIDGE i) Control Condition: Surface + Collagen Gel N o gel contraction occurred, (data not shown) ii) CellGel Condition: (Surface + Cells) + Collagen Gel Fig. 23 represents the process under the Cel lGel conditions. Before plating the gel, the stable culture conformed with the topographic cues resulting in most cells being aligned with the grooves (Fig.23a) (supported by data in Fig. 10). A day after plating the gel overtop, cells were capable of moving into the gel and lost their alignment with the grooves (Fig.23b). Progressing through days 3, 7, 10 (Fig .23 c,d,e, respectively) it is evident that the 3D lattice is becoming more and more compacted and condensed both with cells and fibers. Most of the cells on the surface remain aligned but some do stretch across the ridge (indicated by arrows). Cells above the first Zone showed no preferred orientation. Finally, on day 22 (Fig. 23f) the criss-cross pattern of orthogonal multilayers is clearly visible (arrows indicate the two directions of the layers). 62 iii) Gel Condition: Surface + (Collagen Gel + Cells) Fig . 24 illustrates the observations under the Gel condition. A t plating (DayO, F ig . 24a) the cells were rounded as a result of previous trypsinization. On day 1 (Fig 24b) the cells stretched in different directions, some being aligned but many were not. Following the process thorough days 3, 7, and 10 (Fig24c,d,e, respectively) the gels became more and more compact, as seen by the condensation of cells and fibers. Many cells in the bottom layer (Zone 1) were aligned whereas in Zones 2 and 3 (arrows) they were randomly distributed (supported by data in fig. 10). O n day ten (Fig 24e) the "R ing" has been formed and what is seen here is the very thin, sheet of collagen and cells that eventually would be pulled into the "Ring" , d) 30IP175S: T I T A N I U M S U R F A C E W I T H I N V E R T E D PYRAMIDS i) Control Condition: Surface + Collagen Gel N o gel contraction occurred (data not shown). ii) CellGel Condition: (Surface + Cells) + Collagen Gel Fig . 25 represents the process under Cel lGel condition. Before plating the gel (Fig 25a), the cells formed a number of randomly distributed parallel arrays. A day after plating the collagen lattice overtop the cell culture (Fig. 25b), some cells left the earlier established array and entered the collagen gel (indicated by arrows). During the days 3, 7, andlO (Fig 25c,d,e) more cells entered the gel (on pictures at right angles to the parallel array, indicated by arrows), the parallel arrays encompass more cells within them and the overall structure was a lot more condensed. B y day 22 (Fig. 25f) a criss-cross pattern of orthogonal multilayers can be seen (arrows indicate the two directions). iii) Gel Condition: Surface + (Collagen Gel + Cells) Fig . 26 represents the process under the Gel condition. A t plating cells were spherical (Fig 26a). O n day 1 (Fig 26b) cells were randomly distributed on the surface and throughout the 63 collagen gel. During days 3,7, 10 (Fig 26c, d, e respectively) cells and collagen formed parallel arrays on the substratum, above the substratum cells run across the arrays at different angles (indicated by the arrows), and the overall structure of the gel became more condensed. B y day 14 (Fig 26f) a " R i n g " has formed with a thin sheet of very condensed gel in the middle (seen here). 2. O B S E R V A T I O N O N T H E DISTRIBUTION O F C E L L S B E T W E E N C O L L A G E N AND T I T A N I U M A F T E R M E C H A N I C A L STRIPPING A qualitative test of the relative adherence of cells to the collagen gel or the titanium surface after stripping the gels from the substrata was performed. In the Cel lGel condition nearly all cells were removed from a titanium surface when the collagen gel was mechanically stripped from the surface. In the Gel condition, on the other hand, most of the cells that were in Zone 1, remained on the surface after removing the gel (data not shown). Figure 30 provides a diagrammatic representation of the results of the stripping processes which w i l l be discussed in the next chapter. 3. C O L L A G E N G E L P E N E T R A T I O N O N A G R O O V E D S U R F A C E Collagen gels plated on grooved surfaces were stained with Eosin and examined under C L S M to determine their penetration into the substratum. The fluorescence was observed only on the ridges of the grooves but not at the bottom of the grooves (data not shown). The absence of collagen at the bottom of the grooves indicated that the gel behaved as a whole sheet covering only ridges and not spreading into the bottom of the grooves. 64 4. FIBER A L I G N M E N T IN B O X E S W I T H OPPOSED M I C R O M A C H I N E D SURFACES Gel Condition: "Boxes" with opposed micromachined surfaces + (Collagen Gel + Cells) The boxes did not induce a two centre effect. That is, there was no alignment of collagen fibers and cells between the two opposed surfaces surfaces (smooth, grooved, and pitted). The walls of this system seemed to act like the walls of any other dish. Therefore, the only effect seen under this condition was the vertical gel contraction and ring formation (data not shown). 65 Figure 3. Ce l l Orientation on T C dish substrata by Zones (Zonel : Oum, Zone2: <30um, Zone3: >30um) under three conditions: control: Surface + Cells, Cel lGel condition: (Surface + Cells) + Collagen Gel , and Gel condition: Surface + (Cells + Collagen Gel). A 1 week culture. Orientation was measured with respect to a randomly chosen line and grouped into 10 degree increments (P value< 0.05). Cells under all three conditions and the three Zones show no preferred orientation. Figure 4. Ce l l Orientation on smooth titanium surface by Zones (Zonel : Oum, Zone2: <30um, Zone3: >30um) under three conditions: control: Surface + Cells, Cel lGel condition: (Surface + Cells) + Collagen Gel , and Gel condition: Surface + (Cells + Collagen Gel). A 1 week culture. Orientation was measured with respect to a randomly chosen line and grouped into 10 degree increments (P value< 0.05). Cells under all three conditions and the three Zones show no preferred orientation. Figure 5. Diagrammatic representation of cell orientation by Zones within a three dimensional matrix on T C dish, smooth titanium, and pitted surfaces. Cells in Zone 1 form a patchwork of parallel arrays, while cells in Zones 2 and 3 exhibit no preferred orientation throughout the gel. 66 o 00 Figure 5: CELL DISTRIBUTION AND ORIENTATION 69 Figure 6. Ce l l Orientation on 3G30P grooved surface by Zones (Zonel : Oum, Zone2: <30um, Zone3: >30um) under three conditions: control: Surface + Cells, Cel lGel condition: (Surface + Cells) + Collagen Gel , and Gel condition: Surface + (Cells + Collagen Gel). A 1 week culture. Orientation was measured with respect to the axis of the grooves and grouped into 10 degree increments (P value<0.05). Cells in Zone 1 are mostly aligned, with the highest level of alignment being achieved under the Control condition, followed by the Cel lGel condition, and least alignment seen under the Gel condition. Cells in Zones 2 and 3 show no preferred orientation. Figure 7. Cel l Orientation on 3G30P grooved surface by Zones (Zonel : Oum, Zone2: <30um, Zone3: >30um) under three conditions: control: Surface + Cells, Cel lGel condition: (Surface + Cells) + Collagen Gel , and Gel condition: Surface + (Cells + Collagen Gel). A 2 week culture. Orientation was measured with respect to the axis of the grooves and grouped into 10 degree increments (P value<0.05). Cells in Zone 1 are mostly aligned, with the highest level of alignment being achieved under the Control condition, followed by the Cel lGel condition, and least alignment seen under the Gel condition. Cells in Zones 2 and 3 show no preferred orientation. 70 Figure 8. Ce l l Orientation on 30G40P grooved surface by Zones (Zonel : Oum, Zone2: <30um, Zone3: >30um) under three conditions: control: Surface + Cells, Cel lGel condition: (Surface + Cells) + Collagen Gel , and Gel condition: Surface + (Cells + Collagen Gel). A 1 week culture. Orientation was measured with respect to the axis of the grooves and grouped into 10 degree increments (P value<0.05). Cells in Zone 1 are mostly aligned, with the highest level of alignment being achieved under the Control condition, followed by the Cel lGel condition, and least alignment seen under the Gel condition. Cells in Zones 2 and 3 show no preferred orientation. Figure 9. Ce l l Orientation on 30G40P grooved surface by Zones (Zonel : Oum, Zone2: <30um, Zone3: >30um) under three conditions: control: Surface + Cells, Cel lGel condition: (Surface + Cells) + Collagen Gel , and Gel condition: Surface + (Cells + Collagen Gel). A 2 week culture. Orientation was measured with respect to the axis of the grooves and grouped into 10 degree increments (P value <0.05). Cells in Zone 1 are mostly aligned, with the highest level of alignment being achieved under the Control condition, followed by the Cel lGel condition, and least alignment seen under the Gel condition. Cells in Zones 2 and 3 show no preferred orientation. 73 Figure 10. Ce l l Orientation on 30G175P grooved surface by Zones (Zonel : Oum, Zonel: <30um, Zone3: >30um) under three conditions: control: Surface + Cells, Cel lGel condition: (Surface + Cells) + Collagen G e l , and Gel condition: Surface + (Cells + Collagen Gel). A 1 week culture. Orientation was measured with respect to the axis of the grooves and grouped into 10 degree increments (P value<0.05). Cells in Zone 1 are mostly aligned, with the highest level of alignment being achieved under the Control condition, followed by the Cel lGel condition, and least alignment seen under the Gel condition. Cells in Zones 2 and 3 show no preferred orientation. Figure 11. Ce l l Orientation on 30G175P grooved surface by Zones (Zonel : Oum, Zone2: <30um, Zone3: >30um) under three conditions: control: Surface + Cells, Cel lGel condition: (Surface + Cells) + Collagen Gel, and Gel condition: Surface + (Cells + Collagen Gel). A 2 week culture. Orientation was measured with respect to the axis of the grooves and grouped into 10 degree increments (P value<0.05). Cells in Zone 1 are mostly aligned, with the highest level of alignment being achieved under the Control condition, followed by the Cel lGel condition, and least alignment seen under the Gel condition. Cells in Zones 2 and 3 show no preferred orientation. 76 o O C c 00 Figure 12. Diagrammatic representation of cell orientation by Zones within a three dimensional matrix on grooved surfaces. Cells in Zone 1 are mostly aligned with the groves, while cells above the surface in Zones 2 and 3 are not influenced by the topography and therefore exhibit no preferred pattern in their orientation Figure 13. Ce l l orientation and cell density distribution within the three Zones of the gel on a grooved surface (1 week culture; C L S M images of PI stained cells). Zone 1 with the highest cell density and highest level of cell alignment: a) Cel lGel condition: (Surface + Cells) + Collagen Gel, and e) Gel condition: Surface + (Cells + Collagen Gel); Zone 2: 15um b) Cel lGel condition and f) Gel condition and 3 Oum c) Cel lGel condition and g) Gel condition: cell density decreased and cells do not exhibit a preferred orientation; Zone 3: 45um d) Cel lGel condition and h)Gel condition: cell density decreased further with a diverse cell orientation (Note: only one layer of Zone 3 shown here, Zone 3 extends up to about 200um). Figure 14. Proportion of aligned cells in Zone 1- a 1 week culture (i.e. with orientation angle <10 degrees, increment 1) on the three grooved substrata: 30G175P, 30G40P, 3G30P under the three conditions: control: Surface + Cells, Cel lGel condition: (Surface + Cells) + Collagen Gel , and the Gel condition: Surface + (Cells + Collagen Gel). 79 Figure 12: C E L L DISTRIBUTION AND ORIENTATIONON GROOVES _ Figure 13: CELL ORIENTATION AND DISTRIBUTION CellGel G e l OO Figure 15. Ce l l Orientation on 30IP175S pitted surface by Zones (Zonel : Oum, Zone2: <30um, Zone3: >30um) under three conditions: control: Surface + Cells, Cel lGel condition: (Surface + Cells) + Collagen Gel , and Gel condition: Surface + (Cells + Collagen Gel). A 1 week culture. Note, orientation angle was measured with respect to the two axis created by the sides of pits. Therefore, the maximum angle was 45 and so only 5 possible increments (P value< 0.05). Cells under all three conditions and the three Zones show no preferred orientation. Figure 16. Ce l l distribution in Zones l(0um), 2(<30um), and 3(>30um) on the 6 surfaces types under the two conditions, Cel lGel : (Surface + Cells) + Collagen Gel and Gel : Surface + (Cells + Collagen Gel), after 1 week of culture. Under the Cel lGel condition, most of the cells remained close to the surface after a week (ie. the cells were not distributed throughout the entire gel). Under the Gel condition, cells were allowed to migrate throughout the entire span of the gel during the first week. 83 Figure 17. Process of collagen gel contraction: smooth titanium surface under the CellGel condition: (Surface + Cells) + Collagen Gel a) day 0, cells in the absence of collagen gel form a monolayer of parallel arrays, b) day 1, c) day 3, after plating the collagen lattice, cells escape the monolayer and enter the gel losing their orientation (arrows), d) day 7, e) day 10, condensation of cells and fibers into parallel arrays in Zone 1, random orientation of cells elsewhere in the gel, f) day 22, thin sheet of condensed gel formed (one layer seen here). Figure 18. Process of collagen gel contraction: smooth titanium surface under the Gel condition: Surface + (Cells + Collagen Gel) a) day 0, cells rounded up due to trypsinization, b) day 1, c) day 3, cells with no particular orientation within the gel, d) day 10, parallel arrays of cells established in Zone 1 while cells above the surface have no particular orientation (arrows), e) day 14, parallel arrangement of PSR stained collagen fibers along the axis of cellular arrays (direction indicated by arrow) in the thin sheet of the gel (seen here) located in the middle of the ring structure. Figure 19. Process of collagen gel contraction: TC dish surface under the CellGel condition: (Surface + Cells) + Collagen Gel a) day 0, cells in the absence of collagen gel form a monolayer of parallel arrays, b) day 1, c) day 3, after plating the collagen lattice, cells escape the monolayer and enter the gel losing their orientation (arrows), d) day 7, e) day 10, condensation of cells and fibers into parallel arrays in Zone 1, random orientation of cells elsewhere in the gel, f) day 22, orthogonal multilayering of cells and fiber (the two directions of fiber arrangements indicated by arrows). Figure 20. Process of collagen gel contraction: TC dish surface under the Gel condition: Surface + (Cells + Collagen Gel) a) day 0, cells rounded up due to trypsinization, b) day 1, c) day 3, cells randomly distributed throughout the gel, d) day 7, condensation of cells and fibers, e) day 10, ring formation with a thin sheet of tightly packed cells and fibers seen here, Note bottom layer of the sheet is composed of parallel arrays of cells and fibers. 86 Figure 21. Process of collagen gel contraction: 3G30P surface under the CellGel condition: (Surface + Cells) + Collagen Gel a) dayO, cells without the gel are aligned, b) day 1, c) day 3, after plating a collagen gel, cells enter the matrix and lose their orientation (arrows), d) day 7, PSR stained collagen fibers aligned with the grooves (Note, cells at the bottom are still aligned with the grooves, as indicated by the data in fig.6, and therefore act on the fibers to bring them also into alignment). Figure 22. Process of collagen gel contraction: 3G30P surface under the Gel Condition: Surface + (Cells + Collagen Gel) a) dayO, cells round due to previous trypsinization, b) dayl, c) day3, d) day7, cells at the bottom (i.e. Zone 1) are aligned while cells within the lattice are not (arrows; refer to numerical data in fig.6), e) day 10, condensation of cells and fibers proceeds, f) collagen ring has formed (arrows; lower left corner) with the thin sheet of cells and fibers in the middle). 91 92 Figure 23. Process of collagen gel contraction: 30G175P surface under the CellGel condition: (Surface + Cell) + Collagen Gel a) day 0, cells without collagen are aligned, b) dayl, collagen plated, some cells escape into the matrix and lose their orientation (arrows), c) day3, d) day7, e) day 10, cells at the bottom aligned while cells which have entered the collagen lattice show no pattern of orientation (arrows)(confirmed by data in fig. 10), f) tightly packed contracted gel composed of cells and fibers arranged in orthogonal arrays (the two directions indicated by arrows). Figure 24. Process of collagen gel contraction: 30G175P surface under the Gel condition: Surface + (Cells + Collagen Gel) a) dayO, cells rounded due to trypsinization, b) day 1, c) day3 cells within the gel are not guided to obtain any particular orientation, d) day 7, cells at the bottom are aligned with the grooves, while cells above the surface exhibit a random distribution (arrows, confirmed by numerical data in fig. 10), e) day 10, ring has formed with a tightly condensed thin sheet (seen here)of cells and fibers in the middle. Note, cells at the bottom aligned and cells within the matrix not. 94 Figure 25. Process of collagen gel contraction: 30IP175S surface under the CellGel condition: (Surface + Cells) + Collagen Gel a) day 0, cells in the absence of collagen lattice for a monolayer of parallel arrays, b) day 1, c) day3, after plating the gel some cells escape the arrays and enter the collagen matrix (arrows), d) day 7, e) day 10, condensation of cells and fibers into tightly packed parallel arrays, cells above Zone 1 show no organization at this stage, f) day 22, thin sheet of orthogonal arrays (the two directions indicated by arrows). Figure 26. Process of collagen gel contraction: 30IP175S surface under the Gel condition: Surface + (Cells + Collagen Gel) a) day 0, cells rounded up due to trypsinization, b) day 1, cells show no pattern of their distribution and orientation, c) day 3, d) day 7, cells form parallel arrays at the bottom (Zone 1) and show no particular orientation higher up in the gel (arrows), e) day 10, condensation of the matrix, f) day 14, ring has formed with a thin sheet of compacted cells and fibers (seen here). 97 i *4l • Figure 27. Ring Structure "Ring" formation as the end result of collagen gel contraction under the Gel condition: Surface + (Cells + Collagen Gel). Note, cells and fibers oriented with the ring (PSR staining). Fig27.a: Ring formed on a grooved titanium surface. Fig. 27B: Ring formed on a smooth surface. Figure 28. Ring formation: frames isolated from a time lapse movie Ring formation as a function of time: a, b, c, d: 21 day, e: 22 day, f: 23day. Note cells and fibers oriented with the structure of the ring. Figure 29. Orthogonal Sheet Orthogonal arrangement of cells and collagen fibers as the end result of collagen gel contraction under the CellGel condition: (Surface + Cells) + Collagen Gel. a) lower parallel array of the collagen fibers, b) upper, orthogonally positioned array of fibers (Fiber orientation revealed by PSR staining). 100 Figure 28: RING FORMATION MOVIE FRAMES Figure 30. Distribution of cells between collagen and titanium, a) CellGel condition: cells exert forces on collagen only from one position, collagen gel is only being pulled down towards the cells. Cells make protrusions into the matrix and therefore come off the surface with the gel when the gel is mechanically stripped, b) Gel condition: cells suspended within the 3D lattice exert forces in different directions (depending on cell orientation) and therefore they not only pull the fibers downwards but also upward and sideways (creating a ring) . Cells that have migrated to the surface and landed on it may lose their contact with collagen as the collagen gel continues to be contracted upward and is being detached from the surface. The cells that landed on the surface, remain on the surface when the gel is being stripped off. 104 Figure 30: C E L L DISTRIBUTION BETWEEN TITANIUM AND COLLAGEN (GEL CONTRACTION) CellGel CONDITION B E F O R E A F T E R Gel Cells Surface Gel CONDITION B E F O R E A F T E R Gel and Cells Cells Surface / / / / / / 105 S U M M A R Y O F R E S U L T S I. Cell Orientation 1. T C dish, smooth titanium and pitted titanium surfaces did not induce an alignment of cells in any favored direction. Cells did form a patchwork of parallel arrays. 2. Grooved surfaces induced cell alignment with the axis of the groves. The highest alignment was achieved on the 30G40P surface, followed by the shallow 3G30P surface, and least alignment was found on the wide ridge 30G175P surface. II. Effects of Collagen Matrix 1. The presence of collagen matrix atop a stable cell culture (Cel lGel Condition) disrupts the orientation of cells leading to a lesser alignment of cells with the grooves. 2. The presence of collagen matrix around cells (Gel Condition), leads to less cell alignment with the grooves from that when the collagen is present on top of the culture. III. Effects of Cell Distribution on Collagen Gel Contraction 1. Cells are required for gel contraction to occur. 2. When most cells are found at the bottom of the gel from the beginning to the end of the culture time (CellGel Condition), the gel acts as an attached gel and the contraction proceeds in two dimensions (only vertical contraction). A very thin sheet of orthogonally arranged fibers is formed. Fibers conform with the orientation of cells on a surface and become aligned with the grooved topography. 3. When cells are suspended within the gel (Gel Condition), the gel becomes a floating gel and contraction proceeds in three dimensions (vertical and horizontal-radial contraction). A ring of collagen fibers is formed by the end of 2 weeks. Qualitative Distribution of Cells Between Collagen and Titanium 1. Under Cel lGel condition, cells apparently form attachments with the collagen as they pull it down towards themselves. The attachments between cells and the matrix are maintained even after the gel is mechanically stripped off the titanium surface. It appears that cells leave the titanium surface and enter the collagen matrix. 2. Under the Gel condition, the cells that are located on the titanium surface apparently lose their attachments with the collagen lattice, as the gel is being lifted up by the traction forces. When a mechanical strip is applied only those cells are left behind on the titanium surface. It appears that some cells are also able to leave the collagen matrix and settle on the titanium surface. Collagen Gel penetration into a Grooved Surface 1. Collagen gel did not penetrate into the grooves, it behaved as a sheet covering the ridges. 107 C H A P T E R IV DISCUSSION A wound is created upon an insertion of a dental implant into a tissue creating a lesion in epithelium, basement membrane, and adjacent soft and hard connective tissue. In general, the connective tissue reaction to "injury" may be adequately characterized as inflammatory-reparative response (Gay and Miller, 1978). Soft tissue repair is a complex process involving the integrated action of cytokines, migrating inflammatory cells, and resident fibroblasts. The inflammatory response encompasses the first steps in the healing process in the forms of clot formation, the appearance of necrotic-tissue- resorptive phagocytes, and the increase in the vascularization of the granulation tissue (Walter et al, 1976). The reparative component of the response follows and it includes the migration and proliferation of connective tissue fibroblasts. Fibroblasts play a central role in the process of wound healing and implant integration as they deposit the initial fine network of fibrils of type III collagen. The continued proliferation of fibroblasts causes the subsequent accumulation of Type I collagen fibers. Fibroblasts are not only involved in the production of extracellular matrix but also in its remodeling. The dynamic cell-matrix interactions between fibroblasts and collagen lead to the reorganization of fibers and to the final contraction of connective tissue (Stephens et. al., 1996). The success of implanted devices has been considered previously mainly from the perspective of interactions between cells and the implant surface. This study examined how fibroblast populated E C M (CellGel and Gel Conditions) regulates fibroblasts interaction (attachment and orientation) on synthetic substrata. 108 For the ease of understanding the results, they w i l l be discussed in a similar order as they were presented in the results section, with slightly different grouping to avoid repetition. First, cell orientation on different surfaces in Zone 1 wi l l be discussed and how it was influenced by the presence of a surrounding collagen matrix (whether under the Ce l lGel or the Gel condition). It was found that cells on T C dish, smooth, and pitted titanium form a patchwork of parallel arrays while cells on grooved titanium aligned with the direction of the grooves. Cells in the absence of collagen achieved the highest level of alignment, followed by the Cel lGel condition, with least alignment observed under the Gel condition. Next, cell orientation within a collagenous lattice in Zones 2 and 3 wi l l be discussed. Cells not present on surface showed no preferred orientation. The explanation of the different end results of gel contraction depending on the condition wi l l follow. In the absence of cells no contraction occurred, under the Ce l lGel condition thin sheets of orthogonally oriented fibers were formed, and under the Gel condition a ring was created. A t the end, the distribution of cells between collagen and titanium under the Cel lGel (cells left the titanium and entered the collagen matrix) and the Gel conditions (some cells left the collagen gel and settled on the titanium surface) w i l l be explained by a proposed model. A t low population densities Human Gingival Fibroblasts on smooth titanium substrata have been found to exhibit a behaviour best described as a random walk and therefore have no pattern in their orientation and polarization (Damji et. al., 1996). In dense confluent fibroblast cultures on T C plastic the formation of patchwork of parallel arrays (Elsdale and Bard, 1972) and orthogonal multilayers has been found (Elsdale and Foley, 1969) (refer to figure 5). In the present study similar results have been found for the T C dish, smooth titanium and pitted titanium surfaces. In Zone one on these surfaces a patchwork of parallel arrays has been observed (fig 17-20 and 25-25). The collagen in the Cel lGel condition provided the cells with an 109 opportunity to extend their processes into the overlaying matrix and therefore to leave the arrays and enter the matrix. The collagen in the lattice did not disrupt the parallel arrays. It rather permitted the cells to "build" another layer of cells on top of the first layer. A n orthogonal arrangement of fibers has been observed (e.g. fig.l9f, 25f). In the Gel condition, the cells that settled at the bottom of the gel (in Zone 1) also created parallel arrays. The positive influence of collagenous gel on the creation of orthogonal multilayers o f cells and fibers may be expected as Elsdale and Foley (1969) have found that collagen is required for multilayering to occur. A t the same time, the cells that create parallel arrays can be expected to arrange nearby collagen fibers in the same orientation (fig. 17f, 19f, 25f, 26f show cells and fibers in the same orientation), as they exert traction forces that pull collagen fibers into alignment with the fibroblasts (Harris et. al., 1981). It has been previously found that micromachined grooved substrata w i l l induce the alignment of fibroblasts in cell culture (e.g. Oakley and Brunette, 1995). This study has found that a grooved surface also aligned fibroblast cells that are in contact with a collagen matrix. However, the presence of collagen as an alternate substratum to the titanium decreased the total alignment of cells from that of cells in the absence of collagen. The decrease in alignment seemed to depend on the dimensions of the grooves as well as the sequence at which the cell were in contact with the collagen gel (being either Cel lGel or Gel) (refer to figure 14). Under the Control condition, in the absence of collagen, all three types of grooved surfaces (3G30P, 30G40P, 30G175P) induced a high level of alignment. Previous studies have found that cell alignment decreased with decreased groove depth, increased pitch size, and increased ridge width (Clark et. al., 1990; Brunette, 1986a). Indeed, in this study the wide ridge of the 30G175P grooves decreased the cell alignment slightly. However, the depth of the grooves (3um versus 30um) did not seem to affect the alignment. Since Clark et. al. (1990) used grooved surfaces which depth varied between 0.2-1.9um, with the 1.9 um eliciting higher alignment, it could be 110 assumed that any grooves deeper than 1.9um wi l l induce a high level of alignment, which would be consistent with the findings in this thesis (as 3um and 3Oum deep grooves were used). Under the Cel lGel condition the alignment on all three types of grooved substrata decreased to some extent. Cells were first allowed to be influenced by the topography (as in Control condition) which induced a high level of alignments. It was only after the cells exhibited topographic guidance that the collagen gel was placed atop. Cooper et. al. (1993) have found cell binding to titanium was lower than to collagen gels. Similarly, Lowenberg et. al. (1987) have found that cell attachment to the exposed collagen of a root slice is significantly higher than to a titanium disc. Both of these studies used differently treated titanium surfaces, and yet both reported the cell preference for collagen over titanium. Therefore, it is possible that the haptotactic natural substratum in the form of collagen gel plated atop the cell culture attracted the cells and provided opportunity for the cells to make new attachments with or protrusions into the matrix, thereby altering their orientation. Tarpila et. al.,1998 and Lorimier et. al., 1998 have shown the involvement and the importance of cellular processes in collagen lattice contraction. On the shallow grooves (3G30P) cells were readily exposed to the overlayed collagen as the grooves were not deep enough to provide a shelter for the cells. More cells on this surface lost their orientation as compared with the cells on the deep grooved surfaces. In turn, many cells on the 30G40P and 30G175P surfaces maintained their alignment as they could have been protected from the haptotactic effects of collagen in the deep grooved environment, where it appeared the collagen gel did not penetrate. Slightly more cells, however, on the wide ridged surface (30G175P) lost their alignment as compared to the cells on the narrow grooves surface (30G40P) probably due to the fact that those cells were present on the wide ridge (140um) and therefore exposed to the effects of the overlaying collagen. The level of cell alignment with the direction of the grooves was even less when the cells were suspended within the gel. Under the Gel condition less cells seemed to be aligned on all i l l three grooved surface types as compared to both the Control and Cel lGel conditions. The lesser degree of alignment could be due to the following explanation. The cells' first substratum was the collagen gel and their primary contacts were formed with the attractive collagen matrix as they were suspended in it at plating. These cells were able to orient themselves and make extensions in any direction within the three dimensional matrix by which they were surrounded. It was only after the cells migrated to the bottom of the gel that they could be secondarily influenced by the topography of the titanium surface, yet still being affected by the collagen. The collagen continuously provided an alternate matrix for cell contacts and protrusions resulting in less cell alignment than would be expected in the absence of the collagen. The shallow grooves did induce alignment. Many cells, however, were not well aligned with the grooves (as defined by Clark, 1990). It is possible that those cells were still able to retain their protrusions within the collagenous matrix and therefore stretch across the grooves (figure 13e), as the 3um deep groove did not provide a shelter or present an adequate obstacle in the presence of the haptotactic collagen atop, with which the cells could maintain contacts. The deep grooved surface with a narrow ridge (30G40P) induced a very high level of alignment. This can be due to the fact that the 3 Oum depth of a groove served as a sufficient obstacle to stretching across the groove, instead the cells migrated into the groove where they were protected from the effects of the collagen and were influenced by the topography and became aligned with the direction of the groove. The deep grooved surface with a wide ridge (30G175P), however, did not induce a very high alignment. A s previously, cells were first suspended within the collagen and only after they migrated to the bottom of the gel could they be affected by the topography. Because the ridge was very wide (140um) it is possible that most cells settled there (figure 24b), where their orientation was not restricted by direction of the groove, and where the cells maintained their protrusions within the matrix, which is reflected by the lesser alignment. The cells that did find 112 their way to the bottom of the deep groove would become aligned with the direction of the grooves. Even though the experimental conditions in this study differed in some aspect from those of other studies, the pattern of lesser alignment due to a decreased depth of groove or increased pitch size and ridge width in this study is in accordance with earlier findings (Brunette, 1986b; Dunn and Brown, 1986). The cells that were not in contact with the titanium surface and were found in Zones two and three were not influenced by the topography of the substratum since no features were present that could alter the probability of a cell making a protrusion in any given direction (Brunette, 1988) (refer to figures 5 and 12). Noble (1987) has found that cells locomoting within a three dimensional matrix show approximate random movement. In this study, cells within the 3D lattice had no particular orientational pattern probably because of their random movements and because Of the fact that the fibers in the gels did not have a particular orientation at the onset of the experiments therefore no guidance was provided to the cells. In this study it has been shown that cells are not only the crucial elements of a collagen gel contraction but also their distribution within the gel has an effect on the final form of a contracted gel. Collagen gels without cells did not undergo any changes over time. In these gels no contraction occurred. The same results were observed by Harris et. al. (1981). The gel contraction has been found to be controlled by cell-matrix and cell-cell interactions as well as by cell migration which is related to the effectiveness of the contraction (Andujar et. a l , 1992). Therefore, in the absence of such interactions, the gels remain static. 113 The contraction under the Cel lGel condition resulted in a formation of a very thin sheet of cells and fibers. Under the Gel condition, on the other hand, the contraction resulted in a formation of a strand of cells and fibers that created a "ring". These results confirm previously reported observations on the attached and floating gels in relation to the substratum on which they were plated. Previously it has been determined that contraction of gels which were not attached to a substratum (in other words: floating) resulted in a decreased thickness and decreased diameter (Bell et. al., 1979) while the contraction of gels attached to substratum resulted only in decreased thickness (Guidry and Grinnell, 1985). Results in this thesis wi l l be explained by a proposed model discussed below and illustrated in figure 30. In the Cel lGel condition cells first formed contacts with the titanium surface and later made contacts with or protrusions into the collagen gel placed atop. I f it was the protrusions that were being made into the gel from titanium, they would not be parallel to the surface, rather they would be at a slight angle. Lorimier et. al. (1998) have observed that re-orientation of collagen fibers was evidenced only at the proximity of the pseudopodia, which in our case would be entering the gel at an angle (fig. 30). The forces exerted by cells, then, also are not parallel to the surface, and therefore can pull the collagen fibers down towards the cells. The forces exerted from the bottom part of the gel are the only forces present, as most cells are located there, and so the collagen gel is brought into tighter contact with the cells and therefore the titanium. It seems that cells maintain their contacts with our titanium surfaces (which have been R F G D treated to increase cell binding) and cell attachments to both the collagen and titanium reached an equilibrium, cells being sandwiched in between the two substrata. A s a result these gels become "attached" to the titanium through the cell layer and their contraction results only in the decreased thickness forming sheets. In this study it has been found that fibers in these sheets were arranged in parallel arrays and in orthogonal multilayers, which can be explained by the fact that fiber orientation in attached gels is guided by the plane of cell spreading (Nakagawa et. 114 al., 1989), cells also having such orthogonal arrangements (figure 29: two orthogonal layers of cells and fibers oriented together). In this thesis, the data on cell orientation in different Zones does not indicate an orthogonal multilayering of cells. This difference between numerical data on cell orientation and observed data on fiber arrangements may be explained by the fact the quantitative data was collected after one and two weeks, the cells not being in their final orientation yet. While the qualitative observations of orthogonal arrangements of fibers were made after a longer period of time: 3-4 weeks (The qualitative data on cell orientation was collected from experiments which lasted only two weeks so that the data from Cel lGel conditions and Gel conditions were comparable, two weeks being the time required for the formation of the "ring" structure under the Gel condition). Even though the gels act as attached, it should be noted that when a mechanical strip was administered, the cells lost their contacts with the titanium and were taken off the surface together with the collagen. The stripping of the cells together with the gel off the titanium surface could be explained by the cells making more contacts with the matrix or by making a number of protrusions into the collagen gel and therefore being strongly adhered to it. Be l l et. al. (1979) have found that a decreased gel thickness and diameter occurred in floating gels. The "ring" formation under the Gel condition can be possibly explained by the following model. In the Gel condition the cells are distributed throughout the gel at the onset of the experiments. Before some of the cells settle on a surface, they not only can make many protrusions within the three dimensional matrix (protrusions being required for matrix contraction: Tarpila et. al., 1998) but they are also fairly motile as they are not restricted by a high density of neighbours within the gel (motile cells required for matrix contraction: Harris and Stopak, 1981). A s the cells change their position and the position of their fillopodia, the traction forces exerted by those cells also are located in different parts of the gel, thereby affecting a number of collagen fibers. The cells exert traction forces in different directions as 115 their orientations are not uniform. The fibers are not only being pulled down but also up, rendering the gel to float (refer to figure 3 0 and next paragraph regarding cell distribution between titanium and collagen under the Cel lGel and Gel conditions). A s these gels were not attached anymore to the substratum on which they had been plated, their contraction proceeds in a radial fashion creating a ring. A s mentioned previously, ring formation was dependent on the cell-matrix and cell-cell interactions and on the motility of the cells, all of which are present under the Gel conditions. When the qualitative distribution of cells between collagen and titanium was considered, it has been found that in Zone 1 the maintenance of cell attachments to collagen (as opposed to the attachments to titanium) depended on the direction and the strength of the total forces that are pulling the fibers (refer to figure 3 0 ) . A n d so, in the Cel lGel condition, since most of the cells are found at the bottom of the gel throughout the duration of the experiments (figure 16), all the forces are exerted from that locations and pull the collagen down toward the cells resulting in adhesion between cells and collagen. When a mechanical strip of the gel is applied, the cells attached to it also come off the titanium surface. In the Gel condition, since cells are distributed throughout the collagen matrix (figure 16), the forces are exerted in many directions and pull the collagen fibers sideways, as well as down and up. A number of cells do settle on the surface. However, there are many more cells found within the gel than on the surface, and so the forces exerted by the cells in the gel pulling it upwards wi l l be of greater magnitude than the total forces exerted by the cells on the surface. Therefore, the gel w i l l be slightly lifted off the surface (becomes a floating gel) leaving a small space between the collagen lattice and the substratum. A s a result, cells on the titanium surface lose their attachments with the collagenous matrix. It should be noted, that such a gap between the gel and the titanium was not noticed visually. However, since the cells did lose their attachments with the overlaying matrix, it can be 116 postulated that the size of the gap was very small, only big enough to allow loss of contact between the cells and the collagen. The titanium surfaces in this study have been treated by the R F G D prior to plating cells and gels. Therefore, these surfaces had a high surface energy inducing a high level of cell attachments. These attachments seemed to be maintained (CellGel) even in the presence of collagen matrix. However, the collagen did act as a haptotactic agent since some cells did lose their earlier achieved orientation. Also, some of the cells which settled on titanium after migrating through the collagenous matrix (Gel) created new strong attachments with the titanium. The conclusion that can be made here is that cells w i l l attach strongly to a R F G D surface with high surface energy but w i l l also adhere to a haptotactic agent such as collagen. How do these findings, than, relate to possible approaches to improving implant performance? First of all, it has been found that grooved surfaces elicit an almost uniform orientation of cells even in the presence of collagen. The surface that seemed to be least affected by the "collagen effect" (i.e. decreased cell alignment due to cells extending their protrusions into collagen) was the 30G40P. Cells were located in the deep grooves (30um) as the collagen did not penetrate into the grooves. The aligned cell orientation was retained. Although the narrow ridge (5 pm) did not provide large a surface area for the cells to orient, cells were still able to form attachments with the collagen at this location (CellGel condition). Secondly, it seemed that the cell distribution throughout the gel may also affect the success of an implanted surface. The Cel lGel condition produced uniform sheets of "tissue" covering all of the surface. This is similar to the in vivo process of wound healing where cells migrate first to the site of injury and only later the Extracellular Matrix is produced and organized into a new connective tissue (Walter and Path, 1976). These results may suggest that a technique of placing 117 an implant in vivo could involve a prior seeding of the implant device with cells or cells overlayed by a collagen gel, which could lead to a better integration of the implant within the host tissue. Understanding how cells react to a topography of an implant device in the presence of the components of extracellular matrix w i l l lead to a better understanding of the long term success of an implant. It w i l l also aid in designing surfaces with precisely defined topographies which can be used on the implant devices that w i l l then be properly accepted by the host without encapsulation or rejection. 118 F U T U R E W O R K This study presented data on cell orientation by Zone within a three dimensional collagen matrix on different topographies. However, as this study involved fairly thick gels (3mm) with a relatively high collagen concentration (1.95mg/ml) it was not possible to observe the orientation of individual fibers but rather sheets of collagen were observed. A future study may involve very thin gels with a lower collagen concentration so the effect of cells on orienting the individual collagen fibers on different topographies can be studied. This study showed that cells found in Zone 1 right on the surface w i l l be stripped off that surface together with the collagen gel when under the Cel lGel condition and w i l l remain on the surface when under the Gel condition. A model explaining this phenomenon was presented. It would be interesting to test this hypothesis by visualizing cell contacts with the collagen gel and with the titanium surfaces. A specific staining should be utilized to see whether cells in Zone 1 under the Cel lGel conditions do indeed have a high number of contacts with the overlaying gel and whether the cells in Zone 1 under the Gel condition lack those attachments after a two week period. A s the interactions between cells surface receptors and components of the Extracellular Matrix are mediated through integrins, it may be appropriate to stain the integrins to determine the cell contacts with the collagen lattice. There are a number of different integrins, however the ones that are of interest to us would be the Beta-1 integrins which are largely involved in interactions between cell surface and extracellular matrix molecules such as collagen, laminin, and fibronectin To be more specific the cc ip i and oc2Bl integrins are the ones that interact with collagen. Beta-1 integrins can be visualized with indirect immunofluorescence by using primary antibody directed against an integrin subunit and a secondary antibody conjugated with FITC or Rhodamine (Reszka and Horowitz, 1992). 119 The two different end results of gel contraction: orthogonal arrays o f fibers and the ring under the Cel lGel and the Gel condition respectively were explained by a model based on the direction and strength of force exerted by cells on the matrix producing either an attached or floating gel. Using a Culture Force Monitor ( C F M measures the contractile forces generated within a gel; Eastwood et. al., 1996) one could test the total strength and the overall direction of the forces exerted by cells on the matrix under the two conditions, thereby leading to a more viable explanation of the process of gel contraction. Orthogonal multilayers of collagen fibers were found, however, the data on cell orientation did not indicate such arrangements. This can be explained by the fact that the data on cell orientation was collected on experiments lasting up to two weeks while the qualitative observations on gel contraction were done on experiments lasting three weeks and longer. It would be interesting to see whether cells can also achieve such orthogonally arranged multilayers, by staining the cells with Propidium Iodide after 3-4 week culture and examining their orientation within a gel under C L S M at that time. This study was mainly concerned with results of experiments on collagen gels plated on horizontally positioned micromachined surfaces. The next step ought to involve studies on collagen gels plated around vertically positioned surfaces. It would be interesting to see whether cells within a gel could arrange collagen fibers in such a way that the collagen fibers would become inserted into the vertical micromachined surface, mimicking the structure of periodontal ligament. In this study, "boxes" with opposed microfabricated surfaces were used under the Gel condition. However, the only restructuring of the gel that occurred involved the formation of a "ring". The future studies could use such "boxes" with additional changes to the design. The opposed sides could be brought closer together (i.e. less than 1cm away). A Cel lGel condition could be used to mimic the previously studied fiber alignment between two explants. Cells could be first plated on the vertical sides and allowed time to be affected by the topography. Only then 120 a collagen gel by itself or with cells suspended within it could be plated in between the two surfaces. B y understanding the cascade of biological events and a number of interactions of cells with an artificial surface, other cells, and extracellular matrix we can improve the process of wound healing after an insertion of an implantable device, thereby hopefully increasing the success rates of implants integrating with the connective tissue. 121 B I B L I O G R A P H Y Abercrombie M . , The crawling movement of metazoan cells, Proc. Roy. Soc. London 207(SeriesB):129-147, 1980. Abercrombie M . and Heaysman J . E . M . , Observations on the social behaviour of cells in tissue culture: I. speed of movement of chick heart fibroblasts in relation to their mutual contacts, Exp. Cel l Res., 5:111-131, 1953. Abercrombie M . and Heaysman J . E . M . , Observations on the social behaviour of cells in tissue culture: II. Monolayering of fibroblasts, Exp. Cel l Res., 6:293-306, 1954. Abercrombie M . and Heaysman J . E . M . , and Karthauser H . M . , Social behaviour of cells in tissue culture: III. Mutual influence of sarcoma cells and fibroblasts, Exp. Ce l l Res., 13:276-291, 1957. A l l en T .D. and Schor S.L., The contraction of collagen matrices by dermal fibroblasts, J. Ultrastruct. Res., 83(2):205-219, M a y 1983. Andujar M . B . , M e l i n M . , Guerret S., and Grimaud J .A. , Ce l l migration influences collagen gel contraction, J. Submicrosc. Cytol . Pathol., 24(2): 145-154, A p r i l 1992. Auger F . A . , Rouabhia M . , Goulet F. , Berthod F., Mou l in V . , Germain L . , Tissue-engineered human skin substitutes developed from collagen populated hydrated gels: clinical and fundamental applications, Med. B i o l . Eng. Comput , 36(6):801-812, 1998. Baes C.F. and Mesmer R . E . , The hydrolysis of cations, John Wiley, N e w York, N . Y . , pp. 147-52, 1976. Baier R .E . , Surface preparation, J. Oral Implant., 12:389-395,1986. Baier R .E . , Surface properties influencing biological adhesion, In Adhesion in biological systems (ed. Manly R.S.), N e w York: Academic Press, pp. 15, 1970. Baier R . E . and Meyer A . E . , Implant surface preparation, Int. J. Oral Maxillofac. Implants, 3:9-20, 1988. Baier R .E . , Surface proteins determine bioadhesive outcomes: methods and results, J. Biomed. Mater. Res., 18:337-355, 1984. Bard J. and Elsdale T., The three-dimensional culturing of fibroblasts in collagen, In Biology of Fibroblasts (ed. Kulonen E . and Pikkarainen J.), Acad. Press Inc. (London) L T D , 1973. Be l l E . , Ivarsson B . , and Merr i l l C , Production of tissue-like structure by contraction of collagen lattice by human fibroblasts of different proliferature potentials in vitro, Proc. Natl . Acad. Sci. U S A , 76:1274-1278, 1979. Bellows C . G . , Melcher A . H . , and Aubin J.E., Association between tension and orientation of periodontal ligament fibroblasts and exogenous collagen fibers in collagen gels in vitro, J. Ce l l Sci. , 58:125-138, December 1982. Bellows C . G . , Melcher A . H . , and Aubin J.E., Contraction and organization of collagen gels by cells cultured from periodontal ligament, gingiva and bone suggest functional differences between cell types, J. Cel l Sci . , 50:299-314, 1981. Bennett V . and Gil l igan D . M . , Spectrin-based membrane skeleton and micro-scale organization of plasma membrane, Annu. Rev. Ce l l B i o l . , 92:27-66, 1993. Ben-Ze'ev A . , Differential control of cytokeratins and vimentin synthesis by cell-cell contact and cell spreading in cultured epithelial cells, J. Ce l l B i o l . , 99:1424-1433, 1984. Berthiaume F., Moghe P .V . , Toner M . , and Yarmush M . L . , Effect of extracellular matrix topology on cell structure, function, and physiological responsiveness: hepatocytes cultured in a sandwich configuration, The F A S E B J., 10:1471-1484, 1996. Berry D.P. , Harding K . G . , Stanton M . R . , Jasani B . , and Ehrlich H.P. , Human wound contraction: collagen organization, fibroblasts, and myofibroblasts, Plast. Reconstr. Surg., 102(1):124-131,1998. Bothe R.T. , Beaton L . E . , Davenport H . A . , Reaction of bone to multiple metallic implants, Surg. Gynecol. Obstet, 71:598-602, 1940. Boyko G . A . , Brunette D . M . , and Melcher A . H . , Ce l l attachment to demineralized root surfaces in vitro, J. Periodont. Res., 15:297-303, 1980. Branemark P .L , Breine U . , Lindstrom J., Ade l l R , Hansson B . O . and Ohlsson A , Intraosseous anchorage of dental prostheses. I Experimental studies, Scand. J. Plast. Reconstr. Surg., 3:81, 1969. Bretscher M . S . , Moving membrane up to the front of migrating cells: minireview, Ce l l , 85:465-467, M a y 1996. Brunette D . M . , Kenner G.S. , Gould T .R .L . , Grooved titanium surfaces orient growth and migration of cells from human gingival explants, J. Dent. Res., 62:1045-1048, 1983. Brunette D . M . , Fibroblasts on micromachined substrata orient hierarchically to grooves of different dimensions, Exp. Cel l Res., 164:11-26, 1986a. Brunette D . M . , Spreading and orientation of epithelial cells on grooved substrata, Exp. Cel l Res., 167:203-217, 1986b. Brunette D . M . , The effects of implant surface topography on the behaviour of cells, Inter. J. Oral Maxillofacial Imp., 3:231-246, 1988. Camporese D . L . , Lester T.P., and Pulfrey D . L . , A fine line silicon shadow mask for inversion layer solar cells, I E E E Electron Device Letters, E D L - 2 (3);61-63, 1981. Chandrakasan G . , Torchia D. , and Piez K . A . , Preparation of intact monomeric collagen from rat tail tendons and skin and the structure of the nonhelical ends in solution, J. B i o l . Chem., 251(19):6061-6067,1976. Chehroudi B . , The effects of Surface topography on the behaviour of cells attached to percutaneous implants, PhD Thesis, U B C , Vancouver, B . C . , Canada, 1991. 124 Chevallay B . , Abdul-Malak N . , and Herbage D. , Mouse fibroblasts in long-term culture within collagen three-dimensional scaffolds: influence of crosslinking with diphenylphosphorylazide on matrix reorganization, growth, and biosynthetic and proteolytic activities, J. Biomed. Mater. Res., 49:448-459, 2000. Clark P., Connolly P., Curtis A . S . G . , Dow A . T . , and Wilkinson C . D . W . , Topographical control of cell behaviour: I simple step cues, Develop. 99:439-448, 1987. Clark P., Connolly P., Curtis A . S . G . , D o w A . T . , and Wilkinson C . D . W . , Topographical control of cell behaviour: II multiple grooved substrata, Develop. 108:635-644, 1990. Cooper L . F . , Handelman B . , McCormack S . M . , and Guckes A . D . , Binding of murine osteoblastic cells to titanium disks and collagen I gels: implications for alternative interpretations of osseointegration, Int. J. Oral Maxillofac. Implants, 8:264-272, 1993. Couchman J.R. and Rees D . A . , The behaviour of fibroblasts migrating from chick heart explants: changes in adhesion, locomotion and growth, and in the distribution of actomyosin and fibronectin, J. Cel l Sci. , 39:149-165, 1979. Curtis A . S . G . and Clark P., The effects of topographic and mechanical properties of materials on cell behaviour, Crit. Rev. Biocompat, 5(4):344-362, 1990. Curtis A . S . G . and Forrester J .V. , The competitive effects of serum proteins on cell adhesion, J. Cel l . Sci. , 71:17-35, 1984. Curtis A . S . G . and Wilkinson C .D.W. , Topographical control of cells: Review, Biomaterials 18:1573-1583,1997. Damji A . , Weston L . , and Brunette D . M . , Directed conformations between fibroblasts and epithelial cells on micromachined grooved substrata., Exp. Ce l l Res. 228:114-124, 1996. Dickinson R . B . , Giudo S., and Tranquillo R.T. , Biased cell migration of fibroblasts exhibiting contact guidance in oriented collagen gels, Annals of Biomedical Engineering, 22:342-356, 1994. 125 Doundoulakis J .H. , Surface analysis of titanium after sterilization: role in implant-tissue interface and bioadhesion, J. Prosthet. Dent., 58:471-478, 1987. Dunn G . A . and Heath J.P., A new hypothesis of contact guidance in tissue cells, Exp. Cel l . Res., 101:1-14, 1976. Dunn G . A . , Contact guidance of culture tissue cells: a survey of potentially relevant properties of the substratum, In Cel l Behaviour (ed. Bellairs R., Curtis A . and Dunn G.). Cambridge University Press, 1982. Dunn G . A . and Brown A . F . , Alignment of fibroblasts on grooved surfaces described by a simple geometric transformation, J. Cel l Sci. , 83:313-340, 1986. Dunn G . A . and Ebendal T., Contact guidance on oriented collagen gels, Exp. Cel l Res., l l l (2) :475-479, 1978. Eastwood M . , Porter R., Khan U . , McGrouter G. , and Brown. R., Quantitative analysis of collagen gel contractile forces generated by dermal fibroblasts and the relationship to cell morphology, J. of Ce l l Physiol., 166:33-42, 1996. Eastwood M . , Mudera V . C . , McGrouther D . A . , and Brown R . A . , Effect of precise mechanical loading on fibroblast populated collagen lattices: morphological changes, Cel l Mot i l . Cytoskeleton, 40:13-21, 1998. Eckert R., Jeney S., Horber J . K . H . , Understanding intercellular nteractions and cell adhesion: lessons from studies on protein-metal interactions, Ce l l B i o l . Internat., 21(11):707-713, 1997. Ehrlich H.P. and Rajaratnam J . B . M . , Cel l locomotion forces versus cell contraction forces for collagen lattice contraction: an in vitro model of wound contraction, Tissue Ce l l , 22(4):407-417,1990. 126 El-Ghannam A . , Starr L . , Jones J., Laminin-5 coating enhances epithelial cell attachment, spreading, and hemidesmosome assembly on Ti-6A1-4V implant material in vitro, J. Biomed. Mater. Res., 41:30-40, 1998. Elsdale T. and Bard J., Collagen substrata for studies on cell behaviour, J. Ce l l B i o l . , 54:626-637, 1972. Elsdale T. and Bard J., Cellular interactions in mass cultures of human diploid fibroblasts, Nature, 236:152-155, 1972. Elsdale T. and Foley R., Morphogenetic aspects of multilayering in patri dish cultures of human fetal lung fibroblasts, J. Cel l B i o l . , 41:298-311, 1969. Emerman J.T., Enami J., Pitelka D. , and Nandi S., Hormonal effects on intracellular and secreted casein in cultures of mouse mammary epithelial cells on floating collagen membranes (cell differentiation), Proc. Natl . Acad. Sci. U S A , Cel l B i o l . , 74(10):4466-4470, October 1977. Emerman J.T. and Pitelka D.R. , Maintenance and induction of morphological differentiation in dissociated mammary epithelium on floating collagen membranes, In Vitro, 13(5):316-328, 1977. Erickson C . A . , Analysis of the formation of parallel arrays by B H K cells in vitro, Exp. Cel l Res., 115:303-315,1978. Fernyhough W. , Aukh i l I, and Link T., Orientation of gingival fibroblasts in simulated periodontal spaces in vitro containing collagen gels, J. Periodontal., 58(11):762-769, November 1987. Folkman J. and Moscona A . , Role of cell shape in growth control, Nature, 273:345-349, 1978. Frey J., Chamson A . , Raby N . , and Rattner A . , Collagen bioassay by the contraction of fibroblast-populated collagen lattices, Biomaterials, 16(2):139-143, January 1995. 127 Gay S. and Mi l l l e r E.J . (eds.), Collagen in the physiology and pathology of connective tissue, Gustav Fischer Verlag, Stuttgard, N e w York, 1978. Gelman R . A . , Will iams B.R. , and Piez K . A . , Collagen fibril formation: evidence for a multistep process, J. B i o l . Chem., 254(1): 180-186, 1979. Grinnell F. , Focal adhesion sites and the removal of substratum bound fibronectin, J. Ce l l . B i o l . , 103:2697-2706, 1986. Grinnell F. and Lamke C.R. , Reorganization of hydrated collagen lattice by human skin fibroblasts, J. Ce l l Sci. , 66:51-63, 1984. Grinnell F. and Minter D . , Attachment and spreading of baby hamster kidney cells to collagen substrata: effects of cold-insoluble globulin, Proc. Natl . Acad. Sci . U S A , 75:4408-4412, 1978. Guido S. and Tranquillo R.T. , A methodology for the systematic and quantitative study of cell contact guidance in oriented collagen gels; Correlation of fibroblast orientation and gel birefringence, J. Cel l Sci. , 105:317-331, 1993. Guidry C. and Grinnell F., Contraction of hydrated collagen gels by fibroblasts: evidence for two mechanisms by which collagen fibrils are stabilized, Collagen Rel . Res., 6:515-529,1986. Guidry C . and Grinell F., Studies on the mechanism of hydrated collagen gel reorganization by human skin fibroblasts, J. Cel l Sci. , 79:67-81, 1985. Gullberg D. , Tingstrom A . , Thuresson A . C . , Olsson L . , Terracio L . , Borg T . K . , and Rubin K . , _1 Integrin-medated collagen gel contraction is stimulated by P D G F , Exp. Ce l l Res., 186:264-272, 1990. Harris A . K . , Locomotion of tissue culture cells considered in relation to ameboid locomotion, Int. Rev. Cytol . , 150:35-68,1994. Harris A . K . and Dmytryk J. J., In The Biological Mechanism of Tooth Eruption and Root Resorption (ed. Davidovitch Z.) pp. 181-186, E B S C O Media, Brimingham, A L , 1988. Harris A . K . , Stopak D. , and W i l d P., Fibroblast traction as a mechanism for collagen morphogenesis, Nature, 290:249-252, March 1981. Harris A . K . , Stopak D. , and Warner P., Generation of spatially periodic patterns by a mechanical instability: a mechanical alternative to the Turning model, J. Embryol . Exp. Morph., 80:1-20, 1984. Harrison R . G . , The reaction of embryonic cells to solid structures, J. Exp. Zool . , 17(4):521-544, 1914. Hayman E . G . , Pierschbacher M . D . , Suzuki S., and Ruoslahti E . , Vitronectin- a major cell attachment-promoting protein in fetal bovine serum, Exp. Ce l l Res., 160:245-258, 1985. Heath J.P. and Peachey L . D . , Morphology of fibroblasts in collagen gels: A study using 400 keV electron microscopy and computer graphics, Cel l Mot i l . Cytoskeleton, 14:382-392, 1989. James J. and Tanke H.J . (editors), Biomedical light microscopy, Kluwer Academic Publishers, 1991. Keller J.C., Stanford C M . , Wightman J.P., Draughn R . A . , and Zaharias R , Characterization of titanium implant surfaces III, J. Biomed. Mater. Res., 28:939-946, 1994. Klebe R.J . , Caldwell H . , and M i l a m S., Cells transmit spatial information by orienting collagen fibers, Matrix, 9:451-458,1989. Kono T., Tanii T., Furukawa M . , Mizuno N . , Kitajima J., hamada T., and Yoshizato K . , Parallel arrangement, growth inhibition and cell cycle phase analysis of human dermal fibroblasts cultured in collagen lattice, J. Dermatol., 17(1):2-10, 1990. Kothari S., Hatton P .V . , and Douglas C.W.I . , Protein adsorption to titanium surfaces, J. Mater. Sci. Mater. Med. , 6:695-698,1995. Layfer M . , Ashkenazi C , Katz D. , and Wolman M . , Orientation of collagen in wound healing, Br. J. Exp. Pathol., 55(3):233-236, 1974. 129 Lorimier S., Hornebeck W. , Godeau G. , Pellat B . , Gillery P., Maquart F . X . , and Laurent-Maquin D . , Morphometric studies of collagen and fibrin lattices contracted by Human Gingival Fibroblasts; comparison with dermal fibroblasts., J. Dent. Res., 77(9): 1717-1729, 1998. Lowenberg B .F . , Pill iar R . M . , Aubin J.E., Fernie G.R. , and Melcher A . H . , Migration, attachment, and orientation of human gingival fibroblasts to root slices, naked and porous-surfaced titanium alloy discs, and zircalloy 2 discs in vitro, J. Dent. Res., 66(5): 1000-1005, May 1987. Lowenberg B .F . , Aubin J.E., Deporter D . A . , Sodek J., and Melcher A . H . , Attachment, migration, ad orientation of human gingival fibroblasts to collagen-coated, surface demineralized, and untreated root slices, J. Dent. Res. 64(9):1106-1110, Sept. 1985. Lowenguth R . A . , Poison A . M . , and Caton J.G., Oriented cell and fiber attachment systems in vivo, J. Periodontol. 64:330-342, 1993. MacNei l R . L . , D 'Err ico J., Strayhorn C , Pickrum H . , and Somerman M . J . , Agents with periodontal regenerative potential regulate cell-mediated collagen lattice contraction in vitro, J. Dent. Res., 75(3):903-911, 1996. Meachim G . and Will iams D.F . , Changes in nonosseous tissue adjacent to titanium implants, J. Biomed. Mater. Res., 65B:555, 1973. Merritt K . , Role of medical materials, both in implant and surface and surface applications, in immune response and resistance to infection, Biomaterials, 5:47-53, 1984. M i o T., Adachi Y . , Romberger D.J . , Ertl R.F. , and Rennard S.I., Regulation of fibroblasts proliferation in three-dimensional collagen gel matrix, In Vitro Ce l l . Dev. Bio l . -Animal , 32:427-433, July/August 1996. Nakagawa S., Pawelek P., and Grinnell F., Long term cilture of fibroblasts in contracted collagen gels: effects on cell growth and biosynthetic activity, Dermatol., 93(6):792-798, 1989. 130 Nakagawa S., Pawelek P., and Grinnell F., Extracellular matrix organization modulates fibroblasts growth and growth factor responsiveness, Exp. Ce l l Res., 182:572-582, 1989. Noble P .B. , Extracellular matrix and cell migration: locomotory characteristics of MOS-11 cells within a three-dimensional hydrated collagen lattice, J. Ce l l Sci . , 87:241-248, 1987. Nusgens B . , Merr i l l C , Lapiere C M . , Collagen biosynthesis by cells in a tissue equivalent matrix in vitro, Collagen Relat. Res., 4:351-364, 1984. Oakley C , Jeager N . A . F . , Brunette D . M . , Sensitivity of fibroblasts and their cytoskeletons to substratum topographies: topographic guidance and topographic compensation by micromachined grooves of different dimensions, Exp. Ce l l Res., 234:413-424, 1997. Oakley C. and Brunette D . M . , The sequence of alignment of microtubules, focal contacts and actin filaments in fibroblasts spreading on smooth and grooved titanium substrata, Journal of Ce l l Science, 106:343-354, 1993. Oakley C. and Brunette D . M . , Topographic Compensation: Guidance and directed locomotion of fibroblasts on grooved micromachined substrata in the absence of microtubules, Cel l Mot i l . Cytoskeleton, 31:45-58, 1995. Ohara P.T. and Buck R . C . , Contact guidance in vitro. A light, transmission, and scanning microscopic study, Exp. Cel l Res., 121:235-249, 1979. Pierard G.E . , Sirius red polarization method is useful to visualize the organization of connective tissue but not the molecular composition of their fibrous polymers, Matrix, 9(1):68-71, 1989. Puchtler H . , Waldrop F.S. , and Valentine L .S . , Polarization microscopic studies of connective tissue stained with picro-sirius red F B A , Beitr. Pathol., 150(2): 174-187, 1973. Rae T., Cel l biochemistry in relation to the inflammatory response to foreign materials, In C R C Fundamental Aspects of Biocompatibility, volume 1, ed. Wil l iams D . F. , C R C Press, Boca Raton, F L . p p . 159-81, 1981. Raikar G . N . , Gregory J.C., Ong J.L. , Lucas L . C . , Lemons J.E., Kawahara D . and Nakamura M . , Surface characterization of titanium implants, J. Vac . Sci . Technol. A 13(5):2633-2637, Sept/Oct 1995. Ratner B . D . , N e w Ideas in biomaterials science- a path to engineered biomaterials, J. Biomed. Mater. Res., 27:837-850,,1993. Register A . A . and Burdick F . A . , Accelerated reattachment with cementogenesis to dentin, Demineralized in situ. I. Optimum range, J. Periodontol, 46:646-655, 1975. Reszka A . and Horowitz A . , Integrin mutagenesis, in Cel l -Cel l Interactions (editors: Stevenson B.R. , Gal l in W.J . , and Paul D .L . ) , Oxford University Press, Oxford, 1992. Revel J.P., and Wolken K . , Electron microscope investigations of the underside of cells in culture, Exp. Ce l l Res., 78:1-14, 1973. Ridley A . J . and Hal l A . , The small GTP-binding protein rho regulates the assembly of focal adhesion and actin stress fibers in response to growth factors, Ce l l , 80:389-399, 1992. Schakenraad J . M . , Busscher H.J . , Wildevuur C . R . H . and Arends J., The influence of substratum surface free energy on growth and spreading of human fibroblasts in the presence and absence of serum proteins, J. Biomed. Mater. Res., 20:773-784, 1986. Schakenraad J . M . , Arends J., Busscher H.J . , Dijk F. , van Wachem P .B . , Wildevuur C . R . H . , Kinetics of cell spreading on protein precoated substrata: a study of interfacial aspects, Biomaterials, 10:43-50,1989. Sheetz M . P . , Felsenfeld D.P. , and Galbraith C . G . , Ce l l migration: regulation of force on extracellular matrix-integrin complexes, Trends in Ce l l B i o l . , 8(2):51-54, 1998. Sheetz M . P . , Felsenfeld D.P. , Galbraith C . G . , and Choquet D. , Ce l l migration as a five-step cycle, Biochem. Soc. Symp., 65:233-243,1999. Sheetz M . P . and Dai J., Trends in Ce l l B i o l . , 6:85-89, 1996. Smith D. , "Discussion Surface Factors", J. Oral Imp., Special issue, 2(XIV):417-422,1988. Souren J . E . M . and Wijk R., Fibroblastic cell cycling in collagen gels, Ce l l Prolif., 26:13-23, 1993. Stanford C. M . , Keller J.C., Solursh M . , Bone cell expression on titanium surfaces is altered by sterilization treatments, J. Dent. Res 73(5):1061-1071, M a y 1994. Steinberg B . M . , Smith K . , Colozzo M . , and Pollack R., Establishment and transformation diminish the ability of fibroblasts to contract native collagen gel, J. Ce l l B i o l . , 87:304-308, 1980. Stephens P., Davies K . J . , Al-Khateeb T., Shepherd J.P., and Thomas D . W . , A comparison of the ability of intra-oral and extra-oral fibroblasts to stimulate extracellular matrix reorganization in a model of wound contraction, J. Dent. Res., 75(6): 1358-1364, 1996. Stingl J., Eaves C.J. , Kuusk U . , and Emerman J.T., Phynotypic and functional characterization in vitro of a multipotent epithelial cell present in the normal adult human breast, Differentiation, 63:201-213, 1998. Stopak D . and Harris A . K . , Connective tissue morphogenesis by fibroblast traction. I. Tissue culture observations, Dev. B i o l . , 90:383-398, 1982. Stopak D. , Wessels N . K . , Harris A . K . , Morphogenetic rearrangement of injected collagen in developing chicken limb buds, Proc. Natl . Acad. Sci . U S A , 82:2804-2808,1985. Stuart E.S. , Garber B . , and Moscona A . A . , A n analysis of feather germ formation in the embryo and in vitro in normal development and in skin treated with hydrocortisone, J. Exp. Zool . , 179:97-118, 1972. Swart K . M . , Keller J.C., Wightman J.P., Draughn R . A . , Stanford C M . , and Michaels C M . , Short term plasma cleaning treatments enhance in vitro osteoblast attachment to titanium, J. Oral Implantol, 18:130-137, 1992. Tarpile E . , Ghassemifar R . M . , and Franzen L . E . , Fibroblasts movement during contraction of collagen lattices: a quantitative study using three-dimensional time-lapse technique with 133 phase-contrast laser scanning microscopy., In Vitro Ce l l Dev. Bio l . -Animal , 34:640-645, 1998. Takakuda K . and Miya i r i H . , Tensile behaviour of fibroblasts cultured in collagen gels, Biomaterials, 17(14):1393-1397, July 1996. Tengvall P. and Lundstrom., Physico-chemical considerations of titanium as a biomaterial, E . C . M . 9:115-134, 1992. Thompson G.J . and Puleo D . A . , Effects of sublethal metal ion concentrations on osteogenic cells derived from bone marrow stromal cells, J. App l . Biomater., 6(4):249-258, 1995. Tranquillo R.T. , Self-organization of tissue equivalents: the nature and role of contact guidance, Biochem. Soc. Symp. 65:27-42, 1999. Trinkaus J.P., Cells into organs, Englewood Cliffs. N . J . , U S A : Prentice Hal l Inc., pp. 179-263, 1984. Uitto V . J . and Larjava H . , Extracellular matrix molecules and their receptors: an overview with special emphasis on periodontal tissue, Crit. Rev. Oral B i o l . Med. , 2(2):323-354, 1991. Walter B . and Path F .R .C . , Wound healing, J. Otolaryng., 5(2): 171-176, 1976. Watt F . M . , The extracellular matrix and cell shape, Trends Biochem. Sci . , 11:482-485,1986. Weiss L . , Cel l adhesion, Int. Dent. J., 28(1):7-17, 1975. Weiss P., In vitro experiments on the factors determining the course of outgrowing nerve fiber, J. Exp. Zool . , 68:393, 1934. Weiss P. and Garber B . , Shape and movement of mesenchyme cells as functions of the physical structure of the medium. Contributions to quantitative morphology, Proc. Natl . Acad. Sci. , 38:264-280, 1952. Weiss P., Interactions between cells, Rev. Modern Physics, 31:449-454, 1959. Weiss P., Experiments on cell and axon orientation in vitro: The role of colloidal exudates in tissue organization, J. Exp. Zool . , 100:353-386, 1945. 134 Weiss P, and Taylor A . , Fish scales as substratum for uniform orientation of cells in vitro, Anat. Res., 124:381, 1956. Weiss P., Nervous system, In Analysis of Development (ed. Wil l ie r B . H . , Weiss P., and Hamburger V . ) , pp.346-401, Saunders, Philadelphia, 1955. Werb Z . , Tremble P . M . , Behrendtsen O., Crowley E . and Damskey C . H . , Signal transduction through the fibronectin receptor induces collagenase and stromelysin gene expression, J. Ce l l B i o l . , 109:877-889, 1989. Wilkinson P.C. , Shields J . M . , and Haston W.S. , Contact guidance of human netrophil leukocytes, Exp. Cel l Res., 140:55-62, 1982. Will iams D.F. , Biocompatibility considerations in the use of orthopedic implant alloys, Current Perspectives on Implantable Devices, 1:47-70, 1989. Will iams D.F. and Meachim G. , A combined metallurgical and historical study of tissue-prosthesis interactions in orthopeadic patients, J. Biomed. Mater. Res., Symp. 5 (Part 1):1, 1974. Yamada K . M . and Olden K . , Fibronectin: adhesive glycoproteins of cell surface and blood, Nature, Lond. 275:179-185, 1978. Yamato M . , Adachi E . , Yamamoto K . , and Hayashi T., Condensation of collagen fibrils to the direct vicinity of fibroblasts as a cause of gel contraction, J. Biochem. (Tokyo), 117(5):940-946, M a y 1995. Zitter H . and Plenk Jr. H . , The electrochemical behaviour of metallic implant materials as an indicator of the biocompatibility, J. Biomed. Mater. Res., 21:881-896, 1987. 135 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0089536/manifest

Comment

Related Items