Conocimientos Previos para el aprendizaje Significativo

Protein attachment precedes cellular interaction with the substratum. Cell adhesion to a substrate includes the steps of serum-protein adsorption, cell contact, attachment and spreading (Vogler and Russian, 1987). For a cell inoculum attaching to a substrate from a static fluid, there is a brief lag phase followed by a rapid increase in attached-cell numbers, then a declining rate of attachment to a plateau or equilibrium- adherence level. Cell spreading begins well after initial contact, continuing during and after the first hour o f attachment, while adhesion patches and plaques form Avith progressive adherence to the substrate. By contrast, steps prior to spreading occur over a shorter time period. Protein adsorption is nearly spontaneous, depositing from serum a 2-5 nm layer within the first minute of contact. Cells in suspension can gravitate to within 5-8 nm of a substrate in less than 5 minutes, although maximal attachment seldom occurs in this time frame. Rather there is a time-dependent cellular interaction with adsorbed proteins, probably involving both microextensions of the cell surface that break electrostatic barriers and cell-membrane receptors specifically directed towards various adsorbed factors (Sharefkin and Watkins, 1986). Thus, cell contact and adhesion is a time-dependent phenomenon.

The primary step in cell-substrate interactions is the rapid and differential adsorption o f proteins from the media to form a substrate more enriched in some proteins than others. However, the significance of these differences is not so well understood. Differences in the adsorbed protein layer on different substrates might affect cell behaviour in a variety o f ways. The simplest theory is the 'composition' mechanism, whereby increased amounts of adsorption of a protein for which the cell has an affinity cause that cell to interact more strongly with the substrate. An alternative theory is that

the 'state' or conformation of the adsorbed protein is more important than its amount in affecting cell attachment (Grinnell & Feld, 1981).

Cell attachment to surfaces begins with protrusion of ruffled edges or lamellipodia, followed by cellular spreading. Initial attachment, spreading, and flattening are active, energy-requiring processes. Cell flattening and spreading appear to be necessary, though not sufficient, events for later DNA synthesis and cell division. Most eukaryotic cells have networks of 70Â diameter microfilaments containing the contractile protein actin, which functions together with myosin-like proteins to cause cell movement and shape changes. Attachment and spreading on plane substrata involves the dynamic reorganisation of three filamentous structures: microtubules, microfilaments and intermediate filaments (Ireland et al., 1987). Several types of cell-substratum adhesion have been described, including focal contacts, close contacts and extracellular matrix contacts (Puleo & Bizios, 1992). Of particular interest is the interaction between the cytoskeleton and a number of adhesive glycoproteins in the extracellular matrix. Fibronectin and vitronectin are the two most well studied. They play an important role in linking the cells to surfaces, inducing cell spreading, and possibly regulating later cell growth and phenotype expression (Alberts et al., 1989). These proteins can attach in an active form to plastic surfaces, to collagen and to the fibrin strands of fresh clots. They enhance attachment of cell types as diverse as fibroblasts, osteoblasts, endothelial cells to collagenous matrices or plastic surfaces (Hewlett et al., 1994). They do so by binding to transmembrane glycoprotein receptors which are in turn attached to the cytoskeleton (Yamada et al., 1985). These and other matrix receptors including some that bind collagen and laminin belong to the family of transmembrane glycoproteins called integrins. They are heterodimers with a and p chains and recognise the arginine-glycine-aspartic acid (RGD) tripeptide sequence in the matrix components they bind. The integrins link the extracellular matrix to the cytoskeleton and provide a mechanism by which matrix proteins can influence cellular activities via changes in cell shape or morphology (Sauk et al., 1991).

In bone matrix there are multiple glycoproteins that contain the integrin-binding RGD sequence: fibronectin (FN), thrombospondin (TSF), osteopontin (OPN), bone sialoprotein (BSP), type I collagen (COL I), and vitronectin (VN). Grzesik & Robey

(1994) localised TSP, FN, VN, and several integrins within developing human long bone using immunohistochemical methods. They went on to assess the effect o f all bone RGD proteins on the adhesion of human osteoblastic cells. Thrombospondin, fibronectin, and vitronectin showed distinct localisation patterns within bone tissue. TSP was found mainly in osteoid and the periosteum; VN appeared to be present mainly in mature bone matrix. FN was present in the periosteum as well as within both mature and immature bone matrix. Using a panel of anti-integrin antibodies, it was found that bone cells in vivo and in vitro express a^, a^, a^Pj, and p/p^ integrins, and these receptors were expressed on all bone cells at different stages of maturation with quantitative rather than qualitative variations, with the exception of a^, which was expressed mainly by osteoblasts. Cell attachment assays performed using primary human bone cells under serum-fi-ee conditions revealed that COL I, TSP, VN, FN, OPN, and BSP promoted bone cell attachment in a dose-dependent manner and were equivalent in action when used in equimolar concentrations. In the presence of GRGDS peptide in the medium, the adhesion to BSP, OPN, and VN was almost completely blocked and attachment to FN, COLL I, and TSP was only slightly reduced. These results suggest that human bone cells may use RGD-independent mechanisms for attachment to the latter glycoproteins.

1.3.7 Cell attachm ent assays

Many different cell types may potentially attach to the surface of dental implants. In the absence of infection, the desired goal is to achieve a stable interface with adjacent bone and the connective tissue. Although it is not clear that bone cells are actually attaching to the surfaces of endosseous implants in vivo, many research groups have used cell attachment studies to explore the primary interaction between implant surfaces and osteoblast-like cells.

Measurement of cell attachment is usually carried out by means o f counting at different time intervals either the number of cells not attached to the material surfaces or dissociating the attached cells firom the substrata and resuspending them for counting. Cell number can also be assessed by radioisotope labelling and scintillation counting o f labelled cells.

1.3.7.1 Bone RGD proteins and cellular responses

Hewlett et al. (1994) investigated the contribution that serum fibronectin (Fn) or vitronectin (Vn) make to the attachment and spreading of cells cultured from explanted human bone (bone-derived cells) during the first 90 min of culture on metallic and ceramic surfaces. The requirement for Fn or Vn for attachment and spreading of bone- derived cells onto stainless steel 316 (SS), titanium (Ti) and alumina (AI2O3) and to

polyethyleneterephthalate (PET) was directly tested by selective removal o f Fn or Vn from the serum prior to addition to the culture medium. Attachment and spreading of bone-derived cells onto SS, Ti and AI2O3 surfaces were reduced by 73-83% when the

cells were seeded in medium containing serum from which the Vn had been removed. Cell attachment and spreading on these surfaces when seeded in medium containing Fn- depleted serum (which contained Vn) were not reduced to the same extent as in the medium containing Vn-depleted serum. The bone-derived cells failed to attach to the surfaces to the same extent when seeded in medium containing serum depleted o f both Vn and Fn. The results show that for human bone-derived cells, the attachment and spreading of cells onto SS, Ti and AI2O3 as well as PET during the first 90 min o f a cell

culture attachment assay are a function of adsorption of serum Vn onto the surface.

1.3.7.2 Cytoskeletal Organisation

Sinha et al. (1994) hypothesised that the nature o f cell substratum attachment may determine subsequent cellular behaviour and in the context of bone-implant interfaces, may be a predictor of bone growth on clinically relevant surfaces of varying chemical composition and topography. Apart from measuring the number of cells attached to a variety o f metal surfaces, they examined other characteristics including cell spreading, cytoskeletal organisation and focal contact formation. Different patterns of actin filament distribution were found and focal contact areas varied between cells grown on different materials.

An alternative assay for cell attachment was proposed by Hunter et al. (1995). They assessed fibroblast and osteoblast attachment on a variety of metals and polymers for orthopaedic use by indirect immunofluorescent labelling of vinculin, a component of

the cell's focal adhesion plaque. The degree of cell attachment was quantified on the materials by determining the mean number of adhesion plaques and using an image analysis system to determine the mean total area o f plaques per cell. Fibroblasts and osteoblasts responded differently to the materials tested. Scanning electron microscope (SEM) observations of cells on the materials correlated with the morphometric data. Cells with the greatest number and area of adhesion plaques were well spread and flattened whilst those with the least number of adhesion plaques were more rounded and less spread.

13,7,3 Integrin-mediated adhesion

The question of whether integrins are involved in the attachment o f osteoblasts to implant materials has been addressed only recently. Sinha & Tuan (1996) found that the integrin subunits, a j, a^, a^, a^, a^, p, and pg, were expressed by primary human osteoblasts cultured on polystyrene coated with various extracellular molecules. However, and were notably absent in cells attached to the orthopaedic metal alloys. Also, ttj was not present on rough TiAlV, polished CoCr, or rough CoCr, and P3 was not

expressed by cells on rough CoCr. The ability of cells to adhere to and receive messages from the extracellular matrix may also be influenced by the substratum. These differences in integrin expression may underlie previously observed differences in degree of cell attachment to these metals.

Gronowicz & McCarthy (1996) looked into the details of how the human osteoblast-like cell line SAOS-2 adhere to orthopaedic implant materials. To determine if integrins were involved in cell attachment to implant materials, the peptide GRGDSP (Gly-Arg-Gly-Asp-Ser-Pro), which blocks integrin receptors through the Arg-Gly-Asp sequence, was added to the cells in serum-ffee medium. This peptide inhibited cell adhesion on the metal alloys. Inhibition of protein synthesis and enzymatic removal of surface proteins did not affect the ability of Arg-Gly-Asp peptides to inhibit cell attachment to the implant materials. These results suggest that integrins are able to bind directly to metals. Western blot analysis of integrin proteins revealed different levels of many integrin subunits, depending on the substrate to which cells attached. To determine specifically which integrins may be involved in adhesion, antibodies to integrins were

added. An antibody to the fibronectin receptor, significantly inhibited binding of cells to the metals but the antibody to vitronectin receptor, a^pj/pj, did not alter cell adhesion. The conclusion was that osteoblast-like cells appear to be capable o f attaching directly to implant materials through integrins. The type of substrate determines which integrins are expressed by osteoblasts.