Although early histologists suggested that the cytoplasm of cells possessed a characteristic ar- chitecture, the full impact of these observations did not become clear until the demonstration of cytoskeletal components, microtubules, contractile microfilaments, and intermediate (10-nm) filaments in cells. The most common method of demonstration of these structures has been by fluorescent antibody techniques.
Microtubules are ultramicroscopic tubular structures found in virtually all cells. They comprise a family of proteins called tubulins α, β, and γ (cf. Burns, 1991). The α- and β-tubulins form heterodimers and assemble into microtubules. γ-Tubulin is primarily associated with the poles of the microtubule structure. Originally, microtubules were felt to be primarily involved in mitosis, where they make up a major portion of the mitotic spindle. However, it is now apparent that microtubules have definitive patterns of organization within cells, depending on the cell type, and are associated with a variety of other proteins that are important in a number of cellular functions. A listing of some of the structure/function relationships of microtubules in cells is given in Table 16.3. A number of proteins have been associated with microtubules and are essen-
Table 16.3 Structure/Function Relationships of Microtubules in Cells
tial for several of the functions noted in Table 16.3. As expected, different cells express different protein families for such microtubule expression (cf. Lane and Allan, 1998; Avila, 1992). Mi- crofilaments are the smallest of the components of the cytoskeleton and consist of one or more of several contractile proteins and their associated molecular species related to actin and myosin (cf. Goldman et al., 1979).
Intermediate filaments are a more diverse group of cytoskeletal elements, which have been divided into a number of groupings including keratins, vimentin, desmin, and neurofilament pro- teins (Table 16.4; Weber and Osborn, 1982). These cytoskeletal elements are members of a large multigene family, many of which are differentially expressed in different tissues (Fuchs and Ha- nukoglu, 1983; Moll et al., 1982). Such variation in the expression of components of the multi- gene family of intermediate filaments has also been seen in neoplastic cells, both in vivo and in vitro (Wada et al., 1992; Skalli et al., 1988; Caulín et al., 1993), and this finding has been used diagnostically to distinguish general classes of neoplasms (cf. Miettinen et al., 1983). Summer- hayes et al. (1981) demonstrated that rat bladder epithelium in culture as well as nontumorigenic foci altered by exposure to benzo[a]pyrene demonstrated no detectable vimentin filaments. However, tumorigenic cell lines from this tissue did express this protein. In mouse epidermal keratinocytes, both viral and chemical transformation resulted in abnormalities in the expression of keratins (cf. Caulín et al., 1993). Specifically, keratin K8 was upregulated in transformed epi- dermal cell lines and in neoplasms resulting from injection of the cells back into appropriate mouse hosts.
One of the most striking changes exhibited in the cytoskeleton of transformed cells is ex- emplified by Figure 16.3. The upper figure is an artist’s conception of the patterns of microfila- ments forming networks of “actin cables” extending in a parallel fashion throughout the cytoplasm of the cultured cell resting on a surface. Below this is shown the microfilament pat- tern in a transformed cell. Here the filaments display very little parallel arrangement, and there is diffuse fluorescence to indicate depolymerization of microfilaments into their constitutive mole- cules, which react to give the diffuse pattern. The normal patterns of microtubule (Brinkley and Fuller, 1978) and intermediate filaments (Ben-Ze’ev, 1984) lose the organization seen in non- transformed cells when they become transformed.
Both microfilaments and microtubules interact directly with the surface membrane of the cell. Thus, both the morphology of a cell and the mobility of its lectin-binding sites are in all likelihood directly related to microtubule and microfilament organization. This probability has been demonstrated by a direct relationship between concanavalin A capping and a redistribution of microtubules as well as the changes induced by a temperature-sensitive Rous sarcoma virus (cf. Nicolson and Poste, 1976). In the latter instance, both cell morphology and surface topogra- phy are altered concomitantly with changes in the organization and assembly of microfilaments. Therefore, it is quite likely that the mobility of lectin-binding sites in neoplastic cells is more directly related to changes in the microfilament-microtubule system governing cell architecture than to membrane microviscosity. In transformed fibroblasts exhibiting this disruption of mi- crofilaments, receptor proteins that are normally associated with a relatively rigid organization of the structures would become more mobile, allowing for the lectin induction of clustering and capping (Wang and Goldberg, 1976). In the case of the lymphocyte, capping induced by lectins and by antibodies to specific surface receptors may be directly related to the interaction of mi- crofilaments with surface membrane receptors and other proteins (Bourguignon and Bourgui- gnon, 1984). Ultimately in the normal lymphocyte, as discussed in Chapter 20, interaction with lectins or antibodies stimulates cell replication, but such stimulation may not always be effective in neoplastic lymphocytes.
Figure 16.3 Patterns of microfilaments (actin cables or fibers) in normal and transformed fibroblasts growing in vitro. The regular pattern of parallel microfilaments in the normal cell can be contrasted with the disarray and diffuseness of the cables and fibers in the transformed cell.