There are two major approaches that can be used to generate cell lines of interest. The first of these, and still the most frequently used, is to transduce an immortalizing gene construct into cultures of dividing cells. This can be accomplished by transfection or by use of retroviruses to insert the immortalizing gene randomly into the cellular genome.
Despite the undoubted importance of in vitro gene insertion in the creation of cell lines, there remain certain drawbacks intrinsic to this approach. For example, transfection requires a large number of target cells in order to ensure that some cells of interest stably integrate the chosen DNA in a position suitable for expression. Viral-mediated gene transfer can be carried out with smaller numbers of cells by cocultivation of target cells with virus-producing feeder layers, but still requires the division of target cells of the desired cell type to achieve the necessary goal of integration of the selected DNA into the genome. Moreover, both of these technologies require the growth of cells for
From: Methods in Molecular Biology, Vol. 97: Molecular Embryology: Methods and Protocols Edited by: P. T. Sharpe and I. Mason © Humana Press Inc., Totowa, NJ
extended periods of time in culture under selective pressure to obtain sufficient numbers of immortalized cells for experimentation.
An additional problem associated with the use of immortalized cell lines is that the introduction of immortalizing genes into cells can alter normal cellular physiology in two different ways. First, because the functional definition of an immortalizing gene is that it prevents cells from terminally differentiating, expression of characteristics of differentiated cells may be prevented in immortalized cell lines. In addition, expression of immortalizing genes can also alter the response of cells to exogenous signals, such as mitogens or regulators of differentiation. Such alterations can be so dramatic that they actually cause cells to respond to particular signals in a manner fundamentally different from that which occurs in the normal counterpart of the immortalized cell. One example of this problem is offered by studies on cooperative interactions between SV40 large T-antigen (TAg, an immortalizing oncogene) and a muta-tionally activated ras gene (a transforming oncogene, the product of which is thought to be involved in the control of cell division). When expressed in Schwann cells harboring TAg, ras converts slowly dividing cells dependent on exogenous mitogens to rapidly dividing cells capable of growing in mitogen-free conditions, an effect characteristic of a transforming oncogene. In contrast, when ras is expressed in Schwann cells that are not also expressing an immortalizing oncogene, then the constitutively activated ras protein rapidly induces the diametrically opposite effect of proliferation arrest (4). Since it is thought that the ras protein normally functions in nontransformed cells as a principal component of signal transduction systems (for review, see ref. 5), the ability of an immortalizing oncogene to alter the effects of ras activation raises the possibility that expression of immortalizing oncogenes can also alter the effects of other components of the signal transduction apparatus.
The potential alterations in cellular physiology associated with expression of immortalizing genes theoretically can be overcome through the use of conditional immortalizing genes, which allow the generation of cell lines in which the activity of the product of the experimentally introduced gene can be turned off by manipulation of the cellular environment. For example, the tsA58 mutant of TAg (6) encodes a thermolabile protein capable of immortalizing cells only at the permissive temperature of 33°C, and has been successfully used in the generation of a variety of conditionally immortal cell lines, as discussed below. In addition, it has turned out that cell lines rendered immortal by overexpression of the c-myc gene are, in at least some cases, able to differentiate normally in vivo.
The construction of cell lines able to differentiate has proven to be a powerful approach for studying the developmental biology of various neural cell types, including both neurons and glia (7-11). In these studies, it has been possible to create cell lines that are able to participate in normal development after injection into the developing brain. The range of genes used to generate cell lines is large, and includes the SV40 large T-gene (8,12-16), polyoma virus large T and adenovirus 5 E1A (17), src (18,19) and c-and v-myc (10,20). Recently, Louis and coworkers also have described a spontaneously generated cell line called CG-4 (21); since this cell line arose spontaneously, the immortalizing mutation is unknown.
The capacity of conditionally immortalized neural cells to regain normal patterns of behavior in certain conditions (e.g., after isTAg is inactivated at nonpermissive temperatures) is indicated by both in vitro and in vivo experimentation. In vitro, in the studies on Schwann cells referred to earlier, the effects of expression of activated ras protein were identical in normal Schwann cells and in Schwann cells harboring tsTAg and switched to nonpermissive temperatures (4). More dramatically, studies by McKay and colleagues (8,9,22) have demonstrated that a hippocampal cell line immortalized with isTAg in vitro can undergo neuronal differentiation and apparent integration into normal tissue when injected into the rodent central nervous system (CNS) (where the normal body temperature is sufficiently high to cause isTAg to be degraded rapidly and thus inactivated). The capacity of c-myc-expressing cells to integrate normally into CNS tissues is elegantly demonstrated by the recent studies of Snyder et al. (10). In addition, oligodendrocyte-type-2 astrocyte (O-2A) progenitor cell lines made with either ísA58 TAg or c-myc are useful tools for the in vitro and in vivo study of this lineage, because these cells are able to differentiate in a similar manner to O-2A progenitor cells from primary cultures (11,20). The in vitro differentiation potential of the O-2A/TAg cell line correlates well with its differentiation potential in vivo. Cell lines transplated into white matter lesions depleted of host glial cells are able to remyelinate axons and also to generate astrocytes in vivo (11,20).
Despite the merits of using conditionally active oncogenes to immortalize cells, these approachs still share many of the problems associated with the introduction of constitutively active immortalizing cells by in vitro means of gene insertion. Thus, the inability to target rare cells, the need to promote cell division to achieve integration and continued function of the immortalizing gene, and the need for extended growth of immortalized cells in vitro before enough cells exist for experimental use are all problems that are not affected by whether oncogene activity is conditional or constitutive.
One final problem associated with the generation of cell lines by in vitro gene insertion is that every cell line produced by these procedures has the immortalizing oncogene integrated into a different site in the genome. For reasons presumably associated with these differing sites of integration, putatively identical cell lines can express markedly different levels of oncogene product and markedly different behaviors. This problem can even confound the use of conditional immortalizing genes. For example, introduction of teTAg into rat embryo fibroblasts leads to the ready isolation of cell lines that can be grown indefinitely when maintained at 33°C (the permissive temperature for function of this gene), but that rapidly senesce when switched to 39.5°C (the nonpermissive temperature). However, the degree of conditionality expressed by different fibroblast lines varies over several orders of magnitude, with some lines showing only a modest reduction in growth at 39.5°C, whereas other cell lines derived from the same infected plate of fibroblasts exhibited an almost complete cessation of growth when switched to nonpermissive temperatures (23). As a consequence of this variability, it is obviously difficult to compare the properties of cell lines made in different laboratories, or even of lines made within a single laboratory.
Was this article helpful?