The ability of mammalian embryos to incorporate foreign cells and develop as chimeras has been exploited for a variety of purposes including the elucidation of cell lineages, the investigation of cell potential, the perpetuation of mutations produced in embryonic stem (ES) cells by gene targeting, and the subsequent analysis of these mutations. The extent of contribution of the foreign cells depends on their developmental synchrony with the host embryo and their mitotic and developmental potential, which may be severely restricted if the cells bear mutations. If the goal in making chimeras is the transmission of a mutation produced by gene targeting to the next generation, the mutant ES cells must have the capacity to undergo meiosis and gametogenesis. Cells from two different mammalian embryos were first combined experimentally to produce a composite animal, dubbed a chimera, nearly four decades ago. Pairs of cleaving, pre-implantation embryos were mechanically associated in vitro until they aggregated together to make single large morulae; these in turn resulted in chimeric offspring (1). Genetic markers were used to distinguish the contributions of the two embryos in these animals. Since then, various methods for making chimeras have been explored to address different types of questions (2). In 1972 it was reported that highly asynchronous embryonic cells, which had been cultured in vitro, could contribute to chimeras upon re-introduction into pre-implantation embryos (3). Not long afterward, several groups working with teratocarcinomas, tumours derived from germ cells of the gonad, discovered that stem cells from these tumours, known as embryonal carcinoma cells, could contribute to an embryo if introduced into pre-implantation stages (4-6). It appeared that the undifferentiated stem cells of the tumour had enough features in common with early embryonic cells that they could respond to the embryonic environment, differentiating in a normal manner, even after long periods in vitro. Their embryonic potential was limited, however, and many teratocarcinoma cell lines made only meagre contributions to the developing fetus or even produced tumours in chimeras (7). Either their derivation from tumours or their extended sojourn in vitro rendered these cells so dissimilar from early embryonic cells that they rarely, if ever, had full embryonic potential.
Use of double-replacement gene targeting to replace the murine alpha-lactalbumin gene with its human counterpart in embryonic stem cells and mice.