PubMed1989
For the lymphocytic descendants of the hematopoietic differentiation pathway, characteristic gene rearrangements result in deletions of significant portions of chromosome regions specifying lymphocyte-specific gene products on either or both chromosomes of each involved pair. Molecular mechanisms facilitating the rearrangement/deletion events have been elegantly and fruitfully elaborated in the years since the first documentation of their occurrence by Hozumi and Tonegawa (N. Hozumi and S. Tonegawa, Proc. Natl. Acad. Sci. USA, 73:3628-3632, 1976). Numerous genetic phenomena observed in experiments or the literature suggest to us that specific genome rearrangement/deletion may be characteristic of and necessary for many, perhaps all, differentiating lineages. Thus we propose that on the way to terminal differentiation, cells within a particular lineage must rearrange specific chromosome regions characteristic for that lineage in order to switch off or on, perhaps concomitantly, requisite genes for that differentiation program. Such rearrangements and their ensuing deletions may have been undetected cytogenetically because of small size and/or because normal terminally differentiated cells do not enter mitosis. However, footprints of these rearrangements, most likely in aberrant form, may be preserved in preneoplastic cells and cancer cells of various lineages which exhibit characteristic deletions. An understanding of the cellular recombinational machinery involved in normal physiological genome rearrangements such as we propose may clarify some puzzling aspects of current theories concerning retinoblastoma, Wilms' tumor, and other "deletion syndromes" and the role of parental genome imprinting [B. M. Cattanach and M. Kirk, Nature (Lond.), 315: 496-498, 1985; C. Sapienza et al., Nature (Lond.), 328: 251-254, 1987; D. Solter, Annu. Rev. Genet., 22: 127-146, 1988]. The recombinatorial activity, when inappropriately expressed in dividing cells (i.e., cells which should be terminally differentiated but are still cycling for various reasons) could be responsible for such diverse phenomena as large deletions; chromosomal translocations into commonly deleted regions; amplicons; apparent nonrandom chromosome integration of viral genomes such as hepatitis B, human papilloma virus, papovaviruses, and retroviruses; and the observation of fragile sites. It could explain why these various phenomena often involve the same restricted regions of the genome. Some clues and consequences integral to the proposal are discussed.(ABSTRACT TRUNCATED AT 400 WORDS)