According to a recent study, however, cell fusion triggered by viruses is a possible contributing factor in the development of human cancer. The study also raises concerns about the use fusogenic viruses as vectors for human gene therapy or in other clinical applications owing to the possibility that such viruses might cause cancer.

The idea that aberrations in the number or structure of chromosomes can spur tumor formation is more than a century old. Such aberrations--known collectively as "aneuploidy"--arise in two principal ways: as a consequence of abnormal cell division, or as a result of cell fusion. By either mechanism, the resulting aneuploid cells no longer have the proper genetic makeup and frequently die. But researchers now know that tumor cells are often aneuploid--and very much alive. Whether aneuploidy is a cause or a consequence of a cancerous cellular state is the crux of a current debate.

In a recent study, Dr. Yuri Lazebnik and his colleagues at Cold Spring Harbor Laboratory observed, fortuitously, that normal cultured human cells are fused by the action of the Mason-Pfizer Monkey Virus (MPMV), but that the resulting hybrid cells do indeed fail to proliferate. However, the researchers discovered that if one of the fusion partners carried a particular "predisposing" gene mutation (in the oncogenes E1A or Myc, or in the tumor suppressor gene p53), then a significant proportion of the resulting hybrid cells were highly proliferative and thus potentially cancerous.

Whether such proliferating hybrid cells are produced by viruses in the human body, whether they can lead to cancer, and which of the many known and candidate human fusogenic viruses (for example, endogenous retroviruses, whose DNA sequences comprise at least 8% of the human genome) might contribute to cancer remain to be determined.

In addition to revealing that common viruses might contribute to cancer by fusing cells, the researchers report that the use of fusogenic viruses as vectors for gene therapy or in other clinical applications should be carefully evaluated.

cshl/