Advances in Recombination Research

I N the ten years since I last addressed an International Congress of Genetics, considerable progress has been made in understanding the process of recombination, and yet many of its central features still elude us. I would like to summarize briefly some of the main advances of the last decade, and to indicate where these discoveries seem to be leading us. An outstanding advance was the demonstration by TAYLOR (1965) that chiasmata correspond to points of breakage and rejoining, as shown by 3H-thymidine labelling. This experiment provided the first direct evidence that crossing over was by breakage and joining, more than 50 years after MORGAN had first demonstrated the process. Any suspicion that TAYLOR’S conclusion was falsified through the occurrence of sister-chromatid exchange has been eliminated by the experiments of PEACOCK (1970) and JONES (1971). Of course, breakage and rejoining had already been established for prokaryote recombination, initially in the classic experiments by MESELSON and WEIGLE ( 1961 ) and by KELLENBERGER, ZICHICHI and WEIGLE (1961) with phage lambda of Escherichia coli, and subsequently in transformation studies by Fox and ALLEN (1964) and by BODMER and GANESAN ( 1964) with Diplococcus pneumoniae and Bacillus subtilis, respectively. That recombination has been shown to occur by breakage and rejoining does not mean that break-and-copy is ruled out. Such models invoke breakage of a nucleotide chain followed by its extension using a complementary chain from the other parent as template, and have been favored by BOON and ZINDER (1971) for phage f l of E. coli, and by STAHL et al. (1972) for the general recombination in lambda. On the other hand, the idea that break-and-copy mechanisms operate in eukaryotes (e.g., PASZEWSKI 1970) has little evidence to support it, as the studies by LEBLON and ROSSIGNOL (1973) with Ascobolus immersus have shown. Much new evidence has been obtained recently for correction of mismatching of bases in hybrid DNA. First suggested by HOLLIDAY (1962), it has been favored ever since as an explanation for conversion in eukaryotes, but in prokaryotes its occurrence has been questioned until recently. EPHRUSSI-TAYLOR and GRAY (1966) proposed such correction to account for the low integration efficiency of some mutants in pneumococcal transformation, and HOGNESS and co-workers (HOGNESS et al. 1967; DOERFLER and HOGNESS 1968) obtained evidence for it in phage lambda from ingenious experiments using heteroduplex molecules prepared in vitro. But the idea of mismatch repair in prokaryotes did