Background Many investigations have reported that advantageous mutations occurred more frequently under selective conditions than those under non-selective conditions. This phenomenon is referred to as adaptive mutation. Their characteristics are that adaptive mutations are directed and growth-independent. The idea of directed adaptive mutation had been objected by some reports, however, the idea of growth-independent adaptive mutation has been held till today.

      Results
      In this paper, we have observed that under leucine starvation conditions, leu+ revertants accumulated as a function of time; leu- to leu+ reverse mutation rates and frequencies were higher than those under non starvation conditions; and no divided cells could be monitored by the penicillin method. These results were similar to the time-dependent manner of adaptive mutation from previous reports. However, leucine concentration determinate experiments revealed that certain traces of leucine, which leaked from the E. coli cells, was almost always present in the culture. More numbers of leu+ revertants appeared when the similar cultures were dropped in small areas on the selective plates than when spread on the whole selective plates. These results have shown that mutations under leucine starving conditions are growth-dependent. Fluctuation analysis of leu+ revertants indicated that leu-leu+ mutation occurred spontaneously and randomly. In addition, the spectra of leuB gene in the revertants proved that mutations under selective conditions were not specific or directed.


      Conclusions
      The above investigations led to the conclusion (1) that the occurrence of leu+ mutations under starvation conditions was growth-dependent. The occurence mutations was also similar to that under non-starvation conditions (2). Under starvation conditions the mutation rates were higher, and was not constant during the long process.