Volume 1

Gene expression / Benjamin Lewin.

  • Lewin, Benjamin.
Date:
1974-1977
Catalogue details

Licence: Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)

Credit: Gene expression / Benjamin Lewin. Source: Wellcome Collection.

    148/668 (page 126)
    to 23S rRNA, which indicates that these two genes are adjacent. An order of 16S-23S-5S rRNA genes would be consistent with these results. After extracting DNA from E.coli cells in stationary phase—which should contain only one copy of the genome in each cell—Spadari and Ritossa (1970) found that the amount of 16S and 23S rRNA which can hybridize corresponds to between 5-7 and 7-4 genes; this suggests that there are about six or seven genes for each rRNA on the genome. By hybridizing fragments of DNA of different lengths with rRNA, they showed that the extent of clustering is very high, with all the rRNA genes probably located at one site. The order of genes for 16S and 23S rRNA within each gene pair can be detected by following the synthesis of molecules of ribosomal RNA after transcription is blocked by addition of an inhibitor. Bleyman et al. (1969) used actinomycin, which probably inserts a random block to transcription. This means that the sensitivity of a unit of transcription to the antibiotic should depend on its length and genes which are far from the point where RNA polymerase starts transcription should have a greater risk that a block will prevent RNA synthesis from proceeding through them. In B.subtilis, the synthesis of 23S rRNA is almost twice as sensitive to actinomycin as that of 16S rRNA; since this is the same ratio as the lengths of the genes, synthesis of these two molecules appears to be independent. The synthesis of 5S rRNA is very sensitive to actinomycin, which supports the idea that it is located at the end of a large precursor molecule; one model is to suppose that the 5S rRNA is cleaved from the 23S precursor. Rifampicin provides a better tool for transcriptional mapping, for this antibiotic specifically inhibits the initiation of transcription without influencing the completion of RNA chains already under synthesis. This means that longer units of transcription can function for greater lengths of time after the addition of rifampicin to bacterial cells, for they possess a greater number of RNA polymerase enzymes already engaged in RNA synthesis. On average, if there are n polymerases transcribing 16S rRNA, there should be 2 n polymerases transcribing the 23S rRNA gene, so that twice as much 23S rRNA should be produced compared with 16S rRNA after the addition of rifampicin. But if the two genes are transcribed together as one unit with, say, the gene for 16S rRNA preceding that for 23S rRNA, then 23S rRNA will be synthesised by all the polymerase molecules located on the 16S rRNA gene as well as those located on the 23S rRNA gene itself. A similar prediction can be made for the situation in which the 23S rRNA gene precedes the 16S rRNA. By adding P 32 together with rifampicin and measuring its incorporation into the 16S and 23S rRNA molecules made in the presence of the antibiotic, Doolittle and Pace (1971) were able to show that the most likely arrangement of genes is one in which 16S rRNA immediately precedes the gene for 23S rRNA, with each molecule of RNA polymerase transcribing both genes. An analysis of the incorporation of H 3 -uridine into 5S RNA in similar experiments