Amino acid control over RNA synthesis: a re-evaluation.

Wheni a strinigent strain of Escherichia coli is starved of aii amirno acid, RNA synthesis, as usually measured, as well as protein synthesis, stops.' Several explanations have been proposed for this observation. It has been suggested (1) that the discharged transfer RNA, accumulating during amino acid starvation, inhibits the RNA polymerase;1 2 (2) that traniscriptioni anld translation are coupled, through a mechanism in which the ribosomes in the process of protein synthesis pull RNA off the I)NA template;'-' and (3) that the synthesis of one of the substrates for RNA synthesis, UTP, is subject to amino acid control.6 All these theories also reflect the genieral belief that the synthesis of all species of RNA, messenger, transfer, and ribosomal, are equally subject to the same conltrol. However, clear support for or againist this theory of "coordiniate control" has beeni lackinig due to difliculty in obtaining an unambiguous measure of the amount of messenger RNA or unistable RNA made under conditions of amino acid starvation;7 it is to this questioni that this work is addressed. In previous work it has been established that unider conditionis where there is little net RNA synthesis, as is the case during amino acid starvationl, the entry of a radioactive base into the initracellular nucleotide pool is severely restricted.'0 It is therefore possible that unider these coniditions the synthesis of unstable forms of RNA, not contributing to net RNA sytithesis, may go undetected due to the failure of the labeled precursors to eniter the intracellular pools.8' 10 That this is in fact the case, ancl that colisiderable RNA synithesis can be measured during amino acid limitation, is shown in this paper. Mllethods.-The bacterial strains uded ill this study were <,. coli K-12, E. coli NP-2 (formerly called KBI), and a mutant of the latter, NP1-29 (formerly 1-9), carrying a temiperature-sensitive valyl--tRNA syvnthetase.11 These cells were cultured in a Trisbuffered medium,12 supplemnelited with 0.001 111 K2HPO4, 0.4% glucose, 20 ,ug/ml adenine (which inihibits, the interconversioin reactionbetween guaniine anid adeniniel3), and 0.05 gg/ml thiamniie-HCI. For the growthof strains NP-2 an-d NP-29, this medium was further supplemnevited with 20 tg/ ml uracil an-d 10 nig/gnl vitanmin-free case amino acids (Difco). To measure the total rate of synthesis of RNA, H3-guatiine oir H3-guailosine was added to cultuires as desciibed in t,he Results section. Samples of I ml were re-mnoved and added to 0.5 ml of cold 0.75 211 Lerehloric acid. The cells were collected onmembrane filters (the filt;rate saved f'or isolatioti of GcTP) ani.d washed six timles with 2 ml of 5% tr'ichloiooacetic aci(I containing 100 8t/tml guanine (oI' guan-osime), followed by two washiiigs with 2 ml H20. 'T'he dried samiaples wer e counted in the liquid scintillation spectrometer in a toluene-based cou-rtitmg mixture. Determination of the sp(cifi( activity of the tiitiuin-labeled GTP' pools requires a measurement of the total armounit of G''P p)resenlt in the samples. 'I'his was obtained by adding to the cultures, 1.5 to 25 mnin prior to the beginninig of the experiments, either C14guanine in the case of the experiment with E. coli K-12, or H3P"204 in the case of E. coli NP-29. The GTP was isolated from the perchloric acid extracts of the cells, after addition of 0.2 Amole of unlabeled GTP to serve as carrier. Each sample was absorbed onto