Consequences of Expression of the “Relaxed” Genotype of the RC Gene

Two pairs of strains that dither in the rel-1 locus were compared for their ability to synthesize protein, RNA, and lipid in the presence and absence of a required amino acid. In the absence of a required amino acid, protein synthesis ceased in both relaxed control (RF’) and stringent control (RF*) cells, and RNA synthesis ceased only in the RCF cells. Lipid synthesis continued at an approximately equal rate in the presence or absence of a required amino acid in both RF1 and RF’ cells. Analysis of the lipid extracts showed a decrease in cardiolipin in RCY’ cells deprived of a required amino acid. Otherwise, the distribution of the lipids was not affected by the absence of a required amino acid in either the RC’“’ or RCStr cells and was quite similar in both types of cells. The fatty acid distribution also ap-peared unaffected. These results are consistent with the view that the control of RNA synthesis exercised by the rel-1 locus cannot be simply at the level of nucleoside triphosphate availability.


SUMMARY
Two pairs of strains that dither in the rel-1 locus were compared for their ability to synthesize protein, RNA, and lipid in the presence and absence of a required amino acid. In the absence of a required amino acid, protein synthesis ceased in both relaxed control (RF') and stringent control (RF*) cells, and RNA synthesis ceased only in the RCF cells. Lipid synthesis continued at an approximately equal rate in the presence or absence of a required amino acid in both RF1 and RF' cells. Analysis of the lipid extracts showed a decrease in cardiolipin in RCY' cells deprived of a required amino acid. Otherwise, the distribution of the lipids was not affected by the absence of a required amino acid in either the RC'"' or RCStr cells and was quite similar in both types of cells. The fatty acid distribution also appeared unaffected. These results are consistent with the view that the control of RNA synthesis exercised by the rel-1 locus cannot be simply at the level of nucleoside triphosphate availability.
When amino acid-requiring bacteria are cultured in media lacking necessary amino acids, there is a cessation of net RNA synthesis.
The stringent dependence on amino acids for net RNA synthesis can be relaxed genetically by the presence of the rel-1 locus (l-3) or phenotypically by the presence of chloramphenicol in the growth medium (4,5). It has recently been proposed that the difference in the control of net RNA synthesis may be related to the availability of nucleoside triphosphates (6,7). Several different laboratories have tried to analyze directly the intracellular nucleoside triphosphate pools and to show a correlation between pool size and RNA synthesis (6)(7)(8)(9)(10)(11). At this time, these results appear in conflict with one another, and no clear picture emerges. The recent reports that the regulation of RNA synthesis is noncoordinate (12-15) appear to contradict the possibility that the control is due to nucleoside * This work was supported by a grant from The City University of New York.
$ Recipient of a City University Predoctoral Research Assistantship. triphosphate availability. However, the observation that net lipid synthesis continues in RCrell cells deprived of amino acids but not RCstr cells (16) would certainly be consistent with control by nucleotides or by some other intermediate common to both synthetic pathways.
We have therefore compared two pairs of strains that differ in the rel-1 locus for their ability to synthesize RNA, lipid, and protein in the presence and absence of a required amino acid and have characterized the lipids synthesized by one of the pairs cultured under these conditions.

Bacteria and Culture Conditions
Escherichia coli PA1 (arg-, rel-) and E. coli PA2 (arg-, rel+), an isogenic pair except for the rel locus, belong to the TLBl family of E. coli K-12. They were kindly provided by Dr. R. Lava&& E. coli Tb6 (arg*+, his-, rel-) and E. co& B77 (argAtS, his-, rel+), an isogenic pair except for the arg, and rel loci, are derivatives of E. co& B isolated by Dr. E. Ron. They were generously provided by Dr. B. Davis. The bacteria were cultured in standard Davis medium (17). The growth medium for PA1 and PA2 was supplemented with thiamine-HCl, 0.5 mg per liter, and where indicated with n-arginine, 100 mg per liter.
The growth medium for Tb6 and B77 was supplemented with n-arginine, 100 mg per liter, and where indicated with Lhistidine, 50 mg per liter. A fully supplemented overnight culture was diluted 25-fold into fresh supplemented minimal medium.
Growth was followed on a Klett-Summerson colorimeter. When the turbidity reached 80 to 90 Klett units, the cells were chilled in an ice bath and then harvested at 4' in the Sorvall RC-2B centrifuge in the type SS-34 rotor at 15,000 rpm for 5 min. The cells were washed in 75% of the volume of cold 'Relaxed control of RNA synthesis is indicated by the notation RCF, stringent control by RCF.

856
Consequences of RCrel Express&m Vol.245,No. 4 minimal medium lacking the required amino acid. The cells acetate, 0.09 PCi per ml (specific activity, 17 mCi per mmole) were then resuspended in a volume of cold culture medium as well as the necessary additions for growth. The samples equivalent to 1.25 times the volume of the initial aliquot harwere dried in scintillator vials either by overnight evaporation vested.
or by heating at temperatures below 50". The second method Assay for Protein Synthesis was a more extensive modification of the procedure of Sokawa et al. (16). Aliquots of 2 ml each were removed and mixed with The culture medium was supplemented with nn-isoleucine, chilled unlabeled carrier cells and centrifuged in the cold. 200 pg per ml, and aH-n-isoleucine, 0.1 &i per ml (specific After all the samples had been collected, the pellets were washed activity, 1.0 mCi/0.087 mg), as well as the necessary additions once with 2 ml of cold Davis medium. The washed pellets were for growth.
Incorporation of label into protein was determined then extracted overnight with 4 ml of a 3: 1 chloroform-methanol by a slight modification of the procedure of Byfield and Scher-solution instead of the 1:l chloroform-methanol solution used baum (18). At intervals, 0.10 ml was spotted onto Whatman in the first method. The remainder of the procedure was identi-No. 3MM filter paper discs. The discs were immediately cal with the first method. The radioactivity was determined by placed in 5% trichloracetic acid. The old trichloracetic acid was dissolving in toluene scintillator fluid and counting on the Beckpoured out 3 hour after the last disc was collected, and the discs man LS-150 scintillation counter. were washed for an additional 20 min with fresh 5% trichlor-Lipid Analysis-After the bacteria were incubated for 80 acetic acid. The process was repeated twice more, after which min, the lipids were isolated for analysis by scaling up the the filters were washed twice with acetone. Approximately 10 second method to 40 ml and omitting the addition of unlabeled ml of 5% trichloracetic acid were used for each of the discs in carrier cells. Total lipid extracts were examined by thin layer each wash and approximately half as much acetone was used. chromatography on Silica Gel H (S. Merck, Darmstadt, Ger-The dry discs were counted in toluene scintillator fluid in the many), with chloroform-methanol-water (65: 25 :4). The dis-Beckman model LS-150 scintillation counter. tribution of labeled compounds on the plates was determined by Assay for RNA Synthesis the zonal scanning technique of Snyder (19). In this procedure, the silica is mechanically scraped from the plates into counting The culture medium was supplemented with uracil, 10 pg vials, in 2-mm increments, starting below the origin and proper ml, and aH-6-uracil, 0.1 PC1 per ml (specific activity, 3.1 ceeding through the solvent front. A l-ml sample of 10% Ci per mmole) as well as the necessary additions for growth. glacial acetic acid in absolute ethanol and 15 ml of toluene-base The filter paper disc method described for measuring protein scintillation fluid is added to each vial for counting.  TABLE  I  TABLE   II Total fatty acids in strains PA1 and PA2 after incubation with and without arginine Cells of strains PA1 and PA2, after being incubated on supplemented Davis medium to a turbidity of about 85 Kletteeo units, were divided into three equal portions (40 ml), centrifuged, and washed. Two pellets of each strain were resuspended in 50 ml of medium supplemented with or lacking n-arginine as indicated. These cells were incubated at 37" for 80 min and centrifuged. During this period of time, the turbidity of the cells cultured in the presence of arginine increased by 30 Klett6r.o units. Those cultured in the absence of arginine did not increase in turbidity. The pellets were then extracted with 10.0 ml of chloroform-methanol (3:l) containing 0.50 mg of eicosanoic (C&J acid. Zero time controls were extracted in the same way without incubation. Methyl esters prepared after saponification of the lipid extracts were examined by gas-liquid chromatography, and the combined weight of C&, C17, Cr8, and Cl9 fatty acids was calculated from the relative amount of the COO internal standard. The experiment was performed in duplicate. Distribution of '4C from 14C-&acetate in total lipids Cells of strains PA1 and PA2 cultured in supplemented medium were harvested, resuspended, and incubated for 80 min in media containing r*C-2-acetate and arginine where indicated, as described in the text. After incubation, the total lipid extract was separated by thin layer chromatography, and the distribution of 14C was determined by zonal scanning, as described in the text. The methyl esters were separated by gas chromatography at NO", on a glass column (6 ft. x 2 in) packed with 10% EGSS-X on 100 to 120 mesh Gas Chrom P (Applied Science Laboratories, Inc., State College, Pennsylvania).
Peak areas were measured with a CRS-104 digital integrator (Infotronics Corporation, Houston, Texas).
When radioactivity was to be measured, the effluent gas of the chromatograph was bubbled into a flowing stream of toluene-base scintillation fluid and collected automatically in serial vials at 45-see intervals, as described by Dutton (20). The vials were counted in a scintillation counter and the counts plotted to provide a profile of the distribution of 14C in the methyl esters.

RESULTS
In t'he absence of the required amino acid, arginine, E. coli PA1 continues to synthesize RNA but E. coli PA2 does not (Fig. I). Both E. co& PA1 and PA2 stop making protein under these same conditions (Fig. 2). As expected, similar results were obtained in the absence of histidine when E. coli Tb6 and B77 were studied. Lipid synthesis was not affected by the removal of arginine in E. coli PA1 or PA2 (Fig. 3). Similar results were obtained in the absence of histidine when E. coli Tb6 and B77 were studied.
Extraction of the lipids with either 3 : 1 or 1: 1 chloroform-methanol yielded similar results regarding acetate incorporation.
Total fatty acids of E. coli PA1 and PA2, as measured by gas-liquid chromatography with an internal standard (Table I) Table III, was considered identical within the error of the analysis, and was not significantly altered by the presence or absence of arginine in either E. coli PA1 or PA2.

DISCUSSION
The present studies were designed to investigate whether the rel locus controls lipid synthesis as well as RNA synthesis.
It is clear that RNA synthesis abruptly ceases upon the removal of a required amino acid in RCsti cells but not in RCre' cells. Lipid synthesis continues in both types of cells. The results concerning lipid synthesis are in disagreement with those of Sokawa et al. (16). However, even in their studies, abrupt cessation of lipid synthesis was not observed.
These workers dried their samples by heating under an infrared lamp which we at first thought might account for the differences observed in total lipid synthesis provided that some component of the lipid fraction made in RCstr cells was more volatile than that present in RCre' cells. The lipid analyses do not support this idea because the lipids made by RCStr and RCrel cells in the absence of a required amino acid appear to be nearly the same. Decreased cardiolipin incorporation for the relaxed cells during arginine starvation seems to be the only reproducible difference. We have no explanation for this deviation, and its significance is not apparent.
The observation that lipid synthesis continues in both RCstr and RC'"' cells in the absence of the required amino acid would seem to make it unlikely that the control of RNA synthesis is at the level of nucleoside triphosphate synthesis. This is particularly true for ATP and CTP which are necessary for the continued lipid synthesis observed.
Starvation of an RCstr cell for a required amino acid does not appear to affect the syn-thesis of cyclopropane fatty acids. This must mean that Xadenosylmethionine continues to be synthesized in these cells. RNA synthesis may be controlled by the combined effects of the concentration of several nucleotides including the mono-and diphosphates or be sensitive to the concentration of GTP (11). To date, no consistent picture concerning the role of the rel locus has emerged.
The independence of lipid synthesis from the rel locus is consistent with recent observations concerning noncoordinate control of RNA synthesis (12-15) and the reports that nucleoside triphosphates are not the limiting factors for RNA synthesis (9,10).