VOL. 47, 1961 BIOCHEMISTRY: STENT AND BRENNER 2005

7 Agren, G., 0. Zetterqvist, and M. Orjamae, Acta Chem. Scand., 13, 1047 (1959).8 Engstrbm, L., Acta Soc. Med. Uppsala, 64, 214 (1959); Biochim. et Biophys. Acta, 52, 49

(1961); ibid. (in press).9 Garen, A., and C. Levinthal, Biochim. et Biophys. Acta, 38, 470 (1960).

10 Levinthal, C., in Structure and Functioni of Genetic Elements, Brookhaven Symposia in Biology,No. 12 (1959), p. 76.

11 Garen, A., in Microbial Genetics, Tenth Symposium of the Society for General Microbiology,ed. W. Hayes and R. C. Clowes (Cambridge: Cambridge University Press, 1960), p. 239.

12 Echols, H., A. Garen, S. Garen, and A. Torriani, J. Mpl. Biol., 3, 425 (1961).13 Engstrom, L., Biochim. et Biophys. Acta (in press).14 Nathans, D., and F. Lipmann, these PNOCEEDINGS, 47, 497 (1961).11 Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall, J. Biol. Chem., 193, 265 (1951).16 Moore, S., personal communication.17 Rothman, F., and R. Byrne (in preparation).18 Electrophoresis systems: pH 4.5 and paper chromatography, Dintzis, H. M., these PRO-

CEEDINGS, 47, 247 (1961); pH 6.5 buffer, Ingram, V. M. Biochim. et Biophys. Acta, 28, 539 (1958);pH 1.9 buffer, 88 per'cent formic acid:acetic acid:water (25:87:888), electrophoresis in a tank(Servonuclear Corp., New York) under Varsol (Standard Oil).

19 Jones, M. E., and L. Spector, J. Biol. Chem., 235, 2897 (1960).20 Cohn, M., J. Cell. Comp. Physiol., 54, Sup. 1, 17 (1959).21 Lipmann, F., and L. C. Tuttle, Biochim. et Biophys. Acta, 4, 301 (1950).22 Levinthal, C., personal communication.23 Cohen, J. A., R. A. Oosterbaan, H. S. Jansz, and F. Berends, J. Cell. Comp. Physiol., 54, Sup.

1, 231 (1959).24 Porter, G. R., H. N. Rydon, and J. A. Schofield, Nature, 182, 927 (1958).

A GENETIC LOCUS FOR THE REGULATION OFRIFONUCLEIC ACID SYNTHESIS

BY GUNJTHER S. STENT* AND SYDNEY BRENNER

MEDICAL RESEARCH COUNCIL UNIT FOR MOLECULAR BIOLOGY, CAVENDISH LABORATORY, CAMBRIDGE,ENGLAND

Communicated by'*,. M. Stanley, October 23, 1961

More than 80 per cent of the total ribonucleic acid (RNA) of bacteria. is con-tained in ribosomes, the particles which appear to be the site of protein synthesis inthe cell, and most of the remainder is accounted for by the "soluble" acceptorRNA involved in the activation of amino acids for protein synthesis.1-3 A veryminor fraction of the bacterial RNA is represented by the ephemeral messengermolecules, which, according to recent hypotheses on the mechanism of protein syn,thesis, are the primary gene products of the cistrons of they bacterial deoxyribo-nucleic acid that harbor the sequence information for the ordered copolymerizationof amino acids into specific polypeptides.44 Whereas the formation of messengerRNA, and hence of the quality of protein synthesis, is controlled by various metab-olites that act as* either inducers Qr repressors of the functional expression ofspecific parts of the bacterial genome,4,the formation of ribosomal RNA, andhence of the quantity of protein' synthesis, seems to depend on the availability ofamino acids. Thus, auxotrophic bacterial mutants requiring an amino acid forgrowth stop synthesizing not only.-protein but- (1so bulk, and hence ribosomal,

Dow

nloa

ded

by g

uest

on

Mar

ch 2

2, 2

020

2006 BIOCHEMISTRY: STENT AND BRENNER PROC. N. A. S.

RNA as soon as they are deprived of an exogenous source of the amino acid. How-ever, if synthesis of protein is inhibited in such bacteria, e.g., by addition of theantibiotic chloramphenicol to the growth medium, then a very small concentrationof the required amino acid-far below that otherwise necessary for optimal growth-suffices to promote RNA synthesis at a nearly maximal rate. That is, the re-quirement for amino acids in ribosomal RNA synthesis appears to be catalyticrather than stoichiometric.7' 8An exception to the general finding of stringent amino acid control of RNA

synthesis was discovered by Borek, Rockenbach, and Ryan, who showed that themethionine-requiring (Met,-) mutant 58-161 of Escherichia coli, strain K12, con-tinues to synthesize RNA even after being deprived of all exogenous methio-nine. 9 10 During methionine starvation, the bacteria more than double theirrelative content of RNA and, without losing the ability to give rise to colonies onnutrient agar, undergo a progressive physiological deterioration that delays theresumption of normal anabolic activities after restoration of methionine to thegrowth medium. The reason for the methionine-independent RNA synthesis inthis strain is not that methionine, unlike other amino acids, does not happen toparticipate in the regulation of RNA synthesis, since methionine-requiring auxo-trophs of E. coli other than strain 58-161 do cease net synthesis of RNA as soon asmethionine is removed from the growth medium. Instead, as the work to be re-ported here shows, the exceptional behavior of strain 58-161 reflects its possessionof a chromosomal, genetic character that relaxes the stringency of the amino acidcontrol on RNA synthesis. Amino acid auxotrophs carrying this genetic char-acter synthesize RNA in the absence of not only methionine but also other aminoacids that may be required for growth.

Ribonucleic Acid and Protein Synthesis under Methionine Starvation.-Figure 1presents the results of an experiment in which C"4-uracil has been added to a cul-ture of strain 58-161 immediately after removal of methionine from the growthmedium. It can be seen that, compared to a control culture supplied with methi-onine, the incorporation of the radioactive pyrimidine into the trichloroaceticacid-insoluble fraction of the methionine-starved culture proceeds at about 66per cent of the normal rate during a period of time corresponding to one generationof the control culture and then comes to a gradual halt. Fractionation of theincorporated radioactivity into alkali-soluble and alkali-insoluble portions, accord-ing to the method of Schmidt and Thannhauser,1' showed that at any time ap-proximately 88 per cent of the total radioactivity was incorporated into the al-kali-soluble portion in both methionine-starved and methionine-supplementedcultures, i.e., that the bulk of the radioactivity in both cultures enters ribonucleicrather than deoxyribonucleic acid. Also presented in Figure 1 are the results of asimilar experiment carried out with the methionine auxotroph J53 (Met2-) ofE. coli K12. It can be seen that immediately after removal of methionine from thegrowth medium, incorporation of labeled uracil into the acid-insoluble fraction ofthese bacteria proceeds at less than 10 per cent the rate of incorporation observedwhen methionine is supplied. This experiment thus confirms the previous findingsthat whereas strain 58-161 continues to synthesize RNA after removal of methi-onine from the medium, other methionine auxotrophs do not. Figure 1 showsalso the assimilation of C'4-labeled leucine into the proteins of strain 58-161 under

Dow

nloa

ded

by g

uest

on

Mar

ch 2

2, 2

020

VOL. 47, 1961 BIOCHEMISTRY: STENT AND BRENNER 2007

Ia. b.

/800-Uptam of +Met -- Uptake ofC14-UroCil C14-Leucine

+Met ~~~~~+M~et/200 _MW;

1.0 1.5 2.0 25 1.0 1.5 2.0

Relative increase in Optical Density of fully supplemented aliquot

FIG. 1.-(a) The assimilation of labeled uracil into the acid-insoluble fraction of two differentmethionine auxotrophs, 58161 (Met,-) and J-53 (Met2-), (from the laboratory of W. Hayes) inthe presence and absence of methionine. (b) The uptake of labeled leucine into the acid-insolublefraction of 58-161 (Met,-) in the presence and absence of methionine.

Procedure: Cultures of either strain were growth with aeration in a glucose salts medium supple-mented with 20 pg/ml methionine to a density of 2 X 108 cell/ml. The bacteria were then centri-fuged out of this medium, washed in buffered saline, and resuspended in fresh medium, withoutmethionine but supplemented with either (a) 10 pg/iml uracil and about 0.05 Atc/ml 2-C'4-uracilor (b) 10 ,g/ml leucine and about 0.05 pc/ml C14-leucine. Immediately after their resuspension inthe radioactive medium, the bacteria were divided into two aliquots, methionine was added to oneof these aliquots to a concentration of 20 pg/ml, and both aliquots were incubated with aeration.After various periods of growth, two samples were removed from each aliquot: one sample wasdiluted sixfold into 6 per cent ice-cold trichloroacetic acid, and the other sample was diluted two-fold into 10 per cent formalin. The trichloroacetic acid sample was filtered through an "Oxoid"membrane filter, and the filter, on which the acid-insoluble bacterial precipitate had been de-posited, was placed before a "thin-window" radiation counter. The formalin sample was used forspectrophotometric determination of the optical density at 650 mpA. (N.B. Strain J-53 is actu-ally an amino acid double auxotroph, requiring, besides methionine, also proline for its growth.In the experiment with this strain, 20 pg/ml proline was also present in the growth medium at alltimes.)

methionine starvation; it is evident that as long as it is deprived of methionine,strain 58-161 does not incorporate significant amounts of labeled leucine, showingthat the amino acid independence of the ribonucleic acid synthesis in strain 58-161is not to be attributed to its being a "leaky" methionine auxotroph capable ofcontinuing some protein synthesis in the absence of exogenous methionine.

Multiple Amino Acid Auxotrophs.-In order to test whether RNA synthesis instrain 58-161 can proceed also in the absence of required amino acids other thanmethionine, a number of multiple auxotrophic derivatives of this strain were ex-amined. The results of these tests are presented in Figure 2, where it may be seenthat mutants requiring methionine and leucine, methionine and isoleucine, methi-onine and histidine, and methionine and isoleucine and valine all synthesize RNAafter being deprived of either one or all of the required amino acids. The relativeamount of RNA formed during amino acid starvation, however, depends on whichone of the required amino acids is absent and is always least when none of the re-quired amino acids is furnished.12 In addition to these four strains, an isoleucine-

Dow

nloa

ded

by g

uest

on

Mar

ch 2

2, 2

020

2008 BIOCHEMISTRY: STENT AND BRENNER PROC. N. A. S.

a. +Met b.5000 +Leu 2500_

#W13051 1J-1341

4000_ 2000_

-Met +/Mt(~.3000_ +Leu /500 +5

+ met2000 - LEu/ /000 +Met

-/SO- met

/000 -Lev - 500 /So

I I 7r/s0/.0 2D 30 1.0 2.0 3D

C. +Met+/so+V0/ d.

Fi3 /G-151 +Met+His

/000_ 500- +Met-/ffi -

-Met+His

500 - 250- -Met-His-Met+/so+Vo/Met-/so-Vo/Met-/so-Vol

m; I /' I1.0 2.0 3.0 /.0 2.0 3.0

Rel/tive increase in Optico/Density of fully supplemented aliquotFIG. 2.-The assimilation of labeled uracil into the acid-insoluble fraction of four multiply-

auxotrophic derivatives of the Met, - strain 58-161 in the presence and absence of the requiredamino acids. (a) Strain W1305, requiring methionine and leucine (isolated by E. A. Adelberg);(b) strain J-134, requiring methionine and isoleucine (from the laboratory of W. Hayes); (c)strain G-11, requiring methionine, isoleucine, and valine; (d) strain G-15, requiring methionineand histidine. (G-11 and G-15 were isolated after ethyl methane sulfonate mutagenesis of HfrCavalli.)

Procedure: The bacteria were grown, washed, and resuspended in 2-C'4-uracil-labeled medium,as described under Figure 1, except that here the original growth medium contained also 20 ;&g/mof the other required amino acids. Immediately after their resuspension in radioactive medium,the bacteria were divided into four aliquots; one aliquot was supplemented with 20 ig/ml of bothmethionine and the other required amino acid, one with only methionine, one with only the otherrequired amino acid (in the case of G-11, this aliquot received both isoleucine and valine), and onereceived no further additions. After various periods of growth, samples were removed from eachaliquot and treated as described under Figure 1.

valine-requiring, methionine-independent strain (G-1 1-R1) was examined, inwhich the Metj- locus of the methionine and isoleucine-valine auxotroph G-1 1had been replaced by its Met+ allele through transduction with phage P1. Thisisoleucine-valine-requiring strain continued RNA synthesis during isoleucine-valine starvation in exactly the same manner as the original G-11 double auxo-troph.

These results show that not only methionine but also leucine, isoleucine, valine,and histidine (and, as shall be seen presently, threonine) fail to exert their normalregulatory power in RNA synthesis in the 58-161 strain. The fact that thisaberration persists even after replacement of the Metj- mutant locus by its wild

Dow

nloa

ded

by g

uest

on

Mar

ch 2

2, 2

020

VOL. 47, 1961 BIOCHEMISTRY: STENT AND BRENNER 2009

Met+ allele indicates, furthermore, that the relaxation of the amino acid control ofRNA synthesis is not a direct consequence of the Met,- mutation in the bacterialgenome, a conclusion made even more certain by the results of genetic recombina-tion experiments to be presented in the following.

Transfer of the Regulation Locus by Conjugation.-Strain 58-161 is one of the two"classical" strains employed by Lederberg and Tatum in their original experimentson genetic recombination in bacteria.13' 14 Later work revealed that in theseexperiments 58-161 acted as the F+ donor strain156 16 and that in cultures of strain58-161 F+ there are present Hfr fertility mutants than can transfer with highefficiency parts of their chromosomes to the bacterial zygote formed upon con-jugation with F- recipient cells.17' 18 The two Hfr strains first discovered, HfrHayes18 and Hfr Cavalli,17 as well as a later strain Hfr Reeves1,19 are all methionine-requiring descendants of 58-161. We have, therefore, examined the synthesis ofRNA in these three Hfr strains under conditions of methionine starvation, by meansof an experiment identical to that presented in Figure 1. This experiment showedthat all three strains, Hfr Hayes, Hfr Cavalli, and Hfr Reeves1, behave exactlylike 58-161 F+: after removal of methionine from the growth medium, the bacteriastill more than double their content of RNA. The chromosomal change thatproduced these Hfr fertility mutants has thus not affected the condition respon-sible for the regulatory difference manifested by the F+ parent strain 58-161.

It seemed possible that it is the presence in strain 58-161 of the F fertility epi-some20 itself that is responsible for relaxing the amino acid control of RNA syn-thesis, since F episomes are known to be capable of affecting metabolic functionsof the bacterium.2' Accordingly, we have examined by means of an experimentexactly like that presented in Figure 1, whether the F- variant 58-161 (Spicer)that has lost its F episomel' continues to synthesize RNA during methioninestarvation. This experiment showed that RNA synthesis is as methionine-inde-pendent in the F- variant as in the F+ parent strain. The presence of the fertilityepisome cannot, therefore, be responsible for relaxing the stringency of amino acidcontrol of RNA synthesis in strain 58-161.The other of the two "classical" K12 lines, the F- recipient, is the threonine-

and leucine-requiring strain Y-10, from which multiple mutants, such as W945and W677, have descended.'4' 22 Part of Figure 3 presents an experiment in whichthe incorporation of C'4-uracil into the acid-insoluble fraction of W677 bacteriahas been followed in the presence and absence of threonine and leucine. It isevident from this result that, unlike in strain 58-161, RNA synthesis in strainW677 is under stringent amino acid control, since W677 bacteria incorporatepractically no labeled uracil whatsoever as soon as threonine and leucine are re-moved from the growth medium.

This difference in behavior between Hfr donor and F- recipient strains makes itpossible to investigate whether there exists some genetic locus on the bacterialchromosome that takes part in the amino acid-governed regulation of RNA syn-thesis. For after Hfr and F- cells have been allowed to conjugate, one can readilytest whether the donor property of relaxed amino acid control has been transferredto any auxotrophic recombinants that may issue from the cross. Such bacterialrecombination experiments have been carried out; Figure 4 shows the dispositionon the E. coli chromosome of the various genetic loci used, as well as the point of

Dow

nloa

ded

by g

uest

on

Mar

ch 2

2, 2

020

2010 BIOCHEMISTRY: STENT AND BRENNER PROC. N. A. S.

a. b.6000 6000-600W 6771 _ 6000 lRecombinant T9 Ae ThrLeu

ThrL eu4000 4000

2000- 2000-

no odditions1.0 210 5.0 /.0 2.0 30 4.0 5.0

Relative increase in Optical Density of fully supplemented aliquot

FIG. 3.-The assimilation of labeled uracil into the acid-insoluble fraction of the "classical" F-strain and of a genetic recombinant in the presence and absence of amino acids required for growth.The amino acids supplied to the minimal medium are indicated on each curve. (a) The F- strainW677 requiring threonine and leucine. (b) The methionine-, threonine-, and leucine-requiringrecombinant T9 produced by the cross Hfr Cavalli X W945 (F-). The experimental procedurewas identical to that described under Figure 2.

ALeu2

Meet a lJL ~Lac FIG. 4.-Outer circle: Location on the E. colichromosomal map of the genetic markers em-

Mt/ XGa2 ployed 28 29 (the Gal5 locus has not been preciselyXyl 4_ > y mapped&'). Inner circles: Structure of the three

Hfr donor strains; each arrow shows the pointof origin from and the direction in which eachstrain introduces its chromosome into the recipi-

Str L / / \ \ \ \ ent cell. Abbreviations: Str, streptomycin;Mal, maltose; Xyl, xylose; Mtl, mannitol;Met, methionine; B1, thiamin; Thr, threonine;Ara, arabinose; Leu, leucine; Az, sodium azide;Lac, lactose; Gal, galactose; + and -, abilityand inability respectively to synthesize an amino

\\\Hfr Reevesl / / / acid or vitamin or to ferment a sugar; R and s, re-\\~Hfr Hayes /sistance and sensitivity, respectively, to a drug.

origin from and the direction in which each of the Hfr donor strains introduces itschromosome into the recipient cell. The results of these crosses show that therelaxed control character of the Hfr donor bacterium can be transferred to thebacterial zygote and integrated into the genome of a recombinant cell. As anexample, the second part of Figure 3 presents the assimilation of labeled uracil, inthe absence and presence of required amino acids, by a recombinant produced in across of Hfr Cavalli against W945 (F-). This recombinant is auxotrophic formethionine, threonine, and leucine, having received the genetic loci Lac, Az, Leu,

Dow

nloa

ded

by g

uest

on

Mar

ch 2

2, 2

020

VOL. 47, 1961 BIOCHE.MIISTRY: STENT AND BRENNEER 2011

Ara2, Thr, B1, from the F- parent and the loci Met, Mtl, Xyl, Str, Gal2 from theHfr parent. As can be seen in Figure 3, in the absence of all three required aminoacids, the recombinant bacteria assimilate labeled uracil at about 50 per cent thenormal rate during a period corresponding to one generation of the fully supple-mented culture. Partial supplementation of the culture with one or two of thethree required amino acids can be seen to produce a stimulation of RNA synthesis,though each of the three amino acids behaves in a quantitatively different manner,methionine being the most and leucine the least effective stimulant.The amino acid-independent RNA synthesis in this recombinant shows that

there must exist a chromosomal locus, which we shall call RC, that in its RCrelallelic state of the 58-161 donor line produces less stringent, i.e., "relaxed," aminoacid control of RNA synthesis than its RCSt, or "stringent," allele of the F- re-cipient line.The results and methods of procedure of a number of crosses involving the RC

locus are shown in Table 1. The following conclusions can be drawn from thesecrosses:

(1) Of nineteen different- auxotrophic recombinants descended from zygotesthat received the chromosomal segments Thr-Leu Gal2 - Str from Hfr Hayes, or Leu-Thr - Str - Gal2 from Hfr Cavalli, nine synthesize RNA when starved either forleucine and threonine or for methionine: the others do not. These nine recom-binants, therefore, received the RCrel locus of the Hfr parent, whereas the otherten recombinants retained the RC't locus of the F- parent. The appearance ofthe RCre" character in about half of the recombinants indicates that on the trans-ferred chromosomal segment, the RC locus is not closely linked to any locus usedas a selective marker.23

(2) Not one of 20 different auxotrophic recombinants that received from theHfr donor strains genetic loci (including Met,-) in only that segment that extendsfrom Xyl through Thr, Leu to Gal2 shows RNA synthesis in the absence of re-quired amino acids. Thus, all of these recombinants retained the RCSt locus ofthe F- parent. It is highly probable, therefore, that RC does not lie in the Xyl -Thr - Leu - Gal2 segment and is, instead, situated somewhere in the Xyl - Gal5 - Gal2region.Mechanism of the Regulation.-The intervention of amino acids in the regulation

of RNA synthesis allows the bacterium to adjust its steady-state level of RNA inaccordance with the nutritional limitations imposed on the over-all rate of proteinsynthesis: as long as all of the 20 common amino acids are in abundant supply,the cell can utilize the maximum number of ribosomes to effect protein synthesisat the maximum rate; as soon as the supply of any one amino acid falls to a sub-optimal level, protein synthesis must necessarily decelerate and render super-numerary some of the ribosomal RNA molecules. The "balanced growth" ex-periments of Maal0e and collaborators24' 25 illustrate this point: an immediateacceleration of RNA synthesis follows the transfer of a bacterial culture from aminimal to an amino acid-supplemented medium, and, conversely, an immedi-ate halt of RNA synthesis ensues upon transfer of a culture from a rich to a mini-mal medium.

It is possible to understand the nature of this regulation in terms of recentlydeveloped notions of genetic control mechanisms4 by considering the 20 common

Dow

nloa

ded

by g

uest

on

Mar

ch 2

2, 2

020

2012 BIOCHEMISTRY: STENT AND BRENNER PROC. N. A. S.

0 0 ~~~~~~~~~~~~~~~ ~~~~~~ NO

E-4 0~~~~~~~~~~~~~~~~~~~I).15

0 0930

Q) too~~o~

0~~~~~~~

0.o a)to Q'a)a) -~~~~~~~~~ ~ ~~~ ~~~~~~~~~~~~~~~~~~~oC.RAa~~w 0 .~~~~~aV1. ~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~0C)~ 5 5 ~

z~~~~~~~~~~~- - ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~3 4-t

z .5'o~~~..

C)-~~~~~ ~ ~ ~ ~ ~ ~ ~~~~~~~~~~~~~~~~~~~ 0 .00~~~~~~~~~~~~~~~~~~~~~~~~~~~r.0'.'

Z-~~~~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~ 04 ..

0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~000 O~~~~~~~~~~~~~~~~~~~~~~~~

00D -l0*. oZ;30

03 S-0ir-~~Ia 0.2a)~~~~~~~~~~~~~-~l1. C4-.4-4 0344- '~~~~~~~~~~~~o 0.6.M 4

o0 0~~~~~~~~~~~ 10 ~ ~ -a 0

z t-,) to~0 a)

o- 40O.~ 1~t-O-1-0

0~~~~~~~~~~~~~~~~~~~~~..00 a10 a) G)~~~~~~~~~~~~~~~~~~~~~~4 V" 100

0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~14*.C) 0 ao

a).- ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ O

-4

Dow

nloa

ded

by g

uest

on

Mar

ch 2

2, 2

020

VOL. 47, 1961 BIOCHEMISTRY: STENT AND BRENNER 2013

amino acids as inducers for RNA synthesis. Thus, one would suppose that thereexists within the bacterium a set of twenty species of repressors of RNA synthesis,each repressor species being antagonized or "neutralized" by one particular kindof amino acid. Hence, the lower the intracellular concentration of amino acids,the higher the effective repressor concentration, and hence, the smaller the rateof RNA synthesis. Furthermore, as soon as the concentration of any one aminoacid falls to the very low level that obtains in an auxotroph starved for its growthfactor, the effective concentration of the repressor homologous to that amino acidrises high enough to cut off RNA synthesis altogether. The set of 20 acceptorRNA species, each of which is known to enter into combination with only onekind of amino acid,26 would commend itself as a likely possibility for the repressorspostulated here. This is an image of the regulatory role of amino acids, whichhas been developed also by C. G. Kurland27 on the basis of his observations on theeffect of chloramphenicol on bacterial RNA synthesis.From this point of view, at least two different, alternative functions could be

conceived for the RC locus in the regulation process: (1) RC could be a regulatorgene4 that governs the synthesis of the repressor molecules; the relaxation ofamino acid control engendered by the RCrel allele would then correspond to amutation in this regulator gene that produces either fewer or less effective re-pressors than the stringent control allele RC't. (2) RC would be an operatorgene4 governing the function of an operon that either serves as the direct chromo-somal template for the synthesis of ribosomal RNA or generates an ephemeralcytoplasmic product whose continual replenishment is required for RNA synthesis;the relaxed control allele RCrel would then correspond to a state of the operatorgene that has less affinity for the repressor molecules than the stringent controloperator gene RCst. Under either alternative, starvation for one, two, or threerequired amino acids in an auxotrophic bacterium carrying the RCrel allele wouldnot elevate the effective repressor concentration to a sufficiently high level to cutoff RNA synthesis. Nevertheless, even in RCre' genotypes, RNA synthesismust still subject to some amino acid control, since, as was seen in Figures 2 and 3,partial amino acid supplementation stimulates RNA synthesis in such bacteria.Furthermore, as N. 0. Kjeldgaard27 has found, transfer of the RCrel strain 58-161from a poor to a rich amino acid-supplemented medium still engenders an immediateacceleration of RNA synthesis.Summary.-The methionine-requiring strain 58-161 of E. coli K12 continues to

synthesize ribosomal RNA during methionine starvation, in contrast to aminoacid auxotrophs of other E. coli lines, which stop such synthesis as soon as they aredeprived of the required amino acid. Study of multiple auxotrophic derivativesof strain 58-161 requiring additional amino acids for growth revealed that in thisstrain the normally stringent control of ribosomal RNA synthesis by amino acids isrelaxed, 'is a' vis not only methionine but also other amino acids. Conjugationexperiments in which Hfr variants of strain 58-161 were crossed to F- bacteriasubject to normal, stringent amino acid control showed that there exists a locus,RC, on the bacterial chromosome that in its "relaxed" RCrel allelic state of the58-161 line produces less stringent amino acid control of ribosomal RNA synthesisthan its normal, "stringent" RCSt allele. The RC locus could be either a regulatorgene concerning the synthesis of repressors of RNA synthesis or an operator gene

Dow

nloa

ded

by g

uest

on

Mar

ch 2

2, 2

020

2014 BIOCHEMISTRY: STENT AND BRENNER PROC. N. A. S.

governing the function of an operon that serves as the direct chromosomal tem-plate of ribosomal RNA or generates an ephemeral cytoplasmic product whosecontinual replenishment is required for RNA synthesis.We are greatly indebted to W. Hayes and R. C. Clowes for good advice and for supplying us

with most of the bacterial strains with which this work has been carried out.* Senior Postdoctoral Fellow of the National Science Foundation. Permanent address: Virus

Laboratory, University of California, Berkeley.I Schachman, H. K., A. B. Pardee, and R. Y. Stanier, Arch. Biochem. and Biophys., 28, 245

(1952).2 McQuillen, K., R. B. Roberts, and R. J. Britten, these PROCEEDINGS, 45, 1437 (1957).3 Lacks, S., and F. Gros, J. Mol. Biol., 1, 301 (1959).4 Jacob, F., and J. Monod, J. Mol. Bio!., 3, 318 (1961).6 Brenner, S., F. Jacob, and M. Meselson, Nature, 190, 576 (1961).6 Gros, F., H. Hiatt, W. Gilbert, C. G. Kurland, R. W. Riseborough, and J. D. Watson, Nature,

190, 581 (1961).7 Gros, F., and F. Gros, Biochim. et Biophys. Acta, 22, 200 (1956); Exptl. Cell Res., 14, 104

(1958).8 Pardee, A. B., and L. Prestidge, J. Bacteriol., 71, 677 (1956).9 Borek, E., A. Ryan, and J. Rockenbach, J. Bacteriol., 69, 460 (1955).

10 Borek, E., J. Rockenbach, and A. Ryan, J. Bacteriol., 71, 318 (1955).11 Schmidt, G., and S. J. Thannhauser, J. Biol. Chem., 161, 83 (1945).12 Pardee and Prestidge (ref. 8) had already studied RNA synthesis in the methionine-leucine

double auxotroph W1305 and found less RNA synthesis under leucine starvation than undermethionine starvation. From this they inferred that the aberrant behavior of strain 58-161 maybe confined to methionine, but even their results suggest that in the absence of leucine, strainW1305 synthesizes more RNA than the "normal" amino acid-starved auxotrophs investigatedby them.

13 Lederberg, J., and E. L. Tatum, in Heredity and Variation in Microorganisms, Cold SpringHarbor Symposia on Quantitative Biology, vol. 11 (1946), p. 113.

14 Lederberg, J., Genetics, 32, 505 (1947).15 Hayes, W., J. Gen. Microbicl., 8, 72 (1953).16 Cavalli, L. L., J. Lederberg, and E. M. Lederberg, J. Gen. Microbiol., 8, 89 (1953).17 Cavalli, L. L., Boll. It. Sierotera. Mfilano, 29, 1 (1950).18 Hayes, W., in Viruses, Cold Spring Harbor Symposia on Quantitative Biology, vol. 18 (1953),

p. 75.19 Reeves, P., Ph.D. thesis, University of London (1959).20 Jacob, F., and E. L. Woilman, C. R. Acad. Sci. (Paris), 247, 154 (1958).21 Jacob, F., and E. A. Adelberg, C. R. Acad. Sci. (Paris), 249, 189 (1959).22 Lederberg, J., J. Bacteriol., 59, 211 (1950).23 Wollman, E. L., and F. Jacob, La Sexualitg des Bacterigs (Paris: Masson et Cie., 1959).24 Schaechter, M., 0. Maal0e, and N. 0. Kjeldgaard, J. Gen. Microbiol., 19, 529 (1958).-26 Kjeldgaard, N. O., 0. Maal0e, and M. Schaechter, J. Gen. Microbiol., 19, 607 (1958).26 Berg, P., and E. J. Ofengand, these PROCEEDINGS, 44, 78 (1958).27 Personal communication.28 Jacob, F., and E. L. Wollman, in Recent Progress in Microbiology, ed. G. Tunevall (Stockholm:

Almqvist och Wiksell, 1958).29 Taylor, A. L., and E. A. Adelberg, Genetics, 45, 1233 (1960).3 Lederberg, E., in Microbial Genetics, Tenth Symposium of the Society for General Micro-

biology, ed. W. Hayes and R. C. Clowes (Cambridge University Press, 1960).

Dow

nloa

ded

by g

uest

on

Mar

ch 2

2, 2

020