Expression of the histidine ( his ) operon in Salmonella typhimurium is exerted predominantly at the level of mRNA production ( 22 ) and appears to involve two major control sites , the primary promoter ( P1 ) and a transcriptional barrier called the attenuator ( Atn ) ( 3 , 22 ) .
Thus , overall operon expression involves both the frequency of RNA chain initiations at P1 and the frequency ofpremature RNA chain terminations at Atn ( Fig. 1 ) .
Data obtained from in-vivo and in-vitro experiments indicate that mutation hisO2355 at least in part defines the P1 site and deletion hisO1242 defines the Atn site ( 13 , 15 , 22 ) .
The model also hypothesizes the existence of a positive factor ( 46 ) , acting to relieve the transcriptional terminations otherwise occurring at the attenuator ( anti-Atn ) .
Three mutational sites , hisO2965 , hisO2966 , and hisO3148 , seem to affect relief of attenuation ( 13 , 22 ) ; thus , the site of positive-factor interaction is represented by the broad bracket labeled `` Anti-Atn '' in Fig. 1 .
Suggestive evidence implicates a conformer of the histidyl-tRNA synthetase ( HRS in Fig. 1 ) as the positive factor ( 46 ) .
It is also possible that translation , possibly of a portion of a his mRNA `` leader '' sequence extending from P1 to Atn , may play a role in the activation process ( 3 , 4 ) .
his operon regulation in Escherichia coli appears to proceed by mechanisms identical to those effective in S. typhimurium ( 8 , 18 , 24 , 28 ) .
t Contribution no. 931 of the Department of Biology , Johns Hopkins University .
8 There is no direct evidence that a classical system of negative control ( repression ) is involved in his operon regulation ( 3 , 13 , 15 , 22 ) .
Two sites of unknown function ( sites his 01828 and hisO3155 ) could , however , be involved in a negative-control system .
The hisG gene product , which has been suggested as playing a role in negative control of the his operon ( 20 ) , clearly is not an obligatory , autogenous repressor ( 36 ) .
Histidine input into the operon-attenuated control system depends strictly upon the concentration of charged , pseudouridine-modified tRNAHIS and not on total tRNAHiS or on free histidine ( 7 , 26 ) .
The regulatory system is essentially histidine independent in the absence of pseudouridine modifications in the tRNAH '' ' anticodon loop ( 7 , 26 ) , modifications mediated by the hisT gene product ( 8 , 11 , 12 , 24 , 37 ) .
Thus , while the unmodified tRNAH ' i apparently is un-impaired in protein synthesis and is charged normally with histidine ( 26 ) , it appears relatively inefficient in histidine regulation .
In Fig. 1 we have depicted a role for effector tRNA in conversion of positive factor to a state inactive in stimulation of his operon expression ( 3 , 22 ) .
In this model , operon-attenuated control in a hisT background should depend only on the absolute intracellular concentration of positive factor .
In addition to histidine-specific regulation , the his operon responds to metabolic regulation in the sense that enzyme levels are lower in amino-acid-rich media than in minimal salts + glucose + histidine medium ( 7 , 39 ) .
Stephens et al. pro ¬ H i s-t RNA ( , , ( AFTER ELY , KASAI , ARTZ , BROACH ) pose that the relAU gene-mediated product , gua-nosine 3 ' ,5 ' - bis ( diphosphate ) ( MSI or ppGpp ) , serves as the effector of metabolic regulation at the level of transcription ( 39 ) .
They conclude that : ( i ) ppGpp is required for maximal his op-eron expression in-vitro , particularly when S-30 extracts from a relaxed ( reA ) strain are used ; ( ii ) in-vitro , ppGpp stimulates expression of at-tenuator-deleted ( his01242 ) DNA about as much as wild-type ( hisO + ) DNA , suggesting that ppGpp does not act at the Atn site ; ( iii ) the ppGpp stimulation in-vitro is on transcription and not on translation ; ( iv ) his operon expression is defective in-vivo in a relA strain starved for and operon histidine ; ( v ) his transcription is maximally stimulated at lower than the maximum intracellular ppGpp concentration .
Experiments reported in this paper examine several aspects of the regulatory model ( Fig. 1 ) .
We further characterize the specificity , mechanism , and site of his operon metabolic regulation in-vivo .
Our results consolidate and extend previous conclusions ( 3 , 22 , 39 ) .
MATERLALS AND METHODS Bacterial strains .
All strains were derivatives of S. typhimurium LT2 ( Table 1 ) .
Most strains were constructed by P22 transduction to histidine prototro phy of strains TA471 and TR127 , assuring relatively isogenic backgrounds .
P22 transduction was mediated by int4 and int4 cly2 phages ( 16 ) or by HT int phage obtained from J. R. Roth .
The media used included : broth Media .
( i ) tryptone ( TB ) ( 1 % [ wt/vol ] tryptone [ Difco ] + 0.5 % [ wt/vol ] NaCl ) ; ( ii ) Vogel-Bonner minimal E medium ( 44 ) ; ( iii ) Cold Spring Harbor minimal A medium containing , per liter , 1 g of ( NH4 ) 2SO4 , 10.5 g of K2HPO4 , 4.5 g of KH2PO4 , and 0.1 g of MgSO4 7H20 ; ( iv ) modified Cohen-Rickenberg medium ( CR1/2P ) ( 6 ) ; and ( v ) modified Kaempfer-Magasanik medium ( KM ) , containing 1 mM K2HPO4 instead of 0.1 mM K2HPO4 ( 21 ) .
Concentrations of primary carbon sources , amino are noted with individual acids , and vitamins experiments .
CAA was 0.2 % ( wt/vol ) Casamino Acids ( Difco ) .
Concentrated ( 15 % , wt/vol ) CAA and single-strength A medium tended to precipitate slightly upon storage at room temperature .
Precipitated solutions were discarded , and fresh solutions were prepared .
The relatively low content of salt in CR1/2P medium contained insufficient micronutrients ; however , dilution of 40 % ( wt/vol ) glucose stock solution prepared in singly deionized water provided sufficient micronutrients for optimal growth .
For : 12 p labeling , overnight cultures were prepared in CR1/2P to which a final concentration of 0.2 % ( wt/vol ) glucose in singly deionized water was added .
A sample of 0.05 ml of this culture was used to inoculate the growth flask , whic contained 5.0 ml of CR1/2P supplemented with 0.2 % ( wt/vol ) glucose prepared from ultrapure water .
Bacteria were pregrown with aeration at 37 °C overnight in the particular steady-state medium .
A small sample of this culture was used to inoculate fresh , prewarmed medium contained in a 300-nil Klett flask .
The bacteria were grown at 37 ± 1 °C with vigorous reciprocal shaking .
Growth was monitored on a Klett-Summerson color-imeter ( red filter , 660 nm ) .
Samples for HisB and Gnd ( see below ) enzyme assays were removed from cultures that had clearly been in exponential-growth for several mass doublings .
Cultures continued exponential-growth after samples were removed .
At least two samples were removed from a given culture at well-separated times ( about 0.25 to 1 h , depending upon growth-rate ) .
Each sample was assayed in duplicate .
Samples from separate growth points had the same enzyme specific activities within ± 10 % .
Care , and several repetitions , guaranteed that slowly growing cultures were in steady-state growth when assayed .
Bacteria were grown under steady-state conditions as described above .
The amino-acid-rich preshift media , noted in individual experiments , supported exponential-growth well above a Klett reading of 40 units .
At a Klett reading of about 40 units , a sample for the HisB or Gnd enzyme assay was taken and defined as t = 0 .
Simultaneously , the remainder of the culture was transferred under sterile conditions to a centrifuge tube and sedimented at 7,700 x g for 10 min at 5 °C .
The culture temperature remained greater than 15 °C over this time period .
The pellet was carefully washed once at 25 °C with the postshift medium , and then 10 ml of postshift medium at 37 °C was added .
The bacteria were briefly stirred into suspension with a Vortex mixer , and another sample was often taken for enzyme assay .
The resuspended bacteria were quickly poured into a Klett flask containing postshift medium prewarmed to 37 °C .
The flask was shaken vigorously at 37 °C , and growth and enzyme activities were monitored .
HisB ( histidinolphosphate phosphatase ; EC 3.1.3.15 ) enzyme activity was used to measure his operon expression .
Specific activities were calculated from corrected initial velocity data for assays on toluenized cells by the method of Ely ( 13 ) .
The TEA assay buffer was 0.1 M triethanola-mine-1.5 mM MgC12 , adjusted to pH 7.5 at 25 °C .
Between 1 and 15 ml of cell suspension was collected by filtration onto membrane filters ( 0.45-pm pore size , HAWP ; Millipore Corp. ) prewetted with TEA buffer .
Filters were rinsed with six 1.5-ml samples of ice-cold TEA buffer , and the filters were immediately suspended in 3 ml of ice-cold TEA buffer .
` Bacteria were suspended by agitation with a Vortex mixer , and the suspension was placed at -15 °C for storage .
Neither freezing nor storage over a 3-day period affected enzyme activity , and enzyme activity was not appreciably affected by treatment of the bacterial suspension with a chloramphenicol-azide mixture before filtration .
Th suspensions were thawed and diluted in TEA buffer to obtain absorbancy at 650 nm ( Ao60 ) readings ofabout 0.5 .
Toluene ( 100 1d ) was added to 1.5 ml of thawed suspension , and the assay was performed essentially as described by Ely ( 13 ) .
A 5-min preincubation at 37 °C preceded substrate addition .
Enzyme activity was stable during preincubation .
Each step of the assay was precisely timed , and assay tubes were read immediately after color development at 45 °C .
In cases where toluenized cell suspensions contained high enzyme activities , samples were diluted for assay , values were corrected to initial velocities , and the corrected values were multiplied by the dilution factor .
Different dilutions yielded identical specific activities .
Specific activities are expressed in units per milliliter per A6w for a 15-min assay as defined by Ely ( 13 ) .
The standard deviation of the assays was generally less than 15 % over the entire range of activities .
That Am60 was a true measure of protein content ( see footnote to Table 1 in reference 33 ) was substantiated by controls ( see Results ) .
Gnd ( 6-phosphogluconate dehydrogenase ; EC 1.1.1.43 ) enzyme activity was used to measure gnd gene expression ( 31 ) .
The TEA-Gnd assay buffer consisted of 0.2 M triethanolamine + 1 mM MgCl2 , adjusted to pH 8.3 at 25 °C .
Stock solutions of color-reagents were prepared in the dark ( 46 ) , and assays were performed with room lights off to minimize photoreduction of the dye mix .
Nicotinamide adenine dinucleotide phosphate ( NADP ) stock solution was prepared in water at 4 °C , adjusted to pH 7.5 , and frozen .
Gelatin stock solutions were prepared weekly in warm water and then refrigerated .
Triso-dium 6-phosphogluconate ( Sigma Chemical Co. ) was dissolved in water at 4 °C , adjusted to pH 8.5 , and frozen .
Bacteria were harvested and toluenized in TEA-Gnd buffer as described above .
Assays and A , ; & ) readings were performed on toluenized cell suspensions .
A 10 % decrease in specific activity occurred after overnight storage at -15 °C .
In addition , Gnd activity in toluenized cell suspensions decreased about 35 % during a 5-min preincubation at 37 °C in the absence of substrate .
Therefore , thawed suspensions were toluene treated for exactly 20 min and chilled , and assay tubes were charged with enzyme in the cold .
Freshly prepared NADP-gelatin-dye mixture was added , followed by substrate addition .
The reaction was initiated by immediate transfer to 37 °C .
Generally , six or more time points were assayed for each toluene-treated sample , and a blank lacking substrate prepared for each time point .
Activity moniwas was tored at 520 nm ( 39 ) immediately upon completion of the assay .
The kinetics of product formation were strictly linear for at least 45 min after a variable lag of about 1 min , probably due to temperature equilibration .
Substrate saturation was achieved in all cases .
Specific activities are expressed as A ( A52o per milliliter per Ao60 ) per minute .
Contents of ppGpp and pppGpp were determined by established procedures ( 25 ) .
A 25-ml Klett flask containing 5 ml of CR1/2P + 0.2 % ( wt/vol ) glucose + supplements , as noted , was inoculated to a barely detectable level with bacteria pregrown on medium of the same composition .
H , ) ` PO4 was added after bacterial mass had increased exponentially to 5 to 10 Klett units ( about 2.5 x 107 to 5 x 107 bacteria/mi ) .
The carrier-free H : PPO4 in 0.02 N HCI ( New England Nuclear Corp. ) showed no polyphosphate contamination in tests by direct spotting or Norit treatment of formate-ex-tracted bacteria ( 10 ) .
After more than one mass doubling had occurred in ' 2P-labeling medium , samples of bacteria were removed for formate extraction .
Four to eight samples were taken at 10-to 15-min intervals in each experiment .
One-dimensional thin-layer chromatography on polyethyleneimine-cellulose sheets ( Baker-flex ) was carried out in 1.5 M NaH2PO4 buffer , pH 3.4 ( 32 ) .
A region between the pppGpp and ppGpp spots of an area approximately equal to these spots served as a background blank .
Counting efficiency was 100 % , and specific activity was calculated assuming a total phosphate concentration of 0.5 mM ( 25 ) .
Isotope concentration was checked for each culture by direct counting .
Data are expressed in nanomoles of phosphate per milliliter per Klett unit .
RESULTS Correlation of ppGpp concentration and histidine enzyme levels .
Implication of ppGpp as the primary effector of his operon metabolic regulation in-vivo has relied on a comparison of his enzyme levels in S. typhimurium grown on high-phosphate-containing nutrient broth and E medium versus published levels for ppGpp in E. coli grown in low-phosphate 32P-labeling me-dium ( 39 ) .
However , some reIAU E. coli strains show only about a 1.2-to 1.3-fold-higher ppGpp concentration in cells grown on a minimal me-dium than those grown on amino-acid-rich me-dium ( 25 ) .
Therefore , we measured directly in S. typhinuriwn concentrations of ppGpp , and his enzyme levels in bacteria grown pppGpp , on several media .
HisB activity did , indeed , correlate with ppGpp concentration .
Furthermore , this correlation was found in strain SB3436 , which carries the hisT1504 mutation and presumably is uncoupled in operon-attenuated control ( Table 3 ) .
This observation is consistent with involvement of ppGpp primarily in RNA chain initiation rather than in attenuation .
The intracellular concentration of ppGpp in cells grown on minimal + glucose medium ( Table 2 ) was estimated as 0.16 mM , using 4.7 X 106 cells/ml per Klett unit and an assumed cellular volume of 10-15 liters/cell ( 1 , um3 ) for the calculations .
his enzyme levels could vary over an almost twofold range in various minimal media with glucose as sole carbon source , although growth-rates did not vary widely ( Table 2 ) .
In particular , about the KM 32P-labeling medium allowed only half the his operon expression of the commonly used A medium .
Comparable HisB enzyme levels in CR1/2P and E media suggest that the ppGpp-pppGpp levels measured in CR1/2P me Medium Medium dium are likely representative of the E-medium levels .
Mutation in gene strB , which increases his operon expression approximately threefold ( 46 ) , had only a small effect on the ppGpp level ( Table 2 ) .
Finally , the ratio of ppGpp to pppGpp was dependent on both strain and growth me-dium .
Downshift elicits metabolic regulation in reIA + bacteria .
Two hisO + strains carrying the hisT1504 allele , and thus uncoupled for operon-attenuated control , were examined in downshift experiments for metabolic regulation of enzyme synthesis .
One strain ( SB3436 ) was relA + , and the second strain ( SB3437 ) was relA .
Bacteria were grown on very rich-medium ( TB + 0.2 % CAA + 0.2 % glucose ) and then downshifted to minimal E medium + 0.2 % glucose .
Under the conditions of this severe transition in medium composition , growth proceeded in the relA + strain after a lag ( Fig. 2A ) .
Concomitantly , histidine enzyme levels increased 3.7-fold in the relA4 strain ( Fig. 2B ) .
This increased operon expression was histidine independent , a result consistent with a lack of operon-atten-uated control .
The same basic result was found in replicate experiments with three different sin-gle-colony isolates of strain SB3436 .
Enrichment for six amino-acids ( legend to Fig. 2 ) reduced the increase in operon expression ( Fig. 2B ) .
This reduction is consistent with a relationship between amino-acid supply and metabolic regulation .
Constitutive Gnd ( 31 ) specific activity was not stimulated after downshift in the relA + strain ( Fig. 2 ) , nor was carnosinase specific activity ( pepD gene product ) increased ( data not shown ) .
These results establish a specificity in the metabolic regulation of the his operon .
After extreme downshift , the reA strain failed to resume growth ( Fig. 2A ) or to increase histidine enzyme levels ( Fig. 2B ) .
In subsequent tests , the downshift illustrated in Fig. 2 was adapted as a standard method for examination of his operon metabolic regulation in reUA bacteria .
P1 and P2 as sites of metabolic regulation .
To gain information pertaining to the site ( s ) at which metabolic regulation was exerted , we surveyed in downshift experiments a number of operator-promoter ( hisO ) point and deletion mutant strains .
The combined evidence indicated that metabolic regulation was achieved principally by enhancement of mRNA chain initiations involving specific DNA sequences .
These included the primary ( P1 ) and internal ( P2 ) promoters of the histidine operon ( Fig. 1 ) .
Entries 1 and 2 in Table 3 summarize the data from the downshift experiment depicted in Fig. 2 .
All other strains reported in Table 3 , except strain TA520 ( see below ) , exhibited the same growth recovery and time course of his operon expression as did the strains depicted in Fig. 2 .
The coupled entries 1,2 ; 5,6 ; and 13,14 in Table 3 illustrate apparent dependence of his operon expression on the reUA + gene product ( see Discussion ) .
Entries 1 and 3 show that the same degree of metabolic regulation was exerted in an atten-uator-deleted strain ( hisO1242 ) as in a hisO + hisT1504 mutant strain .
The metabolic regulation of the attenuator-deleted strain was histidine independent ( Table 3 , entries 3a and 3b ) .
This independence of metabolic control from the presence of the attenuator DNA sequence also can be seen in a number of additional entries in Table 3 .
These include entry 19 , in which it also was demonstrated that deletion of the entire hisG gene did not affect metabolic regulation .
Deletion of the primary ( P1 ) promoter in strains carrying hisO2321 ( Fig. 1 ) in the presence of the attenuator sequence allowed expression only of a promoter ( P2 ) internal to the histidin VOL .
Entry 4 shows that transcriptions from the internal promoter were subject to met-abolic regulation , although the kinetics of enzyme increases following downshift were novel , ( footnote e , Table 3 ) .
Additional deletion in hisO2321 of the attenuator sequence , hisO1242 , removed all sites implicated in operon-atten-uated control ( Fig. 1 ) and permitted constitutive his operon expression from a weak promoter .
Variation in low-level operon expression in strains carrying different hisO mutations ( Table 3 , entries 5 , 9 , and 11 ) might suggest that this weak promotion was not the result of read-through from upstream genes .
Possibly this weak promoter was located in what remained of the hisO region .
Entries 5 and 6 show that this low-level promotion was subject to meta-bolic regulation .
Deletion of the entire operator-promoter region ( hisOG1302 ) led to constitutive production of the remaining histidine enzymes by read-through from a highly active upstream promoter of unknown function ( 1 ) which was not under metabolic regulation ( entry 7 ) .
Enzyme increase upon downshift was small and of the same order as found in relA strains ( entries 2 and 14 ) and for the constitutive enzyme Gnd ( Fig. 2 ) .
Entry 8 shows results obtained with the hisOG203 deletion ( only P2 present ) in combination with a base-substitution mutation that created a new transcriptional promoter within the hisG gene ( 35 ) .
No metabolic regulation occurred at the new promoter site .
Entries 7 and 8 again illustrate the specificity of metabolic regulation and support the view that metabolic regulation involves mRNA synthesis rather than his mRNA stability .
Mutant hisO2355 carries a single base substitution in the primary promoter region and has been assumed to affect the RNA polymerase binding or entry site ( 13 , 22 ) .
This strain was subject to an exceptionally high degree of met-abolic regulation ( entry 9 ) , indicating a strong relationship between metabolic regulation and initiation specificity .
In addition , a similarly high response was found for two additional mutants , hisO2965 and hisO2966 .
Properties of these two mutants have indicated that relief of attenuation can not occur , although transcription initiation can proceed normally ( 13 , 22 ) .
These two mutations are clustered on either side of the primary promoter ( his02355 site in Fig. 1 ) .
We surmise that met-abolic regulation affects both initiation specific-ity and positive-factor affinity .
In this view , ppGpp induces a change in RNA polymerase that affects both promoter-site and positive-fac-tor recognition .
Consistent with this interpretation is the observation that combination of ATION OF THE HISTIDINE OPERON 837 hisO2966 with the Atn deletion , hisO1242 , thus eliminating a requirement for positive factor , allowed normal enzyme increase after downshift ( entry 12 ) .
Another type of anti-Atn-deficient mutation ( strains containing hisO3148 in Table 3 ) did not show this high metabolic response .
Entries in the remainder of Table 3 summarize metabolic regulation for characteristic single and double mutants from the other hisO classes ( Fig. 1 ) .
These strains all showed the same relative increase in his enzyme levels during downshift as did the isogenic hisO + strain ( entry 1 ) , although an extremely wide range of his operon expression was encountered in the various strains .
Limitation of amino-acids other than histidine elicits metabolic regulation .
As shown in Fig. 2 and Table 3 ( entries 1 and 3 ) , his operon metabolic regulation between steady states was histidine independent ( see reference 7 ) .
That is , the presence or the absence of excess histidine had no effect on the degree of meta-bolic regulation in strains uncoupled for operon-attenuated control .
This apparent independence from histidine and from attenuation is consistent with the involvement of a general effector , like ppGpp , as the cause of metabolic regulation .
We would then predict that starvation of relA + bacteria for amino-acids other than histidine should cause increased operon expression , even in the presence of excess histidine and of operon-atten-uated control .
The his + strain SB257 , carrying a leaky aroD mu ` tation , was first grown in minimal + glucose + 20 amino-acids medium and then shifted to the same medium lacking the three aromatic-amino-acids .
The his enzyme levels increased significantly upon starvation for the aromatic-amino-acids even though excess histidine was present throughout the experiment ( Fig. 3B ) .
In contrast , no change in specific activity was noted for the constitutive enzyme ( 31 ) , Gnd ( Fig. 3B ) .
The data indicated that operon expression in both steady-state and starved conditions was in part regulated by availability of amino-acids other than histidine .
Since strain SB257 was his + , indirect effects of amino-acid starvation on the level of charged histidyl-tRNA could not be excluded entirely .
The increased operon expression for strain SB257 ( Fig. 3 ) , however , did fullfill a necessary consistency with predictions derived from steady-state , attenuator-un-coupled metabolic regulation ( Fig. 2 and Table 3 ) .
Metabolic regulation during steady-state growth .
It has been proposed that metabolic regulation of his operon expression is maximal at a ppGpp concentration characteristic of bacteria grown in minimal + glucose medium ( 39 ) In E. coli the intracellular ppGpp concentration varies inversely with growth-rate on different media ( 25 , 38 ) .
Stringent control in E. coli and S. typhimurium appear to be phenomenologically identical ( 27 , 39 ; Table 2 ) .
Therefore , we measured his operon expression as a function of steady-state growth-rate to further characterize his metabolic regulation and the role of ppGpp in this regulation ( Table 4 , Fig. 4 ) .
The data indicate that his metabolic regulation does indeed saturate at a growth-rate characteristic of bacteria grown in minimal + glucose medium ( i.e. , at about one doubling per hour ) .
In addition , an anomalous effect of growth in alanine medium suggests that his metabolic regulation can involve medium-dependent effects on antiattenuation under special circumstances of culture .
Specific activity of constitutive Gnd did not vary with growth-rate in S. typhimurium ( strain SB3436 , Table 4 ; Fig. 4 ) .
This observation has been confirmed by an entirely different enzyme assay procedure utilizing sonically treated cell extracts ( Richard E. Wolf , Jr. , personal communication ) .
his enzyme levels did not vary with growth-rate in mutant hisOG1302 , deleted for the entire hisO region and subject to read-through from an upstream promoter ( strain SB2046 , Table 4 ; Fig. 4 ) .
This observation substantiated the conclusion that promotion in hisOG1302 was not under metabolic control ( Table 3 , entry 7 ) .
In contrast , his enzyme levels increased with decreasing growth-rate up to a saturation level achieved at about one doubling per hour ( strain TA1003 with his attenuator deleted , Table 4 ; Fig. 4 ) .
There was no further relative increase in enzyme levels for strain TA1003 bacteria grown at slower rates ( serine , alanine , or alanine + histidine as primary carbon sources ) .
Internal ( P2 ) promotions ( strain TA520 , Table 4 ; Fig. 4 ) behaved very similarly to the unattenuated , primary ( P1 ) promotions .
Strains containing hisO + hisT1504 behaved similarly to the unattenuated P1 and P2 promotions in all media except alanine medium ( strains SB3436 and SB3437 , Table 4 ; Fig. 4 ) .
Although one of these strains was reA , ppGpp concentrations do not vary appreciably in isogenic relA + and reA strains under steady-state growth-conditions in various media ( 25 , 38 , 39 ) .
The constant , maximal increase in his enzyme levels in strains uncoupled for op-eron-attenuated control is interpreted as reflecting a maxmum level of metabolically induced mRNA chain initiations .
When grown in alanine medium , hisO 0.1 02 03 0.4 0.5 0.6 0.7 0 0.9 1.0 RELATIVE DOUBLINGS PER HOUR FIG. 4 .
Relative HisE and Gnd enzyme levels as a function of relative steady-state growth-rate .
The data are from Table 4 , and numbers designate the entry number in the table .
Different growth-rates were generated by variation in carbon source primary and nutrient enrichment .
For a given strain , spe-in cific activities and growth-rates are relative to those found during-growth in TB + 0.2 % glucose + 0.2 % CAA medium .
HisB specific activities : ( 0 ) SB3436 in A medium ; ( 0 ) SB3436 in E medium ; ( x ) SB3437 in A or E medium ; ( A ) TA520 in A medium ( P2 promotions ) ; ( A ) TA1003 ( deletion of attenuator sequence ) in A medium ; ( U ) SB2046 in A or E medium .
Gnd specific activities are for SB3436 in A or E medium ( i ) .
tion of carnosine to media already containing histidine might influence levels by enzyme an effect , for example , on expansion of the internal histidine pool and a higher steady-state level of charged tRNA ' 5 .
Steady-state enzyme levels in two his + strains grown in minimal , histidine , and histidine + carnosine media ( Table 5 ) showed strain-specific differences in basal his enzyme levels , as noted for some of our strains previously ( R. G. Martin , personal communication ; P. E. Hartman , unpublished data ) .
However , addition of carnosine at the concentration found in nutrient broth did not significantly affect enzyme levels in either histidine-grown strain .
Therefore , exogenous Lhistidine alone acts maximally , and it seems likely that low enzyme levels found in nutrient broth medium , in fact , are another reflection of metabolic regulation .
This conclusion is supported by entries 1 and 8 in Table 4 ; hisT1504 bacteria grown in TB + CAA medium exhibited an enzyme level about one-half that of cells grown in a less rich but otherwise comparable medium .
TB + CAA medium is not expected to contain appreciable carnosine .
DISCUSSION A number of conclusions pertaining to meta-bolic regulation of his operon expression in S. typhimurium are suggested by the results reported above .
In our limited sampling , there is a close correlation in-vivo between his operon expression and intracellular ppGpp level ( Table 2 ) , lending support to the view that ppGpp is an effector of his operon metabolic regulation ( 39 ) .
Previously , this correlation had not been directly measured .
The intracellular concentration of ppGpp in cells grown on CR1/2P + 0.2 % glucose medium was 0.16 mM ( Table 2 ) , i.e. , twice the value reported for Salnonella grown in KM + 0.4 % glucose medium ( J. C. Stephens , Ph.D. thesis , University of California , Berkeley , 1976 ) .
As noted above , his operon expression in the former medium was higher than in the latter medium ( Table 2 ) .
A strain carrying a mutation in gene strB possessed a slightly higher ppGpp level ( Table 2 ) .
Mutations in strB cause other pleiotropic effects , including increased his operon expression ( 46 ) .
This increase in operon expression has been ascribed to an indirect , strB-induced increase in histidyl-tRNA synthetase , a suspected positive factor of the his operon ( 46 ) .
133 , 1978 METABOLIC REGU the small increase in ppGpp levels that we found in the strB mutant was inadequate to fully account for the strB-associated increase in his operon expression .
Downshift experiments utilizing various hisO mutant strains uncoupled for operon-attenuated control ( Fig. 2 , Table 3 ) suggested that his op-eron metabolic regulation resulted primarily from changes in the initiation specificity of RNA polymerase .
The data show that metabolic control was independent of the presence or the absence of the attenuator DNA sequence ( Fig. 1 ) .
Rather , metabolic control apparently was achieved by increased mRNA chain initiations at specific DNA sequences which included the his operon primary ( P1 ) and internal ( P2 ) promoters .
Interestingly , the intemal promoter of the E. coli tryptophan operon also has been reported to be subject to some form of metabolic regulation ( 33 ) .
Significantly , three different hisO mutations in the primary promoter region ( Fig. 1 ) showed exceptionally high degrees of metabolic regulation ( Table 3 ) .
Representative members of all other classes of hisO mutations ( 13 , 15 ) were examined , but none influenced metabolic regulation .
The hisG gene product plays an uncertain role in his operon control , and HisG enzyme activity has been shown to be inhibited in-vitro by ppGpp in the presence of histidine ( 30 ) .
Downshift experiments clearly indicate that hisG gene product plays no obligatory role in his metabolic regulation ( Table 3 ) .
The specificity of the his metabolic regulation elicited by downshifts or steady-state growth in different media was established by a number of controls .
Genes for phosphogluconate dehydrogenase ( ggnd ) and dipeptide-specific peptidase ( pepD ) remained constitutive under our conditions of culture ( Fig. 2 , Table 4 ) .
In addition , his operon expression in a hisOG1302 deletion strain ( providing readthrough into the his op-eron ) and in a deletion strain ( hisOG203 ) containing a new promoter in the hisG gene ( hisG1306 ) exhibited no metabolic regulation ( Tables 3 and 4 ) .
These controls eliminate trivial explanations for metabolic regulation , such as changes in the overall availability of cellular RNA polymerase ( 33 ) .
They also argue against changes in mRNA stability or translation frequency as bases for his operon metabolic regulation .
Downshift experiments revealed an apparent dependence of his metabolic regulation on the reUA gene product ( Fig. 2 , Table 3 ) .
However , an argument for dependence on the reUA function can not be compelling based on our data .
The relA strains did not recover from the extreme downshift ( Fig. 2A ) and thus might not ATION OF THE HISTIDINE OPERON 841 have been capable of extensive protein synthesis .
Furthermore , relA mutants carry out erroneous translation ( 19 ) .
It is tempting to speculate that the reason why the prototrophic relA strains did not recover in the period examined after downshift was in part due to their inability to metabolically stimulate synthesis of critical amino-acid biosynthetic enzymes .
Dependence in-vivo of his operon expression on the reUA allele has been more convincingly shown using downshift conditions different from those used here ( 39 ) .
his operon metabolic regulation between steady-state growth-conditions was histidine independent ( Table 3 ) .
The degree of metabolic regulation depended on the overall availability of amino-acids other than histidine .
Consist-ently , starvation of fully his ' reU + A bacteria for amino-acids other than histidine caused increased operon expression , even in the presence of excess histidine and of operon-attenuated control ( Fig. 3 ) .
The latter results are in sharp contrast to observations reported on the E. coli tryptophan ( trp ) operon , which shows some form of metabolic regulation in repressorless ( trpR ) strains ( 33 ) .
Tip operon expression in such t7pR strains was unaffected by starvation for amino-acids other than tryptophan , although operon-attenuated control was intact and a reUA + - dependent stringent-response was elicited ( 5 , 29 ) .
Histidine-independent involvement of meta-bolic regulation at specific initiation sites and the suggested role of ppGpp in this regulation fit together nicely with a current model of ppGpp as a primary allosteric effector of RNA polym-erase .
The interaction of ppGpp with highly purified RNA polymerase holoenzyme is well established ( 41 , 43 ) .
As first proposed by Cashel ( 9 ) and later extended by others ( reviewed in reference 42 ) , ppGpp can alter the in-vitro initiation specificity of RNA polymerase depending on how the polymerase interacts with a given promoter sequence .
A given initiation can therefore be stimulated , inhibited , or unaffected by the ppGpp-polymerase-promoter interaction .
Many aspects of his operon metabolic regulation can be understood in terms of this model , including metabolic regulation at a number of his operon-related promoters distinct from the primary ( P1 ) promoter .
A ppGpp-induced change in RNA polymerase 's ability to bind positive factor at mutant promoter regions can explain the high degree of metabolic regulation elicited in two anti-Atn-deficient mutants ( hisO2965 and hisO2966 ) in the presence of the Atn sequence ( Table 3 ) .
Likewise , dependence of met-abolic regulation on overall amino-acid availa-bility is consistent with a general effector-polym erase-promoter model .
Saturation of metabolic regulation at growth-rates characteristic of bacteria grown in miniimal + glucose medium ( Fig. 4 ) is consistent with the notion that his operon transcription is maximally stimulated at lower than maximum ppGpp concentrations .
In strains uncoupled for operon-attenuated control , the maintenance of constant his enzyme specific activities at growth-rates slower than those in miniimal + glucose medium is itself a very general form of metabolic regulation ( 33 ) .
This control system , which most likely maintains a proportionality between transcription initiations and growth-rate , has been observed for the repressorless trp operon ( 33 ) , the repressorless and unattenuated trp operon and the decontrolled lac ( 45 ) of E. ( 47 ) , operon coli .
An anomalous effect observed after growth in alanine medium suggests that his metabolic regulation can additionally involve medium-de-pendent , histidine-independent anti-attenuation ( Table 4 , Fig. 4 ) .
Two trivial explanations for this alanine effect can be excluded .
The effect was detected by comparison of his operon expression in a hisO + hisTl504 strain ( SB3436 ) and a his01242 hisr strain ( TA1003 ) .
An indirect effect of the hisT15O4 allele is excluded because the effect is not observed for gnd gene expression in strain SB3436 or for his expression in strains TA520 and SB2046 , which both contain the hisT1504 mutation ( Table 4 , Fig. 4 ) .
A gene duplication selected by growth in alanine medium ( 2 , 40 ) can not be responsible for the approximate doubling in enzyme specific activity , since duplications also should be selected in strains TA1003 , TA520 , and SB2046 .
Furthermore , gene duplication in strain SB3436 would be expected to affect both his and gnd expression , since all duplications so far characterized are large and would include the his operon and the nearby gnd gene ( 2 , 40 ) ; yet gnd enzyme does not increase in strain SB3436 .
The mechanism for this accessory alanine-de-pendent anti-attenuation is unknown .
The two-site , promoter-attenuator model ( Fig. 1 ) would most simply predict that alanine-induced relief from attenuation is caused by an increased concentration of positive factor .
Several different positive factors conceivably might exist .
In an opposite vein , a decreased concentration of a factor necessary for termination at the attenuator could explain the alanine effect .
For example , rho factor has been implicated in enhancing termination of transcription at the attenuator region of the E. coli trp operon ( 23 ) .
One alternate , and highly speculative , explanation for the alanine effect proposes that aberrant initiations may take place at the attenuator sequence itself .
We thank J. C. Stephens and R. C. Wolf , Jr. , for communicating unpublished results , B. N. Ames and J. R. Roth for strains , Marc Rhoades and Eric Weinberg for access to equipment , and Geraldine Chester for glassware and media preparation .
This work was supported in part by Public Health Service research grant AI01650 from the National Institute of Allergy and Infectious Diseases ( to .
M.E.W. was a trainee on Public Health Service grant HD139 from the National Institute of Child Health and Human Development .
Ames , B. N. , P. E. Hartman , and F. Jacob .
Chromosomal alterations affecting the regulation of histidine biosynthetic enzymes in Sabnonella .
Anderson , R. P. , C. G. Miller , and J. R. Roth .
Tandem duplications of the histidine operon observed following generalized transduction in SabnoneUa typhi-murium .
Artz , S. W. , and J. R. Broach .
Histidine regulation in Salmonella typhimurium : an activator attenuator model of gene regulation .
Bertrand , K. , L Korn , F. Lee , T. Platt , C. L. Squires , C. Squires , and C. Yanofsky .
New features of the regulation of the tryptophan operon .
Bertrand , K. , and C. Yanofsky .
Regulation of transcription termination in the leader region of the tryptophan operon of Escherichia coli involves tryptohan or its metabolic product .
Boquet , P. , MI A. Devynck , C. Monnier , P. Fromageot , and R. Roschenthaler .
Inhibition of stable RNA synthesis by levallorphan in Escherichia coli .
Implication of compounds MSI and MSII .
Brenner , M. , and B. N. Ames .
The histidine operon and its regulation , p. 349-387 .
In H. J. Vogel ( ed .
Academic Press Inc. , New York and London .
Bruni , C. B. , V. Colantuon , L Sbordone , R. Cortese , and F. Blasi .
Biochemical and regulatory properties of Escherichia coli K-12 hiwT mutants .
Inhibition of RNA polymerase by ppGpp , a nucleotide accumulated during the stringent-response to amino-acid starvation in E. coli .
Cold Spring Harbor Symp .
Cashel , M. , R. A. Lazzarini , and B. Kalbacher .
An improved method for thin-layer chromatography of nucleotide miztures containing ` P-labelled orthophosphate .
Chang , G. W. , J. R. Roth , and B. N. Ames .
Histidine regulation in SabnoneUa typhimurium .
Mutation of the hw8T gene .
Kammen , S. J. Spengler , and B. N. Ames .
Biosynthesis of pseudouridine in transfer ribonucleic acid .
Physiological studies of Salmonella histidine operator-promoter mutants .
Ely , B. , and Z. Ciesla .
Internal promoter P2 of the histidine operon of Salnonella typhimurium .
Ely , B. , D. B. Fankhauser , and P. E. Hartman .
A fine structure map of the Salmonella histidine oper-ator-promoter .
Ely , B. , R. M. Weppelman , H. C. Massey , Jr. , and P. E. Hartman .
Some improved methods in P22 transduction .
Fink , G. R. , T. Klopotowski , and B. N. Ames .
1967 Histidine regulatory mutants in Salmonella typhimu-rium .
A positive selection for polar histidine-requir-ing mutants from histidine operator constitutive mutants .
Fink , G. R. , and J. R. Roth .
Histidine regulatory mutants in Salnonella typhimurium .
Gallant , J. , and R. A. Lazzarii .
The regulation of ribosomal RNA synthesis and degradation in bacteria , p. 309-359 .
In E. McConkey ( ed .
) , Protein synthesis : a series of advances , vol .
Marcel Dekker , New York .
Autogenous regulation of gene expression .
R. , and B. Magasanik .
Effect of infection with T-even phage on the inducible synthesis of B-galactosidase in Escherichia coli .
Regulation of expression of the histidine operon in Salmonella typhimurium .
Korn , L. J. , and C. Yanofsky .
Polarity suppressors increase expression of the wild-type tryptophan operon of Escherichia coli .
Lawther , R. P. , and G. W. Hatfield .
Biochemical characterization of an Escherichia coli hisT strain .
Lazzarini , R. A. , M. Cashel , and J. Gallant .
On the regulation of guanosine tetraphosphate levels in stringent and relaxed strains of Escherichia coli .
Lewis , J. A. , and B. N. Ames .
Histidine regulation in Salmonella typhimurium .
The percentage of transfer RNAH '' charged in-vivo and its relation to the repression of the histidine operon .
Polarity in relaxed strains of Salmo-nella typhimurium .
Martin-Del-Rio , R. , V. M. Fernandez , A. R. Tebar , and A. 0 .
Genetic characterization of histidine regulatory mutants of E. coli K12 .
Morse , D. E. , and A. N. C. Morse .
Dual-control of the tryptophan operon is mediated by both trypto-phanyl-tRNA synthetase and the repressor .
Morton , D. P. , and S. M. Parsons .
Synergistic inhibition of ATP phosphoribosyltransferase by gua-nosine tetraphosphate and histidine .
Murray , M. L , andT .
Genetic map position of the gluconate-6-phosphate dehydrogenase gene in SalmoneUa typhimurium .
Raue , H. A. , and M. Gruber .
Control of RNA synthesis in Escherichia coli .
Effect of levallorphan on guanosine 3 ' - diphosphate 5 ' - diphosphate metabo-lism in spoT + and spotT-strains .
Rose , J. K. , and C. Yanofsky .
Metabolic regulation of the tryptophan operon of Escherichia coli : re-pressor-independent regulation of transcription initiation frequency .
Roth , J. R. , D. N. Ant6n , and P. E. Hartman .
Histidine regulatory mutants in Salmonella typhimu-rium .
I. Isolation and general properties .
St. Pierre , M. L. 1968 .
Mutations creating a new initiation point for expression of the histidine operon in Salmo-nella typhimurium .
Scott , J. F. , J. R. Roth , and S. W. Artz .
Regulation of histidine operon does not require hisG enzyme .
Singer , C. E. , G. R. Smith , R. Cortese , and B. N. Ames .
Mutant tRNAHI ineffective in repression and lacking two pseudouridine modifications .
Nature ( London ) New Biol .
Sokawa , Y. , J. Sokawa , and Y. Kaziro .
Regulation of stable RNA synthesis and ppGpp levels in growing cells of Escherichia coli .
Stephens , J. C. , S. W. Artz , and B. N. Ames .
Guanosine 5 ' - diphosphate 3 ' - diphosphate ( ppGpp ) : positive effector for histidine operon transcription and general signal for amino-acid deficiency .
Straus , D. S. , and G. R Hoffmann .
Selection for a large genetic duplication in Salmonella typhimurium .
Modulation of RNA polymerase spec-ificity by ppGpp .
RNA polymerase specificity and the control of growth .
van Ooyen , A. J. J. , M. Gruber , and P. Jorgensen .
The mechanism of action of ppGpp on rRNA synthesis in-vitro .
Vogel , H. J. , and D. M. Bonner .
Acetylornithinase of Escherichia coli : partial purification and some properties .
Wanner , B. L , R. Kodaira , and F. C. Neidhardt .
Physiological regulation of a decontrolled lac operon .
Wyche , J. H. , B. Ely , T. A. Cebula , M. C. Snead , and P. E. Hartman .
Histidyl-transfer ribonucleic acid synthetase in positive control of the histidine operon in Salnonella typhimurium .
Control sites in the tryptophan op-eron , p. 149-163 .
) , Control of ribosome synthesis .