DAVID C. HAGEN AND BORIS MAGASANIK * ABSTRACT In Salmonella typhimurium the structural genes of the enzymes responsible for histidine utilization ( hut ) are clustered in two adjacent operons .
A single repressor regulates both operons .
The repressor itself is a member of one of the hut operons and , thus , regulates its own synthesis .
We have assayed the hut repressor by its ability to bind radioactive DNA to nitrocellulose filters .
The binding is specific for DNA bearing the hut operons , and the binding is-abolished by the inducer , urocanate .
As a member of one of the hut operons , the repressor is inducible , subject to catabolite-repression , and affected by a promoter mutation .
Department of Biology , Massachusetts Institute of Technology , Cambridge , Mass. 02139 Contributed by Boris Magasanik , January 4 , 1973 Earlier work from this laboratory has shown that the genes coding for the enzymes of histidine utilization ( hut ) in Salmonella typhimurium are clustered in two adjacent operons : hutMIGC hutPUH ( 1-5 ) ( Fig. 1 ) .
A single repressor , the hutC gene product , regulates both operons .
Urocanate , the first degradation product of histidine , is the inducer and presumably acts by inactivating the repressor .
A unique feature of the hut repressor is that it regulates its own synthesis , i.e. , it is a member of one of the operons that it regulates ( 6 ) .
The evidence for this hypothesis is twofold : strains in which M , the left-hand operon promoter , has been deleted simultaneously lack I and G , the left-hand operon enzymes , and are constitutive for U and H , the right-hand operon enzymes .
`` Super-promoter '' mutations of M with 3-to 5-fold increased amounts of I and G enzymes have depressed amounts of U and H enzymes .
The depression of U and H is a result of the increase in the amount of repressor and can be relieved by point mutations in the C gene .
A method developed by Riggs et al. ( 7 , 8 ) has made it possible to assay repressors in cell extracts .
The principle of this method is retention by repressor of labeled DNA containing the appropriate operator region on a nitrocellulose filter .
Smith ( 5 ) had previously isolated Xphut transducing phages , and we were thus in a position to prepare the labeled DNA required for the assay of the hut repressor .
The results reported in this paper show that a repressor protein with the properties indicated by our earlier studies is present in extracts of cells carrying the hutC + allele .
MATERIALS AND METHODS Chemicals .
The following chemicals were purchased from Calbiochem : urocanic acid * H20 , dihydrourocanic acid ( imid ¬ Abbreviations : Symbols in roman type indicate enzymes or phenotype , while symbols in italic type indicate genes or genotype .
* Author to whom reprint requests should be addressed .
azole propionic acid ) , ihistidine-HC1 * H20 , chicken-blood DNA , bovine-pancreatic DNase , bovine-pancreatic RNase , Pronase , and Cleland 's reagent ( dithiothreitol ) .
The Lhistidine * HCl * H20 used in growth media and Tris were products of Sigma Chemical Co. .
Whatman P11 cellulose phosphate was purchased from Reeve Angel ; fine particles were removed by settling and decantation , and the ion-exchanger was washed with acid and base and equilibrated with buffer before use .
3 ' ,5 ' - Cyclic adenosine monophosphate ( P. L. Biochemicals ) , Bacto casamino-acids , vitaminfree ( Difco ) , bovine-serum albumin ( Miles Laboratory ) , and carrier-free H332P04 ( New England Nuclear ) were also commercial preparations .
All other chemicals used were of reagent grade .
Strains of Bacteria and Bacteriophage .
Escherichia coli strain GS156 , Gal-StrRT1RXR [ Xb2cI857Sam7 , Xphut36 ( hut + ) cI857Sam7 ] , was isolated in our laboratory .
This Strain , a derivative of the nonlysogen M65 obtained from E. Signer , is simultaneously lysogenic for bacteriophages Xb2cI857Sam7 , obtained from E. Signer , and Xphut36 ( hut + ) - cI857Sam7 , from our laboratory ( 5 ) .
Both phages carry cI857 , a thermoinducible X allele , and Sam7 , a lysis-defective X allele .
E. coli strain M5107 , Gal-StrRT1RXR ( XcI857Sam7 ) , was also a gift of E. Signer .
The S. typhimurium strains NE413 , NE446 , NE449 , NE547 , NE606 , and NE607 are described in Table 1 .
These Salmonella strains were constructed in this laboratory by techniques reported elsewhere ( 1,2,4 ) .
LPC broth contains 1.0 % Bacto vitamin-free Casamino-acids , 0.25 % NaCl , 1 mM MgSO4 , and 0.5 , tg of vitamin-B1 per ml .
The phosphorus content was reduced to 4 , ug/ml by precipitation of phosphate as NH4-MgPO4 , and the pH was adjusted to 7.0 .
Succinate-ammonium minimal-medium , SN , has been described ( 3 ) .
SNH medium is SN plus 0.2 % histidine , and GSNH is SN plus 0.4 % glucose and 0.2 % histidine .
Extract buffer , PM , is 10 mM potassium phosphate-10 mM 2-mercaptoethanol ( pH 7.5 ) ; PMY is PM plus 5 % glycerol ; and PMSY is PM plus 5 % glycerol and 0.4 M KCl .
TDY contains 10 mM Tris HCl-0.1 mM dithiothreitol-5 % glycerol ( pH-7.6 ) .
SMO is 10 mM Tris HCl-86 mM NaCl-1 mM MgSO4 ( pH 7.4 ) .
The DNA binding buffer , BBD , is identical in composition to the BB buffer developed by Riggs et al. ( 8 ) , except that 200 jug of chicken-blood DNA per ml was added and the pH is 7.6 .
FB buffer is BBD minus-chicken blood DNA , bovineserum albumin , and dithiothreitol .
Both buffers were filtered through a 0.45 - , um Millipore filter before use .
GS156 and M5107 served , respectively , as sources of Xphut [ 32P ] DNA and X [ 32P ] DNA .
Cells growing exponentially at 320 in LPC broth were centrifuged and resuspended at a density of 5 X 108 cells per ml in 100 ml of LPC broth plus 0.4 % glucose and 5 mCi of carrier-free [ 82P ] phosphate .
The cells were incubated at 430 for 20 min for thermal induction of phage production , and then grown for 4 hr at 38 ° .
Then , after centrifugation and resuspension in 5 ml of SMO buffer , the cells were lysed with chloroform at 370 and treated with bovine-pancrQatic DNase ( 0.05 iug/ml ) for 15 min at 370 ; the cellular debris was removed by centrifugation .
The phage were purified by block and equilibrium CsCl centrifugation , and the DNA was extracted by a modification of the procedure of Adesnik and Levinthal ( 9 ) .
DNA concentrations were determined from A260 using a conversion factor of 54 , ug/ml per unit A260 determined for chicken-blood DNA .
The experiments reported here were performed with [ 32P ] DNA that had been stored at 40 for 2-3 weeks after extraction .
2-1 Flasks containing 500 ml of SN , SNH , o GSNH medium were inoculated with 5-ml cultures-of cells grown overnight in the same medium .
The flasks were shaken at 370 until the cells reached a density of about 6 X 108 cells per ml and then chilled on ice .
The cells were collected by centrifugation , washed with PM buffer , and resuspended in 5 ml of PMY buffer .
The cells were disrupted with five 30-sec sonic treatments with an MSE Ultrasonic Power Unit .
Cellular debris and ribosomes were removed by centrifugation for 20 min at 35,000 X g and then for 90 min at 100,000 X g .
The final supernatant , about 5 ml , was taken as `` crude extract .
'' The crude extracts contained an average of 14 mg of protein per ml , as determined by the method of Lowry et al. ( 10 ) , with bovine-serum albumin for standardization .
The crude extracts were applied directly to PMY-equilibrated , 1.0 X 3.0-cm columns of cellulose phosphate at 4 ° .
Fractions were eluted from the columns first with a total of 4 ml of PMY buffer and then with 0.5-ml portions of PMSY .
Most of the protein eluted by PMSY was contained in the fourth and fifth PMISY fractions .
These fractions were pooled and dialyzed overnight at 40 against 400 ml of TDY buffer .
The dialyzed fractions were termed `` cellulose phosphate-purified extract '' and contained an average of 1.5 mg of protein per ml .
The degree of purification was about 50-fold .
The hut repressor was assayed by the DNA-binding technique of Riggs et al. ( 8 ) .
Chicken-blood DNA was added to the binding buffer to reduce backgrounds due to nonspecific DNA-binding .
Cellulose phosphate-puri-fied extracts ( 0-25 , l ) were mixed with 1.00 ml of BBD buffer , then prefiltered Xphut [ j2P ] DNA at 14 ng/ml ( 10 , l of a 1.4 Mg/ml stock solution ) was added , and the mixtures were incubated at room , temperature ( 220 ) for 30 min .
USA 70 ( 1973 ) Triplicate portions of 0.30 ml were filtered through FB-soaked , 25-mm Schleicher and Schuell , B6 nitrocellulose filters and then washed with 0.30 ml of FB .
Filters were immersed in 15 ml of water in scintillation vials and the Cherenkov radiation was counted directly on a Nuclear Chicago liquid scintillation counter .
Histidase ( `` H enzyme '' ) , urocanase ( `` U enzyme '' ) , and formiminoglutamate hydrolase ( `` G enzyme '' ) were assayed in crude extracts by reported procedures ( 2 , 3 ) .
RESULTS Assay of the hut repressor and the effect of inducers For isolation of the repressor we used cells of strain NE449 , a strain diploid for hutll145 , a 5-fold super-promoter mutation in the hutMHIGC operon .
The cells were grown in SNH medium to obtain full induction of the hut operons .
Attempts at assaying the hut repressor in crude extracts of this strain proved fruitless .
No binding of Xphut [ 32P ] DNA could be detected over the relatively high background of nonspecific DNA binding , i.e. , over binding in the presence of 2 mM urocanate or binding of X [ 32P ] DNA in place of Xphut [ 32P ] DNA .
Purification on cellulose phosphate greatly lowered the nonspecific binding and revealed the hut repressor activity illustrated in Fig. 2 .
With increasing amounts of purified extract , as much as 40 % of the input Xphut [ 32P ] DNA could be bound .
Essentially no binding was detected in the presence of 2 mM uroca-nate or if X [ 32P ] DNA was substituted for Xphut [ 32P ] DNA .
With the repressor concentration adjusted such that the amount of hut DNA bound was one-half that at saturation , urocanate at 0.07 mM reduced the standard level of binding to 50 % .
Imidazolepropionate , a urocanate analogue and a gratuitous inducer of the hut enzymes ( 1 , 2 ) , was need ~ ed at 2 mi \ I to reduce the binding to the same extent .
Histidine ( 0.5-10 mM ) , glucose ( 50-500 mM ) , or cyclic AMP ( 0.002-0.2 mM ) had no effect on the binding reaction ( data not presented ) .
Incubation of the repressor extract with Pronase ( 5 , g/ml for 15 min at room temperature ) completely destroyed the DNA-binding activity ( results not presented ) .
A comparable incubation with pancreatic RNase ( 5 ug/ml ) had no effect .
By these criteria , the repressor is a protein .
Induction and catabolite-repression of the hut repressor Strain NE449 was grown in three different conditions : un-induced ( SN medium ) , induced ( SNH medium ) , and inducedcatabolite repressed ( GSNH medium ) .
The DNA-binding capacities of the cellulose phosphate-purified extracts are illustrated in Fig. 3 .
The level of repressor is seen to be highest in induced cells .
Cells grown either uninduced or catabolite repressed have lower repressor levels .
The G and H enzyme concentrations of these cells , assayed at the crude extract stage , and their repressor levels , assayed after cellulose phosphate purification , are presented in Table 1 , Exp .
It can be seen that the repressor is inducible about to the same degree as the G enzyme , but to a lesser degree than the H enzyme .
Similarly , the repressor is repressed by glucose in parallel with the G enzyme and not with the H enzyme .
Genotype variation and the hut repressor Cells with different hut genotypes grown under identical conditions on SNH medium yielded different levels of repressor ( Fig. 4 and Table 1 , Exp .
Strains lacking the hutC gene by point mutation ( NE606 ) or by deletion of the entire hut region ( NE547 ) are devoid of hut DNA-binding activity .
A strain ( NE446 ) with a super-promoter mutation of M has increased levels of repressor as well as of G enzyme compared to wild type ( NE413 ) .
The hut diploid strain , NE449 , has about twice the repressor content as the corresponding haploid strain ( NE446 ) .
Although strain NE449 has greater repressor specific activity ( a steeper slope in the initial portion of the binding curve ) than strain NE446 , the plateau of the binding curve is lower and degenerates at high protein concentrations ( see Fig. 4 ) .
Extracts of strain NE449 were found to have greater DNase contamination than those of all other strains .
This additional nuclease is possibly the uvrB gene product , for NE449 is uvrB + while all other strains are uvrB-by deletion .
The uvrB gene product is known to be involved in the excision of pyrimidine dimers ( see ref .
Two classes of mutations in the hut operons have been found to bring about full constitutive expression of both operons : repressor-negative ( hutC - ) and urocanase-negative ( hut U - ) mutations ( 1 , 2 ) .
Urocanase-negative cells are constitutive for the remaining hut enzymes ( H , I , and G ) due to an accumulation of endogenously produced urocanate .
Repressor and enzyme levels from induced and uninduced wild-type ( NE-413 ) , repressor-negative ( NE606 ) , and urocanase-negative ( NE607 ) cells are compared in Table 1 , Exp .
It can be seen that , whereas wild-type cells have an inducible repressor and hutC-cells lack repressor altogether , hutU-cells have a repressor that is synthesized constitutively .
Coordinacy of the repressor and G enzyme activities In Fig. 5 the data of Table 1 are presented graphically to show the coordinacy of repressor with G enzyme and the lack of coordinacy of repressor with H enzyme .
The ratio of repressor specific activity to G enzyme specific activity is about constant for several growth-conditions and for several genotypes .
DISCUSSION In this paper we have reported the partial purification of the hut repressor protein and have demonstrated that the repressor binds specifically to hut DNA and that this binding is abolished by the presence of the inducer , urocanate .
In keeping with our earlier evidence that the repressor is a member of one of the operons that it regulates , we have found that the repressor is inducible , subject to catabolite-repression , and affected by a promoter mutation .
Purification by chromatography on cellulose phosphate is a convenient and rapid means of eliminating 98 % of the protein present in the crude extracts .
We have made the assumption that the is amount of repressor eluted from the columns proportional to the amount of repressor applied .
The procedure can be performed without bias to repressor-negative extracts .
The preparations , however , remain quite heterogeneous , and quantitations of repressor activity are necessarily approximate .
The incubation period for the repressor assays was arbitrarily set at 30 min .
At high protein concentrations of cellulose phosphate-purified repressor extracts , 30 min i ample time for repressor-DNA binding to reach equilibrium ; however , contaminant DNases reduce the quantity of DNA retained the filter .
At low extract concentrations , the on on other hand , 30 min is not long enough for the repressor-DNA binding reaction to equilibrate .
The sigmoidal shape of the saturation curves ( Figs. 3 and 4 ) is probably a result of sub-equilibria at low repressor concentrations .
Other possible explanations of the sigmoidicity ( see ref .
12 ) , however , can not be ruled out at this time .
Fitting a straight line to the initial portion of the binding curves and taking the slope as an estimate of repressor specific activity proved to be sufficiently accurate for to demonstrate the coordinacy of us repressor to G enzyme specific activity ( Fig. 5 ) .
The hut system has a complex regulatory mechanism ; the operons are strongly buffered against premature induction .
The repressor is inducible ; hence , as induction is initiated , further induction becomes difficult .
The inducer-forming more enzyme ( H ) is in balance with the inducer-destroying enzyme ( U ) , and these enzymes , because they belong to the same operon , are synthesized coordinately .
In the induced state , the intracellular concentration of urocanate is presumabiy high enough to inactivate the existing repressor completely .
Once histidine is exhausted from the medium , the concentration of urocanate drops , and the high level of repressor rapidly shuts off hut enzyme synthesis .
The hut operons are thus very cautious in their expression .
This work was supported by Public Health Service Research Grants GM-07446 from the National Institute of General Medical Sciences and AM-13894 from the National Institute of Arthritis and Metabolic Diseases , and Grant GB-5322 from the National Science Foundation .
D.C.H. is supported by a predoctoral fellowship from the National Science Foundation .