THE JOURNAL OF BIOLOGICAL CHEM~TRY Vol. 247, No. 4, Issue of February 25, pp. lOS(t1088, 1972

Printed in U.S.A.

Histidine Regulation in Salmonella typhimurium

IX. HJSTIDINE TRANSFER RIBONUCLEIC ACID OF THE REGULATORY MUTANTS*

(Received for publication, September 23, 1971)

MICHAEL BRENNER$ AND BRUCE N. AMESS From the Biochemistry Department, University of California, Berkeley, California 947.20

SUMMARY

Four classes of mutants constitutive for the histidine operon of Salmonella iyphimurium, hisR, hisT, hisU, and hisW, were examined for a possible role in the production of histidine tRNA. The content of tRNAHi” in tRNA isolated from each of the mutant types was compared to that of the wild type, and the chromatographic behavior of the tRNAHiB obtained from the mutants and the wild type was examined. Chromatography of the tRNAHiB of the wild type gave no evidence for multiple species.

Two factors suggest that hisR is a structural gene for histidine tRNA. First, hisR strains were found to have 25 to 45% less tRNA His than the wild type; and second, intro- duction of an episome carrying the hisR gene into the wild type resulted in a 2.5-fold increase in the quantity of tRNAHiB.

The tRNAHi” of a hisT mutant chromatographed differ- ently from that of the wild type. Since the hisT gene has previously been shown to code for a dispensible protein, we conclude that this protein is involved in tRNAHiB maturation. The tRNAHis from hisT mutants is present in normal amounts, is charged normally, and appears to behave nor- mally in protein synthesis, although not in repression control.

The tRNA from a cold-sensitive hisW mutant was found to have an altered acceptance for several amino acids. This indicates that hisW also codes for a tRNA maturation en- zyme. No increase in tRNAHiS levels was obtained when an episome carrying the hisW gene was inserted into the wild type, and the tRNAHiS from a hisW mutant behaved as the tRNA from the wild type on chromatography.

No difference was found between the tRNAHis of hid mutants and that of the wild type.

The histidine operon of Salmonella typhimurium LT-2 is composed of nine contiguous genes which specify the enzymes of histidine biosynthesis. The operon is known to be under re- pression control (reviewed in Reference 2), but the mechanism

* Paper VIII in this series is Reference 18. $ Predoctoral Fellow supported by United States Public Health

Service Training Grant 5TOlGM31. Present address, Department of Biology, Harvard University, Cambridge, Mass. 02138.

0 This work was supported by Grant AM 12092 from the Na- tional Institutes of Health.

of the control is not clear. Although mutations in any of six loci (hisO, his&‘, hisR, hisT, hisU, and hisW) are capable of pro- ducing constitutivity, the functions of only two of the six genes are known. His0 is an operator gene (or promoter gene, or both) : mutations in his0 are &dominant (3, 4)) and map at the beginning of the operon (5). The second gene, his&‘, codes for the histidyl-tRNA synthetase (1, 6), the enzyme which at- taches histidine to tRNAKiB. The properties of the his8 mu- tants, and also some analogue studies (7), have shown that his- tidine itself does not act directly in the repression mechanism. Instead, charged histidine tRNA may provide the controlling signal.

Because of the probable importance of histidine tRNA in the repression mechanism, a series of experiments were undertaken in an attempt to clarify a suggested relationship (2) between histidine tRNA and the remaining regulatory mutants, hisR, hisT, hisU, and hisW. Using column chromatography, a search was made for multiple (isoaccepting) species of histidine tRNA in the wild type. Column chromatography was also used in an attempt to detect alterations in the tRNAHis from mutant strains. Finally, a more thorough study was made of the level of histidine-accepting activity of tRNA isolated from the mutant strains.

MATERIALS AND METHODS

The crystalline sodium salt of ATP, and bovine serum albu- min, type V, were obtained from Sigma. Glass fiber filters (type A, 1 inch in diameter) were purchased from Gelman In- strument Co. and filter paper discs (25 mm) from Schleicher and Schuell. [3H]- and [%]histidine, [14C]glycine, t3H]lysine, and [3H]valine were purchased from New England Nuclear; [3H]- and [14C]valine were also obtained from ICN; [14C]arginine was from Tracerlab; [3H]arginine from Schwarz BioResearch; and [3H]leucine from Amersham - Searle. Homogeneous histidyl- tRNA synthetase (specific activity 5000 units per mg) was sup- plied by F. De Lorenzo. Aminoacyl-tRNA synthetases for argi- nine, glycine, leucine, lysine, and valine were obtained as by-products of the purification of the histidyl-tRNA synthetase (8) : the column fractions containing no histidyl-tRNA synthetase activity were pooled, concentrated by vacuum dialysis, and stored at -20” in 50% glycerol. Benzoylated DEAE-cellulose was a gift of W. Holloman. This material was superior in packing, flow rate, and resolution to some commercial preparations tried (Schwarz BioResearch). Chromosorb W, acid washed, dimeth- yldichlorosilane treated (100 to 120 mesh), was purchased from

1080

by guest on March 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Issue of February 25, 1972 M. Brenner and B. N. Ames 1051

Varian *Aerograph. Trioctylpropylammonium bromide was obtained from Eastman Organic Chemicals.

Bacterial &rains-JL250, a cold-sensitive hisW mutant, was supplied by J. E. Brenchley. At 37”, the mutant has a normal growth rate in nutrient broth, but grows slightly more slowly

than the wild type in minimal salts-glucose. When shifted to 20” the organism stops growth immediately, regardless of the medium (9). His01828 was obtained from P. E. Hartman, and hisC527 metESS ilvC401 purFS4?‘/F’14 (TR566) from J. R. Roth. All other strains used were from the collection of this laboratory. Strains LT-2, ara-9, his01242, hisU1817, hisU1820, hisWiSL1, hisWl825, hisRibfZ3, hisRiM (TA796), hisHlO7, hisRl2OS (TA778), and hisTl504 (same origin as hisTl503) are described in Reference 5. Episomes F’14 and F’32 and the episome-carrying strains hisRld23 metESS8/F’14 and ilvC401 metESSK/F’l4 are described in Reference 3. The mutant hisSl595 was isolated by G. W. Chang by mutagenesis of LT-2 with diethyl sulfate. The strain aroD purFi45 his-I?46i/F’S.f? (TAl784) was constructed by J. Lever by transferring the epi- some from aroD cysCllld/F’Sd (TRll, obtained from J. R. Roth) into aroD purF145 his-2461 (TA297).

Growth of Bacteria-The various strains of S. typhimurium LT.2 were grown in 15-liter New Brunswick fermentors with vigorous aeration in a minimal salt medium (10) containing 0.5% glucose. When required, histidine was added to the growth medium at a concentration of lop4 M (TR566, TA778, and TA1536), and adenine and thiamine added at 2 X low4 M and 10h6 M, respectively (TR566). The cells (in late exponential phase of growth) were harvested by centrifugation and stored at -20” for up to a month.

Preparation of tRNA-The tRNA was prepared by a modifi- cation of the method of Silbert et at. (11) and stripped according to the method of Sarin and Zamecnik (12). The following pro- cedure is given for 100 g of cells, wet weight, but has been scaled up and down with comparable results. All steps are carried out at 4”, unless otherwise noted. The frozen bacterial paste is cut into chunks and suspended in 200 ml of 10 mM Tris-HCl, pH 7.5, and 10 mM MgC12. Then 200 ml of phenol (Malinc- krodt) saturated with the same buffer are added, and the mix- ture shaken vigorously for at least 1 hour. The resulting SUS-

pension is centrifuged at 14,000 x g for 30 min, and the aqueous (top) phase sucked off and passed directly onto a column of DEAEcellulose (Selectacel 70, Schleicher and Schuell Co.) with a bed volume of 100 ml (2.3 x 24 cm).

The column had been packed under about 50 p.s.i. in a buffer of 1 mM Tris-HCl, pH 7.5, and 10 mM MgC& (TM buffer) con- taining 1 M NaCI, and then initially equilibrated with TM buffer containing 0.02 M NaCl. The aqueous phase from the phenol extraction is passed onto the column at a flow rate of 2 to 4 ml per min. The column is washed with the equilibrating buffer at 5 ml per min until the A2c0 of the effluent decreases to about 0.1, and is then washed at the same flow rate with TM buffer containing 0.1 M NaCl until the A260 of the effluent declines to about 0.05. The tRNA is then eluted at a flow rate of 1 ml per min with TM buffer containing 1 M NaCl, and fractions of about 20 ml are collected. The specific activity of tRNA”” remained constant until the AzGO fell below 0.8, at which point the purity decreased rapidly. By pooling those fractions having an A260 greater than 1.0, over 95% of the total A260 is obtained, and material with constant specific activity is insured. The pooled fractions are precipitated with 2 volumes of ethanol, left in the

cold overnight, and then most of the supernatant removed by suction and the precipitate collected by centrifugation at 14,000 x g for 30 min. The precipitate is dried under vacuum at room temperature, and then dissolved in 10 ml of 1.8 M Tris-acetate, pH 8.0. The solution is incubated at 37” for 90 min, filtered through a glass fiber filter, and the tRNA precipitated by the addition of 25 ml of 95yo ethanol kept at - 20”. After 60 min at -2O”, the precipitate is collected by centrifugation at 27,000 x g for 10 min. The precipitate is washed twice by resuspension in cold 75% ethanol. The final precipitate is dried under vacuum at room temperature. Yields are from 150 to 250 mg per 100 g of cells, wet weight, with there being about a 15% loss in A260 units between the pooled column fractions and the final yield. The tRNA has 15 to 18 AzgO units per mg, with approximately 15 pmoles of tRNAHis per AZ60 unit. The purity of all tRNA’s is increased about 30% if fresh, rather than frozen, cells are used, presumably because freezing and thawing the cells produces some cell breakage, leading to greater contamination of the tRNA with ribosomal RNA. Growing the cells on nutrient broth (Difco), rather than in minimal me- dium, produced no significant change in the tRNA contents (done for wild type only). The inclusion of 2 mM thiosulfate in the buffers has no detectable effect on the kinetic or chromato- graphic properties of a number of tRNA species tested; however, Singer and Smith’ have found that the 4thiouridine present in Salmonella tRNAHi’ is about 20% oxidized in the absence of the thiosulfate.

Preparation of Charged tRNA-Charged tRNA was prepared as described elsewhere.2

Operation of Benzoylated DEAE-cellulose Column-A column bed, 1.5 x 84 cm, was packed by gravity flow of a slurry of the adsorbent in a buffer of 10 mM MgClz and 10 mM sodium acetate at pH 4.5, which was 1 M in NaCl. Columns were initially equilibrated with the MgClz-sodium acetate buffer 0.5 M in NaCl, and the tRNA applied in a volume of several milliliters. A salt gradient running from 0.5 to 1 M NaCl in the MgC&sodium acetate buffer was used to develop the columns. The gradient was followed by a wash of 2 M NaCl in 150/, ethanol. When used, nonlinear gradients were generated with a Dialagrad model 190 (ISCO, Lincoln, Neb.). All column operations were con- ducted at room temperature. Recovery of Am units was usu- ally above 90%, and recovery of histidine-accepting activity was about 85%.

Operation of Reversed Phase Column No. S-The adsorbent was prepared and the columns packed and operated as described by Weiss et al. (13). When tRNA was chromatographed un- charged, columns were operated at 37” using a buffer of 10 mM MgClz and 10 mM Tris-HCI, pH 7.0. When charged tRNA was chromatographed, columns were operated at either room tem- perature or 37”, with a buffer of 10 mM MgClz and 10 mM sodium acetate, pH 4.5. Columns were developed with a NaCl gradient generated with the Dialagrad model 190. Recovery of histidine- accepting activity from columns run with the tRNA uncharged was about 80%. Recovery of charged histidine tRNA from columns was roughly that expected from the half-lives for de- acylation of histidyl-tRNA His in the column buffer: 17.3 hours at room temperature, and 5.5 hours at 37”. Recoveries from specific runs are given in the figure legends.

1 C. E. Singer and G. R. Smith, J. Biol. Chem., in press. 2 M. Brenner, J. A. Lewis, D. S. Straus, F. De Lorenzo, and B.

N. Ames, in preparation.

by guest on March 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

1082 Histicline tRNA of the Regulatory iiIuta?lts Vol. 247, So. 4

Analysis of Double Labeled Columns-The tRNA was precipi- tated by the addition of 0.1 volume of 100GjO trichloroacetic acid (1 g per ml), and the precipitate collected on a glass fiber filter. Filters were washed several times with ethanol, dried about 10 min at 110”, and counted in 7.5 ml of a mixture of toluene, 2,5- diphenyloxazole (PPO), and 1,4-bis[2-(5-phenyloxazolyl)] ben- zene (POPOP) (SpectraUuor, Amersham-Searle) in a Kuclear Chicago Mark I liquid scintillation counter.

Concentration and Desalting of Column l+actions-Fractions of columns run with unlabeled tRNA were simultaneously con- centrated and desalted by precipitating the tRNX with 2.5 vol- umes of ethanol. The alcoholic solutions were left overnight at -2O”, and the precipitates then collected on Millipore filters

type HA, 0.45 pm (Xllipore Filter Corp., New Bedford, Mass.).* The filters were air dried, and the tRNh eluted into 1 ml of buffer containing 0.167 M sodium cacodylate at pll 7.5 and 13.3 IllM

1&$X2. These buffer components were chosen so that they would satisfy the standard assay requirements for sodium caco- dylate (0.1 M) and MgCIZ (8 m&l) when the tRNh sample made up fZOc/, of the final reaction Tolume. The elution was carried out in glass scintillatidn vials by shaking for 30 min at 37”. Recovery of AZso units, which was monitored in each case, was generally between 85 and 95%.

Assays for, t&VA--Two methods were used to process the re- action mixture. In the glass fiber filter assay, the reaction was stopped by the addition of 3 ml of cold 10% trichloroacetic acid, and the resulting suspension was chilled in ice. The precipitate was collected on a glass fiber disc and washed four times with 5 ml of 10yO trichloroacctic acid, four times with 5 ml of 95y0 ethanol, and twice with 5 ml of ether. The air-dried filters were counted in the tolucne-2,5-diphcnyloxazole-l,4-bis[2-(5-phenyl- osazolyl)]benzenc scintillation fluid described above in a Nuclear- Chicago Mark I liquid scintillation counter at an efficiency of 30% for 3H and 90% for l*C.

The second assay method utilized filter paper discs (14). Aliquots of reaction mixture were pipetted onto filter paper discs which wcrc held on a straight pin. The wet filter was immedi- ately immersed in cold 10% trichloroacetic acid. The filters were washed three times for 10 min in cold 1O70 trichloroacetic acid, and then 5 min each in ethanol and ether. Stirring was accomplished by a magnetic stirring bar separated by a wire mesh from the filter paper discs. The air-dried discs were counted as above. This assay was not used for precise quantitative work, as it was found washing in trichloroacetic acid slowly removed counts from the filters (about 10% in 10 min).

Assay of Column Fractions for II&i&e Acceptance-A par- tially purified histidyl-tRNA synthetase, supplied by J. R. Roth, was used in the assays of the benzoylated DEAE-cellulose col- umns. The enzyme had been purified about B-fold by prota- mine sulfate precipitation of nuclric acids and Sephadex G-150 chromatography. Assays of the reversed phase columns No. 3 were conducted with pure histidyl-tRNA synthetasp, supplied by De Lorcnzo and Ames (8).

111 assays for histidine acceptance the column fractions were either assayed directly or after they had been concentrated and desalted as described above. When the column fractions were used directly, the aliquot of fraction assayed constituted up to 5070 of the final reaction volume. This gave NaCl concentra- tions of up to 0.4 M in the reaction mixture. However, the aminoacylation reaction with histidine is reasonably insensitive

to salt, with no reduction in the final level of iuLu,rporation being found at 0.25 RI KaCl, and less than a 25C; reduction at 0.5 M. In assays of either the concentrated or ullcc,llcentrated column fractions, the reaction mixture contained P mM MgCll, 4 rnnl ATP, 50 p@H]- or [14C]histidine, 0.1 RI sodium cacodylate at pH 7.5, and sufficient histidyl-tRNA synthetase to bring the reaction to completion in the time allotted. The reactions were run at 37”, usually for 20 min. The reaction volume ~ZIS 0.25 ml for assays of the benzoylated DIME-cellulose column.~, and 0.1 ml for assays of the reversed phase columns No. 3. The re- action mixtures were processed by either the filter paper or glass fiber filter methods.

Assays of Column Fractions for Glycine, Lysirre, and Voli,te ;lc- ceptance-Due to their greater sensitivity to salt concentl,ation assays for glycine, lysine, and valine acceptance were done with the desalted, concentrated fractions only. In a total of 0. I ml the reaction mixtures contained 8 rnh1 MgCll, 4 mM ATP, 0.1 hl sodium cacodylate at $1 7.5, 50 PM %- or 14C-amino acid, and sufficient aminoacyl-tRNA synthetase to bring the reactions to completion in the time allotted. The reactions were conducted at 37”, usually for 20 min. Aliquots of 75 11 were takrn for the filter paper assay.

Assay of Bulk tRSA for Ilistidine, ArginiiLe, Leucinr, Lysine, and Valine Acceptance-111 a final volume of 0.1 ml the reaction mixtures contained 0.1 M sodium cacod\-lat.? at pH 7.5, 8 rnnl AIgCl,, 4 111hI ,1Tl’, 4 to 5 Az~,, units of tRS.2; and sufficient en- zyme to drive the reaction to completion (as judged by no further incorporation of label with time and proportionality of counts incorporated to tRNA added). The reaction contained the appropriate amino acid at a concentration of 50 PM, except for histidine, which was at 15 ,UM. The specific activities of the amino acids were adjusted to yield a minimum of 10,000 cpm incorporated per assay. The reactions were performed :tt 3i” for 20 min, then quenched by the addition of several milliliters of cold 10% trichloroacetic acid. The tR?JA precipitate was collected on glass fiber filters and counted as described above.

Determination of tRNAHi” Content: Method of Standardixution- In the course of prrparing a number of batches of wild type tRKX it had been observed that the purity of the preparations varied significantly. Hence, acceptance per AQGO nilit was not a good basis for comparing the relative amount of tRNAHis in the several regulatory strains. Instead, the acceptance of four other amino acids served as internal standards. These four amino acids were arginine, leucine, lysine, and raline.

The acceptances of the four standard amino acids were used to correct for differences in the purity of the tRNA preparations. This was done by choosing a factor to multiply the acceptance for each of the four amino acids so that the adjusted values would be in the closest possible agreement, with the wild type standard values. This same factor was then used to multiply the acceptance for histidine, thereby providing a value adjusted for the relative purity of the preparation. The basis for calcu- lating the correction factor was that the adjusted acceptance of each of the four standard amino acids should be as close as pos- sible to their acceptance in the wild type ,standard. Hence, if the measured values are represented as JI,,.,, MLru, MLYs, and Mval; and the standard values as S,,, to Sval, then thr factor f was chosen such that f-l1 = X, or by simple rearrangement, so that j;ILr/S = 1. Since four standards were used, the factor f was selected so that the a\-erage f-V/S equaled one.

by guest on March 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

llf. Brenner and B. N. Ames lOS3

+ (f~V,I/SV.dl = 1.00 .A simple rwxr:mgcmcxt of this equation provided the formula for determining f.

In the tables the tKN-1 contents arc presented as t,he adjusted acceptance relative to the wild type standard, that is, as fX/X. The values for histidille acceptance were not included in the cal- culation of j so that no constrzrint would be placed on the values of hixtidinc ac~c~cl)t:ulc~c relative to the wild types.

Relntizle Purity oj tn.\\--1 Preparations--The factor j is a meas- ure of the tRNIZ content of the preparations: the greater f, the lowvcr the t,RK.\ czoncentration. The tRNh concentration varies from solution to solution depending upon the purity of the tRNA prq)arntiolI :Illd the nuclric acid roncentration (A260 units per ml) of the solutiorl. Khcll/ is multiplied by the 11260 of the solu- tiolls, a fac+or ,I’ is obtained which depends only on the purity of the preparation. The reciprocal of this factor, y-1, is directly l)roportiollal to the purit,y. Each of the values of y-l has been tiividcd by the, standard wild type j’-’ to obtain the relative purity of the 1;RN;\. These values are also listed in the appro- priate tables.

RIBULTS

Ilistidine tEATA Content of II&dine Regulatory Mutants

Reproducibility oj Assays and of tRXA Preparations: Selection of TVil:ild Type Standard-Table I shops data obtained from assays

tlL\‘A conlents: wild type Salmonella

Independent preparations of tRNA are denoted by repetition

of the strain designation. Values have been normalized to the lrild type standards as described under “Materials and Methods.” The various preparations are as follows: “standard,” prepared as

described under “Rlaterials and Methods”; “thiosulfate.” Dre-

of several different prepamtions of tRKA from wild type Sal-

monella. As can be seen, very good reproducibility is obtained for the same preparation assayed at different times. However, the different preparations show some variation from each other. This variation might arise during either the growth of the bac- teria or the isolation of the tRNh, as different batches of cells were used for each preparation. These variat,ions determine the sensitivity with which the tRNA content of the various mu- tants can be distinguished, any difference less than &15% is probably not significant. Because of this variation me have used as a standard value an average of the values obtained from all the wild type tRNA preparations.

tRNAHiS Content of ara-3, hid, hid’, hisT, hisU, and hisW strains-The relative histidine acceptances of tRNA isolated from the histidine regulatory mutanbs, hisO, hid, hisT, hisU, and hisW, are shown in Table II. The tRNA isolated from his0 and hi.6 strains may be considered as derepressed controls, as these two genes are known not to be involved in tRNA produc- tion. The tRNAniB content of the his0 and hisS tRNA falls in the normal range of variation.

The hisW mutant displaying the severest phenotypic effects is JL250, a cold-sensitive hisW strain isolated in the laboratory of J. L. Ingraham (9). Analysis of the tRNA content of this

T:WLE II

tRNA contents: ma-9, hisO, hiss, hi.sT, hisli, and hisW

Independent preparations of tRN;A are denoted by repetition of the strain designation. Values have been normalized to the

wild type standards as described under “Materials and Methods.” The various preparations are as follows: “standard,” prepared as described under “Materials and Methods”; “thiosulfate,” pre- pared as “standard” except 2 rnsz thiosulfate xas present in all

buffers; “Silbert,” prepared by method of Silbert et al. (11). -

Strain Preparation Relative acceptance of

pared as “standard” except 2 m&c thiosulfate was present in all buffers; “Silbrrt ” prepared by method of Silbert et al. (11). Oi242

’ , ” _ - ,-

LT.2 Silbcrt

LT.2 Silbert

LT.2 Standard

LT.2 Thiosulfate

LT-2 Thiosulfate

Preparation

- (II a4s

Relative acceptance of 01828

His 4rg

1.02

0.98 0.99 1.02 0.96

0.85 1.10

0.86 1.11

0.84 1.09 0.84 1.10 0.83 1.09

0.96 1.10 1.08 0.92 1.15 1.08 0.94 1.10 1.06

LyS

1.07

1.18

1.01 0.98

1.04 1.03 0.99 0.96 0.99 1.03

1.00 0.99

1.15 0.83

1.16 0.81

1.18 0.83 1.18 0.80

1.17 0.83

0.87 0.96 0.8G 0.95 0.84 0.99

- -

Val

0.97

1.00

0.99 0.95 1.03 0.99

1.00

0.92 0.91 0.91 0.92

0.91

1.09 1.12 1.11

s1595

JL250 (Cold-sensitive

hisW) JL250 Wf 821 (ara-9) WI825 (m-9)

181825 (aru-9)

U1817(ura-9)

Ul817(ara-9)

Ul820 (ura-9)

ara-9 Standard 1.08 1.181.021.041.04~0.90

TX04 Standard 0.90 T1504 Thiosulfate 0.86

-

Standard 0.97 Thiosulfate 0.84 Thiosulfate 0.76

Standard

Thiosulfate

0.82 1.200.951.041.13i0.88

O.G9 0.72

Thiosulfatc Standard Silbert

Standard

Silbert

Standard

Standard

0.75 1.021.001.61j0.420.97 0.85

0.82 1.17 1.00 1.2810.88’0.84 1.070.881.00~1.011.12

0.82 1.02’0.90’0.970.951.17 0.75 1.111.021.290.830.85

0.83 1.110.861.040.941.lG 1.00 i1.230.891.211.010.89 1.03 1.200.96!1.120.950.99 0.87 1.231.00’1.280.860.85 0.89 1.121.041.170.830.97

0.960.961.590.600.85 1.001.001.490.64;0.88

~1.401.100.831.130.94 1.201.10,0.98i1.050.86

I

by guest on March 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

1054 Histidine tRNA of the Regulatory Mutants Vol. 24;. so. 4

TABLE III

LRNA contents: hi& und episome-carrying strains

Independent preparations of tRNA are denoted by repetition of the strain designation. Values have been normalized to the wild type standards as described under “Materials and Methods.” The various preparations are as follows: “standard,” prepared as described under “Materials and Methods”; “thiosulfate,” pre-

pared as “standard” except 2 mM thiosulfate was present in all buffers; “Silbert,” prepared by method of Silbert el al. (11).

Strain

121200

1:1.203”

1<1203(~ R1223 1~1’1813 (ala-Q)

111813 (ala-Q)

Rl223/F”14 (hisR+)O

HisIZ+/F’l.J (hisR+)” TKX%” [hisI<+/F’l& (hisR”)I

TA1784a [hislY+/F’32 (hisIT’+)] T-417840 [hisII.+/F’32 (hisIV+)l

R&l- Relative acceptance of Preparation tive

plH- ity His ’ Arg Leu Lys Val]

__~_

Silbert 0.770.751.060.980.931.03 0.730.731.001.060.901.03

Silbert 1.10O.FO0.880.941.141.04

1.030.5G0.92~0.971.051.06 Thiosulfate 1.000.681.061.140.840.97

Thiosulfate 0.930.58’0.921.080.961.04 Standard 0.630.800.990.921.180.91

0.780.621.000.921.120.96 Thiosulfate 0.990.581.001.080.940.99

Silbert 0.901.530.940.861.181.02

Silbert 0.892.450.901.121.070.92

Thiosulfate 0.852.491.101.170.930.81

Thiosulfate 0.741.131.031.240.830.90

I I Thiosulfate 0.801.071.09~1.150.900.85

i I

n Complete genotype given under “Materials and Methods.”

strain (grown in minimal salts-glucose at 37”) reveals the relative

amount of leucine tRNA to be about 50% higher than normal, and the relative amount of lysine tRNA to be about 500/, lower than normal (Table II). These values are only approximate, since the normalization procedure is not valid when tRNA con- tents other than that for histidine are changed. On the basis of acceptance per A260 unit, leucine acceptance is only about 10% above normal, whereas the acceptance of the other amino acids falls considerably below normal. The same relative tRNA contents are found whether the assays are performed at 20 or 37”. These variations from the norm are greater than for any

of the other strains tested (except for tRNAHis in an episome- carrying strain. Set below). This multiple effect of the hisW mutation suggests it may encode a tRNA maturation enzyme.

The remaining hisW mutants, and also the hisU mutants, have

tRNA contents similar to that of their ura-9 parent.3

3 A preliminary report that hisU and h&W strains had a reduc- tion in their tRNAHis levels was based on comparisons of histidine acceptance to the AZ80 of the preparations (G. R. Fink and J. R. Roth, unpublished experiments referred to in Reference 15). The same’preparations used in that study were also analyzed in the nresent one [Table I. Line 1: Table II. Lines 8 and 10). with the result that ihen the’histidige acceptance was normal&ed to the internal standards no significant difference was found between the mutant and wild type preparations. The apparent reduction in the tRNAniS level in the hisU and hisW mutants had resulted from those tRNA preparations being less pure than that made from the wild type.

The hisT gene codes for an enzyme which forms pseudouridine in the anticodon loop of histidine tRNA (see below). Tile data in Table II show that this modification is not required for in viva stability of tRNAniB, or for its charging.

tRNAHiS Content oj hisR Mutants and of Some Episomc-carry- ing Strains-A previous study by Silbert et al. (11) showed that hisR mutants have a reduced quantity of histidine tRNA. That result is confirmed here (Table III). In addition, the histidine tRNA content has been determined of strains carrying an extra dose of the hisRf gene on an episome. If hisR is a structural gene for histidine tRNA, then such strains should have an in- creased level of tRNAHi8. Introduction of the episome into either a hisR- or a hisR+ strain results in a large increase in the level of histidine tRNA (Table III). On the other hand, inser- tion of an episome covering the hisW region into a hisW strain has no effect on the tRNAHis levels (compare Table III with Table II). These data suggest that hisR is a structural gene for histidine tRNA, while hisW is not.

Chromatographic Properties of tRNA jrom Wild Type and Regulatory Mutanis

Chromatography on Benzoylated DEAE-cellulose-Chromatog- raphy of tRNA from wild type Salmonellu on a benzoylated DEAE-cellulose column gave the elution pattern shown in Fig. 1. As shown in the figure, there is no evidence for multiple species of histidine tRNA.

Chromatography on Reversed Phase Colwt~n No. S--The re- versed phase column No. 3 of Weiss et al. (13) was used to analyze

Salmonella tRNA, as the column had been shown to resolve two species of histidine tRNA from Escherichia coli B, an organism very similar to Salmonella. However, as shown in Fig. 2, wild type Salmonella tRNA yielded only one major peak of tRKAHiS. A very small second peak contained a few per cent of the total histidine-accepting activity. The tRNA of this second peak was not studied further because it was present in such a small amount, and because it was present in tRiYA isolated from each of the regulat.ory strains as well as from the wild type. Experi- ments in which the tRNAHi8 was previously labeled with radio- active histidine before being applied to the column suggest that the small peak is a column artifact, and not a different species of tRNAHi8 (see below).

The large number of sharp A2e0 peaks indicated that the col- umn was resolving well; however, as a control, tRNA from E. coli B was fractionated to show the presence of two tRNAHi’ peaks. Surprisingly, commercially prepared tRNA from E. coli B yielded only a single peak, whereas E. coli B tRNA supplied by A. D. Kelmers gave two peaks, exactly as published (13). The reason for this discrepancy is not known.

In addition to tRNA from wild type Salmonella, tRNA from hisR, hisU, hisW, and hisT mutants was analyzed on the reversed phase column. Since the aminoacyl linkage of histi- dine to tRNA is one of the most labile, it was considered possible that if chromatography was performed with the tRNA previously labeled an altered species of tRNAHiB might be missed if it were preferentially deacylated. Accordingly, an initial set of chro- matographic analyses were conducted in which the elution posi- tion of uncharged tRNAHi8 was determined by charging column fractions with radioactive histidine. Since the A260 pattern varied from column to column, the peak positions of three nearby tRNA’s, glycine, lysine, and valine, were used as internal stand- ards. No significant difference was found between the elution

by guest on March 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Issue of February 25, 1972 Jf. Brenner and B. N. Ames

r

10

I 08

A ii ‘: p ii// Histidine Acceptor Activity

b

i i z= ! i 5 ‘D

FIQ. 1. Fractionaction of tRNA L from Salmonella typhimurium LT-2 on a benzoylated DEAE-cellulose col- umn. 4 sample of 182 4260 units was applied to the column, which was then developed with a linear gradient of NaCl in a buffer of 10 mM MgClp and 10 mM sodium acetate, pH 4.5. The initial volumes of the reservoir and mixing chambers were each 500 ml. The flow rate wasa0.5 ml per min, and fractions of 12 ml were’$ollected.

02

0

__-- ----o __-- ----- ii? 0.6 :

9 ; 5 J

n ;I

,,,’ -1100

Absorbance a, 260 nm ? / ,!’ d

TUBE NUMBER (15 ml/lube)

FIG. 2. Fractionation of tRNA from Salmonella typhimurium LT-2 on a reversed phase column No. 3. A sample of 1200 AzeO units was applied to the column, which was then developed with a gradient of NaCl in a buffer of 10 mM MgClz and 10 rnM Tris-HCl, pH 7.0. The flow rate was 1.5 ml per min, and fractions of 15 ml were collected. Fractions were assayed for histidine acceptance as described under “Materials and Methods.”

position of tRNAHiB from the wild type and that from a hisR, a hisU, or a hisW mutant. Only the histidine tRNA from a hisT mutant (hisTl604), had a suggestion of being altered in its chro- matographic behavior, perhaps eluting two fractions late with respect to the internal standards.

Comparison of Histidine tRNA from Wild Type and Mutant Strains by Double Label Chromatography-Except for the ubiq- uitous minor peak, the tRNA His of each of the mutant strains migrated as a single peak in about the same position as as the tRNAHi” from the wild type. To more critically compare the elution positions of the tRNAHi” from the mutants to that from the wild type, double label columns were run. In these experiments tRNA isolated from a mutant strain was charged with [aH]histidine, that from the wild type with [Wlhistidine

TUBE NUMBER (10 ml per Tube)

-- //I I-

-~~ ~~ - T- Absorbance at 260 “m ,-,

Oi -10 z 20 40 60 80 100 120 140

FRACTION NUMBER

FIQ. 3. Comparison of the elution position of histidyl-tRNAHi8 from wild type (LT-2) and hisTlbO4 on a reversed phase column No. 3. The applied sample contained 0.85 mg of tRNA from LT-2 charged with [aH]histidine, 2 mg of tRNA from hisT1604 charged with [Wlhistidine, and 4.5 mg of uncharged LT-2 tRNA as carrier. The column was developed at room temperature with a gradient of NaCl in 10 mM MgCl, and 10mM sodium acetate. pH 4.5. The flow rate was 1.5 ml per min, and 10 ml fractions were collected. The tRNA was precipitated from the fractions and counted as de- scribed under “Materials and Methods.” Peak tubes of aH (LT-2) and 14C (hisTlbO4) had 2390 and 7360 cpm, respectively. Re- coveries were 40y0 for 3H (LT-2) and 45% for 1% (hisTZ604). The greater sharpness of the hisT1604 peak is probably a result of the salt gradient being steeper during its elution.

(or vice oersa), and the two preparations cochromatographed on a reversed phase column No. 3. The tRNAn” from the hisU and hisW strains tested (hisUl80 and the cold-sensitive hisW, JL250) eluted exactly with the wild type. The tRNAHi* from hisTf604, however, eluted after the wild type tRNAHi* (Fig. 3). This difference in elution positions remained when independent preparations of wild type and hisT1604 tRNA were analyzed, and also when the labels used to charge each tRNA were inter- changed. Histidine tRNA prepared from a hisR1813 mutant eluted very slightly ahead of the wild type tRNA (Fig. 4). How- ever, the sequence of the tRNAnis of hisR181S has subsequently

by guest on March 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Hi&dine tRNA of the Regulatwy Mutants Vol. 247, iTo. 4

Absorbance at 260 nm n

FRACTION NUMBER

FIG. 4. Comparison of the elntion position of histidyl-tltNAHis from 1 he wild type (I,T-2) alrd his121813 011 areversed phase column ?r’o. 3. The applied sample cotltaincd 1.7 mg of tRNA from LT.2 charged with [14C]histiditle, 0.2 rng of tRNA from hisRf813 charged Tvith [,lH]histidine, and 5 mg of uncharged LT.2 tRNA as carrier. The column was developed at 37” with a gradient of NaCl in 10 mM RIgClz and 10 rn~ sodirun acetate, pII 4.5. The flow rate was 1.5 ml per min, and lo-ml fractions were collected. The t,RNA was precipitated from the fractious aud counted as described under “Muterials and Ylethods.” Peak tubes of 14C (LT-2) and 3H (hisf21813) had 2055 and 3054 cpm, respectively. Recoveries were 8.7yo for W (LT.2) and G.7y0 for 3H (hrisRl813).

been found to be identical w&h that of the wild type.* The rea- son for the different clution position found here is not known.

Some difficulty TWS encountered in the performance of the re- versed phase (Lolumn No. 3 in the double label experiments. Freshly packed columns yielded two peaks of radioactivity, one near the position at which unlabeled tRNAniS elutes, and a much larger pea,k at a higher salt concentration. On succeeding runs, however, the late peak steadily diminished, until after about five runs the histidyl-tRNAHis eluted entirely in the early position. In the experiment illustrated in Fig. 5 (the fourth run of that particular column), labeled tRNA was run with a 20-fold excess of unlabeled tRKA. The tRNA in even-numbered tubes was precipitated with trichloroacctic acid to determine the position of the previously labeled tRNAHis, and the tRNA in odd-num- bercd tubes ~vas aminoacylated with [3H]histidine to determine the l)osition of thrx unlabeled tRNA”“. As shown previously (Fig. 2), unlabeled tRNAHis yielded an early major peak and a late minor peak. The proportion of histidine tRNA eluting in the second peak Teas greatly increased, however, when the tRNA was aminoacylated before chromatography. Since the magni- tude of l,his late peak depends on whcthcr or not the tRNA is charged, and since it decreases (with a corresponding increase in the early peak) as the column ages, me consider it to be an arti- fact of the column, rather than representing multiple isoaccept- ing spccics for histidinc t,RKh. The experiment also shows that the position of the early peak is sensitive to aminoacylation of t,he tRNA.

4 C. E. Singer, G. R. Smith, R. Cortese, and B. N. Ames, sub- mittcd to A’ature IVew Biol.

I

uncharged tRNAHiS i

1.0

charged tRNAH’* ;

,

E 06)

40 60 80 100 120

FRACTION NUMBER

FIG. 5. Comparison of the elution position of charged and un- charged tENAxis from h&T1504 on a reversed phase column No. 3. The applied sample contained 2 mg of tRNA from hisTl50Q charged with [“Hlhistidine, and 18 mg of the same tltNA carried through the identical charging procedure except that the histidine was omitted. The column was developed at, 37” with a gradient of NaCl in 10 mM MgCl* and 10 rnl\l sodium acetate, pH 4.5. The flow rate was 1.5 ml per min, and lo-ml fractions were collected. The tRNA in even fractiolrs was precipitated and collected on glass fiber filters to determine the elution positjion of charged tRNAHis. The tRNA of odd fractions was assayed by the standard procedure (“Materials and Methods”) to determine the elution position of uncharged tRNAnis. The recovery of counts applied to the col- umn was UC&, and the recovery of histidine acceptance was 94%. The peak tube of precharged tRNAHiB had 3069 cpm, and the peak tube of postcharged tl’LNAniS (50.~1 samples assayed) had 4710 cpm.

Absorbance at i

DISCUSSION

As a result of the experiments presented here, the function of three of the four histidine regulatory genes examined can be more clearly delineated: hisR appears to be a structural gene for histi- dine tRNA, and his?” and hisW appear to be genes involved with tRNA maturation. No positive evidence has been obtained re- garding the nature of hisU.

It had previously been shown by Silbert et al. that hisR mu- tants have 40 to 50 % less histidinc tRNA than the wild type (11). Those data have been essentially confirmed here, with reductions of from 25 to 45yo being found in the histidine tRNB content of hisR mutants. The reduced histidine tRNA content of hisR mu- tants shows that the hisR gene is required for formation of active histidine tRNA. The finding does not prove that hisR is a struc- tural gene for histidine tRNA, however, since the histidine tRNA levels could be reduced if hisR were a gene regulating the ex- pression of a structural gene located elsewhere on the chromo- some, or if hisR encoded a tRNA-modifying enzyme whose activ- ity was necessary to produce the full complement of active tRNA”“. These alternate possibilities are made less likely by the finding of an increnscd tRNd”” level in strains carrying an

by guest on March 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

ill. Brenner and B. N. Ames lOS7

additional wild type IzisZZ gene on an episome. This increase in the tRKA”‘” level suggcsta that even in the wild type Salmonella expression of the hisR gent must be rate limiting in the produc- tion of histidine tRKJ4. Such a rate-limiting role is not expected for a modifying enzyme or an unlinked regulatory unit.

Below, reasons will be given for bclicving that the hisR locus is the OIIIJ. one codiirg for histidine tRNl4. Accepting this to be true, it remains to csl)lnin the reductioii in tRNAHiS content of kisR mutants. We consider two possibilities likely. One pos- sibility is that the h,isR locus is comprised of two histidinc tRNA gents in tandem, similar to the case for tyrosinc tRNA (16). In this c:ibe the hisR mutations would result in one of the tandem gents being inact,ive. The other possibility is that then: is a single hisR gene, and that mutations in the gene render the tRNA\ unstable. Mutations in a t,yrosine suppressor tRNA gent of i?. coli have been found to have such an effect (17).

It was originally thought that hisU and hisW were also struc- tural genes for histidinc tRNA. As a result, a variety of chro- matographic systems have been employctl in an attempt to de- tect multiple apccics of histidinc tRN-2. In addit,ion to the studies reported hen, with the reversed 1)hase column No. 3 and benzoylated nE:,~E:~cc,lllllosc, less extensive analyses have been performed on rcrcrscd phase columns Nos. 1 and 2,5 and on col- umns of mcthylated albumin kieselguhr, L)E4E-cellulose-Scpha- des, llytlrosylal,atitc, and Scl)hadcs G-100 (11). In no case has evidence been obtniucd for more than one species of hi&dine tRKA. Singer and Smith’ of this laboratory have recently determined a unique scqucnce for the tRNAniS, and have set an upper limit for a possible second spccics as 157, of the total tRNAnis. It is possible that a minor species of histidine tRNA does exist, but that the fractionation methods used to date have failed t,o resolve it; however, other csl)crimcnts reported here and elsewhere indicate that hisU and IlisW, at least, arc not tRNA structural genes.

,4n association between hisW and histidine tRNA was first suggested by a cross-dominance study of 3Iartin et al. (15). They showed that an additional copy of a hisR+ gcnc (on an episomc) permitted repression in a kisW strain. Iu this paper we show that a cold-sensitive IzisW, strain JL250, has aberrant levels of several species of tRNA relative to the wild type. While this confirms a relationship between kism and tRNA, the pleiotropic effect argues against hi.sIV being a structural gene for tRNA”“. This is also indicated by the finding that a strain carrying an extra dose of the wild type /&TV gene on an epi- some retained the normal level of tRiY;iHi”. The plciotropic effect could result from the hisW gene being involved in tran- scription of tRNA genes. More likely is the suggestion that hisTV codes for a tRNh maturation enzyme. The cold sensi- tivity of an unmodified tRNA could be responsible for the im- mediate cessat,ion of growth when the mutant is shifted to 20”.

The remaining hisW mutants, and all the hisU mutants, were found to have normal levels of histidine tRNA. This result differs from those of a recent study of Lewis and Ames.” They found the level of histidine acceptance in hisU and hisW mu- tants to be reduced 20 to 30% relative to the acceptance of valine and arginine. In addition, they found the acceptance of valine and arginine to be reproducibly lowered about 20% relative to

5 IVI. Brenner, G. It. Fink, and B. N. Ames, unpublished experi- merits

6 .J. A. Lewis sllti U. N. Ames, (1972) J. Mol. briol., in press.

the AlGo of the f&U and 1aisW extracts. The conflicting results appear to be due to the different way in which the cells were proc- essed before the tRKA was extrsct.ed. The lnimar>- objective of the experiments of Lewis and Ames was to esamine tlie level of charging in vivo; consequently, cells were killed very rapidly by the addition of trichloroacetic acid to exponentially growing cul- tures. The preparations used in this paper were made from cells which were frozen slowly by being placed in a freezer at -20". Lewis and Ames have found that if they freeze cells in

a similar manner before preparing tRNA by their uhual method, then the relative acceptance of histidinc by hisU and hislV preparations approaches that of the wild type. A pos>ible es- planation for the change upon freezing would be that the hisli and kisTV lesions are lcaky, and that the mutant phenotype is repaired during the period of slow cooling. If such is the case, then the results reported by Lewis and Ames more correctly represent the tRNA concentrations in vivo, and the l&U gene, as well as the &TV gene, must be involved with tRXX produc- tion.

Evidcncc that hisT codes for a tRNA-modifying enzyme is now v-cry st,rong. Since hisT is known to specify a protein (18), the altered chromatographic mobility for hisT t,RK;I”‘” found here is likely R result of a change in a modified base. This srrg- gestion has prompted a comparison of the sequence of histidine tRNA from the wild type and from hisTl~O4. IUnpublishcd results* indicate that 2 pscudouridine residues found in the anticodon loop of the wild type remain as uridinr residues in the mutant; the pseudouridinc in the ribothymidiue-pscudouridine- cytosine loop, however, is intact.

Both this study and that of Lewis and Ames6 find a normal quantity of histidine tRNA to be present in hisT mutants. In addition, the tRNXHiS of hisT mutants gives the wild type R,,, and VmaX in the aminoacylation reaction2 and is charged i7~ vivo to the same extent as is the wild type.” ilpparcntly the hisT modification is required for charged histidinc tRKA to interact properly in the repression mechanism.

The hisT modification may also be required for regulation of other metabolic pathways. ElisT mutants are resistant to the lcucine analogues trifluoroleucinc and P-hydrosyleucine, and to the tyrosine analoguc amino tyrosine,4 suggesting that some clement of negative control of the two biosynthetic pathways has been lost. Significantly, a pseudouridine is found in the anti- codon loop of a tRNATY’ (19) and a tRNALe” (20) of E. coli, an organism which is a very close relative of S. typhimurium.

Acknowledgment-JVe arc happy to acknowledge the cheerful assistance of Mrs. W. L. Chang.

1.

2.

3. 4. 5.

F. 7.

8.

~)nLOR~~SZO,F.,STRAUS,D.S.,ilND hl;S,B.?;. (197%) J.,%Ol. Chem., 247, in press.

BRENNER, M., AND AMES, B. N. (1971) II. J. VOGEL (P;ditor), Melabolic Pathways V, Academic PI&, New York.

FINK. G. R.. AIiD lto~rr. J. R. (1968) ,/. Mol. EoZ.. 33. 547. Fas~&~us&, 1>. B. (1969) Genetics,‘61, ~17. ’ ROTH, J. It., AIST~S, D. N., .\piD HARTMAS, P. E. (1966) J. nfoZ.

Biol., 22, 305. ROTII, J. R., ASD AMES, B. ?j. (19GG) .I. Mol. Hiol., 22, 325. SCHLESII\TGER, S., AND Rl.~~nsasr~c, H. (19G4) J. Mol. Biol., 9,

670. DE LORENZO, F., AND AMES, B. iV. (1970) J. Biol. Chem., 246,

1710.

by guest on March 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

10% Histidine tRNA of the Regulatory Mutants Vol. 247, No. 4

0. BREXHLEY, J. E. (1970) Ph. D. thesis, University of Califor- nia, Davis.

10. VOGEL, H. J., .~ND BONNER, D. M. (1956) J. Biol. Chem., 218, 97.

11. SILBERT, D. F., FINK, G. R., AND AMES, B. N. (1966) J. Mol. Biol., 22, 335.

12. S.IRIN, P. S., -&ND ZAMECNIK, P. C. (1964) Biochim. Biophys. Acta, 91, 653.

13. WEISS, J. F., PEARSON, R. L., AND KELMERS, A. D. (1968) Bio- chemistry, 7, 3479.

14. NISHIMURA. S., AND NOVELLI, G. D. (1964) Biochim. Biophys. Acta, 80, 574.

15. MARTIN, R. G., BAGDASARIAN, M., AMES, B. N., AR’D ROTH, J. R. (1969) Fed. Eur. Biochem. Sot. Symp., 19, 1.

16. RUSSELL, R. L., ABELSON, J. N., LANDY, A., GEFTER, M. L., BRENNER, S., AND SMITH, J. D. (1970) J. Mol. Biol., 4’7, 1.

17. SMITH, J. D., BARNETT, L., BRENNER, S., AND RUSSELL, R. L. (1970) J. Mol. Biol., 64, 1.

18. CHANG, G. W., ROTH, J. R., AND AMES, B. N. (1971) J. Bacte- rioZ.;lO8, 4i0.

19. GOODMAN. H. M.. ABELSON. J.. LANDY. A.. BRENNER. S.. AND SMITH, j. D. (i968) Nat&e, i17, 1016 ’

I

20. DUBE, S. K., MARCHER, K. A., AND YUDELEVICH, A. (1970) Fed. Eur. Biochem. Sot. Lett., 9, 168.

by guest on March 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Michael Brenner and Bruce N. AmesRIBONUCLEIC ACID OF THE REGULATORY MUTANTS

: IX. HISTIDINE TRANSFERSalmonella typhimuriumHistidine Regulation in

1972, 247:1080-1088.J. Biol. Chem. 

  http://www.jbc.org/content/247/4/1080Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/247/4/1080.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on March 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from