effect of utp and gtp pools on attenuation at the pyre gene of escherichia coli

Mol Gen Genet (1987) 208:152-158 MGX © Springer-Verlag 1987

Effect of UTP and GTP pools on attenuation at the pyre gene of Escherichia coil

Peter Poulsen and Kaj Frank Jcnsen University Institute of Biological Chemistry B, Solvgade 83, DK-1307 Copenhagen K, Denmark

Summary. We have used the gaIK gene, minus its promoter, to quantitate transcription of the orfE--pyrE operon of Escherichia eoli in front of and after the intercistronic attenuator. Expression of the hybrid genes was studied in a bac- terium with mutations that permit changes in the UTP and GTP pools during exponential growth. It was found that the greater part of pyrE gene regulation by the nucleotides takes place at the intercistronic attenuator and that promoter control contributes only little, ca. twofold. When pools of both UTP and GTP were high only 5%-6% of the m R N A chains were continued into the pyre gene. However, when the UTP pool was reduced (from 1.3 to 0.2 gmol/g dry weight) nearly 100% of transcription passed the attenuator. Likewise, a reduction in the GTP pool (from 3.2 to 0.8 lamol/g dry weight) resulted in 25%-30% escape of attenuation. Regulation by attenuation disappeared when a premature stop-codon was introduced near the end of orfE such that translational coupling to transcription was prevented in the attenuator area. Therefore, we attribute the modulation of attenuation to nucleotide-induced variations in the kinetics of mRNA chain elongation. In support for this it was found that an RNA polymerase mutant with reduced RNA chain growth rate transcribed past the pyrE attenuator at a high frequency in the presence of a high UTP pool, but only when coupling of translation to transcription was allowed at the end of orfE.

Key words: Pyrimidine biosynthesis - pyrE gene - Attenua- tion - RNA polymerase mutants

Introduction

Six consecutive enzymatic reactions catalyse the de novo biosynthesis of UMP. In Escherichia eoli the corresponding genes are scattered on the bacterial chromosome (O'Dono- van and Neuhard 1970). The pyrE gene encoding the enzyme orotate phosphoribosyltransferase is located counter- clockwise at 81 map units (Poulsen et al. 1983) and is the second gene of a bicistronic operon, orfE--pyrE. The first gene (orfE) encodes a 238 amino acid protein of unknown function (Poulsen et al. 1984).

Expression of the pyrE gene is inversely correlated with the size of the cellular pool of UTP and is high when this pool is low and vice versa (Schwartz and Neuhard 1975). In addition guanine nucleotides, GTP or ppGpp, influence pyrE expression (Jensen 1979; Turnbough 1983).

Offprint requests to: P. Poulsen

Regulation by the UTP pool involves attenuation in the intercistronic orfE--pyrE region (Poulsen et al. 1983, 1984), and translation at the end of the orfE message, which ends eight nucleotide residues prior to the symmetric attenuator (Poulsen et al. 1984), is a prerequisite for attenuator modulatiofi by UTP (Bonekamp et al. 1984).

The regulatory signal is not the free UTP pool. Rather it is the mRNA chain growth rate dictated by the saturation of the elongating RNA polymerase by UTP. This conclu- sion was reached because a mutant of Salmonella typhimurium with a defect in binding nucleoside triphosphates at the elongation site (Jensen et al. 1986a) showed strongly enhanced levels of pyre and pyrBI gene expression in the presence of a high cellular concentration of UTP (Jensen et al. 1982, 1986a). Furthermore, since the codon composi- tion during leader translation influences the frequency of m R N A chain termination in front of pyrE (Bonekamp et al. 1985), the true regulatory signal appears to be the kineti- cally determined length of "naked" mRNA between the transcribing RNA polymerase and the first coupled ribo- some (Jensen et al. 1986b).

In previous studies the pyrE promoters were neglected. Attenuation was investigated using a hybrid gene transcribed from the lac promoter. The present study with the galK gene transcriptionally fused to the pyrE promoters both upstream and downstream of the attenuator permits evaluation of promoter regulation and attenuation indepen- dently. Furthermore, the m R N A chain attenuation frequencies under different growth conditions can be calculated. It was found that attenuation is responsible for the greater part of regulation by UTP and that a reduction in the pool of guanine nucleotides (GTP) also regulates pyrE gene expression by influencing the attenuation process in the intercistronic orfE--pyrE region.

Materials and methods

Materials. Restriction endonucleases were from Boehringer or New England Biolabs. T4 DNA ligase, DNA polymerase I large fragment (Klenow enzyme), and calf intestinal phos- phatase were obtained from Boehringer. Protein assay dye reagent was purchased from Bio-Rad. Nitrocefin was from BBL Microbiology System (USA). [l~C]-galactose (56.5 mCi/mmol) was from New England Nuclear.

Construction of bacterial strains. The strains used in the present work are listed in Table 1. The temperature-sensitive guaB mutant S01784 was selected as previously de-

153

Table 1. Strains and plasmids

Strain no. Genotype Source of reference

Escherichia coli K12 MC4100 araD139 A(lac)U169 rpsL thi S00003 F-, rpsL relA metB S01784 MC4100, A carAB guaB S01792 S01784, galK S03806 S01792, recA56 srlC3OO::TnlO S03811 S03806, 2PP101 S03812 S03806, 2PP102 N6455 F- thi-1 thr-1 leuB6 lacY1 tonA21 supE44 hsdR pro galK nad: : TnlO JCt0240 Hfr(PO45), thr-300 ilv-318 rpsE300 srlC3OO: : TnlO recA56

Salmonella typhimurium LT-2 SL4213 metA22 metE55 galE496 rpsL120 (ilv) xyl-404(Fels2)-Hl-b nml- H2-enx hsdL6 hsdSA29 KP1469 edd-9 cod-8 deoD201 udp-ll KP1475 KP1469, rpo-FU r KP1492 KP1469, rpo-FU s

M. Casadaban E. coli K12 58-161 This work This work This work This work This work M. Gottesman S. Short

B.A.D. Stocker K.F. Jensen et al. (1982) K.F. Jensen et al. (1982) K.F. Jensen et al. (1982)

Plasmid Relevant genotype Source or reference

pPP1 orfE + pyrE + Poulsen et al. (1983) pPP6 orfE + pyrE- Poulsen et al. (1984) pGD4 galK + Dandanell and Hammer (1985) pPP101 orfE' -galK This work pPP102 pyrE' --galK This work pPP112 pyrE ' -ga lK This work

scribed (Jensen 1979): after mutagenesis of strain S01781 with nitrosoguanidine a penicillin counterselection was performed at 40 ° C in minimal medium containing hypoxanthine. The surviving cells were plated on agar containing guanine. Colonies able to grow at 40 ° C when supplied with guanine or xanthine, but not with hypoxanthine, were picked. Subsequently the colonies were tested for temperature-sensitive growth in the presence of hypoxanthine.

S01792 was constructed from S01784 by Plvir transduction (Silhavy et al. 1984). First S01784 was transduced to G a l K - and tetracycline resistance by Plvir lysates prepared on N6455; next, S01792 was obtained by a new transduction using Plvir lysates from S00003, and selection for Nad + transductants. S03806, a RecA derivative of S01792, was constructed by Plvir transduction using JC10240 lysates. S03806 was lysogenized with 2PP101 and 2PP102 yielding S03811 and S03812, respectively, as described (Silhavy et al. 1984) using 2W30 (lb2, cI) as selector phage. The recA mutation ensured that 2 integrated only into the 2 attachment site of the chromosome. Single lysogens were scored by their sensitivity to 2W60 (2ci90, c17) in cross-streaking tests, using known single and multiple lysogens as controls (Shimada et al. 1972).

Recombinant D N A techniques. Plasmid DNA was isolated by the alkaline extraction method (Birnboim 1983); large- scale preparations were additionally purified by centrifugation in ethidium bromide/CsC1 gradients (Maniatis et al. 1982). DNA fragments were isolated from agarose gels using DEAE-cellulose membranes. Filling in of T-recessed ends resulting from cutting with restriction endonucleases by using Klenow enzyme were carried out in 30 gl volumes that contained: 0.5 gg DNA, 6 mM Tris-HC1 pH 7.5, 6 mM MgC12, 1 mM DTT, 0-50 mM NaC1, 60 gM dNTPs, and 0.5 units of Klenow enzyme. Incubation was for 15 min

at room temperature. Terminal 5' phosphates were removed from DNA by treatment with calf intestinal alkaline phos- phatase as recommended by the supplier. The procedure for linker tailing and ligation was that of Lathe et al. (1984). Transformation was carried out by the calcium chloride procedure (Maniatis et al. 1982). When plasmids propa- gated in E. coli were introduced into S. typhimurium, the strain SL4213 (restriction-, modification +) was used as the primary recipient. Plasmid constructions were verified by restriction mapping.

Plasmids. pPP1 (Poulsen et al. 1983) is a pBR322 derivative containing the entire pyrE operon on a 2 kb PvuII fragment inserted into the PvuII site of pBR322, pPP6 (Poulsen et al. 1984) contains orfE and approximately one-third of pyrE. pGD4 (Dandanell and Hammer 1985) is a derivative of the promoter cloning vector pKB-1 (McKenney 1982). pKB-1 is a modified version of pKB2000 (McKenney et al. 1981), having the 2tI terminator inserted upstream of the cloning region. The pKB vector system allows the in vivo transfer of promoter galK fusions to a 2 phage derivative (2galS) containing the entire gal operon (for details see McKenney et al. 1981).

Transfer o f galK gene fusions to ,~gal8. The procedure for transfer of promoter-galK fusions from plasmid to 2gal8 by in vivo recombination was that of McKenney et al. (1981). The resulting 2 phages were checked by restriction mapping.

Growth procedures. Cells were grown with shaking at 32 ° C in phosphate-buffered AB medium (Clark and Maaloe 1967) supplemented with 0.2% glucose, 0.1 gg/ml thiamine and 0.2% casamino acids. Guanine (30 gg/ml), uracil (20 lag/m1), uridine (40 gg/ml) or UMP (100 gg/ml) was added as indicated. For plasmid-containing cells 50 gg/ml

~54

ampicillin was used. Growth was monitored by measuring absorbance at 436 nm in an Eppendorf photometer.

Cell extracts and enzyme assays. At A436 = 1.0, cells were rapidly chilled, harvested by centrifugation, washed in AB basal medium, resuspended in lysis buffer (100 mM Tris- HC1 pH 7.9, 3 mM EDTA, 3 mM DTT), and disrupted by sonication. Cell debris was removed by centrifugation. Oro- tate phosphoribosyltransferase activity was assayed following the conversion of orotate to OMP by measuring the absorbance decrease at 295 nm as previously described (Poulsen et al. 1983). Aspartate transcarbamoylase activity was determined by the method of Gerhart and Pardee (1962). Galactokinase activity was assayed by measuring the accumulation of phosphorylated galactose in a radio- chemical assay. The procedure was that of McKenney et al. (1981) with the modifications introduced by Dandanell and Hammer (1985). Beta-lactamase activity was assayed as described by O'Callaghan et al. (1972); the hydrolysis of nitrocefin was followed by measuring the increase in absorbance at 482 nm in assays containing a total volume of 1 ml. This assay served for copy number correction by assuming a direct relationship between the/%lactamase level and the number of plasmids in the cell (Uhlin and Nordstr6m 1978).

One unit of enzyme activity is defined as the amount of enzyme catalysing the conversion of I nmol substrate to product per minute in the prescribed conditions. Protein concentration was determined by the Coomassie Blue dye- binding method (Bradford 1976), using bovine serum albu- min as protein standard.

Nucleotide pool determination. The cells were grown as described above and were labelled for two generations with [32p]orthophosphate (2Ci/mol) prior to sampling at OD¢36 =0.8. At this time 4.5 ml of the cultures were harvested on Millipore filters by vacuum filtration and rapidly suspended in 1 ml ice-cold 0.33 M formic acid. After 0.5 h cell debris and precipitated material were removed by centrifugation. The nucleotide pools were determined by two- dimensional thin layer chromatography on polyethylenei- mine-impregnated cellulose plates as described previously (Jensen et al. 1979).

Results

Construction of galK gene fusions

The galK gene minus its promoter was fused to transcription of the orfE-pyrE operon both in front of and after the intercistronic attenuator. The fusions were constructed using the vector pGD4 (Fig. 1), which permits transfer of the galK fusion to a lambda phage and integration in the bacterial chromosome (Dandanell and Hammer 1985).

The 760 bp PvuII-MluI restriction fragment, which contains the pyrE promoter region (Fig. 2) was isolated from plasmid pPP1 (Poulsen et al. 1984), blunt-ended with Klenow enzyme and inserted into the unique SmaI site of pGD4. This formed plasmid pPPl 01 (orfE' -- galK). Similar- ly, the 1390 bp PvuII-BamHI restriction fragment, which carries both the promoter region and the attenuator of the pyrE operon (Fig. 2), was cloned into the SmaI site of pGD4, resulting in plasmid pPPlO2(pyrE'--galK). The galK fusions were transferred from the plasmids to 2gal8 by homologous recombination and the two phages, 2PP101

Hind TIf SmaI

f ,~ti "B-region" _gaLK

~ ~ a pGD4 5.5 kb

EcoRF

ori !,

Fig. t. Structure of pGD4. The vector is a member of the pKO-1 family of vectors (McKenney et al. 1981) (see Materials and methods). The thin line represents the EeoRI-PvuII fragment of pBR322 containing the bla gene. The heavy lines represent, respectively, the galK gene and the "B-region" containing bacterial DNA upstream of the gal operon. The tI terminator of lambda is indicated. The direction of transcription of galK, bla, and ori is shown by arrows

pPP102 f 1390bp

pPP101 F 760 bp

Pvufl Mlul Ava I Bam HI

Att

mRNA

Fig. 2. Construction of orfE'--galK and pyrE'-galK transcrip- tional fusions. The structure and regulation of the pyrE operon is outlined: the orfE gene product is expressed from transcripts initiated at the promoter region (PI and P2), while the pyrE gene product is expressed from the transcripts that result from read- through of the attenuator (Att) in the intercistronic region of the operon. The fragments cloned into the transcription fusion vector pGD4 (Fig. 1) and the resulting plasmids are indicated. In addition, the AvaI site used in the construction of pPP112 is shown

and 2PP102, were integrated in the 2 attachment site of strain Sg3806, a recA derivative of S01792 (Table 1). The resulting strains, S03811 (orfE'--gaIK) and S133812 (pyrE'-galhO, were employed in the analysis of promoter regulation and attenuation at the pyrE operon.

Another plasmid, pPPI12, was made analogous to pPP102 (pyrE'--galK), but with stop codons in all three translational reading frames at the end of orfE. This open reading frame (orfE) serves as a "leader protein" for the pyrE attenuator (Fig. 3B) (Poulsen et al. 1984; Bonekamp et al. 1984). Plasmid pPP6 (Poulsen et al, 1984) was opened at its unique AvaI site at the end of orfE (Fig. 2). The protruding ends were blunt-ended with Klenow polymerase and linker tailed with the symmetric 22-mer oligonucleotide d (AGATCTAATCTAGATTAGATCT) that contains stop

codons at the indicated positions (***). The PvuII--BamHI

A P2

rACIGCGGATCATAGACGIICCIGTIIATAAAA~GGIGGAAGGAIIATAGCC -lOO * -80 . -60

Pl A[CGAIGCCTTGrAAGGATAGGAATAACCGCCGGAAGTCCGFAIAArGCGC~CA

• -40 • -20 • ÷1

B A P a l

ATCTTGTTGGCTCTGGCCCGAGGGGGAATCGAATCCATTGTAGCGACGCAGAAGGCGGCG I L e L e u L e u A ~al, euA 1.aA r'f?G ZyG LIt I LeG Lu[;er ' l Le Va LA ~a'l 'hrG 7.nLysA ~aA ~a

CTGGCAAACTGATTTTTAAGGCGACTGATGAGTCGCCTITTTTTTGTCTGTAGAAAAGTA

AGATGAGGAGCGAAGGCATG M e 6 . . . ( p y r E )

Fi~. 3A, B. Promoter and attenuator regulatory regions of the pyrE operon. A The DNA sequence of the promoter regulatory region indicating the Pribnow boxes (underlined) and the transcription start sites (encircled}. Previous experiments revealed that Pl is the major promoter of the operon (Poulsen et al. 1984). Numbers give position relative to the first transcribed nucleotide ( + I) from promoter PI. B The DNA sequence of the intercistronic region of the pyrE operon drawn to indicate the position of the coding regions relative to the attenuator structure, orfE ends at a TGA codon (***) just prior to the dyad symmetry of the attenuator (arrows), while pyrE starts at the ATG methionine 32 nucleotides downstream from the row of eight Ts. The position of the AvaI site used in the construction of pPP112 is also shown

fragment of the l igated plasmid, pPP62, was isolated and inserted in the SmaI site of p G D 4 as described above, thereby forming pPP112.

Manipulation of nucleotide pools

The background for all physiological experiments was strain S01792 (Table 1) which requires pyrimidine (and arginine) because of the carAB deletion and shows a par t ia l requirement for guanine due to the guaB(ts) mutat ion. By growing the strain in the presence or absence of guanine, and with either uracil, uridine or 5 ' U M P as source of pyrimidine nucleotide synthesis, both the UTP pool and the GTP pool could be varied (Table 2). Note fur thermore that both a stress in the supply of guanine nucleotides and of uddine nucleotides tends to reduce the pool of ppGpp .

155

Regulation of pyrE gene expression by UTP and GTP pools

Table 3 (lines 1-3) shows the effect of the UTP pool on t ranscript ion of the orfE-pyrE operon. Strains S03811 (orfE' -galK) and S03812 (pyrE' -galK) were grown in medium containing either uracil, uridine, or U M P as pyrimidine source. Guanine was present in order to have normal pools of guanine nucleotides in these experiments.

Pyrimidine l imitation, established by feeding U M P , resulted in a 1.6-fold increase in the level of galactokinase in S03811 (orfE'--galK), indicating a minor effect of UTP concentrat ion on t ranscr ipt ion initiation. The level of pyrE- encoded orotate phosphoribosyl t ransferase, however, was increased 32-fold in U M P - g r o w n cells relative to that observed in cells grown with uridine. This high regulatory increase was also seen for the synthesis of galactokinase in strain S03812 (pyrE'-galK), where the fusion poin t is after the a t tenuator (Table 3). F r o m the levels of galactokinase in the two strains (Table 3) it was calculated that regulat ion of pyrE gene expression in response to a fall in the UTP pool from 1.3 gmol /g (with uridine) to 0.2 ~tmol/ g (with UMP) is the composite result of 1.6- to 1.8-fold p romoter s t imulat ion and 15- to 20-fold increase in transcription pas t the at tenuator . In the presence of uridine 5 % - 6 % of the R N A polymerases continue pas t the pyrE attenuator, but with U M P as source of pyrimidine nucleotide synthesis close to 100% of all t ranscribing R N A polymerase molecules continue into the structural pyrE gene. Uraci l does not boost the UTP pool as much as uridine (Table 2) and gives rise to ca. 11% escape of a t tenuat ion (Table 3).

Compar i son of lines 1 and 4 in Table 3 reveals the effect of a l imited synthesis of guanine nucleotides on pyre operon expression. Strains S03811 and S03812 were grown with uridine as pyrimidine source either in the presence (Table 3, line 1) or in the absence (Table 3, line 4) of guanine where the pool of G T P is reduced 2.5-fold (Table 2). F r o m the levels of galactokinase in the cells it is clear that the total 17-fold regulat ion of pyrE gene expression stems from 2.3-fold st imulat ion of p romote r activity when GTP synthesis is l imited and 8- to 9-fold increase in t ranscr ipt ion past the at tenuator . The increase in a t tenua tor read- through in this experiment is p robab ly due to reduction in both the GTP and the UTP pools (Table 2). A similar experiment was performed in the Tris-buffered medium of Edlin and

Table 2. Pools of ribonucleotides in S01792

Addition Doubling time Pools (min)

GTP ATP CTP UTP ppGpp a

(pmol/g dry weight)

Uridine, guanine 40 1.23 1.96 0.66 1.30 ND Uracil, guanine 40 1.20 2.03 0.53 0.96 0.07 UMP, guanine 76 4.39 5.09 0.48 0.20 0.01 Uridine 68 0.49 9.19 0.44 0.87 ND Uracil 71 0.46 6.72 0.33 0.73 0.03

S01792 was grown at 32 ° C in AB medium supplemented with glucose, thiamine, casamino acids, guanine, and the indicated pyrimidine source (see Materials and methods)

a Because of the low specific activity of 3zp radioactivity in AB medium the determination of small ppGpp pools was rather uncertain. However, in the presence of both uracil and guanine a distinct ppGpp spot was visible in the autoradiogram. This spot could not be seen when either uracil or guanine was omitted

156

Table 3. Enzyme levels in S03811 and S03812 containing orfE'--galK and pyrE'--galK fusions

Addition S03811 (orfE' -galK) S03812 (pyrE' -galK) Chromosomal pyrE

Galacto- Promoter Galacto- Total Attenuator OPRTase kinase regulation kinase regulation read-through" (units/rag) (units/rag) ( x control) (units/mg) ( x control)

Regulation ( x control)

Uridine, guanine 9.3 1.0 0.57 1.0 6% 9 1.0 Uracil, guanine 9.3 1.0 1.02 1.8 11% 23 2.5 UMP, guanine 15.6 1.6 15.5 27 99% 286 32 Uridine 21.4 2.3 11.2 20 52% 150 17

Stb3811 and S03812 were grown at 32°C in AB medium supplemented with glucose, thiamine, casamino acids and the indicated purine and pyrimidine sources. The crude extracts were assayed for galactokinase and orotate phosphoribosyltransferase (OPRTase) as described in Materials and methods. Since the specific activities of OPRTase in S03811 and S03812 did not differ significantly, when grown under the same conditions, the average specific activities are given

Percentage attenuator read-through is calculated as the ratio (galactokinase level in S03812/galactokinase level in S03811) x 100

Table 4. Effect of guanine limitation on enzyme levels in S03811 and S03812 grown in Tris-buffered medium

Table 5. Enzyme levels in S01792 harbouring plasmids pPP101, pPP102 and pPPll2

Addition OPRTase" Galactokinase Pools b (gmol/g) Addition Galactokinase (units/mg) in OPRTase" (units/mg) (units/mg) in (units/rag)

pPPI01 pPPI02 pPPI12 S03811 S03812 GTP UTP

Uracil, guanine 361 57 (15%) 24 (5%) 24 Guanine 9 10.6 1.0 3.28 1.69 Uracil 608 395 (52%) 13 (1%) 111 None 36 23.6 5.7 0.81 1.89 UMP, guanine 566 519 (85%) 73 (7%) 294

The strains were grown at 32 ° C in Tris-buffered medium supplemented with glucose, thiamine, arginine, and uracil; guanine was added when indicated

As in Table 3 the average specific activities of orotate phosphoribosyltransferase (OPRTase) are shown b Pools were determined in S01792 and are given as micromoles per gram (dry weight) of bacteria

Maa loe (1966) using uracil as source of pyrimidine. In this case omit t ing guanine from the medium caused a slight increase in the U T P pool, and then 24% read- through of t ranscr ipt ion past the a t t enua tor was seen (Table 4).

Effect of translation on attenuation at pyrE

The regulat ion pa t te rn seen above using lysogenic l ambda- phages as carriers of the gene fusions was also observed with the corresponding mul t icopy plasmids, pPP101 (orfE'--galK) and pPPI02 (pyrE'-galK) (Table 5). How- ever, with plasmid p P P l l 2 , where the fusion point to the gaIK gene is after the pyrE a t tenua tor (as for pPP102), but where the leader t ransla t ion is in terrupted 59 nucleotide residues earlier than the native end of the orfE gene, only ca. twofold s t imulat ion of galK expression was seen when the UTP pool was reduced. This parallels the p romote r regulat ion seen with pPP101 (orfE'--galK) and corresponds to ca. 7% escape o f a t tenuat ion regardless of the U T P pool with this plasmid. When the G T P pool was reduced no s t imulat ion of galK expression was seen with p P P l l 2 . In contrast , t ranscr ipt ional read- through past this " n a k e d " a t tenua tor was reduced from 5% to 1% when the pool of G T P was lowered. We can give no good explanat ion, for this potent ial ly interesting result. However, in any case, the behaviour of p lasmid p P P I I 2 shows that t ransla t ion at the end o f orfE is needed for increase in t ranscr ipt ion

The crude extracts of the three plasmid-harbouring strains grown at 32 ° C in supplemented AB medium were assayed for galactokinase and orotate phosphoribosyltransferase (OPRTase) activity as described in Materials and methods. Numbers in parentheses indicate the attenuator read-through calculated as the ratio: [galactokinase directed by pPP102 (or pPPll2)/galactokinase directed by pPP101] x 100. The specific activity of galactokinase was normal- ized relative to the specific activity of fl-lactamase prior to calcula- tion of attenuator read-through

a Since the chromosomally encoded OPRTase in the three plasmid-containing strains did not differ significantly, when grown under the same conditions, the average OPRTase specific activities are given

past the pyrE a t tenua tor above the 5 % - 7 % seen with normal nucleotide pools (Table 5).

Increased read-through of the pyrE attenuator by a mutant RNA polymerase

With plasmids pPP101, pPP102 and pPP112 to hand, we used the oppor tuni ty to analyse the pa t te rn of pyrE gene expression in KP1475, an rpoBC mutan t of S. typhimurium, and its parent KP1469(rpo+). The mutan t shows high levels of pyrBI and pyrE gene expression in the presence of a high internal concentrat ion of UTP (Jensen et al. 1982, 1986a). The KP1475 R N A polymerase elongates m R N A chains slowly both in vivo and in vitro due to defective binding of nucleoside t r iphosphates at the elongat ion site (Jensen et al. 1986a). Experiments were also carried out with KP1492, a pseudorever tant of KP1475 which still shows high pyrE expression. The results are listed in Ta- ble 6.

Compar i son of the galactokinase levels in KP1469, KP1475 and KP1492 harbour ing pPP101, pPP102 and p P P l l 2 shows that the activity of the orfE promoters is

157

Table 6. Enzyme levels in RNA polymerase mutants harbouring plasmids pPP101, pPP102 and pPP112

Strain Galactokinase (units/rag) in OPRTase a (units/rag)

pPP101 pPP102 pPPI12

KP1469 1736 182 (10%) 130 (8%) 12 wt

KP1475 3597 1536 (40%) 155 (7%) 307 rpo-FU r

KP1492 1532 1174 (60%) 155 (9%) 204 rpo-FU ~

The crude extracts of the plasmid-harbouring strains grown at 37 ° C in O3P medium supplemented with glucose, thiamine, uracil, and ampicillin, were assayed for galactokinase and orotate phosphoribosyltransferase (OPRTase) activity as described in Materials and methods. Numbers in parentheses indicate the attenuator read- through calculated after normalization relative to the specific activity of fl-lactamase

a Since the chromosomally encoded OPRTase in each of the three strains was not affected by the presence of plasmids pPP101, pPP102, and pPPt12, the average OPRTase specific activities are given

affected only little by the RNA polymerase mutations, but that transcription past the attenuator is four- to sixfold higher in the RNA polymerase mutants than in the parent strain. However, this requires coupled translation at the end of the orfE message, since the level of galactokinase encoded by plasmid pPP l l2 is low in both mutant and parent strains. Thus, the mutant RNA polymerases are not defective in the recognition of the termination signal per se. Rather their behaviour of reading past the attenuator is a consequence of altered kinetic properties, which tends to tighten the coupling between transcription and translation.

Since the pyre attenuator regions of S. typhimurium and E. colt, although qualitatively similar in organization, differ somewhat from each other in details (Poulsen et al. 1984; Neuhard et al. 1985) parallel regulation of the galK fusion on the plasmids and the chromosomal pyre gene cannot be expected in this experiment and is not seen.

Discussion

In this study we used the galK gene, encoding the enzyme galactokinase, to investigate transcription frequencies in the orfE--pyrE operon of E. colt both upstream of and downstream from the intercistronic attenuator. In contrast to previous experimentation (Poulsen et al. 1984; Bonekamp et al. 1984) this approach permits evaluation of both promoter control and attenuation in regulation of pyre gene expression.

When cells were grown with UMP as pyrimidine source they contained 30-35 times more of the pyre gene product, orotate phosphoribosyltransferase, than after growth on uridine. This regulation, also seen for the galK gene transcriptionally fused after the pyrE attenuator, appeared to be composed of 1.6- to 1.8-fold stimulation of promoter activity and 15- to 20-fold higher frequencies of mRNA chain elongation past the attenuator. The read-through of the attenuator approached 100% when the UTP pool was low.

We also studied the effect of reduced synthesis of guanine nucleotides on transcription of the operon. This treatment had previously been found to enhance pyrE gene expression in S. typhimurium (Jensen 1979). As observed for the UTP pool, a limitation in the synthesis of GTP causing pool reduction to 25% of normal, enhanced pyrE promoter activity by approximately twofold and increased transcription past the attenuator to ca. 30% of the initiated m R N A chains.

The minor promoter control can perhaps be attributed to secondary changes in the concentration of ppGpp (see below). However, we ascribe regulation by attenuation to altered mRNA chain elongation kinetics stemming from the perturbed pools of the nucleoside triphosphates, UTP and GTP, since this control disappeared when the ribosomes were prevented from following closely after RNA polymerase to the attenuator. The elongating RNA polymerase has very high Kms for UTP and GTP (Kingston et al. 1981 ; Jensen et al. 1986a), and it is therefore conceiv- able that reductions in the cellular pools of these two nucleotides will cause RNA polymerase to elongate the mRNA chains more slowly. Thereby the coupling between transcription and translation will become tighter and the ribosomes more prone to prevent the pyre attenuator from adopting the terminating conformation in the m R N A chain.

A mutant RNA polymerase of S. typhimurium showing an increased Km for saturation of the elongation site by nucleoside triphosphates and a reduced RNA chain growth rate given normal nucleotide pools, causes highly increased levels of pyrBI and pyre gene expression in the presence of a high cellular concentration of UTP (Jensen et al. 1982, 1986a). This mutant was analysed using the described fusions of galK to parts of the E. colt orfE--pyrE operon. The mutant RNA polymerase showed a considerably enhanced frequency of transcription elongation past the pyre attenuator in vivo but this behaviour depended upon an intact orfE reading frame allowing the coupled ribosomes to follow RNA polymerase close to the attenuator. There- fore, the mutant enzyme is not intrinsically unable to recog- nize the termination site. Rather its behaviour of reading past the pyre attenuator is due to the altered kinetic properties which are expected to tighten the coupling between RNA polymerase and the ribosomes.

We do not know much about the physiological signifi- cance of the small (twofold) promoter regulation observed. However, both Levin and Schachman (1985) and Roland et al. (1985) noted a similar regulation of attenuatorless galK and lacZ fusions to the pyrBI promoters of E. colt. We observed that both limitation in the UTP pool and the GTP pool tends to reduce the cellular concentration of ppGpp (Table 2). Furthermore, the defective RNA polymerase mutant of the S. typhimurium strain KP1475, which showed a slightly enhanced mRNA chain initiation frequency at pyre promoters, also contains less ppGpp (Jensen et al. 1982). The stringent hypothesis interprets a high pool of ppGpp as a signal for amino acids being limiting for macromolecular synthesis (Gallant 1979). Conversely, a very low ppGpp pool might perhaps be interpreted as a signal for amino acids being present in excess over the nucleotide substrates for RNA synthesis. The major promoter (P1) of the orfE--pyrE operon (Fig. 3 A) contains four GC pairs between the - 1 0 region and the m R N A start point. This sequence motif has been found in several other pro-

158

rooters negatively affected by ppGpp (Travers 1984). These include the carAB upstream promoter (Bouvier et al. 1984) and the p y r B I promoter region (Turnbough 1983).

Bouvier et al. (1984) have reported that the carAB upstream promoter is repressed by pyrimidines, and they have proposed the existence of a pyrimidine repressor. Our experiments do not indicate the involvement of a repressor in the regulation of the p y r E promoters. However, a faint

-10 homology to part of a repeated sequence (GAAT-

+1 G C C G C C G ) in the upstream promoter of carAB suggested by Bouvier et al. (1984) to participate in the recognition of the putative repressor is present in promoter P1 in front

--31 --20

of orfE ( G A A T A A C C G C C G ) (Fig. 3 A). The interpreta- tion of this homology must await mutat ional analysis.

Acknowledgements. We thank Gert Dandanell and Karin Hammer for donating the fusion vector pGD4 and for helpful advice. The valuable technical help of Lise Schack is appreciated. We also thank Torkild Christensen at Nordic Gentofte for making the oligonucleotide linker with the stop codons. The work was supported by grants from the Danish Natural Science Council.

R e f e r e n c e s

Birnboim HC (1983) A rapid alkaline extraction method for the isolation of plasmid DNA. Methods Enzymol 100:243-255

Bonekamp F, Clemmesen K, Karlstr6m O, Jensen KF (1984) Mechanism of UTP-modulated attenuation at the pyrE gene of Escherichia coli: an example of operon polarity control through the coupling of translation to transcription. EMBO J 3 : 2857-2861

Bonekamp F, Andersen HD, Christensen T, Jensen KF (1985) Codon-defined ribosomal pausing in Escherichia coli detected by using the pyrE attenuator to probe the coupling between transcription and translation. Nucleic Acids Res 13 : 4113-4123

Bouvier J, Patte JC, Stragier P (1984) Multiple regulatory signals in the control region of the Escherichia coli carAB operon. Proc Natl Acad Sci USA 81:4139-4143

Bradford MM (1976) A rapid and sensitive method for quantita- tion of microgram quantities of protein using the principle of protein-dye binding. Anal Biochem 72:248-254

Clark D J, Maaloe O (1967) DNA replication and the division cycle of Escherichia coli. J Mol Biol 23 : 99-112

Dandanell G, Hammer K (1985) Two operator sites separated by 599 base pairs are required for deoR repression of the deo operon. EMBO J 4:3333-3338

Edlin G, Maaloe O (1966) Synthesis and breakdown of mRNA without protein synthesis. J Mol Biol 15:428-434

Gallant JA (1979) Stringent control in E. coli. Annu Rev Genet 13:393-415

Gerhart JC, Pardee AB (1962) The enzymology of control of feed- back inhibition. J Biol Chem 237 : 891-896

Jensen KF (1979) Apparent involvement of purines in the control of expression of Salmonella typhirnuriurn pyr genes: Analysis of a leaky guaB mutant resistant to pyrimidine analogs. J Bac- teriol 138 : 731-738

Jensen KF, Houlberg U, Nygaard P (1979) Thin-layer chromato- graphic methods to isolate 3zP-labeled 5-phosphoribosyl-l- pyrophosphate (PRPP) : Determination of cellular PRPP pools and assay of PRPP synthetase activity. Anal Biochem 98 : 254-263

Jensen KF, Neuhard J, Schack L (1982) RNA polymerase involvement in the regulation of expression of Salmonella typhimuriurn pyr genes. Isolation and characterization of a fluorouracil- resistent mutant with high constitutive expression ofpyrB and pyre genes due to a mutation in rpoBC. EMBO J 1 : 69-74

Jeusen KF, Fast R, Karlstr6m O, Larsen JN (1986a) Association

of RNA polymerase having increased Km for ATP and UTP with hyperexpression of the pyrB and pyrE genes of Salmonella typhimurium. J Bacteriol 166:857-865

Jensen KF, Bonekamp F, Poulsen P (1986b) Attenuation at nucleotide biosynthetic genes and amino acid biosynthetic operons of Escherichia coli. Trends Biochem Sci 11:362-365

Kingston RE, Nierman WC, Chamberlin MJ (1981) A direct effect of guanosine tetraphosphate on pausing of Escherichia coli RNA polymerase during RNA chain elongation. J Biol Chem 256 : 2787-2797

Lathe R, Kieny MP, Skory S, Lecocq JP (1984) Linker tailing: Unphosphorylated linker oligonucleotides for joining DNA ter- mini. DNA 3:173-182

Levin HL, Schachman HK (1985) Regulation of aspartate transcarbamoylase synthesis in Escherichia coli: Analysis of deletion mutations in the promoter region of the pyrBI operon. Proc Natl Acad Sci USA 82:4643-4647

Maniatis T, Fritsch EF, Sambrook J (1982) Molecular cloning. A laboratory manual. Cold Spring Harbor Laboratory Press, New York

McKenney K (1982) PhD Thesis, The Johns Hopkins University, Baltimore, MD

McKenney K, Shimatake H, Court D, Schmeissner U, Brady D, Rosenberg M (1981) A system to study promoter and terminator signals recognized by Escherichia coli RNA polymerase. In: Chirikjian JC, Papas TS (eds) Gene amplification and analysis, vol II : Analysis of nucleic acids. Elsevier-North Holland, Amsterdam, pp 383-415

Neuhard J, Stauning E, Kelln RA (1985) Cloning and characterization of the pyrE gene and of pyrE::Mudl(Aplac) fusions from Salmonella typhirnuriurn. Eur J Biochem 146:597-603

O'Callaghan CH, Morris A, Kirby SM, Shingler AH (1972) Novel method for detection of/Mactamases by using a chromogenic cephalosporin substrate. Antimicrob Agents Chemother 1 : 283-288

O'Donovan GA, Neuhard J (1970) Pyrimidine metabolism in mi- croorganisms. Bacteriol Rev 34:278-343

Poulsen P, Bonekamp F, Jensen KF (1984) Structure of the Escherichia coli pyrE operon and control of pyrE expression by a UTP modulated intercistronic attenuation. EMBO J 3:1783-1790

Poulsen P, Jensen KF, Valentin-Hansen P, Carlson P, Lundberg LG (1983) Nucleotide sequence of the Escherichia coli pyrE gene and of the DNA in front of the protein-coding region. Eur J Biochem 135:223-229

Roland KL, Powell FE, Turnbough CL Jr (1985) Role of translation and attenuation in the control ofpyrBI operon expression in Escherichia coli K-12. J Bacteriol 163:991-999

Schwartz M, Neuhard J (1975) Control of expression of the pyr genes in Salmonella typhirnuriurn: Effects of variations in uridine and cytidine nucleotide pools. J Bacteriol 121:814-822

Shimada K, Weisberg RA, Gottesman ME (1972) Prophage lambda at unusual chromosomal locations. I. Location of secondary attachment sites and the properties of the lysogens. J Mol Biol 63:483-503

Silhavy TJ, Berman ML, Enquist LW (1984) Experiments with gene fusions. Cold Spring Harbour Laboratory Press, New York

Travers AA (1984) Conserved features of coordinately regulated E. coli promoters. Nucleic Acids Res 12:2605-2618

Turnbough CL Jr (1983) Regulation of Escheriehia coli aspartate transcarbamylase synthesis by guanosine tetraphosphate and pyrimidine ribonucleoside triphosphates. J Bacteriol 153:998-1007

Uhlin BE, Nordstr6m K (1978) A runaway-replication mutant of plasmid R1 drd-19: Temperature-dependent loss of copy number control. Mol Gen Genet 165:167-179

Communicated by A. B6ck

Received November 26, 1986

effect of utp and gtp pools on attenuation at the pyre gene of escherichia coli

Documents