Control of the Escherichia coli rrnB P1 Promoter Strength by ppGpp

Fusions of the rrnB P1 and P2 promoters, and of the tandem P1-P2 combination, to a wild-type lacZ gene were constructed on plasmids and recombined into the mal region of the bacterial chromosome, close to the normal location and in the normal orientation of rrnB. The upstream activator region (Fis-binding sites) was always present with the P1 promoter, and all constructs contained the box A antitermination site of rRNA genes. Using these constructs, β-galactosidase specific activities were measured in Escherichia coli strains carrying either both ppGpp synthetases, PSI and PSII (relA+ spoT+), or only PSII (ΔrelA spoT+), or neither (ΔrelA ΔspoT), using different media supporting growth rates between 0.6 and 2.8 doublings/h at 37 °C. The β-galactosidase activities were used to estimate the relative strength of the rrnB P1 promoter in comparison to the isolated rrnB P2 promoter. Promoter strength (transcripts initiated per min per promoter per free RNA polymerase concentration) was distinguished from promoter activity (transcripts initiated per min per promoter). In ppGpp-synthesizing (wild-type) bacteria, the relative strength of the rrnB P1 promoter increased nearly 10-fold with increasing growth rate from 0.17 to 1.5, but in the ppGpp-less double mutants it decreased by 20% from 1.7 to 1.5. Thus, at low or zero levels of ppGpp, the P1 promoter was 1.5–1.7 times stronger than the isolated P2 promoter. These results indicate that the normal growth rate control of the rrnB P1 promoter strength requires ppGpp, and that the strength is reduced at basal levels of ppGpp found during exponential growth. No additional ppGpp-independent control of the rrnB P1 promoter strength was evident. From the β-galactosidase data and previously determined values of rRNA gene activities, the activities of the isolated rrnB P1 and P2 promoters, and of the P2 promoter in the tandem combination, were estimated. With increasing growth rate, the activity of the isolated P2 promoter increased 6-fold from 6 to 33 initiations/min, while the activity of the isolated P1 promoter increased 24-fold from 2 to 54 initiations/min. The increasing activity of the isolated P2 promoter is assumed to reflect the increasing RNA polymerase concentration at constant promoter strength, whereas the steeper increase in P1 promoter activity reflects increases in both polymerase concentration and promoter strength. When in tandem with P1, the P2 promoter activity is inferred to decrease as the P1 promoter activity increases. Fusions of the rrnB P1 and P2 promoters, and of the tandem P1-P2 combination, to a wild-type lacZ gene were constructed on plasmids and recombined into the mal region of the bacterial chromosome, close to the normal location and in the normal orientation of rrnB. The upstream activator region (Fis-binding sites) was always present with the P1 promoter, and all constructs contained the box A antitermination site of rRNA genes. Using these constructs, β-galactosidase specific activities were measured in Escherichia coli strains carrying either both ppGpp synthetases, PSI and PSII (relA+ spoT+), or only PSII (ΔrelA spoT+), or neither (ΔrelA ΔspoT), using different media supporting growth rates between 0.6 and 2.8 doublings/h at 37 °C. The β-galactosidase activities were used to estimate the relative strength of the rrnB P1 promoter in comparison to the isolated rrnB P2 promoter. Promoter strength (transcripts initiated per min per promoter per free RNA polymerase concentration) was distinguished from promoter activity (transcripts initiated per min per promoter). In ppGpp-synthesizing (wild-type) bacteria, the relative strength of the rrnB P1 promoter increased nearly 10-fold with increasing growth rate from 0.17 to 1.5, but in the ppGpp-less double mutants it decreased by 20% from 1.7 to 1.5. Thus, at low or zero levels of ppGpp, the P1 promoter was 1.5–1.7 times stronger than the isolated P2 promoter. These results indicate that the normal growth rate control of the rrnB P1 promoter strength requires ppGpp, and that the strength is reduced at basal levels of ppGpp found during exponential growth. No additional ppGpp-independent control of the rrnB P1 promoter strength was evident. From the β-galactosidase data and previously determined values of rRNA gene activities, the activities of the isolated rrnB P1 and P2 promoters, and of the P2 promoter in the tandem combination, were estimated. With increasing growth rate, the activity of the isolated P2 promoter increased 6-fold from 6 to 33 initiations/min, while the activity of the isolated P1 promoter increased 24-fold from 2 to 54 initiations/min. The increasing activity of the isolated P2 promoter is assumed to reflect the increasing RNA polymerase concentration at constant promoter strength, whereas the steeper increase in P1 promoter activity reflects increases in both polymerase concentration and promoter strength. When in tandem with P1, the P2 promoter activity is inferred to decrease as the P1 promoter activity increases. Transcription of rRNA and tRNA ("stable RNA") genes in E. coli is initiated from two tandem promoters, about 120 bp 1The abbreviations used are: bpbase pair(s)UARupstream activator region apart, named P1 and P2 (Gilbert et al., 1979Gilbert S.F. de Boer H.A. Nomura M. Cell. 1979; 17: 211-224Abstract Full Text PDF PubMed Scopus (64) Google Scholar. Upstream of P1, binding sites for the DNA-bending protein Fis (Nilsson et al., 1992Nilsson L. Verbeek H. Vijgenboom E. van Drunen C. Vanet A. Bosch L. J. Bacteriol. 1992; 174: 921-929Crossref PubMed Google Scholar; Finkel and Johnson, 1992Finkel S.E. Johnson R.C. Mol. Microbiol. 1992; 6: 3257-3265Crossref PubMed Scopus (248) Google Scholar; Gosink et al., 1993Gosink K.K. Ross W. Leirmo S. Osuna R. Finkel S.E. Johnson R.C. Gourse R. J. Bacteriol. 1993; 175: 1580-1589Crossref PubMed Google Scholar in the upstream activator region (UAR) enhance the strength of the P1 promoter (Gourse et al., 1986Gourse R.L. de Boer H.A. Nomura M. Cell. 1986; 44: 197-205Abstract Full Text PDF PubMed Scopus (200) Google Scholar. When rrnB or rrnE P1 and P2 promoters are isolated and linked to lacZ, β-galactosidase activity expressed from the P1 promoter increases with increasing growth rate of the bacteria, while the activity expressed from the P2 promoter remains constant (Gourse et al., 1986Gourse R.L. de Boer H.A. Nomura M. Cell. 1986; 44: 197-205Abstract Full Text PDF PubMed Scopus (200) Google Scholar. Thus, rrn P1 promoter activity, but not rrn P2 activity, is under growth rate control. A similar conclusion has been reached earlier from direct observations on electrophoresis gels of transcripts initiated at the P1 and P2 promoters of a shortened rrnA gene present on multicopy plasmids (Sarmientos and Cashel, 1983Sarmientos P. Cashel M. Proc. Natl. Acad. Sci. U. S. A. 1983; 80: 7010-7013Crossref PubMed Scopus (53) Google Scholar; Sarmientos et al., 1983Sarmientos P. Sylvester J.E. Contente S. Cashel M. Cell. 1983; 32: 1337-1346Abstract Full Text PDF PubMed Scopus (163) Google Scholar. base pair(s) upstream activator region By removing upstream or downstream sequences of the rrnB P1 promoter, it was demonstrated that neither upstream (e.g. UAR) nor downstream (e.g. antitermination) sites affect the growth rate control of the rrn P1 promoter, suggesting that the regulation takes place at the P1 promoter itself (Gourse et al., 1986Gourse R.L. de Boer H.A. Nomura M. Cell. 1986; 44: 197-205Abstract Full Text PDF PubMed Scopus (200) Google Scholar. This notion is supported by work from another laboratory in which 1 bp in the GCAC sequence following the –10 region of the rrnB P2 promoter was changed to produce the sequence GCGC, as in the rrnB P1 promoter. This base substitution confers a growth rate regulation to the normally unregulated P2 promoter which is identical to that of the P1 promoter (Zacharias et al., 1989Zacharias M. Göringer H.U. Wagner R. EMBO J. 1989; 8: 3357-3363Crossref PubMed Scopus (55) Google Scholar. The DNA sequence downstream of the –10 region of Escherichia coli promoters has been proposed to be part of a "discriminator" responsible for the control of rRNA synthesis by the effector nucleotide guanosine tetraphosphate, or ppGpp (Travers, 1980Travers A. J. Bacteriol. 1980; 41: 973-976Crossref Google Scholar). rRNA promoters are also subject to "stringent control," characterized by a reduction in rRNA synthesis in response to amino acid starvation (Stent and Brenner, 1961Stent G. Brenner S. Proc. Natl. Acad. Sci. U. S. A. 1961; 47: 2005-2014Crossref PubMed Scopus (231) Google Scholar). The stringent response is associated with an activation of ppGpp synthetase I (PSI, relA gene product). This activation requires binding of uncharged tRNA to the A site of the ribosome and leads to an accumulation of high levels of ppGpp (Haseltine and Block, 1973Haseltine W. Block R. Proc. Natl. Acad. Sci. U. S. A. 1973; 70: 1564-1568Crossref PubMed Scopus (350) Google Scholar). Cytoplasmic levels of ppGpp may also be moderately raised by induction of a weak stringent response, using low concentrations of an inhibitor of isoleucyl-tRNA synthetase. This increase in ppGpp levels is accompanied by a reduction in the rate of stable RNA synthesis (rs) relative to the rate of total transcription (rt); i.e. there is a reduction in the ratio rs/rt (Baracchini and Bremer, 1988Baracchini E. Bremer H. J. Biol. Chem. 1988; 263: 2597-2602Abstract Full Text PDF PubMed Google Scholar). The relationship between various levels of ppGpp and rs/rt produced in this manner is the same as when ppGpp levels and rs/rt are measured during exponential growth in different media (Ryals et al., 1982Ryals J. Little R. Bremer H. J. Bacteriol. 1982; 151: 1261-1268Crossref PubMed Google Scholar. Evidently, a unique relationship exists between the cytoplasmic concentration of ppGpp and rs/rt. The proportion of total transcription that is mRNA equals the difference (1 – rs/rt). For example, if 60% of total transcription is stable RNA synthesis (rs/rt = 0.6), then the remaining 40% represent mRNA synthesis (rm/rt = 0.4). Therefore, the level of ppGpp appears to determine the distribution of transcribing RNA polymerases between stable and mRNA genes. This has suggested that stringent and growth rate control of stable RNA synthesis reflect the same ppGpp-mediated inhibition of transcription from stable RNA promoters (Ryals et al., 1982Ryals J. Little R. Bremer H. J. Bacteriol. 1982; 151: 1261-1268Crossref PubMed Google Scholar; Baracchini and Bremer, 1988Baracchini E. Bremer H. J. Biol. Chem. 1988; 263: 2597-2602Abstract Full Text PDF PubMed Google Scholar; Hernandez and Bremer, 1993Hernandez V.J. Bremer H. J. Biol. Chem. 1993; 268: 10851-10862Abstract Full Text PDF PubMed Google Scholar). Similar conclusions have been reached from direct observations of transcripts initiated at rrnA promoters at different levels of ppGpp (Sarmientos and Cashel, 1983Sarmientos P. Cashel M. Proc. Natl. Acad. Sci. U. S. A. 1983; 80: 7010-7013Crossref PubMed Scopus (53) Google Scholar). Like rs/rt, β-galactosidase activity expressed from stable RNA promoters also reflects growth rate-dependent changes in the synthesis rate of bulk mRNA. The specific activity of β-galactosidase (enzyme activity per total protein) depends on the synthesis rates, lacZ-mRNA per total mRNA (rlac/rm), and if β-galactosidase is expressed from a stable RNA promoter, then rlac/rm corresponds to rs/rm (see "Discussion" in Hernandez and Bremer, 1990Hernandez V.J. Bremer H. J. Biol. Chem. 1990; 265: 11605-11614Abstract Full Text PDF PubMed Google Scholar). In ppGpp-less bacterial mutants with deletions in both ppGpp synthetase genes (relA and spoT; Xiao et al., 1991Xiao H. Kaiman M. Ikehara K. Zemel S. Glaser G. Cashel M. J. Biol. Chem. 1991; 266: 5980-5990Abstract Full Text PDF PubMed Google Scholar), the rates of stable and mRNA synthesis both increase in parallel as the growth rate increases, so that the distribution of RNA polymerase between stable and mRNA genes (rs/rt) remains constant (Hernandez and Bremer, 1993Hernandez V.J. Bremer H. J. Biol. Chem. 1993; 268: 10851-10862Abstract Full Text PDF PubMed Google Scholar). In such strains the specific activity of β-galactosidase expressed from an rrn P1 promoter also remains constant, despite an increasing transcription of stable RNA genes at increasing growth rates (Hernandez and Bremer, 1993Hernandez V.J. Bremer H. J. Biol. Chem. 1993; 268: 10851-10862Abstract Full Text PDF PubMed Google Scholar). Although these observations are consistent with a growth rate control of stable RNA synthesis by ppGpp, an alternative interpretation has been proposed, namely that the control of stable RNA synthesis is normal in ppGpp-less bacteria (Gaal and Gourse, 1990Gaal T. Gourse R. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 5533-5537Crossref PubMed Scopus (80) Google Scholar), but obscured by an altered control of mRNA synthesis. In other words, the correlation between ppGpp levels and stable RNA synthesis observed during exponential growth might result from an effect of ppGpp on the synthesis of mRNA which then indirectly causes the apparent growth rate control of rs/rt and β-galactosidase activity expressed from stable RNA promoters. Similar ideas have been proposed by Jensen and Pedersen, 1990Jensen K.F. Pedersen S. Microbiol. Rev. 1990; 64: 89-100Crossref Google Scholar, based on reports that ppGpp affects transcriptional pausing, and thereby the rate of RNA chain elongation, in vitro. New measurements of in vivo RNA chain elongation rates at different levels of ppGpp appear to be consistent with this proposal (Vogel and Jensen, 1994Vogel U. Jensen K.F. J. Mol. Biol. 1994; 236: 441-454Crossref PubMed Scopus (57) Google Scholar). The experiments described here were designed to distinguish between these alternatives; i.e. to determine whether ppGpp directly inhibits transcription from rrn P1 promoters, or whether it only affects mRNA synthesis. With regard to control of transcription, promoter activity and promoter strength have to be distinguished. Promoter activity is defined as the rate of transcript initiation at a given promoter, whereas promoter strength is defined as the rate of initiation per concentration of free RNA polymerase. A constitutive promoter that lacks a specific control has a constant promoter strength as long as it is not saturated with RNA polymerase, but it does not necessarily show a constant activity. Rather, unregulated promoters are expected to be transcribed at an increasing rate with increasing growth rate as a result of an increasing RNA polymerase concentration. In contrast, a promoter whose strength is specifically controlled by a repressor, or by ppGpp, would show the combined effects of this control and changing RNA polymerase concentrations. In the current work, rrnB P1 and P2 promoters were linked to lacZ, similar as in the work of Gourse et al., 1986Gourse R.L. de Boer H.A. Nomura M. Cell. 1986; 44: 197-205Abstract Full Text PDF PubMed Scopus (200) Google Scholar described above. Transcription from these promoters was studied in wild-type ppGpp-synthesizing, as well as in ppGpp-less (ΔrelA ΔspoT) bacteria. In the ratio of β-galactosidase activities expressed from these two promoters, effects of ppGpp on the synthesis and translation of bulk mRNA cancel. Therefore, this ratio measures the strength of the rrn P1 promoter relative to the strength of the P2 promoter. In this manner it was established that the relative strength of the rrnB P1 promoter increases with increasing growth rate in ppGpp-synthesizing (spoT+) bacteria, but not in ppGpp-less bacteria. In bacteria devoid of ppGpp, the P1/P2 activity ratio remained always high. This indicates that the growth rate control of the strength of the rrn P1 promoter requires ppGpp. In the absence of ppGpp, no additional factors that might be involved in the growth rate control of rrn P1 promoters, like Fis or HN-S (Nilsson et al., 1992Nilsson L. Verbeek H. Vijgenboom E. van Drunen C. Vanet A. Bosch L. J. Bacteriol. 1992; 174: 921-929Crossref PubMed Google Scholar; Wagner, 1994Wagner R. Arch. Microbiol. 1994; 161: 100-109Crossref PubMed Scopus (139) Google Scholar), were apparent. Growth Conditions—For construction of rrnB-promoter-box BAC-λ'-lacZ-kan fusions (Fig. 1), bacterial strains were grown at 37 °C in LB medium or on LB agar with 25 µg/ml kanamycin, 15 µg/ml tetracycline, 50 µg/ml ampicillin, or 20 µg/ml chloramphenicol as required, and on MacConkey-agar with maltose or lactose (Miller, 1972Miller J. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1972Google Scholar) when appropriate. For physiological experiments, cultures were grown at 37 °C in Medium C (Helmsetter, 1967Helmsetter C. J. Mol. Biol. 1967; 24: 417-427Crossref Scopus (128) Google Scholar), supplemented with 0.2% succinate, glycerol, or glucose, with or without 0.75% Difco casamino acids, or in LB medium (Miller, 1972Miller J. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1972Google Scholar) plus 0.2% glucose, as indicated. Experimental cultures were started by at least a 1/1000 dilution of the overnight culture. Growth was monitored turbidometrically at 600 nm (1-cm light path). Plasmid Constructions—The rrnB P-λ′-lacZ+-kan fusions (Fig. 1) were constructed on plasmids in several steps, using plasmids from different sources as described in Table I. The strong rRNA promoters could not be cloned on plasmids together with the strong lacZ ribosome binding site unless a spacer of about 1 kilobase was present between the transcription and translation starts. This was not due to excess β-galactosidase activity: even if 96% of lacZ sequences were deleted and the lacZ ribosome binding site was immediately followed by rRNA transcription terminators, plasmids could not be obtained. The nature of this spacer did not matter, as long as it did not contain promoters or binding sites for proteins that affect the translation efficiency. In the final constructs, a promoter- and ribosome binding site-less (1120-bp EcoRI-BamHI) DNA fragment of phage λ (bp 21226–22346) was found to be satisfactory. A replacement of the truncated lacZ gene in initial constructs by a full-length lacZ gene was only tolerated by the host bacterium if the plasmid copy number was reduced. This was achieved by including an incompatible (pBR322-derived) "helper plasmid" without a kanamycin resistance marker and with a different molecular weight (for later separation of the DNAs) in the cell. In the presence of kanamycin, cell survival depended on the presence of the helper plasmid. To suppress integration of the β-galactosidase-expressing plasmid into the host chromosome, a recA host was used. This improved the yield of the plasmid during DNA preparations.Table IPlasmids usedPlasmidSize (kb)GenotypeReference, or constructionpVJH2712.8bla′malE-rrnB UAR P1-ϕX174E′-′lacZ-T1T2-kan-T1T2-malK′Hernandez and Bremer, 1990Hernandez V.J. Bremer H. J. Biol. Chem. 1990; 265: 11605-11614Abstract Full Text PDF PubMed Google ScholarpKT077.8blarrnB UAR P1-P2-boxBAC-16S′-23S′-5S-T1T2K. Tedin. A 1250-bp EcoRI fragment and a 418-bp HincII-PvuII fragment in the 16 S and 23 S regions were deleted from plasmid pNO1301 (Jinks-Robertson et al., 1983Jinks-Robertson S. Gourse R.L. Nomura M. Cell. 1983; 33: 865-876Abstract Full Text PDF PubMed Scopus (140) Google Scholar carrying the whole rrnB genepKT146.4blarrnB P2-boxBAC-16S′-23S′-5S-T1T2K. Tedin. A 1421-bp PvuII-DdeI fragment of pKT07, carrying rrnB UAR-P1, was deleted and the plasmid was religated using EcoRI linkerspDM412.8bla′malE-rrnB P2-ϕX174E′-lacZ-T1T2-kan-T1T2-malK′D. Murray. The EcoRI rrn P1 promoter fragment of pVJH27 was exchanged for the rrn P2 fragment of pKT14pRS41510.8blarrnB (T1)4-lacZYA-tet"Simons et al., 1987Simons R.W. Houman F. Kleckner N. Gene (Amst.). 1987; 53: 85-96Crossref PubMed Scopus (1301) Google ScholarpXZ0212.7bla′malE-rrnB P2-lacZ-T1T2-kan-T1T2-malK′A 2061-bp EcoRI-SacI fragment of pRS415 carrying the N-terminal region of lacZ was substituted for the corresponding portion in pDM4pXZ0712.7Same as pXZ02EcoRI site in pXZ02 between rrnB P2 and lacZ was changed to XmnIpXZ0812.7Same as pXZ02EcoRI site in pXZ02 between kan and malK′ was changed to XmnIpXZ0911.8bla′malE-(EcoRI)-lacZ-T1T2-kan-T1T2-malK′A 922-bp EcoRI fragment of pXZ08 carrying rrnB P2 was deleted.pXZ4112.6bla′malE-box BAC-lacZ-T1T2-kan-T1T2-malK′A 860-bp BamHI-BstUI fragment from pXZ07 carrying box BAC but not P2 was ligated to an EcoRI linker at the BstUI end and inserted into the EcoRI-BamHI cloning site in pXZ09 (between malE′ and lacZ)pXZ4413.6bla′malE-rrnB P1-box BAC-lacZ-T1T2-kan-T1T2-malK′A 950-bp EcoRI fragment from pVJH27 carrying the rrnB UAR-P1 promoter was inserted into the unique EcoRI site of pXZ41pXZ4513.3bla′malE-rrnB P1P2-box BAC-lacZ-T1T2-kan-T1T2-malK′A 1568-bp BanI-EcoRI fragment from pKT07, carrying the rrnB UAR-P1P2 promoter, was inserted into the unique EcoRI site of pXZ09 after addition of an EcoRI linker to the Barel endpXZ5012.9bla′malE-λ′-lacZ-T1T2-kan-T1T2-malK′A 1120-bp EcoRI-BamHI fragment of phage λ (21226–22346) was inserted into the EcoRI-BamHI cloning site of pXZ09 (between malE′ and lacZ)pXZ5113.9bla′malE-rrnB UAR-P1-box BAC-λ′-lacZ-T1T2-kan-T1T2A 1734-bp PstI-MscI fragment was removed from pXZ50 and replaced by a 2772-bp PstI-BsaAI fragment of pXZ44, carrying ′malE-rrnB UAR-P1pXZ5213.0bla′malE-rrnB P2-box BAC-λ′-lacZ-T1T2-kan-T1T2A 1734-bp PstI-MscI fragment was removed from pXZ50 and replaced a 1896-bp PstI-BsaAI fragment from pXZ02, carrying ′malE-rrnB UAR-P2 with boxes BACpXZ5313.7bla′malE-rrnB UAR-P1P2-box βαC-λ′-lacZ-T1T2-kan-T1T2A 1734-bp PstI-MscI fragment was removed from pXZ50 and replaced a 2542-bp PstI-BsaAI fragment from pXZ45, carrying ′malE-rrnB UAR-P1-P2 with boxes BAC Open table in a new tab Bacterial Strains—Bacterial strains used and their constructions are described in Table II. The plasmid-borne rrnB promoter-lacZ fusions were recombined into the malB locus of the recBC sbc strain JC9387, selecting for kanamycin resistance, as described previously (Hernandez and Bremer, 1990Hernandez V.J. Bremer H. J. Biol. Chem. 1990; 265: 11605-11614Abstract Full Text PDF PubMed Google Scholar). This recombination made JC9387 mal–, which was checked on MacConkey-maltose plates (Miller, 1972Miller J. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1972Google Scholar). From JC9387, the fusion genes were transduced with phage P1 into derivatives of E. coli K12 strain MC4100 or the E. coli B/r A strain HB181 (Table II). Both MC4100 and HB181 are deleted for the lac genes. Therefore, these strains were mal+ lac– before transduction and became mal– lac+ after the transduction. This was verified on MacConkey-maltose and MacConkey-lactose plates.Table IIBacterial strainsStrainGenotypeReference, or constructionJC9387recB21 recC22 sbcB15 sbcC201 tkr-1 ara-14 leuB6 lacY1 Δ(gpt-proA)62 tsx-33 galK2 hisG4 rfbD1 mgl-51 rpsI-31 kdgK51 xyl-5 mtl-1 argE3 thi-1Winans et al., 1985Winans S.C. Elledge S.J. Krueger J.H. Walker G.C. J. Bacteriol. 1985; 161: 1219-1221Crossref PubMed Google ScholarCF1693MG1655, Δrelα251::kan ΔspoT207::catXiao et al., 1991Xiao H. Kaiman M. Ikehara K. Zemel S. Glaser G. Cashel M. J. Biol. Chem. 1991; 266: 5980-5990Abstract Full Text PDF PubMed Google ScholarMC4100araD Δ(argF-lac)U169 rpsL150 relA1 flbB5301 deoCl ptsF25 rbsRCasadaban, 1976Casadaban M.J. J. Mol. Biol. 1976; 104: 541-555Crossref PubMed Scopus (1305) Google ScholarVH2732MC4100, malB::malE′-rrnB UAR-P1-ϕX174 E′-′lacZ-kan-′malK ΔrelA251::kan argA::Tn10Hernandez and Bremer, 1991Hernandez V.J. Bremer H. J. Biol. Chem. 1991; 266: 5991-5999Abstract Full Text PDF PubMed Google ScholarXZ132MC4100, relA+ argA−This study. The relA+ allele of E. coli K12 (wild-type) was transduced with phage P1 into MC4100, selecting for Tn10 (Tcr) present in argA of the donor, and screening for the relA+ phenotype. The strain was then cured of Tcr (see text).XZ209XZ132, malB::malE′-rrnB UAR-P1-box BAC-λ′-lacZ-kan-′malKThis study. First, JC9387 was transformed with pXZ51, selecting for Kmr-kan was then transduced from JC9387 into XZ132, using phage P1XZ210XZ132, malB::malE′-rrnB P2-box BAC-λ′-lacZ-kan-′malKThis study. Same as XZ209, but using plasmid pXZ52XZ211XZ132, malB::malE′-rrnB UAR-P1-P2-box BAC-λ′-lacZ-kan-′malKThis study. Same as XZ209, but using plasmid pXZ53XZ251XZ209, ΔrelA251::kan argA::Tn10This study. The ΔrelA251 allele from VH2732 was P1-transduced into XZ209, selecting for Tcr (argA::Tn10) and screening for RelA phenotype (see text).XZ252XZ210, ΔrelA251::kan argA::Tn10This study. Same as XZ251, but using XZ210 as a recipientXZ253XZ211, ΔrelA251::kan argA::Tn10This study. Same as XZ252, but using XZ211 as a recipientXZ255XZ251, ΔspoT207::catThis study. The ΔspoT207 allele from CF1693 was P1-transduced into XZ251, selecting for cat and screen for loss of prototrophyXZ256XZ252, ΔspoT207::catThis study. Same as XZ255, but using strain XZ252 as a recipientHB181E. coli B/r A malK+ Δ(argF-lacIOZYA)phe(Am)thr(Am)hsd/r K-12Y. C. Xu. The lac deletion of MC4100 was transduced into a E. coli B/r A derivative, using a linked Tn10 and then curing the Tcr on Bochner plates (see text).XZ213HB181, malB::malE′-rrnB UAR-P1-box BAC-λ′-lacZ-kan-′malKThis study. kan from XZ209 was transduced into HB181, using phage P1XZ214HB181, malB::malE′-rrnB P2-box BAC-λ′-lacZ-kan-′malKThis study. Same as XZ213, but using XZ210 as a donorXZ231HB181, malB::malE′-rrnB UAR-P1-P2-box BAC-λ′-lacZ-kan-′malKThis study. Same as XZ213, but using XZ211 as a donor Open table in a new tab To transduce a wild-type relA+ allele into MC4100 (which is relA1; see Table II), a phage P1 lysate was first prepared from VH2732 (Table II), carrying Tn10 in the argA gene near ΔrelA::kan (=relA251). Tn10 was then transduced into a K12 wild-type strain (MG1655), selecting for Tcr and screening for Kms. The RelA+ phenotype was verified by the near UV method (Little et al., 1983Little R. Mou S.-W. Bremer H. J. Bacteriol. 1983; 155: 1426-1428Crossref PubMed Google Scholar. From this strain, Tn10 was further transduced into MC4100, selecting for Tcr. By screening the transductants for cotransduction of the RelA+ phenotype with the near UV method, the wanted relA* derivative of MC4100 was obtained. This strain was cured of Tn10 on Bochner plates (Bochner et al., 1980Bochner B.E. Huang H.C. Schieven G.L. Ames B.N. J. Bacteriol. 1980; 143: 926-933Crossref PubMed Google Scholar, and the cured (Tcs) strain was designated XZ132 (Table II). Similar procedures, using an appropriate strain from the Singer collection (Singer et al., 1989Singer M. Baker T.A. Schnitzler G. Deischel S.M. Goel M. Dove W. Jaacks K.J. Grossman A.D. Erikson J.W. Gross C.A. Microbiol. Rev. 1989; 53: 1-24Crossref PubMed Google Scholar, were used to transduce the lac deletion of MC4100 into a derivative of E. coli β/T (HB181, Table II). β-Galactosidase Assays—β-Galactosidase was assayed as described (Miller, 1972Miller J. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1972Google Scholar) with the following modifications. Four 10-μl samples of culture were taken at an A600 between 0.04 and 0.4 and mixed into 40 μl of permeabilization solution (Solution A: 0.8 mg/ml hexadecyltrimethylammonium bromide, 0.4 mg/ml sodium deoxycholate, 200 mM Na2HPO4, 20 mM KCl, 2 mM MgSO4, and 5.4 μl/ml β-mercaptoethanol). The permeabilization mixture was kept from 30 min to 2 h at 30 °C; then 950 μl of 30 °C prewarmed substrate mixture (Solution B: 60 mM Na2HPO4, 40 mM NaH2PO4, 10 mM KCl, 20 μg/ml hexadecyltrimethylammonium bromide, 10 µg/ml sodium deoxycholate, 1 mg/ml o-nitrophenyl-β-D-galactopyranoside, and 2.7 μl/ml β-mercaptoethanol) was added to initiate the reaction. After 1–3 h at 30 °C, reactions were terminated by the addition of 1 ml of 1 M Na2CO3. Enzyme activities were expressed as the rate of change in A420/h per A600 unit of culture present in the assay. To convert these units into Miller units (Miller, 1972Miller J. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1972Google Scholar), they must be divided by 60 (min/h) and multiplied times 1000. For example, in Fig. 2, A and B, 300 units on the ordinate correspond to 5000 Miller units. In all experiments β-galactosidase activity was measured over a period of 1 to 3 generations in the exponential cultures and was found to increase in parallel with the A600 of the culture. rrnB Promoter lacZ Fusions—Fusions of parts of the rrnB promoter region with lacZ were constructed on plasmids and recombined into the mal region of the bacterial chromosome, close to the normal location and in the normal orientation of rrnB. These constructs were obtained in a similar manner as the rrnB P1-lacZ fusion described previously (Hernandez and Bremer, 1990Hernandez V.J. Bremer H. J. Biol. Chem. 1990; 265: 11605-11614Abstract Full Text PDF PubMed Google Scholar), but in contrast to the previous construct which had the weak ribosome binding site and 11 N-terminal amino acids of the ϕX174 E gene at the N terminus of LacZ, the new constructs have the over 100-fold stronger lacZ ribosome binding site and the N-terminal amino acids of wild-type β-galactosidase. In contrast to the enzyme activity derived from the previous construct, β-galactosidase expressed from the new rrn promoter-lacZ fusions was stable at 42 °C or in the presence of chloramphenicol (data not shown). All rrn promoter-lacZ fusions used in this work contain the antitermination site of rRNA genes (Fig. 1A: Li et al., 1984Li S. Squires C.L. Squires C.