Portal de Periódicos da CAPES

Suppression of leu-500 Mutation in topA + Salmonella typhimuriumStrains

1998; Elsevier BV; Volume: 273; Issue: 45 Linguagem: Inglês

10.1074/jbc.273.45.29929

1083-351X

Ming Fang, Hai-Young Wu,

RNA Interference and Gene Delivery

Suppression of leu-500 mutation in Salmonella typhimurium topA − strains has been one of the most fascinating examples for the DNA supercoiling effect on transcription initiation control. Previous studies have indicated possible involvement of transcription-driven DNA supercoiling in the activation of the leu-500 promoter in topA − strains. Our recent studies have shown that ilvIH transcription activity located 1.9 kilobase pairs upstream is the initial supercoiling signal forleu-500 activation via a promoter relay mechanism. In the present communication, we show that the ilvIH transcription activity-initiated promoter relay can result in leu-500activation in topA + strains. In addition, suppression of the chromosomal leu-500 mutation correlates with the transcription activities of ilvIH and leuO rather than the TopA level in thetopA + strain. It appears that theleu-500 suppression in a topA −strain is due to the constant ilvIH transcription activity. Suppression of leu-500 mutation in Salmonella typhimurium topA − strains has been one of the most fascinating examples for the DNA supercoiling effect on transcription initiation control. Previous studies have indicated possible involvement of transcription-driven DNA supercoiling in the activation of the leu-500 promoter in topA − strains. Our recent studies have shown that ilvIH transcription activity located 1.9 kilobase pairs upstream is the initial supercoiling signal forleu-500 activation via a promoter relay mechanism. In the present communication, we show that the ilvIH transcription activity-initiated promoter relay can result in leu-500activation in topA + strains. In addition, suppression of the chromosomal leu-500 mutation correlates with the transcription activities of ilvIH and leuO rather than the TopA level in thetopA + strain. It appears that theleu-500 suppression in a topA −strain is due to the constant ilvIH transcription activity. base pair(s) kilobase pair(s). The leu-500 mutation, originally isolated from 5-bromouracil-treated Salmonella typhimurium LT2, is an A to G transition at the Pribnow box of the promoter of the leucine operon. This mutation significantly reduces the promoter activity and results in leucine auxotrophy (1Margolin P. Genetics. 1963; 48: 441-447Crossref PubMed Google Scholar). The second mutation in an unlinked suppressor gene (supX), located in the cysteine Band tryptophan region of the S. typhimuriumchromosome, was subsequently shown to suppress the leu-500mutation (2Mukai F.H. Margolin P. Proc. Natl. Acad. Sci. U. S. A. 1963; 50: 140-148Crossref PubMed Google Scholar, 3Dubanau E. Margolin P. Mol. Gen. Genet. 1972; 117: 91-112Crossref PubMed Scopus (39) Google Scholar). The supX was later identified to be the mutation in topA (4Trucksis M. Golub E.I. Zabel D.J. Depew R.E. J. Bacteriol. 1981; 147: 679-681Crossref PubMed Google Scholar), the structural gene for topoisomerase I (TopA) which relaxes negative DNA supercoils (5Wang J.C. J. Mol. Biol. 1971; 55: 523-533Crossref PubMed Scopus (518) Google Scholar). The mutation in topA affects the overall DNA superhelicity which is primarily maintained by the counter enzymatic activities of TopA and gyrase in bacteria (6Gellert M. Proc. Natl. Acad. Sci. U. S. A. 1976; 73: 4474-4478Crossref PubMed Scopus (566) Google Scholar, 7DiNardo S. Voelkel K.A. Sternglanz R. Reynolds A.E. Wright A. Cell. 1982; 31: 43-51Abstract Full Text PDF PubMed Scopus (291) Google Scholar, 8Richardson S.M.H. Higgins C.F. Lilley D.M.J. EMBO J. 1984; 3: 1745-1752Crossref PubMed Scopus (82) Google Scholar) (reviewed in Refs. 9Gellert M. Annu. Rev. Biochem. 1981; 50: 879-910Crossref PubMed Scopus (858) Google Scholar and 10Wang J.C. Annu. Rev. Biochem. 1985; 54: 665-697Crossref PubMed Scopus (1644) Google Scholar). The hypernegative DNA supercoiling in the topA mutants was interpreted to be responsible for the restoration of transcriptional initiation from the mutant leu-500 promoter (4Trucksis M. Golub E.I. Zabel D.J. Depew R.E. J. Bacteriol. 1981; 147: 679-681Crossref PubMed Google Scholar). However, the suppression of the leu-500 mutation correlated only with the absence of TopA but not with the degree of overall negative superhelicity in S. typhimurium (11Richardson S.M.H. Higgins C.F. Lilley D.M.J. EMBO J. 1988; 7: 1863-1869Crossref PubMed Scopus (49) Google Scholar). When the minimalleu-500 promoter (−80 to +87 of the operon) was subcloned in an extrachromosomal DNA, the plasmid-borne leu-500promoter failed to be activated in the topA mutant (11Richardson S.M.H. Higgins C.F. Lilley D.M.J. EMBO J. 1988; 7: 1863-1869Crossref PubMed Scopus (49) Google Scholar). These findings have challenged the overall supercoiling explanation and raised the possibility that in addition to thetopA − genetic background, some cis-factors must be required for leu-500 activation. Transcription-driven supercoiling has been shown to be one of the possible cis-factors which are required for activating the leu-500 promoter. However, the transcription-driven supercoiling effect is usually short-ranged (<250 bp)1 (12Chen D. Bowater R. Dorman C.J. Lilley D.M.J. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 8784-8788Crossref PubMed Scopus (65) Google Scholar, 13Chen D. Bowater R.P. Lilley D.M.J. Biochemistry. 1993; 32: 13162-13170Crossref PubMed Scopus (24) Google Scholar, 14Tan J. Shu L. Wu H.-Y. J. Bacteriol. 1994; 176: 1077-1086Crossref PubMed Google Scholar). We have recently shown that the short-range supercoiling effect can be extended up to 400 bp when a stronger promoter, ptac, is used to drive the transcription-induced supercoiling. 2M. Fang and H.-Y. Wu, unpublished data. When searching for any upstream transcription activity responsible forleu-500 activation in the chromosomal context, we have subsequently demonstrated that ilvIH promoter activity located 1.9 kb upstream is responsible for activating theleu-500 promoter in a S. typhimurium topA − strain, CH582 (15Wu H.-Y. Tan J. Fang M. Cell. 1995; 82: 445-451Abstract Full Text PDF PubMed Scopus (44) Google Scholar). Further characterization has revealed that the long range (1.9 kb) interaction between theilvIH and leu-500 promoters is mediated by a promoter relay mechanism whereby transcription activity of theilvIH promoter activates the leuO promoter located within the 1.9-kb intervening sequence, and both theleuO promoter activity and the LeuO protein are required for subsequent leu-500 activation (16Fang M. Wu H.-Y. J. Bacteriol. 1998; 180: 626-633Crossref PubMed Google Scholar). Thus far both the absence of TopA and the promoter relay initiated byilvIH transcription activity are shown to be involved in leu-500 activation in the topA −strain. Mechanistically, it is important to understand how these two factors play roles in activating the leu-500 promoter. Our previous study showed that the mutation in either ilvIH orleuO promoter abolished leu-500 activation in atopA − strain (15Wu H.-Y. Tan J. Fang M. Cell. 1995; 82: 445-451Abstract Full Text PDF PubMed Scopus (44) Google Scholar, 16Fang M. Wu H.-Y. J. Bacteriol. 1998; 180: 626-633Crossref PubMed Google Scholar), suggesting that the absence of TopA (the topA − genetic background) itself is insufficient to activate the leu-500 promoter. In the present study, we examined the effect of the promoter relay alone on leu-500 activation in topA +strains. Strikingly, we found that the ilvIHactivity-initiated promoter relay can result in leu-500activation in topA + strains, while the absence of TopA enhances the promoter relay-mediated leu-500activation by two-fold. Furthermore, suppression of chromosomalleu-500 mutation in a topA + strain correlated only with the expression of ilvIH and leuO genes but not with the cellular TopA level. Theleu-500 activation in a topA − strain also correlated with constant ilvIH transcription activity as determined by Northern analysis. It appears that ilvIH was turned on which resulted in the suppression of leu-500mutation in topA − strains. We therefore conclude that the ilvIH transcription activity-initiated promoter relay mechanism plays a decisive role while TopA plays a negative regulatory role in leu-500 activation. The 3476-bpNotI-NotI fragment containing the β-galactosidase gene (lacZ) isolated from pSVβ (CLONTECH) was used to replace the 1291-bpHindIII-NdeI fragment of pWU804, pWU805, pWU807, and pWU804M (16Fang M. Wu H.-Y. J. Bacteriol. 1998; 180: 626-633Crossref PubMed Google Scholar) to generate plasmids, pWU804LZ, pWU805LZ, pWU807LZ, and pWU804MLZ (plasmid maps in Figs. 1 and 3). The lacZcoding region was, therefore, transcriptionally fused with theleu-500 promoter to report the promoter activity in these plasmids.Figure 3Promoter relay-mediated leu-500activation in CH601. The β-galactosidase activities of pWU804LZ-harboring CV468, CH582, and CH601 were assayed at OD650 of 0.8 as shown in panel A. The relevant regions in pWU804LZ, pWU805LZ, pWU807LZ, and pWU804MLZ are illustrated. The ilvIH promoter is deleted in pWU805LZ and is mutated (X) in pWU807LZ. The leuO promoter is mutated (X) in pWU804MLZ. Otherwise, these plasmids are identical. The β-galactosidase activities of CH601 carrying one of the above plasmids were assayed at OD650 of 0.8. and shown in panel B, columns 1, 2, 3, and 4, respectively.View Large Image Figure ViewerDownload Hi-res image Download (PPT) S. typhimurium LT2 derivatives: CH582, a ΔtopA2726 leu-500 ara9and its isogenic parental strain, CH601, leu-500 ara9 (8Richardson S.M.H. Higgins C.F. Lilley D.M.J. EMBO J. 1984; 3: 1745-1752Crossref PubMed Scopus (82) Google Scholar) were provided by Dr. David Lilley. CH601 was derived from PM596 as described previously (8Richardson S.M.H. Higgins C.F. Lilley D.M.J. EMBO J. 1984; 3: 1745-1752Crossref PubMed Scopus (82) Google Scholar). PM596, an araB9, leu-500strain, was derived from CV468, an araB9, gal205 S. typhimurium LT2 strain (3Dubanau E. Margolin P. Mol. Gen. Genet. 1972; 117: 91-112Crossref PubMed Scopus (39) Google Scholar). Both PM596 and CV468 were provided by Dr. Joseph Calvo. Unless stated otherwise, the strains were grown aerobically at 32 °C in the synthetic medium SSA without the leucine supplement (17Calvo J.M. Freundlich M. Umbarger H.E. J. Bacteriol. 1969; 97: 1272-1282Crossref PubMed Google Scholar). 40 μg/ml leucine was supplemented when necessary. Plasmids were transformed into bacteria by electroporation. Total RNA was isolated from cell cultures as described previously (15Wu H.-Y. Tan J. Fang M. Cell. 1995; 82: 445-451Abstract Full Text PDF PubMed Scopus (44) Google Scholar). RNA concentration was measured by the absorbance at 260 nm. 100 μg of total RNA per sample were fractionated in the denaturing agarose gel as described previously (18Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Smith J.A. Struhl K. Current Protocol in Molecular Biology. John Wiley & Sons, New York1997Google Scholar). The fractionated RNA was then transferred and immobilized to a nitrocellulose membrane (Schleicher & Schuell, BA85). The size and amount of ilvIH or leuO mRNA was determined by hybridizing the membrane with the specific 32P-labeled DNA probe at 42 °C overnight. The DNA probe for detectingilvIH mRNA was the 279-bpAccI-AccI fragment isolated from theilvI coding region. A synthetic 28-bp oligomer consisting of the DNA sequence downstream of the leuO transcription start site was used for detecting leuO mRNA. Nick translation was used to label the ilvIH probe. T4 kinase was used to end-label the leuO probe with [γ-32P]ATP. The hybridized nitrocellulose membrane was washed with high stringency buffer (15 mm NaCl, 0.5 mmNaH2PO4, 0.5 mmNa2HPO4, and 0.1% SDS) twice at 50 °C for 10 min each time. The hybridization results were visualized using autoradiography. Cells were washed and resuspended in Tris-Cl pH7.5, 1 mm EDTA, and 1 mmdithiothreitol. The resuspended cells were then sonicated on ice for 1 min in four 15-s pulses with 30 s of cooling time on ice between pulses. The protein concentration of each sample was determined by BCA assay (Pierce). 25 μg of total protein from each sample was loaded for 8% SDS-PAGE. Two identical gels were prepared. One of the gels was stained with Coomassie Blue to verify protein loading. The other gel was electroblotted to the nitrocellulose membrane. Escherichia coli DNA topoisomerase I antibody (obtained from Dr. Rolf Menzel and Dr. Haiyan Qi) was the primary antibody. A secondary peroxidase-linked anti-rabbit IgG antibody was used for ECL detection (Amersham Life Science Ltd). Due to the high sequence homology of topA between E. coli and S. typhimurium, the rabbit antibody raised against E. coliTopA antibody is equally efficient in immunologically detectingS. typhimurium TopA as previously stated (19Wang, J. C., and Liu, L. F. (1979) in Molecular Genetics (Taylor, J. H., ed) Part III, pp. 65–88, Academic Press, New YorkGoogle Scholar). 3J. C. Wang and L. F. Liu, unpublished data. β-Galactosidase assays were performed as described by Miller (20Miller J.H. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1972: 352-355Google Scholar) with slight modifications. At each time point, the growth of the bacterial cultures were stopped by placing aliquots of the culture in an iced H2O bath for a minimum of 20 min before proceeding with the assay. Cells (4.5 ml) were made permeable by mixing (10-s vortex) them with 0.5 ml of 10× Z buffer (formula as described in Miller (20Miller J.H. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1972: 352-355Google Scholar)), 175 μl of chloroform, and 100 μl of 0.1% SDS. After incubating at 28 °C for 5 min, the permeabilized cells were incubated with 1 ml of 8 mg/mlo-nitrophenyl-β-d-galactoside at 28 °C. The β-galactosidase kinetics was determined by measuring the activity at several time points during the incubation. At each time point, 1 ml of sample was transferred into a tube containing 416 μl of 1 m Na2CO3 stop solution. After centrifugation at 14,000 × g for 10 min, the supernatant was used to measure A 420 and A 550. The Miller units of the β-galactosidase activity were calculated using the formula in Miller's procedure (20Miller J.H. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1972: 352-355Google Scholar). The β-galactosidase activity was normalized according to the plasmid copy number as determined by Southern analysis. All reported β-galactosidase activities are the averages of data from three separate experiments. The standard deviation is included as theerror bar in the graphs. Using a plasmid model, we have demonstrated the long range interaction between the ilvIHand leu-500 promoters in CH582 (15Wu H.-Y. Tan J. Fang M. Cell. 1995; 82: 445-451Abstract Full Text PDF PubMed Scopus (44) Google Scholar) to explain the original observation of the suppression of leu-500 mutation in theS. typhimurium topA − strain (2Mukai F.H. Margolin P. Proc. Natl. Acad. Sci. U. S. A. 1963; 50: 140-148Crossref PubMed Google Scholar, 3Dubanau E. Margolin P. Mol. Gen. Genet. 1972; 117: 91-112Crossref PubMed Scopus (39) Google Scholar). The fact that leu-500 activation is mediated by the upstream promoter activity has led to the suggestion that topA −genetic background may not have a direct effect on leu-500activation. In other words, the ilvIHactivity-dependent leu-500 activation may also occur in topA + strains as long as the upstreamilvIH promoter is active. The plasmid pWU804LZ derived from the previously established pWU800 series (15Wu H.-Y. Tan J. Fang M. Cell. 1995; 82: 445-451Abstract Full Text PDF PubMed Scopus (44) Google Scholar) was used to test this possibility in both topA + and topA − strains, CV468 and CH582, respectively. The pWU804LZ-harboring cells were grown in SSA(−Leu) medium since the absence of leucine in the medium is required for activating theilvIH promoter (21Ricca E. Aker D.A. Calvo J.M. J. Bacteriol. 1989; 171: 1658-1664Crossref PubMed Google Scholar, 22Platko J.V. Willins D.A. Calvo J.M. J. Bacteriol. 1990; 172: 4563-4570Crossref PubMed Google Scholar, 23Wang Q. Sacco M. Ricca E. Lago C.T. DeFelice M. Calvo J.M. Mol. Microbiol. 1993; 7: 883-891Crossref PubMed Scopus (27) Google Scholar, 24Wang Q. Calvo J.M. EMBO J. 1993; 12: 2495-2501Crossref PubMed Scopus (95) Google Scholar). The β-galactosidase assay results showed that the plasmid-borne leu-500 promoter was active in CV468 when cells grew to mid-log phase, and leu-500 promoter activity was 2-fold higher in CH582 than that in CV468 (Fig. 1). This result indicated that, in the presence of active ilvIH transcription, theleu-500 mutation can also be suppressed in atopA + strain. The enhanced activation in thetopA − strain suggests that TopA plays a negative regulatory role in leu-500 activation. To further determine that leu-500 activation is mediated by the upstream ilvIH transcription activity, we measured the plasmid-borne leu-500 promoter activity in pWU805LZ, where the ilvIH promoter was deleted, in bothtopA + and topA − strains. In good agreement with the previously shown primer extension results (16Fang M. Wu H.-Y. J. Bacteriol. 1998; 180: 626-633Crossref PubMed Google Scholar), the deletion of ilvIH promoter abolishedleu-500 activation in CH582 (Fig. 1), indicating that the β-galactosidase level from the lacZ reporter provided a reliable indicator of the leu-500 promoter activity. As shown in the topA − strain, CH582, deletion of the ilvIH promoter also eliminated leu-500activation in CV468 (Fig. 1), suggesting that it is theilvIH transcription activity rather than thetopA − genetic background that plays a decisive role in leu-500 activation. The activation of the plasmid-borne leu-500 promoter in atopA + strain has prompted us to hypothesize that the chromosomal leu-500 promoter can also be activated in topA + strains as long as the ilvIHexpresses under certain growth conditions. This hypothesis was tested by monitoring the growth of CH601 in SSA(−Leu) medium. CH601, the parental strain of CH582, is considered to be leucine auxotrophic, because it carries the leu-500 mutation and the wild typetopA on chromosome (8Richardson S.M.H. Higgins C.F. Lilley D.M.J. EMBO J. 1984; 3: 1745-1752Crossref PubMed Scopus (82) Google Scholar, 11Richardson S.M.H. Higgins C.F. Lilley D.M.J. EMBO J. 1988; 7: 1863-1869Crossref PubMed Scopus (49) Google Scholar). The growth of CH601 in leucine-free medium (leucine prototrophy) is a good indicator of leu-500 activation in the topA +strain. Strikingly, we observed that CH601 grew in SSA(−Leu) after an unusually prolonged lag phase (16 h after 1:250 inoculation or 19 h after 1:500 inoculation) (Fig. 2). In contrast, with the leucine supplement, CH601 started to grow 6 h post 1:500 inoculation (Fig. 2). The fact that grown cells repeated the prolonged lag phase in many runs of reinoculation has ruled out the possibility that mutants were selected during the period of prolonged lag phase. The growth curve shown in Fig. 2 is the result of the second inoculation. This striking result prompted us to test the growth of another S. typhimurium leu-500 strain, PM596, which was the parental strain of CH601 (8Richardson S.M.H. Higgins C.F. Lilley D.M.J. EMBO J. 1984; 3: 1745-1752Crossref PubMed Scopus (82) Google Scholar). Similar leucine-independent growth after a prolonged lag phase was observed (data not shown). These S. typhimurium leu-500 strains did not grow on SSA(−Leu) plates but grew in SSA(−Leu) liquid medium, indicating that some physiological changes must occur during the prolonged lag phase in the liquid medium. Since the plasmid-borne leu-500 activation is mediated by the ilvIH transcription activity-initiated promoter relay in CV468 (Fig. 1), we tested whether the suppression of leu-500mutation in the topA + strain, CH601, is also due to the ilvIH transcription activity. Again, using the reporter plasmid, pWU804LZ, we were able to detect the plasmid-borneleu-500 activation in the presence of ilvIHpromoter activity when pWU804LZ-harboring CH601 were grown in SSA(−Leu) medium. The level of leu-500 activity is comparable with that in CV468 and 50% less than that in CH582 (Fig. 3 A). The ilvIHpromoter activity is crucial for leu-500 activation, since mutation or deletion of the ilvIH promoter in pWU807LZ or pWU805LZ significantly diminished leu-500 activation in these reporter plasmids (Fig. 3 B, compare columns 1 and 2, or columns 1 and 3). We have recently demonstrated that expression of the intermediateleuO gene relays the ilvIH transcription activity to the leu-500 promoters in CH582 (16Fang M. Wu H.-Y. J. Bacteriol. 1998; 180: 626-633Crossref PubMed Google Scholar). To examine whether the promoter relay is also the case in CH601, the reporter plasmid pWU804MLZ, which carries the mutant leuO promoter, was tested. The β-galactosidase activity was shown to be significantly lowered due to the mutation of the leuO promoter (Fig. 3 B, column 4), suggesting that the plasmid-borneleu-500 activation in CH601 is also mediated by theilvIH promoter activity-initiated promoter relay mechanism. The above results led us to speculate that the suppression of leu-500 mutation on the chromosome is likely due to the promoter relay as well. If so, the ilvIH and leuOtranscription activities should correlate with leu-500activation in the prolonged lag phase of CH601 in SSA(−Leu) medium. To investigate this possibility, Northern analysis was performed to monitor the ilvIH and leuO mRNAs at different time points during the prolonged lag phase. CH601 grew in SSA(+Leu) medium until the OD650 reached approximately 0.6. After spinning down and washing with SSA(−Leu) medium, the cells were continued to be incubated aerobically in SSA(−Leu) medium. CH601 cells resumed growth 16 h after the medium change (the growth curve in Fig. 4). Our Northern blot results revealed that both ilvIH and leuO genes were active just before cells resumed growing (15 h after the medium change), but were silent at early lag phase (6 h after medium change) (Fig. 4, A and B). Therefore, theilvIH transcription activity may serve as the supercoiling signal for chromosomal leu-500 activation via activating theleuO gene in CH601. It appears that ilvIHtranscription is highly regulated in CH601 in response to nutrient (leucine) starvation. So far, our data have indicated that the promoter relay is responsible for leu-500 activation regardless of topA genetic background. Previous studies, however, have shown that leu-500 activation was dependent on the absence of TopA, although leu-500 activation did not correlate with the overall DNA superhelicity (11Richardson S.M.H. Higgins C.F. Lilley D.M.J. EMBO J. 1988; 7: 1863-1869Crossref PubMed Scopus (49) Google Scholar). TopA must play some kind of a role in leu-500 activation. TopA may indirectly affect leu-500 activation via either (a) regulating ilvIH transcription activity or (b) suppressing the negative supercoiling signals generated byilvIH and leuO transcription processes. In order to examine the first possibility, we monitored the TopA level and ilvIH transcription activity simultaneously during the growth of CH601 in SSA(−Leu) medium. CH601 were grown in SSA(+Leu) medium to OD650 of 0.6, and the growth resumed in SSA(−Leu) medium as described above (Fig. 4). Cells were harvested at three time points after the medium change (Fig. 5 growth curve, points 1, 2, and 3). The harvested cells from each time point was divided into two parts for monitoring the TopA protein level using immunoblotting (Fig. 5 A) and the ilvIHmRNA using Northern blotting (Fig. 5 C). While the Northern blot data indicated that ilvIH expression was highly regulated, it was active near the end of the prolonged lag phase and silent once cells resumed exponential growth (Fig. 5 C), the immunoblot indicated that TopA level remained constant across the three time points during the CH601 cell growth in SSA(−Leu) medium (Fig. 5 A). This result ruled out the possible effect of TopA levels on ilvIH expression. If TopA does not affect leu-500 activation via regulatingilvIH expression (illustrated in Fig. 6), the remaining possibility is that the promoter relay is affected negatively by TopA on ilvIHand/or leuO transcription-generated negative supercoiling signals (illustrated in Fig. 6). Hence, TopA affects leu-500activation indirectly only when ilvIH and leuOpromoters are active in the promoter relay. When TopA is absent, the supercoiling signals generated from the ilvIH and leuO promoters are expected to be maximal and activation of the leu-500 promoter is thus facilitated. Since the ilvIH transcription activity plays an indispensable role in leu-500 activation, we predicted thatilvIH must be active in CH582 during normal growth conditions. This prediction was confirmed by the Northern blot data that ilvIH was expressed constantly during the log phase of CH582 growth (Fig. 7). This may explain why the suppression of the leu-500 mutation was originally observed in the topA − strains including CH582 (2Mukai F.H. Margolin P. Proc. Natl. Acad. Sci. U. S. A. 1963; 50: 140-148Crossref PubMed Google Scholar, 3Dubanau E. Margolin P. Mol. Gen. Genet. 1972; 117: 91-112Crossref PubMed Scopus (39) Google Scholar, 11Richardson S.M.H. Higgins C.F. Lilley D.M.J. EMBO J. 1988; 7: 1863-1869Crossref PubMed Scopus (49) Google Scholar). We have demonstrated using both plasmid and chromosome systems that in the presence of the 1.9-kb chromosomal fragment upstream of theleuABCD operon, ilvIH activity-initiatedleu-500 activation is independent of topA genetic background, since CH601 (the topA + strain) can overcome its leucine auxotrophy and grow in leucine-free medium while the cellular TopA level remains constant. However, previous studies have shown that activation of the plasmid-borne leu-500promoter is absolutely dependent on the absence of TopA (topA − genetic background) when theleu-500 promoter is flanked by various plasmid sequences (12Chen D. Bowater R. Dorman C.J. Lilley D.M.J. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 8784-8788Crossref PubMed Scopus (65) Google Scholar, 13Chen D. Bowater R.P. Lilley D.M.J. Biochemistry. 1993; 32: 13162-13170Crossref PubMed Scopus (24) Google Scholar, 14Tan J. Shu L. Wu H.-Y. J. Bacteriol. 1994; 176: 1077-1086Crossref PubMed Google Scholar, 25Chen D. Bachellier S. Lilley D.M.J. J. Biol. Chem. 1998; 273: 653-659Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar). These contrasted results suggest that, owing to the difference in DNA sequence context where the leu-500promoter is located, the mechanisms underlying these two types of leu-500 activation are different despite the involvement of transcription-driven DNA supercoiling in both cases. ThetopA-dependent leu-500 activation by an adjacent promoter activity was demonstrated using a different, non-native sequence flanked by a pair of divergent promoters (12Chen D. Bowater R. Dorman C.J. Lilley D.M.J. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 8784-8788Crossref PubMed Scopus (65) Google Scholar, 13Chen D. Bowater R.P. Lilley D.M.J. Biochemistry. 1993; 32: 13162-13170Crossref PubMed Scopus (24) Google Scholar, 14Tan J. Shu L. Wu H.-Y. J. Bacteriol. 1994; 176: 1077-1086Crossref PubMed Google Scholar). It appears that upstream transcription-driven negative supercoiling propagates along the intervening sequence and directly activates theleu-500 promoter. Lilley's group also demonstrated that the coupling of tetA-mediated transcription and translation induced an increase in overall negative supercoiling and activated theleu-500 promoter in a closed circular DNA (25Chen D. Bachellier S. Lilley D.M.J. J. Biol. Chem. 1998; 273: 653-659Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar). TopA relaxes the negative supercoiling, therefore the absence of TopA is required for the direct effect of transcription-driven supercoiling onleu-500 activation. When the leu-500 promoter is associated with the 1.9-kb upstream natural sequence, the long range interaction between the ilvIH and leu-500promoters is somewhat more complex than the direct effect of DNA supercoiling. Although ilvIH transcription-induced supercoiling initiates the long range promoter interaction, other protein factors and DNA elements within the intervening sequence relay the promoter activity over a long (1.9 kb) distance. We have recently shown that ilvIH first activates the leuO gene located within the intervening sequence, and both the leuOpromoter activity and the LeuO protein are required for subsequentleu-500 activation (16Fang M. Wu H.-Y. J. Bacteriol. 1998; 180: 626-633Crossref PubMed Google Scholar). Although the distance between the two divergent leuO and leu-500 promoters is within the 400-bp limit for the direct transcription-driven supercoiling effect, the leuO promoter activity alone was insufficient to activate the leu-500 promoter (16Fang M. Wu H.-Y. J. Bacteriol. 1998; 180: 626-633Crossref PubMed Google Scholar), suggesting that the direct supercoiling effect does not account for the suppression of leu-500 mutation in the chromosomal context. The evolutionarily conserved AT-rich sequence (26Haughn G.W. Wessler S.R. Gemmill R.M. Calvo J.M. J. Bacteriol. 1986; 166: 1113-1117Crossref PubMed Google Scholar) between the two divergent leuO and leu-500 promoters may repress the direct supercoiling effect, and the LeuO protein can derepress and activate the leu-500 promoter in the presence of the cis-acting leuO promoter activity.2 It appears that ilvIH transcription-induced DNA supercoiling is relayed via the 1.9-kb intervening sequence involving protein factors and DNA elements in a very efficient manner, so that TopA negatively regulates but does not efficiently abolish the supercoiling signal which may be constrained by the protein factors and DNA elements during the relaying process. Since ilvIH expression is crucial in initiating the promoter relay, it is important to know what regulates ilvIHexpression. The Northern blot data indicated that ilvIHtranscription is under tight regulation (Fig. 5 C). TheilvIH promoter activity is normally silent until cells are under nutrient (leucine) starvation such as the prolonged lag phase of CH601 in the SSA(−Leu) medium. We have ruled out the possibility thatilvIH expression is affected by the cellular TopA level (Fig. 5). It has been shown that ilvIH promoter activity is under the positive control of Lrp (leucine-responsive regulatory protein), which is a global transcription regulator whose cellular level is up-regulated by cellular guanosine 3′,5′-bispyrophosphate (ppGpp) in response to nutrient limitation (27Cashel M. Gentry D.R. Hernandez V.J. Vinella D. Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. 1. American Society for Microbiology, Washington, D. C.1996: 1458-1496Google Scholar, 28Landgraf J.R. Wu J. Calvo J.M. J. Bacteriol. 1996; 178: 6930-6936Crossref PubMed Google Scholar). Combined with the previous studies, these results have suggested that ilvIHtranscription activity-initiated promoter relay is a transcription regulatory mechanism elicited in response to stress. We are currently conducting experiments to test the roles of promoter relay in this stress response pathway. Unlike CH601, where ilvIH expression is cryptic during exponential growth, ilvIH is constantly expressed in CH582 (Fig. 7). We observed that the growth rate of CH582 is two times slower than that of CH601. The slow growth rate is shown to up-regulatelrp gene expression (28Landgraf J.R. Wu J. Calvo J.M. J. Bacteriol. 1996; 178: 6930-6936Crossref PubMed Google Scholar), which presumably activates theilvIH promoter constantly in CH582. The constantilvIH transcription activity is likely to be the reason why suppression of the leu-500 mutation occurs constantly in thetopA − strains. Co-detection of the normally silent leuO gene expression with ilvIH indicated that LeuO protein may play a role in such a stress response in bacteria. Besides affecting wild typeleuABCD promoter (16Fang M. Wu H.-Y. J. Bacteriol. 1998; 180: 626-633Crossref PubMed Google Scholar), random screening studies have recently shown that LeuO overexpression affects three other unrelated genes at distinct chromosomal locations (29Shi X. Bennett G.N. J. Bacteriol. 1995; 177: 810-814Crossref PubMed Google Scholar, 30Klauck E. Böhringer J. Hengge-Aronis R. Mol. Microbiol. 1997; 25: 559-569Crossref PubMed Scopus (68) Google Scholar, 31Ueguchi C. Ohta T. Seto C. Suzuki T. Mizuno T. J Bacteriol. 1998; 180: 190-193Crossref PubMed Google Scholar), suggesting its global regulatory role in gene expression regulation under stress. The exact physiological implication of leuO expression under the control of the promoter relay at this chromosomal location is very interesting but remains unclear. We may have just begun to unravel the importance of global transcription regulatory functions of LeuO protein through understanding the mysterious leu-500 activation phenomenon in topA mutant which was first reported more than 30 years ago (2Mukai F.H. Margolin P. Proc. Natl. Acad. Sci. U. S. A. 1963; 50: 140-148Crossref PubMed Google Scholar). We thank Dr. Amit Banerjee and Terry Barrette for critical reading and comment on the manuscript. We are grateful to Drs. Rolf Menzel and Haiyan Qi for providing E. coli TopA antibody; Drs. David Lilley and Joseph Calvo for providing S. typhimurium strains.