Growth Phase and Growth Rate Regulation of the rapA Gene, Encoding the RNA Polymerase-Associated Protein RapA in Escherichia coli

Please download to get full document.

View again

of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information Report

How To, Education & Training


Views: 42 | Pages: 10

Extension: PDF | Download: 0

Growth Phase and Growth Rate Regulation of the rapA Gene, Encoding the RNA Polymerase-Associated Protein RapA in Escherichia coli
    10.1128/JB.183.20.6126-6134.2001. 2001, 183(20):6126. DOI: J. Bacteriol. Julio E. Cabrera and Ding Jun Jin  Escherichia coli  Polymerase-Associated Protein RapA in Gene, Encoding the RNA rapA of theGrowth Phase and Growth Rate Regulation information and services can be found at: These include:  REFERENCES This article cites 41 articles, 26 of which can be accessed free CONTENT ALERTS  more»articles cite this article), Receive: RSS Feeds, eTOCs, free email alerts (when new Information about commercial reprint orders: To subscribe to to another ASM Journal go to:  onM ar  c h 2  9  ,2  0 1 4  b  y  g u e s  t  h  t   t   p:  /   /   j   b . a s m. or  g /  D  ownl   o a d  e d f  r  om  onM ar  c h 2  9  ,2  0 1 4  b  y  g u e s  t  h  t   t   p:  /   /   j   b . a s m. or  g /  D  ownl   o a d  e d f  r  om   J OURNAL OF  B  ACTERIOLOGY ,0021-9193/01/$04.00  0 DOI: 10.1128/JB.183.20.6126–6134.2001Oct. 2001, p. 6126–6134 Vol. 183, No. 20 Growth Phase and Growth Rate Regulation of the  rapA  Gene,Encoding the RNA Polymerase-Associated Protein RapA in  Escherichia coli JULIO E. CABRERA   AND  DING JUN JIN*  Laboratory of Molecular Biology, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892 Received 14 May 2001/Accepted 16 July 2001 The  Escherichia coli rapA  gene encodes the RNA polymerase (RNAP)-associated protein RapA, which is abacterial member of the SWI/SNF helicase-like protein family. We have studied the  rapA  promoter and itsregulation in vivo and determined the interaction between RNAP and the promoter in vitro. We have found thatthe expression of   rapA  is growth phase dependent, peaking at the early log phase. The growth phase control of   rapA  is determined at least by one particular feature of the promoter: it uses CTP as the transcription-initiating nucleotide instead of a purine, which is used for most  E. coli  promoters. We also found that the  rapA promoter is subject to growth rate regulation in vivo and that it forms intrinsic unstable initiation complexes with RNAP in vitro. Furthermore, we have shown that a GC-rich or discriminator sequence between the  10and   1 positions of the  rapA  promoter is responsible for its growth rate control and the instability of itsinitiation complexes with RNAP. The transcription machinery in  Escherichia coli  consists of RNA polymerase (RNAP) and RNAP-associated proteins.The RNAP core enzyme is composed of four subunits,   2  ,and is capable of transcription elongation and termination atintrinsic terminators. After binding to any of the seven    fac-tors, the resulting RNAP holoenzyme (  2  ) is able to ini-tiate transcription at specific sites called promoters in the bac-terial chromosome (6, 6a). Several other proteins (NusA,GreA/GreB, and   ) bind core and/or holoenzyme RNAP andmodify specific steps of the transcription cycle (1, 4, 9, 10, 13,17, 34) or facilitate RNAP assembly (23).Previously, we showed that the RNAP-associated proteinRapA (110 kDa) binds both core and holoenzyme RNAP;however, it has a higher affinity to the former (35, 36). TheRapA protein is a bacterial homolog of the SWI/SNF helicase-like protein family which is involved in chromatin remodelingand gene expression (39). RapA has ATPase activity that isstimulated by binding to RNAP, indicating that RapA interacts with RNAP both physically and functionally (36). However,RapA has only a marginal effect on transcription in vitro (24,35), and the role of   rapA  in transcription has been elusive. The  rapA  gene (also called  hepA ) was srcinally identified down-stream of the  polB  gene, which is controlled by DNA damage(18). However, we have shown that RapA is not likely to beinvolved in DNA repair (36), contrary to a previous report(24). The  rapA  promoter and its regulation have never beenstudied. In the present work we have analyzed the  rapA  pro-moter, determined the expression of   rapA  under differentphysiological conditions, and studied the interaction betweenRNAP and the  rapA  promoter in vitro. MATERIALS AND METHODSBacterial strains.  The  E. coli  strains used in this work are listed in Table 1. Thebasic bacterial techniques used have been described elsewhere (22). The  fis ::  kan allele was moved into different strains by phage P1 transduction with a lysatemade from strain RLG1351, and Kan r transductants were selected. StrainDJ2543-47B was constructed in two steps. First the  relA251 ::  kan  allele wasmoved into strain DJ2517-C2A by P1 transduction with a lysate made from strainCF1651, and Kan r transductants were selected. Second, the resulting strain wastransduced with a P1 lysate made from strain CF4943, and Tet r transductants were selected. Because the  spoT203  is linked with the  zib563 ::Tn 10  allele at afrequency of approximately 50%, the resulting Tet r colonies were scored for the  spoT203  phenotype (small colonies). Note that cells harboring the  spoT203  alleleand a wild-type  relA  allele are not viable because the (p)ppGpp synthesized bythe RelA protein cannot be degraded by the mutant SpoT protein, resulting inextremely high concentrations of (p)ppGpp (32). In  relA251 spoT203  cells, the(p)ppGpp is synthesized from the mutant SpoT protein. Chemicals and reagents.  Nucleotides and  32 P-labeled nucleotides were pur-chased from Amersham. Chemicals were from Sigma. RNAP was purified fromstrain MG1655 as described previously (14). Antibodies against RapA andRNAP have been described previously (35). The Fis protein was a gift from R.Johnson (University of California—Los Angeles). Construction of   lacZ  fusions.  All  lacZ  fusions reported in this work wereintroduced into strain DJ480 as    phage monolysogens. Vectors and methods were as described previously (33). Briefly, DNA fragments containing the  rapA promoter were synthesized by PCR and cloned into the  Eco RI and  Bam HI sitesof the vector pRS415. The resulting plasmids were verified by DNA sequencingand recombined in vivo with the   RS45 phage. Blue recombinant phage plaques were purified twice for each fusion, and the resulting phages were used to obtainlysogens using  E .  coli  DJ480 as host cells. Single-prophage integration wasconfirmed by PCR amplification as described previously (29). Bacterial growth and  -galactosidase measurements.  All cultures were grown with vigorous agitation in a water bath at 37°C. For time course experiments, afresh overnight culture was diluted 1/100 into fresh medium. For growth rateexperiments, cells were grown in morpholinepropanesulfonic acid (MOPS) me-dium (25) supplemented with either 0.2% (vol/vol) glycerol or 0.2% (wt/vol)glucose, with or without 0.8% Difco Casamino Acids plus 50   g of tryptophan/ ml, or in Luria-Bertani (LB) medium. Growth was monitored by measuring the  A 600 , and growth rates were calculated using the slopes of the growth curves.  -Galactosidase assays were performed as described previously (44). Culturealiquots were lysed in a microtiter plate and exposed to the chromogenic sub-strate  o -nitrophenyl-  - D -galactopyranoside (ONPG). Kinetic measurements were made using a SpectraMax 250 microtiter plate reader (Molecular Devices)to obtain the  V  max  . Units were presented as specific activities, which were cal-culated by dividing the  V  max   by the  A 600  of the culture and then multiplying by * Corresponding author. Mailing address: Laboratory of MolecularBiology, National Cancer Institute, National Institutes of Health,Building 37, Room 2B16, 9000 Rockville Pike, Bethesda, MD 20892.Phone: (301) 402-9281. Fax: (301) 594-3611. E-mail:   onM ar  c h 2  9  ,2  0 1 4  b  y  g u e s  t  h  t   t   p:  /   /   j   b . a s m. or  g /  D  ownl   o a d  e d f  r  om   25. The speci fi c activities thus obtained have been empirically determined tocorrespond with the standard Miller units. For each culture, triplicate samples were taken at each time course and results were expressed as the means of thethree measurements. The standard deviations of the triplicates were less than5%. In time course experiments where two or more strains were compared,duplicate cultures were assayed for each strain on the same day. The variationsbetween the culture duplicates were less than 10%. Each set of experiments wasrepeated at least three times, and the differences between repetitions were lessthan 10%. For the growth rate experiments in each different medium, a wholetime course covering different growth phases was performed. The maximal ac-tivity obtained at each growth rate was plotted as a function of growth rate. Cloning and DNA manipulations.  All DNA manipulations and cloning tech-niques were carried out as described elsewhere (20). PCR ampli fi cations werecarried out using the High Fidelity Expand system (Roche). Sequencing wasperformed on a Genetic Analyzer using the  D -rhodamine terminator cycle se-quencing ready reaction kit (both from Perkin-Elmer, Applied Biosystems Di- vision) according to the manufacturer ’ s speci fi cations. All plasmids used in the in vitro transcription reactions were constructed byinserting PCR products (digested with the  Eco RI and  Pst I restriction enzymes)into the  Eco RI-  Pst I sites of plasmid pSA508, which contained a very strongRho-independent terminator downstream of the two restriction sites (7). ForPCRs, genomic DNA (MG1655) was used as the DNA template except asmentioned otherwise. The DNA sequences of the two primers used for the insertin plasmid pDJ760 were 5  -AGATCGAATTCGAATTCGGCCCGGAGCCGCTGGACTACCAACGTT (primer DJ144, upper strand) and 5  -CATGGCTGGTATGGTATCTGCAGGGTTGAACACGCGGTCA (lower strand). The twoprimers used for the insert in plasmid pDJ2506 were primers DJ144 andRAPATGPST (5  -ACCAAGTGTAACTGCAGATGTTGTTCGGGTCTATA TCT). The insert in plasmid pDJ2512 containing the discriminator mutations wasobtained in two steps. First, two independent DNA fragments (A and B) whichshare complementary sequence in the discriminator region were each ampli fi edby PCR. The primers used to amplify fragment A were DJ144 for the upperstrand and JC107H (5  -TCAGGTCCAGGAATGGAAAGGAATTTATGGTA CTGGATG) for the lower strand. The primers used to amplify fragment B wereJC107G (5  -GCCATCCAGTACCATAAATTCCTTTCCATTCCTGGACCTG A-3  ) for the upper strand and RAPATGPST for the lower strand. PrimersJC107G and JC107H are partially complementary and have mutations in thediscriminator region (underlined in the above sequences). The second step wasa PCR ampli fi cation with primers DJ144 and RAPATGPST and a mixture of fragments A and B as the DNA template. Primer extension analysis.  Total RNA from  E .  coli  cells and in vitro tran-scription reactions was isolated using an RNeasy kit from Qiagen according tothe manufacturer ’ s instructions. Primer extension reactions were carried out withthe avian myeloblastosis virus primer extension kit (Promega) according to themanufacturer ’ s instructions. Using plasmid pDJ760 as the DNA template, thesame primer used in the primer extension assays was used to generate the DNA sequencing ladder for mapping the transcription start points. In vitro transcription assays.  Reactions were carried out essentially as de-scribed previously (43). Transcription reactions were carried out at   24 ° C in fi nal volumes of 20   l containing 2 to 3 nM concentrations of DNA templatesand 25 nM RNAP. RNAP and DNA templates were preincubated for 15 min,and reactions were started by addition of nucleoside triphosphates (NTPs) (0.2mM for ATP, GTP, and CTP, and 0.02 mM for UTP, including about 5   Ci of [  - 32 P]UTP). Where indicated, heparin (100   g/ml) was added with the NTPs torestrict transcription to a single round by binding to free RNAP molecules. After15 min, reactions were stopped and products were analyzed on an 8% sequencinggel, followed by autoradiography. In the experiments where the kinetics of inactivation of open complexes were analyzed, heparin was added after prein-cubation of RNAP and the DNA template (time zero). At the times indicatedafter addition of the inhibitor, NTPs were added, and reactions were allowed tocontinue for 15 min before being stopped and analyzed as described above. Data were quanti fi ed with an ImageQuant PhosphorImager (Molecular Dynamics). RESULTSMapping the transcription start site of   rapA .  To identify thetranscription start point of the  rapA  gene, we carried outprimer extension analysis using a radiolabeled oligonucleotidethat hybridizes with the translation initiation region of   rapA mRNA (Fig. 1). We puri fi ed total RNA from  E. coli  MG1655cells harboring plasmid pDJ760, which contains the promoter/ regulatory region of   rapA  (Table 1), and from in vitro tran-scription reactions using plasmid pDJ760 as the template. Inboth cases, we found that the start sites of the  rapA  gene wereat two cytosine residues located 94 and 95 bp upstream fromthe initiation codon ATG (Fig. 2). We also performed primerextension analysis using RNA puri fi ed from MG1655 cells andidenti fi ed the same transcription start points, although thesignals were very weak (data not shown). Thus, we have iden-ti fi ed the transcription start points of the  rapA  gene and de- fi ned the second cytosine residue as the   1 position (Fig. 1). The  rapA  promoter is growth phase dependent.  To study the  rapA  promoter activity in vivo, we fused the  rapA  promoterregion (  211 to  77 [Fig. 1]) with the promoterless  lacZ  gene,followed by integration of this fusion in a single copy into the  E. coli  chromosome, resulting in strain DJ2517-C2A. Thus,expression of   rapA  could be monitored by   -galactosidase ac-tivity. First, we determined the expression of   rapA  as a function TABLE 1. Bacterial strains and plasmids used in this work Strain or plasmid Genotype or relevant characteristics Reference or source StrainsMG1655 Wild-type  E. coli  K-12 Laboratory collectionDJ480 MG1655  lacX74  Laboratory collectionCF1651 MG1655  relA251 ::  kan  29CF4943 MG1655  galK2 relA251 ::  kan zib563 ::Tn 10 spoT203  40RLG1351 MG1655  lacX74  fi  s ::  kan    rrnBp I (  88 to   1)::  lacZ  39DJ2517-C2A DJ480    rapA  (  211 to   77)::  lacZ  This workDJ2611-C1 DJ480    rapA  (  57 to   77)::  lacZ  This workDJ2611-C2 DJ480    rapA  (  36 to   77)::  lacZ  This workDJ2524-13E DJ2517-C2A   fi  s ::  kan  This workDJ2621A DJ2611-C1  fi  s ::  kan  This workDJ2611-A1 DJ480    rapA  (  211 to   77;   1C 3   A)::  lacZ  This workDJ2543-47B DJ2517-C2A   relA251 ::  kan zib563 ::Tn 10 spoT203  This workDJ2611-B1 DJ480    rapA  (  211 to   77; discriminator mutant)::  lacZ  This workPlasmidspSA508 pBR322-based vector 10pDJ760  rapA  (  221 to   415) in vector pSA508 This workpDJ2506  rapA  (  221 to   77) in vector pSA508 This workpDJ2512  rapA  (  221 to   77; discriminator mutant) in vector pSA508 This workV OL  . 183, 2001 GROWTH PHASE AND GROWTH RATE CONTROL OF  rapA  6127   onM ar  c h 2  9  ,2  0 1 4  b  y  g u e s  t  h  t   t   p:  /   /   j   b . a s m. or  g /  D  ownl   o a d  e d f  r  om   of cell growth (Fig. 3A). We found that  rapA  activity increaseddramatically during the  fi rst doubling time after cultures werediluted in fresh medium. The peak of promoter activity wasreached approximately during the  fi rst 30 to 45 min of growth. As cells continued to grow, however, the activity decreased,and it became minimal during the stationary phase. It shouldbe noted that there are limitations in using the  rapA-lacZ fusion to monitor  rapA  promoter activity due to the stability of   -galactosidase, levels of which are reduced in the cell only dueto growth and dilution. Thus, our data suggest that after a burstof synthesis of   rapA  during the early log phage, the expressionof   rapA  is essentially shut off, as the reduction of    -galactosi-dase activity can be explained by the growth and dilution of thecell in the cultures. Apparently, this expression pattern is spe- FIG. 1. Nucleotide sequence of the  rapA  promoter  .  The codingregions of the  polB  and  rapA  genes are shaded. The translation initi-ation codon (ATG) for the RapA protein and the translation termi-nation codon for the upstream gene  polB  (TGA) are boxed. The   10and  35 regions of the  rapAp  are underlined, and the two transcriptionstart points are indicated by bent arrows. The solid arrow indicates theoligonucleotide used in the primer extension experiments. Dashedarrows indicate the positions of the Fis binding sites predicted by theinformation theory algorithm (16); all predicted sites had scores be-tween 3 and 7. Arrowheads show the boundaries of the transcriptionalfusions used in this work.FIG. 2. Mapping the transcription start points of the  rapA  gene.Primer extension reactions were carried out with a  32 P-labeled oligo-nucleotide that hybridizes near the ATG region and 100   g of totalRNA from  E. coli  MG1655 cells harboring the pDJ760 plasmid (lane1) or RNA from an in vitro transcription reaction with the pDJ760plasmid as the template (lane 2). DNA sequencing reactions werecarried out with the same labeled oligonucleotide (lanes G, A, T, andC) and electrophoresed in parallel on 8% polyacrylamide – 8 M ureagels. The sequence on the right corresponds to the nontemplate strand.FIG. 3.  rapA  promoter activity is growth phase dependent. (A)DJ2517-C2A cells carrying a fusion between  rapA  (positions   211 to  77) and the  lacZ  gene were grown in LB medium at 37 ° C andmonitored for both growth, expressed as optical density (OD) (trian-gles), and   -galactosidase activity (squares). (B)   -Galactosidase ac-tivity expressed from different fusions as a function of growth. Cellscarrying positions  211 to  77 (strain DJ2517-C2A) (squares),  57 to  77 (strain DJ2611-C1) (triangles), or  36 to  77 (strain DJ2611-C2)(circles) of the  rapA  promoter region fused to the  lacZ  gene weregrown in LB medium at 37 ° C.   -Galactosidase activity was measuredas described in Materials and Methods. No signi fi cant differences inthe growth rate were detected among the different fusions.6128 CABRERA AND JIN J. B  ACTERIOL  .   onM ar  c h 2  9  ,2  0 1 4  b  y  g u e s  t  h  t   t   p:  /   /   j   b . a s m. or  g /  D  ownl   o a d  e d f  r  om   ci fi c for the  rapA  promoter, because the   -galactosidase activ-ity expressed from several other promoters (   pL ,  lacUV5 , and  dsrA ) exhibited no such expression pattern, as reported re-cently (30). We conclude from this experiment that  rapA  pro-moter activity is growth phase dependent.To de fi ne the sequences of the  rapA  promoter that aresuf  fi cient to maintain the promoter activity and the expressionpattern observed above, we constructed two other, similar  lacZ fusions with shorter upstream sequences of the  rapA  promoterregion and determined the   -galactosidase activities of thesefusions similarly. One fusion (DJ2611-C1), containing residues  57 to   77 of the  rapA  promoter (Fig. 3B), was almost indis-tinguishable from the DJ2517-C2A fusion containing residues  211 to   77 (Fig. 3B). This indicates that the determinantsthat provide both full promoter activity and growth phase reg-ulation are located in the   57-to-  77 region. Another fusion(DJ2611-C2), containing residues   36 to   77 of the  rapA promoter (Fig. 3B), still exhibited growth phase dependencefor the expression of   rapA , although it had only about half thepeak activity of DJ2517-C2A. Together, these results indicatethat residues   57 to   36 are required for full promoter activ-ity and that the minimal promoter region from   36 to   77 issuf  fi cient to provide growth phase regulation.To address whether the growth phase regulation of   rapA  isalso re fl ected in RapA protein levels, we determined RapA protein levels as a function of cell growth by Western blotanalysis using polyclonal antibodies against RapA (Fig. 4). Wefound that RapA levels also increased dramatically during the fi rst half-hour after cultures were diluted in fresh medium,reached maximal levels after 1 h, and then decreased andbecame minimal in the late-stationary phase (Fig. 4A). As acontrol, we probed the same samples in a parallel Western blotusing an antibody against core RNAP. We found that the levelsof the    and    subunits remained almost constant duringdifferent growth phases (Fig. 4B). At present, we do not know why there is a difference between the times of highest pro-moter activity (30 to 45 min) and peak protein accumulation (1h). We speculate that it may be due to differences between thehalf-lives of the  lacZ  and  rapA  transcripts and/or differencesbetween the translation ef  fi ciency of the   -galactosidase andRapA proteins. We conclude that RapA protein levels corre-late reasonably well with promoter activity. The growth phase regulation of   rapA  is independent of Fis. Because the Fis protein also increases dramatically immedi-ately after cultures are diluted in fresh medium (26), in amanner very similar to that of RapA described above, we askedif Fis was responsible for the growth phase regulation of   rapA .Thus, we measured the expression of   rapA  in  fi  s  mutant cellsand found that the  rapA  activity was still growth phase depen-dent, like that in the wild-type isogenic cells (Fig. 5). Interest-ingly, promoter activities were more than twofold higher in the  fi  s  cells than in wild-type cells. Taken together, our resultssuggest that Fis negatively regulates  rapA  but is not involved inthe growth phase regulation of the  rapA  promoter.To account for the negative effect of Fis on the expression of   rapA , we searched for putative Fis binding sites in the pro-moter region using an information theory algorithm (16). Weidenti fi ed several potential Fis binding sites in the  rapA  pro-moter region (Fig. 1). However, compared to the well-knownFis binding sites in the ribosomal promoter  rrnB  P1, whichhave high scores between 10 and 15 by the algorithm indicatingstrong binding (16), the putative Fis binding sites in the  rapA promoter are weak, with low scores ranging from 3 to 7. Theresults from gel shift assays with puri fi ed Fis and DNA frag-ments containing the Fis binding sites of the  rapA  and  rrnB  P1promoters were consistent with the prediction (data notshown). However, we do not believe that these Fis binding siteare responsible for the observed negative effect of Fis on theexpression of   rapA  for the following two reasons. (i) All thepredicted Fis binding sites in the  rapA  promoter region arelocated upstream of the nucleotide at   57 (Fig. 1). However,the promoter fusion (  57 to   77) lacking these putative bind-ing sites (or any other putative Fis binding sites in the vectorsequences upstream of the   57 position) still showed the neg-ative effect of Fis on the promoter (Fig. 5). (ii) Puri fi ed Fis hadno effect on RNA synthesis from the  rapA  promoter containingthe Fis binding sites by in vitro transcription assays (data notshown). It is very likely that the in vivo negative effect of Fis onthe expression of the promoter is indirect.  A mutation at the transcription start site alters the growthphase response of   rapA .  Since both the  rapA  and  fi  s  promoters FIG. 4. RapA levels as a function of cell growth. A culture of   E. coli MG1655 cells was monitored for growth in LB medium at 37 ° C bymeasuring the optical density (OD) at 600 nm (C). At the timesindicated by the numbers along the curve, samples were taken andconcentrated by centrifugation. Each sample was used in Western blotanalyses with polyclonal antisera against RapA (A) and core RNAP(B). The same amount of cells (normalized using the OD at 600 nm) was loaded in each lane.V OL  . 183, 2001 GROWTH PHASE AND GROWTH RATE CONTROL OF  rapA  6129   onM ar  c h 2  9  ,2  0 1 4  b  y  g u e s  t  h  t   t   p:  /   /   j   b . a s m. or  g /  D  ownl   o a d  e d f  r  om 
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!