HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Weyand, N. J. Articles by Low, D. A. Search for Related Content PubMed PubMed Citation Articles by Weyand, N. J. Articles by Low, D. A. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 5, 3192-3200, February 4, 2000

Regulation of Pap Phase Variation
Lrp IS SUFFICIENT FOR THE ESTABLISHMENT OF THE PHASE OFF pap DNA METHYLATION PATTERN AND REPRESSION OF pap TRANSCRIPTION IN VITRO^*

Nathan J. Weyand and David A. Low

From the Department of Molecular, Cellular, and Developmental Biology, University of California, Santa Barbara, California 93106

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The pyelonephritis-associated pili (pap) operon in Escherichia coli is regulated by an epigenetic mechanism involving the formation of specific DNA methylation patterns characteristic of transcriptionally active (phase ON) and inactive (phase OFF) cells. The formation of pap DNA methylation patterns in vivo was previously shown to require the leucine-responsive regulatory protein (Lrp) and DNA adenine methylase (Dam). To monitor the binding of Lrp to pap DNA, an in vitro methylation protection assay was developed. Binding of Lrp to a Dam target site proximal to the papBA promoter (designated GATC^prox) blocked methylation of this site and specifically repressed transcription. The DNA methylation pattern and transcription state are identical to those observed in vivo in phase OFF cells. To determine if binding of Lrp at GATC^prox was necessary for repression of papBA transcription, we analyzed a pap mutation (pap-13) that reduced the affinity of Lrp for the GATC^prox region. Binding of Lrp to pap-13 DNA was shifted to a promoter distal Dam target site (designated GATC^dist). Lrp blocked methylation of GATC^dist in the pap-13 mutant, but did not repress papBA transcription. Together, these results show that binding of Lrp to the GATC^prox region is sufficient for the establishment of the phase OFF DNA methylation pattern and repression of papBA transcription.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The pyelonephritis-associated pili (pap)¹ operon codes for pili, which mediate adhesion of uropathogenic Escherichia coli to host uroepithelial cells (1). The expression of Pap pili is regulated by phase variation, a process by which individual cells switch between expression (phase ON) and non-expression (phase OFF) states (2). Pap phase variation is controlled at the transcriptional level by a complex epigenetic mechanism involving the formation of specific DNA methylation patterns, similar to that observed in certain eukaryotic systems (3). DNA methylation has been shown to control the binding of the global regulator leucine-responsive regulatory protein (Lrp) (4, 5) to pap regulatory DNA sequences (6-9). In addition, the pap-encoded co-regulator PapI has also been shown to control the binding of Lrp to pap regulatory DNA (8, 10, 11). Genetic analyses showed that Lrp, PapI, and DNA adenine methylase (Dam) were all necessary for pap transcription (6).

Lrp binds cooperatively to two pap DNA regions in vitro, each containing a GATC sequence, which is the target of Dam (Fig. 1). In vivo DNA analysis showed that, in phase OFF cells, the GATC site proximal to the papBA promoter (GATC^prox) was protected from methylation, whereas the GATC site distal to the promoter (GATC^dist) was fully methylated. Conversely, in phase ON cells, the GATC^prox site was fully methylated and GATC^dist was protected from methylation (6). In the absence of Lrp, GATC^prox and GATC^dist were not protected from methylation, showing that Lrp was essential for the phase ON and phase OFF DNA methylation patterns observed in vivo (12). Based on these data, it was speculated that, in phase OFF cells, Lrp was bound near GATC^prox whereas in phase ON cells Lrp was bound near GATC^dist (13).

DNA footprint analysis was consistent with this interpretation of the DNA methylation pattern data. Lrp bound cooperatively and with highest affinity to pap Lrp DNA binding sites 1, 2, and 3 encompassing GATC^prox in vitro. Addition of PapI shifted the binding of Lrp to pap DNA Lrp binding sites 4, 5, and 6 over 100 base pairs upstream (relative to the papBA promoter) encompassing GATC^dist (8) (see Fig. 1). Moreover, binding of Lrp-PapI to pap Lrp binding sites 4, 5, and 6 was blocked when GATC^dist was fully methylated, providing a means by which DNA methylation could control the Pap phase variation switch (10).

Although Lrp was shown to bind to pap regulatory DNA encompassing the GATC^dist and GATC^prox sites in vitro, it was not clear if this was sufficient for the formation of the DNA methylation patterns observed for phase ON and phase OFF cells in vivo, or how this binding affected pap transcription (14). Here we show that the phase ON and phase OFF pap DNA methylation patterns can be established in vitro by Lrp alone. Moreover, binding of Lrp to GATC^prox was sufficient for repression of papBA transcription. These studies provide an in vitro system for the biochemical analysis of Pap phase variation.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Chemicals and Reagents-- Restriction endonucleases and T4 polynucleotide kinase were purchased from New England Biolabs. AMV-reverse transcriptase, RNasin ribonuclease inhibitor, and ribonuclease (RNase ONE^TM) were purchased from Promega. E. coli RNA polymerase (⁷⁰) holoenzyme was from Epicentre Technologies. Nucleotides were obtained from NEN Life Science Products. Oligonucleotides used for this study were purchased from the Protein-DNA Core Facility (Cancer Center, University of Utah, Salt Lake City, UT) or Genosys Biotechnologies, Inc.

Purification of Proteins-- Lrp was purified (Margareta Krabbe)² from strain JWD3-1 (a gift from Rowena Matthews). Lrp protein concentrations were estimated by absorbance at 280 nm, assuming a molar extinction coefficient of 1197 for tyrosine (15) and five tyrosines per monomer of Lrp. Preparations of Lrp were judged to be equal to or greater than 95% pure by SDS-polyacrylamide gel electrophoresis (PAGE) and Coomassie Blue staining.²

Dam was overexpressed using plasmid pDOX1 containing E. coli dam under control of the phage  P_L promoter (16). E. coli DL3322 containing pDOX1 was grown in Luria-Bertani broth to an A₆₀₀ of 0.6 at 37 °C, shifted to a 42 °C shaking water bath for 1 h, and then transferred back to 37 °C for an additional 2 h. Bacterial cells were centrifuged (3,000 × g, 10 min), and the cell pellet was resuspended in 30 ml of buffer A (50 mM sodium phosphate, 10 mM EGTA, 10% (v/v) glycerol, 10 mM -mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride) containing 1 M NaCl. The cell suspension was broken using a French pressure cell, and cell debris was removed by centrifugation at 20,000 × g for 20 min. Polyethylenimine was added to the supernatant (0.1% v/v), incubated 15 min at room temperature, and centrifuged at 48,000 × g for 20 min. Ammonium sulfate was added to the supernatant (44.2 g/100 ml), stirred at 4 °C for 30 min, and centrifuged at 27,000 × g for 15 min. Cell pellets were resuspended in 16 ml of buffer A and dialyzed against buffer A. Dialyzed cell extract was applied to a 20-ml phosphocellulose P-11 column (Whatman) previously equilibrated with buffer A. A linear gradient (0-1 M NaCl) was applied, and Dam-containing fractions were pooled and dialyzed against 50 mM Tris-HCl, pH 7.5, 50 mM KCl, 10 mM EDTA. Bovine serum albumin (BSA, 200 µg ml¹ final w/v) and glycerol (50% final v/v) were added for storage at 20 °C. Dam protein was found to be 70% pure by SDS-PAGE and Coomassie staining. Dam protein concentrations were estimated by absorbance at 280 nm, based on molar extinction coefficients of 0.7 for phenylalanine, 5559 for tryptophan, and 1197 for tyrosine (15).

Strains, Media, and Plasmids-- The E. coli strains and plasmids used in this study are shown in Table I. Isogenic E. coli strains (Table I) containing dam-16, rpoD800, and katF13::Tn10 were constructed by P1 transduction (17). Cultures used for RNA isolation and subsequent primer extension analyses were grown at the indicated temperatures under aeration in minimal medium M9 (17) supplemented with 0.2% (v/v) glycerol as the sole carbon source. Antibiotics were added to the following final concentrations: kanamycin, 25 µg ml¹; tetracycline, 12.5 µg ml¹. The chromogenic substrate 5-bromo-4-chloro-3-indolyl-D-galactoside (X-gal; Alexis Corp.) was added to a final concentration of 40 µg ml¹ in M9 agar medium as needed. The site-directed mutagenesis protocol of Kunkel (18) was used to introduce a substitution mutation into Lrp binding site 3 of pDAL337 (6) as described previously (8) to create pDAL337-13. Plasmids pDAL337 and pDAL337-13 were transformed into DL3338 (dam-16) to produce nonmethylated plasmid DNAs. Strains harboring plasmids were grown in Luria-Bertani broth (17) containing ampicillin (100 µg ml¹) to an A₆₀₀ greater than 2. Large scale preparations of plasmid DNA of native superhelical densities were isolated by alkaline lysis and two rounds of CsCl gradient centrifugation essentially as described (19). Ethidium bromide was extracted from the DNA using H₂O/NaCl-saturated isopropanol. Plasmid DNA was precipitated with ammonium acetate and ethanol followed by resuspension in Tris-EDTA buffer (10 mM Tris-HCl, pH 8.0, 1 mM EDTA) and dialysis against the same buffer. Plasmid DNA concentrations were determined by optical density (A₂₆₀) and confirmed by gel electrophoresis. Plasmids were 100% non-methylated at GATC sites as determined by comparison of Sau3AI and MboI restriction digests. The supercoiling states of each plasmid preparation were assessed by two dimensional gel electrophoresis on 0.5% (w/v) agarose in the presence of chloroquine diphosphate as described previously (20). The DNA was transferred to Hybond-N membranes (Amersham Pharmacia Biotech) in 3 M NaCl, 0.3 M sodium citrate, using a PosiBlot pressure blotter (Stratagene), and fixed to membranes by UV cross-linking. Blots were probed with ³²P-labeled RNA synthesized from 1 µg of pTZ19U plasmid template using the Maxiscript T7 RNA polymerase in vitro transcription kit (no. 1312; Ambion). Hybridization was performed in ULTRAhyb (Ambion) at 42 °C, followed by washing as directed by the manufacturer. Autoradiography showed that each plasmid preparation contained overlapping distributions of topoisomers; the exact linking numbers were not determined. The monomeric populations of plasmid pDAL337 and pDAL337-13 were estimated to be 85% and 90% supercoiled, respectively. We estimate that approximately 70% of each plasmid preparation was supercoiled monomer.

Calculation of pap Phase Variation Switch Rates and the Fraction of Phase ON Cells-- The pap phase ON to OFF and OFF to ON switch frequencies (number of switch events per cell per generation) and the fraction of bacteria in the phase ON state were determined as described previously (21).

RNA Isolation and Primer Extension Analysis-- The oligonucleotides used for primer extension were: 5'-AGGGAATAAGGGCGACACGGAAATGTTGAATAC-3' for -lactamase and 5'-ATAGTCCTGGCCTGAATCGACAGCGTTATTTTG-3' for csiD. A DNA sequence ladder was constructed for positioning of transcription start sites using plasmid pDAL337 DNA template and the oligonucleotide primers shown above. DNA cycle sequencing was performed as directed (Life Technologies, Inc.).

For RNA isolation, E. coli strains were initially grown on M9 glycerol X-gal plates with the appropriate antibiotic(s) (Table I). Strains DL3319 and DL3321 were grown at 30 °C to an A₆₀₀ of 0.5 and then divided into two aliquots grown at 30 °C and 42 °C for another 1 h prior to RNA isolation. E. coli strains DL1504, DL2121, and DL3033 were grown at 37 °C prior to RNA isolation. To ensure that a high percentage of phase ON cells were present, bacteria were streaked on M9 glycerol medium containing X-gal indicator and blue (Lac⁺) colonies were selected. This process was repeated, and blue colonies were used to inoculate M9 glycerol cultures for RNA isolation (15 ml).

RNA was isolated as follows; 10 ml of each culture was added to 2 ml of lysis buffer (250 mM Tris-HCl, pH 6.8, 10 mM EDTA, 10% SDS) in a 50-ml conical tube and boiled for 2 min. After a 3-min incubation at room temperature, 0.5 ml of 2 M sodium acetate·³H₂O, pH 5.2, plus 10 ml of phenol (buffered with 50 mM Tris-HCl, pH 7.9) were added. After vigorous shaking for 5 min, 10 ml of chloroform was added and mixed for another 5 min. The samples were then centrifuged at ~1400 × g for 10 min. The aqueous layer was transferred to a fresh tube and the phenol and chloroform extractions repeated as above until the interface was clearly demarcated (typically 3-4 extractions). The aqueous layer of each tube was split into two 30-ml Corex tubes, followed by addition (to each) of 0.75 ml of glacial acetic acid and ammonium acetate to a final concentration of 0.2 M. RNA was precipitated by addition of two sample volumes of ethanol and incubated at 20 °C for at least 2 h. Samples were centrifuged at 10,500 × g for 30 min. Pellets were air-dried and suspended in 270 µl of diethyl pyrocarbonate-treated H₂O. DNase buffer (30 µl of 40 mM Tris-HCl, pH 7.9, 10 mM NaCl, 6 mM MgCl₂) was added along with 5 units of RNase-free DNase I (Stratagene). Samples were incubated at 37 °C for 1 h followed by phenol/chloroform (1:1) extraction. RNAs were precipitated with 1/10 volume 10 M ammonium acetate plus two volumes of ethanol (100%). RNAs were pelleted by centrifugation, air-dried, resuspended in diethyl pyrocarbonate double-distilled H₂O, and stored at 70 °C. RNA concentration was determined based on the absorbance at 260 nm.

For each primer extension reaction, 50 µg of RNA was added to 50,000 cpm of end-labeled primer, and precipitated with ethanol. Air dried pellets were resuspended in 30 µl of hybridization buffer (3 M NaCl, 0.5 M HEPES, 1 mM EDTA, pH 7.8). After overnight incubation at 30 °C, the hybridized RNA and primers were precipitated by addition of 170 µl of 0.3 M sodium acetate (pH 5.2) and 500 µl of 100% ethanol. The RNA-primer hybrids were pelleted by centrifugation, air-dried, and resuspended in 25 µl of reverse transcription buffer (50 mM Tris-HCl, pH 8, 50 mM KCl, 5 mM MgCl₂·6H₂O, 5 mM dithiothreitol, 50 µg/ml BSA, 55 mM dNTPs, and 0.8 units/µl RNasin). AMV-reverse transcriptase (40 units) was added to each reaction followed by incubation at 42 °C for 90 min. Reverse transcription reactions were terminated by addition of 1 µl of 0.5 M EDTA and 1 µl of Rnase ONE^TM, followed by a 30-min incubation at 37 °C. Ammonium acetate (100 µl of 2.5 M) was added to each sample followed by extraction with 125 µl of phenol/chloroform (1:1). RNA (aqueous phase) was precipitated in ethanol, cDNAs were resuspended in 3.75 µl of Tris-EDTA buffer, pH 7.4, and 5 µl of primer extension loading buffer (98% formamide (v/v), 1 mg/ml xylene cyanol, 1 mg/ml bromphenol blue, 10 mM EDTA) was added. Samples were separated on polyacrylamide gels (6%) in 1× TBE buffer (89 mM Tris base, 89 mM boric acid, 1 mM EDTA) containing 8 M urea.

In Vitro Transcription Reactions-- In vitro transcription was performed using supercoiled plasmid DNA (2 nM) pre-incubated for 15 min at 37 °C with varying concentrations of purified Lrp. Reactions were performed in 100 mM KCl, 40 mM Tris-HCl, pH 8, 5 mM MgCl₂, 0.1 mM EDTA, 0.1 mg/ml BSA, 2 mM dithiothreitol, 1 mM -mercaptoethanol, 6% glycerol, 80 µM S-adenosyl methionine, 500 µM NTP (125 µM each of ATP, UTP, CTP, and GTP), 0.8 units/µl RNasin. In addition, a radiolabeled, gel-purified murF DNA fragment (12,500 cpm/reaction) was included as a control for recovery of nucleic acids and subsequent quantitation of transcript levels. The 109-bp murF DNA fragment was prepared by PCR using ³²P-end-labeled oligonucleotide 5'-CGGAATTCATGATTAGCGTAACCCTTA-3' and non-labeled oligonucleotide 5'-CTGGATCCTAATACGACTCACTATAGGGAGGGGGTGATATCTGCACCTT-3' with chromosomal DNA template from E. coli strain MC4100. The murF DNA fragment was gel-purified prior to use.

Transcription was initiated by the addition of RNA polymerase (25 nM), and at the same time Dam methylase (100 to 200 nM) was added (titrated previously to completely methylate both pap GATC sites in 5 min). Transcription reactions were terminated after 5 min at 37 °C by addition of 50 µl of stop solution (50 mM Tris-HCl, pH 7.4, 50 mM EDTA, pH 8) followed by heating at 65 °C for 20 min. After the addition of 10 µg of yeast RNA, each reaction was extracted with phenol/chloroform/isoamyl alcohol (25:24:1). A portion of each sample was precipitated with ammonium acetate and ethanol and used as template for primer extension of in vitro synthesized RNAs.

In vitro synthesized RNA transcripts were analyzed by primer extension as follows. RNA transcripts were suspended in reverse transcription buffer with 50,000 cpm each of end-labeled pap or bla primer (see above). AMV-reverse transcriptase was added to each reaction followed by a 90-min incubation at 42 °C. Reactions were terminated by the addition of 1 µl of 0.5 M EDTA, pH 8.0, and 10 units of RNase ONE^TM followed by incubation at 37 °C for 30 min. Ammonium acetate (100 µl of 2.5 M) was added to each sample, and samples were then extracted with phenol/chloroform (1:1). The cDNAs were precipitated with ethanol.

Analysis of primer extension products was carried out by polyacrylamide gel electrophoresis as described above. Transcripts were quantitated using an Alpha Imager 2000 documentation and analysis system (Alpha Innotech Corp., San Leandro, CA). The pap and bla transcript values from each lane were divided by the murF loading control value and plotted relative to two 0 nM Lrp controls (RNA polymerase only).

In Vitro Methylation Protection Assays-- The in vitro methylation protection assay was performed as described for in vitro transcription reactions (see above), with the following modifications. RNA polymerase, radiolabeled murF probe, and RNasin were not added, and reactions were stopped with phenol/chloroform (1:1). Reactions were desalted by gel chromatography (P-30 spin column, Bio-Rad) and digested with EcoRI and MboI. Digested DNA was separated on 3.5% polyacrylamide gels run (1× TBE buffer). Gels were blotted by electrotransfer (ZetaProbe membranes, Bio-Rad) according to standard protocol (19). A PCR-generated DNA template incorporating a bacteriophage T7 promoter sequence was amplified using plasmid pDAL337 DNA template with the following oligonucleotides: 5'-AATTTAATACGACTCACTATAGGGATCAATTTGCCATGATGTTTTTA-3' (the italicized portion represents the T7 RNA polymerase promoter) and 5'-GATCTTTTAACCCACAAAACA-3'. An RNA probe was synthesized using the PCR product as template for T7 RNA polymerase (Maxiscript, Ambion). Briefly, 1 µg of PCR-generated template DNA was combined in transcription buffer (20 µl total volume) with the four nucleotide triphosphates, RNasin (10 units), and T7 RNA polymerase (10 units). The reaction mix was incubated at 37 °C for 4 h, and DNA template was removed by incubation with DNase I at 37 °C for 15 min. Unincorporated nucleotides were removed by gel chromatography (P-30 spin column, Bio-Rad). Blots were hybridized using an RNA probe (87 nucleotides) hybridizing to pap DNA sequences between GATC^dist and GATC^prox (Fig. 4A). Blots were probed in ULTRAhyb (Ambion) at 42 °C with 10⁶ cpm ml¹ of the RNA probe, followed by washes as directed by the manufacturer. Restriction fragments were visualized by autoradiography and quantified (see above).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Transcription at the papBA Promoter Is Dependent upon the RNA Polymerase Sigma Factor ⁷⁰-- The primary sigma factor used for transcription in E. coli during logarithmic growth is ⁷⁰ (22). Since the pap operon is transcribed in E. coli cells growing logarithmically (6), it seemed likely that the RNA polymerase-⁷⁰ holoenzyme transcribes the pap operon. To test this hypothesis, a mutation in the rpoD gene (rpoD800) encoding ⁷⁰ was transduced into E. coli containing a papBA-lacZ operon fusion. At the permissive temperature (30 °C) the rpoD800 allele does not significantly alter the transcription of ⁷⁰-dependent promoters whereas at the nonpermissive temperature (42 °C) transcription is totally blocked (23, 24). Analysis of the effect of rpoD800 on pap gene expression showed that even at the permissive temperature pap transcription was blocked as evidenced by the absence of bacterial cells in the phase ON transcription state (Table II). These results show that ⁷⁰ is required for pap transcription.

Transcription at many promoters is initiated by both the ⁷⁰ and ³⁸ sigma factors, which share overlapping DNA binding specificities (25). Additionally, because a number of virulence genes are dependent on ³⁸ for transcription (26-28), we determined if ³⁸ played a role in activation of pap transcription. As shown in Table II, introduction of a null mutation in rpoS encoding ³⁸ (katF13::Tn10) did not significantly alter pap transcription, as evidenced by measurement of pap phase variation switch frequencies. Together, these data indicate that ⁷⁰ is the principal sigma factor used for pap transcription.

Further analysis of the roles of ⁷⁰ and ³⁸ in pap transcription was carried out by primer extension analysis. For these studies, a pap mutant (pap-13, Fig. 1) containing a substitution mutation in Lrp binding site 3 was used since it confers a constitutive pap transcription phenotype (8). First, wild type pap and pap-13 transcription were analyzed at the permissive and restrictive temperatures in strains containing rpoD800. In log phase, rpoD800 prevented wild type papBA transcription at both the permissive (30 °C) and non-permissive (42 °C) temperatures (Fig. 2A). In contrast, mutant pap-13 was transcribed at 30 °C but not at 42 °C. These results indicate that ⁷⁰ is the primary sigma factor responsible for pap transcription in log phase cells. When pap-13 RNA was isolated from stationary phase cells of strains with a transposon insertion in rpoS, papBA transcription was not reduced (Fig. 2B) whereas transcription from csiD, a known ³⁸-dependent promoter (29), was abolished. Similar results were obtained using a wild type papBA promoter.³ These studies support the hypothesis that ⁷⁰ is the primary sigma factor responsible for papBA transcription.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   The papBA promoter regulatory region. A, Lrp binding sites are numbered 1 to 6 with the distance (base pairs) relative to the papBA transcription start site shown. The GATCdist and GATCprox DNA sequences are located within Lrp binding sites 5 and 2, respectively. The methylated adenines of the GATCdist and GATCprox sites are located at 154 and 52 relative to the papBA transcription start site, respectively. The DNA methylation patterns of phase ON and phase OFF E. coli are shown (nonmethylated GATC (open circle), methylated GATC (filled circle)). B, the nucleotide sequence of the papBA promoter region from 62 to +5 is shown. The promoter consensus for E. coli 70 holoenzyme is also shown (38, 39). The 35 and 10 hexamers of papBA are underlined. The pap-13 mutation is in italics. Lrp binding sites 2 and 3 are boxed.

View larger version (86K): [in this window] [in a new window]   Fig. 2.   Primer extension analysis of in vivo papBA transcripts. Cultures were grown in M9 minimal medium (0.2% v/v glycerol). Radiolabeled oligonucleotide primers complementary to papBA (panels A and B) or csiD (panel B) were hybridized with RNA and used as templates for reverse transcription. Reaction products were analyzed by polyacrylamide gel electrophoresis as described under "Experimental Procedures." A, lanes 1 and 2, DL3319; lanes 3 and 4, DL1504; lanes 5 and 6, DL3321. The pap and rpoD (70) genotypes, growth temperature, and percentage of phase ON cells (see "Experimental Procedures") are shown at the bottom (NA, not applicable). The papBA transcripts are indicated with an arrow. The pap DNA sequence ladder (CATG) was generated with the same oligonucleotide used for primer extension of papBA transcripts. B, lane 1, DL2121; lane 2, DL3033 (katF13::Tn10). Both bacterial cultures were 100% phase ON.

Lrp Specifically Represses papBA Transcription-- Evidence obtained previously indicated that Lrp acts as both an activator and a repressor of pap transcription in vivo (13). To determine if Lrp can directly repress papBA transcription, an in vitro transcription system was developed using RNA polymerase-⁷⁰ holoenzyme (E⁷⁰). Transcription from both the papBA and bla (-lactamase) promoters was simultaneously monitored in the presence or absence of Lrp in vitro using a primer extension assay (see "Experimental Procedures"). The level of papBA transcription was reduced at 8 nM Lrp and abolished at 16 nM Lrp, whereas bla transcription was not reduced until addition of 128 or 256 nM Lrp and abolished at 512 nM Lrp (Fig. 3). These results showed that Lrp was sufficient for specific repression of papBA transcription.

View larger version (46K): [in this window] [in a new window]   Fig. 3.   Lrp specifically represses papBA transcription. A, in vitro transcription was performed at 37 °C with plasmid template pDAL337, and cDNA products were separated by polyacrylamide gel electrophoresis as described under "Experimental Procedures." Equal counts of a radiolabeled PCR product were added to each transcription reaction to serve as a recovery control (RC). The Lrp concentration (nM) is shown at the bottom. Lrp was added 15 min prior to addition of RNA polymerase. A control reaction lacking RNA polymerase is shown in lane 1. B, quantitation of pap (, solid line) and bla (, dotted line) transcripts relative to the 0 nM Lrp controls (A, lanes 2 and 13) were performed as described under "Experimental Procedures."

Lrp-dependent Methylation Protection at the GATC^prox Site Correlates with Repression of papBA Transcription-- Lrp blocks methylation of the GATC^prox site in phase OFF cells and blocks methylation of GATC^dist in phase ON cells (6). Van der Woude et al. (14) showed recently that Lrp specifically blocks methylation of the pap GATC^dist and GATC^prox sites in vitro. Therefore, we developed an in vitro methylation protection assay to monitor Lrp occupancy at the pap GATC sites. Lrp was incubated with nonmethylated pap DNA followed by addition of Dam (see "Experimental Procedures"). The methylation state of each GATC site was assessed by digestion with MboI, which cuts only nonmethylated GATC sites. Sufficient Dam was added to methylate both the GATC^dist and GATC^prox sites as evidenced by the presence of the 1,784-bp pap DNA fragment (see DNA fragment R, Fig. 4A). Lrp preferentially protected methylation of GATC^prox as evidenced by the presence of the 1,142-bp DNA fragment (see DNA fragment S, Fig. 4, A and B). Maximal methylation protection of GATC^prox occurred at 16 nM Lrp, which was the same concentration observed to cause maximal repression of pap transcription (compare Figs. 3B and 4C). In contrast, methylation protection of GATC^dist was not observed until 256 nM Lrp (Fig. 4C). At higher concentrations of Lrp, both GATC sites were protected from methylation as evidenced by the presence of 744-, 1,142-, and 102-bp pap DNA fragments (Fig. 4A, 512 nM Lrp). These data showed that Lrp blocked methylation of the GATC^prox site at the same levels required for specific repression of papBA transcription. In contrast, methylation protection of GATC^dist did not correlate with repression of papBA transcription.

View larger version (37K): [in this window] [in a new window]   Fig. 4.   Binding of Lrp to the pap GATCprox site correlates with repression of papBA transcription. A, DNA fragments that result from MboI restriction digestion of pap regulatory DNA when GATCdist and/or GATCprox are protected from Dam methylation by Lrp. DNA fragments are identified by letter and sizes (base pairs) are shown. B, in vitro methylation protection assays were performed as described in "Experimental Procedures" using wild-type pap DNA (plasmid pDAL337) and the 87-nucleotide RNA probe shown in A. DNA fragment R is fully methylated pap DNA that was not digested by MboI; DNA fragments S and V represent digestion at a nonmethylated GATCprox or GATCdist site, respectively. DNA fragment Z represents cutting at both sites. DNA fragments T, X, and Y were detected by the RNA probe, likely as a result of extension of RNA outside of the 85-nucleotide probe region due to contaminating PCR products. C, quantitation of DNA fragment S, V, and Z from B. DNA fragment S (, nonmethylated GATCprox), V (, nonmethylated GATCdist) and Z (*, nonmethylated GATCdist and GATCprox).

Binding of Lrp at the GATC^prox Site Is Required for Repression of papBA Transcription-- The data shown in Fig. 4 indicated that binding of Lrp to GATC^prox inhibited pap transcription. To determine if binding of Lrp to GATC^prox is required for repression of pap transcription, the pap-13 mutant containing a mutation in Lrp binding 3 was analyzed (Fig. 1B). This mutation reduces the affinity of Lrp for the GATC^prox region (2-3-fold) and increases its affinity for GATC^dist (4-fold) (8). Repression of transcription from wild-type pap DNA template occurred at 8 nM (Fig. 3A), whereas repression of the pap-13 template was observed only at Lrp levels higher than 256 nM (Fig. 5A). At these high levels of Lrp, repression of the bla gene, which is not Lrp regulated, was also observed (Fig. 5A). Thus, repression of pap-13 transcription at the higher levels of Lrp appeared to be due to nonspecific binding of Lrp.

View larger version (54K): [in this window] [in a new window]   Fig. 5.   Lrp does not specifically repress transcription from the phase locked ON pap-13 mutant DNA template. A, in vitro transcription was performed at 37 °C with plasmid template pDAL337-13 (Table I), and cDNA products were separated by polyacrylamide gel electrophoresis as described under "Experimental Procedures." Equal counts of a radiolabeled PCR product were added to each transcription reaction to serve as a recovery control (RC). The Lrp concentration (nM) is shown at the bottom. Lrp was added 15 min prior to addition of RNA polymerase. A control reaction lacking RNA polymerase is shown in lane 1. B, quantitation of pap (, solid line) and bla (, dotted line) transcripts relative to the 0 nM Lrp controls " (A, lanes 2 and 13).

Methylation protection analysis indicated that Lrp bound to the GATC^dist site with highest affinity (K_d = 4-8 nM, Fig. 6B) in contrast to binding of Lrp to GATC^dist of wild type pap (estimated K_d = 256-512 nM, Fig. 4C). Maximal binding of Lrp to GATC^dist of pap-13 occurred at 16 nM Lrp, at which level no repression of pap-13 transcription was observed (Fig. 5). In contrast, binding of Lrp to GATC^prox of pap-13 DNA occurred at concentrations greater than 256 nM Lrp (Fig. 6, A and B). At these high Lrp levels, nonspecific repression of pap-13 transcription was observed (Fig. 5). Thus, Lrp did not specifically repress pap-13 transcription due to its reduced affinity for GATC^prox. Together, these results strongly indicate that binding of Lrp at GATC^prox, but not GATC^dist, represses pap transcription.

View larger version (42K): [in this window] [in a new window]   Fig. 6.   Lrp binds with highest affinity to the GATCdist site of phase locked ON pap-13 DNA. A, in vitro methylation protection assays were performed at 37 °C with plasmid template pDAL337-13. Refer to Fig. 4A for identification of pap DNA fragments. B, quantitation of DNA fragments V, S, and Z from A. DNA fragment V (, nonmethylated GATCdist), DNA fragment S (, nonmethylated GATCprox), and DNA fragment Z (*, nonmethylated GATCdist and GATCprox).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In this study, we have demonstrated for the first time that Lrp alone can establish the phase OFF and phase ON pap DNA methylation patterns in vitro. Formation of the phase OFF DNA methylation pattern by Lrp binding at GATC^prox was shown to result in repression of papBA transcription in vitro. In contrast, no repression was observed when Lrp established the phase ON methylation pattern by binding to GATC^dist in vitro (Figs. 5 and 6). The same relationships between DNA methylation patterns and transcription phenotypes were observed previously in vivo (6). In phase OFF cells GATC^prox was protected from methylation, whereas in phase-locked ON cells containing pap-13, GATC^dist was protected from methylation.

In vivo in phase ON cells PapI and Lrp are required for protection of GATC^dist from Dam methylation and activation of papBA transcription (8). Previously, PapI was shown to increase the affinity of Lrp for the GATC^dist region of pap (8). However, in the pap-13 mutant containing a 6-bp substitution mutation in Lrp binding site 3 (Fig. 1B), PapI is no longer required for methylation protection of GATC^dist or for papBA transcription in vivo (8). Our in vitro analysis showed that the pap-13 mutation reversed the binding affinities of Lrp for the two pap GATC sites, enabling Lrp binding at GATC^dist in the absence of PapI. In wild-type pap, Lrp bound with highest affinity to GATC^prox, whereas in pap-13 Lrp bound with highest affinity to GATC^dist (compare Figs. 4C and 6B). In vitro transcription analysis indicated that binding of Lrp to GATC^prox but not GATC^dist correlated with repression of papBA transcription (Figs. 3 and 4). Analysis of the pap-13 mutant showed that Lrp could not specifically repress papBA transcription (Fig. 6). By disrupting the binding of Lrp to the GATC^prox region with the pap-13 mutation, a direct link between binding of Lrp at GATC^prox and repression of pap transcription was established (Fig. 5).

The pap GATC^prox site also plays a critical role in activation of pap transcription. Previous results showed that mutation of the GATC^prox site to GCTC sequence blocked pap transcription (6). The GCTC^prox mutation, however, did not significantly reduce Lrp's affinity compared with nonmethylated wild-type pap DNA. Thus, the effects of the GCTC^prox mutation were attributed to the inability of GCTC sequence to be methylated by Dam. In support of this hypothesis, transcription in both wild type and the phase-locked ON GCTC^dist mutant was blocked in dam cells (6). These data strongly indicated that methylation of GATC^prox was essential for pap transcription.

The mechanism by which methylation of GATC^prox activates papBA transcription is not clear. In phase ON cells the GATC^prox site is fully methylated (Fig. 1). Since our results here showed that Lrp blocks GATC^prox methylation in vitro (Fig. 4), it seems likely that Lrp is not bound at this site in phase ON cells. Based on our model for Pap phase variation, Lrp moves from promoter proximal papBA DNA sites to promoter distal sites during the transition to the phase ON transcription state (9). Methylation of GATC^prox could assist in this transition by reducing the affinity of Lrp for pap DNA sites proximal to the promoter, similar to the reduction in affinity of Lrp-PapI for promoter distal sites that results from methylation of GATC^dist (10). Alternatively, methylation of GATC^prox could increase the affinity of RNA polymerase-⁷⁰ for the papBA promoter, facilitating transition to the phase ON state or alter the binding of an additional regulatory factor(s).

The mutant allele of ⁷⁰ (rpoD800) blocked transcription at the wild type papBA promoter at the permissive temperature (30 °C) as well as the non-permissive temperature (42 °C) (Fig. 2). This is unusual since rpoD800 had little to no effect on transcription of other genes at the permissive temperature (23). RpoD800 contains a 14-amino acid in frame deletion of residues 330-343 of ⁷⁰, which causes it to become unstable at temperatures above 37 °C (30). Although the levels of RpoD800 appear to be comparable to wild type RpoD at the permissive temperature, RpoD800 appears to have a lower affinity for core RNA polymerase at the permissive temperature, which could result in a lower level of RNA polymerase-⁷⁰ holoenzyme (31). Transcription of papBA may be more sensitive to a lowered RNA polymerase level than other genes since it may compete with Lrp for binding near GATC^prox. Based on this hypothesis, disruption of binding of Lrp to GATC^prox should restore papBA transcription at the permissive temperature in cells containing RpoD800. Analysis of pap-13 indicated that, unlike wild type pap, transcription occurred at the permissive temperature (30 °C) (Fig. 2A). The pap-13 mutation, which is located between the 35 and 10 papBA hexamer RNA polymerase binding sites, significantly reduced the affinity of Lrp for GATC^prox (8). Thus, these data support the hypothesis that RNA polymerase competes with Lrp for binding at the papBA promoter.

Although our data showed that Lrp specifically repressed papBA transcription initiated by RNAP-⁷⁰, the repression mechanism has yet to be determined. Previous work showed that Lrp repressed transcription from the D-amino acid dehydrogenase (dad) promoter in vitro (32). Lrp appeared to bind near the 35 hexamer region of the dad promoter, similar to the binding of Lrp to the GATC^prox region of the papBA promoter. It is possible, therefore, that Lrp could sterically interfere with binding of RNA polymerase to these promoters. Lrp could repress transcription by alternative mechanisms such as interaction with transcriptional regulators or RNA polymerase. For example, work by Rhee et al. (33) showed binding of Lrp to the leader-attenuator region of ilvGMEDA 226-bp downstream of the transcription start site is necessary for repression. Repression of ilvGMEDA occurred at the same concentration of Lrp (16 nM) as repression of papBA transcription (Fig. 3 and Ref. 33).

A number of features of the methylation-dependent regulatory system of pap are shared by other pili operons in E. coli (34) and Salmonella,⁴ which play important roles in virulence. Recently, it was shown that DNA methylation is essential for Salmonella virulence (35). The pap system described here provides a useful paradigm for understanding the mechanisms by which DNA methylation patterns are established, altered by environmental stimuli (36, 37) and control gene expression.

	ACKNOWLEDGEMENTS

We thank Dr. M. Krabbe for purified Lrp, Dr. R. Matthews and Dr. W. Reznikoff for bacterial strains, and Drs. B. A. Braaten and M. Krabbe for plasmids and critical reading of this manuscript.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant AI23348 (to D.A.L.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Dept. of Molecular, Cellular, and Developmental Biology, Rm. 3129, Bioscience II Bldg., University of California, Santa Barbara, CA 93106. Tel.: 805-893-5597; Fax: 805-893-7558; E-mail: low@lifesci.ucsb.edu.

² M. Krabbe, unpublished data.

³ N. J. Weyand, unpublished data.

⁴ Nicholson, B. P., and Low, D. A. (2000) Molecular Microbiology, in press.

	ABBREVIATIONS

The abbreviations used are: pap, pyelonephritis-associated pili; Lrp, leucine-responsive protein; Dam, DNA adenine methylase; GATC^prox, GATC site proximal to the papBA promoter; GATC^dist, GATC site distal to the papBA promoter; PAGE, polyacrylamide gel electrophoresis; BSA, bovine serum albumin; X-gal, 5-bromo-4-chloro-3-indolyl--D-galactoside; PCR, polymerase chain reaction; bp, base pair(s); AMV, avian myeloblastosis virus.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				N. Holden, M. Totsika, L. Dixon, K. Catherwood, and D. L. Gally Regulation of P-Fimbrial Phase Variation Frequencies in Escherichia coli CFT073 Infect. Immun., July 1, 2007; 75(7): 3325 - 3334. [Abstract] [Full Text] [PDF]

				T. K. Wagner and M. H. Mulks Identification of the Actinobacillus pleuropneumoniae Leucine-Responsive Regulatory Protein and Its Involvement in the Regulation of In Vivo-Induced Genes Infect. Immun., January 1, 2007; 75(1): 91 - 103. [Abstract] [Full Text] [PDF]

				J. Casadesus and D. Low Epigenetic Gene Regulation in the Bacterial World Microbiol. Mol. Biol. Rev., September 1, 2006; 70(3): 830 - 856. [Abstract] [Full Text] [PDF]