A method for direct cloning of Fur-regulated genes: identification of seven new Fur-regulated loci in Escherichia coli

Natalia Vassinova1 and Dmitri Kozyrev1

Department of Biophysics, St. Petersburg State Technical University, 29 Polytechnicheskaya Str, St. Petersburg 195251, Russia1

Author for correspondence: Dmitri Kozyrev. e-mail: vikey{at}genet.hop.stu.neva.ru


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
A strain that allows the cloning of Fur-regulated loci was constructed. The strain, named FUR-SEL1, contains a chromosomal fhuA'–'cat transcriptional fusion that is expressed from the Fur-regulated promoter, fhuAp. Therefore, Fur boxes introduced on a multicopy plasmid can cause derepression of the fusion by titrating the Fur repressor and thereby confer chloramphenicol resistance, which can be used as a selectable phenotype for cloning Fur-regulated loci. However, a number of additional mutations had to be introduced before FUR-SEL1 could be used for cloning Fur-regulated genes. The principal approach consisted of introducing a leaky fur mutation that ensures a more than 106-fold increase in chloramphenicol resistance for FUR-SEL1 transformants carrying FUR-box-containing plasmids. To verify that the cloning procedure selects Fur-regulated genes, 10 recombinant plasmids chosen at random among the ones selected with FUR-SEL1 were analysed by FURTA (Fur-titration assay), a method for identification of Fur-regulated genes. In addition, the nucleotide sequences of their chromosomal inserts were determined. Besides known Fur-regulated genes, seven Escherichia coli loci which have not previously been shown to be Fur regulated were found, including the pgmA and nrdHIEF genes, predicted ORF yhhY and four promoters identified first in this study. Three of the promoters preceded the nohA gene, and ORFs ygaC and yhhX. The fourth was located upstream of orf78 predicted in this work. The regulation of the promoter activities by iron and the involvement of Fur in this regulation were shown. Employing FUR-SEL1 for cloning Fur-regulated loci from other enterobacteria is discussed.

Keywords: Fur regulon, Fur repressor, Fur box, iron regulation, Escherichia coli

Abbreviations: FURTA, Fur-titration assay; Lac-FURTA, FUTA using lacZ as a reporter; Cm-FURTA, FURTA using cat as a reporter


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Fur (ferric uptake regulator) is a global transcriptional regulator, using ferrous iron as a corepressor, that controls the expression of a great many operons in E. coli (Hantke & Braun, 1998 ). Fur-like proteins have also been identified in many other bacterial species. By binding regulatory DNA sequences, Fur boxes, located in the promoter regions of Fur-controlled genes, Fur regulates not only the genes responsible for iron uptake, but also genes involved in aerobic respiration, chemotaxis, synthesis of amino acids and DNA precursors, sugar metabolism, protecting the cell from oxidative damage, and genes encoding bacterial toxins (Stojiljkovic et al., 1994 ; Compan & Touati, 1993 ; Prodromou et al., 1992 ; Calderwood & Mekalanos, 1987 ). A number of new Fur-regulated loci of unknown function has recently been found in the Escherichia coli, Salmonella typhimurium and Pseudomonas aeruginosa genomes (Stojiljkovic et al., 1994 ; Tsolis et al., 1995 ; Ochsner & Vasil, 1996 ).

A method, called FURTA (Fur-titration assay), for the identification of Fur-regulated genes has been developed (Stojiljkovic et al., 1994 ) that is based on multiple plasmid-encoded Fur boxes derepressing chromosomal Fur-regulated genes by titrating the Fur protein. Both FURTA and strategies employing tandem lacZ fusions demand labour-intensive analysis of many clones. Identification of seven Fur-regulated genes of S. typhimurium required screening 10000 independent MudJ insertion mutants. Screening 10000 clones of a Salmonella plasmid bank by FURTA allowed the isolation of eight Fur-regulated loci (Tsolis et al., 1995 ).

An alternative in vitro method, the cycle-selection procedure, is based on the specific interaction of the Fur repressor with chromosomal fragments containing Fur boxes, purification of Fur–DNA complexes and PCR amplification of the enriched DNA pool for the next cycle (Ochsner & Vasil, 1996 ). This method was applied to the isolation of Fur targets in P. aeruginosa. This is a very laborious procedure requiring the purification of the Fur protein. In addition, the percentage of isolated fragments for which the specificity of the Fur–DNA interaction was not verified by subsequent analysis was high.

In this study we describe an approach for cloning Fur-regulated loci that is based on direct in vivo selection.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Strains, media and growth conditions.
The bacterial strains and the plasmids used in this study are listed in Table 1. Bacteria were grown in LB broth or on LB agar plates unless indicated otherwise. Antibiotics, when required, were added at the following concentrations: ampicillin, 100 mg l-1; kanamycin, 25 mg l-1; streptomycin, 100 mg l-1; tetracycline, 3·0–12·5 mg l-1; chloramphenicol, 2·5–7·5 mg l-1. M9 minimal medium supplemented with threonine, histidine, arginine (40 mg l-1) and 0·2% (w/v) glucose as the carbon source was used for selection after conjugation.


View this table:
[in this window]
[in a new window]
 
Table 1. E. coli strains and plasmids used in this study

 
For selection of fur mutants, the medium described previously (Hantke, 1987 ; Silver et al., 1972 ) was used except that desferal was removed.

For quantitative assay of ß-galactosidase activity in the FURTA technique, strains were grown on EMBO agar medium (10 g tryptone, 1 g yeast extract, 5 g NaCl, 2 g KH2PO4 and 12 g agar l-1) supplemented with 0·025 mM (NH4)2Fe(SO4)2. The ß-galactosidase and alkaline phosphatase activities of the iron-regulated fusions carried on pUJ10 derivatives were determined using cells grown on iron-rich and iron-deficient agar medium. LB broth was used as iron-rich medium (de Lorenzo et al., 1988 ). To create conditions of iron starvation, LB broth was supplemented with 0·2 mM 2,2'-dipyridyl. ß-Galactosidase activity was determined by the method of Miller (1972) and alkaline phosphatase activity by the method of Brickman & Beckwith (1975) , using cells treated with chloroform and 0·1% SDS.

Lac-FURTA was carried out on Lac-EMBO agar plates (Miller, 1972 ) containing (NH4)2Fe(SO4)2.

Genetic techniques.
Phage P1vir was used for generalized transduction of operon fusions into different genetic backgrounds. Transductants were selected on plates containing 25 mM sodium citrate and the appropriate antibiotic. Selection for fur mutants of E. coli was carried out according to Hantke (1987) .

The lacZ deletion {Delta}(lacZ)M15 was introduced into strain VK-6C using a two-step procedure. First, F'::Tn10 proA+B+ lacIq {Delta}(lacZ)M15 was introduced into VK-6C by conjugation with XL-1 Blue. The VK-6C(F') exconjugants were selected on medium containing streptomycin and tetracycline. To transfer {Delta}(lacZ) from F' to the chromosome, ~103 cells of strain VK-6C(F') were inoculated into 1ml LB broth and 3·8mg ethidium bromide l-1 was added to induce homologous recombination between F' and the chromosome, and then to eliminate F'. After overnight incubation at 37 °C, the bacterial culture was diluted and plated on X-Gal and IPTG-containing medium to allow Lac- colonies to be selected. After streaking on X-Gal medium, several Lac- clones were transformed with pUC19 and their ß-galactosidase activity was found to be complemented by the lac {alpha} peptide of pUC19. The data indicated that the Lac- clones carry only the {Delta}(lacZ)M15 allele of the lac gene. One of them was designated FUR-SEL1.

Recombinant DNA techniques.
Preparations of plasmid and chromosomal DNA, restriction digestions and ligations were performed according to standard techniques (Sambrook et al., 1989 ). To prepare plasmid DNA for digestion with BclI the E. coli dam strain WA321 was used. Plasmid DNA was introduced into bacteria by transformation of CaCl2-treated cells (Cohen et al., 1973 ) or by electroporation (Dower et al., 1988 ).

Construction of lacZ and phoA gene fusion plasmids.
Isolation of transcriptional fusions between the Fur-regulated genes found in this work and the lacZ or phoA genes was accomplished using plasmid pUJ10 (de Lorenzo et al., 1990 ) in strain CC118. This vector contains several unique sites flanked by promoterless ß-galactosidase (lacZ) and alkaline phosphatase (phoA) genes oriented in opposite directions. The Sau3A chromosomal insertions from pEF7, pEF19, pEF20, pEF29, pEF34 and pEF63, and the 585 bp Sau3A fragment containing the promoter of the bla gene encoding ß-lactamase from pUC19 were isolated after separation in 5% acrylamide gels and ligated into BglII-digested and dephosphorylated pUJ10. Transformants were selected on LB medium with ampicillin. The structures of pUJ10 derivatives and the orientations of their Sau3A inserts were verified by restriction enzyme mapping.

Construction of the promoter probe vector pBR-{Delta}.
pBR-{Delta} is a pBR322 deletion derivative with the promoterless tetracycline-resistance gene tet. For construction of this vector, the 29 bp EcoRI–HindIII fragment of pBR322 containing the -35 element of the tet promoter was replaced by the 0·6 kb EcoRI–HindIII fragment including the 3' portion of the kanamycin resistance gene from pUC-4K (Vieira & Messing, 1982 ). In this study, the vector was used for the generation of transcriptional tet fusions by replacing its truncated kanamycin resistance gene by the EcoRI–HindIII fragments including the chromosomal insertions from plasmids of the pEF series.

Construction of pEF-A plasmid series.
To obtain pEF derivatives pEF19-A, pEF34-A and pEF63-A that contain Sau3A chromosomal insertions in the opposite orientation to pEF19, pEF34 and pEF63, respectively, the corresponding Sau3A fragments were recloned in pUC19 with strain FUR-SEL1. The desired constructions were identified by restriction analysis.

Construction of strain VK-0C containing the fhuA'–'cat operon fusion located on the chromosome.
Strain VK-0C carrying the chromosomal transcriptional fusion fhuA'–'cat was constructed in a two-step procedure. First, the fhuA'–'cat fusion was constructed in plasmid pFC2 (Fig. 1) and transferred from the plasmid to the chromosome of strain CA77 by in vivo homologous recombination. To enhance recombination frequency, CA77 cells transformed with pFC2 were UV irradiated. The recombinants resulting from double crossovers were selected with phage T5 which uses FhuA for infection. To distinguish between the recombinants selected in this way and spontaneous fhuA mutants, the fhuA'–'cat fusion was transferred from several T5-resistant clones into AB1157 via conjugation. As the fhuA locus is at about 4 min on the E. coli physical map, it is closely linked with the leu+ (2 min) and pro+ (6 min) markers. Leu+ Pro+ exconjugants were selected and tested for resistance to chloramphenicol, phage T5 and for sensitivity to tetracycline. To obtain a strain more isogenic with AB1157, the fhuA'–'cat fusion of a Cmr/T5r/Tcs exconjugant was transferred into AB1157 by P1vir transduction, yielding strain VK-0C that was used further as the parental strain for constructing FUR-SEL1.



View larger version (21K):
[in this window]
[in a new window]
 
Fig. 1. Construction of fhuA'–'cat gene fusion. Plasmid pFHUA2 was made by cloning the 4·5 kb PstI fragment including the fhuA gene from pLC19-19 into the single PstI site of pBR322. pFC2 is a result of insertion of the promoterless portion of cat gene into the coding sequence of fhuA in plasmid pFHU2. The 0·78 kb BamHI fragment of plasmid pCM4 was used as the source of the cat cartridge. The fragment was inserted into the BclI(b) site of pFHU2 located within the fhuA gene in the appropriate orientation to produce a fhuA'–'cat fusion where the transcription of cat is dependent on the fhuA promoter. To distinguish between the desired construction and the cat insertions into the BclI(a) and BclI(c) sites positioned beyond the fhuA gene, the selection of recombinant plasmids was performed in strain 803-8A (fhuA mutant) with bacteriophage T1 and the appropriate antibiotics. Only the clones harbouring plasmids carrying the disrupted fhuA gene could exhibit the resistance to T1, which uses FhuA for infection. Gene fusion plasmids from several Tcr Cmr T1r clones were analysed by restriction mapping to verify they had the required structure.

 

   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Construction and characterization of strain FUR-SEL1
The procedure used for cloning Fur-regulated loci is based on the FURTA. As shown previously, a Fur box introduced on a high-copy-number plasmid can compete with Fur boxes located on the chromosome for the Fur repressor (Stojiljkovic et al., 1994 ). Therefore, if a plasmid-encoded Fur box is introduced into a strain carrying a reporter gene, whose transcription is under control of Fur, it will titrate the repressor, causing expression of the reporter gene. The FURTA technique employs a Fur-regulated lacZ fusion as a reporter gene that allows the detection of transformants carrying multicopy Fur-binding sites as Lac+ colonies on MacConkey agar plates. In contrast to the traditional FURTA, we used a strain (FUR-SEL1) carrying a fhuA'–'cat fusion expressed from the Fur-regulated fhuA promoter. This approach allows plasmid-encoded Fur boxes to confer chloramphenicol resistance on FUR-SEL1, which can be used as a selectable phenotype for cloning Fur-regulated loci.

Strain VK-0C was used to determine the levels of chloramphenicol resistance achieved for transformants with a high-copy-number plasmid containing a Fur box. The plating efficiency of the strain transformed with pUC19 and pUCFUR1 carrying a synthetic Fur box consensus sequence was tested on iron-rich medium containing chloramphenicol (7·5 mg l-1). The survival of strain VK-0C containing pUCFUR1 was found to be increased by about 4x103-fold compared to that of the strain with pUC19 (Table 2). However, in spite of the fact that VK-0C with pUC19 virtually completely loses its ability to grow on medium with chloramphenicol (its plating efficiency becomes less than 2·0x10-7), which excludes undesirable background in cloning, the 4x103-fold increase in the survival of VK-0C caused by plasmid-encoded Fur boxes is not enough for the strain to be applied successfully for cloning Fur-regulated loci. The reason for this is that the absolute value of survival still remains low (~10-3).


View this table:
[in this window]
[in a new window]
 
Table 2. Plating efficiency (PE) of strains VK-0C and VK-6C

 
To enhance the difference in the level of cat expression between the cells transformed with pUC19 and Fur-box-containing plasmids we tried to find a leaky fur mutation that would reduce the affinity of Fur for the fhuA promoter and, thereby, allow Fur to be titrated more easily by a Fur box introduced on a multicopy plasmid.

On analysing the effect of pUCFUR1 on chloramphenicol resistance for a collection of 24 fur mutants of strain AB1157 (see Methods), we found that strain VK-6C exhibited the greatest increase in resistance. The survival of strain VK-6C containing plasmid pUCFUR1 on chloramphenicol medium was increased by 4·5x106-fold compared to that of the same cells transformed with pUC19 (Table 2). This increased difference in chloramphenicol resistance accompanied by the increase in the plating efficiency (up to 0·7) of the pUCFUR1-containing strain made it likely that strain VK-6C would be suitable for cloning chromosomal Fur boxes.

In addition, the {Delta}(lacZ)M15 mutation was introduced into strain VK-6C to allow screening for recombinants as Lac- clones on X-Gal medium in contrast to clones containing pUC19 exhibiting a Lac+ phenotype. The resulting strain was named FUR-SEL1 and was used for cloning Fur-box-containing fragments of the E. coli chromosome.

Cloning and functional analysis of Fur-regulated loci
Sau3A fragments of the AB1157 chromosome were inserted into the BamHI site of pUC19. FUR-SEL1 cells were transformed by electroporation, the transformants were selected on ampicillin- and chloramphenicol-containing plates and then screened for the presence of recombinant plasmids on X-Gal medium.

The chloramphenicol resistance of the clones selected in this way may be caused not only by recombinant plasmids carrying Fur boxes but also by chromosomal mutations, for instance in cmlAB. Therefore, the recombinant plasmids from the clones selected with chloramphenicol were retransformed into FUR-SEL1 and tested by repeating the FURTA in order to distinguish the clones containing recombinant plasmids with Fur boxes from those whose chloramphenicol resistance is due to a chromosomal mutation. All plasmids examined gave positive results in the second Cm-FURTA. The plating efficiencies of FUR-SEL1 cells transformed by the plasmids were 105–106 times higher than the FUR-SEL1 cells containing pUC19.

It is possible that recombinant plasmids carrying chromosomal fragments other than Fur boxes could cause a nonspecific increase in chloramphenicol resistance, not connected with fhuA'–'cat derepression. To confirm the reliability of the selection technique, fhuF'–'lacZ and fiu'–'lacZ operon fusions, in which expression of ß-galactosidase emanates from Fur-regulated promoters fhuFp and fiup, respectively, were tested in the FURTA. If the recombinant plasmids obtained carry Fur boxes they should cause derepression not only of the fhuA promoter but also of other Fur-regulated operons.

Ten recombinant plasmids (pEF7, 10, 19, 20, 23, 29, 33, 34, 63, 67) chosen at random among those containing putative Fur boxes were introduced into strains H1717 and H1698-1. Plasmids pUCFUR1 and pUC19 were used as positive and negative controls, respectively. Derepression of the operon fusions caused by the pEF plasmids was detected on Lac-EMBO agar medium supplemented with (NH4)2Fe(SO4)2. All 10 recombinant plasmids and the control plasmid pUCFUR1 were positive in this Lac-FURTA. They conferred a Lac+ phenotype (wine-coloured colonies) to cells carrying reporter fusions (Table 3). In contrast, the cells containing pUC19 exhibited a Lac- phenotype (pink colonies on Lac-EMBO plates).


View this table:
[in this window]
[in a new window]
 
Table 3. Induction of fusion operons fhuF'–'lac and fiu'–'lac caused by pEF plasmids

 
In addition, pEF-stimulated derepression of fhuF'–'lacZ was confirmed by measuring the activity of ß-galactosidase in strain H1717 harbouring pEF plasmids. As seen in Table 3, the level of ß-galactosidase activity reached in the strain carrying any of the 10 pEF plasmids was increased by two- to sevenfold compared to that reached in the same strain containing the vector control.

Sequence analysis and mapping of Fur-regulated loci cloned in FUR-SEL1
Automated sequencing of the 10 recombinant plasmids shown to be positive in the Cm-FURTA and the Lac-FURTA was performed. Comparison of their nucleotide sequences with genes known to be Fur-regulated revealed that six of the plasmids contained promoter regions of genes responsible for iron uptake. The inserts of two of these plasmids (pEF23 and pEF33) contained the promoter region of the cir gene, encoding the colicin I receptor responsible for 2,3-dihydroxybenzoate-promoted iron uptake (Hancock et al., 1977 ). Three overlapping Fur-binding sites are located in the cir promoter region (Griggs & Konisky, 1989 ). Two FURTA-positive plasmids (pEF7 and pEF67) contained overlapping, divergently oriented promoters governing the expression of fesentF and fepAentD belonging to the enterochelin gene cluster. A single operator region consisting of two overlapping Fur boxes co-ordinately regulates the expression of the genes (Hunt et al., 1994 ). One plasmid (pEF10) contained another bidirectorial promoter from the enterochelin region that contacts the fepB and entCEBA operons. Three Fur boxes are characterized in the bidirectorial control region: one located in the fepB promoter and a pair of contiguous Fur boxes in the 5' untranslated region of the entC message (Brickman et al., 1990 ). One plasmid (pEF63) contained a Sau3A fragment with the promoter region of fhuF, which is responsible for ferrioxamine B-promoted iron uptake (Patzer & Hantke, 1999 ).

The remaining four plasmids contained promoter regions of genes which are not at present known to be Fur regulated in E. coli. The insert of plasmid pEF20 contained the promoter of the pgm gene encoding phosphoglyceromutase 1. The position of the corresponding Sau3A fragment on the E. coli genome sequence (Blattner et al., 1997 ) is given in Table 3. Two putative overlapping Fur-binding sites matching the consensus sequence in 15 of 19 bases spanned positions -13 to +12 with respect to the pgm transcriptional-start site (Fig. 2b). pEF19 contained the promoter region of the nrdHIEF operon responsible for synthesis of deoxyribonucleotides from the corresponding ribonucleotides (Jordan et al., 1996 ). A sequence matching the consensus site for Fur binding in 11 of 19 bases was found to precede the promoter at position -61 relative to the transcription-start site (Fig. 2c). pEF34 contained two potential promoters located upstream of an ORF, designated yhhY (or o162), defining a gene product of unknown function. Two putative overlapping Fur boxes identical to the consensus sequence in 16 and 13 of 19 bases were present at the periphery of the proposed promoters, spanning positions -68 to -94 with respect to the transcriptional start site of yhhY (Fig. 2a). pEF29 contained a 115 bp chromosomal fragment where no promoter or ORF has previously been identified. Two putative overlapping Fur boxes, each of which matched the consensus sequence in 15 of 19 bases, were present in the Sau3A fragment. The nearest ORF that is preceded by the Fur boxes is the nohA gene belonging to cryptic prophage Qin, which many strains of E. coli K-12 contain (Kotani et al., 1992 ). The Fur-binding region is located approximately 230 bp upstream of the coding sequence of nohA.



View larger version (26K):
[in this window]
[in a new window]
 
Fig. 2. Organization of promoter-regulatory regions in Sau3A chromosomal inserts of FURTA-positive clones. Arrangement of genetic elements in cloned Sau3A fragments and contiguous regions of the E. coli genome is shown. (a), fragment F34 (pEF34); (b), fragment F20 (pEF20); (c), fragment F19 (pEF19); (d), fragment F29 (pEF29); (e), fragment F63 (pEF63). The 5'–3' nucleotide sequences are shown on the DNA strand which is an antisense strand for loci transcribed from left to right and a sense one for loci transcribed in the opposite direction on the E. coli chromosome. The distances between two represented DNA sequences or/and between a regulatory region and the corresponding ORF are given in base pairs and designated with numbers preceded by N. Proposed promoters and putative Fur-protein-binding sites are marked. White arrows denote positions and orientations of ORFs. Figures in parentheses indicate identity (in base pairs) of putative Fur boxes with the Fur box consensus sequence. Fur boxes were identified on the basis of similarity to the consensus sequence for Fur binding. The potential promoters and transcriptional start points for yhhX, ygaC, nohA, orf78 and fhuF, consistent with the positions of the corresponding Fur boxes and statistically most similar to the consensus sequence, were determined on the basis of the algorithm described by Staden (1984) .

 
In summary, sequence analysis of 10 recombinant plasmids chosen at random among the ones selected with FUR-SEL1 permitted us to identify eight distinct Fur-regulated loci, four of which have not previously been shown to be Fur-regulated in E. coli. The promoter regions of fepAfes and cir were found in several plasmids.

Fur-dependent promoter activity of new genes belonging to the Fur regulon of E. coli
To prove that the chromosomal fragments cloned in FUR-SEL1 possess promoter activities regulated by Fur, the inserts of pEF19, pEF20, pEF29 and pEF34 were transferred into the promoter probe vector pBR-{Delta}. This vector is a pBR322 deletion derivative with the promoterless tetracycline resistance gene tet. The orientations of the proposed Fur-regulated promoters in pEF19, pEF20 and pEF29 were determined from sequence data to be opposite to that of the lac promoter of the vector. That permitted us to conclude that transfer of their EcoRI–HindIII fragments, including chromosomal insertions, into pBR-{Delta} would result in transcriptional fusions to tet. The fact that all the resulting constructs (pBR-{Delta}F19, pBR-{Delta}F20 and pBR-{Delta}F29) were selected on tetracycline medium by transforming strain 803-8 confirmed this prediction. The direction of the yhhY promoter in pEF34 coincided with the lac promoter and hence the same strategy could not be employed for creating a yhhY'–'tet fusion. However, the F34 fragment contained a region of the genome upstream of the yhhY promoter where no regulatory sites or ORFs are identified according to the NCBI (National Center for Biotechnology Information) database. Therefore, the pBR-{Delta} derivative of pEF34 was also constructed to examine whether the F34 fragment contains promoter activity oriented oppositely to yhhY in this region. The selection of pBR-{Delta}F34 in strain 803-8 with tetracycline confirmed the existence of the opposing promoter activity. The nearest ORF which is preceded by the promoter activity is the coding sequence for yhhX, which lies 205 bp beyond the F34 fragment on the E. coli genome (Fig. 2a).

In addition, pBR-{Delta} derivatives for known Fur-regulated promoters fesp and fepAp were constructed to be used as positive controls when analysing promoter activity of new Fur-regulated loci. For this purpose, the EcoRI–HindIII fragments of pEF7 and pEF67 were inserted into pBR-{Delta}, giving rise to pBR-{Delta}F7 and pBR-{Delta}F67, respectively. Both plasmids contain the same chromosomal fragment carrying the bidirectional promoter region fesfepA. In pBR-{Delta}F7, the region is oriented in such a way that the transcription of tet emanates from fesp, but in pBR-{Delta}F67, the expression of tet is driven by fepAp.

The expression of the tet gene in pBR-{Delta}F plasmids was compared in Fur+ and Fur- backgrounds, in strains AB1157 and VK-12, respectively. Of the six constructs examined, all six exhibited Fur-responsive promoter activity (Table 4). Under Fur- conditions, plating efficiency on tetracycline medium of the cells containing the plasmids was increased by greater than 104- to 105-fold compared to Fur+ conditions, which confirmed the involvement of Fur in the regulation of the activity of promoters preceding the nrdHIEF, pgm, nohA and yhhX genes.


View this table:
[in this window]
[in a new window]
 
Table 4. Promoter activity of FURTA-positive fragments

 
Analysis of DNA regions selected in FUR-SEL1 for divergent promoter activities
Taking into consideration that divergently oriented promoter regions are an organizational scheme utilized often enough in a broad range of organisms we assumed that some chromosomal fragments cloned in plasmids pEF could also possess such promoter activities.

To examine the chromosomal fragments for possible divergent promoter activities, the Sau3A inserts of pEF19, pEF20, pEF29, pEF34 and pEF7 (as a control) were recloned into the promoter probe vector pUJ10. This vector contains a multiple cloning site flanked by promoterless ß-galactosidase and alkaline phosphatase genes oriented in opposite directions. Therefore, the introduction of a fragment at this site permits analysis of promoter activity of the insertion simultaneously in both directions. The orientations of Sau3A insertions on the pUJ10 derivatives determined by restriction endonuclease mapping are shown in Table 5. By measuring the levels of the ß-galactosidase and alkaline phosphatase expressed from the pUJ10 derivatives in response to various iron concentrations in the growth medium, the directions of promoter activities on the Sau3A inserts were determined and the influence of iron on the regulation of the promoters driving the synthesis was assessed (Table 5).


View this table:
[in this window]
[in a new window]
 
Table 5. Bidirectional promoter activity of FURTA-positive fragments

 
Chromosomal insertions F19 and F34 exhibited bidirectorial iron-responsive promoter activity. Besides the ß-galactosidase expression driven by the promoters of nrdH and yhhX, whose Fur regulation was estimated with the corresponding pBR-{Delta} constructs in previous experiments, pUJ-F19 and pUJ-F34 showed iron-regulated alkaline phosphatase expression. The promoter activity driving phoA expression in pUJ-F19 was regulated negatively by iron: alkaline phosphatase synthesis was consistently lower in iron-rich conditions compared with that at iron limitation. In contrast, the promoter directing phoA synthesis in pUJ-F34 was regulated in a positive fashion by iron: under iron-replete conditions roughly fivefold induction of alkaline phosphatase synthesis was observed while the synthesis fell to near-background levels under low-iron conditions.

The F20 insertion was found not to contain other promoters besides that of the pgm gene: pUJ-F20 carrying the insertion in orientation yielding the pgm'–'lacZ fusion exhibited iron-regulated ß-galactosidase activity (which was consistent with the data from the previous study of the expression of the pBR-{Delta}F20 construct in Fur+ and Fur- background) but the level of the alkaline phosphatase activity of this clone did not significantly exceed that of the vector control.

The F29 chromosomal fragment exhibited low alkaline phosphatase activity, about 10-fold higher than the background level. The phoA synthesis was reproducibly higher under low-iron conditions although the difference never exceeded 60%. The ß-galactosidase synthesis observed in cells with pUJ-F29 did not exceed the background level at both low and high iron which led us to conclude that there is not any promoter activity in this direction.

To confirm that the observed iron-responsive induction of the examined operon fusions was not attributable to a difference in plasmid copy number under iron-limited conditions, the expression of the bla'–'lacZ fusion carried on plasmid pUJ-BLA and governed by the Fur-independent promoter of the gene encoding ß-lactamase was measured under high- and low-iron conditions. The ß-lactamase activities under the two growth conditions were quite similar, which ruled out the involvement of copy number in the modulation of the iron-responsive expression of the fusions carried on pUJ10 derivatives.

To deduce an appropriate coding sequence for the opposing promoter activity observed for the F19 insertion, the DNA sequence of this fragment and the contiguous region of the chromosome were scanned for long ORFs in each of three reading frames in the direction consistent with the promoter activity. Long ORFs other than ygaC (noted in the database) were not found. Accordingly, the ORF lying 145 bp upstream of the F19 fragment on the E. coli chromosome (Fig. 2c) could be transcribed from the promoter situated within the F19 fragment despite the fact that another potential promoter region preceding the ygaC coding sequence and lying beyond the F19 fragment is proposed in the database.

Subsequent transfer of the F19 fragment into pBR-{Delta} in the orientation yielding the ygaC'–'tet fusion and analysis of the tet expression from the pBR-{Delta} derivative in Fur+ and Fur- strains showed that regulation of the ygaC promoter by iron depends on Fur (Table 4; pBR-{Delta}F19-A). pBR-{Delta}F34-A carrying the F34 insertion in the orientation giving rise to the yhhY'–'tet fusion was unstable in a Fur- strain (perhaps because of a deleterious effect of high-level expression of bla from the strong divergently directed yhhX promoter). Therefore, in this case the involvement of Fur in regulation was not confirmed by a more direct fashion.

Finally, the F23 insertion was examined for the presence of promoter activity in the direction opposite to the cir promoter located in this fragment. For this purpose, the corresponding pBR-{Delta} derivative, pBR-{Delta}F23, was constructed and the plating efficiency of strains AB1157 and VK12 transformed by pBR-{Delta}F23 was measured at different concentrations of tetracycline (3·0–12·5 mg l-1). pBR-{Delta}F23 was not found to confer any tetracycline resistance to the Fur+ host or its Fur- derivative: the plating efficiency of the strains carrying pBR-{Delta}F23 did not exceed that of the same cells containing vector pBR-{Delta} (data not shown). The results indicated that the F23 fragment does not contain promoter activity opposed to cir.

Analysis of the F63 chromosomal fragment for additional genetic loci
A detailed study of the F63 chromosomal fragment was performed to ascertain whether the fragment contains regulatory regions other than the fhuF promoter. The basis for such an assumption was the fact that pEF63 provided {alpha} complementation of mutation {Delta}(lacZ)M15 located on the chromosome of FUR-SEL1 (the clones harbouring the plasmid were light blue on X-Gal medium). This observation could be due to the creation of a functional in-frame fusion between lacZ and some coding sequence located on the insertion and containing a functional binding site for ribosomes. However, according to the database, the only coding sequence located in the fragment was the 5' end of the fhuF gene oriented oppositely to lacZ. The analysis of the sequence of the F63 fragment revealed that it contains the 5' end of a 234 bp ORF which had no homology to any entries in the database. The ORF (designated orf78) was oriented in the opposite direction relative to fhuF and its amino-terminal portion (30 aa) created an in-phase fusion with the {alpha} fragment of LacZ.

To determine whether the promoter driving orf78 transcription is contained within the F63 insertion or the transcription of the orf78'–'lacZ fusion is governed by the lacZ promoter of pUC19, the F63 fragment was transferred into the promoter probe vector pUJ10. Among LacZ+ clones the pUJ10 derivatives containing the insertion both in the orientation giving rise to the fhuF'–'lacZ fusion and in the opposite orientation producing orf78'–'lacZ (pUJ-F63) were found. The results pointed to the fact that the F63 insertion contained two divergent promoter activities. Clone pUJ-F63 was chosen to characterize the dependence of the orf78 promoter activity from iron. As shown in Table 5, growth of CC118(pUJ-F63) under low-iron conditions resulted in marked induction of the alkaline phosphatase expression governed by the orf78 promoter compared with growth in the presence of iron. To explore the mechanism by which the iron concentration in the growth medium regulates the expression of orf78 and the involvement of Fur in this regulation, the F63 fragment was inserted in pBR-{Delta} (plasmid pBR-{Delta}F63-A) and the expression of the orf78'–'tet fusion was analysed in Fur+ and Fur- backgrounds. Assay of the tetracycline resistance of the pair of isogenic strains AB157 and VK-12 containing pBR-{Delta}F63-A (Table 4) confirmed that the regulation of orf78 by iron depends on the fur locus.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In this study strain FUR-SEL1 that enables the cloning of Fur-regulated loci is described. Unlike the current methods, including the FURTA technique, for identification of genes belonging to a given regulon which are based on screening, the in vivo procedure allows positive selection of clones. The introduction of a chromosomal leaky fur mutation into FUR-SEL1 allowed cloning of Fur boxes exhibiting weak affinity for the Fur repressor. Indeed, only plasmids pEF10 (entCfepB) and pEF23 or pEF33 (cir) were able to impart considerable chloramphenicol resistance to the Fur+ strain VK-0C. The plasmids carrying the other Fur-regulated loci found in this work, namely: fhuForf78, fepAfes, pgmA, nrdHIEFygaC and yhhYyhhX did not bring about sufficient fhuA promoter derepression for effective growth of Fur+ cells on chloramphenicol medium.

Genes pgm, yhhY (o162), yhhX, ygaC, nrdHIEF, nohA and orf78, whose Fur regulation has previously not been demonstrated in E. coli, were identified during this study. Fur-regulated ORFs exhibiting high similarity with pgm and o162 of E. coli (identity at the level of amino acid sequences was 97% and 85%, respectively) were found previously in S. typhimurium by FURTA screening of a plasmid bank (Tsolis et al., 1995 ). o162 defines a Fur-regulated product of yet unknown function, and no homology to known proteins has been found. The Fur regulation of pgm encoding phosphoglyceromutase, an enzyme catalysing one of the phosphorylation reactions in glycolysis, is one more confirmation of the involvement of Fur in the control of general catabolic pathways. The first evidence for this was provided by the findings that the genes encoding aconitase of E. coli and fumarase of P. aeruginosa contained a Fur box in their promoters (Prodromou et al., 1992 ; Ochsner & Vasil, 1996 ).

The nrdHIEF operon defines a locus which has not previously been described in E. coli or S. typhimurium as Fur regulated. Our data, together with the fact that expression of the purR repressor, a regulatory protein for the pur regulon and for genes involved in pyrimidine synthesis (He et al., 1993 ), is under Fur control (Stojiljkovic et al., 1994 ), give evidence in support of the involvement of Fur in the regulation of the synthesis of ribonucleotides and deoxyribonucleotides.

The analytical study of the sequences of the FURTA-positive chromosomal fragment selected in this work permitted identification of promoter-like sequences and putative Fur boxes that could be responsible for the Fur-regulated promoter activities. The results obtained supported a mechanism for the iron-dependent repression of yhhX, ygaC, pgm and nohA, and orf78 transcription that was similar to that utilized for control of other known Fur-regulated genes: the putative Fur-binding sites overlap the -10 (and -35 in the case of orf78) region of the promoters. However, the positions of the yhhY, nrdH and fhuF promoters are not consistent with the placement of proposed Fur boxes, which might be caused by either inadequacies in the computer algorithms employed for detecting the promoters or a more complex scheme of the regulation of the genes supposing involvement of other regulatory mechanisms. Some assumptions that could remove the noted discrepancy are given below.

Among the iron-regulated loci identified in this work there was a locus, yhhY, which is regulated in a positive fashion. A putative Fur-binding region consisting of two overlapping Fur boxes was located 32 bp upstream of the yhhY promoters and overlapped the opposing yhhX promoter (Fig. 2a). In light of the absence of data suggesting a role for Fur as a positive effector, the activation of the yhhY promoter under conditions when Fur is also active could be explained by participation of another repressor, which in its turn is negatively regulated by Fur, in the regulation of yhhY.

In the F63 fragment, two nearby Fur-binding sites with high similarity (15 and 16 matches) to the Fur box consensus sequence were found. However, neither of them overlapped the promoter region (indicated in Fig. 2e as PfhuF) predicted for fhuF in the database. On the other hand, the results obtained from alkaline phosphatase measurements led to the conclusion that expression of the fhuF'–'phoA fusion is highly sensitive to iron: the iron-mediated induction ratio produced by the fhuF regulatory region reached values attributable to strongly Fur-regulated promoters. In light of the data the possibility of other potential promoters was investigated. A promoter-like sequence (denoted P'fhuF in Fig. 2e) very similar to the consensus promoter sequence and consistent with the position of Fur-binding sites was found. The two putative Fur boxes identified in the F63 fragment encompassed the -10 and -35 elements as well as the transcription-initiation site of P'fhuF. Such an arrangement of the fhuF regulatory region could abolish the discrepancy in the positions of promoter and operator signals and bring them into line with the significant Fur-mediated modulation of fhuF expression.

The Fur-binding regions of all the FURTA-positive fragments identified in this study, except the F19 fragment (nrdHygaC), were represented by two Fur boxes displaced 6 bp in respect of each other. Such arrangement of the operator region is typical for many Fur-regulated genes, including fepAfes (Hunt et al., 1994 ), fur (de Lorenzo et al., 1988 ), sltA (Calderwood & Mekalanos, 1987 ), cir (Griggs & Konisky, 1989 ) and entCfepB (Brickman et al., 1990 ), and indicates that binding of two repressor dimers occurs which contact major groove turns on the DNA helix if Fur conforms to the structural motif observed for other prokaryotic repressors (Pabo & Sauer, 1984 ; Ptashne, 1986 ). Perhaps these overlapping sites increase specificity or help to stabilize binding by increasing the site of the target region.

The fhuForf78 operator region consisted of two Fur boxes separated by 4 bp with good similarity to the consensus iron box (15 and 16 matches) and two Fur boxes (not shown in Fig. 2) with less similarity to the consensus (12 matches) which were displaced 6 bp in respect of the major operators. The placement of the major operators in respect of each other was similar to that for phage {lambda}, where co-operative binding of CI repressor dimers occur (Ptashne, 1986 ). Therefore, the exceptional sensitivity of the fhuF promoter to changes of iron concentration (Stojiljkovic et al., 1994 ) may be due in part to co-operative effects.

The Fur proteins of Gram-negative bacteria exhibit a high degree of sequence similarity. For example, the Fur repressors of E. coli, Vibrio cholerae and Yersinia pestis have 75% identity (Staggs & Perry, 1991 ; Litwin et al., 1992 ). This is the reason why Fur boxes from different Gram-negative bacteria can be detected in E. coli strains suitable for FURTA (Stojiljkovic et al., 1994 ; Tsolis et al., 1995 ). FUR-SEL1 could, therefore, be employed for cloning Fur-regulated loci, among which there are genes responsible for pathogenicity, not only from E. coli but also from other enterobacteria. What is more, the use of cat operon fusions whose coding sequence for cat is transcribed from promoters regulated by other regulatory proteins could be a general strategy to clone the targets for the regulators.


   ACKNOWLEDGEMENTS
 
We thank K. Hantke for strains and for a given chance of sequencing our plasmids with the ALF automated sequencer. Both the authors have made equal contribution to this study. This work was financed by the authors from their wages, at the final stage the work was supported by Russian Fund of Basic Research, grant 98-04-49840, and by programme ‘Russian Universities – Basic Research’ (1998–2000) from Ministry of Education, grant 7-3569.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Blattner, F. R., Plunkett, G., Bloch, C. A. & 14 other authors (1997). The complete genome sequence of Escherichia coli K-12. Science 277, 1453–1474.[Abstract/Free Full Text]

Brickman, E. & Beckwith, J. (1975). Analysis of the regulation of Escherichia coli alkaline phosphatase synthesis using deletions and {phi}80 transducing phages. J Mol Biol 96, 307-316.[Medline]

Brickman, T. J., Ozenberger, B. A. & McIntosh, M. A. (1990). Regulation of divergent transcription from the iron-responsive fepB-entC promoter-operator regions in Escherichia coli. J Mol Biol 212, 669-682.[Medline]

Calderwood, S. B. & Mekalanos, J. J. (1987). Iron regulation of Shiga-like toxin expression in Escherichia coli is mediated by the fur locus. J Bacteriol 169, 4759-4764.[Medline]

Clarke, L. & Carbon, J. (1976). A colony bank containing synthetic ColE1 hybrid plasmids representative of the entire E. coli genome. Cell 9, 91-99.[Medline]

Close, T. J. & Rodriguez, R. L. (1982). Construction and characterization of the chloramphenicol-resistance gene cartridge: a new approach to the transcriptional mapping of extrachromosomal elements. Gene 20, 305-316.[Medline]

Cohen, S. N., Chang, A. C. Y. & Hsu, L. (1973). Nonchromosomal antibiotic resistance in bacteria: transformation of Escherichia coli by R-factor DNA. Proc Natl Acad Sci USA 69, 2110-2114.

Compan, I. & Touati, D. (1993). Interaction of six transcriptional regulators in expression of manganese superoxide dismutase in Escherichia coli K-12. J Bacteriol 175, 1687-1696.[Abstract]

Dower, W. J., Miller, J. F. & Ragsdale, C. W. (1988). High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res 16, 6127-6145.[Abstract]

Griggs, D. W. & Konisky, J. (1989). Mechanism for iron-regulated transcription of the Escherichia coli cir gene: metal-dependent binding of Fur protein to promoters. J Bacteriol 171, 1048-1054.[Medline]

Hancock, E., Hantke, K. & Braun, V. (1977). Iron transport in Escherichia coli K-12; 2,3-dihydroxybenzoate-promoted iron uptake. Arch Microbiol 114, 231-239.[Medline]

Hantke, K. (1983). Identification of iron uptake system specific for coprogen and rhodotorulic acid in Escherichia coli. Mol Gen Genet 191, 301-306.[Medline]

Hantke, K. (1987). Selection procedure for deregulated iron transport mutants (fur) in Escherichia coli K-12: fur not only affectes iron metabolism. Mol Gen Genet 210, 135-139.[Medline]

Hantke, K. & Braun, V. (1998). Control of bacterial iron transport by regulatory proteins. In Metal Ions in Gene Regulation , pp. 11-44. Edited by S. Silver & W. Walden. New York:Chapman & Hall.

He, B., Choi, Y. & Zalkin, H. (1993). Regulation of Escherichia coliglnB, prsA, and speA by the purine repressor. J Bacteriol 175, 3598-3606.[Abstract]

Hunt, M. D., Pettis, G. S. & McIntosh, M. A. (1994). Promoter and operator determinants for Fur-mediated iron regulation in the bidirectorial fepA-fes control region of the Escherichia coli enterobactin gene system. J Bacteriol 176, 3944-3955.[Abstract]

Jordan, A., Aragall, E., Gibert, I. & Barbe, J. (1996). Promoter identification and expression analysis of Salmonella typhimurium and Escherichia coli nrdEF operons encoding one of two class I ribonucleotide reductases present in both bacteria. Mol Microbiol 19, 777-790.[Medline]

Kotani, H., Kawamura, A., Takahashi, A., Nakatsuji, M., Hiraoka, N., Nakajima, K. & Takanami, M. (1992). Site-specific dissection of E. coli chromosome by lambda terminase. Nucleic Acids Res 20, 3357-3360.[Abstract]

Litwin, C. M., Boyko, S. A. & Calderwood, S. B. (1992). Cloning, sequencing, and transcriptional regulation of the Vibrio choleraefur gene. J Bacteriol 174, 1897-1903.[Abstract]

de Lorenzo, V., Wee, S., Herrero, M., Giovannini, F. & Neilands, J. B. (1988). Fur (ferric uptake regulation) protein and CAP (catabolite-activator protein) modulate transcription of fur gene in Escherichia coli. Eur J Biochem 173, 537-546.[Abstract]

de Lorenzo, V., Herrero, M., Jakubzik, U. & Timmis, K. (1990). Mini-Tn5 transposon derivatives for insertion mutagenesis, promoter probing, and chromosomal insertion of cloned DNA in Gram-negative eubacteria. J Bacteriol 172, 6568-6572.[Medline]

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Ochsner, U. & Vasil, M. (1996). Gene repression by the ferric uptake regulator in Pseudomonas aeruginosa: cycle selection of iron-regulated genes. Proc Natl Acad Sci USA 93, 4409-4414.[Abstract/Free Full Text]

Pabo, C. & Sauer, R. (1984). Protein–DNA recognition. Annu Rev Biochem 53, 293-321.[Medline]

Patzer, S. I. & Hantke, K. (1999). SufS is a NifS-like protein, and SufD is necessary for stability of the [2Fe–2S] FhuF protein in Escherichia coli. J Bacteriol 181, 3307-3309.[Abstract/Free Full Text]

Prodromou, C., Artymiuk, P. J. & Guest, J. R. (1992). The aconitase of E. coli. Eur J Biochem 204, 599-609.[Abstract]

Ptashne, M. (1986). A Genetic Switch: Gene Control and Phage {lambda}. Cambridge, MA: Cell.

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Silver, S., Johnseine, P., Whitney, E. & Clark, D. (1972). Manganese-resistant mutant of Escherichia coli: physiological and genetic studies. J Bacteriol 110, 186-195.[Medline]

Staden, R. (1984). Computer methods to locate signals in nucleic acid sequences. Nucleic Acids Res 12, 505-519.[Abstract]

Staggs, T. M. & Perry, R. D. (1991). Identification and cloning of fur regulatory gene in Yersinia pestis. J Bacteriol 173, 417-425.[Medline]

Stojilykovic, I., Baumler, A. J. & Hantke, K. (1994). Fur regulation in Gram-negative bacteria: identification and characterization of new iron-regulated Escherichia coli genes by a Fur titration assay. J Mol Biol 236, 531-545.[Medline]

Tsolis, R. M., Baumler, A. J., Stojilykovic, I. & Heffron, F. (1995). Fur regulation of Salmonella typhimurium: identification of new iron-regulated genes. J Bacteriol 177, 4628-4637.[Abstract]

Vieira, J. & Messing, J. (1982). The pUC plasmids, a M13mp7 derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 9, 259-264.

Yanisch-Perron, C., Vieira, J. & Messing, J. (1985). Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103-109.[Medline]

Received 6 July 2000; revised 31 August 2000; accepted 7 September 2000.