• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jbacterPermissionsJournals.ASM.orgJournalJB ArticleJournal InfoAuthorsReviewers
J Bacteriol. Jan 2011; 193(2): 429–440.
Published online Nov 19, 2010. doi:  10.1128/JB.01341-10
PMCID: PMC3019833

Altered Regulation of the OmpF Porin by Fis in Escherichia coli during an Evolution Experiment and between B and K-12 Strains [down-pointing small open triangle]

Abstract

The phenotypic plasticity of global regulatory networks provides bacteria with rapid acclimation to a wide range of environmental conditions, while genetic changes in those networks provide additional flexibility as bacteria evolve across long time scales. We previously identified mutations in the global regulator-encoding gene fis that enhanced organismal fitness during a long-term evolution experiment with Escherichia coli. To gain insight into the effects of these mutations, we produced two-dimensional protein gels with strains carrying different fis alleles, including a beneficial evolved allele and one with an in-frame deletion. We found that Fis controls the expression of the major porin-encoding gene ompF in the E. coli B-derived ancestral strain used in the evolution experiment, a relationship that has not been described before. We further showed that this regulatory connection evolved over two different time scales, perhaps explaining why it was not observed before. On the longer time scale, we showed that this regulation of ompF by Fis is absent from the more widely studied K-12 strain and thus is specific to the B strain. On a shorter time scale, this regulatory linkage was lost during 20,000 generations of experimental evolution of the B strain. Finally, we mapped the Fis binding sites in the ompF regulatory region, and we present a hypothetical model of ompF expression that includes its other known regulators.

The Fis protein is of crucial importance in bacterial cells, being involved in their phenotypic acclimation to nutritional shifts. Fis was discovered because of its role in stimulating the site-specific DNA inversion reactions mediated by the Hin and Gin DNA recombinases (30, 31), hence its name: factor for inversion stimulation. However, it soon became evident that Fis was involved in many other important functions. First, Fis is involved in the transcriptional regulation of many genes, including the rRNA and various tRNA genes (7, 53, 64), the topoisomerase-encoding genes (69, 82), genes involved in metabolic pathways (23, 58, 72), virulence genes (32, 35, 54), genes involved in quorum sensing (41), and other regulatory genes (26). Microarray studies in Salmonella enterica serovar Typhimurium (32) and Escherichia coli (8) revealed that several hundred genes are directly or indirectly under the control of Fis. The Fis protein has been shown to bind as a dimer (81) to a degenerate consensus sequence (25).

Second, Fis is recognized as a nucleoid-associated protein involved in the global organization of the bacterial chromosome (24, 67, 70, 75, 80). This histone-like function provides a potential mechanism by which Fis achieves its diverse activities. By binding to and bending DNA, Fis facilitates DNA melting and the formation of molecular complexes at specific sites, especially recruitment of RNA polymerase (6), and it also participates in the global control of chromosome supercoiling by defining topological domains (24, 50). Third, Fis has been shown to be involved in the control of the initiation of DNA replication (20, 65).

The amount of Fis in a cell depends on the cell's nutritional state and is subject to complex regulation. The level of Fis peaks during exponential growth, with more than 50,000 molecules per cell, and then drops precipitously in stationary phase to become almost undetectable (1, 2). This signature involves multilevel regulation of Fis expression, including transcriptional control mediated through the stringent response, through Fis itself, and through the negative supercoiling state of DNA (43, 44, 68), and translational control mediated by the ribosome binding protein BipA (56).

Emphasizing its important roles in bacterial cells, we showed previously that Fis and, more generally, DNA supercoiling were targets of selection during an experiment in which E. coli adapted to a simple laboratory environment for tens of thousands of generations (12). In this long-term evolution experiment (LTEE), 12 populations were started from the same ancestral strain, a derivative of E. coli B, and have been propagated by daily 1:100 transfers in fresh minimal media containing limiting glucose (38, 39). Each day, the populations experience a lag phase followed by exponential growth, eventually leading to depletion of the glucose, and finally stationary phase. Upon transfer into fresh medium the next day, the same physiological transitions begin anew. We showed that the negative supercoiling state of the DNA had increased in most of these populations (12). The in-depth analysis of one population, called Ara-1, led us to discover a point mutation in the fis gene that is located within the ribosome-binding site (RBS) and causes a 3-fold decrease in the level of the Fis protein (12). By moving an allele carrying this mutation into the ancestral strain, we showed that it confers both higher fitness and more negative supercoiling. Sequencing fis in all 12 LTEE populations further revealed that this gene had evolved in 5 of them, while 5 others had mutations in an adjacent gene, dusB, that was shown to regulate the expression of Fis (13).

In this work, our aims are to (i) identify target genes of Fis regulation in the ancestral strain and (ii) determine whether their regulation was affected by the evolved fis allele or other genetic changes that occurred during the LTEE. We examined the proteomic profiles of three strains that are isogenic except for their fis alleles. These three strains bear (i) the ancestral fis allele, (ii) the beneficial fis-1 allele, which evolved in the Ara-1 population, and (iii) an in-frame fis deletion, in all cases in the genetic background of the E. coli B derivative (14) that served as the ancestral strain for the LTEE. Among other differences, the proteomic profiles revealed that the expression of the OmpF porin varied according to the fis allele. This finding indicated a regulatory effect that had not previously been reported, raising the additional question of whether it might be specific to the B strain. In E. coli K-12, two major porins have been described, OmpF and OmpC (51). The expression of both is regulated in response to environmental fluctuations, especially changes in osmotic conditions, through the nucleoid-associated protein IHF and the two-component system that includes the sensor kinase EnvZ and the response regulator OmpR (62). Other environmental conditions are also known to influence the expression of these porins; they are regulated during temperature and other stresses by the antisense RNA micF and the global regulator Lrp and during growth by the alternative sigma factor RpoS (for a review, see reference 62). However, a substantial portion of ompC and its upstream regulatory region are deleted in E. coli B strains (49, 71, 77), and therefore OmpF is their sole major porin.

We show here that ompF transcription in E. coli B is directly regulated by Fis, and we map the relevant Fis binding regions. We further demonstrate that changes in the regulation of ompF by Fis occurred during the LTEE and that the regulation we describe is specific to E. coli B; it is not found in E. coli K-12. We also present and discuss a hypothetical model of this differential regulation by Fis.

MATERIALS AND METHODS

Bacterial strains, plasmids, media, and oligonucleotides.

Derivatives of two common strains of E. coli, B and K-12, were used in this study. The B strains have been described previously (12), and they are derived from the LTEE (38, 39). Briefly, 12 populations were all founded from the same ancestral strain, REL606, which we call “Anc” in this paper. The genome of this ancestral strain has been fully sequenced (29), and its historical derivation (14) and various differences from K-12 (77) have been described. The LTEE populations have been propagated by daily serial transfers in Davis minimal medium for many thousands of generations. A mutation in the fis gene, which is beneficial in the environment of the LTEE, arose and was discovered in the population called Ara-1 (12). This evolved mutation affects the RBS of fis but not the sequence of the encoded Fis protein. The mutation was introduced by homologous recombination into the ancestral chromosome, leading to a pair of strains, Anc and Ancfis-1, that are isogenic except for their fis alleles. An in-frame deletion mutant of fis, designated AncΔfis, was also constructed in the ancestral chromosome. An evolved clone, called Evol20K, was isolated from population Ara-1 after 20,000 generations, and it is also used in this study. The E. coli K-12 derivative we used is CF7968 (MG1655 rph+ ΔlacI-lacZ). The same in-frame deletion of fis was introduced into this strain to generate CFΔfis.

The suicide plasmid pKO3 (42) was used during the construction of the in-frame fis deletion mutants. The plasmids pRS551 and pRS550 (73) were used to construct transcriptional fusions. The plasmid pET41 (Novagen) was used to overexpress the Fis protein.

Strains were grown either in rich LB medium (66) or in Davis minimal medium (39) supplemented with 250 μg/ml glucose (DM250) or 1 mg/ml glucose (DM1000). Infections with phage λ to construct single-copy transcriptional fusions were performed after growing bacteria in TBMM medium (10 g/liter tryptone, 5 g/liter NaCl, 0.2% maltose, 10 mM MgSO4, 1 μg/ml thiamine). Kanamycin (50 μg/ml) was added when needed. Agar (Difco) was added at 12 g/liter to LB medium for plating cells. Selection for sucrose-resistant clones during allelic exchange experiments was performed by spreading cells on sucrose plates in which 5% sucrose but no NaCl was added to the LB agar medium. All experiments and procedures were performed at 37°C unless otherwise noted.

Oligonucleotides ODS451 (5′ GCAGGGACGATCACTGCCAG 3′) and ODS453 (5′ AGTCAAGCAATCTATTTGCA 3′) were used to clone the entire promoter region of ompF, including the transcriptional start site. The same primers were used to generate DNA fragments for gel shift and DNase I footprint experiments. In the latter case, primers were also designed for the middle of the ompF promoter region to generate shorter fragments: ODS452 (5′ TGGTGTCTTTATGTGTCTGC 3′), ODS458 (5′ CTAAACGGAAATTTTGTTTCGT 3′), and ODS459 (5′ CAGAAACAAAATTTCCGTTTAG 3′).

Construction of an in-frame deletion of fis.

An in-frame deletion allele of fis was recombined into the Anc and CF7968 chromosomes using the suicide plasmid pKO3, which has a temperature-sensitive replication origin (42), as described previously (12). Briefly, two DNA fragments, one with ~500 bp of the upstream region of the fis gene and one with ~500 bp of the downstream region, were cloned into pKO3 to generate an in-frame deletion allele that has only three codons, including the start and stop codons. The resulting plasmid was electrotransformed into the Anc and CF7968 strains. Constructs with the deletion allele integrated into the chromosome were selected by plating transformed cells on chloramphenicol-LB agar and incubating them at high temperature. Chloramphenicol-resistant cells were subsequently plated on sucrose-containing LB agar to select for plasmid-free segregants; the plasmid carries sacB, which renders cells sensitive to killing by sucrose. Sucrose-resistant and chloramphenicol-sensitive plasmid-free clones were then screened for the presence of the appropriately integrated deletion allele by PCR.

Two-dimensional polyacrylamide gel electrophoresis (PAGE).

Cells were grown in DM1000, which contains 40 times the glucose used in the LTEE, in order to provide denser cultures. Cell pellets were harvested during exponential phase (optical density at 600 nm [OD600] = 0.1) by centrifugation at 5,000 × g, washed once with 50 mM Tris, pH 8, resuspended in 200 μl of 50 mM Tris, 0.2 M dithiothreitol (DTT), 0.3% (wt/vol) SDS, and 1 mM EDTA, and boiled for 5 min. Twenty microliters of 50 mM Tris, pH 7.5-50 mM MgCl2 containing 1 mg/ml DNase I and 0.25 mg/ml RNase A (Roche) was added to the cell extracts, and the resulting suspensions were incubated for 10 min at 4°C. Finally, proteins were solubilized by adding 800 μl of buffer containing 10 M urea, 4% (vol/vol) NP-40, 0.1 M DTT, and 2.2% (vol/vol) Ampholine with a pH range of 3 to 10 (Amersham Pharmacia). Protein samples were stored at −20°C.

Two-dimensional PAGE was performed using a Multiphor II system (Amersham Pharmacia) for isoelectric focusing, and the Protean II xi cell (Bio-Rad, Hercules, CA) was used for SDS-PAGE. Dry strips (laboratory made, 18 cm, pH 4 to 8) were hydrated in the protein samples (500 μg) prepared in 2 M thiourea, 5 M urea, 1.6% (wt/vol) 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate, 100 μM DTT, and 0.8% (vol/vol) Ampholine. Isoelectric focusing was run until pH equilibrium was reached (37). The strips were equilibrated for 10 min in Tris, pH 6.8, containing 2.5% (wt/vol) SDS, 30% (vol/vol) glycerol, 6 M urea, and 50 mM DTT. This step was repeated in the same buffer, with DTT replaced by 100 mM iodoacetamide. For the second-dimension electrophoresis, SDS-PAGE was performed as described elsewhere (34) using 12% acrylamide gels, with the following modifications: cathode and anode migration buffers consisted of 50 mM Tris-0.1% (wt/vol) SDS with 200 mM taurine and 380 mM glycine, respectively. Proteins were stained with Coomassie brilliant blue R-250.

Quantitative analysis of the protein gels was performed using Melanie II software (Genebio, Geneva, Switzerland). Determination of the global protein profiles of the three strains (Anc, Ancfis-1, and AncΔfis) was performed in triplicate from three independent cultures of each strain. Each three-way set of comparisons was performed in parallel on the same day and with the same media and solutions to produce gels with similar background and signal levels for quantitative analysis. This approach provided excellent reproducibility by making gel parameters maximally consistent for each set of comparisons. We viewed as biologically meaningful only those protein spots with levels that differed among strains in the same way for all three replicates.

Protein spots were excised, treated, and sequenced by matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) or nano-liquid chromatography-tandem mass spectrometry (nanoLC-MS/MS) as previously described (59). Consecutive automatic searches against the Swissprot Trembl database were performed for each sample using version 1.9 of Mascot software.

Porin extraction.

Bacteria were grown for about 4 h in DM250 to achieve exponential-phase growth and sufficient cell density (OD600 ≈ 0.2). The stationary-phase density for these cultures typically corresponded to an OD600 of ≈0.6. Cell pellets were washed in 30 mM Tris buffer (pH 7.3), resuspended in the same buffer, and then lysed in a buffer containing 30 mM Tris, pH 7.3, 20% sucrose, 10 mM EDTA, and 10 mg/ml lysozyme. Samples were disrupted by sonication for 20 s (5-s pulses) and centrifuged at 50,000 × g for 1 h at 4°C. Pellets were resuspended in 30 mM Tris buffer, pH 7.3, plus 20% sucrose. Protein concentration was measured using the Bradford protein assay kit (Bio-Rad), with bovine serum albumin (BSA) as a standard. Equal amounts of protein (30 μg) were analyzed after boiling on a 12% SDS-acrylamide gel containing 1.2 M urea. Proteins were stained with Coomassie brilliant blue R-250.

Construction of ompF-lacZ transcriptional fusions.

The ompF promoter region (positions −422 to +148 relative to the transcription initiation site) was amplified by PCR using primers ODS451 and ODS453 and the genomic DNA of either Anc or CF7968. Fragments were then cloned into pCRII-TOPO (Invitrogen), sequenced, cut with BamHI and EcoRV, and transferred into pRS551 (73), which was previously cut first with EcoRI, blunted, and cut again with BamHI, thereby generating transcriptional fusions with lacZ as a reporter. These fusions were then crossed in vivo into the λRS45 bacteriophage derivative, followed by lysogenization of the relevant strains at the chromosomal λ att site, as described previously (73). A PCR assay was performed to confirm that only a single prophage was integrated (61).

The β-galactosidase activities were assayed by using o-nitrophenyl-β-d-galactopyranoside (ONPG) as a substrate, and specific activities are calculated as μkat/mg protein (48). All reported activity values are the averages of three independent assays.

Fis purification.

The fis gene from the Anc strain was cloned into pET41, and the protein was overexpressed in BL21(DE3) cells (Novagen). Fis was purified as previously described (57) with the following modifications. Cell pellets were washed and resuspended in extraction buffer (70 mM Tris, pH 7.4, 1 M NaCl, 10% glycerol). Cells were disrupted using a French press (ThermoSpectronic, Rochester, NY) at 1,000 lb/in2 pressure. After a short sonication step, cell lysates were centrifuged for 15 min at 5,000 × g and 4°C to remove cell fragments and unbroken cells. DNA was removed by the addition of 0.35% polyethylenimine (PEI; Sigma), and the precipitate was eliminated through high-speed centrifugation at 100,000 × g for 1 h. The supernatant was then incubated with 0.2 volume of phosphocellulose (P11; Whatman) to remove the excess PEI, dialyzed into FB buffer (20 mM Tris, pH 7.5, 1 mM EDTA, 300 mM NaCl, 10% glycerol), and passed through an SP Sepharose High Trap column (Amersham). Fis was eluted in a gradient of NaCl, and the protein concentrations of the fractions were determined using the Bradford kit (Bio-Rad) and BSA as a standard.

EMSA.

The same DNA fragments generated for the construction of the ompF-lacZ transcriptional fusions were used for the electrophoretic mobility shift assays (EMSA). The ODS453 primer was labeled with [γ-32P]dATP (6,000 Ci/mmol; ICN) using T4 polynucleotide kinase (Euromedex, France) prior to the PCR. The PCR product (0.6 nM) was incubated for 20 min at 37°C with increasing amounts of Fis (from 0 to 40 nM) in a binding buffer (20 mM Tris, pH 7.5, 1 mM EDTA, pH 8, 80 mM NaCl, 4% glycerol, 0.5 mM DTT, 0.5 mg/ml BSA, and 10 μg/ml sonicated herring sperm DNA). The binding reactions were resolved on 5% polyacrylamide gels in 0.5× Tris-borate-EDTA buffer at 100 V. Gels were dried, exposed to Fujifilm imaging plates, and analyzed using a FLA8000 scanner (Fujifilm). The percentage of shifted DNA was quantified using the Image Gauge software (Fujifilm). The value of Kd (dissociation constant) was calculated as the Fis concentration that corresponds to 50% protection of the higher-affinity binding site. Two control experiments were performed: (i) a supershift reaction by adding a specific Fis antibody and (ii) the addition to the binding reaction mixture of either a specific or a nonspecific unlabeled DNA (from 5- to 1,000-fold excess). The specific DNA is a double-stranded 36-bp oligonucleotide containing high-affinity Fis binding site I, present upstream of the fis promoter (5′ CAGAGGATTGGTCAAAGTTTGGCCTTTCATCTCGTG 3′); the nonspecific DNA is a double-stranded 34-bp oligonucleotide in which the Fis site has been substantially mutated (5′ CAGAGGATTAAAGCTCGGAGCTATTCATCTCGTG 3′).

DNase I footprints.

DNase I footprint experiments were performed using each of four different PCR products with the following primer pairs: ODS453-ODS452, ODS459-ODS452, ODS451-ODS453, and ODS458-ODS453. For each reaction, the first primer listed was labeled with [γ-32P]dATP (6,000 Ci/mmol; ICN) using T4 polynucleotide kinase (Euromedex, France) prior to the PCR (except for ODS453-ODS452, for which both primers were alternately labeled). Each end-labeled DNA fragment (0.6 nM) was incubated in the absence or presence of increasing concentrations of Fis (0 to 40 nM) in the same binding conditions as for EMSA (see above). After 20 min, samples were treated with 0.75 ng DNase I (Worthington) for 35 s at room temperature. The reactions were stopped with 50 μl of phenol and 180 μl of stop solution (12.5 mM EDTA, pH 8, 0.4 M sodium acetate [AcONa], and 25 μg/ml herring sperm DNA). After precipitation, DNA was resuspended in 5 μl formamide, heated at 90°C for 3 min, and loaded onto an 8% polyacrylamide gel (1× TBE, 1.5 M urea). A G+A Maxam-Gilbert sequencing reaction mixture (47) was loaded on the same gel to localize precisely the Fis binding sites in the ompF regulatory region. The gel was then dried and analyzed the same way as for EMSA.

RESULTS

Proteomic comparison of strains with fis mutations.

We performed two-dimensional proteomic electrophoresis on three strains that are isogenic except for their fis alleles: the ancestor of the LTEE with the ancestral fis allele (Anc), a strain with the same ancestral background but carrying the evolved fis allele that arose in population Ara-1 (Ancfis-1), and the ancestral background strain carrying the in-frame fis deletion allele (AncΔfis). Protein extracts were prepared from three sets of independent exponential-phase cultures (OD600 = 0.1 in DM1000; see Materials and Methods) of all three clones. Such protein gels typically allow the visualization of about 300 to 400 protein spots, representing roughly 10% of the total coding capacity of the E. coli genome (assuming each spot represents a different protein), and we observed similar numbers in our gels. The relative abundances for nine protein spots changed consistently (across all three replicate sets of gels) in one or both of the strains with mutated fis alleles compared to those in the strain with the ancestral allele. They were identified by MS (see Materials and Methods) and included four membrane proteins: Pal is a peptidoglycan-associated lipoprotein that stabilizes the outer membrane, although its precise role is unknown (22); SurA is a member of the peptidyl-prolyl isomerase family that may function as a chaperone in the outer membrane protein targeting pathway (74); and OmpA and OmpF are both porins in E. coli (51). The five other proteins with altered levels are H-NS, a histone-like protein and transcriptional regulator of gene expression (15), EF-Tu, a translation elongation factor (76), MetE, the cobalamin-independent methionine synthase (27), Aat, a leucyl/phenylalanyl-tRNA protein transferase (78), and the 50S ribosomal subunit protein L7/L12 (5). The expression levels of two of these proteins, H-NS and EF-Tu, were already known to be regulated by Fis (18, 52, 86).

In E. coli K-12, the levels of the major porins OmpF and OmpC are regulated in response to a variety of environmental signals, with osmolarity being the most extensively studied factor (62). Many global regulators contribute to this regulation, but, to the best of our knowledge, Fis has never been reported to control the expression of OmpF or OmpC. It is important to reiterate that E. coli B encodes only OmpF and not OmpC (71, 77). We therefore sought to confirm that ompF is indeed regulated by Fis, at least in derivatives of E. coli B. Moreover, to understand why the regulatory connection between Fis and ompF had not been previously detected, we hypothesized that it may have changed during evolution. We therefore studied this regulatory connection over two evolutionary time scales, a relatively short one corresponding to the LTEE and a longer one reflecting divergence of two E. coli strains, B and K-12, independently sampled from nature (14). During the first 20,000 generations of the LTEE, a clone from the Ara-1 population accumulated a total of 45 mutations relative to its ancestor across its entire genome (3). In contrast, the B and K-12 genomes differ by over 26,000 point mutations in addition to numerous insertions, deletions, and genomic islands (77). It has been estimated that B and K-12 have diverged for ~70 million generations, or ~360,000 years if one assumes 200 cell generations per year in nature (17).

Regulation of ompF by Fis in E. coli B.

We first measured the amounts of OmpF in the outer membranes of the isogenic strains Anc, Ancfis-1, and AncΔfis, bearing the ancestral, evolved, and deletion fis alleles, respectively. For these same three strains, we previously showed that Ancfis-1 has about one-third the level of Fis as Anc, while Fis is completely absent in AncΔfis (12). Here we observe a clear decrease in the amount of OmpF associated with the lower levels of Fis in Ancfis-1 and AncΔfis (Fig. (Fig.1A).1A). In contrast, the level of OmpA increased in Ancfis-1 and AncΔfis, while the amounts of another typical (unidentified) protein were similar across all three strains (Fig. (Fig.1A1A).

FIG. 1.
Effect of Fis level on OmpF expression in the E. coli B genetic background. (A) Amounts of OmpF in three strains that are isogenic except for their fis alleles. Porins were extracted from exponential-phase cultures after 4 h in DM250 medium for Anc, Anc ...

To investigate whether the positive effect of Fis on OmpF expression occurs at the level of transcription, either directly or indirectly, we constructed single-copy lacZ transcriptional fusions of the ompF promoter cloned from the Anc strain (PompFAnc), which were introduced into the chromosomal λ att sites of strains Anc, Ancfis-1, and AncΔfis. Their β-galactosidase activities were measured during exponential phase, around the transition from exponential to stationary phase, and in stationary phase (Fig. (Fig.1B).1B). The expression of ompF in Anc was nearly constant from exponential to stationary phase. In contrast, the ompF transcription levels were significantly lower in both Ancfis-1 and AncΔfis at all time points, with the greatest differences during exponential phase, when the expression of Fis is maximal. Therefore, Fis positively controls the transcription of ompF in the E. coli B strain used as the ancestor of the LTEE. To determine whether the transcriptional activation of ompF by Fis is direct, we examined whether Fis binds to the ompF promoter region and we then mapped the Fis binding sites, as described in the next section. It should also be emphasized, however, that Fis is not the only regulator of ompF because some transcription occurs in the fis mutants and during stationary phase, when the Fis level is very low. These effects may well be related to the previously reported regulation of ompF by other global regulators, including LRP, IHF, and the alternative sigma factor RpoS (62).

Characterization of the Fis binding properties on the ompF promoter region.

We performed gel mobility shift assays to measure the ability of Fis protein to bind to the PompFAnc region (Fig. (Fig.2A).2A). A shift in electrophoretic mobility was observed even at the lowest Fis concentration (0.2 nM) tested. At least 5 retarded bands were seen at higher Fis concentrations, indicative of several Fis binding sites (84). The dissociation constant (Kd) for the Fis site showing the highest affinity was calculated as ~3 nM, which is comparable to values reported for other specific Fis binding sites (19, 45), with no apparent cooperativity. Binding specificity was checked by using (i) 5- to 1,000-fold excesses of an unlabeled oligonucleotide containing a consensus Fis binding site, (ii) the same unlabeled DNA fragment except with the Fis site mutated, and (iii) an antibody against Fis in the binding reactions. The first assay reduced the observed Fis binding, while the second assay had no detectable effect (Fig. (Fig.2B).2B). The third assay resulted in a supershift (data not shown). These results demonstrate that the Fis binding to the ompF promoter region is specific.

FIG. 2.
Binding of Fis to the ompF promoter region. (A) Electrophoretic mobility gel shift assays with an end-labeled DNA fragment spanning the PompFAnc region from positions −422 to +148 relative to the transcription initiation site. The fragment ...

To map the Fis binding regions in PompFAnc, each of four DNA fragments spanning the promoter region was used in DNase I footprint experiments with increasing concentrations of purified Fis (Fig. (Fig.3).3). Fis protected five regions, designated I to V, upstream of the ompF transcription start site, consistent with the number of shifted bands (at least five) observed in the gel mobility shift assays. The binding of Fis to the promoter also induced enhanced DNase I cleavage (hypersensitivity) in all protected regions, possibly reflecting Fis-induced DNA distortion, as reported for other Fis targets (21).

FIG. 3.
Mapping Fis-protected regions in PompFAnc by DNase I footprint experiments. The entire promoter region was examined using four DNA fragments obtained by PCR amplification with the following primer pairs (shown on Fig. Fig.4):4): ODS453-ODS452, ...

Figure Figure44 shows the sequence of the transcriptional regulatory region of the ompF gene in the E. coli B-derived Anc strain. Fis-protected regions I to V are shown, along with the known OmpR binding sites F1 to F4 (28). Region I is centered 85 bp upstream of the transcriptional start site, and it covers the region from −45 to −118, which could bind several Fis dimers. Only one sequence close to the consensus Fis binding site could be recognized, although the locations of hypersensitive sites leave open the possibility that two or even three Fis dimers may bind in this region (Fig. (Fig.4).4). Region I also covers two of the OmpR binding sites, F1 and F2, and it overlaps a third, F3, which suggests an antagonistic interaction between Fis and OmpR in the ompF promoter region (see Discussion). The other four Fis-protected regions all contain one consensus binding sequence and are located between positions −130 and −363. They are separated by 69 bp (sites I and II), 82 bp (sites II and III), 60 bp (sites III and IV), and 54 bp (sites IV and V).

FIG. 4.
Sequence of the ompF regulatory region in the E. coli B-derived Anc strain. The ompF promoter region is shown from positions −422 to +148 relative to the main transcriptional start site (represented by a bent arrow and an uppercase boldface ...

To the best of our knowledge, no regulatory connection between Fis and major porin expression has previously been described in E. coli, despite many studies on both Fis and these porins. To investigate why this connection was not discovered before, we analyzed changes in the direct regulatory connection between Fis and OmpF across the two time scales corresponding to the LTEE and to the divergence of E. coli strains B and K-12. These analyses are presented in the following two sections, and they show evolutionary changes in this regulatory connection.

Changes in ompF regulation by Fis during 20,000 generations of evolution.

The evolved fis allele from population Ara-1 has been shown to be beneficial under the conditions of the LTEE (12). Here, we showed that this allele also reduces the level of OmpF. This effect of the fis allele is surprising, however, given the fact that the E. coli B-derived Anc of the LTEE cannot produce the other major porin, OmpC, unlike E. coli K-12, which expresses OmpC at higher levels when OmpF is downregulated. To investigate this conundrum, we measured ompF transcription in an evolved clone, Evol20K, that was isolated from population Ara-1 after 20,000 generations and that has the evolved fis allele. The same PompFAnc::lacZ transcriptional fusion was introduced into the same chromosomal site of Evol20K as was done previously for the three isogenic strains with different fis alleles. Recall that ompF transcription declined when the evolved fis allele was introduced into the ancestral genetic background (strain Ancfis-1). However, we saw no difference in ompF transcription between the Evol20K and Anc strains, with the exception of a slight increase in transcription in Evol20K during stationary phase (Fig. (Fig.1B).1B). This was confirmed at the protein level by porin gels (data not shown). Therefore, one or more other mutations that occurred during those 20,000 generations must interact with the evolved fis mutation in a way that either eliminates regulation of ompF transcription by Fis or otherwise restores expression to the ancestral level.

Elsewhere, we recently reported the genome sequence of this Evol20K clone compared to that of the ancestral strain, Anc (3). A total of 45 mutations were discovered, including 29 point mutations and 16 deletions, insertions, and other rearrangements. One of these mutations occurred in the promoter region of ompF, and at first glance it might seem to be a strong candidate. However, the transcription of the PompFAnc::lacZ fusion uses the ancestral ompF promoter, and the levels of that transcription did not differ between the Evol20K and Anc genetic backgrounds. Therefore, the evolved ompF allele is unlikely to be directly responsible for the altered regulation of ompF by Fis that occurred during the experimental evolution. For example, if the evolved ompF promoter region binds the Fis protein more tightly, then there would be less (not more) Fis available to increase expression from the ancestral ompF promoter in the Evol20K background. Similarly, if the evolved ompF region binds some transcriptional inhibitor less tightly, there would be more (not less) inhibitor to reduce expression from the ancestral ompF promoter in the evolved background. The mutation (or mutations) that modifies or counteracts the effect of the evolved fis allele on ompF transcription must therefore be found among the other 44 differences. Given that large number of differences, however, it would be difficult to identify the responsible mutation.

Differential regulation of ompF between E. coli B and K-12 derivatives.

To assess whether the regulation of ompF by Fis was specific to E. coli B, we extended these analyses to a K-12-derived strain, CF7968. A single-copy chromosomal lacZ transcriptional fusion was constructed using the CF7968 ompF promoter (PompFCF::lacZ), and it was introduced, as before, into two strains: CF7968 and a derivative with the in-frame fis deletion (CFΔfis). The β-galactosidase activities were measured at four time points, including early-, mid-, and late-exponential-phase growth and during the transition into stationary phase (Fig. (Fig.5A).5A). We included more time points during exponential growth than for Fig. Fig.11 because normal fis expression is stronger during that phase, and hence we could better ascertain the effect of Fis on ompF transcription.

FIG. 5.
Differential transcriptional regulation of ompF by Fis as a function of genetic context. (A) Effect of Fis on ompF transcription in K-12-derived strain CF7968. The transcriptional fusion PompFCF::lacZ was introduced into the chromosomes of CF7968 and ...

We observed two main differences between the K-12-derived transcriptional fusions and those made in the B-derived Anc strain. First, there is no discernible regulation of PompFCF by Fis in K-12. Second, the transcription of ompF in K-12 increases severalfold during the transition into stationary phase. The differences in ompF regulation between B and K-12 could, in principle, be caused by differences in Fis, in the promoter region of ompF, or in other factors involved in ompF regulation. There are no differences at all between the CF7968 and Anc strains in the fis coding sequence. However, there are three differences in their ompF promoter regions (Fig. (Fig.4),4), including two in the OmpR binding sites F2 (a 1-bp indel) and F4 (a point mutation) and one between the −35 and −10 boxes of the promoter (a point mutation).

To gain further insights into this differential regulation, we ran binding experiments to determine if Fis can bind to the CF7968 ompF promoter region. The experiments were performed exactly as before, except now using DNA corresponding to the PompFCF region. The Fis binding properties were indistinguishable from those observed for the Anc strain (data not shown). Therefore, differential binding of Fis to the B and K-12 ompF promoters is not responsible for the difference in their ompF regulation.

We also performed ompF transcription analyses in which we varied both the genetic background and the genetic source of the ompF promoter region. To that end, we moved each single-copy lacZ fusion, PompFAnc::lacZ and PompFCF::lacZ, into the same chromosomal site of strains Anc and CF7968. The β-galactosidase activities were measured at five time points during their growth (Fig. (Fig.5B).5B). The resulting data confirm the constitutive expression of PompFAnc::lacZ in Anc and the induction of PompFCF::lacZ transcription during stationary phase in CF7968. With PompFCF in the Anc genetic background, we observed the same pattern of inducible transcription as in its native context, which suggests that the three mutations distinguishing PompFAnc and PompFCF might explain the difference in regulation during exponential phase. However, we saw the same inducible pattern in the reciprocal situation when PompFAnc::lacZ was expressed in the CF7968 background. This outcome implies that genetic context, as well as differences in the ompF promoter itself, may contribute to the differential regulation of ompF in the B-derived Anc and K-12-derived CF7968 strains.

We therefore also compared the sequences of envZ and ompR, which encode the two-component sensor kinase EnvZ and response regulator OmpR, respectively, and which are important regulators of ompF in K-12 (62). More generally, this two-component system controls the transcription of more than 100 genes (55). There are two nonsynonymous mutations in ompR (leading to K6N and A130T amino acid changes from K-12 to B) that were previously reported in another B-derived strain (49), and one nonsynonymous mutation in envZ (leading to P41L from K-12 to B). However, the A130T mutation in OmpR is not present in all B-derived strains; it is absent in strain BL21(DE3) and may have been introduced by mutagenesis of an earlier progenitor to the Anc strain of the LTEE (14, 36, 77).

DISCUSSION

The Fis protein is involved with many important functions in E. coli cells such as responding to nutritional shifts, regulating the transcription of numerous genes with diverse roles, controlling the initiation of DNA replication, and acting as a nucleoid-associated protein that helps organize the entire chromosome. Our previous work on experimental populations of E. coli showed that Fis was an important target of selection, resulting in repeated changes in the expression level or sequence of that protein that were evidently advantageous under the study conditions (12, 13). Given the diverse roles of Fis and these evolutionary changes, we sought to determine the proteomic effects of two mutations in the fis gene. One of them affects the RBS of fis and causes reduced Fis expression without changing the protein sequence, and this mutation was beneficial under the conditions of the long-term evolution experiment (LTEE). The other is an artificially constructed in-frame deletion that does not produce any Fis protein at all.

Effects of Fis expression on OmpF level in E. coli B.

We moved each of these mutations separately into the ancestral strain of the LTEE, replacing the ancestral fis allele in the process. Two-dimensional protein gels revealed consistent changes in the levels of nine proteins in one or both of the fis mutants relative to their isogenic counterpart. One of these proteins was OmpF, which was present at lower levels in both the fis deletion mutant and the strain with the evolved allele. To the best of our knowledge, no previous study has shown any regulatory effect of Fis on OmpF, even though both of the associated genes have been well studied, especially in K-12 strains. E. coli K-12 has two major porins, OmpC and OmpF. However, E. coli B strains, including the progenitor of the evolution experiment, do not produce any OmpC (71, 77), and thus their only major porin is OmpF. Both of these facts made the decline in OmpF level caused by the evolved fis allele especially interesting, and so we sought to understand the regulatory effect of Fis on OmpF, including how this regulation may have changed during the evolution experiment and how it might differ between E. coli B and K-12 strains.

Fis binding to the ompF promoter increases transcription in E. coli B but not in K-12.

By analyzing transcriptional fusions with the B-derived promoter PompFAnc, we observed that ompF is apparently constitutively expressed but that the level of its expression was depressed in the absence of Fis protein. Electrophoretic mobility shift assays showed that this activation by Fis was associated with its binding to PompFAnc, while DNase I footprint experiments indicated the presence of five Fis-protected regions in the promoter. The activation of ompF by Fis was lost, however, during the evolution that led to the mutant allele with the reduced level of Fis. The Fis-mediated activation was also not detected in E. coli CF7968, a K-12-derived strain. The expression of ompF during exponential growth was also lower in K-12 than in B, while that expression was induced during the transition into stationary phase in K-12 but not in B.

Sites in the ompF promoter region that differ between B and K-12 contribute to this differential regulation, although other loci also may be involved in this difference. When the transcriptional fusion with the K-12-derived promoter was moved into the B genetic background, we saw the same pattern of inducible transcription as seen in its native K-12 context, indicating the effect of differences in the ompF promoter itself. However, putting the B-derived promoter fusion in the K-12 background did not restore the pattern of regulation observed in the B strain, indicating that other genetic differences between B and K-12 must also contribute to their differential regulation of ompF by Fis.

Turning from our results to speculation, we suggest that envZ and ompR, which encode the sensor kinase EnvZ and response regulator OmpR, respectively, are reasonable candidate genes that might contribute to these differences between B and K-12. This two-component system is known to be a key regulator of ompF expression in K-12. Moreover, there are two differences in OmpR between B and K-12 and one in EnvZ, and all three affect conserved amino acid residues. It was previously suggested that the differences in OmpR contribute to the differential expression of ompF in B and K-12 (49). We suggest that all three differences in these proteins may reduce the ability of OmpR to bind to the ompF promoter region and activate transcription in the B strain. The B-specific activation of ompF by Fis would then compensate for the deficiency in OmpR binding. In the next section, we present, first, the current model of ompF regulation by OmpR and EnvZ and, second, the possible effects of the differences between K-12 and B in the ompF regulatory region, ompR, and envZ. We then integrate these ideas into a model for the transcriptional regulation of ompF that we observed specifically in the B strain. Although this model is hypothetical at present, it is based on and consistent with our own data and previously published observations. We hope this model will stimulate future experiments that may confirm or reject its components.

Models of the transcriptional regulation of ompF.

The current model for transcriptional regulation of ompF and ompC by OmpR comes from studies of osmotic stress performed in E. coli K-12 (87). Figure Figure6A6A shows this model for ompF only, because E. coli B does not express OmpC. In K-12, the OmpF porin is preferentially expressed at low osmolarity and the OmpC porin at high osmolarity. At low osmolarity, EnvZ maintains a low level of phosphorylated OmpR (OmpR-P), a state that favors the activation of ompF but not ompC. Under these conditions, OmpR-P binds to two (F1 and F2) of its four binding sites (F1 to F4) in the promoter region of ompF. At high osmolarity, EnvZ increases the level of OmpR-P, which can then occupy all four binding sites, leading to the repression of ompF; these same conditions activate ompC transcription when that gene is present. Yoshida et al. (87) present a model of hierarchical binding of OmpR-P to the F1 to F4 sites, with binding affinities ranked as follows: F1 > F2 ≈ F3 > F4 (4). Each site is further subdivided into two OmpR-P binding subsites, a and b, with each downstream b subsite having higher affinity for OmpR-P than the upstream a subsite (Fig. (Fig.6A),6A), giving rise to an expanded hierarchy: F1b > F1a ≈ F2b > F2a ≈ F3b > F3a > F4.

FIG. 6.
Model of ompF regulation in E. coli K-12 and B strains. (A) The “galloping” model for K-12 (87). (B) Our model for B-derived strains. Binding sites for OmpR-P and Fis are shown as open boxes labeled F1 to F4 and I to V, respectively. The ...

The mutational differences between B and K-12 strains, both in the ompF regulatory region and elsewhere, must alter this regulatory system. A 1-bp deletion in the ompF promoter region of B is near the junction of the F1b and F2a subsites. This deletion brings the Fis-binding site closer to the consensus sequence by putting the following C at the 15th position, a very conserved position for Fis binding sites (21). We hypothesize that this difference may lead both to higher affinity for Fis and reduced affinity for OmpR-P, which could no longer cover both F1 and F2 sites. These effects would therefore activate ompF transcription in B relative to K-12.

In the ompR gene, two B-specific alleles affect codons 130 and 6, while in envZ a B-specific allele affects codon 41. However, some B strains have the same codon 130 as does K-12, and the difference at that site may have been introduced by mutagenesis of a progenitor to the ancestor of the LTEE (77). The OmpR A130 residue in K-12 is a highly conserved position located in the linker domain between the N-terminal phosphorylation and dimerization domain and the C-terminal DNA-binding domain. The K6 residue in strain K-12 is also well conserved and may be involved in protein dimerization. The P41 EnvZ residue is in the periplasmic domain (flanking the membrane-spanning segment TM1) of the protein, which may be involved in fine-tuning osmotic signal transduction through dimerization of EnvZ, and it is adjacent to a predicted well-conserved alpha helix (33, 85). The P41 residue is also well conserved, and mutations affecting this residue (P41L, P41S) in K-12-derived strains yield a protein that appears to be defective in its phosphatase activity, leading to a higher concentration of OmpR-P, thereby causing an OmpF-deficient, OmpC-constitutive phenotype (79). E. coli B strains do not produce any OmpC, and so it would be important to restore OmpF production given the likely harmful effect of this EnvZ mutation on ompF transcription. The ompR mutation at codon 6, which is evidently common to all B strains, might be such a compensatory mutation. Indeed, a compensatory mutation has been reported in ompR for another envZ mutation that caused an OmpF-deficient, OmpC-constitutive phenotype (46).

Based on these observations, we can propose a new model for ompF transcriptional activation specific to the E. coli B-derived strain Anc that we used in this study (Fig. (Fig.6B).6B). In K-12-derived strains, ompF is activated by OmpR in a so-called “galloping,” or discontinuous, manner that allows fine control over the relative levels of ompF and ompC transcription (87). Thus, ompF transcription is activated only once F1 and F2 are covered by OmpR-P (Fig. (Fig.6A).6A). In E. coli B, we hypothesize that OmpR-P can still bind to the F1 site but that it has a decreased ability to bind cooperatively across adjacent sites owing to the B-specific mutations in the ompF promoter region and in ompR (perhaps including the Anc-specific second mutation in ompR). This reduced OmpR-P binding ability, together with the presence of Fis-binding sites, would allow Fis to bind and activate ompF transcription (Fig. (Fig.6B).6B). The binding of Fis to PompFAnc would also prevent repression of ompF in E. coli B strains (49). This B-specific model can be evaluated by further experiments.

Evolutionary change in ompF regulation.

The regulation of ompF by fis was altered during the evolution of population Ara-1. In particular, the evolved fis allele, with its lower expression, produced reduced levels of ompF transcription when it was moved into the ancestral background. However, the evolved genotype itself had the same high level of ompF expression as the ancestor. This outcome suggests some epistatic interaction between the evolved fis allele and one or more other mutations that arose during the population's history.

The evolved genome has now been sequenced, and it accumulated 45 changes, including 29 point mutations and 16 deletions, insertions, and inversions, relative to the ancestral sequence (3). Several of the mutations affect high-level regulatory networks, including the stringent response and DNA supercoiling, that may have widespread pleiotropic and epistatic effects (60), but the identity of the mutation(s) that interacts with the evolved fis allele is unknown at present.

However, three previous studies found evidence of widespread epistatic interactions in the LTEE (11, 16, 63). Two of them showed that random transposon insertion mutations often had different effects on fitness in the ancestral strain and in an evolved genetic background. The third study introduced an artificial deletion mutation of crp, which encodes a key regulator of catabolite repression, into the ancestor and clones sampled from two of the LTEE populations after 20,000 generations, including the Ara-1 population with the evolved fis allele that we have used here. That study revealed widespread epistatic interactions affecting global transcription profiles, with the crp deletion affecting the expression of hundreds of genes differently in the ancestral and evolved backgrounds (11).

Here we have found epistatic interactions between mutations that occurred during the evolution experiment (as opposed to artificial mutations that were generated to test for epistasis). One interesting candidate for an interaction with the fis mutation is a mutation upstream of ompF that evolved in this same Ara-1 population. Recall, however, that the levels of ompF transcription did not differ between the evolved and ancestral backgrounds based on expression from transcriptional fusions that placed the ancestral ompF promoter into both genetic contexts and on direct measurements of OmpF levels. Hence, that evolved ompF allele is not directly responsible for the altered regulation of ompF by Fis that occurred over the LTEE. However, two other lines of evidence suggest that this ompF mutation is important and likely to be embedded in a network of epistatic interactions. First, two other LTEE populations also evolved mutations in ompF in 20,000 generations (3). This evolutionary repeatability, or parallelism, is a strong indicator of adaptive evolution (3, 83). In contrast, no cases of parallelism were seen when several dozen genes were chosen at random and sequenced in 20,000-generation isolates from all 12 populations (40). Second, the ompF allele that evolved in population Ara-1 is the only one (among those representing nine mutations at different loci so far tested) that is detrimental based on competition assays between isogenic strains that differ only at a single locus. The Ara-1 ompF allele reduced fitness by almost 10% when it was put into the ancestral genetic background. It seems unlikely that such a deleterious mutation could have hitchhiked to fixation; a more plausible explanation is that it is beneficial in the context of the evolved clone in which it first appeared, i.e., in association with some other mutations that arose in this experiment. However, the fis mutation is probably not involved in the fitness change associated with the ompF allele because our experiments showed that the ompF and fis mutations do not interact epistatically (at least not at the level of ompF regulation). This point also suggests that the fitness advantage of the Ara-1 fis allele is independent of its effect on ompF and may be associated with other Fis-related phenotypes as previously suggested (12). For now, an understanding of the evolved ompF alleles must await further research.

Evolutionary rewiring of regulatory networks.

Regulatory networks are generally understood to allow a high level of phenotypic plasticity, such that bacteria and other organisms can quickly acclimate to changing environments. What has been less clear is the degree to which these networks can themselves readily evolve as the frequency and magnitude of fluctuations change over longer timescales. Genes that encode key regulatory proteins or that modify key regulatory molecules are embedded in complex networks. Therefore, mutations in these genes are likely to have widespread pleiotropic effects that might constrain their evolution. On the other hand, and for the same reason of their connectivity to the network, mutations in these key genes might affect the interactions of these genes with many other genes, providing numerous opportunities for refinement, including compensatory changes that ameliorate maladaptive side effects of earlier mutations.

Our analyses of the regulation of OmpF support the latter view, as do many other findings from the LTEE. With respect to OmpF, we documented evolutionary changes over two timescales. Over the many thousands of years that strains K-12 and B have diverged (17), the predecessors to B lost one of their two major porins, OmpC, whose expression had been finely counterbalanced with that of OmpF. As a consequence, B evolved toward constitutive expression of OmpF and, moreover, Fis was co-opted as an activator of ompF transcription that may also block repression by OmpR-P. It is also interesting that comparative genomics analyses indicate that both ompF and ompC have been under selection in uropathogenic E. coli (9). Over the short timescale of a decade of evolution in the laboratory, the role of Fis in regulating OmpF expression was further altered in a B-derived population. In particular, Fis itself evolved to have much lower expression, presumably because of selection on one of its many other cellular roles. That reduction in Fis was associated with further changes in the regulation of ompF such that it was transcribed at a high level even with the lower availability of Fis. More generally, the LTEE populations have undergone many changes in regulatory networks, including their global-level stringent response (10, 59), catabolite repression (11), and DNA supercoiling networks (12, 13) as well as lower-level regulons that affect substrate utilization (59) and porin expression (this study). On balance, then, the results of this long-term study of evolution in action indicate that connectivity among the components of these regulatory networks does not constrain them but, instead, provides flexibility to respond over evolutionary as well as physiological timescales.

Acknowledgments

We thank Michael Cashel for strain CF7968.

This work was supported by the Centre National de la Recherche Scientifique (CNRS), Université Joseph Fourier, and grants from the Agence Nationale de la Recherche (ANR, Program Génomique, Grant ANR-08-GENM-023-001) to D.S. and the U.S. National Science Foundation to R.E.L. E.C. thanks the CNRS for a BDI fellowship.

Footnotes

[down-pointing small open triangle]Published ahead of print on 19 November 2010.

REFERENCES

1. Ali Azam, T., A. Iwata, A. Nishimura, S. Ueda, and A. Ishihama. 1999. Growth phase-dependent variation in protein composition of the Escherichia coli nucleoid. J. Bacteriol. 181:6361-6370. [PMC free article] [PubMed]
2. Ball, C. A., R. Osuna, K. C. Ferguson, and R. C. Johnson. 1992. Dramatic changes in Fis levels upon nutrient upshift in Escherichia coli. J. Bacteriol. 174:8043-8056. [PMC free article] [PubMed]
3. Barrick, J. E., et al. 2009. Genome evolution and adaptation in a long-term experiment with Escherichia coli. Nature 461:1243-1247. [PubMed]
4. Bergstrom, L. C., L. Qin, S. L. Harlocker, L. A. Egger, and M. Inouye. 1998. Hierarchical and cooperative binding of OmpR to a fusion construct containing the ompC and ompF upstream regulatory sequences of Escherichia coli. Genes Cells 3:777-788. [PubMed]
5. Bocharov, E. V., et al. 2004. From structure and dynamics of protein L7/L12 to molecular switching in ribosome. J. Biol. Chem. 279:17697-17706. [PubMed]
6. Bokal, A. J., W. Ross, and R. L. Gourse. 1995. The transcriptional activator protein FIS: DNA interactions and cooperative interactions with RNA polymerase at the Escherichia coli rrnB P1 promoter. J. Mol. Biol. 245:197-207. [PubMed]
7. Bosch, L., L. Nilsson, E. Vijgenboom, and H. Verbeek. 1990. FIS-dependent trans-activation of tRNA and rRNA operons of Escherichia coli. Biochim. Biophys. Acta 1050:293-301. [PubMed]
8. Bradley, M. D., M. B. Beach, A. P. Jason de Koning, T. S. Pratt, and R. Osuna. 2007. Effects of Fis on Escherichia coli gene expression during different growth stages. Microbiology 153:2922-2940. [PubMed]
9. Chen, S. L., et al. 2006. Identification of genes subject to positive selection in uropathogenic strains of Escherichia coli: a comparative genomics approach. Proc. Natl. Acad. Sci. U. S. A. 103:5977-5982. [PMC free article] [PubMed]
10. Cooper, T. F., D. E. Rozen, and R. E. Lenski. 2003. Parallel changes in gene expression after 20,000 generations of evolution in Escherichia coli. Proc. Natl. Acad. Sci. U. S. A. 100:1072-1077. [PMC free article] [PubMed]
11. Cooper, T. F., S. K. Remold, R. E. Lenski, and D. Schneider. 2008. Expression profiles reveal parallel evolution of epistatic interactions involving the CRP regulon in Escherichia coli. PLoS Genet. 4:e35. [PMC free article] [PubMed]
12. Crozat, E., N. Philippe, R. E. Lenski, J. Geiselmann, and D. Schneider. 2005. Long-term experimental evolution in Escherichia coli. XII. DNA topology as a key target of selection. Genetics 169:523-532. [PMC free article] [PubMed]
13. Crozat, E., et al. 2010. Parallel genetic and phenotypic evolution of DNA superhelicity in experimental populations of Escherichia coli. Mol. Biol. Evol. 27:2113-2128. [PubMed]
14. Daegelen, P., F. W. Studier, R. E. Lenski, S. Cure, and J. F. Kim. 2009. Tracing ancestors and relatives of Escherichia coli B, and the derivation of B strains REL606 and BL21(DE3). J. Mol. Biol. 394:634-643. [PubMed]
15. Dorman, C. J. 2004. H-NS: a universal regulator for a dynamic genome. Nat. Rev. Microbiol. 2:391-400. [PubMed]
16. Elena, S. F., and R. E. Lenski. 2001. Epistasis between new mutations and genetic background and a test of genetic canalization. Evolution 55:1746-1752. [PubMed]
17. Elena, S. F., T. S. Whittam, C. L. Winkworth, M. A. Riley, and R. E. Lenski. 2005. Genomic divergence of Escherichia coli strains: evidence for horizontal transfer and variation in mutation rates. Int. Microbiol. 8:271-278. [PubMed]
18. Falconi, M., A. Brandi, A. La Teana, C. O. Gualerzi, and C. L. Pon. 1996. Antagonistic involvement of FIS and H-NS proteins in the transcriptional control of hns expression. Mol. Microbiol. 19:965-975. [PubMed]
19. Feldman-Cohen, L. S., et al. 2006. Common and variable contributions of Fis residues to high-affinity binding at different DNA sequences. J. Bacteriol. 188:2081-2095. [PMC free article] [PubMed]
20. Filutowicz, M., W. Ross, J. Wild, and R. L. Gourse. 1992. Involvement of Fis protein in replication of the Escherichia coli chromosome. J. Bacteriol. 174:398-407. [PMC free article] [PubMed]
21. Finkel, S. E., and R. C. Johnson. 1992. The Fis protein: it's not just for DNA inversion anymore. Mol. Microbiol. 6:3257-3265. (Erratum, 7:1023, 1993.) [PubMed]
22. Gerding, M. A., Y. Ogata, N. D. Pecora, H. Niki, and P. A. J. De Boer. 2007. The trans-envelope Tol-Pal complex is part of the cell division machinery and required for proper outer-membrane invagination during cell constriction in E. coli. Mol. Microbiol. 63:1008-1025. [PubMed]
23. Gonzalez-Gil, G., P. Bringmann, and R. Kahmann. 1996. FIS is a regulator of metabolism in Escherichia coli. Mol. Microbiol. 22:21-29. [PubMed]
24. Hardy, C. D., and N. R. Cozzarelli. 2005. A genetic selection for supercoiling mutants of Escherichia coli reveals proteins implicated in chromosome structure. Mol. Microbiol. 57:1636-1652. [PubMed]
25. Hengen, P. N., S. L. Bartram, L. E. Stewart, and T. D. Schneider. 1997. Information analysis of Fis binding sites. Nucleic Acids Res. 25:4994-5002. [PMC free article] [PubMed]
26. Hirsch, M., and T. Elliott. 2005. Fis regulates transcriptional induction of RpoS in Salmonella enterica. J. Bacteriol. 187:1568-1580. [PMC free article] [PubMed]
27. Hondorp, E. R., and R. G. Matthews. 2004. Oxidative stress inactivates cobalamine-independent methionine synthase (MetE) in Escherichia coli. PLoS Biol. 2:e336. [PMC free article] [PubMed]
28. Huang, K. J., and M. M. Igo. 1996. Identification of the bases in the ompF regulatory region, which interact with the transcription factor OmpR. J. Mol. Biol. 262:615-628. [PubMed]
29. Jeong, H., et al. 2009. Genome sequences of Escherichia coli B strains REL606 and BL21(DE3). J. Mol. Biol. 394:644-652. [PubMed]
30. Johnson, R. C., M. F. Bruist, and M. I. Simon. 1986. Host protein requirements for in vitro site-specific DNA inversion. Cell 46:531-539. [PubMed]
31. Kahmann, R., F. Rudt, C. Koch, and G. Mertens. 1985. G inversion in bacteriophage Mu DNA is stimulated by a site within the invertase gene and a host factor. Cell 41:771-780. [PubMed]
32. Kelly, A., et al. 2004. A global role for Fis in the transcriptional control of metabolism and type III secretion in Salmonella enterica serovar Typhimurium. Microbiology 150:2037-2053. [PubMed]
33. Khorchid, A., M. Inouye, and M. Ikura. 2005. Structural characterization of Escherichia coli sensor histidine kinase EnvZ: the periplasmic C-terminal core domain is critical for homodimerization. Biochem. J. 385:255-264. [PMC free article] [PubMed]
34. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. [PubMed]
35. Lautier, T., and W. Nasser. 2007. The DNA nucleoid-associated protein Fis coordinates the expression of the main virulence genes in the phytopathogenic bacterium Erwinia chrysanthemi. Mol. Microbiol. 66:1474-1490. [PubMed]
36. Lederberg, S. 1966. Genetics of host-controlled restriction and modification of deoxyribonucleic acid in Escherichia coli. J. Bacteriol. 91:1029-1036. [PMC free article] [PubMed]
37. Lelong, C., and T. Rabilloud. 2003. Prokaryotic proteomics, p. 157-171. In M. Blot (ed.), Methods and tools in biosciences and medicine: prokaryotic genomics. Birkhäuser Verlag, Basel, Switzerland.
38. Lenski, R. E. 2004. Phenotypic and genomic evolution during a 20,000-generation experiment with the bacterium Escherichia coli. Plant Breed. Rev. 24:225-265.
39. Lenski, R. E., M. R. Rose, S. C. Simpson, and S. C. Tadler. 1991. Long-term experimental evolution in Escherichia coli. I. Adaptation and divergence during 2,000 generations. Am. Nat. 138:1315-1341.
40. Lenski, R. E., C. L. Winkworth, and M. A. Riley. 2003. Rates of DNA sequence evolution in experimental populations of Escherichia coli during 20,000 generations. J. Mol. Evol. 56:498-508. [PubMed]
41. Lenz, D. H., and B. L. Bassler. 2007. The small nucleoid protein Fis is involved in Vibrio cholerae quorum sensing. Mol. Microbiol. 63:859-871. [PubMed]
42. Link, A. J., D. Phillips, and G. M. Church. 1997. Methods for generating precise deletions and insertions in the genome of wild-type Escherichia coli: application to open reading frame characterization. J. Bacteriol. 179:6228-6237. [PMC free article] [PubMed]
43. Mallik, P., et al. 2004. Growth phase-dependent regulation and stringent control of fis are conserved processes in enteric bacteria and involve a single promoter (fisP) in Escherichia coli. J. Bacteriol. 186:122-135. [PMC free article] [PubMed]
44. Mallik, P., B. J. Paul, S. T. Rutherford, R. L. Gourse, and R. Osuna. 2006. DksA is required for growth phase-dependent regulation, growth phase-dependent control, and stringent control of fis expression in Escherichia coli. J. Bacteriol. 188:5775-5782. [PMC free article] [PubMed]
45. Martin, R. G., and J. L. Rosner. 1997. Fis, an accessorial factor for transcriptional activation of the mar (multiple antibiotic resistance) promoter of Escherichia coli in the presence of the activator MarA, SoxS, or Rob. J. Bacteriol. 179:7410-7419. [PMC free article] [PubMed]
46. Matsuyama, S.-I., T. Mizuno, and S. Mizushima. 1986. Interaction between two regulatory proteins in osmoregulatory expression of ompF and ompC in Escherichia coli: a novel ompR mutation suppresses pleiotropic defects caused by an envZ mutation. J. Bacteriol. 168:1309-1314. [PMC free article] [PubMed]
47. Maxam, A. M., and W. Gilbert. 1977. A new method for sequencing DNA. Proc. Natl. Acad. Sci. U. S. A. 74:560-564. [PMC free article] [PubMed]
48. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
49. Mizuno, T., A. Shinkai, K. Matsui, and S. Mizushima. 1990. Osmoregulatory expression of porin genes in Escherichia coli: a comparative study on strains B and K-12. FEMS Microbiol. Lett. 68:289-294. [PubMed]
50. Muskhelishvili, G., and A. Travers. 2003. Transcription factor as a topological homeostat. Front. Biosci. 8:279-285. [PubMed]
51. Nikaido, H. 1996. Outer membrane, p. 29-47. In F. C. Neidhardt et al. (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology, ASM Press, Washington, DC.
52. Nilsson, L., A. Vannet, E. Vijgenboom, and L. Bosch. 1990. The role of FIS in trans activation of stable RNA operons of E. coli. EMBO J. 9:727-734. [PMC free article] [PubMed]
53. Nilsson, L., and V. Emilsson. 1994. Factor for inversion stimulation-dependent growth rate regulation of individual tRNA species in Escherichia coli. J. Biol. Chem. 269:9460-9465. [PubMed]
54. O Cróinín, T., R. K. Carroll, A. Kelly, and C. J. Dorman. 2006. Roles for DNA supercoiling and the Fis protein in modulating expression of virulence genes during intracellular growth of Salmonella enterica serovar Typhimurium. Mol. Microbiol. 62:869-882. [PubMed]
55. Oshima, T., et al. 2002. Transcriptome analysis of all two-component regulatory system mutants of Escherichia coli K-12. Mol. Microbiol. 46:281-291. [PubMed]
56. Owens, R. M., et al. 2004. A dedicated translation factor controls the synthesis of the global regulator Fis. EMBO J. 23:3375-3385. [PMC free article] [PubMed]
57. Pan, C. Q., et al. 1996. Variable structures of Fis-DNA complexes determined by flanking DNA-protein contacts. J. Mol. Biol. 264:675-695. [PubMed]
58. Pease, A. J., B. R. Roa, W. Luo, and M. E. Winkler. 2002. Positive growth rate-dependent regulation of the pdxA, ksgA, and pdxB genes of Escherichia coli K-12. J. Bacteriol. 184:1359-1369. [PMC free article] [PubMed]
59. Pelosi, L., et al. 2006. Parallel changes in global protein profiles during long-term experimental evolution in Escherichia coli. Genetics 173:1851-1869. [PMC free article] [PubMed]
60. Philippe, N., E. Crozat, R. E. Lenski, and D. Schneider. 2007. Evolution of global regulatory networks during a long-term experiment with Escherichia coli. BioEssays 29:846-860. [PubMed]
61. Powell, B. S., D. L. Court, Y. Nakamura, M. P. Rivas, and C. L. Turnbough, Jr. 1994. Rapid confirmation of single copy lambda prophage integration by PCR. Nucleic Acids Res. 22:5765-5766. [PMC free article] [PubMed]
62. Pratt, L. A., W. Hsing, K. E. Gibson, and T. J. Silhavy. 1996. From acids to osmZ: multiple factors influence synthesis of the OmpF and OmpC porins in Escherichia coli. Mol. Microbiol. 20:911-917. [PubMed]
63. Remold, S. K., and R. E. Lenski. 2004. Pervasive joint influence of epistasis and plasticity on mutational effects in Escherichia coli. Nat. Genet. 36:423-426. [PubMed]
64. Ross, W., J. F. Thompson, J. T. Newlands, and R. L. Gourse. 1990. Escherichia coli Fis protein activates ribosomal RNA transcription in vitro and in vivo. EMBO J. 9:3733-3742. [PMC free article] [PubMed]
65. Ryan, V. T., J. E. Grimwade, J. E. Camara, E. Crooke, and A. C. Leonard. 2004. Escherichia coli prereplication complex assembly is regulated by dynamic interplay among Fis, IHF and DnaA. Mol. Microbiol. 51:1347-1359. [PubMed]
66. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
67. Schneider, R., A. Travers, and G. Muskhelishvili. 1997. FIS modulates growth phase-dependent topological transitions of DNA in Escherichia coli. Mol. Microbiol. 26:519-530. [PubMed]
68. Schneider, R., A. Travers, and G. Muskhelishvili. 2000. The expression of the Escherichia coli fis gene is strongly dependent on the superhelical density of DNA. Mol. Microbiol. 38:167-175. [PubMed]
69. Schneider, R., A. Travers, T. Kutateladze, and G. Muskhelishvili. 1999. A DNA architectural protein couples cellular physiology and DNA topology in Escherichia coli. Mol. Microbiol. 34:953-964. [PubMed]
70. Schneider, R., et al. 2001. An architectural role of the Escherichia coli chromatin protein FIS in organising DNA. Nucleic Acids Res. 29:5107-5114. [PMC free article] [PubMed]
71. Schneider, D., et al. 2002. Genomic comparisons among Escherichia coli strains B, K-12, and O157:H7 using IS elements as molecular markers. BMC Microbiol. 2:18. [PMC free article] [PubMed]
72. Shin, D., N. Cho, S. Heu, and S. Ryu. 2003. Selective regulation of ptsG expression by Fis. Formation of either activating or repressing nucleoprotein complex in response to glucose. J. Biol. Chem. 278:14776-14781. [PubMed]
73. Simons, R. W., F. Houman, and N. Kleckner. 1987. Improved single and multicopy lac-based cloning vectors for protein and operon fusions. Gene 53:85-96. [PubMed]
74. Sklar, J. G., T. Wu, D. Kahne, and T. J. Silhavy. 2007. Defining the roles of the periplasmic chaperones SurA, Skp, and DegP in Escherichia coli. Genes Dev. 21:2473-2484. [PMC free article] [PubMed]
75. Skoko, D., J. Yan, R. C. Johnson, and J. F. Marko. 2005. Low-force DNA condensation and discontinuous high-force decondensation reveal a loop-stabilizing function of the protein Fis. Phys. Rev. Lett. 95:208101. [PubMed]
76. Stark, H., et al. 1997. Visualization of elongation factor Tu on the Escherichia coli ribosome. Nature 389:403-406. [PubMed]
77. Studier, F. W., P. Daegelen, R. E. Lenski, S. Maslov, and J. F. Kim. 2009. Understanding the differences between genome sequences of Escherichia coli B strains REL606 and BL21(DE3) and comparison of the E. coli B and K-12 genomes. J. Mol. Biol. 394:653-680. [PubMed]
78. Suto, K., et al. 2006. Crystal structures of leucyl/phenylalanyl-tRNA-protein transferase and its complex with an aminoacyl-tRNA analog. EMBO J. 25:5942-5950. [PMC free article] [PubMed]
79. Tokishita, S., and T. Mizuno. 1994. Transmembrane signal transduction by the Escherichia coli osmotic sensor, EnvZ: intermolecular complementation of transmembrane signalling. Mol. Microbiol. 13:435-444. [PubMed]
80. Travers, A., R. Schneider, and G. Muskhelishvili. 2001. DNA supercoiling and transcription in Escherichia coli: the FIS connection. Biochimie 83:213-217. [PubMed]
81. Wagner, R. 2000. Transcription regulation in prokaryotes. Oxford University Press, Oxford, United Kingdom.
82. Weinstein-Fischer, D., M. Elgrably-Weiss, and S. Altuvia. 2000. Escherichia coli response to hydrogen peroxide: a role for DNA supercoiling, topoisomerase I and Fis. Mol. Microbiol. 35:1413-1420. [PubMed]
83. Woods, R., D. Schneider, C. L. Winkworth, M. A. Riley, and R. E. Lenski. 2006. Tests of parallel molecular evolution in a long-term experiment with Escherichia coli. Proc. Natl. Acad. Sci. U. S. A. 103:9107-9112. [PMC free article] [PubMed]
84. Xu, J., and R. C. Johnson. 1995. Fis activates the RpoS-dependent stationary-phase expression of proP in Escherichia coli. J. Bacteriol. 177:5222-5231. [PMC free article] [PubMed]
85. Yaku, H., and T. Mizuno. 1997. The membrane-located osmosensory kinase, EnvZ, that contains a leucine zipper-like motif functions as a dimer in Escherichia coli. FEBS Lett. 417:409-413. [PubMed]
86. Yoon, H., S. Lim, S. Heu, S. Choi, and S. Ryu. 2003. Proteome analysis of Salmonella enterica serovar typhimurium fis mutant. FEMS Microbiol. Lett. 226:391-396. [PubMed]
87. Yoshida, T., L. Qin, L. A. Egger, and M. Inouye. 2006. Transcription regulation of ompF and ompC by a single transcription factor, OmpR. J. Biol. Chem. 281:17114-17123. [PubMed]

Articles from Journal of Bacteriology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...