Logo of jbacterPermissionsJournals.ASM.orgJournalJB ArticleJournal InfoAuthorsReviewers
J Bacteriol. 2011 Jan; 193(1): 22–29.
Published online 2010 Oct 29. doi:  10.1128/JB.01143-10
PMCID: PMC3019953

Regulation of Horizontal Gene Transfer in Bacillus subtilis by Activation of a Conserved Site-Specific Protease

Abstract

The mobile genetic element ICEBs1 is an integrative and conjugative element (a conjugative transposon) found in Bacillus subtilis. The RecA-dependent SOS response and the RapI-PhrI cell sensory system activate ICEBs1 gene expression by stimulating cleavage of ImmR, the ICEBs1 immunity repressor, by the protease ImmA. We found that increasing the amount of wild-type ImmA in vivo caused partial derepression of ICEBs1 gene expression. However, during RapI-mediated derepression of ICEBs1 gene expression, ImmA levels did not detectably increase, indicating that RapI likely activates the protease ImmA by increasing its specific activity. We also isolated and characterized mutations in immA (immAh) that cause partial derepression of ICEBs1 gene expression in the absence of inducing signals. We obtained two types of immAh mutations: one type caused increased amounts of the mutant proteins in vivo but no detectable effect on specific activity in vitro; the other type had no detectable effect on the amount of the mutant protein in vivo but caused increased specific activity of the protein (as measured in vitro). Together, these findings indicate that derepression of ICEBs1 gene expression is likely caused by an increase in the specific activity of ImmA. Homologs of ImmA and ImmR are found in many mobile genetic elements, so the mechanisms that regulate ImmA-mediated cleavage of ImmR may be widely conserved.

Integrative and conjugative elements (ICEs), also known as conjugative transposons, are mobile genetic elements found in a wide range of bacteria (reviewed in references 5, 6, 9, 13, 15, and 18). These elements reside in the host chromosome and contribute to genome plasticity. They can facilitate the acquisition of new traits, including antibiotic resistance, symbiosis, and virulence. ICEs can be excised from the host chromosome and transfer to other cells by conjugation. Once inside the recipient, the ICE typically integrates into the host chromosome and is stably maintained and propagated by host replication and cell division. ICEs generally encode proteins that function in regulation, integration, excision, and transfer.

ICEBs1 is an approximately 20-kb element inserted in the 3′ end of a tRNA gene in Bacillus subtilis (2, 7). Genes at the left end of ICEBs1 (Fig. (Fig.1A)1A) are part of a regulatory module that resembles those found in many bacteriophages (2, 7). This module includes immR and immA, encoding the element's immunity repressor and antirepressor, respectively (1, 4). ImmR represses transcription of genes required for excision and transfer and both activates and represses its own expression (1).

FIG. 1.
Regulation of ICEBs1. (A) The four genes and two promoters at the left end of ICEBs1 are shown. immR, immA, and int encode the immunity repressor, the antirepressor, and the recombinase (needed for both integration and excision), respectively, and are ...

ICEBs1 gene expression is derepressed in vivo during the RecA-dependent SOS response or when the ICEBs1-encoded cell-cell signaling regulator RapI is present and active (Fig. (Fig.1B)1B) (2). In both cases, derepression requires the antirepressor ImmA (4). ImmA is a site-specific protease that cleaves ImmR, thereby causing derepression of ICEBs1 gene expression (4). It is not known how RapI or RecA stimulate ImmA to cleave ImmR, but previous results indicated that the mechanism of activation was not through transcriptional control of immA (4).

We found that increasing the amount or the specific activity of ImmA can cause derepression of ICEBs1, even without activation by RecA or RapI. We isolated and characterized mutations in immA that cause derepression of ICEBs1 gene expression in the absence of exogenous inducing signals. We also analyzed the effects of artificially elevating the amount of ImmA in the cell. Our results indicate that there are at least two ways in which ImmA-mediated cleavage of ImmR can be activated: (i) by increasing the activity of ImmA or (ii) by increasing the cellular concentration of ImmA. We also found that ImmA levels did not significantly change during activation by RapI, indicating that RapI-mediated induction likely results from an increase in the activity of ImmA.

MATERIALS AND METHODS

Strains and alleles.

B. subtilis strains used in this study are listed in Table Table1.1. Standard techniques were used for cloning and strain construction (11, 16). The ICEBs10 strain and the cgeD::{(PimmR-immR) kan}, thrC::{(Pxis-lacZΩ343) mls}, and the amyE::{(Pspank-immAΩ218) spc} alleles were described previously (1, 2, 4). Note that mls represents the same gene as the previously used “erm” designation. Strains with recA260 (cat-mls) (8) were kept in the dark whenever possible.

TABLE 1.
B. subtilis strainsa

Pxyl and xylR from pDR160 (14) were cloned into pMMB752 (3) to generate pBOSE508, a tet-containing vector, for introduction of a xylose-inducible gene at lacA. A sequence extending from 33 bp upstream of the rapI start codon to 2 bp downstream of the stop codon was cloned into pBOSE508 to generate pBOSE525, which was integrated at lacA by homologous recombination, selecting for tetracycline resistance.

The sequence from 24 bp upstream of immA to 2 bp downstream of its stop codon was cloned downstream from Pspank into pDR110 (from D. Rudner), a vector that contains spc, lacI, and Pspank and that allows for homologous recombination into amyE, to generate pBOSE540. Mutations in Pspank-immA were made by Quikchange (Stratagene) site-directed mutagenesis of pBOSE540. For higher expression of immA, Pspank in pBOSE540 was converted to Pspank(hy) by Quikchange mutagenesis, generating pBOSE1069. The immA alleles were introduced into wild-type cells by a double crossover at amyE, selecting for spectinomycin resistance.

Escherichia coli strains derived from BL21-AI (Invitrogen) were used to purify ImmR and ImmA. N-terminally His-tagged ImmR was purified from BOSE798, and untagged ImmA was purified from BOSE848 (4). immA mutations were introduced by Quikchange mutagenesis of pBOSE831 (4), and the resulting plasmids were introduced into E. coli BL21-AI to produce BOSE843 (R85G), BOSE844 (V92E), BOSE845 (N93D), BOSE846 (I165*), BOSE1133 (I160*), and BOSE1134 (G169*).

Media and growth conditions.

B. subtilis cells were grown at 37°C with aeration in defined S7 minimal salts medium (17), except that MOPS (morpholinepropanesulfonic acid) buffer was used at 50 mM instead of 100 mM. The medium also contained 1% arabinose, 0.1% glutamate, 40 μg/ml tryptophan and phenylalanine, and 120 μg/ml threonine. Xylose (1%) was used to induce expression of Pxyl-rapI. When appropriate, antibiotics (at the indicated concentrations) were used as follows: chloramphenicol (5 μg/ml), kanamycin (5 μg/ml), tetracycline (10 μg/ml), and spectinomycin (100 μg/ml); erythromycin (0.5 μg/ml) and lincomycin (12.5 μg/ml) were used together to select for macrolide-lincosamide-streptogramin B (MLS) resistance. Mitomycin-C (Roche) was used at a final concentration of 1 μg/ml. X-Gal (5-bromo-4-chloro-3-indolyl α-d-galactopyranoside) in LB agar plates was used at 120 μg/ml.

E. coli cells used to overproduce ImmA and ImmR were grown in LB medium at 30°C or 37°C. When appropriate, ampicillin (100 to 200 μg/ml) and/or chloramphenicol (15 μg/ml) was added. To induce gene expression in BL21-AI cells (Invitrogen), the growth medium was supplemented with ≥1 mM IPTG (isopropyl-β-d-thiogalactopyranoside) and 0.2% arabinose.

ImmAh mutant hunt.

Identification of hyperactive immA mutants was based on screening the expression of a Pxis-lacZ fusion. In the presence of the ImmR repressor, cells containing Pxis-lacZ are white on X-Gal indicator plates because Pxis-lacZ is efficiently repressed. Cells with wild-type immA or a nonfunctional immA are also white. A hyperactive ImmA protein should cleave and inactivate ImmR, causing derepression of Pxis-lacZ and leading to visualization of blue colonies on X-Gal.

immA was amplified from chromosomal DNA (strain JH642) by mutagenic PCR in the presence of MnCl2 by the use of Taq DNA polymerase (1.25 units in a 50-μl reaction mixture) under conditions recommended by the supplier (Roche). Products were cloned downstream of the LacI-repressible, IPTG-inducible promoter Pspank in pDR110 vector. Ligation mixtures were used to transform competent E. coli DH5α cells, selecting for ampicillin resistance. Transformants from a single PCR were pooled, and plasmid DNA was prepared from each pool and used to transform the B. subtilis BOSE533 strain (containing Pxis-lacZ and immR). Transformants were grown on LB plates containing spectinomycin (to select transformants), IPTG (to induce Pspank-immA), and X-Gal (to visualize Pxis-lacZ expression). Blue colonies were picked and restreaked to purify single colonies. Each candidate colony was grown in liquid LB, and genomic DNA was purified from these cells. This DNA was used to transform the BOSE533 indicator strain, selecting for the spc marker associated with Pspank-immA at amyE and screening for blue coloration (in the presence of IPTG when used to express Pspank-immA). Genomic DNA from a candidate that produced blue transformants indicated that the blue phenotype was caused by a mutation linked to spc at amyE, most likely a mutation that made ImmA hyperactive. To verify this, immA was amplified from this DNA by high-fidelity PCR (Platinum Taq DNA polymerase; Invitrogen), and the PCR products were sequenced (MIT Biopolymers Laboratory).

Mutations identified in the candidates were reconstructed by site-directed mutagenesis of immA in pBOSE540 (pDR110-immA) (by use of Quikchange site-directed mutagenesis; Stratagene), and the resulting plasmids were introduced into JH642 by double crossovers at amyE. Genomic DNA from these strains was used to introduce each immAh allele into indicator strains such as BOSE533.

Western blot analysis.

Samples collected from B. subtilis cultures were flash-frozen (liquid nitrogen) and stored at −20°C before being thawed and pelleted, or they were pelleted immediately. Cell pellets were washed with TN buffer (50 mM Tris, 300 mM NaCl, pH 8) and stored at −20°C. Cell pellets were thawed and resuspended in buffer (10 mM Tris, 10 mM EDTA, pH 7) containing 0.1 mg/ml lysozyme and the protease inhibitor 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride (AEBSF) at 1 mM. The volume of buffer used to resuspend each sample of cells was adjusted to the optical density at 600 nm (OD600) in order to normalize the concentration of proteins in each sample. Resuspended cells were incubated at 37°C for 30 min, SDS sample buffer was added, and samples were heated at 100°C for 10 min followed by centrifugation to remove insoluble material.

Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on 15% or 18% gels and transferred to a PolyScreen polyvinylidene difluoride (PVDF) membrane (Perkin Elmer) using a Trans-blot semidry electroblot transfer apparatus (Bio-Rad). Membranes were blocked in 0.2% I-Block (Tropix)-TBST (50 mM Tris, 200 mM NaCl, 0.05% Tween 20, pH 8) either at room temperature for 1 h or overnight at 4°C. The PVDF membranes were incubated in a 1:5,000 dilution of anti-ImmA rabbit polyclonal antisera (made commercially by Covance using ImmA-His6 protein purified from E. coli) in 0.2% I-Block TBST for 1 h at room temperature, washed several times in TBST, incubated in 1:3,000 goat anti-rabbit IgG-horseradish peroxidase (IgG-HRP) conjugate (Bio-Rad) for 1 h at room temperature, and washed several times in TBST. Signals were detected using Western Lightning chemiluminescence reagents (Perkin-Elmer) followed by exposure to Kodak Biomax light film.

The amount of ImmA in B. subtilis lysates was quantified by comparison to purified ImmA on Western blots with anti-ImmA antibodies. Films were scanned and densities were assessed using ImageQuant TL (GE Healthcare). For each blot, a standard curve was generated using at least four different quantities of purified ImmA. ImmA levels that fell within the linear range of the standard curve for each blot were measured for each B. subtilis sample. For each B. subtilis sample, two or more independent measurements of ImmA level by Western blotting were averaged.

β-Galactosidase assays.

β-Galactosidase specific activity was assayed as described previously (12). Specific activity was calculated relative to the optical density at 600 nm of the samples. Results shown are from a single experiment and are representative of results obtained in at least three independent experiments, except for Fig. Fig.2A,2A, which shows the results of one experiment.

FIG. 2.
Effects of ImmA levels on expression of Pxis-lacZ. All strains were cured of ICEBs1 (ICEBs10) and expressed immR, immA, and rapI at ectopic loci. Expression of immA from Pspank or the stronger Pspank(hy) was controlled with different concentrations of ...

Protein purification and in vitro assays.

N-terminally His-tagged ImmR, untagged ImmA, and untagged mutants of ImmA were purified and assayed as previously described (4).

RESULTS

Overexpression of immA can derepress Pxis-lacZ.

Previous experiments indicated that cells with C-terminally epitope-tagged ImmA had elevated ImmA protein levels and a small increase in ICEBs1 gene expression (J. M. Auchtung and A. D. Grossman, unpublished results). Based on these results, we decided to directly test the relationship between levels of native (untagged) ImmA and transcription from the major promoter in ICEBs1. We used transcription of a Pxis-lacZ fusion as an indicator of ICEBs1 gene expression. Pxis is strongly repressed by ImmR (1). Derepression requires ImmA and occurs after overexpression of rapI or during the RecA-mediated SOS response (4). This regulation occurs in both the presence and the absence of all other ICEBs1 genes (1, 4). For simplicity, the experiments presented here were done with cells lacking ICEBs1 (ICEBs10).

We found that increased expression of immA caused increased expression of Pxis-lacZ. We modulated expression of immA by the use of a fusion to the IPTG-inducible promoter Pspank (Pspank-immA) or to the stronger Pspank(hy) [Pspank(hy)-immA] promoter and by adjusting immA expression levels with different concentrations of IPTG. The cells contained immR, expressed from its own promoter, and a copy of rapI fused to the xylose-inducible promoter Pxyl (Pxyl-rapI) but were grown in the absence of xylose. Expression of Pxis-lacZ (Fig. (Fig.2A)2A) was low (∼1 specific activity unit) when cells containing the weaker Pspank-immA fusion were grown with 1 mM IPTG (fully induced Pspank). Expression of Pxis-lacZ was higher (∼10 specific activity units) when cells containing Pspank(hy)-immA were grown with 10 μM IPTG and higher still (∼150 specific activity units) when grown with 100 μM IPTG (Fig. (Fig.2A).2A). Expression of Pxis-lacZ was fully derepressed (∼1,000 specific activity units) when cells containing Pspank(hy)-immA were grown with 1 mM IPTG (data not shown) or when rapI was overexpressed from Pxyl-rapI following the addition of xylose (see below).

We measured relative levels of ImmA protein (in Western blots with anti-ImmA antibodies) under the conditions used for monitoring expression of Pxis-lacZ. As expected, the relative amounts of ImmA increased with higher concentrations of IPTG (Fig. 2B and C). In cells with a >5-fold increase in the amount of ImmA [from Pspank(hy)-immA with 100 μM IPTG compared to the fully induced weaker Pspank-immA], there was an approximately 150-fold increase in ß-galactosidase specific activity (Fig. (Fig.2A).2A). These results indicate that increased levels of ImmA cause increased transcription from Pxis.

Induction of Pxis-lacZ by overexpression of RapI does not cause an increase in the amount of ImmA.

Although increasing the amount of ImmA caused increased expression of Pxis-lacZ, the amount of ImmA did not increase when expression of Pxis-lacZ was induced by overproduction of RapI. Addition of xylose to induce expression of rapI (from Pxyl-rapI), in cells with Pspank-immA grown with 1 mM IPTG, caused an increase in expression of Pxis-lacZ (Fig. (Fig.2A).2A). By approximately 60 min after expression of RapI, there was a >100-fold increase in ß-galactosidase specific activity, and after 90 to 120 min, there was an approximately 1,000-fold increase in ß-galactosidase specific activity (Fig. (Fig.2A).2A). Under these conditions, there was little or no detectable change in the amount of ImmA (Fig. 2B and D). These results indicate that Pxis-lacZ is derepressed by RapI even though immA is expressed from a heterologous promoter and that derepression can occur without a significant increase in the amount of ImmA. Importantly, the increase in transcription from Pxis with no detectable change in ImmA levels was significantly greater than that caused by an approximately 5-fold increase in ImmA.

Isolation of hyperactive mutants of ImmA.

To better understand the mechanisms of derepression of ICEBs1 gene expression, we isolated and characterized mutations in immA (immAh) that cause increased expression of Pxis-lacZ without exogenous induction, i.e., without induction of the SOS response and without overexpression of rapI. In the presence of the repressor ImmR, cells containing Pxis-lacZ are white on X-Gal indicator plates because Pxis-lacZ is efficiently repressed. Cells with wild-type immA or a nonfunctional immA are also white. A hyperactive ImmA should cleave and inactivate ImmR, causing derepression of Pxis-lacZ and leading to the visualization of blue colonies on X-Gal.

We used a strain (BOSE533) that contains Pxis-lacZ, expresses immR from its own promoter (to repress Pxis-lacZ), and has a Pxyl-rapI fusion. We introduced mutagenized immA under the control of Pspank into these cells and screened for mutants that formed blue colonies (in the presence of IPTG), indicating at least partial derepression of Pxis-lacZ. The immA alleles from candidate mutants were backcrossed and cells retested to be sure the mutant phenotype was linked to immA. Mutant immA alleles were then sequenced. Since several alleles contained multiple mutations, we reconstructed each single mutation by site-directed mutagenesis of immA and tested each for effects on expression of Pxis-lacZ (Materials and Methods). Each mutation is designated by the amino acid in wild-type ImmA, its position in the sequence, and the amino acid to which it was changed. Changes to a stop codon are designated with an asterisk (*). Single mutations causing K13I, T23S, R85G, K90R, V92A, V92E, N93D, F144I, F156S, and I165* were remade (Fig. (Fig.3).3). Four of these caused significantly increased expression of Pxis-lacZ (see below), and the others had little or no effect on expression of Pxis-lacZ (data not shown) and were not characterized further.

FIG. 3.
ImmA sequence and mutations. The complete amino acid sequence of ImmA is shown. The horizontal bar indicates the putative zinc-binding motif HELGH. Single amino acid substitutions are shown above and below the sequence. Circled mutations were identified ...

Four mutations in immA (R85G, V92E, N93D, and I165*) caused increased expression of Pxis-lacZ on X-Gal indicator plates, indicating that they somehow caused ImmA to be hyperactive. Three of these mutations (R85G, V92E, and N93D) are in the central region of ImmA, just C terminal to the predicted zinc-binding motif HEXXH (Fig. (Fig.3).3). The I165* mutation truncates the protein by five amino acids at its C terminus. We further explored the effect of truncating ImmA at its C terminus by targeted mutagenesis. Nonsense mutations at positions 155, 159 through 167, and 169 caused a hyperactive phenotype, and nonsense mutations at positions 138, 149, and 168 caused a null phenotype (Fig. (Fig.3),3), as detected on X-Gal indicator plates.

Effects of ImmAh mutants on expression of Pxis-lacZ.

We further characterized the three central missense mutations (R85G, V92E, and N93D) and three of the C-terminal nonsense mutations (I160*, I165*, and G169*). All three caused an increase in expression of Pxis-lacZ in cells grown in defined liquid medium (Fig. (Fig.4),4), although to different extents. R85G had the largest effect (Fig. 4A and C) and G169* the smallest (Fig. 4B and D). Expression of Pxis-lacZ in the R85G mutant appeared to be fully derepressed, and expression in all the other mutants was further increased after overexpression of rapI, with all strains reaching the same maximum level of Pxis-lacZ expression (Fig. 4A and B).

FIG. 4.
Effects of ImmAh mutants on Pxis-lacZ. All strains were cured of ICEBs1 (ICEBs10) and expressed immR, immA, and rapI at ectopic loci. Strains harboring immAh mutants were compared to the wild type. Cultures were grown in defined minimal medium with 1 ...

Expression of Pxis-lacZ in the mutants was also further increased after the addition of mitomycin C (MMC) to induce a DNA damage response. In these experiments, strains with higher uninduced (no MMC) levels of Pxis-lacZ expression had a higher induced level of ß-galactosidase specific activity than those with lower uninduced expression levels (Fig. 4C and D). This is consistent with the typically lower induction of ICEBs1 seen in response to MMC than in response to RapI (2). In addition, the presence of MMC caused a decrease in cell growth and viability as judged by a drop in the OD600 ∼2 h after treatment, indicating that cells may have begun to die before all the strains could accumulate the maximum level of ß-galactosidase.

We constructed several alleles of immA that contained two mutations, each of which causes a hyperactive phenotype. The double mutants V92E G169* (Fig. (Fig.5A),5A), V92E I165* (Fig. (Fig.5B),5B), V92E N93D (Fig. (Fig.5C),5C), and R85G I165* (Fig. (Fig.5D)5D) all caused higher expression of Pxis-lacZ than either of the respective single mutations (Fig. (Fig.5).5). A fifth combination, N93D S155*, showed higher expression of Pxis-lacZ than one of the single mutants (S155*) but less than the other (N93D) (data not shown). These results indicate that, for some of the combinations, the effects on expression of Pxis-lacZ appear to be additive, which is consistent with the notion that the central and C-terminal mutants might have different effects on ImmA.

FIG. 5.
Effects on Pxis-lacZ expression of immA alleles that contain two mutations, either of which causes a hyperactive phenotype. All strains were cured of ICEBs1 (ICEBs10) and expressed immR and immA at ectopic loci. Cultures were grown in defined minimal ...

RapI and RecA are not required for ImmAh phenotypes.

The ImmAh mutants respond to both RapI and DNA damage, and the strain background in which we isolated the immAh mutants was recA+ and contained Pxyl-rapI. Therefore, we tested whether the ImmAh mutants required endogenous RecA and perhaps low-level (leaky) expression of Pxyl-rapI under otherwise noninducing conditions. We found that the immAh mutations tested (R85G, V92E, I165*, and G169*) all caused elevated expression of Pxis-lacZ in a recA null mutant in the absence of rapI (Fig. (Fig.6A),6A), although expression was about 20% of that seen in recA+ cells with Pxyl-rapI (Fig. (Fig.6B).6B). In cells producing wild-type ImmA, expression of Pxis-lacZ was similarly reduced in the recA mutant in the absence of rapI (Fig. (Fig.6),6), indicating that much of this effect was independent of the immAh mutations. These findings indicate that the ImmAh mutants are hyperactive in the absence of either of the known inducers (RecA and RapI).

FIG. 6.
Effects of rapI and recA on Pxis-lacZ expression. All strains were cured of ICEBs1 (ICEBs10) and expressed immR and immA at ectopic loci. Cells harbored wild-type immA or an immAh mutant. Cultures were grown in defined minimal medium with 1% arabinose ...

ImmA C-terminal truncation mutants have increased protein levels in vivo.

Because elevated levels of ImmA caused increased expression of Pxis-lacZ (Fig. (Fig.2),2), we determined whether any of the immAh mutations caused an increase in the amount of ImmA in the cell. We measured relative amounts of wild-type ImmA and six ImmAh mutants by the use of Western blots with anti-ImmA antibodies. Each of the three central ImmAh mutants (R85G, V92E, and N93D) was present in the cell at a level similar to that seen with wild-type ImmA (Fig. (Fig.7).7). In contrast, the three C-terminal mutants examined (I160*, I165*, and G169*) had significantly higher protein levels in vivo compared to wild-type ImmA levels (Fig. (Fig.7).7). There was approximately 3-fold more ImmAG169* and approximately 7-fold more ImmAI160* and ImmAI165* than wild-type ImmA. These relative levels are consistent with the effect of each of the truncation mutations on expression of Pxis-lacZ (Fig. (Fig.4).4). It seems most likely that the truncation mutations affect stability of ImmA, indicating that the wild-type protein is normally unstable. Initial results from screening known cellular protease mutants indicated that ImmA levels were significantly higher in a clpP mutant (B. Bose and A. D. Grossman, unpublished results). We conclude that the truncation mutations likely cause increased expression of Pxis-lacZ due to increased stability of the mutant ImmA protein and that the central mutations likely cause increased expression of Pxis-lacZ by some other mechanism.

FIG. 7.
Cellular levels of ImmAh mutant proteins. Levels of ImmA protein in strains expressing different alleles of immA were determined by Western blotting with anti-ImmA antibodies. For each strain, ImmA levels were quantified, and the averages of the results ...

Central immAh mutations increase the activity of ImmA in vitro.

The ability of the central ImmAh mutant proteins to cleave ImmR in vitro was greater than that of wild-type ImmA or the truncation mutants. We purified the three central mutants (R85G, V92E, and N93D) and three C-terminal truncations (I160*, I165*, and G169*) and compared the abilities of these ImmA mutants to cleave ImmR (His6-ImmR) in vitro. The C-terminal truncations all cleaved ImmR (His6-ImmR) at a rate comparable to that of wild-type ImmA (Fig. (Fig.8).8). In these reactions, less than 50% of the ImmR was cleaved by wild-type ImmA after 180 min, as judged by the appearance of a fragment of ImmR and the ratio of the amount of the fragment to that of full-length ImmR (Fig. (Fig.88).

FIG. 8.
In vitro proteolysis of ImmR by ImmAh mutants. His6-ImmR (180 μM) and ImmA (12 μM) were incubated together at 37°C for the times indicated above each column. Reaction products were visualized by Coomassie-stained SDS-PAGE. Each ...

In contrast, all three central ImmA mutants cleaved ImmR significantly faster than did wild-type ImmA (Fig. (Fig.8).8). In these reactions, >90% of the ImmR was cleaved by each ImmA mutant within 180 min (Fig. (Fig.8).8). The R85G mutant appeared to be the most active, which is consistent with the in vivo effects on expression of the presence of Pxis-lacZ (Fig. (Fig.4).4). Together, our results indicate that the C-terminal truncation mutations affect the amount of ImmA protein in vivo but do not have a significant effect on ImmA specific activity and that the central mutations primarily affect the specific activity of ImmA and do not significantly affect ImmA levels in vivo.

DISCUSSION

Our results indicate that there are at least two ways to increase ImmA-mediated proteolysis of ImmR to cause derepression of ICEBs1 gene expression: (i) by increasing the amount of ImmA and (ii) by increasing the specific activity of ImmA. We found that RapI (and probably RecA) likely activates ICEBs1 by causing an increase in the specific activity of ImmA in vivo.

ImmA levels and derepression of Pxis.

We found that increasing the amount of wild-type ImmA, by increasing its production from a heterologous promoter, caused increased expression of Pxis-lacZ, a reporter for ICEBs1 gene expression. Preliminary measurements indicate that ImmA levels are significantly increased in a clpP protease mutant (Bose and Grossman, unpublished). In addition, truncations of ImmA that removed several C-terminal amino acids also caused an increase in the amount of ImmA and an increase in expression of Pxis-lacZ. These findings indicate that ImmA is normally unstable and that this instability is influenced by C-terminal residues of ImmA. There are many examples of regulatory proteins that are constitutively degraded, allowing for rapid changes in their intracellular concentrations when they are stabilized against proteolysis (10).

Although increasing the amount of ImmA in the cell caused derepression of ICEBs1 gene expression, this does not seem to be the primary mechanism for derepression in response to RapI. The amount of ImmA did not detectably change when Pxis-lacZ expression was induced by production of RapI. Under these conditions, there was an approximately 1,000-fold increase in ß-galactosidase specific activity from Pxis-lacZ. When the amount of wild-type ImmA was increased approximately 4- to 5-fold, there was an approximately 100-fold increase in ß-galactosidase specific activity from Pxis-lacZ, significantly less than the increase in expression caused by RapI. These findings indicate that something other than an increase in ImmA levels causes RapI-dependent derepression of ICEBs1 gene expression.

Mutations that increase the specific activity of ImmA.

We also isolated mutations that affect the central part of ImmA, near the putative metal-binding motif, and that cause an increase in ICEBs1 gene expression. In contrast to the C-terminal truncation mutations that affect the levels of ImmA, these mutations do not detectably affect the cellular concentration of ImmA. Rather, these ImmA mutant proteins were more efficient than the wild type at cleaving ImmR both in vivo and in vitro, and these effects were generally consistent with the effects on Pxis-lacZ expression in vivo. These findings indicate that the mutations cause an increase in the specific activity of ImmA.

There are several mechanisms by which the ImmA mutants could have increased specific activity, and these could be related to the mechanisms by which RapI and RecA increase the activity of ImmA. The ImmA mutants might have altered conformations that enable them to recognize and/or cleave ImmR more efficiently than the wild type. It is also possible that the mutations allow ImmA to fold more easily or enhance incorporation or retention of metal (presumably zinc) into the protein.

Activation of ICEBs1 gene expression by RapI and RecA.

ICEBs1 gene expression is normally induced by production of active RapI or during the RecA-dependent SOS response (Fig. (Fig.1B)1B) (2). The two activation mechanisms are independent of each other in that (i) SOS stimulates induction of ICEBs1 in a rapI null mutant and (ii) RapI production induces ICEBs1 in a recA null mutant (2). Both RapI and RecA stimulate ImmA-dependent cleavage of ImmR, and RapI-mediated stimulation appears to be direct, as it occurs in the heterologous host E. coli (4). For simplicity, we assume that RecA-stimulated cleavage is also direct, although this has not yet been determined. Attempts to activate ImmA in vitro with purified RapI and RecA have not yet been successful.

There are several possible mechanisms by which RapI and RecA could activate ImmA-mediated cleavage of ImmR. Three general possibilities include (i) activation of ImmA, (ii) activation of ImmR, making it a better substrate for ImmA, and (iii) activation by bringing ImmA and ImmR together. It is not clear whether RapI and RecA stimulate ImmA-mediated cleavage of ImmR by the same mechanism. For discussion purposes and for simplicity, we assume that the RapI and RecA work similarly, although this clearly need not be the case.

Isolation of mutations in immA that cause significant derepression of ICEBs1 gene expression points to ImmA as the target of RapI and RecA. Based on our results, we propose that RapI (and likely RecA) causes induction of ICEBs1 gene expression by causing an increase in the specific activity of ImmA and not by causing an increase in its cellular concentration. There are several possible mechanisms by which the specific activity of ImmA might be increased. RapI and RecA might bind to ImmA to make it favor a conformation that cleaves ImmR faster. Alternatively, they might chaperone ImmA to help it fold properly and to prevent it from forming inactive aggregates. It is also possible that RapI and RecA cause a covalent change in ImmA, although we have not detected any such change and do not favor this possibility.

ImmA and ImmR in other mobile genetic elements.

Homologs of ImmA and ImmR are found in many other mobile elements and putative mobile elements (4). The roles of ImmA and ImmR homologs in phage ø105 are similar to those of ImmA and ImmR in regulating ICEBs1 (4). The ImmR homolog of phage ø105 represses transcription to maintain lysogeny, and the ImmA homolog is required for induction of lytic growth when the SOS response is induced. Thus, the mode of regulation involving ImmA and ImmR is likely conserved in a variety of other systems. Alignments indicate that many ImmA homologs are shorter than ICEBs1 ImmA and are missing the C-terminal 10 to 15 residues (data not shown). In the homologs that are similar in length to ICEBs1 ImmA, the C-terminal regions are not conserved. Since the C-terminal residues of ICEBs1 ImmA are dispensable and contribute to the instability of ImmA, we suspect that many of the ImmA homologs are likely to be stable and are regulated by controlling the activity and not the amount of the protein. Further characterization of ImmA and the mechanisms by which it is regulated is likely to provide information about the regulation of many agents of horizontal gene transfer.

Acknowledgments

We thank J. M. Auchtung, C. A. Lee, A. I. Goranov, M. B. Berkmen, and S. E. Cohen for helpful discussions and C. A. Lee, J. Thomas, and K. Menard for comments on the manuscript.

This work was supported in part by an NSF graduate fellowship to B.B. and Public Health Service grant GM50895 from the NIH to A.D.G.

Footnotes

Published ahead of print on 29 October 2010.

REFERENCES

1. Auchtung, J. M., C. A. Lee, K. L. Garrison, and A. D. Grossman. 2007. Identification and characterization of the immunity repressor (ImmR) that controls the mobile genetic element ICEBs1 of Bacillus subtilis. Mol. Microbiol. 64:1515-1528. [PMC free article] [PubMed]
2. Auchtung, J. M., C. A. Lee, R. E. Monson, A. P. Lehman, and A. D. Grossman. 2005. Regulation of a Bacillus subtilis mobile genetic element by intercellular signaling and the global DNA damage response. Proc. Natl. Acad. Sci. U. S. A. 102:12554-12559. [PMC free article] [PubMed]
3. Berkmen, M. B., C. A. Lee, E. K. Loveday, and A. D. Grossman. 2010. Polar positioning of a conjugation protein from the integrative and conjugative element ICEBs1 of Bacillus subtilis. J. Bacteriol. 192:38-45. [PMC free article] [PubMed]
4. Bose, B., J. M. Auchtung, C. A. Lee, and A. D. Grossman. 2008. A conserved anti-repressor controls horizontal gene transfer by proteolysis. Mol. Microbiol. 70:570-582. [PMC free article] [PubMed]
5. Burrus, V., J. Marrero, and M. K. Waldor. 2006. The current ICE age: biology and evolution of SXT-related integrating conjugative elements. Plasmid 55:173-183. [PubMed]
6. Burrus, V., G. Pavlovic, B. Decaris, and G. Guedon. 2002. Conjugative transposons: the tip of the iceberg. Mol. Microbiol. 46:601-610. [PubMed]
7. Burrus, V., G. Pavlovic, B. Decaris, and G. Guedon. 2002. The ICESt1 element of Streptococcus thermophilus belongs to a large family of integrative and conjugative elements that exchange modules and change their specificity of integration. Plasmid 48:77-97. [PubMed]
8. Cheo, D. L., K. W. Bayles, and R. E. Yasbin. 1992. Molecular characterization of regulatory elements controlling expression of the Bacillus subtilis recA+ gene. Biochimie 74:755-762. [PubMed]
9. Churchward, G. 2002. Conjugative transposons and related mobile elements, p. 177-191. In N. L. Craig (ed.), Mobile DNA II. ASM Press, Washington, DC.
10. Gottesman, S. 1996. Proteases and their targets in Escherichia coli. Annu. Rev. Genet. 30:465-506. [PubMed]
11. Harwood, C. R., and S. M. Cutting. 1990. Molecular biological methods for Bacillus. Wiley, Chichester, England.
12. Jaacks, K. J., J. Healy, R. Losick, and A. D. Grossman. 1989. Identification and characterization of genes controlled by the sporulation-regulatory gene spo0H in Bacillus subtilis. J. Bacteriol. 171:4121-4129. [PMC free article] [PubMed]
13. Roberts, A. P., and P. Mullany. 2009. A modular master on the move: the Tn916 family of mobile genetic elements. Trends Microbiol. 17:251-258. [PubMed]
14. Rudner, D. Z., K. S. Breger, and D. C. Rio. 1998. Molecular genetic analysis of the heterodimeric splicing factor U2AF: the RS domain on either the large or small Drosophila subunit is dispensable in vivo. Genes Dev. 12:1010-1021. [PMC free article] [PubMed]
15. Salyers, A. A., N. B. Shoemaker, A. M. Stevens, and L. Y. Li. 1995. Conjugative transposons: an unusual and diverse set of integrated gene transfer elements. Microbiol. Rev. 59:579-590. [PMC free article] [PubMed]
16. Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
17. Vasantha, N., and E. Freese. 1980. Enzyme changes during Bacillus subtilis sporulation caused by deprivation of guanine nucleotides. J. Bacteriol. 144:1119-1125. [PMC free article] [PubMed]
18. Wozniak, R. A., and M. K. Waldor. 2010. Integrative and conjugative elements: mosaic mobile genetic elements enabling dynamic lateral gene flow. Nat. Rev. Microbiol. 8:552-563. [PubMed]

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

Formats:

Save items

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...