Defining the Functionally Important Domain and Amino Acid Residues in Mycobacterium tuberculosis Integration Host Factor for Genome Stability, DNA Binding, and Integrative Recombination

ABSTRACT

The integration host factor of Mycobacterium tuberculosis (mIHF) consists of a single polypeptide chain, the product of the ihf gene. We previously revealed that mIHF is a novel member of a new class of nucleoid-associated proteins that have important roles in DNA damage response, nucleoid compaction, and integrative recombination. The mIHF contains a region of 86 amino acids at its N terminus, absent from both α- and β-subunits of Escherichia coli IHF. However, the functional significance of an extra 86-amino-acid region in the full-length protein remains unknown. Here, we report the structure/function relationship of the DNA-binding and integrative recombination-stimulating activity of mIHF. Deletion mutagenesis showed that an extra 86-amino-acid region at the N terminus is dispensable; the C-terminal region possesses the sequences essential for its known biological functions, including the ability to suppress the sensitivity of E. coli ΔihfA and ΔihfB cells to DNA-damaging agents, DNA binding, DNA multimerization-circularization, and stimulation of phage L5 integrase-catalyzed integrative recombination. Single and double alanine substitutions at positions Arg170 and Arg171, located at the mIHF DNA-binding site, abrogated its capacity to suppress the sensitivity of E. coli ΔihfA and ΔihfB cells to DNA-damaging agents. The variants encoded by these mutant alleles failed to bind DNA and stimulate integrative recombination. Interestingly, the DNA-binding activity of the mIHF-R173A variant remained largely unaffected; however, it was unable to stimulate integrative recombination, thus revealing a separation-of-function allele of mIHF. The functional and structural characterization of this separation-of-function allele of mIHF could reveal previously unknown functions of IHF.

IMPORTANCE The integration host factor of Mycobacterium tuberculosis is a novel nucleoid-associated protein. mIHF plays a vital role in DNA damage response, nucleoid compaction, and integrative recombination. Intriguingly, mIHF contains an extra 86-amino-acid region at its N terminus, absent from both α- and β-subunits of Escherichia coli IHF, whose functional significance is unknown. Furthermore, a triad of arginine residues located at the mIHF-DNA interface have been implicated in a range of its functions. Here, we reveal the roles of N- and C-terminal regions of mIHF and the individual residues in the Arg triad for their ability to provide protection in vivo against DNA damage, bind DNA, and stimulate integrase-catalyzed site-specific recombination.

INTRODUCTION

The nucleoid-associated proteins (NAPs) are a set of small, abundant, and generally basic proteins that not only compact the bacterial chromosome and modulate DNA supercoiling but also regulate its structure and influence various cellular processes, such as DNA replication, recombination, repair, transcription, environmental adaptation, and virulence (1–6). By specifically and nonspecifically binding to the genome, NAPs remodel and organize the bacterial chromosome through DNA bending, bridging, wrapping, and/or stiffening (6). At least 12 distinct types of NAPs have been described in Escherichia coli; among these, integration host factor (IHF), HU, Fis, and H-NS are the best characterized (3, 4, 7). The expression level of NAPs and their relative abundances vary with the growth phase of bacteria. For example, the levels of IHF rise substantially upon entry into the stationary phase of growth, becoming one of the predominant NAPs in E. coli cells (3, 4, 6–9). The adaptation and survival of pathogenic bacteria in different microenvironmental niches depend on their ability to regulate the expression of specific genes (6, 7–9). Whereas the roles played by NAPs in various cellular processes is well recognized, the major challenge is to identify the individual contribution of these different but related proteins in bacterial cells.

Homologous NAPs are widely distributed across bacterial species, including Mycobacteria, Streptomyces, Rhodococcus, and Corynebacteria (10–24); however, they are far less well characterized. The focus of the present study is mainly on the characterization of the IHF of M. tuberculosis; consequently, the existing literature relevant to IHF is discussed here. The IHF protein was first discovered in E. coli as an essential accessory host factor in phage lambda integration (25, 26). It is a small heterodimeric DNA-binding and minor groove-bending protein that specifically binds to a 13-bp core consensus DNA sequence located within the A/T-rich region (27, 28). The subunits of a functionally active IHF heterodimer are designated IHFα (11.1 kDa) and IHFβ (10.7 kDa) and are encoded by the unlinked genes ihfA and ihfB, respectively (29). Evidently, the functionality of IHF relies on its capacity to sharply bend DNA as much as 180°, which is critical for the regulation of certain promoter architectures (30–33). By bending the DNA, IHF acts as a global regulator in various cellular processes, such as replication and integrative recombination (1–6, 32). In addition, IHF represses transcription of virulence-associated genes in a wide range of bacterial pathogens (34–38). Accordingly, mutants deficient in α, β, or both subunits of IHF are often associated with a loss of virulence and bacterial motility (39–42). Notwithstanding these observations, IHF-dependent activation of gene expression at multiple promoters has also been reported. For example, IHF is necessary for the expression of the Vibrio cholerae tcpABCDEF operon (40). Similarly, the transcription of the Vibrio vulnificus vvpE gene is dependent on IHF (43). Fascinatingly, recent studies have revealed that the IHF of E. coli, through binding and bending the CRISPR leader, promotes Cas1-Cas2-mediated spacer integration at the leader-proximal repeat of the CRISPR array (44).

A novel DNA-binding protein was isolated from Mycobacterium smegmatis based on its ability to stimulate mycobacteriophage L5 integrase-catalyzed integrative recombination (45–47). The whole-genome sequence and annotation of Mycobacterium tuberculosis H37Rv revealed the presence of a putative ihf gene (Rv1388) capable of encoding a single polypeptide chain of 20 kDa, designated mIHF (for M. tuberculosis integration host factor) (45, 46). However, the purified mIHF showed an apparent mass of 25 kDa because of its anomalous mobility on SDS-polyacrylamide gels (23). Moreover, our work revealed that the mechanisms underlying binding of mIHF to DNA and integrative recombination are different from those of the E. coli IHFαβ heterodimer (23). Bioinformatics analysis suggested that the IHF of M. tuberculosis contains a region of 86 additional amino acid residues at the N terminus, which is absent from both α- and β-subunits of E. coli IHF. However, the role of the 86-amino-acid-long region remains unknown. Additionally, the contribution of individual residues (R170, R171, and R173) of the Arg triad to its function is less clear. In the present study, a structure/function analysis of mIHF was performed to define the roles of its N- and C-terminal regions and individual residues in the Arg triad for their ability to provide protection against DNA damage, bind DNA, and stimulate phage L5 integrase-catalyzed integrative recombination.

RESULTS

Functional complementation of E. coli ΔihfA or ΔihfB strain with N- and C-terminal deletion variants of mIHF against the effects of DNA-damaging agents.

To understand the functional importance of the extra 86 amino acid residues at the N terminus of mIHF, we prepared two derivatives of mIHF, a truncated derivative containing the N-terminal 86 amino acid residues (pMtihfntd) and a C-terminal derivative lacking the 86 amino acids from the N terminus (pMtihfctd). These vectors encoding either the N-terminal (mIHF-NTD) or C-terminal (mIHF-CTD) regions of mIHF were tested for their ability to suppress the sensitivity of E. coli ΔihfA or ΔihfB cells using a previously established complementation assay (23). Wild-type (WT) E. coli ihfA or ihfB strains treated with methyl methanesulfonate (MMS) and UV served as positive controls. In untreated cells, it was found that the growth characteristics of E. coli wild-type cells was comparable to those of the ΔihfA and ΔihfB strains bearing a variety of M. tuberculosis ihf variants (Fig. 1A).

FIG 1.

mIHF-CTD protects E. coli ΔihfA and ΔihfB strains against the genotoxic effects of UV and MMS. (A) Growth of E. coli ΔihfA and ΔihfB cells expressing wild-type mIHF or its deletion derivatives in the absence of UV and MMS treatment. (B) mIHF-CTD complements the sensitivity of E. coli ΔihfA and ΔihfB cells against UV- and MMS-induced cell death. (C) mIHF-NTD fails to rescue the sensitivity of E. coli ΔihfA and ΔihfB cells from UV- and MMS-induced cell death.

The next test was carried out to see if the mIHF deletion derivatives could restore the survival of E. coli ΔihfA or ΔihfB cells in response to MMS treatment. Consistent with previous studies (23), it was found that the E. coli ΔihfA or ΔihfB cells were highly sensitive to MMS compared to the wild-type cells (Fig. 1B). Notably, when these strains were complemented with mIHF-CTD, they showed a growth phenotype similar to that of the E. coli ΔihfA or ΔihfB strain bearing the full-length M. tuberculosis ihf gene. In contrast, E. coli ΔihfA and ΔihfB strains complemented with the mIHF-NTD showed a growth phenotype similar to that seen in E. coli ΔihfA and ΔihfB strains (Fig. 1B). Thus, it could be concluded that the C-terminal region of mIHF, but not the N-terminal region, is sufficient to restore the impaired ability of E. coli ΔihfA or ΔihfB cells consequent to MMS-induced DNA damage.

In order to validate these findings, the effects of C- and N-terminal derivatives of mIHF on the survival of E. coli ΔihfA or ΔihfB cells in response to UV radiation was tested, using the full-length M. tuberculosis ihf as a control. Representative images of UV-irradiated cells showed that the UV-sensitive phenotypes of both E. coli ΔihfA and ΔihfB strains was partially rescued by the C-terminal (but not the N-terminal) derivative of mIHF (Fig. 1C). These findings support the idea that the C terminus of mIHF alone is sufficient to functionally substitute for the E. coli IHFαβ heterodimer and provide protection in vivo against DNA damage and genome instability.

Functional complementation of E. coli ΔihfA or ΔihfB strain with Arg-to-Ala single, double, and triple mutants of M. tuberculosis ihf gene against the effects of DNA-damaging agents.

We have previously reported that the DNA-binding domain of mIHF is substantially different from that of the E. coli IHFαβ heterodimer (23). The modeled structure of mIHF, based on the cocrystal structure of Streptomyces coelicolor IHF-duplex DNA (20), revealed that Arg170, Arg171, and Arg173, in the C-terminal region, could be involved in DNA binding (23). Consistent with the computational prediction, it was found that E. coli ΔihfA or ΔihfB cells complemented by an mIHF triple mutant with an Arg→Ala substitution (mIHF-R170A/R171A/R173A) were hypersensitive not only to UV light but also to MMS treatment and lacked the ability to stimulate integrative recombination catalyzed by mycobacteriophage L5 integrase. To identify which of these residues (R170/R171/R173) is crucial for mIHF activity, mutants were constructed in pET15b vector bearing base changes rendering single (R170A, R171A, or R173A)- or double (R170A/R171A, R171A/R173A, or R170A/R173A)-amino-acid-substitution variants of mIHF. The mIHF triple (R170A/R171A/R173A) mutant was constructed as previously described (23). In all these variants, the most basic amino acid residue, arginine, was replaced by neutral alanine residues by site-directed mutagenesis.

To analyze the functional significance of M. tuberculosis ihf variants, E. coli ΔihfA or ΔihfB cells were individually transformed with expression vectors encoding single (R170A, R171A, or R173A), double (R170A/R171A, R171A/R173A, or R170A/R173A), or triple (R170A/R171A/R173A) M. tuberculosis ihf variants. These were tested for their ability to provide protection against MMS- or UV-induced cell death in comparison with the wild-type ihf gene. It was found that the wild-type M. tuberculosis ihf, but not its triple mutant variant, counteracted the adverse effects of MMS or UV in both E. coli ΔihfA and ΔihfB cells (Fig. 2). This effect was eliminated in E. coli ΔihfA and ΔihfB cells complemented with various single, double, and triple mutant variants of M. tuberculosis ihf. However, expression of the M. tuberculosis R173A variant partially suppressed the sensitivity of E. coli ΔihfA or ΔihfB cells to MMS and UV, indicating that R173 residue is not very important for mIHF in vivo activity. Based on cell survival assays, it was concluded that the mIHF residues R170 and R171 are important for its functionality in vivo.

FIG 2.

Survival of E. coli ΔihfA and ΔihfB cells expressing wild-type mIHF or constructs carrying single, double, or triple arginine-to-alanine mutations in mIHF following MMS or UV treatment.

mIHF-CTD, but not mIHF-NTD, exhibits robust DNA-binding activity.

To gain further insights into the mechanistic basis underlying the differences among alleles of the M. tuberculosis ihf gene in the complementation assays, N- and C-terminal deletion variants of mIHF were cloned and expressed, along with six other mutant IHF proteins containing Arg-to-Ala substitutions. All of the mIHF recombinant proteins were produced in E. coli strain Rosetta2 (DE3)pLysS and purified from the cell lysates by affinity chromatography on Ni²⁺-nitrilotriacetic acid (NTA)-agarose columns as described in Materials and Methods. However, in the case of mIHF-CTD, passage through a Superdex S75 column was necessary to obtain protein in high yield and purity (>90%) for biochemical purposes. The purity of protein samples was assessed by SDS-PAGE (Fig. 3), and they were found to be free of detectable exo- and endonucleases.

FIG 3.

Purification of N- and C-terminal truncated derivatives as well as six mutant variants of mIHF containing single or double arginine-to-alanine substitutions at positions R170, R171, and R173. SDS-PAGE analysis of protein samples of deleted derivatives and point mutants of mIHF at different stages of purification. Following electrophoresis, the gels are stained with Coomassie brilliant blue R-250 to visualize protein(s) in the gel. (A) Purification of mIHF-CTD. Lanes: 1, protein molecular mass markers; 2, sample from uninduced whole-cell lysate; 3, sample from induced whole-cell lysate; 4, sample from the pooled fractions after affinity chromatography on Ni²⁺-NTA-agarose beads; 5, sample from the pooled fractions after chromatography on a Superdex S75 column. Lanes 2, 3, and 4 contain 10 μg protein, and lane 5 contains 3 μg protein. (B) Purification of mIHF-NTD. Lanes: 1, protein molecular mass markers; 2, sample from uninduced whole-cell lysate; 3, sample from induced whole-cell lysate; 4, sample from the pooled fractions after affinity chromatography on Ni²⁺-NTA-agarose beads. Lanes 2 and 3 contained 10 μg protein, and lane 4 contained 3 μg protein. (C to E) Purification of mIHF single point mutation variants. Lanes: 1, protein molecular mass markers; 2, sample from uninduced whole-cell lysate; 3, sample from induced whole-cell lysate; 4, sample from the pooled fractions after affinity chromatography on Ni²⁺-NTA-agarose beads. Lanes 2 and 3 contain 10 μg protein, and lane 4 contains 3 μg protein. (F to H) Purification of mIHF double point mutation variants. (F) Lanes: 1, sample from uninduced whole-cell lysate (10 μg); 2, sample from induced whole-cell lysate (10 μg); 3, 3 μg protein from the pooled fractions after affinity chromatography on Ni²⁺-NTA-agarose beads; 4, protein molecular mass markers. (G and H) Lanes: 1, protein molecular mass markers; 2, sample form uninduced whole-cell lysate (10 μg); 3, sample from induced whole-cell lysate (10 μg); 4, 3 μg protein from the pooled fractions after affinity chromatography on Ni²⁺-NTA-agarose beads.

The ability of purified mIHF-NTD and mIHF-CTD deletion variants to bind double-stranded linear DNA fragments containing either attB or attP sites of mycobacteriophage L5 was assessed because full-length mIHF displayed robust binding to these sites (23). Similar experiments were performed with curved and noncurved DNA fragments, as full-length mIHF showed a strong preference for binding to curved DNA over noncurved DNA (23). The binding reactions were carried out using 2 nM ³²P-labeled attB or attP sites containing DNA substrates with increasing concentrations of either mIHF-NTD or mIHF-CTD variants. The binding buffer used throughout this study contains 100 mM NaCl, which was added to minimize nonspecific mIHF-DNA interactions. After the binding reaction, the resulting protein-DNA complexes were resolved by native PAGE. In reactions performed with increasing concentrations of mIHF-NTD and a ³²P-labeled probe containing either an attB or attP site, no measurable binding could be discerned with either probe; this was inferred from the absence of slower-moving species of DNA representing protein-DNA complexes (Fig. 4A and B). In marked contrast, under the same conditions, mIHF-CTD formed protein-DNA complexes with both probes in a concentration-dependent manner (Fig. 4E and F). Despite the similarity in the binding mode of mIHF-CTD to the attP and attB sites containing DNA substrates, it showed a distinct preference for binding to the latter. These results are consistent with the DNA-binding mode of the full-length wild-type mIHF for attP and attB sites containing DNA substrates (23), thus validating the current approach.

FIG 4.

mIHF-CTD, but not mIHF-NTD, binds to target DNA substrates. The assay was performed as described in Materials and Methods. (A to H) The reaction mixtures contained 2 nM ³²P-labeled attB, attP, curved DNA, or noncurved DNA (as indicated at the top of the images) in the absence (lane 1) or presence of 0.2, 0.3, 0.4, 0.5, 0.6, 0.75, 1, 1.2, and 1.5 μM mIHF-CTD or mIHF-NTD (lanes 2 to 10, respectively). The triangle at the top of the gel denotes increasing amounts of N- or C-terminally deleted derivatives of mIHF. (I) Graphical representation showing the extent of binding of mIHF-CTD or mIHF-NTD to attB, attP, curved DNA, or noncurved DNA substrates. As mIHF-NTD was devoid of DNA-binding activity with both types of DNA substrates, the curves representing its binding are merged into the baseline. Statistical analysis was performed using two-way analysis of variance. The data shown represent means ± standard deviations from three independent experiments: *, P < 0.05; **, P < 0.01; ***, P < 0.001 versus wild-type mIHF (attB).

To examine the DNA-binding specificity in more detail, the DNA-binding activity of mIHF-NTD and mIHF-CTD to curved and noncurved DNA was tested. Although mIHF-NTD failed to bind both curved and noncurved DNA substrates (Fig. 4C and D), mIHF-CTD formed significantly larger amounts of DNA-protein complexes with curved DNA compared to noncurved DNA (Fig. 4G and H). However, the complexes formed with the latter appeared diffuse and smeared at lower protein concentrations (Fig. 4H). Longer incubation times with noncurved DNA did not result in a higher yield of mIHF-CTD/DNA complexes. The mode of binding of mIHF-CTD to the curved and noncurved DNA substrates is in line with our previous observations with full-length mIHF (23). The amount of protein-DNA complexes was quantified over a range of input protein concentrations; this revealed that the quantitative accumulation of protein-DNA complexes occurred within a narrow range of mIHF-CTD concentrations (Fig. 4I). The binding curves obtained for the attP and attB sites and curved DNA showed a hyperbolic shape that typically characterizes one-site binding curves. On the other hand, the mIHF-CTD bound to noncurved DNA, having mixed sequence with much lower avidity, and the binding did not reach saturation in the range of protein concentrations that were analyzed in this study (Fig. 4I).

To determine the binding affinity of mIHF-CTD, 230 bp curved DNA (derived from plasmid pB16) and 220 bp noncurved double-stranded DNA (dsDNA; derived from plasmid pNB10) were used as described previously (23). The data for mIHF-CTD yielded apparent dissociation constants of 0.21 μM and 0.26 μM, respectively, for attB and attP sites containing DNA substrates, and 0.36 μM and 1 μM, respectively, for curved and noncurved DNA substrates (Fig. 4I and Table 1). These values are in accordance with the affinity of interaction of full-length mIHF with these substrates as previously determined (23). Together, these results clearly demonstrate that mIHF binds to DNA through its C-terminal region.

TABLE 1.

Measured dissociation constants for mIHF-CTD and mIHF-R173A

DNA substrate	Kd (μM)
	mIHF-CTD	mIHF-R173A
Curved DNA (pB16)	0.36 ± 0.022	1.22 ± 0.024
Noncurved DNA (pNB10)	1.00 ± 0.015	1.51 ± 0.031
attP site-containing substrate	0.26 ± 0.016	0.65 ± 0.045
attB site-containing substrate	0.21 ± 0.013	0.52 ± 0.034

DNA-binding activity of six mutant IHF proteins containing 1 to 3 arginine-to-alanine substitutions.

The mIHF triple Ala substitution (R170A/R171A/R173A) mutant was found to be hypersensitive to the effects of UV irradiation as well as MMS, devoid of DNA-binding activity, and failed to stimulate integrative recombination catalyzed by phage L5 integrase (23). In this context, we asked whether this property associated directly with DNA binding or indirectly through a protein-protein interaction. In particular, we wished to identify functionally important amino acid residues in the arginine triad that could be required for discrete functions. As a first approach toward this goal, the effects of single (R170A, R171A, or R173A) and double Ala substitutions (R170A/R171A, R171A/R173A, or R170A/R173A) on the DNA-binding activity of mIHF were tested in parallel. Binding reactions were performed using a fixed amount of the indicated DNA substrate and various concentrations of the mutant mIHF and analyzed as described above. The results revealed the importance of the arginine triad within the DNA-binding site: substitution of alanine for arginine at position 170 (R170A) or 171 (R171A) completely eliminated mIHF binding to attP and attB sites as well as to curved and noncurved DNA substrates (Fig. 5A and B, upper and middle), indicating that they are absolutely essential for the binding activity. Interestingly, however, a similar substitution at position 173 (R173A) showed a DNA-binding activity with both types of DNA substrates similar to that of mIHF-CTD (Fig. 5A and B, lower) and to that of the wild-type mIHF as reported previously (23) but at relatively higher concentrations. None of the double mutants had any measurable DNA-binding activity (Fig. 6). The binding affinities were quantitatively determined and plotted as a function of the percent DNA bound versus mIHF concentration (Fig. 5C and 6C). From a similar assay, the dissociation constant (K_d) was determined using attB and attP as well as curved and noncurved DNA substrates. Although the R173A variant exhibited an ability to bind target DNA substrates, the apparent dissociation constant was approximately 3-fold lower than that of wild-type mIHF (Table 1).

FIG 5.

DNA-binding activity of the mIHF variants containing Arg-to-Ala substitutions at positions R170/R171/R173. (A and B) The assay was performed as described in Materials and Methods. The reaction mixtures contained 2 nM ³²P-labeled attB and attP (A) or curved and noncurved (B) DNA substrates (as indicated on top of the gel images) in the absence (lane 1) or presence of 0.2, 0.3, 0.4, 0.5, 0.6, 0.75, 1, 1.2, and 1.5 μM the indicated mIHF variant protein (lanes 2 to 10, respectively). The triangle at the top of the gel denotes increasing amounts of variant mIHF proteins. (C) Graphical representation showing the extent of binding of mIHF variant proteins to attB, attP, and curved or noncurved DNA substrates. As R170A and R171A mutants lacked the ability to bind DNA, the curves representing their binding are close to or merged into the baseline. Statistical analysis was performed using two-way analysis of variance. The data shown represent means ± standard deviations from three independent experiments: *, P < 0.05; **, P < 0.01; ***, P < 0.001 versus wild-type mIHF (attB).

FIG 6.

DNA-binding activity of the mIHF variants with double point mutations at positions R170, R171, and R173. (A and B) The assay was performed as described in Materials and Methods. The reaction mixtures contained 2 nM ³²P-labeled attB and attP (A) or curved and noncurved (B) DNA substrates (as indicated on top of the gel images) in the absence (lane 1) or presence of 0.2, 0.3, 0.4, 0.5, 0.6, 0.75, 1, 1.2, and 1.5 μM (lanes 2 to 10, respectively) the indicated mIHF variant proteins. The triangle at the top of the gel denotes increasing amounts of the indicated mIHF variant proteins. (C) Graphical representation showing the extent of binding of the indicated mIHF variant protein to attB, attP, and curved or noncurved DNA substrates. As the mIHF double mutants lacked the ability to bind DNA, the curves representing their binding are merged into the baseline.

mIHF-CTD, but not mIHF-NTD or various single or double Arg-to-Ala substitution derivatives, stimulates integrative recombination promoted by phage L5 integrase.

Similar to E. coli IHFαβ, mIHF functions as an accessory factor in integrative recombination promoted by phage L5 integrase (46, 47). To gain further insight into the role of mIHF truncation derivatives and the Arg→Ala mutants, we sought to investigate the in vitro functionality of mIHF-NTD, mIHF-CTD, and mIHF single and double Ala substitution variants in integrative recombination catalyzed by phage L5 integrase. The assay was carried out as previously described (23, 47). Integrative recombination between attP and attB catalyzed by phage L5 integrase was monitored for the generation of a new DNA species with a molecular size of ∼8 kb (47). As shown in Fig. 7A, in reactions performed with increasing amounts of full-length mIHF and a fixed amount of attP and attB containing DNA substrates and phage L5 integrase, the appearance of a DNA product coincided with the concomitant disappearance of bands corresponding to attP and attB sites containing DNA substrates. Under similar conditions, the ability of various mIHF derivatives to stimulate integrative recombination catalyzed by phage L5 integrase was examined. The results showed that among the various mIHF derivatives studied, except for mIHF-CTD (Fig. 7C), the others failed to stimulate integrative recombination at all concentrations tested (Fig. 7B, D, and I). Interestingly, although the R173A variant was able to bind attP and attB sites containing DNA substrates comparable to the wild-type mIHF, it failed to promote integrative recombination (Fig. 7I). These results suggest that the DNA binding and integrative recombination functions of mIHF can be separated by substitution of Arg for Ala at position 173. The extent of formation of recombined product was determined from agarose gels and plotted versus the amount of input mIHF/mIHF variants (Fig. 7).

FIG 7.

mIHF-CTD, but not mIHF-NTD or various single and double point mutant proteins of mIHF, stimulates phage L5 integrase-catalyzed integrative recombination. The assay was performed as described previously (23). The reaction mixtures contained 200 ng of negatively supercoiled plasmid DNA bearing the attB site, 0.5 μg of phage L5 integrase, and 0.05, 0.1, 0.25, 0.5, 0.75, 1, 1.5, 2, and 2.5 μg of wild-type mIHF/mIHF-NTD/mIHF-CTD/single or double point mutant mIHF protein (lanes 4 to 13). Lanes: 1, DNA markers; 2, plasmid DNA containing the attB site; 3, negatively supercoiled plasmid DNA containing the attB site. (A) Positive-control reaction performed using wild-type mIHF; (B) reaction performed with mIHF-NTD; (C) reaction performed with mIHF-CTD. (D to I) Reactions conducted with various single and double point mutant proteins of mIHF. (J) Graphical representation of the extent of recombined DNA product formation stimulated by wild-type mIHF and variants of mIHF. All of the variants, except mIHF-CTD, failed to catalyze integrative recombination; therefore, the curves representing them are merged into the baseline.

mIHF-CTD, but not mIHF-NTD or R173A variant, promotes multimerization-circularization of linear DNA.

One of the hallmarks of the interaction of IHF with DNA is its ability to induce DNA bending, which aids in the activation of IHF-dependent promoters and formation of intasome complexes (6, 32). To test whether mIHF variants bend or induce flexibility in DNA, a ligase-mediated multimerization-circularization assay was performed as described previously (23). The reaction mixtures contained ³²P-labeled 140-bp DNA fragments excised from pUC19 plasmid DNA that is devoid of intrinsically bent sequences, with low concentrations of T4 DNA ligase and increasing concentrations of full-length mIHF/mIHF-CTD/R173A variants. Following incubation with mIHF variants, the reaction mixtures were treated with exonuclease III to degrade the linear DNA. The deproteinized products of such reactions, after separation by gel electrophoresis, were visualized using a phosphorimager. In the absence of ligase or low concentrations of mIHF, circular or linear multimeric DNA products were not detected (Fig. 8A, lanes 1 to 5). In contrast, products appeared in the form of DNA concatemers in reactions performed with increasing concentrations of full-length mIHF (Fig. 8A, lanes 6 to 8). The formation of circular DNA was ascertained in parallel reactions. Two ligation products were found to be resistant to exonuclease III treatment, thus confirming the formation of circular DNA molecules (Fig. 8A, lane 12). Although similar results were obtained with mIHF-CTD, the extent of multimerization-circularization of linear dsDNA was less than that attained with full-length mIHF (Fig. 8B), indicating that the C-terminal region of mIHF is responsible for mediating multimerization-circularization of linear dsDNA. Under similar conditions, however, the R173 variant failed to promote multimerization or circularization of linear dsDNA, consistent with the notion that protein-induced DNA bending is dramatically affected by substitution of arginine at position 173 by alanine (Fig. 8C). Collectively, these findings revealed a separation-of-function mutant of mIHF, which provides a new tool for studying the role of IHF in nucleoid structure and function.

FIG 8.

Wild-type mIHF and mIHF-CTD, but not mIHF-NTD, promote multimerization-circularization of a DNA fragment containing 140 bp. The assay was performed as described in Materials and Methods. The reaction mixtures contained 0.25 nM ³²P-labeled 140-bp DNA in the absence (lanes 1, 2, and 9) or presence (lanes 4 to 8) of 20, 40, 80, 160, and 200 nM wild-type mIHF (A), mIHF-CTD (B), or mIHF-R173 (C) with 20 U of T4-DNA ligase, as specified. Subsequently, the samples in lanes 9 to 12 were incubated with ExoIII. The plus and minus symbols shown on top of the gel represent incubations performed in the presence or absence of Exo III and T4 DNA ligase.

Analysis of the molecular mass of native IHF in M. tuberculosis strains H37Ra and H37Rv.

Previously, we cloned and overexpressed in E. coli the coding sequence corresponding to the M. tuberculosis H37Rv ihf gene (Rv1388) and purified the 25-kDa protein product (23). However, it remained possible that mIHF purified from E. coli cells as a 25-kDa polypeptide; a longer form of the ihf gene was cloned, and the actual molecular mass of mIHF in mycobacterial cells may not be 25 kDa and lacks the N-terminal 86 residues. Moreover, RiboSeq data support the idea that IHF is translated from a downstream translation initiation site in M. smegmatis and M. tuberculosis strains (48).

Western blot analysis was performed to ascertain the molecular size of native mIHF in M. tuberculosis strains H37Rv and H37Ra, the two most commonly used strains in the clinical and research laboratory setting. Aliquots of the whole-cell lysates from cells at different stages of growth were electrophoresed under reducing conditions on an SDS-polyacrylamide gel. The proteins were electrophoretically transferred onto polyvinylidene difluoride (PVDF) membranes and incubated with polyclonal antibodies against mIHF or GroEL, and the blots were developed. The data indicate that ∼25-kDa mIHF exists in M. tuberculosis H37Ra cells whose molecular mass correlated with the purified mIHF from E. coli cells (Fig. 9A). Interestingly, the molecular mass of mIHF in M. tuberculosis H37Rv cells also corresponds to the mIHF in M. tuberculosis H37Ra cells (Fig. 9C). Although weak signals corresponding to the molecular mass of mIHF-CTD were found in both cell lysates, it is not clear whether this is due to degradation or translation initiation at a downstream site. Furthermore, the first 10 amino acid residues at the N terminus [M-L-G-N-T-I-V-(X)-V-P] of mIHF, determined through repeated cycles of Edman degradation, were identical to those deduced from the M. tuberculosis ihf gene sequence. Thus, we do not know at this point the precise reason(s) for using a downstream initiation site in the ribosome profiling experiments in M. tuberculosis. Nevertheless, our results are consistent with the idea that translation of M. tuberculosis ihf mRNA begins primarily at the annotated translation start site.

FIG 9.

Western blot analysis shows that the apparent molecular mass of native IHF is similar to that of purified mIHF. The whole-cell lysates (2.5 μg protein) from M. tuberculosis strains H37Ra (A) and H37Rv (C), which were grown for several days, was used for Western blot analysis with antibodies against mIHF. Purified mIHF, mIHF-CTD, and mIHF-NTD (1 μg each) were used as positive controls. The expression level of GroEL was used as a loading control (B and D). Shown are representatives from three independent experiments.

DISCUSSION

Decades of research have shown that NAPs play important roles in the global organization of the bacterial chromosome (1–6). However, very little is known about NAPs and their potential roles in mycobacterial species. The integration host factor of M. smegmatis was originally characterized by its ability to stimulate phage L5 integrase-catalyzed DNA integration into the mycobacterial chromosome (46, 47). In addition, mIHF is essential for the viability of M. smegmatis (45, 46) and appears to be important for the growth of M. tuberculosis (49, 50), consistent with the idea that mIHF has other roles in the cell beyond its function in phage DNA integration. We have previously demonstrated that IHF expression was maximal during the late exponential growth phase of M. tuberculosis H37Ra, protects E. coli ΔihfA and ΔihfB cells against the DNA-damaging agents, and promotes nucleoid compaction (23). Furthermore, the purified mIHF showed higher binding affinity for the phage L5 attP and attB sites and stimulated integrative recombination catalyzed by phage L5 integrase (23).

One aspect of the current study was to perform systematic structure/function analysis of mIHF to correlate the in vivo phenotypes with the in vitro biochemical activities. Toward this goal, a number of important observations were made regarding the structure and function of mIHF. It was found that mIHF-CTD, lacking the 86-amino-acid region at the N terminus, complemented both E. coli ΔihfA and ΔihfB cells against the effects of DNA-damaging agents, similar to the levels seen with full-length mIHF. Under similar conditions, mIHF-NTD (lacking the C-terminal region) failed to complement the growth defects and sensitivity of either E. coli ΔihfA or ΔihfB cells against UV- and MMS-induced cell death. Furthermore, in contrast to mIHF-NTD, mIHF-CTD bound avidly to two types of DNA substrates in vitro, similar to wild-type mIHF and stimulated phage L5 integrase-catalyzed integrative recombination.

The truncation products are often a source of protein instability, or they can exert toxic influences on cells. The inactivity of mIHF-NTD seen here most likely is not attributable to protein instability or related to its deleterious effects on the cell. Two issues need to be considered: first, the expression efficiency, protein solubility, and yield of mIHF-CTD and mIHF-NTD derivatives are similar; second, the question of whether the complementation defect against UV radiation and MMS observed with mIHF-NTD is due to its toxic effect on E. coli cells. The growth of E. coli ΔihfA or ΔihfB cells transformed with the N-terminal truncation derivative strongly correlates with the C-terminal truncation transformants in the absence of DNA-damaging agents (Fig. 1A), thus excluding the possibility that lack of complementation is related to toxic effects. Hence, we conclude that the complementation defect observed with the M. tuberculosis ihf-NTD derivative of mIHF could be due to its inability to bind DNA.

Previously, we isolated and characterized the mIHF triple mutant in which R170, R171, and R173, the residues that are located in close proximity to each other at the putative DNA-binding site, were replaced with Ala. Alanine mutagenesis was chosen based on the assumption that such a substitution would have no/minimal effects on the protein structure (51, 52). It was found that E. coli ΔihfA or ΔihfB cells bearing the M. tuberculosis ihf triple Ala substitution mutant were hypersensitive to the genotoxic effects of UV irradiation as well as MMS; the substitution abrogated both the DNA-binding activity and ability to stimulate integrative recombination catalyzed by phage L5 integrase. The molecular basis of this phenotype has not yet been fully resolved. In the present study, three single (R170A, R171A, and R173A) and three double (R170A/R171A, R171A/R173A, and R170A/R173A) point mutants were generated and characterized. The double mutation, encompassing combinations of the three residues in the Arg triad, completely eradicated the activity of mutant proteins both in vivo and in vitro. On the other hand, parallel studies revealed several important insights into the function of the individual single point mutants R170, R171, and R173.

A significant relationship was found between the in vivo phenotypes of two mutant alleles (R170A and R171A) and the in vitro DNA-binding and integrative recombination activities of the corresponding proteins. For example, both mutant proteins exhibited complete loss of activity in vivo and in vitro, providing a potential explanation that the R170 and R171 residues are pivotal for the functionality of mIHF. However, the most interesting behavior was noticed in the case of R173. The R173A mutant partially suppressed the growth defects and sensitivity of E. coli ΔihfA and ΔihfB cells to the toxicity induced by DNA-damaging agents and exhibited DNA-binding activity but failed to stimulate integrative recombination catalyzed by phage L5 integrase. These results support the idea that all three Arg residues are essential for a productive integrative recombination event. The quantitative data revealed that interaction of the R173A variant with DNA was weak; this can be seen by comparing the apparent K_d values (Table 1). Therefore, it may not be robust enough for its normal function. The findings are also in good agreement with the DNA multimerization-circularization activity of the R173A variant. In our previous work, it was not possible to glean the potential contribution of arginine 173 in the triple Arg-to-Ala or double alanine substitution mutants (23). Nevertheless, this study uncovered a previously unanticipated separation-of-function R173A allele, which could guide future studies.

In summary, this work provides two valuable insights in the context of extensive knowledge available regarding the properties of E. coli IHF. First, mIHF-CTD is functionally equivalent to both the α- and β-subunits of E. coli IHF, as evidenced by its ability to suppress the growth defects and sensitivity of E. coli ΔihfA and ΔihfB cells to DNA-damaging agents and stimulate phage L5-catalyzed integrative recombination. However, it remains possible that an extra 86-amino-acid region in the full-length protein is involved in the organization of nucleoid structure and in various DNA transactions in vivo. Second, we have gleaned novel insights into the contribution of individual residues in the Arg triad; these studies led to the identification of a separation-of-function allele of M. tuberculosis ihf. Further characterization of the separation-of-function R173A allele may help to uncover previously unanticipated functions of mIHF.

MATERIALS AND METHODS

MMS and UV sensitivity assays.

Wild-type E. coli ΔihfA or ΔihfB strains harboring the M. tuberculosis ihf gene or its genetic variants (truncated constructs or mutants) under the control of the T7 promoter were grown in LB medium supplemented with 100 μg/ml ampicillin. Cells grown to an A₆₀₀ of 0.4 were treated with 0.5% MMS. After 45 min, the cells were pelleted by centrifugation and resuspended in an equal volume of M9 minimal medium. Serial dilutions were prepared in an M9 buffer, and aliquots (5 μl) were spotted onto LB agar plates containing 100 μg/ml ampicillin. Similarly, the indicated serial dilutions of E. coli ΔihfA or ΔihfB cells bearing various genetic variants of the M. tuberculosis ihf gene were plated on LB agar plates and irradiated under 254-nm UV light (UVGL-58, G6T5 lamp). For survival analysis, the petri dishes were incubated at 37°C in the dark for 22 h.

General DNA and protein techniques.

PCR amplification of the M. tuberculosis ihf gene was carried out in a Mastercycler pro S apparatus (Eppendorf) using 100 ng template DNA in the presence of the indicated primers and PfuTurbo DNA polymerase at an annealing temperature of 57.5°C. The resulting PCR products were purified with a QIAquick gel extraction kit per the manufacturer's protocol. Standard procedures were used for DNA digestion, ligation, agarose gel electrophoresis, and transformation (53). The E. coli DH5α strain was used as a cloning host for plasmid construction. The plasmids were isolated by the alkaline lysis method (53). Protein concentrations were estimated using the Bradford protein assay. The protein samples were analyzed by SDS-PAGE using 12.5% acrylamide gel electrophoresis (23).

Preparation of DNA substrates.

Plasmids pMH57 (attB) and pSS19 (attP) were digested with HindIII and BamHI, respectively. The linearized plasmid DNA molecules were labeled at the 5′ end with [γ-³²P]ATP and T4 polynucleotide kinase (53). Subsequently, the labeled DNA was cleaved with EcoRI to release 600-bp and 546-bp fragments as described previously (23). The cleavage reactions were loaded onto 5% native polyacrylamide gels and electrophoresed in 45 mM Tris-borate buffer (pH 8.3) containing 1 mM EDTA at 10 V/cm for 8 h. The bands corresponding to 600-bp and 546-bp fragments were excised from the gel and eluted into the TE buffer (10 mM Tris-HCl [pH 7.5] and 1 mM EDTA).

Intrinsically curved 230 bp of plasmid DNA (pB16) and 220 bp of noncurved DNA (pNB10) were excised from the pB16 and pNB10 plasmids, respectively, by digestion with HindIII and EcoRI. The DNA fragments were labeled at the 5′ end with [γ-³²P]ATP and T4 polynucleotide kinase. The cleavage reaction mixtures were loaded onto 8% polyacrylamide gel and electrophoresed in 45 mM Tris-borate buffer (pH 8.3) containing 1 mM EDTA at 10 V/cm for 6 h. The bands corresponding to 230 bp and 220 bp were excised from the gel and eluted into TE buffer (10 mM Tris-HCl [pH 7.5] and 1 mM EDTA).

Construction of expression vectors encoding the N-terminal or C-terminal region of mIHF.

The sequence encoding the N-terminal 86-amino-acid region of the M. tuberculosis ihf gene was amplified by PCR using pMtihf (23) as the template with the following oligonucleotides (ODNs): forward primer 5′-GAGGGCCATATGTTAGGCAACACTATTCATG-3′ and reverse primer 5′ ATACATGGATCCGATTCCTCCGTCTCT-3′. Similarly, the sequence encoding the C-terminal region of mIHF was amplified from the pMtihf plasmid template with the following ODNs: forward primer 5′-TATATTCATATGGTGGCCCTTCCCCAG-3′ and reverse primer 5′ ATACATGGATCCGATTCCTCCGTCTCT-3′. After digestion with NdeI and BamHI, the resulting 256-bp and 345-bp DNA fragments corresponded in size to the N-terminal and C-terminal regions of mIHF, respectively, and were cloned into pET15b and pET28b expression vectors. The ligation products were transformed into E. coli DH5α competent cells, and the presence of the insert was verified by colony PCR with primers for the T7 promoter and terminator regions. The resulting constructs were designated pMtihfntd and pMtihfctd. The identity of the recombinant constructs was ascertained by determining the nucleotide sequence of the cloned DNA fragments.

Construction of expression vectors encoding mutant mIHF proteins containing single or double amino acid substitutions.

The mutant variants of the M. tuberculosis H37Rv ihf gene (Rv1388) bearing base changes rendering single (R170A, R171A, or R173A) or double (R170A/R171A, R171A/R173A, or R170A/R173A) amino acid replacements (arginine with alanine) were generated individually using pairs of mutagenic primers of defined length with the mutated codon at the center of the primer (Table 2). The triple point mutant of the M. tuberculosis ihf gene (R170A/R171A/R173A) was generated as previously described (23). Site-directed mutagenesis was performed with a QuikChange mutagenesis kit (Stratagene, La Jolla, CA) using PfuTurbo DNA polymerase. To remove the plasmid template, PCR products were digested with DpnI at 37°C for 12 h. The variation in the sequence in each DNA fragment was ascertained by restriction analysis and DNA sequencing. The resulting PCR fragments were purified using a QIAquick gel extraction kit. The purified DNA fragments were directionally ligated into NdeI and BamHI sites of vector pET15b and transformed into E. coli DH5α competent cells. The transformants were selected on LB plates containing 100 μg/ml ampicillin.

TABLE 2.

Oligonucleotide primers used for the generation of mIHF single and double point mutants

Mutant	Primer	Sequencea
R170A	Forward	5′-GCTGGAAATTGCGCCCACCGCCCGCCTTCGTGGCCTCGGTG-3′
R170A	Reverse	5′-CACCGAGGCCACGAAGGCGGGCGGTGGGCGCAATTTCCAGC-3′
R171A	Forward	5′-GGAAATTGCGCCCACCCGCGCCCTTCGTGGCCTCGGTGACCG-3′
R171A	Reverse	5′-CGGTCACCGAGGCCACGAAGGGCGCGGGTGGGCGCAATTTCC-3′
R173A	Forward	5′-GCCCACCGCCGCCCTTGCTGGCCTCGGTGACCGTCAGC-3′
R173A	Reverse	5′-GCTGACGGTCACCGAGGCCAGCAAGGGCGGCGGTGGGC-3′
R170A/R171A	Forward	5′-GCTGGAAATTGCGCCCACCGCCGCCCTTCGTGGCCTCGGTGACCG-3′
R170A/R171A	Reverse	5′-CGGTCACCGAGGCCACGAAGGGCGGCGGTGGGCGCAATTTCCAGC-3′
R170A/R173A	Forward	5′-GCTGGAAATTGCGCCCACCGCCCGCCTTGCTGGCCTCGGTGACCGTCAGC-3′
R170A/R173A	Reverse	5′-GCTGACGGTCACCGAGGCCAGCAAGGCGGGCGGTGGGCGCAATTTCCAGC-3′
R171A/R173A	Forward	5′-GGAAATTGCGCCCACCCGCGCCCTTGCTGGCCTCGGTGACCGTCAGC-3′
R171A/R173A	Reverse	5′-GCTGACGGTCACCGAGGCCAGCAAGGGCGCGGGTGGGCGCAATTTCC-3′

^aUnderlining indicates mutated codons.

Expression and purification of mIHF-NTD.

E. coli Rosetta2 (DE3)pLysS cells harboring the plasmid pMtihfntd were grown in Luria-Bertani broth containing 100 μg/ml ampicillin and 34 μg/ml chloramphenicol at 37°C until an optical density (OD) at 600 nm of 0.5 was reached. mIHF-NTD was induced by the addition of 0.5 mM isopropyl-β-galactopyranoside (IPTG). The culture was incubated while being gently shaken at 37°C for 2.5 h. The cells were collected by centrifugation and washed in STE buffer (10 mM Tris-HCl [pH 8], 100 mM NaCl, and 1 mM EDTA), and the pellets were suspended in buffer A (10 mM Tris-HCl [pH 8], 150 mM NaCl, and 10% [vol/vol] glycerol) and stored at −80°C. The thawed cell suspensions were lysed by sonication (model GEX-750 ultrasonic processor) on ice at 60% duty cycles in pulse mode. The lysed cell suspensions were centrifuged in a Beckman Ti 45 rotor at 30,000 rpm for 1 h at 4°C. Purification was carried out at 4°C. The supernatant was then loaded on to a Ni²⁺-NTA column equilibrated with buffer A. After washing the column with buffer A, mIHF-NTD was eluted with a 30 to 500 mM linear gradient of imidazole in buffer A. The peak fractions containing mIHF-NTD protein as determined by SDS-PAGE were pooled and dialyzed against buffer B (10 mM Tris-HCl [pH 7.5], 1 mM EDTA, 1 mM dithiothreitol [DTT], 150 mM NaCl, and 20% [vol/vol] glycerol). The purity of mIHF-NTD was assessed by SDS-PAGE and found to be >98%. Aliquots of mIHF-NTD were stored at −80°C.

Expression and purification of mIHF-CTD.

E. coli Rosetta 2(DE3)pLysS strain harboring the plasmid pMtihfctd was grown in Luria-Bertani medium containing 100 μg/ml ampicillin and 34 μg/ml chloramphenicol at 37°C until an OD at 600 nm of 0.5 was reached. mIHF-CTD was induced by the addition of 1 mM IPTG. The culture was incubated with gentle shaking at 37°C for 5 h. The cells were collected by centrifugation, washed in STE buffer (10 mM Tris-HCl [pH 8], 100 mM NaCl, and 1 mM EDTA), resuspended in buffer A (10 mM Tris-HCl [pH 8], 150 mM NaCl, and 10% [vol/vol] glycerol), and stored at −80°C. The thawed cells were lysed by sonication (model GEX-750 ultrasonic processor) in pulse mode on ice at 60% duty cycles. The lysed cell suspensions were centrifuged in a Beckman Ti 45 rotor at 30,000 rpm for 1 h at 4°C. The supernatant was then loaded onto a Ni²⁺-NTA-agarose column equilibrated with buffer A. After washing the column with 100 ml of buffer A, bound mIHF was eluted with a 5 mM to 600 mM linear gradient of imidazole in buffer A. The peak fractions containing mIHF-CTD as determined by SDS-PAGE were pooled and dialyzed against buffer B (20 mM Tris-HCl [pH 7.5], 0.5 mM EDTA, 0.5 mM DTT, 500 mM NaCl, and 10% [vol/vol] glycerol). The dialyzed sample was loaded onto a Superdex S75 (GE Healthcare) column that had been previously equilibrated with buffer B. The peak fractions were pooled and dialyzed against buffer C (10 mM Tris-HCl [pH 7.5], 1 mM EDTA, 1 mM DTT, 200 mM NaCl, and 20% [vol/vol] glycerol). The purity of mIHF-CTD was assessed by SDS-PAGE and found to be >98%. Aliquots of mIHF-CTD were stored at −80°C.

Expression and purification of mutant mIHF proteins containing single or double amino acid substitutions.

Six mutant mIHF proteins containing either single (mIHF-R170A, mIHF-R171A, or mIHF-R173A) or double (R170A/R171A, R171A/R173A, or R170A/R173A) amino acid substitutions were individually expressed in the E. coli Rosetta2 (DE3)pLysS strain. The same procedure as that described below was used for the expression and purification of all of the mutant IHF proteins used in this study. The cells were grown in LB broth containing 100 μg/ml ampicillin and 34 μg/ml chloramphenicol at 37°C until an A₆₀₀ of 0.5 was reached. The mutant IHF proteins were induced by the addition of IPTG to a final concentration of 0.5 mM. The culture was incubated while being gently shaken at 37°C for 2.5 h. The cells were collected by centrifugation and washed in STE buffer (10 mM Tris-HCl [pH 8], 100 mM NaCl, and 1 mM EDTA), and the cell pellet was suspended in buffer A (10 mM Tris-HCl [pH 8], 150 mM NaCl, and 10% [vol/vol] glycerol) and stored at −80°C. The thawed cells were lysed by sonication (model GEX-750 ultrasonic processor) on ice at 60% duty cycles in pulse mode. The lysed cell suspensions were centrifuged with a Beckman Ti 45 rotor at 30,000 rpm for 1 h at 4°C. Purification was carried out at 4°C. The supernatant was then loaded onto a Ni²⁺-NTA column equilibrated with buffer A. After washing the column with buffer A, bound mutant mIHF protein was eluted with a 30 to 500 mM linear gradient of imidazole in buffer A. The fractions containing the mutant mIHF protein, as determined by SDS-PAGE, were pooled and dialyzed against buffer B (10 mM Tris-HCl [pH 7.5], 1 mM EDTA, 1 mM DTT, 150 mM NaCl, and 20% [vol/vol] glycerol). The purity of IHF mutant proteins was assessed by SDS-PAGE and found to be >98%. Aliquots of mIHF variants were stored at −80°C.

Electrophoretic mobility shift assay.

The reaction mixtures (20 μl) for electrophoretic mobility shift assays contained 40 mM Tris-HCl (pH 8), 100 mM KCl, 1 mM DTT, 5 mM potassium phosphate, 5% glycerol, 0.5 mM EDTA, 2 nM ³²P-labeled dsDNA, and the indicated amounts of mIHF as specified in the figure legends. After incubation at 37°C for 20 min, 2 μl of 10% loading dye was added to each sample. The samples were electrophoresed through two polyacrylamide gel types: (i) 6% gels for the separation of complexes formed with curved and noncurved DNA (pB16 and pNB10, respectively) and (ii) 4% gels in the case of complexes formed with DNA fragments containing attB and attP sites. However, the same running buffer was used in both cases, namely, 0.5× TAE (20 mM Tris-acetate buffer [pH 7.4]) containing 0.5 mM EDTA. Electrophoresis was carried out at 4°C for 6 h using 12 V/cm. The gels were dried and the bands visualized as described previously (23). The quantification of radioactive fragments was done using a UVI-Tech gel documentation station and UVI-Band Map software (version 97.04).

In vitro integrative recombination assay.

The in vitro integrative recombination assay was performed as previously described (23, 46). In the present study, purified phage L5 integrase was used as described previously (47). Briefly, the reaction mixtures contained 20 mM Tris-HCl (pH 7.5), 1 mM DTT, 25 mM NaCl, 5 mM spermidine, 0.5 μg of phage L5 integrase, 200 ng of negatively superhelical pMH39 DNA (containing the attP site), 200 ng of 3.9-kb linear dsDNA containing the attB site (generated by digestion of pMH57 with HindIII), and increasing concentrations (0.05 to 2.5 μg) of mIHF-CTD/mIHF-NTD/mutant mIHF proteins containing single or double amino acid substitutions. After incubation for 3 h at 24°C, the reaction was terminated by the addition of 1.2 μl of a solution containing 5 mM EDTA, 0.1% SDS, and 0.4 mg/ml proteinase K. The samples were incubated for 5 min at 24°C and electrophoresed through a 0.8% agarose gel at 35 V for 10 h. The gels were stained with ethidium bromide and the products identified following visualization under UV light.

DNA multimerization-circularization assay.

The DNA multimerization-circularization assay was performed as described previously (23). Briefly, the reaction mixtures (20 μl) contained 20 mM Tris-HCl (pH 8), 150 mM KCl, 1 mM DTT, 1 mM potassium phosphate, 5% glycerol, 0.25 nM ³²P-labeled 140-bp duplex DNA (prepared by digestion of pUC19 plasmid DNA with TfiI), and the indicated concentrations of wild-type mIHF/mIHF-CTD/mIHF-R173A. After incubation at 37°C for 30 min, 20 U of T4 DNA ligase (Fermentas) and 1× ligase buffer were added, and the incubation was extended for 30 min. In reactions carried out with Exo III, samples were further incubated at 37°C for 20 min with 5 U of Exo III. The reaction was terminated by the addition of 1 μl of 20% SDS and 1 μl of 10 mg/ml proteinase K, followed by incubation for 20 min at 37°C. After extraction with phenol-chloroform, the DNA was precipitated using ethanol. The DNA pellet, obtained after centrifugation, was resuspended in 5 μl of 6× DNA gel loading buffer. The samples were subjected to electrophoresis through a 5% polyacrylamide gel in 45 mM Tris-borate buffer (pH 8.3) containing 1 mM EDTA at 10 V/cm for 5 h. The gels were dried and the bands were visualized using a Fuji FLA-9000 phosphorimager.

Western blot analysis.

M. tuberculosis strains H37Ra and H37Rv were grown in Middlebrook 7H9 medium (Difco) supplemented with 10% (vol/vol) albumin-dextrose-catalase enrichment and 0.05% Tween 80 in a gyratory shaker incubator at 37°C and 180 rpm. The cells were collected at the indicated time points and lysed in a buffer (20 mM sodium phosphate buffer, pH 7.5, 300 mM NaCl, 5% glycerol, 10 mM imidazole, 5 mM 2-mercaptoethanol, and protease inhibitors). The samples of whole-cell lysates were resolved by 15% SDS-PAGE and transferred onto PVDF membranes (Millipore). The membranes were blocked using 5% skim milk in TBST buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.1% Tween 20) for 90 min at 24°C. The blots were washed six times in TBST and probed with anti-mIHF or anti-GroEL antibodies in TBST for 15 h at 4°C as described previously (23). The blots were subsequently washed three times with TBST and incubated for 90 min at 24°C with horseradish peroxidase-conjugated secondary antibody (Millipore). The blots were developed and the protein bands were visualized using an enhanced chemiluminescence method as previously described (23).

N-terminal sequencing determination.

mIHF from a polyacrylamide gel was electrophoretically transferred onto PVDF (Millipore) in CAPS (N-cyclohexyl-3-aminopropanesulfonic acid) buffer (Sigma), the membrane was stained with Ponceau S, and the mIHF band was sliced. The excised band was subjected to 10 cycles of N-terminal sequencing using the Applied Biosystems Procise sequencer, which operates on solid-phase Edman degradation chemistry. Protein sequence was analyzed by Procise pro 2.1 software. The raw data relating to the determination of mIHF N-terminal sequence is shown in Fig. S1 in the supplemental material.