A novel phase variation mechanism in the meningococcus driven by a ligand-responsive repressor and differential spacing of distal promoter elements

Abstract

Phase variable expression, mediated by high frequency reversible changes in the length of simple sequence repeats, facilitates adaptation of bacterial populations to changing environments and is frequently important in bacterial virulence. Here we elucidate a novel phase variable mechanism for NadA, an adhesin and invasin of Neisseria meningitidis. The NadR repressor protein binds to operators flanking the phase variable tract and contributes to the differential expression levels of phase variant promoters with different numbers of repeats likely due to different spacing between operators. We show that IHF binds between these operators, and may permit looping of the promoter, allowing interaction of NadR at operators located distally or overlapping the promoter. The 4-hydroxyphenylacetic acid, a metabolite of aromatic amino acid catabolism that is secreted in saliva, induces NadA expression by inhibiting the DNA binding activity of the repressor. When induced, only minor differences are evident between NadR-independent transcription levels of promoter phase variants and are likely due to differential RNA polymerase contacts leading to altered promoter activity. Our results suggest that NadA expression is under both stochastic and tight environmental-sensing regulatory control, both mediated by the NadR repressor, and may be induced during colonization of the oropharynx where it plays a major role in the successful adhesion and invasion of the mucosa. Hence, simple sequence repeats in promoter regions may be a strategy used by host-adapted bacterial pathogens to randomly switch between expression states that may nonetheless still be induced by appropriate niche-specific signals.

Figure 1. The P_nadA promoter and transcript level.

(A) Schematic diagram of the P_nadA elements. DR, direct repeat (border of region of horizontal transfer); GPR, growth phase regulatory region; ΔP2-ΔP5 indicates the nucleotide positions of the 5′ deletion mutants (Figure 3). The nucleotide sequence of the promoter is shown with the regions bound and protected in DNase I footprinting shaded according to the regulatory proteins tested in vitro: light grey, RNAP α-subunit; white, IHF; dark grey, NadR. (B) Transcription of each phase variant promoter is growth phase responsive. Cultures of MC-PΔ, MC-P2(x4), MC-P2(x5), MC-P2(x6), MC-P2(x7), MC-P2(x8), MC-P2(x9), MC-P2(x10), MC-P2(x11), MC-P2(x12), MC-P2(x13) strains, carrying single copy transcriptional fusions of the phase variant nadA promoter with a defined number of copies of the tetranucleotide repeat (TAAAxN) and the repeated tract deleted (Δ) (Table 1), were grown to mid-log or stationary growth phase and total RNA was prepared. Quantitative primer extension was performed using a gfp-specific primer as described in materials and methods. Autoradiographs of a representative experiment are shown as well as the quantification of transcript levels as determined by phosphorimaging. The error bars on the graph represent the standard deviations observed for the quantification of transcript levels between at least 2 biological replicates.

Figure 2. Regulatory proteins binding to the nadA promoter.

(A) DNase I footprinting of IHF protein to three different phase variant nadA promoters with 9, 6 and 7 repeats corresponding to low, medium and high transcript level in vivo, respectively, and the PΔ mutant nadA variant with a deletion of the TAAA repeated tract. To 20 fmoles of each radioactively labelled probe, 0, 43 and 172 nM (lanes 1–3) of IHF heterodimer were added. Relevant regions are marked and numbers correspond to nucleotide positions with respect to the transcriptional start site of a promoter with 9 repeats. (B) DNase I footprinting of RNAP or the α-subunit of RNAP to the indicated nadA promoter probe. The probe was incubated with 0, 0.25, 0.5, 1, 2, 4, and 5 U of RNAP (lanes 1–7) or 0, 0.17, 0.68, 2.7, 5.5, 11 µM of purified α-subunit (lanes 8–13).

Figure 3. Identification of a cis-acting element of the nadA promoter determining growth phase regulatory effects (the GPR region) and the GPR-binding protein from cleared cell extracts of MC58.

(A) Schematic representation of the mutant nadA promoter variants (based on the MC58 nadA promoter with 9 repeats) present in single copy transcriptional fusion in the MC58 background in the strains, MC-P1, MC-P2, MC-P3, MC-P4, MC-P5, MC-PΔ. The numbers indicate nucleotide positions with respect to the +1 transcriptional start site. DR, direct repeat; GPR, growth-phase regulatory region; Δ, deletion of the TAAA repeats. (B) Transcription from the mutant promoter variant fusions in log and stationary phases. The MC-P1, MC-P2, MC-P3, MC-P4, MC-P5, MC-PΔ strains were grown to mid-log and mid-stationary growth phase and total RNA was prepared from each sample. Quantitative primer extension was performed as described in materials and methods. Autoradiographs of a representative experiment are shown as well as the quantification of transcript levels. Similar results were found for deletion variants carrying 11 TAAA repeats (data not shown) although the overall transcript levels for promoters containing 11 repeats was higher than that of 9 repeats, as expected. The relative quantities between biological replicates with different numbers of repeats were reproducible within an error of 20% of the absolute value for each mutant promoter. (C) Binding activity towards the nadA promoter in cell extracts of MC58. Cell extracts were prepared from mid-log cultures of MC58 and increasing quantities were incubated with a radioactively labelled DNA probe consisting of the GPR region (−170 to −108) or P5 (−9 to +79) or an unrelated intergenic region Pcon as negative control and submitted to EMSA analysis. To ca. 80 fmoles of radioactively labelled probe, 0, 0.2, 0.6, 1.8, 5.0, 15 µg of cell extract in lanes 1–6 were added, respectively; 0 µg in lanes 10 and 12; and 15 µg in lanes 7–9, 11 and 13, were added; and 130, 400, and 1000 fmoles of cold GPR probe in lanes 7, 8, and 9 were added as specific competitor. (D) The peptide mass fingerprint spectrum of one µl of the eluted fraction after DNA affinity purification of the binding factor of the GPR region. Four of the major ions, labelled, could be assigned to tryptic peptides (positioning of the amino acids indicated above) of the NadR transcriptional regulator protein. In addition, BSA was added during the process of purification, was eluted from the column, since 3 major signals observed in the spectrum corresponded to BSA tryptic peptides (marked with an asterisk). (E) The amino acid sequence of the NMB1843 (NadR) protein showing the peptides that were identified by MS in bold and underlined.

Figure 4. The NadR repressor binds specifically to three operators in the nadA promoter.

(A) DNase I footprinting analysis with purified NadR on the nadA promoter with 9 repeats. The NadR protected regions are indicated (OpI–III) and numbers represent the nucleotide positions with respect to the transcriptional start site. The size of protected regions ranges from 20 bp (OpI and OpII), and 30 bp (OpIII), a size compatible with the binding of a protein dimer. Furthermore, in vitro cross-linking experiments with the purified NadR protein revealed the presence of cross-linked oligomers which migrated on SDS-PAGE with a molecular weight compatible with a dimer (data not shown). Therefore, NadR, similarly to other members of the MarR family of proteins is likely to be a dimer in solution. Binding reactions contained 40 fmoles of probe radioactively labelled at one extremity and 0, 7.5, 15, 30, 60, 120 nM of NadR purified dimer (lanes 1–6, respectively). (B) EMSA with radioactively labelled GPR, TAAA and P5 probes containing the individual OpI, OpIII and OpII operators, respectively, or the entire P2 nadA promoter spanning from −170 to +79 with increasing concentrations of recombinant NadR protein as indicated. The retarded migration of protein DNA complexes are indicated with asterisks.

Figure 5. The NadR repressor contributes to phase variable expression.

(A) Western Blot analysis of the level of expression of NadA and NadR in wild type strains 5/99, BZ83, ISS838, 961–5945 and MC58 carrying nadA promoters with 8, 5, 6, 12, and 9 repeats (lanes 1–5), respectively, and their NadR null mutant derivatives, 5/99- Δ 1843, BZ- Δ 1843, ISS-Δ1843, 961-Δ1843, MC-Δ1843 (lanes 6–10). Cells were recovered from overnight culture on plates and 5 µg of total protein were loaded on SDS-PAGE, blotted and stained with anti-NadA, anti-NadR, or anti-NMB2091 polyclonal antiserum. Migration of the NadA proteins is altered as these strains express NadA proteins with variations in their amino acid sequences , however the promoter sequence in each strain is identical apart from the altered number of repeats. (B) Transcription of phase variant promoters with 8, 6, 9, and, no, repeats, in the MC58 and NadR null mutant backgrounds. Total RNA was prepared from cultures of strains MC-P2(x8), MC-P2(x6), MC-P2(x9), MC-PΔ, Δ1843-P2(x8), Δ1843-P2(x6), and Δ1843-P2(x9), Δ1843- PΔ, grown to mid-log and stationary growth phase. Quantitative primer extension was performed as described in materials and methods. A representative experiment is shown. The experiment was performed on at least 2 biological replicates and the standard deviations between the values did not exceed 20% of the value.

Figure 6. Ligand-responsive regulation of NadA expression.

(A) Induction of expression of NadA by addition of a small molecule ligand 4HPA. Broth cultures of MC58 or Δ1843 were grown to OD₆₀₀ of 0.24 without (lane 1) or with 1 mM or 5 mM (lane 2 and 3) 4HPA; or to OD₆₀₀ of 0.24 and then incubated with 0, 1 or 5 mM 4HPA (lanes 4–6) added for 1 h. Cells were harvested and 5 µg of total protein from each culture was subjected to SDS-PAGE and Western Blot analysis with anti-NadA or anti-NMB2091 antibodies as negative control. (B) EMSA assays demonstrating dissociation of NadR from OpI operator in the GPR probe in vitro following the addition of 4HPA (lanes 3–5) but not the broadly acting salicylic acid ligand (lanes 6–8).

Figure 7. Model of regulation of NadA promoter.

Two promoter variants with 9 and 8 repeats representing low activity and high activity promoter phase variants, respectively, highlighting the ability of NadR to efficiently or less efficiently repress the promoters (top panels) and NadR-independent effects on the derepressed promoter basal levels possibly due to differential contacts with the α-subunit of RNAP (bottom panels) due to different spatial organization of the NadR and RNAP contact points resulting from the different number of repeats.