Bacillus subtilis hlpB encodes a conserved stand-alone HNH nuclease-like protein that is essential for viability unless the hlpB deletion is accompanied by the deletion of genes encoding the AddAB DNA repair complex

Abstract

The HNH domain is found in many different proteins in all phylogenetic kingdoms and in many cases confers nuclease activity. We have found that the Bacillus subtilis hlpB (yisB) gene encodes a stand-alone HNH domain, homologs of which are present in several bacterial genomes. We show that the protein we term HlpB is essential for viability. The depletion of HlpB leads to growth arrest and to the generation of cells containing a single, decondensed nucleoid. This apparent condensation-segregation defect was cured by additional hlpB copies in trans. Purified HlpB showed cooperative binding to a variety of double-stranded and single-stranded DNA sequences, depending on the presence of zinc, nickel, or cobalt ions. Binding of HlpB was also influenced by pH and different metals, reminiscent of HNH domains. Lethality of the hlpB deletion was relieved in the absence of addA and of addAB, two genes encoding proteins forming a RecBCD-like end resection complex, but not of recJ, which is responsible for a second end-resectioning avenue. Like AddA-green fluorescent protein (AddA-GFP), functional HlpB-YFP or HlpB-FlAsH fusions were present throughout the cytosol in growing B. subtilis cells. Upon induction of DNA damage, HlpB-FlAsH formed a single focus on the nucleoid in a subset of cells, many of which colocalized with the replication machinery. Our data suggest that HlpB plays a role in DNA repair by rescuing AddAB-mediated recombination intermediates in B. subtilis and possibly also in many other bacteria.

Fig 1

(A) Growth of Bacillus subtilis strains in the presence or absence of inducer (xylose) driving synthesis of HlpB. pxyl-hlpB, transcription of hlpB (encoding HlpB) depends on presence of xylose; pDGhlpB, plasmid expressing hlpB; wild type, JM11 (scpA-yfp) strain expressing a functional YFP fusion unrelated to HlpB that grows on plates containing chloramphenicol. (B) Growth curves of B. subtilis strain VK01 (pxyl-hlpB) in minimal medium (plus Casamino Acids) in the presence (black line) or absence (gray line) of xylose. Dashed gray line, cells grown in the absence of xylose resuspended into fresh medium lacking xylose. (C) Competent cells transformed with 0.01 μg of chromosomal DNA from MH44 (hlpB::tet, pDGhlpB), selecting for tet resistance. Note that all strains could be transformed by an unrelated marker (comEA::cat).

Fig 2

Fluorescence microscopy of B. subtilis strain VK1 (pxyl-hlpB). (A) Cells growing exponentially in the presence of xylose. (B and C) Single cells (B) and short chains (C) showing growth representing about 12 doubling times after removal of xylose (images show the state of the cells when growth has ceased). Note that growing B. subtilis cells grow in chains rather than as single cells. Membranes were stained with FM4-64, DNA with DAPI. White scale bars, 2 μm; all images are equal in scale.

Fig 3

Purification of HlpB after heterologous expression in E. coli. (A) Coomassie-stained SDS-PAGE of purification steps. Lane 1, marker; lane 2, cell extract from noninduced E. coli strain; lane 3, cell extract after induction of HlpB synthesis; lane 4, elution fraction after immobilized metal ion affinity chromatography (IMAC) on Ni²⁺-Sepharose; lane 5, elution fraction after gel filtration. (B) Elution profile of gel filtration chromatography on superose. Absorption at 280 nm is shown in black. Fractions containing HlpB are designated by the horizontal line; values representing elution of marker proteins are given above the graph.

Fig 4

Gel retardation experiments (0.8% agarose, except for panel E, for which a 5% to 10% native polyacrylamide gradient gel was employed) using purified HlpB. (A and B) Plasmid DNA (0.5 μg/6 nM) was incubated with different amounts of HlpB as indicated above the lanes in the presence of 5 mM ZnCl₂. (C) Plasmid VK1 (0.5 μg/6 nM) was incubated with 100 nM HlpB and different concentrations of Zn²⁺ as indicated. (D) Different DNA molecules were incubated with 5 mM ZnCl₂ and without (lanes 2 to 5) or with (lanes 6 to 9) 200 nM HlpB. Lane 1, marker DNA; lanes 2 and 6, pVK1 (0.5 μg); lanes 3 and 7, pQE60 (0.5 μg); lanes 4 and 8, B. subtilis chromosomal DNA (0.5 μg); lanes 5 and 9, 500-bp PCR fragment (0.5 μg). Note that the shifted band is not in but is underneath the wells of the gel. (E) A 70-bp dsDNA oligonucleotide (3.2 μg) was incubated with increasing amounts of HlpB. Lane 1, no HlpB; lane 2, 1.5 μg HlpB; lane 3, 2.5 μg HlpB; lane 4, 3 μg HlpB; lane 5, 3.5 μg HlpB; lane 6, 4 μg HlpB; lane 7, 4.5 μg HlpB. (F) Different DNA molecules were incubated without (lanes 1 to 4) or with (lanes 5 to 8) 200 nM HlpB and 0.5 mM Zn²⁺. Lane 9, PstI-digested DNA; lanes 4 and 8, plasmid DNA (100 ng); lanes 3 and 7, 500-bp PCR fragment (0.5 μg); lanes 2 and 6, chromosomal Bacillus subtilis DNA (0.5 μg); lanes 1 and 5, ssDNA oligonucleotide (0.5 μg). (G) Plasmid DNA was incubated with 2 mM ZnCl₂, without or with 100 nM HlpB, and without or with 5 mM MgCl₂ as indicated. (H) Plasmid DNA was incubated with or without 100 nM HlpB and with different metal ions (5 mM) as indicated above the lanes. (I) Plasmid DNA was incubated with or without 100 nM H22A mutant HlpB and with different metal ions (5 mM) as indicated above the lanes.

Fig 5

Fluorescence microscopy of Bacillus subtilis cells. (A) HlpB-YFP in exponentially growing cells. (B and C) Localization of HlpB-FlAsH in exponentially growing cells before (B) and 45 min after (C) induction with 100 ng/ml MMC. (D to F) Colocalization experiments with HlpB-FlAsH (green in the overlay) and DnaX-mCherry (red in the overlay) during exponential growth (D) and 45 min after induction with 100 ng/ml MMC (E and F). (G) PY79 cells (lacking HlpB-FlAsH) treated with FlAsH reagent 45 min after addition of MMC. White arrowheads indicate the localization of the fusion proteins. Red arrowheads indicate the localization of DnaX, green the localization of HlpB, and yellow the colocalization of both proteins. White bars, 2 μm.

Fig 6

Survival assays. Strains were grown to mid-exponential phase (OD = 0.5) and were split into three cultures, one of which was treated with 50 ng/ml (dark gray bars), one of which was treated with 70 ng/ml (light gray bars), and one of which was left untreated (black bars). Survival is indicated in percentages of cells relative to the nontreated culture. Data are means of the results of three independent experiments. Note that addAB recJ double-mutant cells are much more sensitive to MMC treatment than recJ or addAB single-mutant cells, showing that AddAB proteins play an important role in DNA repair.

Fig 7

HlpB sequence and consensus sequence of HlpB-like proteins from bacteria and phages. Invariant residues are shaded in black, conserved residues are shaded in gray. Plus signs corresponding to the consensus sequences indicate conserved types of amino acids. Residues usually conserved in the HNH domain are underlined in the consensus sequences, with the HNH motifs highlighted in gray. The conserved arginine is not found in HlpB-like proteins and is indicated in parentheses.