ComE, a competence protein from Neisseria gonorrhoeae with DNA-binding activity

Abstract

Neisseria gonorrhoeae is naturally able to take up exogenous DNA and undergo genetic transformation. This ability correlates with the presence of functional type IV pili, and uptake of DNA is dependent on the presence of a specific 10-bp sequence. Among the known competence factors in N. gonorrhoeae, none has been shown to interact with the incoming DNA. Here we describe ComE, a DNA-binding protein involved in neisserial competence. The gene comE was identified through similarity searches in the gonococcal genome sequence, using as the query ComEA, the DNA receptor in competent Bacillus subtilis. The gene comE is present in four identical copies in the genomes of both N. gonorrhoeae and Neisseria meningitidis, located downstream of each of the rRNA operons. Single-copy deletion of comE in N. gonorrhoeae did not have a measurable effect on competence, whereas serial deletions led to gradual decrease in transformation frequencies, reaching a 4 x 10(4)-fold reduction when all copies were deleted. Transformation deficiency correlated with impaired ability to take up exogenous DNA; however, the mutants presented normal piliation and twitching motility phenotype. The product of comE has 99 amino acids, with a predicted signal peptide; by immunodetection, a 8-kDa protein corresponding to processed ComE was observed in different strains of N. gonorrhoeae and N. meningitidis. Recombinant His-tagged ComE showed DNA binding activity, without any detectable sequence specificity. Thus, we identified a novel gonococcal DNA-binding competence factor which is necessary for DNA uptake and does not affect pilus biogenesis or function.

FIG. 1

(A) Alignment of B. subtilis ComEA and amino acid sequence derived from the GC FA1090 ORF. Identical residues and conservative substitutions are indicated. The predicted signal peptide cleavage site between residues 19 and 20 of the GC ORF is marked by an arrow. The putative flexible hinge in B. subtilis ComEA is underlined. (B) Multiple alignment of amino acid sequences sharing homology with the GC ComE region containing HhH motifs. Residues conserved in at least 5 of the 13 sequences are shown in bold. The sequences (and their accession numbers in GenBank, for protein or nucleotide) are from B. subtilis ComEA (B. sub, AAC36905), S. pneumoniae CelA (S.pne, AF052208), D. nodosus ORF E (partial) (D.nod, AAB41275), E. coli hypothetical protein (E.col, AAB40198), Pseudomonas aeruginosa (P.aer, AF147795), Haemophilus influenzae hypothetical protein (H.inf, HI1008), Haemophilus ducreyi ORF (H.duc, AF087414), Campylobacter jejuni putative protein (C.jej, CAB72504), Deinococcus radiodurans (D.rad, AAF11406), Mycobacterium tuberculosis hypothetical protein (M.tub, AB03744), Synechocystis sp. protein (Synec, BAA10416), and Thermotoga maritima protein (T.mar, AAD36129).

FIG. 2

Genetic organization of the comE copies in GC FA1090. The diagram shows previously characterized genes as solid boxes and putative genes as dotted-line boxes. Correia elements are indicated by gray boxes.

FIG. 3

Sequence of comE1 and flanking sequences in GC. Solid arrows above the sequence indicate the inverted repeats that constitute the extremities of the Correia element. Tsp, transcription start point. The −10 and −35 elements of the derived promoter are underlined. rbs, putative ribosomal binding site. The limits of the deletion in the construct ΔcomE1::Spc, and also for ΔcomE2, ΔcomE3, and ΔcomE4, are marked by dotted arrows, whereas the 3′ limit of the deletion in the construct ΔcomE1::Erm is indicated by the NsiI site. The ORF downstream of comE1 corresponds to the 5′ region of comP.

FIG. 4

Deletion of comE copies in GC strain MS11 and mutant phenotypes. (A) Deletions of comE, shown by Southern blot analysis. (B) Deletions of comE affect genetic transformation of GC MS11. Frequencies are means ± standard deviations from six independent experiments, performed on three different days. (C) Deletion of comE1 affects comP expression. Total RNA was prepared from the indicated strains and used as a template for RT-PCR. Amplification of tbpA, an iron-regulated gene (6), served as an internal control. Negative controls were performed with heat-inactivated RT, with no detectable product (not shown). (D) Deletions of comE affect GC ability to take up DNA. Results are means ± standard deviations of cell-associated radioactivity from one experiment done in triplicate and are representative of at least five independent assays. (E) Detection of ComE by immunoblotting. Purified rComE (25 ng) and whole-cell lysates from GC MS11 (wild type [wt]) and comE deletion strains (ca. 2 × 10⁸ CFU) were loaded on to a 12.5% polyacrylamide–Tris–Tricine–SDS gel, fractionated by electrophoresis, transferred to nitrocellulose filters, and probed with rabbit anti-rComE serum.

FIG. 5

rComE binds DNA. (A) Southwestern blot with lysates from E. coli strains hosting control plasmid pQE30, pQM1, and pQM1 with IPTG induction. The pQM1/IPTG lysate was loaded on a nickel-nitrilotriacetic acid column, and rComE was eluted with imidazole. After electrophoresis through a 15% polyacrylamide–SDS gel, proteins were transferred to a nitrocellulose filter, which was probed with a radiolabeled 0.7-kb DNA fragment containing seven copies of the DUS. (B) Gel retardation assay. A radiolabeled 96-bp DNA fragment containing a copy of the DUS was incubated with increasing amounts of rComE in the presence of 250 ng of calf thymus DNA and loaded on 8% polyacrylamide gel. (C) Gel retardation assay showing binding of rComE to DNA with or without 1 μg of poly (dI-dC). (D) Agarose gel retardation assay. Plasmid pBluescript II SK(−) was incubated with increasing amounts of rComE and loaded on a 0.8% agarose gel.

FIG. 6

Exogenous rComE is able to inhibit GC MS11 transformation in a dose-dependent manner. The transformation assay was carried out with 0.5 μg of pSY6. The inhibition by rComE was alleviated by the presence of 5 μg of poly (dI-dC). Results are means ± standard deviations from one experiment done in duplicate and are representative of at least three independent experiments.