Characterization of comQ and comX, two genes required for production of ComX pheromone in Bacillus subtilis

Abstract

Many microbes use secreted peptide-signaling molecules to stimulate changes in gene expression in response to high population density, a process called quorum sensing. ComX pheromone is a modified 10-amino-acid peptide used by Bacillus subtilis to modulate changes in gene expression in response to crowding. comQ and comX are required for production of ComX pheromone. We found that accumulation of ComX pheromone in culture supernatant paralleled cell growth, indicating that there was no autoinduction of production of ComX pheromone. We overexpressed comQ and comX separately and together and found that overexpression of comX alone was sufficient to cause an increase in production of ComX pheromone and early induction of a quorum-responsive promoter. These results indicate that the extracellular concentration of ComX pheromone plays a major role in determining the timing of the quorum response and that expression of comX is limiting for production of ComX pheromone. We made alanine substitutions in the residues that comprise the peptide backbone of ComX pheromone. Analysis of these mutants highlighted the importance of the modification for ComX pheromone function and identified three residues (T50, G54, and D55) that are unlikely to interact with proteins involved in production of or response to ComX pheromone. We have also identified and mutated a putative isoprenoid binding domain of ComQ. Mutations in this domain eliminated production of ComX pheromone, consistent with the hypothesis that ComQ is involved in modifying ComX pheromone and that the modification is likely to be an isoprenoid.

FIG. 1.

Production of ComX pheromone. Production of ComX pheromone requires expression of the 55-amino-acid (aa) precursor (encoded by comX), modification of the tryptophan at position 53, processing, and export from the cell. The final 10 amino acids, which comprise the peptide backbone of ComX pheromone, are boxed. comP and comA encode the two-component system required for response to ComX pheromone. Known promoters are labeled with an arrow and a P; putative promoters are indicated by (P). orf, open reading frame.

FIG. 2.

Expression of srfA-lacZ in strains overexpressing comQ and comX. Strains were grown in defined minimal medium with appropriate antibiotics for plasmid selection and required amino acids. Samples were harvested at indicated culture densities (OD₆₀₀) for determination of β-galactosidase specific activity. For each set of strains (panels A, B, C, and D), the maximal specific activity attained by the comQ⁺ and/or comX⁺ overexpressor was set at 100; all other data points were expressed relative to this value. All strains carry an srfA-lacZ transcriptional fusion (srfA-lacZΩ682 neo) at the amyE chromosomal locus. (A) Stars, comQ comX diploid (KTB102); triangles, wild type (KTB121). (B) Squares, multicopy comQ comX (TMH326); triangles, wild type containing the cloning vector pHP13 (TMH327). (C) Solid circles, Pspac-hy-comX (KTB360); triangles, wild type (KTB346). KTB360 and KTB346 were grown in the presence of 1 mM IPTG for 2 to 3 generations prior to collection of samples for β-galactosidase activity determination. (D) Solid diamonds, multicopy comQ (KTB349); triangles, wild type with cloning vector pHP13 (TMH327).

FIG. 3.

ComQ protein levels in strains overexpressing comQ and/or comX. Strains were grown in defined minimal medium, and samples were harvested during exponential growth. ComQ protein was detected as described in Materials and Methods. Using purified ComQ protein, we estimate that there are approximately 2,200 molecules of ComQ per cell. Panels on the left show samples from an overproducer strain, and panels on the right show samples from the appropriate control. Serial twofold dilutions of protein samples were run to show the sensitivity and linearity of detection, and numbers above each lane indicate the amount of total protein run in that lane. Data presented are representative of three independent experiments. In all cases, samples for direct comparison were run on the same gel and subjected to the same blotting conditions. Apparent differences in the wild-type controls from panel to panel reflect different exposure times for the enhanced chemifluorescence. (A) comQ comX diploid (KTB345) and wild type (KTB344). (B) Multicopy comQ comX (TMH142) and wild type with the cloning vector pHP13 (TMH143). (C) Pspac-hy-comX (KTB360) and wild type (KTB346). (D) Multicopy comQ (KTB348) and wild type with the vector pHP13 (TMH143).

FIG. 4.

ComX pheromone levels in strains overexpressing comX. Indicated strains were grown in defined minimal medium, and conditioned medium (CM) was harvested at various culture densities. Conditioned medium was fractionated to partially purify ComX pheromone, and ComX pheromone activity was assayed for the ability to induce expression of an srfA-lacZ fusion in a strain unable to produce its own ComX pheromone (JMS424). Graphs present the ComX pheromone activity in conditioned medium as a function of the culture density (optical density at 600 nm) at which the conditioned medium was harvested. (A) Triangles, wild type (KTB344); stars, comQ comX diploid (KTB345). (B) Triangles, wild-type (TMH143); open circles, multicopy comQ comX (TMH142). (C) Triangles, wild type (KTB346); solid circles, Pspac-hy-comX (comX overexpressor, KTB360). Pspac-hy-comX and the wild-type control were grown in the presence of 1 mM IPTG for 2 to 3 generations before the first conditioned medium sample was harvested.

FIG. 5.

Expression of srfA-lacZ in ComX alanine mutants. Functions of the comX mutants were judged by monitoring expression of an srfA-lacZ transcriptional fusion. Strains were grown in defined minimal medium, and samples were taken every 30 min for determination of β-galactosidase specific activity. Data presented are representative of three independent experiments. In panels A to I, KTB172 (ΔcomX) is represented by squares and KTB280 (comX⁺ ΔcomX) is represented by diamonds. Strains containing mutant alleles are represented by solid circles. (A) KTB310 (comXD47A ΔcomX). (B) KTB300 (comXP48A ΔcomX). (C) KTB302 (comXI49A ΔcomX). (D) KTB363 (comXT50A ΔcomX). (E) KTB288 (comXR51A ΔcomX). (F) KTB304 (comXQ52A ΔcomX). (G) KTB308 (comXW53A ΔcomX). (H) KTB282 (comXG54A ΔcomX). (I) KTB306 (comXD55A ΔcomX). (J) Summary of activity of the alanine mutants. Symbols: ++, functional; +, mostly functional; +/−, partially functional; −, nonfunctional. ND, not done.

FIG. 6.

Alignment of region II of IPPases with part of ComQ. ComQ and IPPases were aligned using ClustalW. Conserved residues are shaded light gray, similar residues are gray, and residues identical in ComQ and all IPPases are shaded dark gray. ISPA_BACST, IspA from Bacillus stearothermophilus; YqiDB, YqiD from B. subtilis (putative synthase); ISPA_ECOLI, IspA from E. coli; IDSA_METTM, IdsA from Methanobacterium thermoautotrophicum; GGPP_SULAC, geranylgeranyl pyrophosphate synthetase from Sulfolobus acidocaldarius; Cjejuni, putative polyprenyl synthase from Campylobacter jejuni; GerCCB, GerCC from B. subtilis; CRTE_RHOCA, CrtE from Rhodobacter capsulatus; CRTE_ERWHE, CrtE from Erwinia herbicola; COMQ_BACSU, ComQ from B. subtilis. Conserved region II is labeled as described (4), and the consensus sequence is listed. The two aspartates converted to glutamate by site-directed mutagenesis (D67E and D71E) are indicated in the lower diagram.

FIG. 7.

Functional analysis of comQD67E and comQD71E mutant alleles. (A) Functions of the comQ mutant alleles were assayed by measuring expression of an srfA-lacZ fusion. Cells were grown in defined minimal medium, and samples were taken for determination of β-galactosidase specific activity. All strains carry an srfA-lacZ transcriptional fusion (srfA-lacZΩ682 neo) integrated at the amyE chromosomal locus. Diamonds, comQD67E strain KTB216 (comQ::spc pKB28 [comQD67E in pHP13] thrC::mls); stars, comQD71E strain KTB217 (comQ::spc pKB29 [comQD71E in pHP13] thrC::mls); solid circles, comQ null strain KTB122 (comQ::spc thrC::mls); triangles, comQ⁺ strain KTB131 (comQ::spc thrC::comQ⁺ mls). (B) Immunoblot of total cellular protein probed with antibodies against ComQ. Lanes 1 to 3, KTB217 (comQD71E on pHP13); lanes 4 to 6, KTB216 (comQD67E on pHP13); lanes 7 to 9, KTB131 (thrC::comQ⁺). Total protein loaded was 1 μg (lanes 1, 4, and 7), 2 μg (lanes 2, 5, and 8), and 4 μg (lanes 3, 6, and 9). (C) Alignment of the ComQ putative isoprenoid binding region from B. subtilis (Bsub) RS-B-1, B. mojavensis (Bmoj) RO-H-1, and B. subtilis 168. Sequences were aligned using ClustalW and shaded using MacBoxShade. Conserved residues are shaded light gray, similar residues are gray, and residues identical in all three sequences are shaded dark gray. The aspartates shown to be required for ComQ function in B. subtilis 168 are indicated with asterisks.