pfs-dependent regulation of autoinducer 2 production in Salmonella enterica serovar Typhimurium

Abstract

Bacterial intercellular communication provides a mechanism for signal-dependent regulation of gene expression to promote coordinated population behavior. Salmonella enterica serovar Typhimurium produces a non-homoserine lactone autoinducer in exponential phase as detected by a Vibrio harveyi reporter assay for autoinducer 2 (AI-2) (M. G. Surette and B. L. Bassler, Proc. Natl. Acad. Sci. USA 95:7046-7050, 1998). The luxS gene product mediates the production of AI-2 (M. G. Surette, M. B. Miller, and B. L. Bassler, Proc. Natl. Acad. Sci. USA 96:1639-1644, 1999). Environmental cues such as rapid growth, the presence of preferred carbon sources, low pH, and/or high osmolarity were found to influence the production of AI-2 (M. G. Surette and B. L. Bassler, Mol. Microbiol. 31:585-595, 1999). In addition to LuxS, the pfs gene product (Pfs) is required for AI-2 production, as well as S-adenosylhomocysteine (SAH) (S. Schauder, K. Shokat, M. G. Surette, and B. L. Bassler, Mol. Microbiol. 41:463-476, 2001). In bacterial cells, Pfs exhibits both 5'-methylthioadenosine (MTA) and SAH nucleosidase functions. Pfs is involved in methionine metabolism, regulating intracellular MTA and SAH levels (elevated levels of MTA and SAH are potent inhibitors of polyamine synthetases and S-adenosylmethionine dependent methyltransferase reactions, respectively). To further investigate regulation of AI-2 production in Salmonella, we constructed pfs and luxS promoter fusions to a luxCDABE reporter in a low-copy-number vector, allowing an examination of transcription of the genes in the pathway for signal synthesis. Here we report that luxS expression is constitutive but that the transcription of pfs is tightly correlated to AI-2 production in Salmonella serovar Typhimurium 14028. Neither luxS nor pfs expression appears to be regulated by AI-2. These results suggest that AI-2 production is regulated at the level of LuxS substrate availability and not at the level of luxS expression. Our results indicate that AI-2-dependent signaling is a reflection of metabolic state of the cell and not cell density.

FIG. 1.

Constitutive luxS expression and the correlation of pfs transcription to AI-2 production in Salmonella serovar Typhimurium during growth in LB medium with or without 0.5% glucose. luxS expression (A) and pfs expression (B) in Salmonella serovar Typhimurium during growth in LB medium with or without 0.5% glucose was measured by using low-copy-number luxCDABE reporter fusion to the luxS and pfs promoters. Solid line, LB; dashed line, LB with 0.5% glucose. (C) AI-2 production profile of Salmonella serovar Typhimurium during growth in LB media with or without 0.5% glucose. Supernatants were prepared from samples of growing cultures, filter sterilized, and assayed for AI-2 activity by using the V. harveyi BB170 AI-2 reporter strain. Samples collected from cultures grown in LB are shown by the broken line, and the dashed line represents AI-2 activity from cultures grown in LB with 0.5% glucose. (D) Growth curves of Salmonella serovar Typhimurium 14028 containing promoter-luxCDABE reporter vectors in LB or LB with 0.5% glucose are shown by solid and dashed lines, respectively. Symbols: ▪, 14028/pAB12; ⋄, 14028/pAB13.

FIG. 2.

Effects of different carbohydrate supplements on pfs expression and AI-2 production. pfs expression (using pAB13) during growth in LB medium with 0.5% of various carbohydrates was measured every hour. The AI-2 activity of these cultures was also assayed by using V. harveyi BB170 AI-2 reporter strain on cell-free culture supernatants. The growth conditions (in columns A to I) were LB medium plus a 0.5% concentration of one of the following carbohydrates: A, arabinose; B, galactose; C, glucose; D, mannose; E, raffinose; F, melibiose; G, maltose; H, glycerol; and I, LB. The maximal AI-2 activity (bars) is depicted as a percentage relative to glucose, and pfs expression levels (datum points, ▪) are presented. In each case, they coincided at the same time point.

FIG. 3.

AI-2 does not regulate transcription of luxS. (A) The luxS::lacZ transcriptional fusion in Salmonella serovar Typhimurium CS132 (luxS::MudJ) was used to monitor luxS expression (β-galactosidase activity measured in Miller units [M.U.]). Mid-log-phase (4 h) luxS expression in CS132/pMS234 (pBAD18 containing luxS) was measured when AI-2 was produced during growth in LB plus 0.5% arabinose. The expression of luxS in CS132/pMS234 was determined in the absence of AI-2 during growth in LB without arabinose induction. luxS expression in CS132 without pMS234 and 14028 (negative control) are also shown. (B) Overexpression of LuxS does not result in increased AI-2 production over wild-type levels. AI-2 production levels by Salmonella serovar Typhimurium 14028 (wild type) and Salmonella serovar Typhimurium CS132 (luxS)/pMS234 are indicated by solid and dashed lines, respectively.

FIG. 4.

AI-2 does not regulate the transcription of pfs. The pfs-luxCDABE transcriptional fusion was used to monitor pfs expression in the presence or absence of AI-2 in Salmonella serovar Typhimurium 14028 (luxS⁺) or SS007 (ΔluxS) strain backgrounds. pfs expression (light production) was measured in the two strains during growth in LB with 0.5% glucose. The pfs expression profile in Salmonella serovar Typhimurium 14028/pAB13 is indicated by the solid line and in Salmonella serovar Typhimurium SS007 (luxS::T-POP)/pAB13 by the dashed line.