Direct quantification of N-(3-oxo-hexanoyl)-l-homoserine lactone in culture supernatant using a whole-cell bioreporter

Abstract

The autoinducer N-(3-oxo-hexanoyl)-l-homoserine lactone (3-oxo-C6-HSL) plays a significant role in the quorum-sensing system of the marine bacterium Vibrio fischeri. Upon forming a transcriptional activation complex with LuxR, 3-oxo-C6-HSL induces transcription of the luxICDABEG operon, leading to the increased production of both the 3-oxo-C6-HSL synthase (LuxI) and the bioluminescent proteins. In order to quantitatively analyze this regulatory mechanism, a novel approach was developed to measure 3-oxo-C6-HSL concentrations in V. fischeri cell culture supernatant. A bioluminescent strain of Escherichia coli that responds to 3-oxo-C6-HSL was used as a bioreporter. Although a linear response of the bioreporter to exogenously added synthetic 3-oxo-C6-HSL was found over several orders of magnitude, we show that bioreporter performance was dramatically impacted by variations in the supernatants using samples from a V. fischeri LuxI⁻ strain. However, when maintained in the same supernatant background, the normalized peak bioluminescence maintained a linear response to 3-oxo-C6-HSL concentrations. Therefore, a standard additions technique was developed in which a known concentration of 3-oxo-C6-HSL was added to supernatant samples from wild-type V. fischeri cultures, and the incremental increase of the normalized peak bioluminescence relative to the untreated sample was determined. The concentration of 3-oxo-C6-HSL in the supernatant of the unknown sample was then quantified from the slope of the response between the normalized bioluminescent peaks with and without the addition of 3-oxo-C6-HSL. Advantages of this method are that it is rapid, does not require concentration or extraction, uses a small sample volume (ca. 2 ml), and accounts for effects caused by the composition of the supernatant. Furthermore, the findings can be broadly applicable to other bioreporter systems involving variable background conditions.

Introduction

Many Gram-negative bacteria use acylated homoserine lactone molecules to sense and respond to their own cell density in a process known as quorum sensing (Fuqua et al., 1994). This process, first described for the marine bacterium Vibrio fischeri, has been extensively studied as a model of density-dependent gene regulation (Nealson and Hastings, 1979). The autoinducer N-(3-oxo-hexanoyl)-l-homoserine lactone (3-oxo-C6-HSL) was identified as one of the quorum signaling molecules for V. fischeri (Eberhard et al., 1981). This molecule is the product of the LuxI autoinducer synthase, which catalyzes the reaction between S-adenosylmethionine and acylated-acyl carrier proteins (Acyl-ACPs) to produce 3-oxo-C6-HSL (Schaefer et al., 1996b, Watson et al., 2002). The luxI gene resides in the rightward portion of the bidirectional lux operon (luxICDABEG) containing both luxI and the genes encoding the proteins involved in bioluminescence (Kaplan and Greenberg, 1985, Shadel et al., 1990). The luxR gene, which encodes the 3-oxo-C6-HSL-dependent response regulator, is encoded in the left operon (Egland and Greenberg, 1999). The luxA and luxB genes encode the α and β subunits of the luciferase enzyme, which catalyzes the reaction among reduced flavomononucleotide (FMNH₂), fatty aldehyde and oxygen to produce oxidized FMN, aliphatic acid, water and blue green light (Eberhard et al., 1981, Lupp et al., 2003). The luxC, luxD, and luxE genes encode the aliphatic acid reductase that regenerates the substrate, and luxG encodes an FMNH₂ reductase (Dunlap, 1999, Eberhard et al., 1981, Lupp et al., 2003). When the level of the freely diffusible 3-oxo-C6-HSL reaches a threshold concentration, this molecule binds to the N-terminal domain of the LuxR protein and activates the C-terminal domain of the protein to bind with lux box (Hanzelka and Greenberg, 1994, Schaefer et al., 1996a). The lux box is a 20-bp region of DNA in the promoter region having dyad symmetry and is centered upstream of the transcriptional start site of the luxICDABEG operon (Devine et al., 1989, Fuqua et al., 1996, Shadel et al., 1990). The binding of the 3-oxo-C6-HSL-LuxR complex to the lux box facilitates the binding between RNA polymerase and the promoter region of luxICDABEG, leading to increased transcription and the production of bioluminescence (Hanzelka and Greenberg, 1994, Egland and Greenberg, 1999).

A second autoinducer, N-octanoyl-l-homoserine lactone (C8-HSL), has been reported to influence bioluminescence in V. fischeri by two mechanisms. In one mechanism, thought to be most relevant at low cell densities, C8-HSL forms a complex with LuxR protein and induces bioluminescence via interaction with the lux box (Callahan and Dunlap, 2000, Miyamoto et al., 2000, Kuo et al., 1996). C8-HSL may also regulate the activity of LuxR through a pathway that is analogous to the luxLM-luxN-LuxU-luxO-luxR system in Vibrio harveyi (Miyamoto et al., 2003).

A number of methods have been described to assay bacteria for HSL production or to screen environmental samples for the presence of HSL. Most of these methods use various bioreporters because of their greater sensitivity compared to conventional analytical methods (Brelles-Marino and Bedmar, 2001). Bioassays based on reporter systems such as LuxCDABE, (e.g., Boettcher and Ruby, 1995, Winson et al., 1998), green fluorescent protein (Andersen et al., 2001), violacein production (Blosser and Gray, 2000, Ravn et al., 2001), and lacZ-β-galactosidase (e.g., Shaw et al., 1997) have been used. Bioreporters can be selected to respond specifically to a particular HSL (e.g., Boettcher and Ruby, 1995) or to a broad range of homologies (e.g., Shaw et al., 1997). HSL concentrations may be quantified using a calibration curve with standards of known concentration.

Prior to detection, the HSL is often extracted from the samples using organic solvents (e.g., dichloromethane, ethyl acetate or chloroform). The extraction process improves the sensitivity of detection by removing the background effects caused by the complex sample matrix. Following extraction, high pressure liquid chromatography (e.g., Boettcher and Ruby, 1995) or thin-layer chromatography (e.g., Shaw et al., 1997) is frequently used to separate various homologies and to further purify the HSL from interfering impurities that may have been coextracted by the organic solvent. A recent study has used solid phase extraction as an alternative means of separation and sample purification for environmental samples (Schupp et al., 2005).

In order to quantitatively study the quorum sensing mechanism in V. fischeri, it is necessary to evaluate the concentration of the signaling molecule present at various stages in the growth process. Here we present a simplified method of quantifying concentrations of a known HSL that may be particularly advantageous when a series of samples must be taken. A bioreporter strain with a high specificity to 3-oxo-C6-HSL was employed. A linear response between the peak biomass-normalized bioluminescence and 3-oxo-C6-HSL was found in the physiologically relevant concentration range of 3 to 390 nM. Further, a standard additions method was developed that effectively circumvented artifacts related to the sample supernatant. Although specifically applied in this case, the methodology presented here may be adaptable to other systems using bioreporters for the detection of target analytes.