Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Many of the quorum sensing (QS)-controlled genes, such as in P. aeruginosa and S. aureus, encode known virulence factors. These include the toxins elastase, alkaline protease, chitinases, cyanide, phenazines, lectins, and rhamnolipids by P. aeruginosa (Schuster et al. 2003; Wagner et al. 2003; Hentzer et al. 2003; Vasil 2003; Hassett et al. 1999). In S. aureus QS-regulated virulence factors include alpha, beta, gamma, and delta-hemolysin, triacylglycerol lipase precursor, glycerol ester hydrolase, hyaluronate lyase precursor, staphylococcal serine protease (V8 protease), cysteine protease precursor, cysteine protease, staphopain-cysteine proteinase, 1-phosphatidylinositol phosphodiesterase, zinc metalloproteinase aureolysin precursor, holing-like proteins, and capsular polysaccharide synthesis enzymes (Lowy 1998; Korem et al. 2005). Also controlled are genes involved in iron limitation in biofilms (Hentzer et al. 2005) as well as the adaptive response that limits the deleterious effects of the reduced pH associated with anaerobic growth conditions in biofilms (Beenken et al. 2004).

Given the fact that QS systems control many different unwanted bacterial phenotypes, including toxin production and biofilm formation, and given the fact that they function by means of extracellular signal, they are promising targets for developing novel antimicrobials (Fig. 1). In the following sections, we will review such compounds.

Fig. 1
figure 1

Inhibition of quorum sensing as a novel mode of therapy (Illustration by Mike Beshiri, Tufts University, Cummings School of Veterinary Medicine, Department of Biomedical Sciences, Division of Infectious Diseases, North Grafton, MA, USA)

Full size image

The importance of QS-dependent gene expression for bacterial virulence has been established in several animal models, as shown in the chapter In Vivo Studies: Inhibiting Biofilm-Associated Bacterial Infections Using SQIs, and in clinical studies (discussed in the chapter Clinical Wound Healing Using Signal Inhibitors).

Detection of Quorum Signals

Several strains have been developed for detecting QS activators and inhibitors (Table 1), based on fusing a QS-controlled promoter to a reporter gene. A characteristic of some of these strains is that they lack the ability to produce their own QS but are able to respond to exogenous QS with a visible phenotype, such as violacein pigment production in Chromobacterium violaceum CV026 (McClean et al. 1997; Adonizio et al. 2006), prodigiosin pigment production in Serratia marcescens ATCC 39006 (Glansdorp et al. 2004), light production in Vibrio harveyi D1 (Cao and Meighen 1993), and swarming in Serratia liquefaciens MG44 (Eberl et al. 1996). A bioluminescent strain of V. harveyi BB170 was developed for the bioassay of AI-2 (Bassler et al. 1994), and RNAIII-blaZ transcriptional fusion was developed in S. aureus for detecting agr as a marker for virulence (Novick et al. 1995).

Table 1 Representative quorum-sending (QS) bacteria and plasmids
Full size table

QS-mediated gene expression does not always result in a readily visible phenotype. One example of this is in the plant pathogen Agrobacterium tumefaciens, in which conjugation is mediated by QS (Fuqua and Winans 1996). Two A. tumefaciens biosensor strains have been constructed, A136 (pCF218)(pCF372) (Fuqua and Winans 1996) and KYC55 (pJZ372)(pJZ384)(pJZ410) (Zhu et al. 2003). These two strains overexpress traR, which is the luxR homolog. Although these strains are most sensitive (often in the sub-pmol concentration) to the cognate N-acyl homoserine lactone (AHL) [3-oxo-octanoyl homoserine lactone (3-oxo-C8 HSL)], they are also able to detect a wide range of other AHLs at less sensitivity (μmol to nmol range; Zhu et al. 1998, 2003). A representative example is shown in Fig. 2. Other AHL biosensor strains have also been constructed using other reporter genes, including gfp and lux, that have been fused to luxR homologs (Andersen et al. 2001; Swift et al. 1993). Positive controls for AHL detection include WT strains of the biosensor organism with intact luxI genes and AHL synthetic capability. Alternatively, AHL overproducers such as A. tumefaciens KYC6 (Fuqua and Winans 1996) and C. violaceum 31532 (McClean et al. 1997) can be used for the Agrobacterium and Chromobacterium bioassays, respectively. A number of AHLs are commercially available, or they can be produced via chemical synthesis (Eberhard and Schineller 2000).

Fig. 2
figure 2

A soft-agar overlay can be used for qualitative detection of N-acyl homoserine lactones (AHLs) using a reporter strain such as A. tumefaciens A136 (pCF 218) (pCF372) (Fuqua and Winans 1996). Positive control (a) involves testing with AHL-overproducing A. tumefaciens KYC6. Negative control (b) involves testing with non-AHL-producing A136 reporter strain. An AHL-producing environmental isolate is shown in (c). For comparison, a positive control cross-streaking plate is shown in (d)

Full size image

For increased sensitivity or resolution of different AHLs, biosensor strains have also been incorporated along with other analytical techniques such as high-performance liquid chromatography (Moré et al. 1996; Charlton et al. 2000a,b) and thin-layer chromatography (Shaw et al. 1997). Alternatively, violacein, a QS-regulated pigment in C. violaceum, can be extracted with acetone or ethanol, and measured with a spectrophotometer (Blosser and Gray 2000). There is a recent report of a mass spectrometry technique for AHL detection that rivals the picomolar detection limits of the best biosensors (Makemson et al. 2006).

RNAIII-blaZ transcriptional fusion was developed to detect agr activity, using the agr P3-blaZ fusion plasmid pRN6683 in lab strain RN6390 (Novick et al. 1995). Beta-lactamase activity can be measured by the addition of nitrocefin. A red color indicates activation of agr, and yellow indicates inhibition of agr. This assay can be carried out in microtiter plates (Fig. 3).

Fig. 3
figure 3

a An example of beta lactamase reporter assay in S. aureus. Four yellow columns on left reflect inhibition of RNAIII production by RIP. Four pink columns on right reflect RNAIII production in the presence of saline. b Spectrophotometric analysis (absorbance at 490/650 nm × 1000) of column 1 on far left (yellow, RIP+) and column 8 on far right (pink, RIP−)

Full size image

Detection of Quorum Signal Inhibitors

As these reporter systems are fused to a QS-controlled promoter, they become activated when the bacteria encounter exogenous signal molecules. Conversely, when the bacteria are challenged with quorum signal inhibitors (QSIs), the signal from the reporter systems is reduced. Hence, there is a positive “hit” when expression of the reporter is significantly reduced.

Several strategies can be employed to identify QSIs. A general strategy can involve taking a biosensor strain (Table 1) and exposing it to test compounds that would cause a loss of signal response. Alternatively, a wild-type strain containing a QS-regulated phenotype can be used as a biosensor (Table 1). Bacteria using Gfp or beta-lactamase-based screening systems, for instance, can be grown in liquid media in microtiter dishes in which many different compounds and/or concentrations can be probed at a time. An example of inhibition of S. aureus QS by RIP is shown in Fig. 3.

The Chromobacterium violaceum AHL bioassay (McClean et al. 1997) and pigmented Pseudomonas aureofaciens strain 30–84 (Wood and Pierson 1996), for example, have been used for detecting potential QSIs (McLean et al. 2004). Here, biological material (typically plant components or bacterial cultures) is placed in close proximity to either WT C. violaceum ATCC 12472 or P. aureofaciens 30–84. These QSI indicator strains can be incorporated into a soft-agar overlay (Fig. 4) or, alternatively, can be streaked in close proximity to the test material. Potential QSI activity is observed via a loss of pigmentation (purple violacein production in C. violaceum or loss of orange phenazine in P. aureofaciens).

Fig. 4
figure 4

Bioassay for quorum signal inhibition using C. violaceum ATCC 12472 (McLean et al. 2004). The cognate N-acyl homoserine lactone (AHL) for this strain is C6-HSL (McClean et al. 1997). Organisms such as Pseudomonas aeruginosa PAO1 can be used as a positive control (a) in that their AHLs, 3-oxo-C12-, and C4-HSL will compete with C6-HSL for binding on cviR (luxR homolog) of C. violaceum. The negative control, C. violaceum 12472, is shown in (b). Quorum-sensing inhibitor of aquifer isolate (McLean et al. 2005) is shown in (c)

Full size image

Potential antibiotics or other antimicrobial agents can also be detected by growth inhibition of the QSI strains. The AHL normally used by C. violaceum and P. aureofaciens is N-hexanoyl homoserine lactone (C6-HSL). Other AHLs will competitively bind to the LuxR homologs in these two organisms, CviR and PhzR (McClean et al. 1997; Chancey et al. 1999), but are not otherwise biologically active, thus inhibiting their AHL-regulated pigmentation. Although competitive inhibition by AHLs (other than C6-HSL) can be used as a positive control for this bioassay, one should also test samples with QS biosensor strains such as A. tumefaciens A136 or C. violaceum CV026. In this fashion, one can readily detect potential QSI and also determine whether it is due to AHL-mediated inhibition.

A QSI bioassay was recently modified to discern whether potential inhibitors target AHL synthesis (via LuxI) or AHL response (via LuxR) (Vattem, Bryant, and McLean, unpublished). As shown in Fig. 5, two strains are used: an AHL overproducer such as A. tumefaciens KYC6 or C. violaceum 31532, and an AHL biosensor such as A. tumefaciens A136 or C. violaceum CV026. A test compound is usually placed in an absorbent, sterile filter paper and the AHL overproducer and biosensor strains inoculated in varying proximity to the test material. To test for potential LuxI inhibition, the AHL overproducer is placed in close proximity to the test substance and the AHL biosensor placed distal. To test for LuxR inhibition, the locations of the AHL overproducer and biosensor strains are reversed. In either case, potential QSI activity results in a lowered signal from the AHL biosensor. Although the QSI bioassay at present is a qualitative assay, we are currently developing a quantitative QSI bioassay (Vattem, Bryant, and McLean, unpublished) analogous to a previously published violacein extraction QS assay (Blosser and Gray 2000).

Fig. 5
figure 5

Incorporation of two strains into a quorum-sensing inhibitor bioassay can distinguish whether inhibition is due to N-acyl homoserine lactone (AHL) response (luxR effect; a) or synthesis (luxI effect; b). Negative controls are at the top, and test samples are at the bottom. c Negative control for luxI effect. d Positive reaction for inhibition of AHL production due to green tea (Camellia sinensis) extract

Full size image

Usually, both growth (OD) and expression of the reporter system are monitored over time. One major drawback to this type of screening system is that compounds that either inhibit or slow growth inevitably reduce reporter expression and consequently may lead to scoring of false positives. Hence, growth of the bacterial screen has to be carefully monitored to ensure that the test compounds are not interfering with growth and thereby with general protein synthesis. To circumvent this problem, another type of screen, termed the QSI selector (QSIS), has been developed (Rasmussen et al. 2005).

In this system, the QS-controlled promoter is fused to a gene causing growth arrest when expressed. The screening bacterium does not produce any QS signals by itself, so in the absence of AHL molecules, growth is unrestrained. If the growth medium is supplemented with AHL molecules, the QS-controlled killing system becomes activated, leading to growth arrest (Fig. 6). Further addition of a QSI compound inhibits expression of the QS-controlled killing cassette, and the bacteria are allowed to grow (Fig. 6). Hence, the presence of a hit is indicated by growth. Furthermore, the bacteria express phenotypes that ease identification of growth, such as beta-galactosidase and bioluminescence (Fig. 7).

Fig. 6
figure 6

General layout of the quorum-sensing inhibitor (QSI) selector systems. a The screening bacteria are grown without N-acyl homoserine lactone (AHL). The LuxR homolog is not activated, hence there is no expression from the QS-controlled promoter (P QS), and, in turn, the killing is not expressed, and the cells survive. This situation is used when the bacteria are grown for purposes other than screening. b Exogenously added AHL molecules activate the P QS promoter, and the killing gene is thereby expressed, causing growth arrest of the bacteria. c A QSI blocks QS, and the killing gene is not expressed, allowing growth of the bacteria (Rasmussen et al. 2005)

Full size image
Fig. 7
figure 7

The quorum-sensing inhibitor selector (QSIS) 1 screening system in action. The screening bacteria are cast into an agar plate along with N-acyl homoserine lactone (AHL) and other cofactors. Wells are punched in the plate, into which test compounds are added. The test compounds diffuse into the agar, and where the concentration is appropriate, quorum sensing is blocked, allowing growth of the bacteria. Growth is indicated by a blue ring as the bacteria produce beta-galactosidase turning over X-gal in the plate

Full size image

Briefly, the bacteria are cast into agar along with signal molecules that will activate the killing cassette. Wells are punched in the agar, and compounds or mixtures to be tested are added to the wells. From the wells, compounds diffuse into the semisolid agar, establishing a concentration gradient with the highest concentration closest to the well. This enables the researchers to test the effect of numerous concentrations in just one assay. If the compound has no QSI activity, the killing system in the bacteria is active due to the AHLs present in the agar; hence, the bacteria are killed, no growth is observed, and a negative screen is observed. If the compound is toxic to the bacteria, no growth is observed. This is also scored as a negative outcome of the screen. Only if the test compound has nontoxic properties and exhibits QSI activity will the bacteria be rescued and a positive outcome of the screen be scored (Fig. 7) (Rasmussen et al. 2005).