A former dumping ground in Ayer Hitam (Malaysia; GPS coordinates N03′00′12.1 E101′39′33′1) was selected as the sampling site. Soil specimens were collected at a subsurface depth of 10 cm using a clean 50 mL Falcon tube. Subsequently, soil samples were inoculated into KGm medium and enrichment was done for four cycles [18,19]. In short, soil samples were mixed thoroughly with 10 mL of KGm medium. Subsequent inoculation of soil suspension into fresh KGm medium was performed every 48 h and the inoculum was incubated with agitation. The KGm medium used was supplied with 50 mM 3-oxo-C6-HSL (Sigma-Aldrich, St. Louis, MO, USA). A diluted suspension was then spread on a plate of KGm agar supplemented with 3-oxo-C6-HSL and culture was spread on Luria-Bertani (LB) agar in the last enrichment cycle. The sole growth media used throughout the entire experiment are LB broth and LB agar. All strains were cultured aerobically in 28 °C with exception of E. coli [pSB401], which was cultured in 37 °C.

Short chain AHLs production was screened by using two different biosensors namely C. violaceum CV026 and E. coli [pSB401]. The C. violaceum CV026 biosensor responds to short chain AHLs by producing a purple pigmentation after an overnight incubation at 28 °C. AHL activity was then re-confirmed with a bioluminescence assay involving the use of the lux-based biosensor E. coli [pSB401]. Positive and negative controls for short chain AHLs detection were E. carotovora GS101 and E. carotovora PNP22, respectively. All reporter strains and both positive and negative control strains (Table 1) were gifts kindly provided by Professor Paul Williams (University of Nottingham, Nottingham, UK) (Table 1).

Isolates that showed QS activity were cultured in LB medium buffered to pH 5.5 with 50 mM (MOPS) in order to avoid alkaline-pH-driven deterioration of any AHLs. Log phase culture was subsequently extracted twice via the solvent extraction method. The solvent utilised in this extraction procedure was ethyl acetate supplemented with 0.1% glacial acetic acid [20]. The dried extracts were kept at −20 °C for storage purposes.

Measurement of bioluminescence were determined by using an Infinite M200 luminometer (Tecan, Männerdorf, Switzerland). Dessicated AHL extracts were resuspended with lux-based E. coli [pSB401] biosensor (OD₆₀₀ of 0.1) before dispensing into each well of 96-microtitre well plates in triplicate and the experiments were independently repeated three times. Bioluminescence activities were analyzed as luminescence (relative light unit) at the excitation wavelength of 495 nm (RLU/OD_495nm) over the course of 24 h. Reading of optical density and bioluminescence were automatically and simultaneously recorded at 60 min intervals [18,21].

AHLs produced by the QS isolate were studied using a High Resolution Tandem Liquid Chromatography Quadrupole Mass Spectrometry (LC-MS/MS; Agilent 1290 Infinity LC and Agilent 6490 Triple Quadrupole LC/MS systems, Agilent Technologies Inc., Santa Clara, CA, USA) as described previously [18,19]. The column used was Agilent Zorbax Rapid Resolution High Definition SB-C18 Threaded Column. Precursor ion scan mode was used for AHL detection where the product ion m/z value was fixed at m/z 102. Spectra analysis using the Agilent Mass Hunter software was performed as reported [18,19].

MALDI TOF MS was performed essentially as reported previously [18,19]. Bacterial samples were smeared very thinly on the MSP 96 target polished steel BC plate and MALDI matrix solution was added. A Microflex MALDI-TOF (Bruker Daltonik GmbH, Leipzig, Germany) bench-top mass spectrometer was used for identification of sample isolates [18,19,22] and for analysis of the MALDI-TOF mass spectra, we used the version 3.1 (Build 65) of the Bruker MALDI Biotyper Real Time Classification Classification software. To design the dendrogram for bacteria identification, we then used the MALDI Biotyper MSP creation method (Bruker Daltonics, Bremen, Germany) as reported previously [18,19,22].

To amplify the 16S rDNA genes, we used the reported method [23] with the following universal primers: 27F forward primers, 515F forward primers and 1525R reverse primers. Purification of genomic DNA, 16S rDNA PCR template and PCR condition was performed as reported elsewehere [23] and a phylognetic tree was conducted using MEGA5 [24].

KGm medium is a minimal medium which facilitates growth of non-fastidious bacteria and quorum quenching bacteria [18,19]. Soil samples were inoculated into sterile KGm medium and four cycles of enrichment was conducted. It was noted that KGm medium became turbid after two days. This indicated growth of microorganisms during each enrichment cycle. No growth was detected under our present experimental condition in the control KGm medium where AHL was not supplemented (data not shown). Bacterial isolates were streaked in several passages in order to obtain single, pure colonies and the bacteria were examined for both AHL production and degradation using the two types of biosensors. For screening of AHL activities, a number of AHL biosensors have been constructed, for instance C. violaceum CV026 was used for screening of short chain AHL production [21,25,26]. This AHL biosensor relies on the Cvi-based QS system whereby violacein is produced as a result of the presence of short chain AHLs. This makes CV026 a very convenient biosensor for initial and rapid screening of AHLs based on pigment formation [25]. Based on this preliminary screening, we detected that strain RB-25 produced short chain AHLs (Figure 1). However no significant degradation of AHLs was observed, suggesting strain RB-25 is a non-fastidous bacteria capable of thriving under nutrient limiting conditions.

We further studied the production of AHLs using bioluminescence analysis with E. coli [pSB401] biosensor cells which detect short chain AHLs [21, 26], confirming that strain RB-25 produced short chain AHLs (Figure 2). Production of short chain AHLs was seen as the increment of RLU/OD_495nm value observed over 24 h.

MALDI-TOF has been proven to be a rapid and reliable technique useful for bacteria biotyping [27–29]. In the present work, preliminary identification was done using MALDI-TOF where we could identified isolate RB-25 as S. fonticola with a value of 2.361, which indicated a reliable detection at both the genus and species level. To obtain confident identification at the species level, the MALDI-TOF scoring system value should be higher than 2.3 (Figure 3) [18,19,22].

To further confirm the identity of strain RB-25, we then used a molecular approach by analyzing its 16S rDNA gene nucleotide sequences using phylogenetics. Web-based comparison against the NCBI database and phylogenetic analysis of 16S rDNA sequences of strain RB-25 confirmed that this isolate is indeed S. fonticola (Figure 4). In this work, the Neighbor-Joining method [30] with bootstrap value from 1,000 replicates [31] and Maximum Composite Likelihood method [32] were used.

AHLs production has been documented in some species of Serratia where the most frequently detected AHL is C6-HSL [1]. Production of C4-HSL and C6-HSL had also been documented in Serratia sp. ATCC 39006 [33] and S. marcescens MG1 [34,35], whereas some Serratia spp. like S. plymuthica RVH1, S. marcescens SS-1 and S. proteamaculans B5a have been reported to produce other types of AHLs [8,36,37].

For S. fonticola, although its QS genes had been documented for rapid identification purposes [38], little is known about its AHL profile. Hence, in this work, we strived to identify if this isolate produced any detectable AHLs. Characterization of its AHLs profile is the first step in understanding the QS properties of S. fonticola and its AHL extract was assayed via tandem mass spectrometry platform that confirmed the presence of three AHLs, namely C4-HSL, C6-HSL and 3-oxo-C6 HSL (Figure 5).