The strains and plasmids used in this study are listed in Table 1, the primers in Table 2. The sahH gene was amplified from genomic DNA of the P. aeruginosa strain PAO1 using primers SF and SR, and the luxS gene was amplified from genomic DNA of E. coli W3110 with primers LF and LR. The restriction enzymes chosen in this part were BamH I and EcoR I for sahH and vector, EcoR I and Bgl II for luxS. And Bgl II is the isocaudamer of BamH I. To construct the plasmids pluxS and psahH, amplified products were inserted into the expression vector pGEX4T-1, using the standard method of preparing clones [24]. Generally, DNA amplification products were digested by the corresponding restriction enzymes and ligated with the vector. After sequencing, pluxS, psahH, and pGEX4T-1 were transformed into strain MDAI2 (luxS mutant strain) to construct strains M-L, M-S, and M-P, respectively.

All strains were precultured in a 2 × YT medium at 37 °C overnight with shaking at 220 rpm, and the concentration of ampicillin for selection was 100 μg/mL. These overnight cultures were inoculated into a fresh medium and cultivated at 37 °C with 250-rpm shaking. When the strains had reached the exponential phase, a final concentration of 0.1 mmol/L isopropyl-β-D-thiogalactopyranoside (IPTG) was added, and cultivation was continued for about 6 h.

After cultivation, the OD₆₀₀ of five strains was tested and total proteins were extracted. Briefly, about 2 × 10⁸ E. coli were collected and a corresponding volume of double-distilled water (ddH₂O) and a 5× sample of buffer (20% sodium dodecyl sulfate [SDS], 20% glycerol, 200 mM Tris base, pH 6.8, 0.001% bromophenol blue) was used to subsequently suspend the pellets. The samples were heated at 100 °C for 10 min and centrifuged for 1 min. Equal amounts of total proteins were electrophoresed in SDS–10% polyacrylamide gel electrophoresis. Western blotting was performed as previously described [24]. The antibodies used in this study were monoclonal anti-glutathione-S-transferase (anti-GST) and rabbit antimouse antibody with a green fluorescent group.

After IPTG induction and cultivation for 6 h, the total RNA of these strains was extracted. Culture volumes equivalent to 10 mL with an OD₆₀₀ of 1.0 were harvested by centrifugation. The pellet was suspended with TRIzol (Invitrogen, Carlsbad, CA, USA). The reagent and RNA extraction process followed the manufacturer's specifications. The resulting RNA was dissolved in diethyl pyrocarbonate-treated water and stored at −80 °C. The PrimerScript gDNA eraser RT reagent kit (Takara, Otsu, Shiga, Japan) was used to generate cDNA; about 1 μg of total RNA was used for each strain. To detect the target exogenous gene, cDNA of each strain was used as a template for polymerase chain reaction (PCR). Other compositions were forward/reverse primer (SF, SR, LF, LR), PrimeStar HS DNA Polymerase (Takara) and buffer, dNTP mixture, and ddH₂O. PCR conditions included an initial denaturation at 98 °C for 5 min, followed by 30 cycles amplification consisting of 98 °C for 15 s, 55 °C for 15 s, and 72 °C for 90 s, then final extension at 72 °C for 5 min. The amplified products were electrophoresed in 1% agarose gel.

To determinae whether the AMC-completed strain would restore the biofilm formation-related genes affected by luxS mutant, the expression of these genes was compared with the real-time PCR. The target genes and their primers are shown in Table 2. The amplification efficiency and template specificity of each primer pair were verified, then the amplification was performed with the following 15-μL reaction mixture: 7.5 μL 2× Thunderbird SYBRqPCR Mix (Toyobo, Osaka, Japan), 5 μL cDNA template, 1 μL PCR primers mix (10 μM), and 1.5 μL ddH₂O. PCR conditions included an initial denaturation at 98 °C for 5 min, followed by 40 cycles amplification of 98 °C for 15 s, 55 °C for 15 s, and 72 °C for 30 s. To check DNA contamination, the production of RT-PCR step 1 (without reverse transcriptase) served as a negative control. The rpoA gene was used in this study as the normalising gene for all reactions [26]. Applied Biosystems (Carlsbad, CA, USA) 7900HT Fast Real Time PCR System was used for this test, and fold changes of the expression levels were calculated by SDS Software v2.3 with RQ Study 1.2 (Applied Biosystems). All the assays were conducted with each sample in triplicate. Here, the Kruskal-Wallis test was used to analyse the gene expression difference among the five strains. A P-value less than 0.05 was considered statistically significant.

To test and verify our hypothesis, the QS system and AMC must be independent. Previous investigators have collected supernatant that had cultivated a wild-type strain for some time, believing it to contain AI-2 [27,28]. They then used it as a conditioning medium in which to cultivate the luxS mutant strain. In a sense, this method could make the QS system and AMC independent, but it allowed the other substance in the supernatant—secreted by the bacteria—to influence the results. This study, aiming to restore the obstructed AMC and simultaneously keep the defective QS mechanism, transformed the sahH into an E. coli luxS mutant strain. Consequently, we could now check whether the AMC or the QS system took responsibility for changes caused by the luxS mutant.

The RT-PCR and Western blot were performed to prove the expression of target genes in these five strains. As shown in Figure 2, the bands detectable by anti-GST represented the fusion proteins GST-LuxS and GST-SahH, whose molecular weights were about 44.9 kDa and 77.7 kDa, respectively. For luxS RT-PCR, the primer pairs LF and LR could spot both W3110 and M-L, whereas sahH was detected only in M-S. Moreover, the band amplified from M-L is more remarkable than W3110, which amplification should be attributed to the overexpression of pluxS.

To determine whether AMC regulates gene expression of the luxS mutant strain, a real-time PCR was conducted. Biofilm formation is one of the main causative factors in infectious diseases. And as described before, biofilm formation was reported to be regulated by quorum sensing system. Genes such as fliA, fliC, motA, and motB, which were responsible for flagellar biosynthesis and rotation, also played an important role in the initial stage of the biofilm formation process [29–31]. They were also reported to have different expression levels between the wild-type and luxS mutant strain [14,15], so here these four genes were tested [32–34].

The results of real-time PCR are shown in Figure 3. In general, M-S, relative to M-L, could not restore the gene expression of the luxS mutant strain. In practical terms, all the detected genes showed the trend that the wild-type strain expressed more than the luxS defective strain. They increased about 2.8-fold, 40.8-fold, 2.1-fold, and 5.0-fold for fliA, fliC, motA, and motB, respectively. But M-S, whose AMC disorder may be partly repaired, meanwhile the AI-2 producing process obstructed, had an expression level similar to the luxS mutant strain, MDAI2, and their relative expression folds were 0.73 for fliA, 2.3 for fliC, and 1.1 for motA and motB. That means SahH failed to restore the changes caused by luxS mutant. Under the same experimental conditions, the expression of target genes in the M-L strain was recovered or even up-regulated compared with the wild-type strain. These results showed that there was no significant restoration by expression of SahH which was intended to complete the AMC, thus the luxS-mediated QS system impacted the biofilm-related gene expression.

This study also showed that, the genes, downregulated by the luxS mutant, were even strongly upregulated when LuxS was overexpressed. This demonstrated that the mutant strain with LuxS expression may complete the QS system and cause AI-2 secretion to recover the expression of the target genes. And theoretically, overexpressed LuxS synthetized more active AI-2, so these target genes were upregultaed much more than the wild-type strain. But under the same experimental conditions, the mutant strain with SahH expression, which might recover the AMC system without AI-2 secretion, kept the expression of target genes at a low level (Figure 3), which means that, without integrated QS mediating, the phenotype could not be recovered, although its role in metabolism may be intact.