We successfully constructed a mutation strain ΔluxS, and complemented strain C-luxS. We found that the absence of LuxS does not affect the growth of SEZ under 37°C conditions ( Figure 1A ). Then, we used the Vibrio harveyi BB170 as a reporter strain that responds to AI-2 by inducing luminescence ( Figure 1B ). The ΔluxS strain exhibited the lowest luminescence at 2 hours, indicating the functionality of the AI-2/LuxS quorum sensing system in SEZ. We compared the ability of SEZ, ΔluxS, and the complemented strain C-luxS to adhere to PK-15 cells. These results show no significant differences in adhesion among the three groups ( Figure 1C ). Furthermore, hemolysis assays revealed no differences among the three groups ( Figure 1D ).

We investigated whether LuxS has an impact on hyaluronic acid (HA) production. We first assessed hydrophobicity in SEZ strains since HA is highly hydrophilic. The results indicated that WT SEZ, ΔluxS, and C-luxS strains showed no difference in hydrophobicity and the three strains exhibited very low hydrophobicity compared to the SEZ capsule-deficient strain ΔhasB ( Figure 2A ). Lastly, the HA production in the three groups of strains at 10 hours showed no significant differences ( Figure 2B ).

To investigate whether LuxS influences biofilm formation in SEZ, we initially conducted crystal violet staining using microtiter plates ( Figure 2C ). The ΔluxS mutant of SEZ exhibited a significantly increased biomass compared to wild-type strain. In the case of the complemented C-luxS strain, the biofilm content showed a reduction but still exhibited a significant difference compared to SEZ.

We further visualized biofilm formation through SEM and CLSM ( Figure 3A ). The SEM results allowed observation of the biofilm adhered to the glass surface, revealing that the biofilm formed by the ΔluxS mutant was comparatively denser than SEZ, while SEZ exhibited a more loosely formed biofilm.

Then, we visualized SEZ, ΔluxS, and C-luxS biofilms using CLSM, converting them into z-stacks for analysis. The biofilms were characterized for structural features, thickness, and bio-volume ( Figure 3B, C ). The ΔluxS strain developed a thicker mature biofilm. The application of Comstat2 for biomass and average thickness determination revealed that ΔluxS was significantly higher than the wild-type SEZ strain. Collectively, these results confirm that the LuxS regulates biofilm formation of SEZ.

Previous research has indicated that the quorum sensing system plays a significant role in immune evasion (Cao et al., 2020). To further investigate this, we conducted infection experiments using mouse macrophage RAW264.7 cells. Flow cytometry analysis revealed a significant increase in apoptosis levels in the ΔluxS mutant strain compared to the SEZ-infected cells ( Figures 4A, B ). The ΔluxS induce higher TNF-α secretion compared to SEZ, however the complemented C-luxS strain shows no difference compared to the ΔluxS ( Figure 4B ). RT-PCR analysis of pro-inflammatory cytokines, including IL-1β, IL-6, IL-10, IL-18, TNF-α, and the transcription factor NF-kB, demonstrated higher transcription levels in ΔluxS ( Figure 4C, D ).

To investigate whether the LuxS affects the virulence of SEZ, we infected three groups of mice (n=10 for each group) via the intraperitoneal route. The final mortality of mice infected with SEZ and C-luxS was 80%, while ΔluxS group was 70% ( Figure 5A ). Then, we evaluated the bacterial burden in different organs at 24 hours post-infection. Surprisingly, the ΔluxS strain exhibited higher bacterial loads in the spleen compared to the SEZ wild type. There were no significant differences in bacterial loads between the ΔluxS strain and the SEZ wild type in the lungs, livers and kidneys ( Figure 5B ). The histopathological with H&E showed notable atrophy in the splenic nodules of the ΔluxS and C-luxS groups ( Figure 6 ). The differentiation between white and red pulp was not distinct, particularly in the ΔluxS group, which showed more severe effects. In addition, the SEZ, ΔluxS, and C-luxS groups exhibited various levels of lymphocyte deformation and necrosis. Around these necrotic areas and within the red pulp, there was a significant infiltration of inflammatory cells, mainly neutrophils, with the ΔluxS group experiencing the most severe infiltration. The data suggest that while LuxS might not significantly influence the mortality rates in mice, the tissue-specific phenomena observed highlight a complex and nuanced role for LuxS in the pathogenesis and host-pathogen of SEZ.

A comparative transcriptomic analysis was conducted between the SEZ wild type and the ΔluxS mutant strain, providing a comprehensive insight into the role of the luxS gene in SEZ. Among a total of 242 genes displaying differential expression in the ΔluxS strain compared to SEZ, 158 genes were upregulated, while 84 genes were downregulated, with an adjusted p-value (false-discovery rate [FDR]) of ≤0.05 ( Figure 7A , Supplementary Table 2 ).

Gene ontology (GO) annotation analysis (Ashburner et al., 2000) revealed a downregulation of many genes associated with metabolic and cellular processes, while most of the downregulated genes were linked to biosynthesis, cellular localization, and structural functions in the ΔluxS mutant strain when compared to the SEZ ( Figure 7B ).

To provide a clearer understanding the differentially expressed genes in the ΔluxS mutant, we constructed a network ( Figure 8 ). Cluster analysis showed that the majority of genes formed clusters related to carbohydrate metabolism, transport, iron-sulfur cluster and selenocompound metabolism, as well as the maintenance of DNA repeat elements and antiviral defense, all of which were upregulated. Downregulated genes were clustered into modules associated with ribonucleotide metabolism, peptidoglycan biosynthesis, DNA replication, and repair.

The bacterial strains and plasmids used in this study are detailed in ( Supplementary Table 1 ). The SEZ ATCC35246 bacterial strain, originally isolated from an infected pig in Sichuan (Ma et al., 2011), was preserved in our laboratory for research purposes. SEZ strain were grown in Todd–Hewitt broth (THB) at 37°C. Vibrio harveyi strains BB170 were used as biosensors of the AI-2 signal and as the positive control for AI-2 bioassay, respectively. The V. harveyi strains were cultured in autoinducer bioassay medium (AB) as described (Bassler et al., 1993). The RAW264.7 macrophages and PK-15 cells were also maintained in our laboratory. These cell lines were cultured in DMEM (Gibco, USA), supplemented with 10% fetal bovine serum (FBS), and maintained in a humidified 5% CO₂ incubator at 37°C.

The deletion of the luxS gene in SEZ was conducted following a previously established protocol (Takamatsu et al., 2001; Wei et al., 2012b). In sum, we initiated the process by using PCR amplification to generate upstream and downstream fragments of the luxS gene with specific primers ( Supplementary Table 1 ). Then, these PCR products were ligated into the temperature-sensitive suicide vector, pSET-4s, using appropriate restriction enzymes. The resulting plasmid, denoted as pset4s-luxS, was introduced into SEZ through electroporation.

Bacterial cultures were under 28°C in the presence of 100 μg/mL spectinomycin (Spc). When the cultures reached the mid-logarithmic growth phase, bacteria were sub-cultured in THB supplemented with 100μg/mL Spc and further incubated at 28°C until they reached the early logarithmic phase. The culture was then shifted to 37°C for a duration of 4 hours. Subsequently, cells were plated on TSA plates and kept at 28°C. Colonies that did not exhibit resistance to Spc at 37°C were selected, indicating the loss of spectinomycin resistance conferred by the plasmid. Confirmation of the double-crossover homologous recombination mutant, designated as ΔluxS, was carried out using PCR analysis with primers luxSL-F and luxSR-R.

As for the complemented strain C-luxS, we amplified fragments that included the promoter region and the luxS gene, carried out the necessary digestion, and incorporated them into the pSET2 vector. This resulting construct was then introduced into the ΔluxS strain through electroporation to achieve complementation.

To assess the growth characteristics of SEZ, the ΔluxS mutant, and C-luxS, a single colony from each strain was cultured overnight at 37°C and adjusted to an OD600 of 0.3 using fresh THB medium. Subsequently, the cells were inoculated into fresh THB medium at a 1:1000 ratio and incubated at 37°C with shaking at 180 rpm for 24 hours. Cell densities were measured every 2 hours at OD₆₀₀.

For AI-2 detection, established protocols were followed (Buck et al., 2009). The SEZ strain, ΔluxS, and the complemented C-luxS were cultured in THB medium until they reached an optical density of 0.3 at 600 nm. To obtain cell-free culture fluid (CF), the supernatant was filtered through a 0.22-μm-pore-size filter (Millipore, Bedford, MA). The reporter strain V. harveyi BB170 was diluted at a ratio of 1:5,000 in autoinducer bioassay (AB) medium, and AB medium was added to the diluted BB170 culture at a 1:10 ratio (vol/vol). This mixture was then incubated at 28°C. Aliquots of 100μL were dispensed into white, flat-bottomed, 96-well plates for AI-2 activity detection. CFs from BB170 and AB medium served as positive and negative controls, respectively. Luminescence measurements were conducted using a microplate reader.

To investigate the impact of LuxS on SEZ adhesion to cells, we conducted the following procedure (Tang et al., 2017). Cultured bacterial strains (SEZ, ΔluxS, C-luxS) were harvested, washed with sterile PBS, and then added to PK15 cells cultured in a 12-well dish at a multiplicity of infection (MOI) of 10. The cell-bacteria mixtures were co-incubated for 2 hours. Following incubation, non-adherent bacteria were removed by rinsing the cells with PBS. Subsequently, the cells were resuspended in sterile water, appropriately diluted, and plated onto tryptic soy agar (TSA) plates for colony counting.

To assess hemolysin activity in the ΔluxS mutant, we conducted a hemolytic assay following a previously described protocol (Jusuf et al., 2021). In brief, bacterial cells were washed and resuspended in PBS. Approximately 10⁶ bacterial cells in 100 μL were added to a 96-well microtiter plate. Fresh pig blood was processed to remove serum and washed with physiological saline 2-3 times, then diluted to a 5% concentration, with 100 μL added to each well. The plate was incubated for 2 hours at 37°C. After incubation, the supernatant was collected following centrifugation, and the absorbance was measured at 420 nm. A 0.1% Triton X solution was used as a positive control to calculate the percentage of bacterial hemolysis activity.

To assess the hydrophobic properties of the bacteria, an equal volume of bacterial suspension was combined with xylene and mixed for 2 minutes. The mixture was then left to stand for 30 minutes, allowing the phases to separate, after which the OD₆₀₀ of the aqueous phase was measured. The hydrophobicity percentage (H%) was calculated using the formula H% = [(H0 − H)/H0] × 100%, where H0 represents the initial absorbance before mixing with xylene, and H represents the absorbance after xylene extraction (Del Re et al., 2000). The SEZ capsule-deficient strain ΔhasB was used as the control group in this assay.

The cell-associated hyaluronic acid (HA) of SEZ was measured using a procedure inspired by existing methods, with some modifications (Velineni and Timoney, 2015). In brief, 10 ml of SEZ culture in THB, grown at 37°C, was centrifuged at 8000 g for 10 minutes. The resulting pellet was gently washed with PBS and then resuspended in 2 ml of distilled water. This suspension was transferred to a capped glass tube, to which 5 ml of chloroform was added. The mixture was then vigorously mixed for 1 hour. Following this, the sample was centrifuged at 8,500 g for 10 minutes, and the aqueous phase was carefully transferred to a new container. To this aqueous phase, two volumes of cetyltrimethylammonium bromide (CTAB) buffer (2.5 g L⁻¹) were added, and the mixture was gently stirred. The reaction was allowed to proceed for 10 minutes at room temperature, after which the OD₆₀₀ was measured (Xie et al., 2019). The concentration of HA in the sample was determined based on the OD600 value and a standard curve. This standard curve was established using a stock solution of HA prepared with standard HA (Sigma, Shanghai, China).

Biofilm formation was quantified by the crystal violet assay (Bonifait et al., 2008; Wang et al., 2011). Bacterial cultures were grown until they reached an optical density (OD600) of approximately 0.3. Then, 10 μL of the bacterial suspension was added to 96-well flat-bottomed plastic culture plates containing 90 μL of THB supplemented with 10 mg/mL of fibrinogen. The plates were incubated at 37°C for 24 hours, with each bacterial strain tested in triplicate. The culture medium was discarded, and the wells were gently washed three times with sterile PBS to remove loosely adherent cells. The remaining bacteria attached to the wells were fixed with methanol for 30 minutes and air-dried. A 1% crystal violet solution was added to each well, and the plates were stained for 10 minutes at room temperature. Excess crystal violet was removed and the wells were rinsed multiple times with distilled water. The biofilm-associated crystal violet was then solubilized with 100 μL of 95% (v/v) ethanol, and the optical density at 570 nm was measured using a microplate reader.

For scanning electron microscopy (SEM), SEZ was inoculated into a 24-well culture plate containing sterile glass slides and incubated at 37°C for 48h. The slides were then removed, washed three times with sterile PBS, and fixed in a solution of 5% glutaraldehyde, 4.4% formaldehyde, and 2.75% picric acid in 0.05% sodium cacodylate buffer for at least 1 h. The sample underwent sequential dehydration with 25%, 50%, 70%, 80%, and 95% ethanol. SEM was performed as previously described (Petruzzi et al., 2017), using a scanning electron microscope.

To study biofilms using confocal laser scanning microscopy (CLSM), each biofilm was stained with 0.3% SYTO-9 (obtained from Sigma-Aldrich, China) and incubated in darkness for 15 minutes for fluorescent labeling. The analysis of the biofilm images was conducted using Zeiss confocal software, enabling the acquisition of z-stack images. The thickness and biovolume of the biofilms were quantified using the COMSTAT2 software available at www.comstat.dk.

In the experiment with macrophages, we used RAW264.7 cells and followed a method that was a bit different from what others have done before. We put the macrophages in six-well plates. After that, we added SEZ bacteria to 12-well plates with a MOI of 1, and co-incubate for 12 h. For the negative control, we just added 100 μL of PBS.

To evaluate apoptosis, the treated cells were collected, washed twice with ice-cold PBS, and then resuspended in 1× Annexin V binding buffer at a concentration of 1×10⁶ cells/mL. Apoptosis detection was carried out using the Annexin V-FITC Apoptosis Detection Kit (Sigma, Shanghai, China), strictly adhering to the manufacturer’s instructions. The stained cells were analyzed via flow cytometry within 1 hour of staining.

The cell supernatant was collected by centrifugation at 3,000g, and the secretion level of TNF-α was determined using an ELISA assay kit (Enzyme-linked Biotechnology, Shanghai, China).

Quantitative real-time reverse-transcription polymerase chain reaction (qRT-PCR) was used to assess the transcription levels of pro-inflammatory cytokines (IL-1β, IL-6, IL-10, IL-18, TNF-α) and NF-κB. Briefly, post-infection, the cells were washed twice with sterile PBS. Total RNA was extracted using TRIzol reagent (Sigma, Shanghai, China), followed by cDNA synthesis using the Takara reverse transcription kit. Real-time quantitative PCR was performed using SYBR Green Premix (Takara, Beijing, China) on a LightCycler 480 fast real-time cycler. Fold changes were calculated using the 2^-ΔΔCt formula (Livak and Methods, 2000). The data are presented as the means ± standard deviation (SD) of triplicate reactions for each gene transcript, and the primer sequences are provided in ( Supplementary Table 1 ).

Four-to-six-week-old BALB/c mice were obtained from Beijing Vetonlitech Experimental Animal Technology Co. Bacterial suspensions were prepared by diluting the bacteria in PBS, 1 × 10⁶ CFU of bacteria were administered via intraperitoneal injection.

Tissue bacterial burden assays were conducted as previously described (Wei et al., 2012a). Briefly, mice were euthanized 24 hours after infection, and their spleens, livers, kidneys, and lungs were collected and weighed. Then, the tissues were homogenized and diluted in sterile saline (g/mL). The homogenized tissue samples were plated onto TSA plates for culturing, and the colonies were counted.

Briefly, the mice were infected with 1 × 10⁶ CFU of bacteria. At 24 h post-infection, the infected mice were euthanized, and the spleen were collected and placed into 10% neutral buffered formalin. After fixation, the tissues were dehydrated and embedded in paraffin, and 4-mm sections were taken and stained with hematoxylin and eosin (H&E) for histological evaluation.

The ΔluxS mutant and wild-type strains were incubated in THB on a shaker at 37°C for 10 hours. After incubation, the cells were harvested by centrifugation at 6,000 g for 10 minutes at 4°C. Total RNA samples were extracted using the TRIzol method. RNA sample quality was assessed using Thermo NanoDrop One and Agilent 4200 Tape Station. Once qualified, the ribosomal RNA (rRNA) was removed using the Epicentre Ribo-Zero rRNA Removal Kit.

Library preparation followed a standardized protocol using the NEBNext Ultra II Directional RNA Library Prep Kit for Illumina, involving mRNA fragmentation, cDNA synthesis, double-stranded cDNA generation, purification, enzymatic degradation, end repair, adapter ligation, fragment size selection, and final PCR amplification. Libraries meeting quality control criteria underwent PE150 sequencing on the Illumina platform. Raw sequencing data were processed into FASTQ files, including read sequences and corresponding quality scores. The data sets generated for this study can be found in the NCBI Sequence Read Archive (SRA) repository under BioProject accession number PRJNA1048089.

Before conducting the differential expression analysis, we conducted quality control on the sequencing reads and trimmed the adapter sequences. The sequencing reads from each library were then used for differential expression analysis with the read counts. As a reference genome sequence, we used the complete genome sequence of Streptococcus equi subsp. zooepidemicus (ATCC_35246) and its genome annotation from the NCBI nucleotide database (accession number: NC_017582.1).

To compare gene expression between SEZ and ΔluxS. Genes with absolute log2 fold change values of ≥1 and false discovery rate (FDR) P values of ≤0.05 were considered differentially expressed. For further analysis of these differentially expressed genes, we utilized the STRING Consortium 2020, which provides functional enrichment analysis of protein-protein interaction networks through the STRING mapper tool (Szklarczyk et al., 2023).

The data in this study are expressed as means ± standard deviation (SD). For the statistical analysis, we used two-way ANOVA, conducted with Graph Pad Prism 9.0 (GraphPad Software Inc., USA). We marked significant differences with asterisks: * for p < 0.05, ** for p < 0.01, and *** for p < 0.001. Different letters represent significant differences within each strain, with p < 0.05 as the threshold for significance. All data are presented as means ± standard error of the mean (SEM), and the results are compiled from a minimum of three independent experiments.