The quorum sensing system luxS gene contributes to the environmental tness of Streptococcus suis

Background: Streptococcus suis type 2 (SS2) is an important zoonotic pathogen. We have previously reported the structure of LuxS protein and found that the luxS gene is closely related to biolm, virulence gene expression and drug resistance of SS2. However, the mechanism of luxS mediated SS2 stress response is unclear. Therefore, this experiment performed stress response to luxS mutant (ΔluxS) and complement strain (CΔluxS), overexpression strain (luxS+) and wild-type SS2 strain HA9801, and analyzed the differential phenotypes in combination with transcriptome data. Results: The results indicate that the luxS gene causes a wide range of phenotypic changes, including chain length. RNA sequencing identied 278 luxS-regulated genes, of which 179 were up-regulated and 99 were down-regulated. Differential genes focus on bacterial growth, stress response, metabolic mechanisms and drug tolerance. Multiple mitotic genes were down-regulated; while the ABC transporter system genes, cobalamin /Fe 3+ -iron carrier ABC transporter ATPase and oxidative stress regulators were up-regulated. The inactivation of the luxS gene caused a signicant reduction in the growth and survival in the acid and iron stress environments. However, the mutant strain ΔluxS showed increased antioxidant activity to H 2 O 2 . Conclusions: The luxS gene in SS2 appears to play roles in iron metabolism and protective responses to acidic and oxidative environmental conditions.


Background
Streptococcus suis (SS) is a major pathogen in pigs, and it is also a zoonotic agent of a variety of diseases for swine and humans. Among the thirty-three serotypes of SS (serotypes 1-32 and serotype 1/2), Streptococcus suis serotype 2 (SS2) are generally considered to be the most virulent serotypes found so far. It can cause a variety of life-threatening infections, including meningitis, arthritis and sepsis [1,2]. The LuxS/AI-2 quorum sensing (QS) system is considered a process by which bacteria communicate using autoinducers 2 (AI-2). It is widespread in Gram-positive and Gram-negative bacteria.
LuxS-mediated QS mechanism is based on the production of AI-2 that regulates various important biological properties in different bacteria. In our previous study, we have shown that the loss of luxS gene can reduce the bio lm formation ability, hemolytic activity, adhesion to human laryngeal carcinoma (HEp-2) cell line and virulence genes transcription, and the luxS gene also related to drug-resistant e ux gene expression of SS [3][4][5][6]. Furthermore, by analyzing the crystal structure of LuxS protein, it was found that Zn 2 + and two important amino acids, Phe80 and His87, can affect the catalytic activity of S-ribose homocysteine (luxS) [7]. The ability of bacteria to resist environmental stress is one of the important factors for their survival. Studies have shown that LuxS protein is involved in regulating changes in bacterial resistance to stress [8]. However, the relevant mechanisms remain unclear. The quorum sensing system is an important protective mechanism for bacteria to adapt to the environment [9]. Previous studies have shown that luxS and AI-2 involved in bacterial regulation of a series of important stress responses, including heat shock, anti-gamma radiation, H 2 O 2 and other oxidative stress responses [10][11][12][13][14]. With the deepening of luxS research, researchers found that luxS showed different phenotypes in some bacteria after mutation. For example, the luxS mutation in Helicobacter pylori leads to a decrease in the expression of agella transcription regulator hA and a decrease in motility [15]. Compared with the wild-type strain, the luxS mutant of Escherichia coli has increased survival in the environment with pH < 3.2 [13]. In addition, the ABC transporter gene of radiation Deinococcus radiodrans luxS mutant is up-regulated under oxidative stress, and ABC transporter participates in the adaptation mechanism of stress environment, transporting damaged nucleotides and polypeptides to vitro. And it's proved that the luxS mutant of Borrelia burgdorferi has reduced pathogenicity in mice [16]. Moreover, Liu et al. found that overexpression of luxS gene could improve the stress resistance of Lactobacillus paraplantarum L-ZS9 [8]. We have found luxS gene as an important regulator in many aspects. However, it is still unknown that the impact of the luxS genes on SS growth and stress responses. In order to understand the luxS gene more fully, in this study, we wanted to know the differences of wild-type strain, ΔluxS, CΔluxS and luxS + in growth characteristics and stress responses.

Cell morphology
Through SEM observations, mutant strain ΔluxS tended to aggregate into clusters without any apparent chain formation, and exhibited abnormal morphology relative to the wild-type strain. In addition, the capsule of ΔluxS cells was signi cantly thinner than that of wild-type strain. These morphology phenotypes can be restored in part by CΔluxS. However, the differences between the overexpression strain luxS+ and wild-type strain were not signi cant (Fig. 1). Gram stain results showed that the morphological characteristics of the four strains consistent with the results of SEM (data not shown).

Growth curves
The growth curves of wild-type, ΔluxS, CΔluxS and luxS+ SS2 strains are presented in Fig. 2. Compared with the wild-type strain, the mutant strain ΔluxS did not show growth defects. For complementary strains, the difference between CΔluxS and overexpression strain luxS+ is not signi cant.
ΔluxSmutant and wild-type strain transcriptome analysis Analysis of transcriptome data shows that luxS gene mutation has a wide-ranging effect on the gene expression of SS. There were 1978 identically expressed genes in the SS wild-type strain and the mutant strain ΔluxS. In addition, there were 179 up-regulated genes and 99 down-regulated genes (Fig. 3A). Obviously, the amino acid ABC transporter permease gene expression was up-regulated most and acetyltransferase gene, cell division genes was down-regulated most. Gene Ontology (GO) analysis identi ed the biological functions of differentially expressed genes, and found that the main enrichment of differential genes was in biological processing and molecular functions. There was no signi cant enrichment of differential genes in cellular components, as shown in Fig. 3B.

qRT-PCR to verify changes in gene expression
In order to verify the reliability of the RNA-seq results, 6 differentially expressed genes were randomly selected for veri cation (Fig. S1). The results showed that SSU05_2024, SSU05_1111, and SSU05_1069 were signi cantly up-regulated, and SSU05_0050, SSU05_0087, and SSU05_0302 were signi cantly down-regulated. The above results are consistent with the sequencing results (P 0.05).

Acid tolerance of SS
The acid tolerance assay suggested that compared with pH values of 5.0, 6.0, 7.0, the OD 600nm values of the four strains (HA9801, ΔluxS, CΔluxS, and luxS+) at pH = 3.0 and pH = 4.0 decreased signi cantly. In addition, there is no difference in growth status among the HA9801, CΔluxS, and luxS+ strains in acidic environments. However, the mutant strain ΔluxS showed signi cantly decreased survival in acidic environments ranging from pH 3.0 to pH 5.0 at tested times compared with wild-type strain (Fig. 4). Moreover, complementation of luxS gene restored the acid resistance for the complemented strain. These results indicated that luxS gene contributed to the acid tolerance of SS2.

Fe stress response
The effects of exogenous Fe 2+ and Fe 3+ on the growth of SS were determined. It was observed that wildtype and luxS+ strains, in the presence of 3 mmol/L iron chelator were not signi cantly impacted by the metabolic stressor (P > 0.05). However, mutant strain ΔluxS was observed to have a signi cantly reduced OD 600nm value (P < 0.001) compared with wild-type strain. Moreover, the growth capacity of CΔluxS was retored by the luxS gene complementation. The effect of Fe 2+ on the growth of ΔluxS was more pronounced than that of Fe 3+ (Fig. 5).

Oxidative stress response
To assess the capability of the luxS gene to manage oxidative stresses, survival of the SS2 wild-type, ΔluxS, CΔluxS and luxS+ strains were measured after 1 h, 2 h, 3 h or 4 h of H 2 O 2 treatment. The results indicated that the wild-type strain were more susceptible to H 2 O 2 treatment (58.8 mmol/L) than the mutant strain ΔluxS. Survival rate of the CΔluxS and wild-type strains were signi cantly different from that of ΔluxS (P 0.05), and it was very signi cant with luxS+ strain (P 0.01). Therefore, it can be concluded that luxS is associated with the antioxidant activity of SS2 (Fig. 6).

Discussion
The highly conserved luxS gene has been extensively studied in recent years due to its involvement in the regulation of the expression of various growth and virulence-related genes [17,18]. The related AI-2 is a compound that plays a key role in bacterial cell-to-cell communication [19]. In this study, we compared the cell morphology and response of mutant strain ΔluxS and wild-type strain HA9801 to different stress conditions. The transcriptome differences between HA9801 and ΔluxS were also determined. Our results showed that mutant strain ΔluxS showed a large transcriptional difference and signi cantly different tolerance to stress environments.
We found that the growth rate of mutant strain ΔluxS was similar with those of wild-type, CΔluxS and luxS + strains. This experimental results corroborate the ndings of Zhang et al. [20] and Van et al. [21], in which the absence or overexpression of the luxS gene no affecting the expression of other downstream genes important for bacterial growth. Previous studies also reported that luxS + did not increase the level of AI-2 production, and no affected the growth of SS [1].
Acid resistance is a necessary for bacterial survival in acidic environments and during infection of the host through the digestive tract. In the present experiment, the acid stress test revealed an overall downward trend in vitality in acidic environmental conditions. The downward trend of ΔluxS was more pronounced than that of wild-type strain, and the viability decreased precipitously when the pH < 5.0. Acid stress on luxS + showed slightly stronger acid resistance than wild-type strain. The loss of the luxS gene results in many altered traits, including thinning of the bacterial capsule, which may be a cause of the altered acid resistance. In addition, with the down-regulated of the cell division protein genes SSU05_0760 and SSU05_0761, the ΔluxS strain showed a thin capsule structure. The results were similar to Cao et al [22]. At same, previous studies have also found that luxS gene is involved in bio lm formation [9]. At pH = 3.0, the wild-type and ΔluxS strains had not yet been completely killed, suggesting that the luxS gene is part of a redundant system through which SS2 creates its acid resistance. These observation suggest that acid regulation is a complex process that may be in uenced through a variety of regulatory pathways [23], the SS2 luxS gene appears to be part of one such regulatory pathway.
Iron is actually an essential element in all cells because iron is a cofactor in many enzymes, especially central metabolism and respiratory enzymes, so it is a coenzyme. Wen et al. [24] found that the growth inhibition of ΔluxS in the medium containing 0.3 mmol/L 2,2'-bipyridine was alleviated by adding exogenous iron and culture supernatant regulated by wild-type strains. In addition, Lee et al. [25] used to the gene chip to analyze and compare the transcription data of wild-type and mutant ΔluxS strains of Streptococcus pneumoniae at 4 different growth phages. It was found that two TonB systems are involved in iron absorption. Furthermore, tonB1-exbB1-exbD1 and tonB2-exbB2-exbD2 are affected by the luxS gene in all four growth stages [26]. In the present study, the results showed that the addition of exogenous Fe 2+ resulted in reduced growth of the wild-type strain with increasing ion concentration at 6 h and 12 h. The effects caused by Fe 3+ were more pronounced than those of Fe 2+ on growth promotion of the ΔluxS mutant strain, although the difference was not signi cant (P > 0.05). There was no signi cant difference observed in the growth status between wild-type and luxS + strains (P > 0.05). In comparison to the wild-type and luxS + strains, the growth of the ΔluxS mutant was observed to be signi cantly decreased (P < 0.001). The growth ability CΔluxS did recover, albeit not to the level of the wild-type strain.
Besides, in the mutant strain, the expression of the cobalamin / Fe 3+ -iron carrier ABC transporter ATPase gene SSU05_0650 was up-regulated, up to 1.976 times. We speculate that the reason why the mutant strain is more sensitive to Fe 2+ than Fe 3+ in the experiment may be due to the enhanced ability of the mutant strain to transport Fe 3+ ions [27]. Taken together, the data suggest that the luxS gene is involved in SS iron absorption, and the regulation of bacterial growth.
Oxidative stress is one defense mechanism of the host against an invading pathogen. Bacterial infections must rst overcome the host's stress defense mechanism. Yu et al. [28] knocked out the luxS gene of Yersinia pestis, and observed that the mutant exhibited a reduced resistance to H 2 O 2 . However, the Porphyromonas gingivalis ΔluxS strain exhibited increased survival in the presence of H 2 O 2 [29], this nding is similar to what was previously reported by Cao et al. [30] with respect to a H 2 O 2 stress test of ΔluxS of SS2 strain 05ZYH33. The results of this experiment showed that mutant strain ΔluxS were more tolerant to H 2 O 2 than the wild-type strain. This suggests that luxS genes have different roles in different bacterial species. After analyzing comprehensive transcriptome data, we believe that the differences in stress resistance of ΔluxS and wild-type strains may result in phenotypic differences due to the abnormal transcription of some genes after the luxS gene is mutant. Transcriptome data showed that the SS spxA transcriptional regulator SSU05_1111 and the possible oxidative stress-related gene SSU05_1069 were found in ΔluxS strains, and their expressions were up-regulated by 2.5 and 2.1 fold, respectively. It has a certain tolerance to H 2 O 2 compared with the wild strain; research shows that spxA transcriptional regulators play an important role in oxidative stress response of SS. spxA mutant strains are more sensitive to the oxidative environment of SS [31,32].

Conclusion
In summary, we compared the transcriptome differences and tolerance to various stress environments between mutant strain ΔluxS and wild-type strain. It has been indicated that quorum sensing system luxS gene is of great signi cance to the morphological structure as well as stress resistance of SS2.

Bacterial strains and culture conditions
The SS2 virulent wild-type strain HA9801 was isolated from pigs in the Jiangsu Province in 1998. The mutant strain ΔluxS, complemented strain CΔluxS and overexpression strain luxS+ of HA9801 was constructed previously [1,2]. The SS strains were grown at 37 °C in Todd Hewitt broth (THB) (Becton, Dickinson and Company, USA) medium or plated on THB agar with 5% (vol/vol) sheep blood.

Morphological characteristics
The morphological differences were determined by scanning electron microscopy according to the method previously described [33]. Brie y, coverslips with SS2 cultures were rinsed 3 times with a phosphate buffered solution. The samples were then post-xed for 90 minutes with 1% (w/v) osmium tetroxide in a 0.1 M sodium cacodylate buffer. After staining, the specimens were dehydrated in increasing concentrations of acetone (10%, 30%, 50%, 70%, 90%, and 100%). The specimens were then air-dried for 60 minutes, and were then adhered to metal holders with double-sided tape for coating with gold and palladium in an evaporator. All specimens were positioned with the apices facing up for proper visualization by scanning electron microscopy (SEM) in a vacuum at 5 kV electron beam energy (Hitachi S4700 FESEM; Hitachi Ltd, Tokyo, Japan). The bacterial morphology was also observed by gram staining and optical microscope (OM).

Growth curve
The logarithmic growth phase SS2 cultures were diluted 1: 200 to achieve an optical density at 600 nm (OD 600 nm ) of approximately 0.05. These cultures were incubated at a constant temperature shaking incubator at 37 ℃, shaking at 120 rpm. The OD 600nm values of the cultures were measured at 1 h interval using a spectrophotometer.

RNA-seq analysis
The experimental operation was performed as previously described with minor modi cations [34]. The strains SS HA9801 and ΔluxS were cultured in THB medium for 6 h, and harvested at 8000×g at 4 °C for 10 min. Then, the total RNA was extracted with the Trizol Reagent kit. Three biological replicates were set for each sample, and all samples were sent to Beijing Novogene Co., Ltd. for sequencing by Illumina Hiseq platform. Quality control was performed on the clean reads, and mapping was performed with reference to the SS 05ZYH33 genome. Gene function annotation was performed through the orthologous group (COG) database [35].

Validation of mRNA-Seq by qRT-PCR
The qRT-qPCR method was used to verify the expression results of mRNA-Seq in the transcriptome. Use Total RNA Extraction Kit (Solarbio, R1200) to extract total RNA, RNase-free DNase I to remove genomic DNA. The cDNA was ampli ed using MagicSYBR mix (CoWin Biosciences Co., Ltd., China). The volume of the ampli cation mixture was set to 20 μl (2× MagicSYBR Mixture 10 μl, each primer of 0.5 μM, cDNA 1 μg, nally add RNase-free water to 20 μl). The PCR reaction conditions were as follows: at 95 °C for 30 s, then at 95 °C for 5 s, and then at 60 °C for 30 s for 40 cycles. Randomly select 6 genes, and use 16S rRNA as internal reference to verify the original data. Table S1 lists all primers.

Acid stress assay
To assess the sensitivity of SS strains to acid stress conditions, we carried out an acid stress assay as previously described [36], with slight modi cation. Liquid THB media was prepared with pH values ranging from 3.0 to 7.0 (adjusted with 6N HCl). Approximately 1 3×10 6 CFU SS2 bacterial suspension were inoculated at a ratio of 1:10 (v/v) at 37 ℃ for 12 h or 24 h under aerobic conditions. Growth kinetics of each strain was measured by monitoring OD 600nm values under various conditions.

Fe stress assay
The Fe stress response assay was performed as previously described [37], with some modi cations. All strains were sub-cultured at the same original cell density from two subculture in iron restricted THB medium (containing 100 mM iron chelator 2,2'-dipyridyl). Then, transfer all the strains (HA9801, ΔluxS, CΔluxS, +luxS) grown in iron-restricted THB medium to the same volume of fresh THB medium. Then all the strains cultures were diluted 1: 200 (v/v), and different concentrations of iron ions were added to THB broth, of which Fe 2+ (0.1, 0.2 and 0.3 mmol/L) and Fe 3+ (0.1, 0.2 and 0.3 mmol/L). The cultures were incubated at 37 ℃ with aeration. The above cultures were incubated for 6 h (exponential phase) or 12 h (stable period), and centrifuged at 3000 rpm for 10 min at 4 ℃. The bacterial cell pellets were resuspended in the same volume of PBS buffer, and 250 μL of the mixture was added to a 96-well microtiter plate. The assays were performed in triplicate and the OD 600nm values were measured.
Oxidative stress assay Assessment of the bacterial cells abilities to withstand H 2 O 2 challenges was determined as previously described [38]. Brie y, SS2 cultures were incubated in THB media until mid-exponential phase (OD 600nm ≈ 0.8). For H 2 O 2 challenge, bacterial cells were prepared similarly, and then incubated in THB containing 58.8 mmol/L H 2 O 2 for 1 h, 2 h, 3 h or 4 h. After exponentially reasonable dilution, 10 μL of the dilution was spread on THA medium, incubated at 37 °C for 24 h and counted, the bacterial concentration was calculated, and the survival rate was calculated.

Statistical analysis
The Graphad Prism software was used to perform statistical analyses for all data. All data points for the experiments, performed in triplicate, were analyzed using the single factor analysis of variance (One-Way ANOVA), where P < 0.05 was considered to be statistically signi cant.

Declarations
This work was supported by the National Natural Science Foundation of China (31902309 31772761).

Availability of data and materials
The data sets used during the current study are available from the corresponding author on reasonable request.
Ethics approval and consent to participate Not applicable.

Con ict of Interest
The authors declare no con ict of interest.